burdenko of neurosurgery · issn 0042-8817 (russian ed. print) issn 2313-8254 (english ed. online)...

ISSN 0042-8817 (Russian ed. Print)ISSN 2313-8254 (English ed. Online)

ISSN 2309-1681 (Russian ed. Online)

FUNDAMENTAL AND PRACTICAL JOURNAL

Burdenko′s Journal of Neurosurgery

Vol. 82 1'2018

Burdenko Neurosurgical Institute, Moscow, RussiaOfficial journal of the Association of Neurosurgeons of Russia

«Zhurnal voprosy neirokhirurgii imeni N N Burdenko» (Burdenko′s Journal of Neurosurgery) is a bimonthly peer-reviewed medical journal published by MEDIA SPHERA Publishing Group. Founded in 1937.

Sponcored by fund «Neuro»

Journal is indexed in RSCI (Russian Science Citation Index), Web of Science (Russian Science Citation Index — RSCI), Scopus, PubMed/Medline, Index Medicus, Chemical Abstracts, Ulrich’s Periodicals Di-rectory, Google Scholar.

MEDIA SPHERA Publishing Group:Dmitrovskoe sh. 46-2, Moscow, 127238 RussiaTel. +7 (495) 482 4329Fax: +7 (495) 482 4312E-mail: [email protected]

Correspondence address:Moscow, P.O. Box 54, 127238 RussiaMedia SpheraAdvertising department: +7 (495) 482 0604E-mail: [email protected] department: +7 (495) 482 5336E-mail: [email protected]

Address of the editorial office:4-ya Tverskaya-Yamskaya ul., 16, Moscow, 125047RussiaBurdenko Neurosurgical InstituteTel. +7 (499) 972 8566E-mail: [email protected] EditorV.K. IvannikovaE-mail: [email protected]

Art and Layout: MEDIA SPHERA Publishing Group

The Editorial Board is not responsible for the content of advertising materials. Editorial opinion does not always coincide with the opinion of the authors. Only the articles prepared in compliance with Authors’ guidelines are accepted for publication. When submitting an article to the Editorial Board, the authors accept the terms and conditions of the public offer agreement. Authors’ guidelines and the public offer agreement can be found on website www.mediasphera.ru. Complete or partial repro-duction is allowed by written permission of the Publisher (MEDIA SPHERA Publishing Group).

EDITORIAL BOARD

Editor-in-Chief A.N. Konovalov Deputy Editor-in-Chief O.N. Dreval’

Executive Editor A.V. KozlovScience Editors B.A. Kadashev, O.B. Belousova

EDITORIAL COUNCIL

S.K. Akshulakov (Astana, Kazakhstan), S.R. Arustamyan (Moscow, Russia), A.Yu.Belyaev (Moscow, Russia), V.P. Bersnev (St. Petersburg, Russia), O.A. Gadzhieva (Moscow, Russia), B.V. Gaydar (St. Petersburg, Russia), A. Grigoryan (Macon, USA),S. Dambinova (Atlanta, USA), J.-M. Derlon (Paris, France), V.N. Dobzhansky (Moscow, Russia), G.F. Dobrovol’sky (Moscow, Russia), G. Enikolopov (Cold Spring Harbor, USA) Yu.A. Zozulya (Kyiv, Ukraine),D.N. Kapitanov (Moscow, Russia), E.N. Kondakov (St. Petersburg, Russia), A.N. Kondratyev (St. Petersburg, Russia), A.A. Lutsik (Novokuznetsk, Russia), V.N. Nikolenko (Moscow, Russia), E.G. Pedachenko (Kyiv, Ukraine), D.A. Ptashnikov (St. Petersburg, Russia), A.A. Reutov (Moscow, Russia), Sh.M. Safin (Ufa, Russia), D.V. Svistov (St. Petersburg, Russia), N.K. Serova (Moscow, Russia), K. Slavin (Chicago,USA), A.F. Smeyanovich (Minsk, Belarus), A.V. Smolin (Moscow, Russia) A. Spallone (Rome, Italy), S. Spektor (Tel-Aviv, Israel), M.V. Talabaev (Minsk, Belarus), R.V. Fanardgyan (Erevan, Armenia), A.P. Fraerman (Nizhny Novgorod, Russia), V.A. Khil’ko (St. Petersburg, Russia), V.I. Tsymbalyuk (Kyiv, Ukraine), V.Yu. Cherebillo (St. Petersburg, Russia), O.I. Shcherbenko (Moscow, Russia), Yu.A. Sherbuk (St. Petersburg, Russia), I.V. Yakovenko (St. Petersburg, Russia), S.B. Yakovlev (Moscow, Russia), M. Yankelevich (Detroit, USA)

G.I. Antonov, A.Kh. Bekyashev, Sh.Kh. Gizatullin, A.V. Golanov, S.K. Gorelyshev, Yu.A. Grigoryan, A.A. Grin, A.O. Gushcha, G.V. Danilov, G.Yu. Evzikov, A.M. Zaitsev, O.S. Iskhakov, P.L. Kalinin, V.B. Karakhan, G.M. Kariev, G.L. Kobyakov, N.A. Konovalov, V.N. Kornienko, A.G. Korshunov, A.L. Krivoshapkin, V.V. Krylov Yu.V. Kushel, V.A. Lazarev, L.B. Likhterman, A.Yu. Lubnin, A.G. Melikyan, A.G. Nazarenko, V.V. Nazarov, V.E. Oliushin, A.L. Parfenov, S.S. Petrikov, A.A. Potapov, I.N. Pronin, D.A. Rzaev, A.S. Saribekyan, Zh.B. Semenova, M.A. Stepanyan, A.A. Sufianov, S.V. Tanyashin, V. V. Timirgaz, T.P. Tissen, A.A. Tomsky, D.Yu. Usachev, V.A. Khachatryan, V.A. Cherekaev, G.G. Shaginyan, V.N. Shimanskiy, L.V. Shishkina, Sh.Sh. Eliava

© Media Sphera, 20182

ORIGINAL ARTICLES

Aleksandrova E.V., Batalov A.I., Pogosbekyan E.L., Zakharova N.E., Fadeeva L.M., Kravchuk A.D., Pronin I.N., Potapov A.A. New Possibilities Of Magnetic Resonance Imaging: Tractography CSD-HARDI Algorithm In Brainstem Reticular Formation Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4Kadashev B.A., Konovalov A.N., Astaf’eva L.I., Kalinin P.L., Kutin M.A., Klochkova I.S., Fomichev D.V., Sharipov O.I., Andreev D.N. Endocrine Disorders Prior and After Surgery for Pituitary Stalk Lesion With Suprasellar Tumors . . . . . . . . . . . . . . . . . . . .11Klimov V.S., Kel’makov V.V., Chishchina N.V., Evsyukov A.V. Efficiency of Intraoperative Monitoring of Motor Evoked Potentials in Predicting Early Postoperative Neurological Status in Patients With Cervical Spinal Cord Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . .19Lvov I.S., Grin’ A.A., Nekrasov M.A., Kordonskiy A.Yu., Sytnik A.V. Ventral Decompression in Patients With Traumatic and Non-Traumatic Atlanto-Axial Dislocations . . . . . . . . . . . . . . . . . . .28Lisitskiy I.Yu., Kiselev A.M., Kiselev S.E. Rheumatoid Atlantoaxial Dislocations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35Kalinovskiy A.V., Rzaev D.A., Yoshimitsu K. Opect Optical System in Neurosurgical Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41Moskalev A.V., Gladkikh V.S., Al’shevskaya A.A., Kovalevskiy A.P., Sakhanenko A.I., Orlov K.Yu., Konovalov N.A., Krut’ko A.V. Evidence-Based Medicine: Propensity Score Matching (PSM) Approach to Eliminate Systematic Selection Bias in Retrospective Neurosurgical Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44Pilipenko Yu.V., Konovalov An.N., Eliava Sh.Sh., Belousova O.B., Okishev D.N., Sazonov I.A., Tabasaranskiy T.F. Advisability and Effectiveness of Decompressive Craniectomy in Patients With Subarachnoid Hemorrhage After Microsurgical Repair of Aneurysmsn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49Kasparova Evg.A., Sobkova O.I., Kasparova E.A., Kasparov A.A. Surgical Treatment of Suppurative Corneal Ulcers Due to Neuroparalytic Keratitis . . . . . . . . . . . . . . . . . . . . . . . . . . .60

CASE REPORTS

Kadyrov Sh.U., Konovalov A.N., Pronin I.N. Mr-Tractography in Diagnosis and Choice of Neurosurgical Approach for Subcortical Nuclei Tumors . . . . . . . . . . . . . . . . . .64Sharipov O.I., Kutin M.A., Bayuklin A.V., Imaev A.A., Abdilatipov A.A., Kurnosov A.B., Fomichev D.V., Mikhaylov N.I., Kalinin P.L. Platelet Gel for Repair of Skull Base Liquoric Fistula (Case Report and Literature Review) . . . . . . . . . . . . . . . . . . . . . . .71

REVIEWS

Lubnin A.Yu. Awake Neurosurgery: Forward to the Past . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77Gol’bin D.A., Cherekaev V.A. Variability and Age-Related Anatomic Features of Midline Structures of Anterior Skull Base . . . . . . . . . . . . . . . . . . . . . .85Akulov M.A., Orlova O.R., Tabashnikova T.V., Karnaukhov V.V., Orlova A.S. Neurosurgery Followed by Facial Nerve Injury: Rehabilitation Potential of Botulinum Therapy . . . . . . . . . . . . . . . . . . . . .94

CONTENTS

3

Topics to be covered in our next issue

● Professional satisfaction of neurosurgeons working in the Russian Federation

● WHO classification of Primary Tumors of the Central Nervous System 2016: Clinician's View

● Spinal cord epidermoids

In accordance with the resolution of the Higher Attestation Commission of the Ministry of Education and Science of the Russian Federation, the Problems of Neurosurgery named after N.N. Burdenko was included in the List of Leading Peer-Reviewed Journals and Periodicals issued in the Russian Federation where the main results of Candidate and Doctor Theses are recommended to be published.

4 BURDENKO’S JOURNAL OF NEUROSURGERY 1, 2018

ORIGINAL ARTICLES

© Group of authors, 2018

Long-term brain researches have shown that recov-ery of consciousness, cognitive functions and emotional response is determined by severity of cerebral dysfunc-tion as a result of direct structural damage (trauma, isch-emia, hypoxia) as well as neurochemical and metabolic disorders. It is known that consciousness is ensured by ascending reticular activating system (ARAS) of the brainstem which includes ventral and dorsal pathways. The first one passes through the hypothalamus and basal parts of forebrain (histamine-, choline-, dopamine-, glutamatergic systems). Dorsal pathway activates cortex through the thalamus (mainly dopamine-, serotonin-, noradrenaline-, cholinergic systems) (Fig. 1). ARAS is not only a part of brain stem reticular formation from pons to thalamus, but also nuclei set of different neu-rotransmitter systems [1]. ARAS injury is followed by de-creased number of afferent projections of many neu-rotransmitter systems to cerebral cortex including those activating and maintaining consciousness.

Neuroimaging is very important to investigate mech-anisms of cerebral functions restoration in current neuro-science. It is able to determine in vivo focal and diffuse cerebral lesion, impaired blood flow in certain structures, qualitatively and quantitatively assess conductive paths by using of 3D-reconstruction, to analyze local or diffuse biochemical changes (MR-spectroscopy) [2—6]. In re-cent years, special attention has been paid to MR-trac-tography allowing 3D-reconstruction of various cerebral conductive pathways with creation of so-called connec-tome. Well-equipped projects "Human Brain Connec-tome" in the USA and "Human Brain Project" in the

*e-mail: [email protected]

ttps://doi.org/10.17116/engneiro20188214-10

New Possibilities Of Magnetic Resonance Imaging: Tractography CSD-HARDI Algorithm In Brainstem Reticular Formation MappingE.V. ALEKSANDROVA*, A.I. BATALOV, E.L. POGOSBEKYAN, N.E. ZAKHAROVA, L.M. FADEEVA, A.D. KRAVCHUK, I.N. PRONIN, A.A. POTAPOV

Burdenko Neurosurgery Institute, 4-ya Tverskaya-Yamskaya Str., 16, Moscow, Russia, 125047

The study purpose was to develop a technique for intravital visualization of the brainstem reticular formation fibers in healthy volunteers using magnetic resonance imaging (MRI). Material and methods. The study included 21 subjects (13 males and 8 females) aged 21 to 62 years. The study was performed on a magnetic resonance imaging scanner with a magnetic field strength of 3 T in T1, T2, T2-FLAIR, DWI, and SWI modes. A CSD-HARDI algorithm was used to identify thin intersecting fibers of the reticular formation Results. We developed a technique for reconstructing the reticular formation pathways, tested it in healthy volunteers, and obtained standard quantitative indicators (fractional anisotropy (FA), apparent diffusion coefficient (ACD), fiber length and density, and axial and radial diffusion). We performed a comparative analysis of these indicators in males and females. There was no difference between these groups and between indicators for the right and left brainstem. Our findings will enable comparative analysis of examination results in patients with brain pathology accompanied by brainstem injury, which may help predict the outcome. This work was supported by a grant of the Russian Foundation for Basic Research (№16—04-01472).

Keywords: reticular formation, ascending activating system, consciousness, tractography, CSD-HARDI, magnetic resonance imaging.

countries of the European Union confirm relevance of human conductive pathways research [7, 8].

Patterns of multivariate brain cleavage with degener-ation of commissural (interhemispheric), associative (hemispheric) and corticospinal conductive pathways have been previously shown on diffuse axonal damage model [9, 10]. At the same time, MR-tractography of re-ticular formation is under relatively little attention mainly due to difficult visualization of fine intersecting fibers of this structure. The first works for visualization of fibers and nuclei of reticular formation have been started re-cently [11—14]. Previous publications are based on post-humous cerebral examination of three patients with se-vere brain damage and several healthy volunteers. In these works, the authors used deterministic method of tractography on MR-tomograph (3 T) with Constrained Spherical Deconvolution-High Angular Resolution Dif-fusion Imaging (CSD-HARDI) algorithm. The advan-tage of this method is visualization of two or more axonal beams intersecting in different directions.

Currently, prognosis and treatment of severe cranio-cerebral trauma (CCT) has become particularly relevant due to socio-economic importance of the problem, since trauma is the leading cause of disability among employ-able persons. Definition of probability and grade of men-tal activity recovery is the key to assessing rehabilitation potential. However, domestic and foreign clinical criteria for unconscious states assessment do not always allow us to interpret correctly patient's consciousness. Recent studies have shown that about 40% of patients with clini-cal picture of vegetative status have signs of higher level of

5BURDENKO’S JOURNAL OF NEUROSURGERY 1, 2018

consciousness – state of minimal consciousness mani-festations [15].

At present time correction of impaired consciousness and other mental functions are aimed at changing of neu-rotransmitter systems’ functions. Recently, US Depart-ment of Defense have published review of 12 the most perspective pharmacological agents for treatment of CCT patients: 5 of them directly affect neurotransmitter sys-tems, and another 5 indirectly modulate their activi-ty [16]. It is important that there are still no objective guidelines for neurotransmitter disorders detecting. Clin-ical syndromes of glutamatergic, dopaminergic, cholin-ergic neurotransmitter systems dysfunction have been determined [17]. However, their instrumental confirma-tion with MR-spectroscopy and MR-tractography is ab-sent. By the way these methods are useful for intravital evaluation of cerebral neurotransmitter systems.

The purpose of this study is to obtain the normative data of MR-tractography of reticular formation (frac-tional anisotropy-FA, diffusion coefficient, fiber density and length, axial and radial diffusion), their comparative analysis in men and women for subsequent comparison with similar data in patients with severe brain damage fol-lowed by primary or secondary brain stem injury.

Material and methodsThe study included 21 healthy volunteers (13 men

and 8 women) aged 21 to 62 years (mean 31.3±9.6 years). All of them had higher education, one was left-hander. Informed consent to participate in the study was obtained from each volunteer.

CSD-HARDI MR-tractography. Scanning was carried out on 3.0 T MRI scanner General Electric Signa HD (GE Healthcare) with 8-channel head coil. Following

protocols were used in the study: T1 FSPGR BRAVO with an isotropic voxel 1×1×1 mm and zero gap, axial T2 with cut-off thickness 5 mm and gap between slices 1.5 mm, axial T2-FLAR with cut-off thickness 5 mm and gap 1 mm, DWI ASSET with cut-off thickness 5 mm and gap 1 mm, spectroscopy (single voxel, 2D CSI and 3D MRS), and HARDI protocol. HARDI data were ob-tained by using of SE DWI EPI sequence with the follow-ing parameters: TR=12,000 ms, TE-min, pixel band-width 3,906.25 Hz, 256×256 or 128×128 matrix, the number of slices is 30, 32, 34 or 36. Zero gap was estab-lished between slices, cut-off thickness was 2.5 mm, FOV=22×22, 24×24, 25×25 or 26×26 cm2, b-factor was 2,000, 3,000 or 4,000 s/mm2, the number of diffusion gradient directions — 120 or 110 with corresponding quantity of non-diffusion volumes (b=0 s/mm2): 1 or 11. In the second case non-diffusion volumes scanning was performed in every 10 directions of diffusion gradients. Overall time of scanning for HARDI algorithm was 24 min 25 s. Then diffusion images obtained were inter-polated to voxel size of 1×1×2.5 mm.

Pre-processing of diffusion images was carried out by using of FSL 5.0 (http://fsl.fmrib.ox.ac.uk/fsl/) and Ex-ploreDTI (http://www.exploredti.com/) software. Tracks were constructed in ExploreDTI program.

Paxinos and Huang atlas was used to identify the zone of interest ( brain stem cholinergic nuclei localiza-tion — pedunculopontinous and laterodorsal tegmen-tal) [18]. Image of such zone near 28 mm2 is shown in Fig. 2. All quantitative parameters were measured: mean FA and diffusion coefficient over entire length of fibers, number and length of fibers, axial and radial diffusion.

Statistical processing was carried out by using of Sta-tistica 8.0 software (Statsoft, USA). In all cases nonpara-metric criteria were used. Spearman rank correlation co-efficient and Mann-Whitney test were used to estimate correlations and to compare two independent samples, respectively. Differences were considered significant at p<0.05.

ResultsWe have developed the technique for visualization

and quantitative assessment of fine intersecting fibers of brain stem reticular formation (ARAS) by using of MRI CSD-HARDI algorithm (Fig. 3). Group of healthy volunteers of middle age was selected and analyzed for this purpose.

For each healthy volunteer two-sided 3D-recon-structions of ascending tracts were performed. They started from dorsolateral pons at the level of two closely located cholinergic nuclei (pedunculopontinous and lat-erodorsal tegmental) (Fig. 4). It can be seen that in all cases fibers are represented by two main bundles which pass through mesencephalon and divide into smaller beams extending to thalamic (dorsal tegmental tract) and

Fig. 1. ARAS nuclei and projections scheme superimposed on sagittal T1-MR-scan (24).


ORIGINAL ARTICLES

hypothalamic nuclei, basal parts of frontal lobes (ventral tegmental path) (Fig. 3c).

Quantitative characteristics including FA, diffusion coefficient, length and number of fibers, axial (L1) and radial (L2) diffusion for left (Table 1) and right (Table 2) tracts were obtained. It was shown that all these variables do not depend on the side of tract (left or right) and gen-der.

Thus, we have obtained control normal quantitative characteristics for evaluation of bilateral tracts of reticular formation arising from cholinergic nuclei in young em-ployable men and women.

DiscussionFor the first time we have developed the technique of

brain stem ARAS formation in healthy volunteers. We obtained averaged quantitative data for pons fibers of re-ticular formation where cholinergic nuclei are localized (FA, diffusion coefficient, fibers density and length, axial and radial diffusion). The last will allow further compar-ative analysis of reticular formation state in patients with brain injury by using of statistical processing. Firstly, ARAS visualization with CSD-HARDI and determinis-tic processing was demonstrated in 2 pathoanatomical cases and one healthy volunteer [11]. Also there were at-tempts to visualize fibers via formation through two tar-get points (brain stem and thalamus/hypothalamus) in healthy volunteers [19, 20]. However, these trials were performed on tomograph 1.5 T without CSD-HARDI algorithm, interesting zone included entire region of re-

ticular formation rather separate nuclei. So, it was shown that fibers of reticular formation are mainly projected onto certain areas of prefrontal cortex (lateral and ventro-medial), quantitative variables of reticular formation are similar for both hemispheres and in males and fe-males [19, 20].

Currently, there are more and more publica-tions [21—25] confirming relationship between impaired

Fig. 2. Region of interest (ROI) choice in 3D-visualization of cholinergic fibers of the reticular formation on 2D DTI color map. Pontine ROI at the level of maximum dimension of pedunculo-pontinous and laterodorsal tegmental nuclei (slice 35 mm above medullary obex, area of 28 mm2).

Fig. 3. Example of 3D-reconstruction of ascending and descending fibers of pontine cholinergic nuclei in healthy volunteer (female, 31 years old). a — sagittal plane; b — coronary plane; c — axial plane. VTT — ventral tegmental tract, DTT — dorsal tegmental tract.


consciousness and other mental functions including emotions with injury of some brain stem neurotransmit-ter systems. It is known that cholinergic system partici-pates in short-term memory, attention, emotions regula-tion, wakefulness level.

Acetylcholine is produced in following cerebral nu-clei: pontine reticular formation nuclei (pedunculoponti-nous (PPN) and laterodorsal (LDT) tegmental, inter-neurons of neostriatum and basal forebrain (Meynert nucleus, medial septal nucleus and Brock's diagonal strip which are together forming substantia innominate). PPN and LDT fibers as a part of paramedian pontine reticular formation are associated with vestibular nuclei, locus

coeruleus, cerebellum, thalamus and giant-cell preoptic nucleus [26].

Cholinergic nuclei of basal forebrain are more active in wakefulness than in sleep, while cholinergic neurons of PPN nucleus are predominantly activated during rapid eye movement sleep (REM-sleep) and are responsible for pontine-geniculo-occipital EEG complexes. It is consid-ered that PPN neurons provide decreased muscles tone during REM-sleep through activation of glycinergic in-hibitory systems [27].

Near 85—90% of brain stem cholinergic afferents are involved into innervation of various (specific and non-specific) thalamic nuclei. The most pronounced cholin-

Fig. 4. 3D-images of ascending and descending tracts of pontine reticular formation in projection of cholinergic nuclei in 20 healthy volunteers.

Table 1. Quantitative features of left-sided tracts of reticular formation from pontine cholinergic nuclei

Variable All sample (n=21)

Males (n=13)

Females (n=8)

р (Mann-Whitney test)

FA 0.31—0.49 (0.38±0.05)

0.31—0.49 (0.38±0.05)

0.33—0.49 (0.4±0.05)

0.3

Diffusion coefficient 0.45—0.97·10–3

(0.62±0.13·10–3)0.47—0.76·10–3

(0.59±0.11·10–3)0.45—0.97·10–3

(0.68±0.16·10–3)0.15

Length of fibers, mm 28—64.6 (47.8±11.1)

28—64.6 (46.2±12.2)

35.4–62.9 (50.5±9.1)

0.21

Density of fibers, U 13—40 (27.1±7.3)

21—40 (28.3±7.3) 13—36 (25.1±7.2)

0.5

L1 0.67—1.29·10–3 (0.9±0.2·10–3)

0.67—1.16·10–3

(0.84±0.18·10–3)0.67—1.29·10–3

(0.99±0.2·10–3)0.14

L2 0.39—0.87·10–3

(0.53±0.11·10–3)0.4—0.64·10–3

(0.5±0.1·10–3)0.39—0.87·10–3

(0.58±0.1·10–3)0.3

Footnote. Here and in Table 2: FA — fractional anisotropy; L1 — axial diffusion index; L2 — radial diffusion index.


ORIGINAL ARTICLES

Table 2. Quantitative features of right-sided tracts of reticular formation from pontine cholinergic nuclei

Variable All sample (n=21)

Males (n=13)

Females (n=8)

р (Mann-Whitney test)

FA 0.33—0.48 (0.39±0.04)

0.33—0.48 (0.38± 0.04)

0.33—0.48 (0.39±0.05)

>0.05

Diffusion coefficient 0.46—0.94·10–3

(0.623±0.13·10–3)0.46—0.82·10–3

(0.58±0.12·10–3)0.47—0.94·10–3

(0.68±0.15·10–3)>0.05

Length of fibers, mm 35.3—64.4 (48.2±8.3)

35.3—64.4 (48.4±9.6)

38.2—57.1 (47.9±6.1)

0.97

Density of fibers, U 15—35 (27.3±5.8)

17—35 (27.5±5.8)

15—34 (27±6.1)

0.86

L1 0.67—1.2·10–3

(0.9±0.2·10–3)0.67—1.2·10–3

(0.84±0.19·10–3)0.67—1.3·10–3

(0.98±0.2·10–3)0.21

L2 0.38—0.84·10–3 (0.5±0.1·10–3)

0.38—0.7·10–3

(0.5±0.095·10–3)0.41—0.84·10–3

(0.58±0.13·10–3)0.14

ergic innervation (from pontine PPN and LDT nuclei) is obtained by anterior intralaminar nuclei and additional paralaminar areas of thalamus, which are key for wake-fulness regulation [28]. Cholinergic afferents to thalamic nuclei have a double effect on their activity. On the one hand, they inhibit GABAergic neurons of thalamic re-ticular nucleus through M-cholinergic receptors. These neurons inhibit activating neurons of front intralaminar nuclei. So, cortex is activated. On the other hand, direct cholinergic projections activate front intralaminar nuclei via nicotinic receptors (N-cholinoreceptors) [28]. Brain stem cholinergic fibers also activate anteroventral tha-lamic nucleus that has projections to retrosplenial cortex (part of posteromedial cortex) [29]. Increased metabolic activity of this cortex is one of the neuroimaging markers of consciousness recovery onset. Thus, concomitant acti-vation of brain stem and forebrain cholinergic projections during wakefulness and REM-sleep ensures integrative thalamic function and can modulate conscious process-es. It is evidenced by some trials which have shown that cholinergic neurons of basal forebrain are responsible for cognitive evoked P300 potentials [30].

Besides the thalamus, brain stem cholinergic neu-rons innervate other cerebral structures providing wake-fulness, in particular glutamatergic (reticular formation of mesencephalon, pontine oral nucleus), cholinergic (basal forebrain) and prefrontal cortex [31]. It is impor-tant to note that cholinergic system is able to maintain active cerebral state without activation of monoaminer-gic (catecholaminergic, serotonergic) systems [32], but with obligatory activity of glutamatergic system [30].

Thus, we have obtained reticular formation tracts originating from dorsolateral pons where cholinergic nu-clei are localized. Our data correspond to concept of ana-

tomical connections of this structure. Further evaluation of cholinergic system in brain lesions may be important to predict consciousness and cognitive functions recov-ery. In general, non-invasive diagnosis of structural and metabolic biomarkers reflecting damage of various neu-rotransmitter systems in severe cerebral injury may be key to assess recovery of consciousness, intellectual-mnestic functions and emotional-volitional disorders. The last is necessary to develop personalized therapeutic-predictive model.

Limitations of HARDI should be considered. Identi-cal diffusion characteristics of all cerebral fibers are ac-cepted to simplify calculations [33]. CSD-HARDI method is sensitive to noise on the image [34, 35] and can give erroneous information about the tracts within grey matter and cerebrospinal fluid [36]. Tractography depends on various methodological factors: scanning pa-rameters such as b-factor and the number of diffusion gradients directions [37], and post-processing of data in-cluding noise and artifacts correction. It cannot be ruled out that pathways constructed can contain both descend-ing and ascending fibers since ARAS is polysynaptic structure with many internal and external associations.

The work is supported by RFFI grant No. 16—04-01472.

Acknowledgment to radiologist A.S. Tonoyan for par-ticipation in MR-tractography variables selection.

Authors declare no conflict of interest.


REFERENCES1. Zeman A. Consciousness. Brain. 2001;7:1263-1289. https://doi.org/10.1093/brain/124.7.12632. Pronin IN, Fadeeva LM, Podoprigora AE, Zakharova NE, Serkov SV, Ro-

dionov PV, Shul’ts EI, Korshunov AE, Usachev DYu, Lukshin VA, Celik A, Potapov AA, Kornienko VI. Spinovoe markirovanie arterial’noi krovi (ASL) — metod vizualizatsii i otsenki mozgovogo krovotoka. Luchevaya diagnostika i terapiya. 2012;3(3):64-78. (In Russ.).

3. Zaharova NE, Potapov AA, Kornienko VN, Pronin IN, Aleksandrova EV, Danilov GV, Gavrilov AG, Zajcev OS, Kravchuk AD, Sychev AA. Novaya klassifikatsiya travmaticheskikh porazhenii golovnogo mozga, osnovannaya na dannykh magnitno-rezonansnoi tomografii. Vestnik RFFI. 2016;2(90):12-19. (In Russ.).

4. Danilov GV, Zaharova NE, Potapov AA, Kornienko VN, Pronin IN, Gavrilov AG, Aleksandrova EV, Oshorov AV, Sychev AA, Polupan AA. Krovotok v stvole golovnogo mozga u patsientov s cherepno-mozgovoi trav-moi. Vestnik RFFI. 2016;2(90):33-40. (In Russ.).

5. Potapov AA, Zakharova NE, Kornienko VN, Pronin IN, Aleksandrova EV, Zaitsev OS, Likhterman LB, Gavrilov AG, Danilov GV, Oshorov AV, Sy-chev AA, Polupan AA. Neiroanatomicheskie osnovy travmaticheskoi komy: klinicheskie i magnitno-rezonansnye korrelyaty. Voprosy neirokhirurgii. 2014;1:4-13. (In Russ.).

6. Haacke EM, Duhaime AC, Gean AD, Riedy G, Wintermark M, Mukher-jee P, Brody DL, DeGraba T, Duncan TD, Elovic E, Hurley R, Latour L, Smirniotopoulos JG, Smith DH. Common data elements in radiologic im-aging of traumatic brain injury. J Magn Reson Imaging. 2010 Sep;32(3):516-543. https://doi.org/10.1002/jmri.22259

7. Bardin J. Neurodevelopment: unlocking the brain. Nature. 2012 Jul 4;487(7405):24-26. https://doi.org/10.1038/487024a

8. Toga AW, Clark KA, Thompson PM, Shattuck DW, Van Horn JD. Map-ping the human connectome. Neurosurgery. 2012 Jul;71(1):1-5.

https://doi.org/10.1038/mp.2017.929. Zakharova N, Kornienko V, Potapov A, Pronin I. Neuroimaging of trau-

matic brain injury. Springer International Publishing Switzerland. 2014;35-68. https://doi.org/10.1007/978-3-319-04355-5_3

10. Potapov AA, Konovalov AN, Kornienko VN, Kravchuk AD, Likhter-man LB, Pronin IN, Zakharova NE, Aleksandrova EV, Gavrilov AG, Go-ryainov SA, Danilov GV. Sovremennye tekhnologii i fundamental’nye issle-dovaniya v neirokhirurgii. Vestnik RAN. 2015;85:4: 299-309. (In Russ.). https://doi.org/10.7868/S086958731504009X

11. Edlow BL, Takahashi E, Wu O, Benner T, Dai G, Bu L, Grant PE, Gre-er DM, Greenberg SM, Kinney HC, Folkerth RD. Neuroanatomic con-nectivity of the human ascending arousal system critical to consciousness and its disorders. J Neuropathol Exp Neurol. 2012 Jun;71(6):531-546.

https://doi.org/10.1097/nen.0b013e318258829312. Edlow BL, Haynes RL, Takahashi E, Klein JP, Cummings P, Benner T,

Greer DM, Greenberg SM, Wu O, Kinney HC, Folkerth RD. Disconnec-tion of the ascending arousal system in traumatic coma. J Neuropathol Exp Neurol. 2013 Jun;72(6):505-523.

https://doi.org/10.1097/nen.0b013e3182945bf613 Edlow BL, McNab JA, Witzel T, Kinney HC. The Structural Connectome

of the Human Central Homeostatic Network. Brain Connect. 2016 Apr;6(3):187-200. https://doi.org/10.1089/brain.2015.0378

14. Bianciardi M, Toschi N, Edlow BL, Eichner C, Setsompop K, Polime-ni JR, Brown EN, Kinney HC, Rosen BR, Wald LL. Toward an in vivo neuroimaging template of human brainstem nuclei of the ascending arous-al, autonomic, and motor systems. Brain Connect. 2015 Dec;5(10):597-607.

https://doi.org/10.1089/brain.2015.034715. Schnakers C, Laureys S. Coma and disorders of consciousness. Springer.

2012;169. https://doi.org/10.1007/978-1-4471-2440-516. Diaz-Arrastia R, Kochanek PM, Bergold P, Kenney K, Marx CE,

Grimes CJ, Loh LT, Adam LT, Oskvig D, Curley KC, Salzer W. Pharmaco-therapy of traumatic brain injury: state of the science and the road forward: report of the Department of Defense Neurotrauma Pharmacology Work-group. J Neurotrauma. 2014 Jan 15;31(2):135-158.

https://doi.org/10.1089/neu.2013.301917. Aleksandrova EV, Zaitsev OS, Potapov AA. Klinicheskie sindromy dis-

funktsii neiromediatornykh sistem pri tyazheloi travme mozga. Zh Nevrol Psikhiatr im S.S. Korsakova. 2015;7:22-28. (In Russ.).

https://doi.org/10.17116/jnevro20151157140-4618. Paxinos G, Huang X. Atlas of the human brainstem. San Diego: Academic

Press, 1995.

19. Jang SH, Kwon HG.The ascending reticular activating system from pontine reticular formation to the hypothalamus in the human brain: a diffusion tensor imaging study. Neurosci Lett. 2015 Mar 17;590:58-61.

https://doi.org/10.1016/j.neulet.2015.01.07120. Jang SH, Kwon HG.The direct pathway from the brainstem reticular for-

mation to the cerebral cortex in the ascending reticular activating system: A diffusion tensor imaging study. Neurosci Lett. 2015 Oct 8;606:200-203.

https://doi.org/10.1016/j.neulet.2015.09.00421. Arciniegas DB. The cholinergic hypothesis of cognitive impairment caused

by traumatic brain injury. Curr Psychiatry Rep. 2003;5:391-399. https://doi.org/10.1007/s11920-003-0074-522. Bales JW, Wagner AK, Kline AE, Dixon CE. Persistent cognitive dysfunc-

tion after traumatic brain injury: A dopamine hypothesis. Neuroscience and Biobehavioral Reviews. 2009;33:981-1003.

https://doi.org/10.1016/j.neubiorev.2009.03.01123. Schiff ND. Recovery of consciousness after brain injury: a mesocircuit hy-

pothesis. Trends Neurosci. 2010;33:1-9. https://doi.org/10.1016/j.tins.2009.11.00224. Aleksandrova EV, Zaitsev OS, Potapov AA. Neiromediatornye osnovy soz-

naniya i bessoznatel’nykh sostoyanii. Voprosy neirokhirurgii. 2014;78(1):26-32. (In Russ.).

25. Venkatraman A, Edlow BL, Immordino-Yang MH. The Brainstem in emo-tion: a review. Front Neuroanat. 2017;1-12.

https://doi.org/10.3389/fnana.2017.0001526. Webster RA. Neurotransmitters, drugs and brain functions. Edited by Web-

ster RA. 2001;535. https://doi.org/10.1002/047084657727. Chase MH. Confirmation of the consensus that glycinergic postsynaptic

inhibition is responsible for the atonia of REM sleep. Sleep. 2008;31:1487-1491. https://doi.org/10.1093/sleep/31.11.1487

28. Schiff ND. Central thalamic contributions to arousal regulation and neuro-logical disorders of consciousness. Ann N Y Acad Sci. 2008;1129:105-118. https://doi.org/10.1196/annals.1417.029

29. Vogt BA, Laureys S. Posterior cingulate, precuneal and retrosplenial corti-ces: cytology and components of the neural network correlates of con-sciousness. Prog Brain Res. 2005;150:205-217.

https://doi.org/10.1016/s0079-6123(05)50015-330. Perry E, Ashton H, Young A. Neurochemistry of consciousness. Forework

by Greenfield S. John Benjamins publishing company. 2002;344. https://doi.org/10.1075/aicr.3631. Franks NP. General anaesthesia: from molecular targets to neuronal path-

ways of sleep and arousal. Nature reviews. 2008;370-386. https://doi.org/10.1038/nrn237232. Buzsaki G, Bickford RG, Armstrong DM, Ponomareff G, Chen KS,

Ruiz R, Thal LJ, Gage FH. Electric activity in the neocortex of freely mov-ing young and aged rats. Neuroscience. 1988 Sep;26(3):735-744.

https://doi.org/10.1016/0306-4522(88)90095-433. Tournier J-D, Calamante F, Gadian DG, Connelly A. Direct estimation of

the fiber orientation density function from diffusion-weighted MRI data us-ing spherical deconvolution. Neuroimage. 2004;23:1176-1185.

https://doi.org/10.1016/j.neuroimage.2004.07.03734. Tournier J-D, Yeh C-H, Calamante F, Cho K-H, Connely A, Lin C-P.

Resolving crossing fibres using constrained spherical deconvolution: valida-tion using diffusion-weighted imaging phantom data. Neuroimage. 2008;46:617-625. https://doi.org/10.1016/j.neuroimage.2008.05.002

35. Dell’Acqua F, Scifo P, Rizzo G, Catani M, Simmons A, Scotti G, Fazio F. A modified damped Richardson—Lucy algorithm to reduce isotropic back-ground effects in spherical deconvolution. NeuroImage. 2010;49:1446-1458. https://doi.org/10.1016/j.neuroimage.2009.09.033

36. Roine T, Jeurissen B, Perrone D, Aelterman J, Leemans A, Philips W, Si-jbers J. Isotropic non-white matter partial volume effects in constrained spherical deconvolution. Front Neuroinformatics. 2014;8.

https://doi.org/10.3389/fninf.2014.0002837. Tournier J-D, Calamante F, Connelly A. Determination of the appropriate

b value and number of gradient directions for high-angular resolution diffu-sion-weighted imaging. NMR Biomedicine. 2013;26:1775-1786.

https://doi.org/10.1002/nbm.3017

Received: 10.11.17


ORIGINAL ARTICLES

Commentaries

Modern technique of CSD-HARDI for visualization of conductive fibers passing through pontine cholinergic nuclei (pedunculopontinous and laterodorsal tegmental) is used in healthy volunteers. The authors obtained quantitative data about pontine reticular formation fibers where their cholinergic nuclei are localized. These data may be later used for compara-

E.V. Alexandrova et al. reported successful non-invasive visualization of brain stem reticular formation. Reticular for-mation is a complex anatomical structure consisting of thin in-tersecting descending and ascending fibers from the nuclei of various neurotransmitter systems (cholinergic, noradrenergic, serotonergic, glutamatergic, dopaminergic). There are 98 nu-clei in humans. For the first time this system of brain stem cells was described by Lenhossek in 1855, and Deiters called it re-ticular formation in 1865. Further trials showed that reticular formation’s volume decreases during phylogenetic develop-ment from lower forms to higher ones. Thus, in hedgehog it oc-cupies 39% of brainstem volume, in people — only 9%. There-fore, it is not surprising that such difficult and small system

could be only researched post mortem for a long time. Howev-er, current neuroimaging techniques including MRI HARDI algorithm make it possible to assess reticular formation fibers in vivo and to calculate some quantitative values for each person in certain area of interest. These data are undoubtedly important for further study of reticular formation in various types of cere-bral pathology, especially those accompanied by brain stem damage due to traumatic or vascular events, as well as in neuro-degenerative processes.

The article is relevant, characterized by scientific novelty and is of great importance for understanding structural organi-zation of human reticular formation under various pathological conditions.

I.N. Bogolepova (Moscow, Russia)

tive analysis of patients with brain stem injury and impaired consciousness. The article follows organically from previous literature for this issue and greatly advances our knowledge in this field. There are single works about this problem in the world literature at present time.

A.I. Kholodniy (New York, USA)


*e-mail: [email protected]© Group of authors, 2018

https://doi.org/10.17116/engneiro201882111-18

Endocrine Disorders Prior and After Surgery for Pituitary Stalk Lesion With Suprasellar TumorsB.A. KADASHEV, A.N. KONOVALOV, L.I. ASTAF’EVA*, P.L. KALININ, M.A. KUTIN, I.S. KLOCHKOVA, D.V. FOMICHEV, O.I. SHARIPOV, D.N. ANDREEV


The pituitary stalk (PS) is a relatively thin bundle connecting the hypophyseal stalk to the pituitary gland; it consists of both axons of the hypothalamic nuclei (terminating in the neurohypophysis) and the system of portal vessels. Compression of the PS by a space-occupying lesion or its transection (forced or intended) during surgery may lead to the development of endocrine disorders: hypopituitarism, diabetes insipidus, and hyperprolactinemia. The modern literature lacks studies evaluating the severity of endocrine disorders depending on the PS condition before and after surgery. Purpose. The study purpose was to investigate endocrine disorders in patients with sellar region (SR) tumors and the PS that was compressed before surgery and preserved or transected during a neurosurgical intervention. Material and methods. The study included 139 patients with various SR tumors. In 82 patients, a preoperatively compressed PS was preserved (41 patients with hormonal inactive adenoma (HIA) and 41 patients with suprasellar meningioma); in 57 patients, the PS was transected during surgery (46 patients with pituitary stalk craniopharyngioma and 11 patients with hormonally inactive endosuprasellar pituitary adenoma). The hormonal status (PRL, TSH, LH, FSH, fT4, cortisol, testosterone, or estradiol) was examined in all patients both before and after surgery. Results. Hyperprolactinemia was preoperatively detected in 37% of patients with tumors compressing the PS. Elimination of PS compression (tumor resection) led to normalization of the PRL level in most patients and was not accompanied by aggravation of hypopituitarism symptoms. Transection of the PS caused panhypopituitarism in 100% of patients and diabetes insipidus in 93% of cases. After transection of the PS, hyperprolactinemia did not develop in 59% of patients with craniopharyngiomas (CPs) and 82% of patients with HIA. Conclusions. Given the difference in symptoms associated with compression and surgical transection of the PS, we believe that these two concepts should be clearly distinguished. The PS compression syndrome includes primarily hyperprolactinemia (37% of cases); elimination of PS compression leads to normalization of the PRL level in most patients and is not accompanied by aggravation of hypopituitarism symptoms. The PS transection syndrome in patients with CP and HIA led to the development of panhypopituitarism in all patients and permanent diabetes insipidus in most of them. The causes of the absence of hyperprolactinemia in many patients with PS transection require further research. The surgeon planning intraoperative PS transection to increase the radicality of surgery should be well informed about the consequences of this procedure for the patient’s endocrine status.

Keywords: pituitary stalk, pituitary stalk compression, pituitary stalk transection, diabetes insipidus, prolactin.

Abbreviations:PA — pituitary adenomaADH — antidiuretic hormoneACTH — adrenocorticotropic hormoneHIA — hormone-inactive adenomaCP — craniopharyngiomaLH — luteinizing hormoneDI — diabetes insipidusPRL — prolactinFr.T4 — free thyroxinePS — pituitary stalkTSH — thyroid stimulating hormoneFSH — follicle-stimulating hormoneCSR — chiasmatic-sellar regionPSIS — pituitary stalk interruption syndrome (congenital absence of pituitary stalk)

Moreover, attitude of many neurosurgeons to PS is often incorrect due to unclear understanding of its func-tion.

We have compared endocrine disorders in patients with PS compression by tumor and/or its intraoperative intersection.

So-called intersected PS syndrome can occur in var-ious cases: compression with suprasellar tumors (PA, meningioma, CP, etc.), intraoperative iatrogenic inter-section, traumatic rupture, PSIS.

Despite similar symptoms (hyperprolactinemia, DI, hypopituitarism) treatment and prognosis are signifi-cantly different depending on the pathogenesis.


ORIGINAL ARTICLES

erative normalization and 5 patients with normal preop-erative PRL level had hyperprolactinemia after surgery. Maximum postoperative PRL level was 3,600 IU /l. 2 pa-tients had hypoprolactinemia (35 and 50 IU/l).

The same outcomes were observed in the 2nd group (transcranial removal of hormone-inactive PA followed by PS intersection).

All 11 patients had hypopituitary disorders de novo or their aggravation. DI developed in 10 out of 11 patients. All patients in this group needed for postoperative hor-mone replacement therapy.

Moderate preoperative hyperprolactinemia was ob-served in 4 cases. PA excision followed by PS intersection of SG was associated with increased level of PRL only in 2 patients. Hypoprolactinemia occurred in 1 patient (33 IU/L).

In group 3 (transnasal PS-sparing PA excision) some patients with preoperative hypopituitary symptoms had their regression after surgery: sexual function recovery was observed in 5 (15%) out of 33 patients, hypothyroid-ism relief — in 3 (21%) out of 14 cases, and hypocorti-cism regression — in 2 (18%) out of 11 patients.

In this group incidence of hypopituitary disorders in patients without previous ones was relatively small: hypo-gonadism developed in 1 (12.5%) case, hypocorti-cism — in 1 (3%), hypothyroidism — in 1 (4%) patients. There were no cases of DI. Moderate preoperative hyper-prolactinemia was detected in 16 (39%) cases; tumor ex-cision was followed by decrease of blood PRL in majority of cases.

In 13 (32%) patients with CSR meningiomas (group 4) preoperative hyperprolactinemia was detected; maxi-mum PRL level was 4050 IU/l. Moderate postoperative hyperprolactinemia occurred in 2 (15%) out of 13 pa-tients with increased preoperative level of PRL.

Surgery for suprasellar meningiomas without PS in-tersection in our sample was practically not accompanied by occurrence/aggravation of hypopituitary disturbanc-es. Hypogonadism was observed in 1 case, DI remained in 1 (2%) patient, DI de novo was absent.

DiscussionPituitary stalk is anatomical part consisting of portal

vessels and hypothalamic nuclei axons which terminate in posterior pituitary lobe (Fig. 2).

Current ideas about anatomy and pathophysiology of hypothalamic-pituitary system were comprehensively described by I.I. Dedov, I.G. Akmayev, A.A. Voitkevich, V.N. Babichev [1—4] and some other domestic and for-eign authors.

It is known that pituitary gland supply is provided by small branches of internal carotid artery — superior and inferior pituitary arteries [5]. Superior pituitary arteries enter median hypothalamic eminence and form capillary network (Figs. 2 and 3). These capillaries have a contact with axons of neurosecretory hypothalamic cells and

The aim of the study was an analysis of pre- and postoperative endocrine disorders in patients with supra-sellar CSR tumors followed by PS lesion.

Material and methodsRetrospective study included 139 patients with CSR

tumors who have undergone surgery at the Institute of Neurosurgery from 2000 till 2016.

Patients were divided into 4 groups depending on tu-mor histology, its localization, pituitary gland state, and also whether PS was intraoperatively intersected (Fig. 1):

— group 1: 46 patients with CP of stalk (surgery was carried out via transcranial approach). In these cases, initial place of tumor growth was PS per se that required its intersection. Pituitary gland was intact prior to and af-ter surgery;

— group 2: 11 patients with hormone-inactive endo-suprasellar PA (transcranial surgery was applied). PS was compressed by adenoma that required its intersection during suprasellar capsule resection in order to improve resection quality and reduce likelihood of recurrence. Pi-tuitary gland was anatomically altered before and after surgery;

— group 3: 41 patients with hormone-inactive endo-suprasellar PA followed by compression of pituitary gland and PS (PS-sparing tumor excision through endoscopic endonasal transsphenoidal approach). Tumor’s capsule was not excised, pituitary gland was anatomically altered before and after surgery.

— group 4: 41 patients with suprasellar meningioma followed by PS compression, stalk was intraoperatively preserved. Pituitary gland remained anatomically intact before and after transcranial procedure in these patients.

PRL, TS, LH, FSH, Fr.T4, cortisol, testosterone or estradiol have been analyzed prior to surgery and after 6 months. These data and clinical signs were used by neu-roendocrinologist to confirm hypothyroidism, adrenal insufficiency, hypogonadism, DI.

ResultsSo, there were four groups of patients with different

state of PS and pituitary gland before and after surgery. Outcomes are presented in the Table.

CP removal followed by PS intersection was accom-panied by panhypopituitarism in all patients of the 1st group.

DI was noted in 43 patients (11 of them had DI prior to surgery). Herewith, DI was not observed in 3 patients from this group despite PS intersection.

All patients of the 1st group needed for postoperative hormonal replacement therapy (glucocorticoid, thyroid, sex hormones, desmopressin).

Postoperative hyperprolactinemia occurred in 19 (41%) patients. At the same time, 5 out of 19 patients with increased PRL level prior to surgery had its postop-


then gather in portal veins descending to adenohypophy-sis through PS. In pituitary gland they are again divided into capillary network. Thus, blood is enriched with re-leasing hormones in hypothalamus and then passes into adenohypophysis. Blood outflow with pituitary hor-mones saturated occurs through venous system to venous sinuses of dura mater. Portal supply to pituitary gland with descending blood flow from hypothalamus is a mor-phofunctional component of neurohumoral control of tropic pituitary functions. Pituitary hormones secretion depends on interaction of hypothalamus, portal vessels and secreting cells of adenohypophysis.

Inferior pituitary arteries participate in neurohy-pophysis blood supply, contact with neurosecretory ax-ons of large-cell hypothalamic nuclei and are located below sellar diaphragm (Fig. 3).

PRL secretion is under inhibitory effect of dopamine produced by hypothalamus. All other pituitary hormones are under stimulating effect of hypothalamic releasing hormones.

It is believed that PS rupture or mechanical com-pression of portal vessels disrupts transport of dopamine and releasing hormones. The last is followed by reduced hypothalamic control over pituitary hormones secretion and leads to hyperprolactinemia and impaired secretion of all other pituitary hormones.

Besides vessels PS consists of axons of supraoptic and paraventricular neurons of hypothalamus which are responsible for antidiuretic hormone (ADH) secretion. Complexes of ADH and neurophysin protein are trans-ported along these axons to neurohypophysis which is reservoir for ADH. Therefore, ADH release into blood

flow is observed (Fig. 2). Damage of these axons is asso-ciated with DI [6].

Patients with congenital pituitary stalk interruption syndrome (PSIS) are described in the world literature [7]. There are no PS and posterior pituitary lobe on MRI scans in these cases. Similar phenomenon is interpreted by scientists as neurohypophysis ectopy and PS agenesia due to embryogenesis disorders or rupture within birth trauma. Rare mutations of HESX1, LH4, OTX3 and SOX3 genes may be also the cause of PSIS [8, 9]. Clinical picture is characterized by pituitary dwarfism. Large-scale trial of 55 children with PSIS syndrome revealed somatotropic hormone deficiency in all patients, hypo-gonadotropic hypogonadism in 95.8%, ACTH deficien-cy in 81.8% and secondary hypothyroidism in 76.3% of patients. Hyperprolactinemia was noted only in 36.4% of cases [10].

Some researches [11—15] reported impaired secre-tion of pituitary hormones due to PS compression by dif-ferent CSR tumors including pituitary adenoma, menin-gioma, CP, germinoma, glioma, inflammatory processes (tuberculosis, sarcoidosis), metastases, carotid artery aneurysms. Patients with PS and portal vessels compres-sion have moderate hyperprolactinemia and impaired release of pituitary hormones. PRL level in such cases usually does not exceed 2,000 IU/l [16—18].

B. Arafah et al. [19] reported increased intrasellar pressure and consequently impaired blood flow through portal vessels as important mechanism of hyperprolac-tinemia and hypopituitarism. Intrasellar pressure is near 28.8±13.5 mm Hg in patients with pituitary adenoma (normal intracranial pressure 10—15 mm Hg); herewith

Fig. 1. Scheme of CSR tumors before (a) and after (b) surgery. Group 1: a – stalk CP prior to surgery; b — transcranial removal of stalk CP.Group 2: a — endo-suprasellar pituitary adenoma before surgery; b — state after transcranial removal of endo-suprasellar pituitary adenoma. Group 3: a — endo-suprasellar pituitary adenoma prior to surgery; b — state after intracapsular transnasal endoscopic removal of endo-suprasellar pituitary adenoma. Group 4: a — suprasellar meningioma before surgery; b – state after transcranial removal of suprasellar meningioma.


ORIGINAL ARTICLES

Patients’ characteristics and endocrine disorders before and after surgery

Variable

Group 1 Group 2 Group 3 Group 4

Stalk CP followed by PS intersection

Pituitary adenoma with PS intersection

Pituitary adenoma without PS intersection

Suprasellar meningioma without

PS intersectionPituitary compression – + + –PS compression + + + +Number of patients 46 11 41 41Age, min—max (median), years 15—64 (36) 18—57 (45) 32—67 (53) 35—68 (48)Gender:

Females 31 7 25 37Males 15 4 16 4

Prior to surgeryPRL, min—max (median), IU/l

115—3000 (440) 185—1085 (500)140—1980 (550)

113—4050 (377)Hyperprolactinemia, n (%) 19 (41) 4 (36) 16 (39) 13 (32)Hypothyroidism, n (%) 12 (26) 4 (36) 14 (34) 3 (7)Hypocorticism, n (%) 7 (15) 2 (18) 11 (27) 2 (5)Hypogonadism, n (%) 38 (83) 8 (73) 33 (80,5) 4 (10)DI, n (%) 11 (23) — — 1 (2)

After surgeryIntraoperative PS intersection + + – –Pituitary gland preservation** Preserved ? ? PreservedPRL, min—max (median), IU/l

35—3600 (380)33—1080 (335)

132—610 (235) 93—740 (280)Hyperprolactinemia, n (%) 19(41) 2 (18) 1 (2) 2 (5)Hypothyroidism, n (%) 46 (100) 11 (100) 12 (29) 3 (7)Hypocorticism, n (%) 46 (100) 11 (100) 10 (24) 2 (5)Hypogonadism, n (%) 46 (100) 11 (100) 29 (71) 5 (12)DI*, n (%) 43 (93.5) 10 (91) 0 1 (2)

Footnote. * — transient postoperative DI followed by regression within 6 months was not taken into account in analysis. ** — in groups 2 and 3 preservation of pituitary gland (compressed? atrophied?) is not exactly known. Moreover, it could be additionally damaged during surgery.

higher PRL level is associated with higher intrasellar pressure. Pituitary function recovery after adenomecto-my may be caused by not only pituitary compression elimination but also normalization of intrasellar pressure and restoration of portal blood flow.

Persistent postoperative pituitary dysfunction may be associated with ischemic necrosis followed by circulatory disturbances in portal system [19, 20].

In our sample preoperative hyperprolactinemia was observed in 52 (37%) patients. In most cases PRL level did not exceed 2,000 mI/l and only in 2 cases (CP and meningioma) reached 3,000 and 4,050 IU/l, respectively.

23% of patients with CP followed by partial PS de-struction (group 1) had DI due to interrupted ADH transport or death of hypothalamic ADH-secreting cells.

PS compression in patients with pituitary adenomas and suprasellar meningiomas (groups 2, 3, 4) did not lead to DI. PS-sparing removal of neoplasm (groups 3 and 4) obviously led to restoration of dopamine and hypotha-lamic releasing hormones transport. This is evidenced by normal PRL level in most cases and regress of hypopitu-itarism in some patients (Table).

Similar data were presented by H. Zaidi et al. [21] in analysis of pituitary function in 276 patients undergoing transsphenoidal surgery. Normal PRL level was noted in

72 (77.8%) patients with preoperative hyperprolac-tinemia.

P. Nomikos et al. [22] reported improved pituitary function in 50% of patients with HIA and hypopituita-rism prior to transsphenoidal surgery. Moreover, it was proved that elevated preoperative PRL is a predictor of postoperative restoration of pituitary function. Partial or complete regression of pituitary disorders was noted in 80% of patients with PRL over 500 IU/l, in 52.3% with PRL 100 to 500 IU/l and only in 30% with PRL less than 100 IU/l.

There are small number of reports in the world litera-ture for pituitary stalk interruption syndrome; experi-mental works are also limited.

There is severe atrophy of adrenal glands, gonads, thyroid gland, DI in animals with pituitary stalk inter-ruption. This syndrome is followed by impaired somatic growth in young specimens [23]. L. Vaughan et al. [24] analyzed hormone secretion in 20 monkeys with experi-mental PS intersection. It was accompanied by increased PRL level in most cases throughout follow-up (3 years).

J. Hume Adams [26] reported advanced necrosis of anterior lobe of pituitary gland after PS intersection in animals due to injury of portal vessels. However, glandu-lar pituitary tissue was partially preserved due to blood


Fig. 2. Scheme of the structure and function of "hypothalamus-pituitary gland-peripheral endocrine glands" system (explanations in the text).


ORIGINAL ARTICLES

supply via inferior pituitary arteries. The same author in 1966 reported 21 patients who underwent transcranial in-tersection of PS (article in the journal J Neurol Neurosurg Psychiat). There were 20 cases of breast or prostate gland cancer and 1 case of diabetic retinopathy). In 50s and 60s of the 20th century hypophysectomy and/or PS intersec-tion were used for advanced breast and prostate cancer, as well as in patients with progressive diabetic retinopathy. It was believed that PS intersection was followed by re-gression of pain syndrome and inhibited tumor growth. All those patients died within 30 hours — 11 months after surgery. Morphological survey revealed infarctions in ad-enohypophysis (large as a rule) but complete necrosis of adenohypophysis has been never noted. Posterior pitu-itary lobe remained intact but decreased in dimen-sions [26].

J. Honegger [27] analyzed endocrine disorders after CP removal; it was shown that PS intersection was asso-ciated with panhypopituitarism and DI in all cases. Those after partial intersection of PS had lower risk hy-popituitarism and DI. However, PRL level was not ana-lyzed in this work.

In our study PS intersection in preserved or com-pressed pituitary gland (CP in group 1 or HIA in group 2) led to panhypopituitarism in all patients and DI in most cases. At the same time, there was no DI in 4 (7%) pa-tients from these groups. It is possibly caused by normal ADH secretion and its release into systemic circulation bypassing neurohypophysis. Ectopic neurohypophysis within proximally enlarged PS cannot be also excluded. This phenomenon is described in patients with PS inter-

sected close to sellar diaphragm. DI regression is unlikely if PS intersection occurs near hypothalamus [28, 29].

PS intersection in our patients did not cause expect-ed apparent hyperprolactinemia. In pituitary adenoma it may be explained by intraoperative damage of residual adenohypophysis, but occurrence of this phenomenon in CP with preserved pituitary gland is unclear. Pituitary ischemia and necrosis may be assumed. However, it would be more likely to expect hypoprolactinemia due to pituitary cells death including lactotrophic cells in this case that was not noted in our study. Hypoprolactinemia was detected only in 3 cases. It cannot be ruled out that dopamine transport to adenohypophysis occur not only through PS vessels or partial revascularization of inter-sected vessels is possible.

ConclusionAs a result of our research, it is worthwhile to distin-

guish two concepts: "PS compression syndrome" and "surgically intersected PS syndrome".

PS compression syndrome implies pressure by neo-plasm and includes hyperprolactinemia as a rule (37%); incidence of hypopituitarism and DI is low in preserved pituitary gland. Hyperprolactinemia is usually asymp-tomatic but may be followed by hyperprolactinemic hy-pogonadism in some cases. Therapy with cabergoline leads to regression of hyperprolactinemia and restores sexual function. Tumor excision eliminates PS compres-sion and normalizes PRL level. The last is accompanied by physiologic dopamine transport. Compression of PS

Fig. 3. Hypophysis blood supply and vascular injury in PS intersection (explanations in the text).


and pituitary gland (in pituitary adenoma with suprasel-lar growth) is associated with hypopituitarism besides hy-perprolactinemia. Hyperprolactinemia is also regressed after adenoma removal. Surgical elimination of PS and pituitary gland compression repairs pituitary function in 15—21% of cases. Other patients have partial or total hy-popituitarism due to possible atrophy of hormone-se-creting pituitary cells.

Surgically intersected PS syndrome is manifested by panhypopituitarism in 100% of cases, DI — in 93%, hy-perprolactinemia — in 37% of cases.

Thus, the role of PS in transfer of hypothalamic hor-mones through vessels and axons of secretory nuclei from hypothalamus to adeno- and neurohypophysis is con-firmed. At the same time, expected increase of PRL level after PS intersection was not observed. On the contrary, incidence of hyperprolactinemia decreased in pituitary adenoma and was the same in CP. So, comprehensive

analysis of hyperprolactinemia in patients with CSR tu-mors is required.

Following conclusion is the most relevant for neuro-surgeons: PS intersection is followed by appearance or progression of panhypopituitarism in all patients and DI in most patients. PS should be preserved in patients with pituitary adenoma if it is possible.

In patients with CP this question is individually solved depending on intraoperative features, age and consideration that risk of recurrence will be increased in this case.


REFERENCES1. Dedov II, Balabolkin MI, Marova EI. Bolezni organov endokrinnoi sistemy

(Rukovodstvo po vnutrennim boleznyam). М.: Meditsina, 2000;568. (In Russ.).

2. Akmaev IG. Strukturnye osnovy mekhanizmov gipotalamicheskoi regulyatsii endokrinnykh funktsii. M.: Nauka, 1979. (In Russ.).

3. Voitkevich AA, Dedov II. Ul’trastrukturnye osnovy gipotalamicheskoi nei-rosekretsii. M.: Meditsina, 1972. (In Russ.).

4. Babichev VN. Osnovnye predstavleniya o mekhanizme funktsionirovaniya sistemy «gipotalamus-gipofiz-perifericheskie endokrinnye zhelezy». V kn.: Adenomy gipofiza, klinika, diagnostika, lechenie. Pod red. Kadasheva BA. Tver’. 2007;23-31. (In Russ.).

5. Frank H. Atlas of human anatomy. Third Edition. Netter. 2003 ICON Learning systems, Teterboro, New Jersey.

6. Melmed S. The pituitary. Second edition. Blackwell Science Inc 2002. Chapter 7.

7. Fujisawa I, Kikuchi K, Nishimura K, Togashi K, Itoh K, Noma S, Minami S, Sagoh T, Hiraoka T, Momoi T. Transection of the pituitary stalk: develop-ment of an ectopic posterior lobe assessed with MR imaging. Radiology. 1987 Nov;165(2):487-489. https://doi.org/10.1148/radiology.165.2.3659371

8. Gutch M, Kumar S, Razi SM, Saran S, Gupta KK. Pituitary stalk interrup-tion syndrome: Case report of three cases with review of literature. J Pediatr Neurosci. 2014 May-Aug;9(2):188-191.

https://doi.org/10.4103/1817-1745.1393639. Pinto G, Netchine I, Sobrier mL, Brunelle F, Souberbielle JC, Brauner R.

Pituitary stalk interruption syndrome: a clinical-biological-genetic assess-ment of its pathogenesis. J Clin Endocrinol Metab. 1997;82:3450-3454.

https://doi.org/10.1210/jcem.82.10.429510. Guo Q, Yang Y, Mu Y, Lu J, Pan C, Dou J, Lv Z, Ba J, Wang B, Zou X,

Yang L, Ouyang J, Yang G, Wang X, Du J, Gu W, Jin N, Chen K, Zang L, Erickson BJ. Pituitary stalk interruption syndrome in Chinese people: clin-ical characteristic analysis of 55 cases. PLoS One. 2013;8(1):e53579.

https://doi.org/10.1371/journal.pone.005357911. Zhang F, Huang Y, Ding C, Huang G, Wang S. The prevalence of hyperp-

rolactinemia in non-functioning pituitary macroadenomas. Int J Clin Exp Med. 2015 Oct 15;8(10):18990-18997.

12. Kwancharoen R, Blitz AM, Tavares F, Caturegli P, Gallia GL, Salvatori R. Clinical features of sellar and suprasellar meningiomas. Pituitary. 2014 Aug;17(4):342-348. https://doi.org/10.1007/s11102-013-0507-z

13. Halac I, Zimmerman D. Endocrine manifestations of craniopharyngioma. Childs Nerv Syst. 2005 Aug;21(8-9):640-648.

https://doi.org/10.1007/s00381-005-1246-x14. Wildemberg LE, Vieira Neto L, Taboada GF, Moraes AB, Marcondes J,

Conceição FL, Chimelli L, Gadelha MR. Sellar and suprasellar mixed germ cell tumor mimicking a pituitary adenoma. Pituitary. 2011 Dec;14(4):345-350. https:/doi.org/10.1007/s11102-008-0161-z

15. Bouznad N, El Ansari N. Intra and latero-sellar carotid aneurysm mimick-ing a pituitary adenoma. Pan Afr Med J. 2015 Oct 16;22:150.

https://doi.org/10.11604/pamj.2015.22.150.804316. Astaf’eva LI, Marova EI, Kadashev BA, Korshunov AG. Sravnitel’noe

issledovanie prolaktinom i gormonal’no-neaktivnykh adenom gipofiza s umerennoi giperprolaktinemiei. Problemy endokrinologii. 2006;3:30-33. (In Russ.).

17. Karavitaki N, Thanabalasingham G, Shore HC. Do the limits of serum prolactin in disconnection hyperprolactinaemia need re-definition? A study of 226 patients with histologically verified non-functioning pituitary mac-roadenoma. Clin Endocrinol (Oxf). 2006 Oct;65(4):524-549.

https://doi.org/10.1111/j.1365-2265.2006.02627.x

18. Hong JW, Lee MK, Kim SH, Lee EJ. Discrimination of prolactinoma from hyperprolactinemic non-functioning adenoma. Endocrine. 2010 Feb;37(1): 140-147. https://doi.org/10.1007/s12020-009-9279-7

19. Arafah BM, Prunty D, Ybarra J, Hlavin mL, Selman WR. The dominant role of increased intrasellar pressure in the pathogenesis of hypopituitarism, hyperprolactinemia, and headaches in patients with pituitary adenomas. J Clin Endocrinol Metab. 2000 May;85(5):1789-1793.

https://doi.org/10.1210/jcem.85.5.661120. Zayour DH, Selman WR, Arafah BM. Extreme elevation of intrasellar pres-

sure in patients with pituitary tumor apoplexy: relation to pituitary function. J Clin Endocrinol Metab. 2004 Nov;89(11):5649-5654.

https://doi.org/10.1210/jc.2004-088421. Zaidi HA, Cote DJ, Castlen JP, Burke WT, Liu YH, Smith TR, Laws ERJr.

Time course of resolution of hyperprolactinemia after transsphenoidal sur-gery among patients presenting with pituitary stalk compression. World Neurosurg. 2016 Sep 23;pii: S1878-8750(16)30896-30898.

https://doi.org/10.1016/j.wneu.2016.09.06622. Nomikos P, Ladar C, Fahlbusch R, Buchfelder M. Impact of primary sur-

gery on pituitary function in patients with non-functioning pituitary adeno-mas — a study on 721 patients. Acta Neurochir (Wien). 2004 Jan;146(1):27-35. https://doi.org/10.1007/s00701-003-0174-3

23. Hume Adams J, Daniel PM, Marjorie mL. Prichard some effects of tran-section of transection of the pituitary stalk. Br Med J. 1964 Dec 26;2(5425):1619-1625. https://doi.org/10.1136/bmj.2.5425.1619

24. Vaughan L, Carmel PW, Dyrenfurth I, Frantz AG, Antunes JL, Ferin M. Section of the pituitary stalk in the rhesus monkey. I. Endocrine studies. Neuroendocrinology. 1980;30(2):70-75. https://doi.org/10.1159/000122978

25. Anderson LL, Hard DL, Trenkle AH, Cho SJ. Long-term growth after hy-pophyseal stalk transection and hypophysectomy of beef calves. Endocrinol-ogy. 1999 May;140(5):2405-2414. https://doi.org/10.1210/endo.140.5.6735


ORIGINAL ARTICLES

Authors’ work is devoted to analysis of pre- and postopera-tive endocrine disorders in patients with CSR tumors followed by compressed, preserved or forcedly intersected pituitary stalk during neurosurgical procedure. Neuroendocrine secretion de-pending on tumor growth features and surgical effect on PS, adeno- and neurohypophysis have been assessed. The role of PS in transfer of hypothalamic hormones through vessels and axons of secretory nuclei from hypothalamus to adeno- and neurohypophysis was confirmed. Moreover, the authors have analyzed hormonal secretion in different patients depending on grade of injury of PS, adeno- and neurohypophysis. The re-search is interesting for neuroendocrinologists, neurosurgeons and fully corresponds to the journal’s profile.

Surgical outcomes of CSR tumors, pre- and postoperative hormonal secretion are widely represented in the literature. At the same time, there are only few works devoted to research of PSIS in humans.

Commentary

The authors have evaluated hormonal secretion in 57 pa-tients with PS intersection during surgery and compared these results with group of PS compression. The results determine the significance of this work.

Patients who underwent surgery for CSR tumors were di-vided into 4 groups depending on effect of tumor and subse-quent surgery on PS, adeno- and neurohypophysis. Data are presented as quantitative and percentage dynamics of secretory function. Conclusions are based on these findings.

There are 29 references including 5 domestic and 24 for-eign. Seven of them are not more than 5 years old that empha-sizes work’s relevance.

Statistical analysis is limited in some ways for reliable con-clusions. However, it does not detract relevance and quality of the work. The work will be interesting for a wide range of physi-cians, in particular neurosurgeons and endocrinologists.

A.Yu. Grigoriyev (Moscow, Russia)

26. Hume Adams J, Daniel PM, Marjorie mL. Prichard transection of the pi-tuitary stalk in man: anatomical changes in the pituitary glands of 21 pa-tients. J Neurol Neurosurg Psychiatry. 1966 Dec;29(6):545-555.

http://doi.org/10.1136/jnnp.29.6.54527. Honegger J, Buchfelder M, Fahlbusch R. Surgical treatment of craniopha-

ryngiomas: endocrinological results. J Neurosurg. 1999 Feb;90(2):251-257. https://doi.org/10.3171/jns.1999.90.2.025128 Hayashi Y, Kita D, Watanabe T, Fukui I, Sasagawa Y, Oishi M, Tachibana O,

Ueda F, Nakada M. Prediction of postoperative diabetes insipidus using mor-phological hyperintensity patterns in the pituitary stalk on magnetic reso-nance imaging after transsphenoidal surgery for sellar tumors. Pituitary. 2016 Dec;19(6):552-559. https://doi.org/10.1007/s11102-016-0739-9

29. Fujisawa I. Magnetic resonance imaging of the hypothalamic-neurohy-pophyseal system. J Neuroendocrinol. 2004 Apr;16(4):297-302.

https:/doi.org/10.1111/j.0953-8194.2004.01183.xReceived: 09.06.17




Efficiency of Intraoperative Monitoring of Motor Evoked Potentials in Predicting Early Postoperative Neurological Status in Patients With Cervical Spinal Cord Tumors V.S. KLIMOV, V.V. KEL’MAKOV, N.V. CHISHCHINA, A.V. EVSYUKOV

Federal Center of Neurosurgery, Nemirovich-Danchenko Str., 132/1, Novosibirsk, Russia, 630087

Aim. The study aim was to evaluate the effectiveness of intraoperative monitoring of motor evoked potentials (MEPs) for predicting changes in the neurological status of patients with cervical spinal cord tumors in the early postoperative period.Material and methods. The study included 74 patients with intradural cervical spinal cord tumors who were operated on using motor evoked potential monitoring in the period from 2013 to 2016. There were 29 (39%) males and 46 (61%) females. Group 1 included 26 (35%) patients with intramedullary tumors; group 2 included 48 (65%) patients with intradural extramedullary tumors. The neurological status was assessed by using a six-grade muscle power MRC scale; a modified McCormick scale was used to evaluate the functional status. Transcranial electrical stimulation of the precentral gyri was performed. Recording electrodes were located in the peripheral target muscles of the upper and lower limbs. Total intravenous anesthesia with propofol and fentanyl was used.Results. In Group 1, MEPs decreased in 19 (73%) of 26 patients; MEPs remained unchanged in 7 (27%) patients. Among the patients with decreased MEPs, 14 (74%) patients had postoperative deterioration of the neurological status; 6 (32%) patients had a preoperative severe neurological deficit; 4 (21%) patients had no changes in the neurological status. The sensitivity and specificity of MEPs from the upper limb muscles were higher than those from the lower limb muscles. In Group 2, improvement of the neurological status occurred in all 48 patients. There was no case of a true positive decrease in the MEP amplitude.Conclusions. 1. Registration of MEPs is a highly sensitive and highly specific method for diagnosing corticospinal tract dysfunction in patients with intramedullary tumors of the cervical spinal cord. The sensitivity and specificity of MEPs recorded from the upper limbs are higher than those from the lower limb muscles.2. The sensitivity of MEPs in patients with extramedullary intradural tumors was 0%, the diagnostic effectiveness of MEPs amounted to 86% from the upper limb muscles and 93% from the lower limb muscles.3. When the MEP amplitude in patients with extra- and intramedullary tumors of the cervical spinal cord decreases, a change in the surgion’s approach may reduce or completely eliminate surgically-induced damage to the spinal cord.

Keywords: cervical spinal cord, motor evoked potentials, intradural extramedullary tumors, intramedullary tumors.

Intramedullary tumors are rare and account for about 2—8.5% of all tumors of central nervous system (CNS) and 15% of all primary intradural spinal cord tu-mors [1—4]. Extramedullary tumors are observed in 10—12% among all CNS tumors and outnumber intra-medullary neoplasms as 4:1 [4—7].

Surgery is the main treatment of spinal cord tumors. Intramedullary tumors are the least favorable for their surgical excision although they are benign with slow growth as a rule [8]. Treatment of malignancies is usually limited by biopsy and radiotherapy [1, 9].

There are several methods of intraoperative neuro-physiological monitoring of spinal cord functions: so-mato-sensory evoked potentials (SSEP), motor evoked potentials (MEP) — muscle response registration (actu-ally MEP) and responses from spinal cord surface (D-wave). Changes of SSEP is not currently criterion deter-mining the need for cessation of surgical excision of in-tramedullary tumors [10]. MEP method allows to assess functional state of corticospinal tract and to observe the first signs of surgical injury of spinal cord. The last makes it possible to carry out appropriate measures to prevent or reduce postoperative neurological disorders [10, 11].

The aim of the work was to evaluate an effectiveness of intraoperative monitoring (IOM) of MEP for predict-ing early postoperative neurological status in patients with cervical spinal cord tumors.

Material and methodsThere were 74 patients including 28 (38%) men and

46 (62%) women aged 49±12 years who underwent MEP-controlled surgery at the Federal Center for Neu-rosurgery (Novosibirsk) for intradural tumors and other neoplasms of cervical spinal cord for the period from 2013 till 2016. Histological diagnosis was established in view of criteria of WHO classification (2007) of CNS tu-mors [12]. Schwannoma was observed in 24 (33%) cases, meningioma — in 18 (24%), ependymoma — in 15 (20%), neurofibroma — in 6 (8%), astrocytoma — in 1 (1%), he-mangioblastoma — in 4 (6%), cavernous angioma — in 4 (6%), mature teratoma — in 1 (1%), dermoid cyst — in 1 (1%).

Neurological status was evaluated by using of 6-point scale of muscle strength (MRC-scale), patient’s func-tional state – with McCormick modified scale [13—15].


ORIGINAL ARTICLES

sensitivity=[TP/(TP + FN)] · 100; specificity=[TN/(TN + FP)] · 100; positive predictive value (PPV)=[TP/(TP + FP)] · 100; negative predictive value (NPV)=[TN/(TN + FN)] · 100; diagnostic effectiveness=[(TP + TN)/(TP + TN + FP + FN)] · 100 [16].

All patients were divided into two groups: group 1 with intramedullary tumors: 26 (35%) patients aged 45±11 years, 12 (46%) men and 14 (54%) women; group 2 with extramedullary tumors: 48 (65%) patients aged 50±12 years, 15 (31%) men and 33 (69%) women.

ResultsIn group 1 tumor’s mean transverse size was

12±4 mm, tumor spread along 1—2 vertebrae in 10 pa-tients, in 7 cases – 3—4 vertebrae and in 5 patients – along 6—7 vertebrae. Ependymoma was observed in 15 (58%) cases, hemangioblastoma — 4 (15%), cavern-ous angioma — 4 (15%), astrocytoma — 1 (4%), mature teratoma — 1 (4%), dermoid cyst – 1 (4%). Mean preop-erative MRC-score was 4.5±0.2, McCormick modified scale value — 2.5±0.9.

Patient’s position on operating table depended on tu-mor’s location: sitting position in 19 (73%) patients, lat-eral decubitus position in 5 (19%) ones, prone position in 2 (8%) patients. Mean time of surgery was 240±53 min; mean blood loss — 115±57 mL.

Mean postoperative MRC-score was 3.6±0.1. Dete-riorated neurological status by this scale was noted in 15 (58%) patients who had tumor’s transverse size 4—23 mm. Further improvement of neurologic state was observed in 3 (8%) patients with tumor’s transverse di-mension 9—18 mm. Unchanged neurological status compared to that prior to surgery was in 6 (27%) patients (tumor’s transverse size 5—12 mm). Mean postoperative McCormick modified score was 2.9±1. Deteriorated, unchanged and improved functional outcome was noted in 12 (46%), 11 (42%) and 3 (12%) patients, respectively.

Decrease of MEP amplitude was revealed in 19 (73%) out of 26 patients, unchanged MEPs — in 7 (27%) ones. 14 (54%) patients with decreased MEP had also deterio-ration of postoperative neurological status, 6 (23%) of them had severe preoperative neurologic disturbances, and 4 (15%) patients had not any neurological disorders. Ependymoma was histologically predominant in these patients (14 cases). Hemangioblastoma was detected in 3 patients, cavernous angioma in 1 and diffuse astrocy-toma Grade II in 1 patient.

Neurosurgeon made intermission as soon as MEPs were decreased, changed tumor’s dissection side, applied irrigation of dissection area with warm saline, reduced tumor’s volume or administered methylprednisolone. As a result, deterioration of neurological status within only 1—2 scores was observed.

Informational values of MEP for the 1st group are presented in Table 1. Sensitivity and specificity of MEP

All patients underwent contrast-enhanced magnetic resonance imaging (MRI) prior to surgery to assess tu-mor’s dimensions and location and within the 1st postop-erative day to analyze surgical effect.

Surgical approach, bone resection type, patient's po-sition on the operating table were planned taking into ac-count MRI data about solid tumor. All patients were op-erated under rigid head fixation in Mayfield bracket. Re-moval of intradural tumor was accompanied by MEP monitoring.

We performed transcranial electrical stimulation of precentral gyrus in order to obtain MEP, recording elec-trodes were placed in peripheral target muscles. Triceps and biceps shoulder muscles, tenar and hypotenar mus-cles, quadriceps femoris muscle, anterior tibial muscle and gastrocnemius muscle were selected for this purpose. Bilateral monitoring was applied in all patients. Stimula-tion parameters were various: strength 400—800 V, length of stimulus 0.5—0.75 ms, number of stimuli 4 to 9, inter-val between stimulation 2 ms. Stimulating electrodes were placed in C3/C'1 and C4/C'2 by 10—20 interna-tional system [16].

Baseline transcranial MEP were recorded after mus-cle relaxants’ end effect before skin incision, prior to and after dura mater dissection. Later, intraoperative stimula-tion was continued regularly as often as it was possible in surgical situation. Responses were recorded every 2—10 minutes as a rule. Any significant decrease (≥ 50%) of MEP amplitude was recorded, these data were report-ed to surgeon.

Nicolet Viking and Cadwell Cascade Elite systems were used for monitoring. General intravenous anesthe-sia with propofol and fentanyl was administered in all cases.

Sensitivity, specificity, prognostic value of positive and negative results and diagnostic effectiveness were analyzed to assess MEP efficacy. Above-mentioned vari-ables were evaluated separately for arms and legs muscles. Sensitivity included proportion of truly positive results among patients with decreased MEP, specificity — pro-portion of truly negative results among patients with un-changed postoperative neurological status. Prognostic value of positive result was assessed as probability of neu-rologic disorders in decreased MEP, prognostic signifi-cance of negative result — probability of absent neuro-logic changes in unchanged amplitude of MEP. the Number of true results obtained by using of MEP deter-mined diagnostic efficiency. True positive (TP) result was defined as postoperative deterioration of neurological status with intraoperative decrease of motor evoked re-sponse (MER), true negative (TN) – unchanged MER followed by normal postoperative neurological status. False positive (FP) result implied normal postoperative neurological status with decreased intraoperative MER, false negative (FN) – postoperative neurological disor-ders with normal intraoperative amplitude of MER. Fol-lowing formulas were used to calculate these values:


registered from the arms were higher compared with leg muscles.

According to contrast-enhanced MRI of cervical spine within the 1st postoperative day there was no abnor-mal contrast accumulation. So, tumors were radically removed.

Case report

Patient Sh., 35 years old, was hospitalized at the neu-rosurgery department №2 with the diagnosis of "C2 in-tramedullary tumor", complaints of weakness and numb-ness in right hand.

Neurological status: paresis of flexor and extensor of the right hand (3 scores), flexors and extensors of the right forearm (4 scores).

McCormick scale — 2 scores. Contrast-enhanced MRI of cervical spine revealed intramedullary postero-lateral tumor at the level of C2-vertebra (Fig. 1).

Laminectomy of C2-vertebra, intramedullary tumor removal under neurophysiological control were carried out. There was sitting position of patient on the operating table with head fixed in Mayfield bracket. Before and af-ter dura mater dissection unchanged MER was obtained from all target muscles.

Severe deterioration of MER amplitude from the right hand’s muscles (by 90%) and disappearance of MER from the right leg’s muscles were observed in 7 minutes after tumor dissection onset (19:10) (Fig. 2). These data could indicate impaired right-sided motor conduction through anterior and lateral spinal cord col-umns due to surgical injury of corticospinal tract. At 19:15 irrigation of the wound with warm saline, methyl-prednisolone (1000 mg) administration and change of tumor dissection side were applied.

During hemostasis after tumor removal (19:30) re-covery of MER amplitude from the right leg’s muscles was observed, at 19:39 — restoration of MER amplitude from the right hand’s muscles. Preserved MER was ob-tained from all target muscles at the end of hemostasis (Fig. 3).

Thus, there was a marked transient decrease of MER from the right hand and leg during tumor removal that can indicate surgical injury of spinal cord.

In early postoperative period aggravation of right hand paresis followed by decreased muscle strength up to 2 scores was noted, also there was mild paresis of left hand associated with impaired muscle strength up to 4 scores (initially 5 scores), McCormick score — 3.

Contrast-enhanced MRI of cervical spine within the 1st postoperative day confirmed radical tumor excision (Fig. 4).

Positive in-hospital dynamics of neurological status was noted: muscle strength recovery in left hand up to 5 scores, in right hand – up to 3 scores, McCormick score — 2. Patient was discharged in 8 days after surgery. Histological conclusion: hemangioblastoma, Grade I.

In group 2 mean transverse dimension of tumors was 17±6 mm, tumor spread along 1—2 vertebrae was in 31 patients, 3 vertebrae — in 9 ones and 6—7 verte-brae — in 7 cases. Schwannoma, meningioma and neu-rofibroma were diagnosed in 24 (50%), 18 (38%) and 6 (12%) cases, respectively. Mean preoperative MRC-score was 4.6±0.1, McCormick score — 2.15±0.9. Pa-tient’s position on the operating table depended on tu-mor’s lateralization: 37 (77%) patients were operated in decubitus position, 5 (11%) – in sitting position and 6 (12%) in prone position. Mean time of surgery and blood loss were 296±113 min and 417±350 mL, respec-tively.

There was significantly improved postoperative MRC-score in all 48 patients, mean muscle strength score — 4.8±0.1. Modified McCormick score was 2±0.8 after tumor removal. Unchanged functional status was observed in 38 (79%) patients, improved — in 8 (17%), aggravated – in 2 (4%) patients due to decreased sensitivity of limbs. Intraoperative impaired MEP ampli-tude from hands’ muscles was recorded in only 6 (13%) patients, from legs — in 3 (6%) ones. In other cases, MEP amplitude was the same. Among all patients with de-creased MEP schwannoma Grade 1 was in 5 cases (tu-mor’s transverse size 5, 16, 20, 21 mm), meningioma Grade I — in 3 (transverse dimensions 8, 12 mm), neuro-fibroma — in 1 (transverse size 10 mm).

During intradural extramedullary tumor excision surgeon's actions were aimed at reducing spinal cord traction via enlarged bone approach, use of ultrasonic destructor for debulking, methylprednisolone adminis-tration, irrigation with warm saline.

There were no cases of true positive decrease of MEP in group 2. In 9 (19%) patients with decreased MEP am-plitude normal postoperative muscle strength was ob-served. The last was considered false positive result, so sensitivity of MEP monitoring is 0%. MEP monitoring variables are presented in Table 2.

Contrast-enhanced MRI in the first postoperative day confirmed radical tumor dissection in this group.

Case report

Patient T., 60 years old, was hospitalized at the neu-rosurgery department №2 with the diagnosis of C1—C4 intradural extramedullary tumor, type II by K. Srid- har et al. [17].

At admission: complaints of pain in cervical spine, weakness and numbness in hands and legs.

Neurological examination: spastic tetraparesis (3 scores) conductive hypoesthesia from C3 level. Preop-erative McCormick score 4. Contrast-enhanced MRI of spine cord: intradural extramedullary tumor at C1—C4 lev-el with right-sided lateralization occupying most of verte-bral canal (Fig. 5).

Surgery included C1—C4 laminectomy, intradural tumor dissection under neurophysiological control. Pa-tient’s position is left decubitus with head fixed in May-


ORIGINAL ARTICLES

Table 1. Diagnostic efficacy of intraoperative MEPs monitoring for intramedullary tumors

Method Sensitivity, %

Specificity, %

Positive predictive value, %

Negative predictive value, %

Diagnostic efficacy, %

Hand MEPs 89 100 100 80 92Foot MEPs 83 14 45 50 46

Fig. 1. Contrast-enhanced MRI-scan of cervical spine in patient Sh., sagittal and axial planes. There is intramedullary tumor at the level of C2-vertebra (arrows) in posterior regions with right-sided lateralization. Sagittal size — 6.5 mm, frontal — 8 mm.

Fig. 2. Intraoperative MER from target muscles and variables of transcranial electrical stimulation of motor cortex in patient Sh. Here and in Fig. 3, 6, 7: impaired or disappeared MER is marked with oval. Signatures of channels (superiorly in columns): AP L — left hand: tenar and hypotenar; AP R — right hand: tenar and hypotenar; Brah L — left shoulder: biceps and triceps; Brah R — right shoulder: biceps and triceps, Tib L — left leg: anterior tibial and gastrocnemius muscles; Tib R — right leg: anterior tibial and gastrocnemius muscles. Time of MER registration is indicated on the right from each channel (chronologically). Initial MEPs amplitudes prior to tumor removal are presented under all traces. Arrows indicate the time of MER deterioration.


field bracket. Before and after dura mater dissection un-changed MER was obtained from all target muscles.

At 19:07 severe deterioration of MER amplitude from the right hand’s muscles (by 90%) was recorded (Fig. 6), that could indicate impaired right-sided motor conduction through anterior and lateral spinal cord col-

umns due to surgical injury of corticospinal tract. At 19:08 irrigation of the wound with warm saline, methyl-prednisolone (1,000 mg) administration and change of tumor dissection side were applied.

At 19:11 recovery of MER amplitude from the right hand’s muscles was recorded. At the end of hemostasis

Fig. 3. Postoperative MER from target muscles and variables of transcranial electrical stimulation of motor cortex in patient Sh.Initial MEPs amplitudes prior to tumor removal are presented under all traces. Arrows indicate the time of MER deterioration.

Fig. 4. Postoperative contrast-enhanced MRI-scan of cervical spine in patient Sh., sagittal and axial planes. Tumor is not visualized.


ORIGINAL ARTICLES

normal MER was obtained from all target muscles (Fig. 7).

Contrast-enhanced MRI within the first postopera-tive day confirmed radical removal of tumor (Fig. 8).

Improvement of neurological state in early postop-erative period was observed: tetraparesis up to 4 scores (initially 3 scores), McCormick score 3. Patient was dis-charged in 7 days after surgery. Histological conclusion: schwannoma, Grade I.

DiscussionAccording to V. Deletis and F. Sala [18], unfavorable

outcome after spinal cord surgery is associated with isch-emic disorders and prolonged intraoperative traction. MEPs reflect normal motor function, useful for intraop-erative assessment of conductive function of spinal cord and to determine possibility of further surgical dissection and excision of tumor. In our trial surgeon changed sur-gical tactics if aggravation of MEPs occurred: change of tumor dissection side, waiting for MEPs recovery, irriga-tion of the wound with warm saline, debulking and meth-ylprednisolone administration. F. Sala et al. [18—23] re-ported recovery of potentials and possible following tu-mor dissection after temporary discontinuation of ma-nipulations if significant decrease or disappearance of

MEPs occurred. Tumor dissection despite aggravation of MEPs may be followed by irreversible neurological dis-orders. In our study in both groups this approach led to MEPs recovery followed by radical excision of tumor with preserved movements in limbs. The last confirms high diagnostic efficiency of the method in group 1 (MEP of hand — 92%, MEP of leg — 46%) and in group 2 (MEP of hand — 86%, MEP of leg — 93%).

F. Sala et al. [19] reported similar early postoperative functional outcomes after surgery with/without intraop-erative monitoring. In early postoperative period im-paired functional status was noted in 10 patients with se-vere neurological disorders prior to surgery (McCormick 3 and 4). However, long-term functional outcome (≥3 months) was significantly better in group of intraop-erative monitoring. In our sample impaired neurological status in early postoperative period was noted in patients with intramedullary tumors (14 out of 26) and intraop-erative decrease of MEPs. Long-term neurologic state was not assessed in these patients. Histologically ependy-moma was predominant (9 cases), hemangioblastoma (3), cavernous angioma (1) and astrocytoma (1) were less common. In patients with intradural extramedullary tu-mors impaired neurological status was not detected in either case.

Table 2. Diagnostic efficacy of intraoperative MEPs monitoring for extramedullary tumors

Method Sensitivity, %

Specificity, %

Positive predictive value, %

Negative predictive value, %

Diagnostic efficacy, %

Hand MEPs 0 87 0 100 87Foot MEPs 0 93 0 100 93

Fig. 5. Contrast-enhanced MRI-scan of cervical spine in patient T., sagittal and axial planes. There is intradural extramedullary tumor occupying almost entire vertebral canal with right-sided lateralization (arrows). Sagittal size — 49 mm, transverse size — 22 mm.


S. Peker et al. [24] reported tumor’s transverse di-mension as a predictor of pre- and postoperative neuro-logic disorders. Length of tumor affects postoperative dysesthesia. There was more severe neurological status in patients with large tumors in pre- and postoperative pe-riod. P. Velayutham et al. [25] showed that spinal cord tumor had not effect on postoperative functional status. In our trial deterioration of neurologic state was noted in patients with tumor’s transverse size 4—23 mm that can indicate absent correlation between these variables. However, such data may be associated with small sample size and heterogeneity.

E.A. Burkova [26] concluded that intraoperative MEP monitoring in 41 patients with intramedullary tu-mors was highly sensitive method to diagnose distur-

bances of motor activity. Our data also confirm high sen-sitivity and specificity that justifies MER registration in spinal cord surgery.

P. Velayutham et al. [25] argued that early intraop-erative correction of surgical manipulations for decreased MEP prevent severe postoperative deterioration of motor functions of limbs. The authors observed postoperative neurological deficit in muscles with intraoperative im-pairment of MEP. Complete disappearance of MEP was followed by 3-fold augmentation of the incidence of neu-rologic disorders. Tumor’s placement (intramedullary, extramedullary) did not influence predictive value of MEP monitoring which had high sensitivity (100%) and specificity (98.1 %) in patients with both intra- and extra-medullary tumors.

Fig. 6. Intraoperative MER from target muscles and variables of transcranial electrical stimulation of motor cortex in patient T.

Fig. 7. Postoperative MER from target muscles and variables of transcranial electrical stimulation of motor cortex in patient T.


ORIGINAL ARTICLES

MEP monitoring during intradural tumors excision is associated with opportunity to correct surgical tactics in accordance with intraoperative situation. Evaluation of early postoperative neurological status is useful to de-termine rather intraoperative information is objective for surgeon.

Conclusions1. MEP monitoring is highly sensitive and specific

method to diagnose corticospinal tract dysfunction in patients with intramedullary tumors of cervical spinal

cord. Sensitivity and specificity of MEPs from hands are higher than those from lower extremities.

2. Sensitivity of MEPs in patients with extramedul-lary intradural tumors is 0%, diagnostic efficiency for hands’ muscles — 86%, legs’ muscles — 93%.

3. Correction of surgical tactics for impaired MEPs in patients with intramedullary and extramedullary tu-mors of cervical spinal cord makes it possible to reduce or completely exclude surgical injury of spinal cord.


Fig. 8. Contrast-enhanced MRI-scan of cervical spine in patient T in the 1st day after surgery, axial and sagittal planes. Tumor is not visualized.

REFERENCES1. Bruneau M, Fischer G, Mahla K, Brotchi J, Quiñones-Hinojosa A. Surgi-

cal management of intramedullary spinal cord tumors in adults. Schmidek and sweet operative neurosurgical techniques: indications, methods, and techniques. 2012;6.

2. Constantini S, Miller DC, Allen JC, Rorke LB, Freed D, Epstein FJ. Radi-cal excision of intramedullary spinal cord tumors: surgical morbidity and long-term follow-up evaluation in 164 children and young adults. J. Neuro-surg. 2000;93(2 Suppl):183-193.

3. Sharma M, Sonig A, Ambekar S, Nanda A. Discharge dispositions, compli-cations, and costs of hospitalization in spinal cord tumor surgery: analysis of data from the United States Nationwide Inpatient Sample, 2003-2010: clinical article. Journal of Neurosurgery: Spine. 2014;20:2:125-141.

4. Van Goethem JWM, Van den Hauwe L, Özsarlak Ö, De Schepper AMA, Parizel PM. Spinal tumors. European journal of radiology. 2004;50:2:159-176.

5. Angevine PD, Kellner C, Haque RM, McCormick PC. Surgical manage-ment of ventral intradural spinal lesions: clinical article. Journal of Neuro-surgery: Spine. 2011;15:1:28-37.

6. Chamberlain MC, Tredway TL. Adult primary intradural spinal cord tumors: a review. Current neurology and neuroscience reports. 2011;11:3:320-328.

7. Beall DP, Googe DJ, Emery RL, Thompson DB, Campbell SE, Ly JQ, DeLone D, Smirniotopoulos J, Lisanti C, Currie TJ. Extramedullary intra-

dural spinal tumors: a pictorial review. Current problems in diagnostic radiol-ogy. 2007;36:5:185-198.

8. Kushel’ YuV. Intramedullary tumors of the spinal cord. Part I. (epidemiol-ogy, diagnosis, principles of treatment). Neurosurgery. 2008;3:10. (In Russ.).

9. Isaacson SR. Radiation therapy and the management of intramedullary spi-nal cord tumors. J Neurooncol. 2000;47(3):231-238.

10. Sala F, Bricolo A, Faccioli F, Lanteri P, Gerosa M. Surgery for intramedul-lary spinal cord tumors: the role of intraoperative (neurophysiological) monitoring. European Spine Journal. 2007;16:2:130-139.

11. Merton PA, Morton HB. Stimulation of the cerebral cortex in the intact human subject. Nature. 1980.

12. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, Scheithauer BW, Kleihues P. The 2007 WHO classification of tumours of the central nervous system. Acta neuropathologica. 2007;114:2:97-109.

13. Van der Ploeg RJO, Oosterhuis H, Reuvekamp J. Measuring muscle strength. Journal of neurology. 1984;231:4:200-203.

14. McCormick PC, Stein BM. Intramedullary tumors in adults. Neurosurgery Clinics of North America. 1990;1:3:609.

15. Epstein FJ, Farmer JP, Freed D. Adult intramedullary astrocytomas of the spinal cord. Journal of neurosurgery. 1992;77:3:355-359.


The problem of surgical treatment of intramedullary spinal cord tumors is extremely important for current medicine since disease is characterized by aggressive course, often affects young and middle age people and followed by severe disability in most cases. Currently, intraoperative monitoring (IOM) is obligatory in surgery for intramedullary tumors. At the same time there are still no data about efficacy of this approach in national literature as well as clear algorithm of monitoring that makes this publication interesting.

Yu.V. Kushel, J. Klekamp, J. Brotchi have shown that in-tramedullary tumors should be operated only at specialized clinics where at least 20—30 procedures per year are performed by one surgeon. Fewer number of interventions is followed by advanced morbidity while technical inaccuracies can lead to se-vere disability. According to J. Klekamp surgeons with experi-ence of more than 70 surgeries for intramedullary tumors made total removal in 95 (88.8%) out of 107 cases. At the same time total removal was achieved in 19 (61.3%) out of 31 cases in less skilled surgeons with experience less than 20 surgeries.

At present time MEPs monitoring is one of routine meth-ods for spinal cord functions control although it is necessary to remember shortcomings of this method. One of them is high variability in MER amplitude from study to study that makes it difficult to determine clear criteria of impaired and complete absence of responses. Moreover, sensitivity of these potentials to spinal cord ischemia is high. As a result, false-positive results will be more common in MEPs monitoring alone. Therefore, qualitative changes should be considered during monitoring rather quantitative disorders. So, MEP with spinal D-wave reg-istration is advisable to assess motor pathways of spinal cord.

D-wave (direct wave) reflects direct axonal activation of spinal cord and functional integrity of fast-moving neurons of cortico-spinal tract. According to Morota (1997), D-wave is one of the best predictors of favorable functional outcome along with pre-operative status of patient. At the same time, MEP monitoring significantly improves long-term outcomes and predicts motor disorders at the end of surgery. The last is also noted by the au-thors.

In our opinion, authors' statement about ineffectiveness of somatosensory evoked potentials (SSEP) for IOM seems un-founded. There are confirmations that MEP and SSEP should be used together. Thus, S. Hyun and S. Rhim reported higher sensitivity and positive predictive value for simultaneous MEP and SSEP monitoring rather their separate application.

Thus, it is difficult to disagree with authors that neuro-physiological IOM prevents early postoperative neurological disorders after surgery for intramedullary tumors of cervical spinal cord. It is especially true for preoperative mild and mod-erate neurologic deficit.

In our data preserved M-response amplitude up to 50% by the end of surgery is predictor of unchanged muscle strength, M-response 65—70% in preoperative period is predictor of ad-verse long-term outcome. Combination of monitoring (MEP, SSEP, D-wave) is appropriate. In our opinion, IOM for extramedullary tumors of cervical spinal cord may be only used to ascertain excessive spinal cord compression. Adequate surgi-cal approach and technique should prevent neurological deficit in surgery for extramedullary neoplasms.

A.O. Gushcha (Moscow, Russia)

Commentary

16. Jasper HH. The ten twenty electrode system of the international federation. Electroencephalography and clinical neurophysiology. 1958;10:371-375.

17. Sridhar K, Ramamurthi R, Vasudevan MC, Ramamurthi B. Giant invasive spinal schwannomas: definition and surgical management. Journal of Neu-rosurgery: Spine. 2001;94:2:210-215.

18. Deletis V, Sala F. Intraoperative neurophysiological monitoring of the spi-nal cord during spinal cord and spine surgery: a review focus on the cortico-spinal tracts. Clinical Neurophysiology. 2008;119:248-264.

19. Sala F, Palandri G, Basso E, Lanteri P, Deletis V, Faccioli F, Bricolo A. Motor evoked potential monitoring improves outcome after surgery for in tramedullary spinal cord tumors: a historical control study. Neurosurgery. 2006;58:6:1129-1143.

20. Quiñones-Hinojosa A, Lyon R, Zada G, Lamborn KR, Gupta N, Par-sa AT, McDermott MW, Weinstein PR. Changes in transcranial motor evoked potentials during intramedullary spinal cord tumor resection corre-late with postoperative motor function. Neurosurgery. 2005;56:5:982-993.

21. Pelosi L, Lamb J, Grevitt M, Mehdian SMH, Webb JK, Blumhardt LD. Combined monitoring of motor and somatosensory evoked potentials in orthopaedic spinal surgery. Clinical neurophysiology. 2002;113:7:1082-1091.

22. MacDonald DB. Safety of intraoperative transcranial electrical stimulation motor evoked potential monitoring. Journal of clinical neurophysiology. 2002;19:5:416-429.

23. Costa P, Bruno A, Bonzanino M, Massaro F, Caruso L, Vincenzo I, Ciaram-itaro P, Montalenti E. Somatosensory and motor-evoked potential monitor-ing during spine and spinal cord surgery. Spinal cord. 2007;45:1:86-91.

24. Peker S, Ozgen S, Ozek MM, Pamir MN. Surgical treatment of intramed-ullary spinal cord ependymomas: can outcome be predicted by tumor pa-rameters? Clinical Spine Surgery. 2004;17:6:516-521.

25. Velayutham P, Rajshekhar V, Chacko AG, Babu SK. Influence of tumor location and other variables on predictive value of intraoperative myogenic motor-evoked potentials in spinal cord tumor surgery. World neurosurgery. 2016;92:264-272.

26. Burkova EA. Methods of intraoperative control and postoperative recovery of patients with intramedullary tumors. [dissertation]. M. 2015. (In Russ.).] Dostupno po: http://arhiv.neurology.ru/professional/Disser_Burkova.pdf Ssylka aktivna na 24.05.2016

27. Korn A, Halevi D, Lidar Z, Biron T, Ekstein P, Constantini S. Intraopera-tive neurophysiological monitoring during resection of intradural extra-medullary spinal cord tumors: experience with 100 cases. Acta neurochirur-gica. 2015;157:5:819-830.

Received: 15.06.17


ORIGINAL ARTICLES

*e-mail: [email protected] © Group of authors, 2018


Ventral Decompression in Patients With Traumatic and Non-Traumatic Atlanto-Axial DislocationsI.S. LVOV1, A.A. GRIN’1, M.A. NEKRASOV2, A.YU. KORDONSKIY*1, A.V. SYTNIK1

1Sklifosovsky Research Institute of Emergency Care, B. Sukharevskaya Sq., 3, Moscow, Russia, 129090; 2Pirogov City Clinical Hospital №1, Leninskiy Ave., 8, Moscow, Russia, 117049

Compression of the caudal medulla oblongata and ventral portions of the spinal cord is the most dangerous complication of atlanto-axial dislocation (AAD).Objective: the study objective was to improve surgical management of patients with ventral compression of the spinal cord in the setting of AAD of various genesis.Material and methods. We analyzed treatment outcomes in 250 patients with C1 and C2 injuries and diseases for the period between 2002 and 2016. Persistent ventral compression of the neural structures in the setting of AAD was detected in 34 (13.6%) patients. Anterior or posterior dislocation was in 21 patients, vertical dislocation occurred in 7 patients, and mixed (anterior and vertical) occurred in 6 cases. The causes of AAD included odontoid fractures (21 patients, 61.8%), Jefferson fractures (6 patients, 17.6%), atlas transverse ligament rupture (1 patient, 2.9%), rheumatoid arthritis (4 patients, 11.8%), and nonspecific spondylitis (2 patients, 5.9%). Results. All dislocations were divided into Halo-tractable and Halo-intractable ones. In 24 cases, ventral decompression was achieved due to Halo reposition. Additional resection of a compressing substrate was performed through the submandibular approach in 4 patients, through the transoral approach in 5 patients, and through the transnasal approach in 1 case. In the postoperative period, complications in the form of pharyngeal edema developed in 1 patient after transoral decompression. In the other cases, there were no postoperative complications. All patients had improvement in their condition in the form of regression of a neurological deficit.Conclusion. Halo reposition is a technique eliminating, completely or partially, ventral compression in certain traumatic and non-traumatic dislocations. The choice of a surgical corridor should be performed after preliminary Halo correction. If the nasopalatine line runs in the odontoid neck projection, the submandibular approach may be used in the case of a Halo-tractable dislocation, and the endonasal approach may be used in the case of a Halo-intractable dislocation.

Keywords: odontoid fracture, C1 fracture, C2 fracture, C1 transligamentous dislocation, C1—C2 rheumatoid lesion, C1—C2 non-specific spondylitis, atlanto-axial dislocation, C1—C2 ventral decompression, transoral decompression, transnasal decompression, odontoidectomy, Halo reposition, Kassam line.

Material and methodsThere were 250 patients including 165 men and

85 women aged 15—92 years (mean 40.2) with injuries and diseases of C1—C2 vertebrae for the period from 2002 to 2016. All patients were operated at the Sklifos-ovsky Institute for Emergency Care. Odontoid fractures were diagnosed in 108 patients, Jefferson fractures — in 34, traumatic spondylolisthesis — in 61, multiple C1—C2 fractures — in 34 patients. Injury of atlas trans-verse ligament alone was observed in 1 patient, rheuma-toid polyarthritis — in 8, granulomas after previous non-specific spondylitis — in 2, craniovertebral junction ab-normality — in 2 patients.

At admission permanent ventral compression of spi-nal cord and/or medulla oblongata due to AAD was de-tected in 34 (13.6%) patients. Dislocation due to tran-scendental luxation was present in 21 of them (Fig. 1a), C1 vertebral fractures — in 6. In one case old transliga-mentous luxation of atlas was diagnosed. Twenty patients were hospitalized with acute and subacute trauma, 8 — in intermediate and late period after injury.

Dislocation due to rheumatoid arthritis was verified in 4 patients (Fig. 1b), ventral compression with granu-

Odontoid process, cruciate ligament and secondary ligamentous apparatus (pterygoid ligaments, articular capsule) are the main elements ensuring C1—C2 verte-brae stability [1, 2]. Their injury due to trauma (odontoid fractures, disruption of atlas transverse ligament) or dis-ease (rheumatoid arthritis, spondylitis, tumor) may be followed by C1—C2 instability. Further displacement of atlas is called atlantoaxial dislocation (AAD). There are 5 kinds of dislocation depending on its vector: anterior, posterior, lateral, rotational and axial [3—5]. Combina-tion of these types is possible, for example lateral and ro-tational or front and axial [6, 7].

Compression of ventral spinal cord and/or caudal medulla oblongata is the most dangerous complication of AAD. Advanced neurological deficit along with high risk of vital disorders require prompt elimination of vertebro-medullary conflict with subsequent reliable spine stabili-zation.

The purpose of the work is to improve surgical treat-ment of patients with ventral spinal cord compression due to various AAD.


loma after previous nonspecific spondylitis and AAD — in 2 cases (Fig. 1c).

Anterior or posterior dislocation was diagnosed in 21 patients, axial in 7, combined (front and axial) in 6 ones.

Pain syndrome and neurological deficit were pre-dominantly noted. There was pain syndrome Downie score 3 in 14 patients, score 2 in 19, and score 1 in 1 pa-tient. Among patients with traumatic injuries neurologic symptoms ASIA grade B was observed in 3 patients, C in 6, D in 4 patients. Despite cerebral compression in MRI data there was no neurological deficit. In 6 patients with

non-traumatic ventral compression neurologic disorders consisted of progressive spastic tetraparesis and sensory disturbances from C2-level. So, they were urgently hos-pitalized to Sklifosovsky Institute for Emergency Care.

ResultsFirstly, we have assessed rather dislocation would be

eliminated by using of Halo-apparatus. In this regard all dislocations were divided into Halo-corrected and Halo-uncorrectable [8]. After anesthesia administration Halo-system was deployed. Further, attempts of partial or com-

а b

c

Fig. 1. Examples of AAD in C1—C2 injuries and diseases.a — transdental fracture and luxation of atlas in late period after trauma, ventral spinal cord compression by C2 and fibrous conglomerate within fracture region, dorsally – atlas arch (arrow). T1-weighted MRI-scan; b – axial AAD due to rheumatoid polyarthritis, medulla oblongata compression (arrow). T2-weighted MRI-scan; c — anterior AAD after previous nonspecific spondylitis. Ventral spinal cord compression by granuloma and dorsal compression by posterior arch of dislocated C1 vertebra (arrows) are visualized. T1-weighted MRI-scan.


ORIGINAL ARTICLES

plete reposition of vertebrae dislocated were applied un-der two-plane X-ray control. One of the methods of an-terior or posterior spondylodesis was performed if cranio-vertebral junction was successfully restored (Halo-cor-rected dislocation). If complete reposition was not achieved (Halo-uncorrectable dislocation) or compres-sion was caused by tumor then anterior spinal cord de-compression was applied via transoral, transnasal or sub-maxillary approach.

Halo-corrected dislocation without anterior spinal cord decompression was observed in 24 patients with

traumatic injuries and in 1 case of rheumatoid arthritis followed by odontoid basilar invagination.

Submandibular approach was used in 4 patients: mo-bilization of dislocated vertebrae followed by Halo-cor-rection was performed in 3 cases of old odontoid frac-tures, 1 patient with rheumatoid polyarthritis required ventral decompression.

Transoral access was used in 5 patients: odontoid im-pression due to rheumatoid polyarthritis (2), ventral compression with granulomas after previous nonspecific spondylitis (2) and chronic odontoid fracture (1).

bа

c d

Fig. 2. Halo-correction for AAD. a — patient P., 53 years old. Odontoid fracture followed by anterior AAD up to 2 cm is observed on lateral X-ray image of cervical spine (arrows); b — postoperative X-ray images: dislocation is completely eliminated, dorsal combined spondylodesis with cannulated screws by F. Magerl and using a hooked plate; c – patient Ya., 55 years old, rheumatoid polyarthritis. Preoperative sagittal CT-scan, odontoid apex is in large occipital orifice, posterior atlantodental interval (PADI) is less than 14 mm that indicates spinal cord compression; d — postoperative sagittal CT-scans: C2 is completely repaired, ventral compression is eliminated, C1 laminectomy and posterior occipitospondylodesis are carried out for dorsal decompression.


Tranasal repair was used in 1 patient with transliga-mentous luxation of atlas.

Odontoid resection was intraoperatively complicated by liquorrhea within apical ligament in 2 patients that re-quired dura mater repair and sealing with bioglue. There were no postoperative Halo-associated complications. One patient developed upper respiratory tract edema fol-lowed by tracheostomy. One elderly patient died due to decompensation of chronic diseases and multiple organ failure. Regression of neurological deficit followed by complete or partial social and labor adaptation was ob-served in other patients.

DiscussionDecompression and craniovertebral junction repair

are preferred for traumatic neural compression of upper cervical spine. This may be achieved by Halo-correction, skeletal traction, manual or instrumental intraoperative repair of vertebrae, etc. As a rule, reposition is possible in acute traumatic dislocations and sometimes in rheuma-toid polyarthritis. In case of correctable dislocation, we preferred Halo-correction for the first stage followed by ventral or dorsal spondylodesis (Fig. 2).

а b

c d

Fig. 3. Transoral decompression in patient with AAD after nonspecific spondylitis.a — Preoperative sagittal CT-scan: odontoid sclerosis, contours are unclear, basion-dens interval (BDI) is reduced to 4 mm; b – preoperative T2-weighted MRI-scan: C2 granuloma causes spinal cord compression (arrow); c – postoperative sagittal CT-scan: odontoid process, C1 anterior arch and 1/2 of C2 body are removed, occipitospondylodesis of craniovertebral junction; d — postoperative T1-weighted MRI-scan: spinal cord without compression.


ORIGINAL ARTICLES

Halo-correction is usually ineffective in chronic in-juries or rheumatoid pannus due to rigid chronic verte-bral displacement. Closed correction is also inappropri-ate in case of ventral compression. In these situations, decompression is achieved by compressing substrate re-section via ventral surgical approach.

The main criterion determining decompression technique was placement of C2 vertebra relative to naso-palatine line (Kassam line) [9, 10]. Transnasal odontoid-ectomy is preferred if axis is located above this line (type A). Both transnasal and transoral approaches may be used for decompression if the line passes through odon-toid base plane (type B). If Kassam line is projected onto

odontoid apex (type C) then transoral decompression is optimal.

In our study transoral access was used in 5 patients. 2 patients had axial AAD (types B and C) due to nonspe-cific inflammatory process and granuloma followed by atlantodental joints destruction (Fig. 3). Halo-uncor-rectable impression and Kassam line corresponding to odontoid apex (type C) were observed in 1 case of rheu-matoid arthritis. In another one there was impression type B but accurate Halo-traction was followed by partial recovery of axial displacement. So, odontoid process was more accessible for transoral resection. This approach

Fig. 4. Intraoperative endoscopic images (b, c) and T2-weighted MRI-scans (a, d) in chronic transligamentous AAD. a — MRI-scan prior to surgery: spinal cord compression with invaginated odontoid process (arrow); b — endonasal approach to odontoid process, C1 arch is partially resected; c — odontoid process is removed, dura mater rupture followed by liquorrhea in the area of apical ligament attachment (arrow); d – postoperative MRI-scan: spinal cord compression is eliminated. D – odontoid process, Ax — C2 body.

а b

c d


was also used in 1 patient with chronic odontoid fracture and compressing substrate below nasopalatine line.

The main disadvantages of transoral surgery are postoperative edema of upper respiratory tract and velo-pharyngeal insufficiency due to dissection or trauma of soft palate [9]. In our sample edema followed by trache-ostomy occurred in one patient after granuloma resection (Kassam line type B).

Endonasal approach has some advantages compared with transoral access [10, 11]: absence of no soft palate injury, postoperative edema of tongue and pharynx, less incidence of infectious complications due to wound con-tamination by less aggressive flora of nasal cavity. In our work, endonasal access was used in 1 patient with chron-ic transligamentous cleavage of atlas. Dislocation was rigid (Halo-uncorrected), and Kassam line correspond-ed to odontoid neck (type B). Transnasal decompression was applied due to some patient’s features (severe cicatri-cial deformation of anterior cervical surface and face with impaired mouth opening). The first stage included pos-terior decompression via C1 laminectomy and large oc-cipital orifice edge resection followed by subsequent oc-cipitospondylodesis. After that endonasal odontoid re-section was carried out. It should be noted that odontoid-ectomy without C1 anterior laminectomy was performed due to sufficient distance between clivus and atlas arch for endoscope and instruments placement (Fig. 4).

Submandibular access is not widely used for ventral decompression. C2 — vertebra and C1 anterior arch are visualized under acute angle via this approach if endos-copy is additionally used [11]. This procedure was made in 1 patient with rheumatoid arthritis in our work. There was Kassam line type B, but impression was corrected up to type C by using of Halo-system. Long thin neck and highly placed lower jaw allowed to visualize C1—C2 ver-tebrae via transcervical approach by using of microscope.

Submandibular access may be also used to mobilize bone fragments in chronic transdental atlas dislocations with subsequent Halo-reposition [8]. In our work it was done in 3 patients.

ConclusionHalo-reposition is followed by complete or partial

repair of ventral compression in certain traumatic and non-traumatic dislocations. It is advisable to define sur-gical approach after preliminary attempt of Halo-reposi-tion. In case of Halo-corrected dislocation and Kassam line within odontoid neck (type B) submandibular access may be used. At the same time, endonasal approach is preferable for Halo-uncorrectable dislocations.


REFERENCES1. Puttlitz CM, Goel VK, Clark CR, Traynelis VC, Scifert JL, Grosland NM.

Biomechanical rationale for the pathology of rheumatoid arthritis in the craniovertebral junction. Spine. 2000;25(13):1607-1616.

https://doi.org/10.1097/00007632-200007010-000032. Maiman DJ, Yoganandan N. Biomechanics of cervical spine trauma. Clin

Neurosurg. 1991;37:543-570. https://doi.org/10.1097/00007632-200007010-000033. Pissonnier mL, Lazennec JY, Renoux J, Rousseau MA. Trauma of the up-

per cervical spine: focus on vertical atlantoaxial dislocation. Eur Spine J. 2013;22(10):2167-2175. https://doi.org/10/1007/s00586-013-2814-2

4. Payer M, Wetzel S, Kelekis A, Jenny B. Traumatic vertical atlantoaxial dis-location. J Clin Neurosci. 2005;12(6):704-706.

https://doi.org/10.1016/j.jocn.2004.03.0435. Ramaré S, Lazennec JY, Camelot C, Saillant G, Hansen S, Trabelsi R. Ver-

tical atlantoaxial dislocation. Eur Spine J. 1999;8(3):241-243. https://doi.org/10.1007/s005860050166

6. Spoor AB, Diekerhof CH, Bonnet M, Cumhur Oener F. Traumatic com-plex dislocation of the atlanto-axial joint with odontoid and C2 superior articular facet fracture. Spine. 2008;33(19):708-711.

https://doi.org/10.1097/brs.0b013e31817c140d7. Botelho RV, Souza Palma AM, Balieiro Abgussen CM, França Fontou-

ra E.A. Traumatic vertical atlantoaxial instability: the risk associated with

skull traction. Case report and literature review. Eur Spine J. 2000;9(5):430-433. https://doi.org/10.1007/s005860000166

8. Nekrasov MA, Lvov IS, Green AA. Surgical treatment of patients with frac-tures of the odontoid process of vertebra C2 in the acute and subacute peri-ods of trauma. Russian Journal of Neurosurgery. 2012;4:17-24. (In Russ.).

9. El-Sayed I.H., Wu J-C, Ames C.P., Balamurali G., Mummaneni P.V. Combined transnasal and transoral endoscopic approaches to the cranio-vertebral junction. J Craniovertebr Junction Spine. 2010;1(1):44-48.

https://doi.org/10.4103/0974-8237.6548110. Zenga F, Villaret AB, Fontanella MM, Nicolai P. Endoscopic transnasal

odontoidectomy using ultrasonic bone curette: technical case report. Neurol India. 2013;61(1):69-72. https://doi.org/10.4103/0028-3886.108015

11. Dasenbrock HH, Clarke MJ, Bydon A, Sciubba DM, Witham TF, Go-kaslan ZL, Wolinsky JP. Endoscopic image-guided transcervical odontoid-ectomy: outcomes of 15 patients with basilar invagination. Neurosurgery. 2012;70(2):351-359.

Received: 23.05.17


ORIGINAL ARTICLES

This article is devoted to one of the most important and urgent problems of current neurosurgery and orthopedics — at-lantoaxial dislocation followed by spinal cord compression. There is no universal treatment strategy for this pathology de-spite modern technical achievements. Therefore, the question of preoperative correction to restore normal anatomical rela-tionship between neural and bone structures is extremely rele-vant.

The authors present a great material consisting of their own experience of AAD treatment. Halo-correction prior to surgery has definitely entered routine surgical practice. There are 34 pa-tients with permanent spinal cord compression in the article. As a result, 24 (70%) out of 34 patients underwent successful Ha-lo-correction with repair of craniovertebral junction and C1—C2 segment. The authors emphasize absent Halo-associated complications.

It was noted that complete correction was achieved only in patients with traumatic AAD. Unfortunately, authors do not indicate terms of injury in Halo-corrected group, grade of cor-

Commentary

rection and type of fixation, time of spondylodesis in long-term period. Was the injury isolated? Of course, all these facts would enrich this study.

Halo-correction can improve treatment of AAD. It would be interesting in the future to compare intraoperative Tong-traction of parietal knolls and Halo-correction in order to con-firm the advantage of the last technique.

Absence of general surgical complications in all cases is ad-mirable. At the same time, meta-analysis of Elliott et al. (69 ar-ticles for 2002—2011, 31,46 patients, outcomes of С1—С2 stabi-lization by Harms/Magerl) reported vertebral artery injury in 6%, screws displacement in 9.5%.

Considering volume of material and novel view on the problem the article is certainly complete research and can be extremely valuable for readers of our magazine.

N.A. Konovalov, V.A. Korolishin (Moscow, Russia)




Rheumatoid Atlantoaxial DislocationsI.YU. LISITSKIY1*, A.M. KISELEV1, S.E. KISELEV2

1Vladimirsky Moscow Regional Research Clinical Institute. Shchepkina Str., 61/2, Moscow, Russia,129110, 2Dolgoprudny Central City Hospital, Pavlova Str., 2, Dolgoprudny, Moscow Region, Russia, 141704

Pathological processes in the craniovertebral region (CVR) are usually accompanied by dislocation complications leading to gross neurological disorders. One of the diseases that affect the CVR and lead to atlanto-axial dislocation (AAD) is rheumatoid arthritis. Errors in the diagnosis of rheumatoid disease and in the choice of a treatment approach may cause adverse outcomes.Objective — to define the approach for surgical treatment of AAD associated with rheumatoid disease of the CVR, with allowance for the rigidity of deformity.Material and methods. Five patients with rheumatoid AAD, 4 females and 1 male, aged 54 to 73 years underwent surgery. All dislocations were anterior ones. Three patients had mobile pannus-associated dislocation. In 2 cases, AAD was rigid and combined with odontoid invagination into the foramen magnum (FM).Results. In all mobile AAD cases, decompression of the brainstem and restoration of the normal anatomical relationships in the CVR were achieved by dislocation correction and atlanto-axial fusion performed from the posterior approach, avoiding transoral interventions. In this case, spontaneous resorption of the pannus occurred within several months after surgery. In the postoperative period, all patients achieved significant regression of pain and neurological disorders. Complications in the form of wound infection developed in 1 case.Conclusion. A decision algorithm for choosing a surgical treatment option was based on the degree of deformity stability. Mobile AADs serve as an indication for indirect decompression using instrumental correction of dislocation and atlantoaxial fixation from the posterior approach. In the case of fixed AAD, posterior fixation is complemented by anterior decompression via the transoral approach.

Keywords: craniovertebral region, atlanto-axial dislocation, transoral approach, atlanto-axial fusion, occipitocervical fusion.

Atlantoaxial dislocation (AAD) is pathomorphologi-cal substrate of various craniovertebral injuries and dis-eases rather independent pathology. Rheumatoid arthri-tis is often cause of AAD besides traumatic, inflamma-tory, tumoral reasons [1—3].

Rheumatoid arthritis is chronic autoimmune sys-temic connective tissue disease affecting small joints as a rule. Its incidence is 0.8—2% [4]. Rheumatoid arthritis is predominantly observed among women (3 times more often than in men) [1, 4]. Manifestation occurs on the 4th-6th decade of life [5, 6]. According to different au-thors [5—7], 40—80% of patients with rheumatoid ar-thritis have clinical symptoms of cervical spine lesion with x-ray signs in most of them. Herewith, in 65% of patients with cervical spine subluxations and deformities rheumatoid arthritis destroys atlantoaxial junction fol-lowed by AAD.

Craniovertebral junction (CVJ) is unique anatomical complex aimed at protection of vital cerebral and vascu-lar structures and various motor functions. Complex bio-mechanics of CVJ is provided by atlantooccipital and at-lantoaxial joints and their ligaments. While paired atlan-tooccipital joint is quite stable, atlantoaxial junction con-sisting of paired lateral atlantoaxial and median atlanto-axial joints (Craveilhier’s joint) is more mobile complex providing 60% of head movements [8]. However, the last is followed by certain vulnerability and dislocation due to CVJ trauma and diseases including rheumatoid process. AAD pathogenesis in rheumatoid arthritis is associated

with destruction of capsular-ligament and bone struc-tures of atlantoaxial junction and its instability. Neuro-logical complications are caused by brainstem compres-sion due to odontoid dislocation or rheumatoid granulo-ma (pannus) [9].

There are several types of dislocations in CVJ rheu-matoid lesions: anterior (49%), lateral (20%), posterior (7%) and rotational (extremely rare). They may be non-fixed or rigid. Rigidity of dislocation is determined by stage of arthritis. Late stage is followed by anchylosis and rigid AAD. In 38% of cases AAD is combined with basi-lar invagination because of destruction of lateral atlanto-axial joints [10].

Clinical symptoms of rheumatoid cervical spine le-sion are not always observed. Some authors [11, 12] re-ported pain syndrome alone in 40—88% of patients with x-ray signs of rheumatoid lesion of cervical spine and neurological disorders in only 7—34% of them. Clinical picture of AAD is accompanied by various neurological disorders due to brainstem and superior spinal cord com-pression and impaired vertebrobasilar circulation.

Material and methodsIn 2015—2016 we have operated 5 patients with rheu-

matoid AAD including 4 women and 1 man aged 54—73 years. Rheumatoid arthritis with anamnesis over 20 years was diagnosed in all patients. There were ante-rior AAD in all of them. Three patients had pannus dislo-


ORIGINAL ARTICLES

cation. In 2 cases AAD was rigid and combined with odontoid invagination into large occipital orifice. AAD severity was determined by anterior atlantodental interval and varied from 6 to 10 mm.

All patients with rheumatoid disease underwent functional spondylography besides conventional preop-erative survey in order to assess grade of AAD rigidity. It was necessary to define treatment strategy.

In some cases, ataxia, neurogenic dysfunction of pel-vic organs, bulbar disorders, transient vertebrobasilar ischemic circulatory disorders were observed besides

pain syndrome, spastic tetraparesis and conductive sensi-tivity disorders. Advanced neurological deficit rather pain syndrome was a cause for recourse of medical care as a rule.

Severity of neurological disorders was directly related to AAD grade. Indication for surgery was anterior atlan-todental interval over 6 mm.

Surgical approach was determined by stability of de-formation. Mobile AAD was an indication for indirect decompression via instrumental repair of dislocation and atlantoaxial fixation through posterior access. Anterior

Table 1. Ranawat classification of neurological deficit

Grade SignGrade I Pain, no neurological deficitGrade II Moderate myelopathy with weakness, hyper-

reflexia, dysesthesiaGrade IIIA Severe conductive disorders, patient moves with

difficulties, ambulatoryGrade IIIB Extremely severe neurological deficit, patient

cannot move by himself or move with external support, non-ambulatory

Table 2. Ranawat pain scale

Pain ValueNo pain 0Mild 1 (responsive to analgesics)Moderate 2 (responsive to immobilization)Severe 3 (unresponsive to both analgesics and

immobilization)

Fig. 1. Case report №1. Preoperative survey.a, b — functional spondylograms: non-fixed AAD (arrows); c, d — MRI-scan: severe brainstem compression by dislocated odontoid process followed by focal myelomalacia at this level and pannus (arrows); e, f – CT-scan: anterior AAD and basilar impression (arrows).


decompression through transoral access in addition to posterior fixation was applied for rigid AAD.

We have used C. Ranawat scales (Tables 1 and 2) for assessment of neurological disorders and pain syndrome severity in case of rheumatoid lesion of CVJ [13].

Three patients with unfixed dislocations underwent instrumental repair of deformation and atlantoaxial fixa-tion with screws-based metal construction. Screws were inserted into C1 lateral masses and transpedicularly into C2 body by J. Harms followed by instrumental correction of dislocation and metal construction deployment. One patient with insufficient AAD correction required ar-throlysis of lateral atlantoaxial joints in order to increase mobility. Bilateral exposure of C2 spinal radixes was ap-plied to visualize these joints.

In case of rigid AAD combined with basilar invagina-tion we performed combined simultaneous procedures. There were C0—C2—C3 occipitospondylodesis in one case and posterior fixation of C1—C2 by Harms in an-other one followed by transoral decompression (resec-tion of atlas anterior arch and odontoid process).

Case report № 1

Patient A., 54 years old. Diagnosis: rheumatoid le-sion of CVJ. Non-fixed AAD. Spastic tetraparesis. Pelvic neurogenic dysfunction. Transient vertebrobasilar isch-emic disorders.

Patient has suffered from rheumatoid arthritis for 20 years with lesion of hands and feet joints, knee joints followed by their secondary deformation. There were pain syndrome, severe spastic tetraparesis followed by arms and legs weakness up to 3 scores, cerebellar and sen-sitive ataxia, neurogenic pelvic dysfunction of central type, transient circulatory disorders ("drop-attacks") in sharp change of head position (Ranawat III B, pain 2 scores). Lhermitte's symptom was observed. Neurolog-ical symptoms occurred 2 years before hospitalization with gradual progression.

Functional spondylography revealed non-fixed de-formation (Fig. 1a, b). Magnetic resonance imaging (MRI) and computed tomography (CT) confirm AAD followed by severe brainstem compression with dislocat-ed odontoid, focal myelomalacia and rheumatoid granu-loma (pannus) at the C1—C2 level (Fig. 1c—f).

Fig. 2. Case report №1. Survey data after instrumental repair of dislocation and atlantoaxial spondylodesis by J. Harms.a, b — functional spondylograms, c, d – CT-scan: normal anatomical interrelations of CVJ; e, f — MRI-scan: normal anatomical interrelations of CVJ, no inflammation and spontaneous resorption of rheumatoid granuloma.


ORIGINAL ARTICLES

Surgery: in prone position with head fixation in May-field clamp the screws were inserted into C1 lateral mass-es and transpedicularly into C2 body, dislocation was in-strumentally repaired followed by metal construction deployment.

Postoperative period was uneventful. There were arms and legs strength recovery up to 4—5 scores, relief of

pain syndrome, regression of sensitive disorders and nor-malization of extremities’ muscle tone, no vertebrobasi-lar circulatory disorders (Ranawat I, pain 0 scores).

Control CT and spondylography confirmed restora-tion of correct anatomical interrelations in CVJ and ab-sent brainstem compression (Fig. 2a—d). MRI in 8 months after surgery revealed complete spontaneous resorption of rheumatoid pannus (Fig. 2e, f).

Case report № 2

Patient A., 68 years old. Diagnosis: rheumatoid le-sion of CVJ. Rigid AAD. Spastic tetraparesis.

Anamnesis of rheumatoid arthritis followed by sec-ondary deformation of joints is over 20 years. Neurologi-cal disorders included severe spastic tetraparesis with arms and legs weakness up to 3 scores, bulbar disorders (Ranawat III B, pain 0 scores). Neurological disorders have appeared several months before hospitalization.

CT and functional spondylography revealed rigid AAD with odontoid dislocation into large occipital ori-fice and severe brainstem compression (Fig. 3a—d).

Surgery: patient’s prone position, we have firstly car-ried out resection of large occipital orifice’s margin and posterior hemiarch of atlas, arthrolysis of lateral atlanto-axial joints, occipitospondylodesis with screw-based metal construction and repair of deformation. Therefore, in supine position patient underwent transoral resection of lower clivus, anterior hemiarch of atlas, odontoid pro-cess up to dura mater.

Event-free postoperative period was observed. There was favorable neurological dynamics including augmen-tation of hands and legs strength up to 4 and 3 scores, respectively, regression of sensitive disorders and nor-malization of extremities’ muscle tone (Ranawat III A, pain 0 scores).

Fig. 3. Case report №2. Preoperative survey.a, b — functional spondylograms: rigid AAD; c, d – CT-scan: anterior AAD and basilar impression.

Fig. 4. Case report №2. Survey data after instrumental repair of deformation, occipitospondylodesis and transoral odontoidectomy.a, b – CT-scan: no medulla oblongata compression and normal anatomical relationships of CVJ.


According to control CT there was no brainstem compression due to both decompression and instrumen-tal repair of CVJ deformation (Fig. 4).

Results1-year outcomes were followed-up. Significant post-

operative regression of pain syndrome and neurological disorders were observed in all patients (Table 3). Postop-erative wound infection occurred in 1 patient after sur-gery through posterior approach. Drainage and irrigation of the wound were effective. Brainstem decompression followed by CVJ repair via correction of dislocation and atlantoaxial spondylodesis through posterior approach were successfully achieved in all cases of non-fixed AAD. Herewith, transoral interventions were not neces-sary and spontaneous resorption of pannus in few months after surgery occurred. Grade of cervical spine rotation became physiological within follow-up in patients after atlantoaxial spondylodesis. Significantly limited rota-tion, flexion and extension of cervical spine were ob-served only in 2 cases after occipitospondylodesis.

DiscussionEarly AAD is characteristic for rheumatoid injury of

CVJ. Risk of severe neurological complications deter-mine indications for surgery. Correction should be car-ried out as early as possible in these cases in view of unfa-vorable prognosis.

The purpose of surgery for rheumatoid AAD is brain-stem decompression and reliable fixation of affected ar-ea. Decompressive laminectomy and advanced posterior fixation is now necessary to be considered redundant in treatment of rheumatoid injury of CVJ. Instrumental re-pair of AAD and reliable "short" fixation via posterior ap-proach are followed by brainstem decompression, resto-ration of normal CVJ anatomy, minimum surgical trau-ma and significant improvement of functional outcomes. It is possible due to not only current equipment, but also features of AAD including stability of deformation that determines surgical strategy.

AAD rigidity depends on duration of rheumatoid ar-thritis. Dislocation is non-fixed in manifestation of rheu-matoid lesion. Neurological complications are caused by dislocated odontoid process and rheumatoid granuloma. In these cases, repair of deformation is possible via poste-rior fixation by Harms. This technique avoids prolonged fixation and provides optimal functional outcomes. Moreover, C1—C2 fixation is followed by discontinu-ance of inflammation and spontaneous resorption of pannus.

Late stage of rheumatoid arthritis is accompanied by rigid AAD due to anchylosis. Indications for surgery in these cases are determined by neurological disorders.

Transoral decompression is not justified if AAD is even minimally mobile. In this situation, arthrolysis of atlantoaxial joints, posterior fixation of C1—C2 by Harms and instrumental correction of AAВ are associat-ed with indirect brainstem decompression while transoral approach is avoided.

Transoral decompression is indicated only for rigid AAD after occipitospondylodesis or posterior fixation of C1—C2 by Harms.

ConclusionAll patients with rheumatoid arthritis should be spe-

cifically screened for early diagnosis of CVJ damage. Dis-location is an absolute indication for surgery due to risk of neurological disorders.

Whether deformation is stable is key factor to deter-mine surgical approach.

Instrumental correction of deformation and atlanto-axial spondylodesis via posterior approach are indicated for non-fixed AAD even in neurological symptoms.

Rigid AAD followed by neurological disorders is in-dication for surgery. Occipitospondylodesis or posterior fixation of C1—C2 and transoral decompression are re-quired in these cases.

Our approach for AAD is accompanied by restora-tion of normal anatomy of CVJ, brainstem decompres-sion and minimal postoperative functional disorders.


Table 3. Surgical outcomes

Patients Gender Age, years Pain (Ranawat score) Neurological deficit (Ranawat classification)Prior to surgery After surgery Prior to surgery After surgery

1 F 54 2 0 III B I2 F 68 0 0 III B III A3 F 62 2 1 III A II4 М 58 2 0 III A I5 F 73 1 0 III B III A


ORIGINAL ARTICLES

REFERENCES1. Vetrile ST, Kolesov SV. Kraniovertebral’naya patologiya. M.: Meditsina;

2007. (In Russ.).2. Lutsik AA, Ratkin IK, Nikitin MN. Kraniovertebral’nye povrezhdeniya i

zabolevaniya. Novosibirsk: Izdatel’; 1998. (In Russ.).3. Clark CR. Cervical spine involvement in rheumatoid arthritis. Iowa Med.

1984;74(2):57-62.4. Linos A, Worthington JW, O’Fallon WM, Kurland LT. The epidemiology of

rheumatoid arthritis in Rochester, Minnesota: a study of incidence, preva-lence, and mortality. Am J Epidemiol. 1980;111(1):87-98.

https://doi.org/10.1093/oxfordjournals.aje.a1128785. Weissman BN, Aliabadi P,Weinfeld MS, ThomasWH, Sosman JL. Prog-

nosticfeatures of atlantoaxial subluxation in rheumatoid arthritis patients. Radiology. 1982;144(4):745-751.

https://doi.org/10.1148/radiology.144.4.71117196. Lee DM, Weinblatt ME. Rheumatoid arthritis. Lancet. 2001;358:903-911. https://doi.org/10.1016/s0140-6736(01)06075-57. Boden SD, Dodge LD, Bohlman HH, Rechtine GR. Rheumatoid arthritis

of the cervical spine: a long-term analysis with predictors of paralysis and recovery. J Bone Joint Surg Am. 1993;75(9):1282-1297.

https://doi.org/10.2106/00004623-199309000-00004

8. Panjabi MM, Crisco JJ, Vasavada A, Oda T, Cholewicki J, Nibu K, Shin E. Mechanical properties of the human cervical spine as shown by three-di-mensional load-displacement curves. Spine. 2001;26(24):2692-2700.

https://doi.org/10.1097/00007632-200112150-000129. Delamarter RB, Bohlman HH. Postmortem osseous and neuropathologic

analysis of the rheumatoid cervical spine. Spine. 1994;19(20):2267-2274. https://doi.org/10.1097/00007632-199410150-0000410. Reiter MF, Boden SD. Inflammatory disorders of the cervical spine. Spine.

1998;23:2755-2766. https://doi.org/10.1097/00007632-199812150-0001711. Nguyen HV, Ludwig SC, Silber J, Gelb DE, Anderson PA, Frank L, Vacca-

ro AR. Rheumatoid arthritis of the cervical spine. Spine. 2004;4(3):329-334. https://doi.org/10.1016/j.spinee.2003.10.006

12. Krauss WE, Bledsoe JM, Clarke MJ, Nottmeier EW, Pichelmann MA. Rheumatoid arthritis of the craniovertebral junction. Neurosurgery. 2010;66(3):A83-A95. https://doi.org/10.1227/01.neu.0000365854.13997.b0

13. Ranawat CS, O’Leary P, Pellicci P, Tsairis P, Marchisello P, Dorr L. Cer-vical spine fusion in rheumatoid arthritis. J Bone Joint Surg Am. 1979;61(7):1003-1010.

https://doi.org/10.2106/00004623-197961070-00006

Received: 31.05.17

Five cases of rheumatism-associated atlantoaxial disloca-tion are described in the article. The authors use generally ac-cepted tactics depending on type of instability (anterior, poste-rior) and duration of disease. All variants of surgical approaches

Commentary

(including transoral) as well as various types of local and ad-vanced stabilization are used.

The article is of interest due to rare pathology.

A.O. Gushcha (Moscow, Russia)


Currently, surgeon's intraoperative work is associated with the need for constant access to neuroimaging data (MRI and CT). Availability of computer in operating theatre which is integrated into local hospital network with possible visualization of DICOM-files is integral component of modern surgical units. Thus, on-line in-traoperative data about certain patient are available for surgeon. These systems are controlled by additional em-ployee who does not directly participate in surgery. In our opinion, development and introduction of equip-ment with out-of-touch control is advisable due to some advantages for surgeon and other fellow workers. In this situation surgeon is able to obtain independently neces-sary intraoperative information while other participants are free from this process. Results of Opect system ap-plication are described in the article. The aim of this work is to evaluate system’s effectiveness, learning curve and quality of neuroimaging data.

There were no purposes to calculate cost of this equipment and its comparison with intraoperative work of personnel.

The objective of the work is assessment of contactless control of MRI and CT-scans viewing in neurosurgical theatre.

Material and methodsIn December 2014 innovative equipment was de-

ployed in one of operational theatres of Novosibirsk Fed-eral Neurosurgery Center — Opect optical system for in-traoperative out-of-touch control of MRI and CT-scans viewing. The last was developed at the Tokyo Women's Medical University.



Opect Optical System in Neurosurgical PracticeA.V. KALINOVSKIY*1, D.А. RZAEV1, K. YOSHIMITSU2

1Federal Center of Neurosurgery, Nemirovich-Danchenko Str., 132/1, Novosibirsk, Russia, 630087, 2Tokyo Women’s Medical University, Tokyo, Japan

Introduction. Modern neurosurgical practice is impossible without access to various information sources. The use of MRI and MSCT data during surgery is an integral part of the neurosurgeon’s daily practice. Devices capable of managing an image viewer system without direct contact with equipment simplify working in the operating room. Aims and objectives. To test operation of a non-contact MRI and MSCT image viewer system in the operating room and to evaluate the system effectiveness. Material and methods. An Opect non-contact image management system developed at the Tokyo Women’s Medical University was installed in one of the operating rooms of the Novosibirsk Federal Center of Neurosurgery in 2014. In 2015, the Opect system was used by operating surgeons in 73 surgeries performed in the same operating room. The system effectiveness was analyzed based on a survey of surgeons. Results. The non-contact image viewer system occurred to be easy-to-learn for the personnel to operate this system, easy–to-manage it, and easy-to-present visual information during surgery. Conclusions. Application of the Opect system simplifies work with neuroimaging data during surgery. The surgeon can independently view series of relevant MRI and MSCT scans without any assistance.

Keywords: Opect system, non-contact neuroimaging data management.

There are 3 components in the system (Fig. 1). The main part is a tablet computer (Windows OS) integrated into local computer network and equipped with special software for images viewing. Necessary DICOM (or JPEG) files are previously loaded into the tablet from lo-cal network. Screen for displaying of images is located on one of the walls in operating theatre so surgeon standing near operating table is able to see them. Kinect infrared antenna fixing surgeon's movements is directed toward operating table. Surgeon controls images on the screen by

Fig. 1. Opect system’s elements (explanations in text).


ORIGINAL ARTICLES

these movements. Kinect (Microsoft) is out-of-touch touchscreen game controller introduced for personal computers (Windows OS). It allows the user to interact with computer without contact device (joystick or key-board). Infrared emitter of antenna creates a beam of rays towards the surgeon which are reflected from him and return to device. Then these rays are perceived by camera and analyzed. System assessed human movements in three planes (left-right, up-down, forward-backward). Optimal distance from antenna to surgeon is 2—3 m (Fig. 1).

System is ready as soon as images are downloaded into tablet computer. Surgeon can intraoperatively assess any files at any time. For this purpose, he must be within coverage area of Kinect infrared antenna. Program is au-tomatically activated, color of image’s frame on the screen changes from red to green. Surgeon’s silhouette appears on the screen and physician is able to control process from this moment by using his hand. The system responds to the hand movements like a pendulum when the palm is uplifted and turned to the antenna. Surgeon’s hand fixed by the antenna is displayed on the screen as green sphere that is analog of cursor. Primary slide of the first MRI or CT scans series downloaded into system is displayed on the monitor. Image is identical to primary DICOM-scan obtained by MRI or CT. Whole informa-

tion about patient is presented on the image including its number in series and magnification in relation to original file. Images may be changed by using of interface of the right half of screen with corresponding arrows (Fig. 2).

Sphere navigation on one of these icons activates se-ries scrolling forward or backward. 2-, 4- or 8-fold mag-nification is available. Grade of magnification is dis-played in upper right part of monitor. Smaller original image with subsequently enlarged area is located in lower right part of monitor for convenience. Palm’s movement towards the screen is needed to change image magnifica-tion. Icon of four arrows appears after that (up, down, left, right) to analyze entire picture in different directions (Fig. 3).

Cursor’s movement towards upper screen is followed by change of series of files. Thus, surgeon can receive all necessary information while he is aseptic and without help of other persons.

At the Novosibirsk Federal Neurosurgical Center Opect system has been used in 73 procedures for the pe-riod from January to December 2015. Patients’ charac-teristics are presented in the Table.

Three surgeons of the oncology department of the Novosibirsk Federal Neurosurgical Center took part in the study. All Opect system-assisted interventions were carried out in the same operating theatre. In each case

Fig. 2. MRI-data scrolling within one series.

Fig. 3. Transition within magnified MRI-scan towards pathological area. x 4.0.

Application of Opect system in various cerebral procedures

Disease NumberIntracranial meningioma 26Glial cerebral tumor 19Pituitary adenoma 17Craniopharyngioma 6Microvascular decompression for trigeminal neuralgia 3Vestibular schwannoma 2Total 73


surgeon decided to view MRI and CT-images in neces-sary manner by himself. Postoperatively surgeon ex-pressed his opinion about system regarding intraopera-tive visualization of the images. Herewith, he pointed out the most relevant factors for us:

— learning time required for working with system in the future;

— simplicity or complexity of training;— quality of data;— system’s drawbacks.

ResultsSimple and convenient application of system were

noted in all cases. Learning time was 3—5 minutes in all 3 surgeons who have used equipment in following inter-ventions. Re-training or further clarification for work with system were not required during the 2nd and subse-quent procedures. Interface is clear and easy for under-standing. System is able to view any DICOM or JPEG data including angiographic scans. However, intraopera-tive 3D-modeling and viewing of 3D-format are impos-

sible. Drawback of this system is need to duplicate imag-ing data from storage server to tablet by using of special program that takes few minutes. Since system activated it is necessary to exclude foreign moving objects (equip-ment, personnel) in camera’s coverage field to prevent false triggering. Number of series is limited (≤3) that may be unsatisfactorily for surgeon and requires more careful preoperative assessment and selection of the most infor-mative scans.

Despite above-mentioned shortcomings there was no need to use additionally stationary computer for MRI and CT-scans viewing among 73 procedures. In our opinion, the last fact indicates adequate work of system regarding intraoperative neuroimaging.

ConclusionIn our opinion, Opect system can greatly simplify in-

traoperative neuroimaging without additional help. We consider that introduction of intraoperative out-of-touch equipment is followed by great prospects for further im-provement of surgical interventions.


Report of A.V. Kalinovskiy et al. is dedicated to current technology of out-of-touch control of images viewing in neuro-surgical operating theatre. This technology is realized via infra-red camera tracking surgeon's hands movements. The problem of DICOM-files control within aseptic surgical environment is obviously relevant. Routine practice requires additional staff or technical specialists sometimes to use image viewer software during surgery that increases the number of people in operating theatre. Surgeon gives instructions verbally about necessary images that can disrupt work under stressful conditions when rapid decision-making is required and quick response is impor-tant.

It is described modern system of out-of-touch interaction with computer («Opect») which was developed at the Tokyo Women's Medical University (Japan). This product was created on the basis of computer games technologies and is designed to improve safety of medical images viewing within neurosurgical

operating theatre. In most cases, only one surgeon is required to work with system, while computer is controlled by gestures without additional input devices. The authors summarize their positive experience of Opect system application. The number of Opect-assisted interventions (n=73) is more than 2 times higher than that in report of K. Yoshimitsu et al. in 2014 (n=30). For the first time ever this report has shown an effectiveness of this system. Russian authors confirmed the results Japanese col-leagues.

Advanced computer and information technologies are fol-lowed by extended capabilities to work with medical informa-tion in neurosurgery. Adequate representations about effective-ness and convenience of such tools are essential for neurosurgi-cal community. Opect system demonstrates the possibility to improve efficiency, safety and comfort of operating team’s work through the use of modern computer solutions.

S.A. Goryainov (Moscow, Russia)

Commentary


ORIGINAL ARTICLES



Evidence-Based Medicine: Propensity Score Matching (PSM) Approach to Eliminate Systematic Selection Bias in Retrospective Neurosurgical TrialsA.V. MOSKALEV1, 2*, V.S. GLADKIKH1, 2, A.A. AL’SHEVSKAYA1, 3, A.P. KOVALEVSKIY4, A.I. SAKHANENKO4, 5, K.YU. ORLOV6, N.A. KONOVALOV7, A.V. KRUT’KO8

1Research Center for Biostatistics and Clinical Research, Acad. Lavrentieva Ave., 6/1, Novosibirsk, Russia, 630090; 2Institute of Computational Mathematics and Mathematical Geophysics, Siberian Branch of the Russian Academy of Sciences, Acad. Lavrentieva Ave., 6, Novosibirsk, Russia, 630090; 3Research Institute of Fundamental and Clinical Immunology, Yadrintsevskaya Str., 14, Novosibirsk, Russia, 630099; 4Novosibirsk State University, Pirogova Str., 1, Novosibirsk, Russia, 630090; 5Institute of Mathematics, Siberian Branch of the Russian Academy of Sciences, Universitetskiy Ave., 4, Novosibirsk, Russia, 630090; 6Meshalkin Siberian Federal Biomedical Research Center, Novosibirsk, Russia, 630055; 7Burdenko Neurosurgical Institute, 4-ya Tverskaya-Yamskaya Str., 16, Moscow, Russia, 125047; 8Tsiv’yan Novosibirsk Research Institute of Traumatology and Orthopedics, Frunze Str., 17, Novosibirsk, Russia, 630091

To date, a large amount of retrospectively collected data about treatment of neurosurgical pathology have been accumulated. Modern methods of medical statistics are necessary for correct interpretation of the data. The article purpose is to demonstrate application of one of the modern methods, Propensity Score Matching (PSM), in neurosurgery. Results and discussion. The use of PSM avoids misinterpretation of retrospectively collected data and obviates errors in planning further prospective studies. For the past 10 years, the number of published international PSM-based studies has increased more than 10-fold, with the number of articles by Russian authors accounting for less than 0.2%. In line with the tendencies of international studies, application of PSM in analysis of retrospectively collected data will enable testing of a number of hypotheses and correct planning of prospective randomized studies.

Keywords: Propensity Score Matching, neurosurgery, medical statistics, selection bias.

Correct randomization is followed by equivalent distri-bution of known and unknown factors affecting out-comes. This is the fundamental condition for analysis of outcomes in both groups. In other cases, RCTs are either impossible or extremely time-consuming due to various reasons. These causes are too large sample size, limited budget and/or time, impossible randomization because of ethical considerations or significantly different inter-ventions, etc. [3].

PSM for data analysis

If RCTs are impossible for any reasons, other types of trials are preferred. Usually, it is a retrospective analysis of large sample size for a long time. In this case, system-atic selection bias may be followed by significant inaccu-racies or incorrect conclusions. That is why editors of authoritative journals try to avoid publishing of retro-spective studies with potential systematic selection bias.

PSM is effective to achieve equivalent distribution by certain key characteristics. Key characteristics of the groups include the most significant variables (clinical and demographic). PSM is followed by reduced system-atic selection bias and equivalent groups [4]. So, qualita-tive retrospective analysis and highly evidenced conclu-sions are obtained.

There are 2 steps in PSM algorithm: correction of se-lection bias in order to obtain comparable groups and di-rectly comparative analysis of modified data.

PSM foundations were considered in the middle of the last century [1, 5], detailed development has occurred

Experience of national neurosurgeons over the last years is a unique basis for retrospective analysis. It is nec-essary for comparative evaluation of efficacy and safety of various interventions, objectification of results from large heterogeneous samples of patients (including different centers) and planning of subsequent prospective studies. However, considerable heterogeneity of initial data due to systematic sampling error can significantly hinder di-rect comparison of results and influence conclusions. In this regard correct methodology of retrospective analysis and consequently adequate statistical processing are needed.

The design of most clinical trials is based on com-parison of different effects and identifying relationships between approaches and outcomes [1]. Effect can consist of not only different treatment approaches but also influ-ence of environmental parameters, attachment to partic-ular social group or risk group. As a rule, the main pur-pose of medical researches is assessment of correlation between some factors and observed outcomes. Require-ment of comparable groups in analysis is fundamental, so equivalence of groups by key factors should be controlled. It is may be achieved either by randomization (random-ized controlled trials, RCTs) or by using of Propensity Score Matching (PSM) and its analogues [2] followed by elimination of systematic selection bias. Usually only two groups (control and main) are considered, so we will use such design for analysis.

RCTs are "gold standard" to correct systematic selec-tion bias in treatment safety and efficacy assessment.


Fig. 1. Increasing number of articles with PSM in SCOPUS database.

Fig. 2. Part of the table from the article "Impact of Residents on Surgical Outcomes in High-Complexity Procedures" [17] showing different results before and after PSM application.

Fig. 3. Part of the table from the article "Therapeutic modulation in severe or moderate traumatic brain injury: a propensity score analysis of data from the Nationwide Japan Neurotrauma Data Bank" [20] showing different results before and after PSM application.


ORIGINAL ARTICLES

in 1970—1990 [6—8]. Advanced computer technology at the end of the last century was followed by development of software products to implement this approach. So, PSM has become available for wide range of research-ers [9]. Nevertheless, technical features of PSM can cause difficulties in medical researchers. However, spe-cialist for biostatistics can facilitate this problem.

PSM is based on conditional probability of each re-search object falling into intervention group. Clinically relevant factors must be determined to create propensity scores (PS). Since PS is a probability, its values are in the range from 0 to 1. Multivariate logistic regression model is usually applied for PSM.

Conditional probability may be balanced between intervention and control groups by using of PS values. Various PSM approaches are used for this pur-pose [10] — stratification/subclassifications, inverse probability of treatment weighting (IPTW) and covariate adjustment. Choice of the most effective method to cor-rect systematic selection bias becomes the object for par-ticular trial. None of available methods is sufficiently universal and applicable in all situations. For example, PSM has some advantages in contrast to regression ad-justment: simple analysis of models’ quality, closer to RCTs idea due to separate phases of subgroup selection and analysis and other advantages which are comprehen-sively presented in P. Austin’s report [11].

PSM is associated with reduced systematic selection bias, so this approach is considered as a kind of random-ization for retrospective data. However, there is principal difference between both methods: randomization elimi-nates systematic selection bias for both considered and unaccounted factors, while PSM corrects selection bias only for factors considered and measured.

To date, the number of publications with PSM for analysis and retrospective data interpretation is steadily growing [12]. There are about 3,000 publications in 2016 according to SCOPUS database (Fig. 1). At the same time, for the period 2007—2017 Russian publica-

tions have accounted only 0.15% among more than 15 000 trials [13, 14] in international databases.

Publications for neurosurgical problems are also few among them [15—30]. We see a steady interest of re-searchers to use PSM in neurosurgical studies. For ex-ample, PSM-assisted trial was published in Spine journal in 2017 [31]. We expect that retrospective analysis will become an integral part of researches [8]. Considering advanced ethical requirements and accumulation of large amount of these data relevance and effectiveness of their analysis will be increased. It is not surprisingly, that de-velopment of PSM and its analogs is one of the most im-portant areas of medicine.

PSM is used the report of V. Ferraris et al. [17] where the authors investigated the effect of surgical team on the outcomes of highly difficult general surgery, neurosur-gery, vascular surgery, cardiothoracic surgery. Since lit-erature data about the role of trainee surgeon in interven-tions are extremely controversial, the authors have ana-lyzed treatment of 266 411 patients.

Large database was retrospectively analyzed by using of PSM to minimize systematic sampling bias. Funda-mentally different conclusions were obtained with PSM approach despite initial assumption about assistant’s negative impact on the outcomes according to prelimi-nary analysis (Fig. 2).

In particular, analysis of bias-uncorrected data re-vealed significantly higher trainee-associated intraopera-tive and overall mortality. There were other data if PSM was applied: participation of trainee surgeon positively affects surgical outcomes (less mortality from postopera-tive complications) and does not influence intraoperative and overall mortality.

This example demonstrates that bias-corrected ap-proach is followed by other outcomes and conclusions.

The article of K. Miyata et al. (Japan) is an example of systematic selection bias minimization with PSM can radically change results of observational trial [20]. Effect of therapeutic temperature modulation (TTM) on neu-

Fig. 4. Part of the table from the article "Risks associated with preoperative anemia and perioperative blood transfusion in open surgery for intracranial aneurysms" [23] showing different results before and after PSM application.


rological outcomes and mortality rate was assessed in this article (Fig. 3).

There were 687 patients with brain injuries. Prelimi-nary analysis of bias-uncorrected data confirmed that TTM procedure was significantly associated with lower risk of poor neurological outcome and death. However, trivial negative effect of this procedure was observed as soon as groups were equalized by age, severity of injury, terms and conditions of medical care.

A. Seicean et al. [23] analyzed outcomes of surgery for intracranial aneurysms in patients with/without se-vere anemia requiring transfusion. There were 668 pa-tients who underwent open surgery in 2006—2012. PSM was applied due to preoperative heterogeneity of groups (age, features of aneurysms, comorbidities and risk fac-tors for adverse outcomes and complications).

Different incidence of early major complications be-fore and after PSM application was demonstrative. In groups with/without anemia incidence of complications changed from 20 (25% vs. 45% prior to PSM) to 10% (34% vs. 44% after PSM).

Incidence of complications is often used to deter-mine sample size needed for subsequent randomized controlled trials. If we will accept difference near 20% in above-described example, then predicted sample size

will be at least 3 times less than necessary to obtain reli-able data (Fig. 4).

As examples, we have considered some publica-tions [20, 23, 24] for 2015—2016 and significance of PSM for analysis has been presented.

ConclusionPSM significantly affects interpretation of retrospec-

tive data and design of subsequent studies. Correction of systematic selection bias is followed by more accurate es-timation of sample size for further trials and adequate interpretation of results. PSM for analysis of retrospec-tive data of national neurosurgical centers will allow us to test some hypotheses and to schedule further RCTs.

The research is partially supported by grants of the Rus-sian Foundation for Basic Researches 17-01-00683 A, 17-41-540759 p_а.


REFERENCES1. Fisher SRA. The Design of experiments. Hafner Publishing Company.

1971;1-27.

2. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399-424. https://doi.org/10.1080/00273171.2011.568786

3. Concato J, Shan N, Horwitz R. Randomized, controlled trials, observa-tional studies, and the hierarchy of research designs. N Engl J Med. 2000; 342(25):1887-1892.

4. Brookhart MA, Wyss R, Layton JB, Stürmer T. Propensity score methods for confounding control in nonexperimental research. Circ Cardiovasc Qual Outcomes. 2013;6(5):604-611.

https://doi.org/10.1161/CIRCOUTCOMES.113.0003595. Cochran WG. Matching in analytical studies. Am J Public Health.

1953;43(6 :1):684-691.6. Rubin DB. Matching to remove bias in observational studies. Biometrics.

1973;29(1):159-183. https://doi.org/10.2307/25296847. Rosenbaum PR, Rubin DB. The central role of the propensity score in ob-

servational studies for causal effects. Biometrika. 1983;70(1):41-55. https://doi.org/10.1093/biomet/70.1.418. Rubin DB. Estimating causal effects of treatments in randomized and non-

randomized studies. J Educ Psychol. 1974;66(5):688-701. https://doi.org/10.1037/h00373509. Bergstralh EJ, Kosanke JL. Computerized matching of cases to controls.

Technical Report Series. 1995;56.10. D’agostino RB. Tutorial in biostatistics propensity score methods for bias

reduction in the comparison of a treatment to a non-randomized control group. Stat Med Stat Med. 1998;17:2265-2281.

https://doi.org/10.1002/(SICI)1097-0258(19981015)17:19<2265::AID-SIM918>3.0.CO;2-B

11. Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: A Monte Carlo study. Stat Med. 2007;26(4):734-753. https://doi.org/10.1002/sim.2580

12. Hartnett ME, Tinkham N, Paynter L, Geisen P, Koch G, Cohen KL. A re-view of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates com-pared with conventional multivariable methods. J Clin Epidemiol. 2006;59(5):437-447. https://doi.org/10.1016/j.ajo.2009.07.014

13. Russo GI, Kurbatov D, Sansalone S, Lepetukhin A, Dubsky S, Sitkin I, Salamone C, Fiorino L, Rozhivanov R, Cimino S, Morgia G. Prostatic ar-terial embolization vs open prostatectomy: a 1-year matched-pair analysis of functional outcomes and morbidities. Urology. 2015;86(2):343-348.

https://doi.org/10.1016/j.urology.2015.04.03714. Caus T, Sirota D, Nader J, Lyashenko M, Chernyavsky A. Associated bare

stenting of distal aorta with a Djumbodis system versus conventional surgery in type A aortic dissection. Ann Cardiothorac Surg. 2016;5(4):336-345.

15. Ayling OGS, Ibrahim GM, Drake B, Torner JC, Macdonald RL. Operative complications and differences in outcome after clipping and coiling of rup-tured intracranial aneurysms. J Neurosurg. 2015;123(3):621-628.

https://doi.org/10.3171/2014.11.JNS14160716. Byrne JP, Mason SA, Gomez D, Hoeft C, Subacius H, Xiong W, Neal M,

Pirouzmand F, Nathens AB. Timing of pharmacologic venous thromboem-bolism prophylaxis in severe traumatic brain injury: a propensity-matched ohort study. J Am Coll Surg. 2016;223(4):621-631.e5.

https://doi.org/10.1016/j.jamcollsurg.2016.06.38217. Ferraris VA, Harris JW, Martin JT, Saha SP, Endean ED. Impact of resi-

dents on surgical outcomes in high-complexity procedures. J Am Coll Surg. 2016;222(4):545-555. https://doi.org/10.1016/j.jamcollsurg.2015.12.056

18. Kato S, Chikuda H, Ohya J, Oichi T, Matsui H, Fushimi K, Takeshita K, Tanaka S, Yasunaga H. Risk of infectious complications associated with blood transfusion in elective spinal surgery — a propensity score matched analysis. Spine J. 2016;16(1):55-60. https://doi.org/10.1016/j.spinee.2015.10.014

19. McCutcheon BA, Chang DC, Marcus L, Gonda DD, Noorbakhsh A, Chen CC, Talamini MA, Carter BS. Treatment biases in traumatic neuro-surgical care: a retrospective study of the nationwide inpatient sample from 1998 to 2009. J Neurosurg. 2015;123(August):1-9.

https://doi.org/10.3171/2015.3.JNS131356


ORIGINAL ARTICLES

One of the most important and nontrivial objectives of clinical neurosurgical researches is adequate selection of pa-tients in order to analyze relationships between surgical out-comes and certain variables. Because of ethical and medical limitations groups analyzed (in observational studies) are not comparable by many factors influencing outcomes as a rule. Main and control groups formation through routine accumula-tion of clinical data leads to systematic errors including so-called "selection bias". As a result, these groups turn out to be disparate by basic characteristics which affect the outcomes. Therefore, conclusions about some effects and relationships may be incorrect. For example, better results of minimally inva-sive approaches in neurosurgery may be associated with im-proved condition of patients who undergo these procedures rather "minimal surgical invasiveness" alone.

Maximally equivalent groups by all apparent and unknown characteristics except that under analysis are necessary to assess adequately the effect of certain factor. In prospective studies it is achieved via randomization. There are certain methods to re-duce selection bias in retrospective trials. This article is devoted to one of these methods — Propensity Score Matching (PSM).

The authors have demonstrated an importance of PSM in neurosurgical researches. The article is of interest for certain specialists because mathematical technologies followed by re-duced selection bias and substantially increased evidence level are described. These methods are very useful to ensure high quality of researches within evidence-based medicine precisely in neurosurgical studies. This is due to impossible randomiza-tion as a rule for medical and ethical reasons.

Retrospective analysis is the most common in neurosurgi-cal researches. It seems the simplest due to routine accumula-tion of data and no need for special selection protocol. Howev-er, retrospective analysis is much more difficult due to insuffi-cient data quality control and systematic selection bias. The authors concluded very important statement about need for correct retrospective analysis with involvement of specialists for biostatistics who understands clearly methodology of clinical researches, control of clinical data and systematic errors. This is one of the cornerstones of ensuring high-quality researches within evidence-based medicine.

G.V. Danilov (Moscow, Russia)

Commentary

20. Miyata K, Ohnishi H, Maekawa K, Mikami T, Akiyama Y, Iihoshi S, Wani-buchi M, Mikuni N, Uemura S, Tanno K, Narimatsu E, Asai Y. Therapeu-tic temperature modulation in severe or moderate traumatic brain injury: a propensity score analysis of data from the Nationwide Japan Neurotrauma Data Bank. J Neurosurg. 2016;124(2):527-537.

https://doi.org/10.3171/2015.3.JNS14189521. Quigley MR, Bello N, Jho D, Fuhrer R, Karlovits S, Buchinsky FJ. Esti-

mating the additive benefit of surgical excision to stereotactic radiosurgery in the management of metastatic brain disease. Neurosurgery. 2015;76(6):707-712. https://doi.org/10.1227/NEU.0000000000000707

22. Salottolo K, Carrick M, Levy AS, Morgan BC, Mains CW, Slone DS, Bar-Or D. Aggressive operative neurosurgical management in patients with ex-tra-axial mass lesion and Glasgow Coma Scale of 3 is associated with sur-vival benefit: A propensity matched analysis. Injury. 2016;47(1):70-76.

https://doi.org/10.1016/j.injury.2015.10.00223. Seicean A, Alan N, Seicean S, Neuhauser D, Selman WR, Bambakidis NC.

Risks associated with preoperative anemia and perioperative blood transfu-sion in open surgery for intracranial aneurysms. J Neurosurg. 2015;123(1):91-100. https://doi.org/10.3171/2014.10.JNS14551

24. Shofty B, Neuberger A, Naffaa ME, Binawi T, Babitch T, Rappaport ZH, Sviri G, Paul M. Intrathecal or intraventricular therapy for post-neurosur-gical Gram-negative meningitis: Matched cohort study. Clin Microbiol In-fect. 2016;22(1):66-70. https://doi.org/10.1016/j.cmi.2015.09.023

25. Zangbar B, Pandit V, Rhee P, Khalil M, Kulvatunyou N, O’Keeffe T, Tang A, Gries L, Green DJ, Friese RS, Joseph B. Clinical outcomes in pa-tients on preinjury ibuprofen with traumatic brain injury. Am J Surg. 2015;209(6):921-926. https://doi.org/10.1016/j.amjsurg.2014.05.027

26. Zheng J, Li H, Zhao H-X, Guo R, Lin S, Dong W, Ma L, Fang Y, Tian M, Liu M, You C. Surgery for patients with spontaneous deep supratentorial intracerebral hemorrhage: a retrospective case-control study using propen-sity score matching. Medicine (Baltimore). 2016;95(11):e3024.

https://doi.org/10.1097/MD.000000000000302427. Martyn D, Kocharian R, Lim S, Meckley LM, Miyasato G, Prifti K,

Rao Y, Riebman JB, Scaife JG, Soneji Y, Corral M. Reduction in hospital costs and resource consumption associated with the use of advanced topical hemostats during inpatient procedures. J Med Econ. 2015;18(6):474-481.

https://doi.org/10.3111/13696998.2015.101750328. Alan N, Seicean A, Seicean S, Selman WR, Bambakidis NC. Impact of age

on 30-day postoperative outcome of surgery for ruptured and unruptured intracranial aneurysms. J Neurointerv Surg. 2014;1-7.

https://doi.org/10.1136/neurintsurg-2014-01113529. Chou C-L, Lee W-R, Yeh C-C, Shih C-C, Chen T-L, Liao C-C. Adverse

outcomes after major surgery in patients with pressure ulcer: a nationwide population-based retrospective cohort study. PLoS One. 2015;10(5): e0127731. https://doi.org/10.1371/journal.pone.0127731

30. Yasunaga H. Effect of Japanese herbal kampo medicine goreisan on reop-eration rates after burr-hole surgery for chronic subdural hematoma: analy-sis of a national inpatient database. Evidence-based Complement Altern Med. 2015;2015. https://doi.org/10.1155/2015/817616

31. Seicean A, Seicean S, Neuhauser D, Benzel EC, Weil RJ. The Influence of Race on Short-term outcomes after laminectomy and/or fusion spine sur-gery. Spine (Phila Pa 1976). 2017;42(1):34-41.

https://doi.org/10.1097/BRS.0000000000001657

Received: 05.04.17


*e-mail: [email protected] © Group of authors, 2018


Advisability and Effectiveness of Decompressive Craniectomy in Patients With Subarachnoid Hemorrhage After Microsurgical Repair of AneurysmsYU.V. PILIPENKO*, AN.N. KONOVALOV, SH.SH. ELIAVA, O.B. BELOUSOVA, D.N. OKISHEV, I.A. SAZONOV, T.F. TABASARANSKIY

Burdenko Neurosurgery Institute, str. 4-ya Tverskaya-Yamskaya, 16, Moscow, Russia, 125047

In recent years, the so-called primary or preventive decompressive craniectomy (DC) has been increasingly used in patients with aneurysmal subarachnoid hemorrhage (SAH). The main goal of the technique is prevention of refractory intracranial hypertension (ICH) and its consequences.Purpose. The study purpose was to define the CT criteria for reasonability and efficacy of DC as well as clarification of the indications for preventive DC in patients with SAH after microsurgical aneurysm exclusion.Material and methods. The study included 46 patients who underwent microsurgical clipping of aneurysms and DC in the period between 2010 and 2016. All patients underwent surgery in the period of 1 to 12 days after SAH. Preventive DC (imultaneously with clipping of aneurysms) was performed in 38 patients. Secondary (delayed) DC was performed in 8 patients.Results. Mortality in a group of all patients with DC was 15.2%. Preventive DC was considered as “reasonable” when the patient had signs of cerebral edema in the postoperative period. The X-ray criteria of reasonable DC included a more than 5 mm brain prolapse into the trephination defect or a lateral dislocation of more than 5 mm. If the patient had no prolapse and dislocation in the postoperative period, DC was considered «unreasonable». Among patients with ICH in the postoperative period, including 20 patients with reasonable preventive DC and 8 patients with delayed DC, mortality was 25%. The CT signs of efficient DC were found to be a more than 5 mm brain prolapse into the trephination defect in combination with a decrease in the lateral dislocation less than 5 mm. All seven patients with inefficient DC in our group died. To clarify the indications for preventive DC, we analyzed various preoperative factors in patients with reasonable and unreasonable DC.Conclusion. In most cases, preventive DC in microsurgical aneurysm exclusion is indicated for patients in an extremely grave condition (Hunt—Hess Grade V), a lateral displacement of the mline structures of more than 5 mm, an intracranial hematoma of over 30 mL, and symptoms of acute cerebral ischemia (pronounced cerebral vasospasm and emerging ischemic foci).

Keywords: decompressive craniectomy, craniectomy, acute period of hemorrhage, cerebral aneurysms, vasospasm.

Material and methodsThere were 46 patients who underwent microsurgical

clipping of aneurysms and DC in 2010—2016. 23 (50%) patients were operated within 72 hours after subarach-noid hemorrhage, 17 patients – in 4—7 days after hemor-rhage, 6 patients – in 7—12 days after hemorrhage. In all cases aneurysms were placed in anterior parts of Willis circle.

DC was performed in standard fashion [4]. Skin inci-sions are shown in Fig. 1.

Bone borders of DC were: 1) anteriorly — segment between zygomatic bone and middle of supraorbital arch; 2) superiorly — line 2 cm lateral to sagittal suture (mid-line); 3) posteriorly — line 3—4 cm behind external audi-tory canal, lambdoid suture in lower parts; 4) inferior-ly — line parallel to zygomatic arch (Fig. 2).

Dura mater repair with periosteum flap or artificial patch was made in all cases to enlarge subdural cavity.

Preventive DC (simultaneously with aneurysms clip-ping) was performed in 38 patients.

Indications for preventive DC were:— severe intraoperative cerebral edema in neurosur-

geon’s subjective assessment (n=7);— Hunt-Hess grade V (n=11);

Some reports demonstrated effectiveness of decom-pressive craniectomy (DC) in patients with aneurysmal intracranial hemorrhage followed by intracranial hyper-tension [1—3].

In the previous publication we have discussed indica-tions for DC in aneurysmal subarachnoid hemorrhage by using of current reports [4]. It is generally accepted that main indication for DC is refractory to therapy augmen-tation of intracranial pressure (ICP) over 20 mm Hg.

According to literature data [4], so-called primary or preventive DC has been often used by many clinicians in recent years to prevent refractory intracranial hyperten-sion (ICH) and its consequences. As a rule, such proce-dures are performed in patients with high risk of advanced ICH.

At present time some signs have been identified to predict postoperative ICH [5, 6]. Nevertheless, severe ce-rebral edema is not observed in some patients who un-dergo preventive DC. So, this procedure may be consid-ered unjustified in these situations. On the other hand, severe ICH followed by death is possible even after DC in some cases, that indicates inefficient surgery.

The aim of the work is to determine CT-criteria of advisability and effectiveness of DC, to clarify the indica-tions for preventive DC in patients with subarachnoid hemorrhage after microsurgical repair of aneurysms.


ORIGINAL ARTICLES

— several risk factors of ICH (n=20): Hunt-Hess grade III-IV, hemorrhage Fisher grade 3—4, intraopera-tive complications (blood loss over 1 l, prolonged forced blood flow cessation in major artery due to ruptured an-eurysm (over 10 min), ultrasonic signs of preoperative angiospasm (Vsys over 240 cm/s in M1-segment of mid-dle cerebral artery) (Fig. 3).

These factors were chosen as indications for preven-tive DC in view of results of I.A. Sazonov et al. [7]. Their research was carried out at the Burdenko Research Insti-tute for Neurosurgery in 2006.

Secondary (delayed) DC was performed in 8 pa-tients.

Indications for delayed DC were:— postoperative hematoma in CT-scans followed by

advanced cerebral dislocation syndrome and impaired consciousness (n=3);

— impaired consciousness combined with severe (over 20 mm Hg) ICH resistant to therapy due to CT-confirmed cerebral ischemia (n=5). Preoperative charac-teristics of patients with preventive and delayed craniec-tomies are given in Table 1.

Previous reports justified DC by its effectiveness. The last was determined by overall outcomes of aneu-rysms repair and their comparison with those without craniectomy in similar groups [1—3].

We suggested postoperative CT-criteria to evaluate advisability of DC. CT-data were processed in Inobitec DICOM Viewer program. We used following concepts:

1) lateral dislocation — transverse displacement mag-nitude of median cerebral structures that was measured at the level of septum pellucidum;

2) prolapse – severity of cerebral protrusion through cranial defect. Prolapse grade was measured along the line between outer cerebral boundaries at the level of nu-clei caudate heads perpendicular to midline. Caudate nu-clei heads were chosen due to the fact that the line through these anatomical formations usually corresponds to center of cranial defect circumference. The greatest ce-rebral prolapse is noted here (Fig. 4). Such measurement we used in 42 patients with unilateral DC.

The second, more difficult calculation of prolapse is software fusion of pre- and postoperative scans (Fig. 5). This approach was applied in 4 cases of bilateral craniec-tomy.

Lateral dislocation and prolapse were estimated by using of CT in 2—6 days after surgery. In cases of redo CT-imaging maximum values within this this period were taken into account. Prolapse and lateral dislocation were measured in millimeters. Postoperative 30-day neu-rological outcomes were estimated according to Glasgow coma scale (GCS).

Results

Preventive DC

Cerebral prolapse through cranial defect and lateral dislocation were observed in 25 (65.8%) and 14 (36.8%) patients with preventive DC, respectively.

Outcomes of patients with preventive craniectomy correlated with prolapse and lateral dislocation severity.

1. Patients with prolapse and lateral dislocation less than 5 mm.

Prolapse and lateral dislocation less than 5 mm were observed in 18 patients. In 10 of them prolapse was ab-sent, moreover cerebral compression within cranial de-fect occurred due to edema and hematomas in soft tis-sues. 8 patients had prolapse within 1—4 mm (mean 3 mm). In most of patients lateral dislocation (n=13) was absent or did not exceed 5 mm (n=5).

In our opinion, both these factors indicated absent severe cerebral edema in postoperative period. Accord-ingly, DC was unjustified in these patients since the goal was not realized (empty space filling with edematous

Fig. 2. Bone boundaries of craniectomy.

Fig. 1. Skin incisions in preventive (a) and delayed (b) craniectomy.


brain). An example of unjustified craniectomy is shown in Fig. 6.

Other instrumental and clinical data have also con-firmed no severe cerebral edema in these patients. In par-ticular, ICP was invasively monitored in 8 patients. There were short-term postoperative elevations of ICP over 20 mm Hg in 4 cases followed by successful medication. ICP was normal in other cases.

In 30 days after subarachnoid hemorrhage good or satisfactory outcomes were observed in this group: GCS 5 in 5 cases, 4 in 6 and 3 in 7 patients. Persistent vegetative status and lethal outcomes were absent.

Severe postoperative neurological deficit was associ-ated with ischemic cerebral accidents as a rule. According to postoperative CT, ischemic foci within the same cere-bral lobe were revealed in 6 patients.

2. Patients with prolapse over 5 mm and lateral disloca-tion less than 5 mm.

Cerebral prolapse within 6—22 mm (mean 11.4 mm) was noted in 17 patients. Cerebral edema predicted prior to surgery was confirmed in postoperative period, that indicates advisability of DC in these cases. Lateral dislo-cation was absent in 11 patients and did not exceed 5 mm in 6 patients.

Invasive postoperative monitoring of ICP was per-formed in 13 out of 17 patients of this subgroup. ICP within 20 mm Hg has been observed in 9 patients for 7 days after DC. There were single short-term elevations of ICP above this threshold value in 4 patients.

Among 17 patients, neurological outcomes GCS score 5 were in 1 case, score 4 in 4 cases, score 3 in 8 cases, score 2 in 2 cases and score 1 in 2 cases. In 1 woman veg-etative state in 30 days after subarachnoid hemorrhage was caused by multiple foci of ischemic circulatory disor-ders in both hemispheres, in other one — severe cerebral lesion due to 5 recurrent hemorrhages with parenchymal component prior to surgery. Two patients died from in-fectious complications: purulent meningoencephalitis in one patient and pneumonia followed by sepsis in another one.

Thus, DC was effective in this subgroup in order to prevent ICH and dislocation syndromes. An example of advisable and effective craniectomy is presented in Fig. 7.

3. Patients with prolapse and lateral dislocation over 5 mm.

Severe postoperative cerebral edema was observed in 3 patients that justified preventive DC. In 1 case of ad-vanced hemispheric ischemia contralateral to craniecto-my cerebral protrusion was 19 mm, in another patient with ipsilateral hemispheric ischemia — 13.5 mm. Pro-lapse near 16 mm was noted in patient with edema due to postoperative hematomas.

Despite DC persistent lateral dislocation 7—19 mm (mean 12.7 mm) was observed in these patients, and am-bient cistern was compressed. In 2 patients with advanced

Fig. 3. Combinations of ICH risk factors used to determine the indications for preventive craniectomy.

Table 1. Preoperative patients’ characteristics

Parameter Preventive DC Delayed DCNumber of patients 38 8Mean age, years 46.9 (20—69) 45.7 (25—72)Male/Female 17:21 4:4

Hunt—Hess gradeII, n (%) 5 (13.2) 1 (12.5)III, n (%) 12 (31.6) 2 (25)IV, n (%) 10 (26.3) 5 (62.5)V, n (%) 11 (28.9) —

SAH Fisher grade2, n (%) 1 (2,6) 1 (12.5)3, n (%) 14 (36,8) 1 (12.5)4, n (%) 23 (60,5) 6 (75)

Severity of preoperative vasospasm (Vsyst М1 segment of MCA) [8]

Mild + moderate (120—230 сm/s), n (%) 28 (73.7) 7 (87.5)Severe (240—290 сm/s), n (%) 9 (23.7) 1 (12.5)Critical (over 300 сm/s), n (%) 1 (2.6) 0

Localization of ruptured aneurysmsACA, n (%) 13 (45.4) 5 (62.5)MCA, n (%) 16 (39.3) 1 (12.5)ICA, n (%) 9 (15.1) 2 (25)

Time from SAH to craniectomy0—3 days, n (%) 23 (60.5) 1 (12.5)4—7 days, n (%) 11 (28.9) 5 (62.5)8—14 days, n (%) 4 (10.5) 2 (25)

Footnote. ICA — internal carotid artery.


ORIGINAL ARTICLES

ischemia persistent elevations of ICP up to 45 and 70 mm Hg were observed. There was no direct measurement of ICP in patient with intracerebral hematomas. All 3 pa-tients died due to severe ICH and dislocation syndrome that indicates ineffectiveness of DC. An example of justi-fied but inefficient craniectomy is shown in Fig. 8.

Thus, there were 3 types of craniectomy in this group:1) unjustified;2) justified and effective;3) justified, but ineffective.Criteria of advisability and effectiveness of DC are

presented in Table 2.

Fig. 4. Calculation of brain protrusion through trepanation defect by using of axial CT-scans. a — AC — line through the heads (H) of caudate nuclei. A and C — extreme points corresponding to outer boundaries of the brain; B – point of AC and midline intersection. Protrusion = AB-BC (for the right side) or BC-AB (for the left side); b — an example of calculation of brain prolapse through trepanation defect: no lateral dislocation, pronounced prolapse through trephination defect on the left (74.8—57.8 = 17 mm).

Fig. 5. Calculation of brain prolapse by using of fusion of pre- and postoperative CT-scans. a — axial CT-scan after bilateral craniectomy; b — the same image after fusion with preoperative CT-scan at the same level (Inobitec program, bone mode): measurements of brain protrusion from internal bone lamina (preoperative CT-scan) to external point of brain prolapse (postoperative CT-scan). Prolapse to the right 11.6 mm, on the left — 11.3 mm, right-sided lateral dislocation — 10.7 mm.


Preoperative factors associated with justified DC

Various preoperative factors in groups of justified and unjustifiable craniectomy were analyzed to clarify in-dications for preventive DC (Table 3). Differences were statistically assessed with non-parametric criteria: Mann-Whitney U-test, Fisher`s exact test.

Following reliable factors of justified DC were estab-lished: Hunt-Hess grade V, increased blood flow velocity over 240 cm/s, dislocation over 5 mm and intracerebral hematoma over 30 mL.

Thus, advisability of preventive DC was confirmed in 10 (90.1%) out of 11 patients with Hunt-Hess grade V (Table 3), that is maximum compared with less severe conditions (Fisher`s exact test, p=0.003). At the same time, procedure was effective in 9 patients who had no clinical signs of brain strangulation intraoperatively. It is noteworthy that all patients with Hunt-Hess grade V un-derwent surgery within 1 day after aneurysm rupture.

There were 80% of patients with blood flow velocity over 240 cm/s and justifiable DC (Table 3) compared with 42.9% of justified procedures at velocity ≤240 cm/s (Fisher's exact test; p=0.007). It should be also noted that 2 patients underwent surgery on background of progress-ing ischemic disorders. Preventive craniectomy was justi-fied in both cases.

Intracerebral hematoma was preoperatively diag-nosed in 19 patients with preventive craniectomy. There-fore, DC was advisable in 73.7% of cases (Fisher`s exact test; p=0.02). At the same time, number of justifiable craniectomies became maximum (86%) if the volume of hematoma was over 30 mL (Table 3). Hematoma less than 30 mL was associated with advisable craniectomy in

33% of cases (Fisher`s exact test, p=0.002). Localization of intracerebral hematoma and advisability of craniecto-my were not significantly correlated (Table 3).

Lateral dislocation was able to be preoperatively as-sessed by CT-imaging in 32 patients (Table 3). DC was justified in all patients with dislocation over 5 mm (n=7). The same procedure was advisable only in 36% of cases among 25 patients with no dislocation or its value less than 5 mm (Fisher`s exact test; p=0.00003).

Delayed DC

Delayed craniectomy was performed in 8 patients.Delayed craniectomy is not necessary to be discussed

since it was performed for severe ICH and dislocation in all cases. Therefore, we can only determine its effective-ness.

In 3 cases postoperative intracerebral hematomas (30—35 mL) occurred. They were diagnosed and re-paired in early postoperative period after aneurysms clip-ping. Delayed craniectomy was effective in this case (ce-rebral protrusion 6—16 mm, mean 11 mm). Lateral dis-location was absent in one of these patients and did not exceed 5 mm in 2 of them. Satisfactory 30-day outcome after subarachnoid hemorrhage was noted in 2 patients (GCS score 4—5), one had disabling symptoms (GCS score 3).

Five patients with ischemic disturbances underwent craniectomy within 1—6 days (mean 3 days) after prima-ry surgery with following results: 2 patients – GCS 3, 1 pa-tient – GCS 2, 2 patients – GCS 1. In 1 patient vegetative state was caused multiple ischemic foci in both brain hemispheres. 2 patients died from advanced ischemic le-

Fig. 6. Unjustified craniectomy. There are no lateral dislocation and brain protrusion on the left.

Fig. 7. Justified effective craniectomy. Pronounced brain prolapse through trephination defect on the left (10 mm), no lateral dislocation.


ORIGINAL ARTICLES

sion contralateral to primary trepanation. Delayed crani-ectomies were ineffective.

It should be noted that effective delayed DC for isch-emia (n=3) was associated with cerebral protrusion 6—9 mm (mean 7 mm) and dislocation ≤4 mm. In 2 patients with prolapse 12 and 10 mm and dislocation 6 and 8 mm respectively delayed procedure was ineffec-tive.

Effectiveness of craniectomy depending on cause of ICH

There was no severe cerebral edema in patients who underwent unreasonable preventive DC. Therefore, causes of ICH were analyzed only in group of justified preventive (n=20) and delayed procedure (n=8).

Edema was associated with intracerebral hematoma in 15 patients, cerebral ischemia in 11 and other disorders (liquorodynamic) in 2 patients (Table 4).

DC was more effective (in 14 out of 15 patients) in case of intracerebral hematoma (Table 4). One woman with hematoma and effective preventive DC died due to infectious complications. In one dead patient with inef-fective preventive craniectomy ICH was caused by 2 he-matomas. One of them was removed simultaneously with clipping of aneurysm, the other (imbibition, volume of about 30 mL) occurred in temporal lobe near posterior

margin of trepanation. DC with anteroposterior dimen-sion 9.5 cm did not ensure elimination of advanced dislo-cation syndrome (Fig. 9). There was acute aggravation of patient’s state in 2 days after primary surgery (up to GCS 3—4). Redo surgery consisted of excision of edem-atous cerebral tissue and enlargement of craniectomy. However, intervention was ineffective.

Effective craniectomy was noted in 7 (63.6%) out of 11 patients with ICH and cerebral ischemia. Four pa-tients died from progressive ischemic cerebral edema de-spite craniectomy. There were preventive decompressive craniectomies 9.4 and 12 cm respectively in 2 of them. 2 patients had bilateral delayed craniectomies (11 and 10.5 cm in one patient and 10.8 and 9.7 cm in the other one). In 3 out of 4 cases of ineffective craniectomy ad-vanced ischemia occurred in hemisphere contralateral to primary trepanation (Fig. 10).

Efficacy of DC was separately assessed in patients with cerebral ischemia and ICH depending on terms of surgery.

Six patients with ischemia underwent simultaneous clipping of aneurysm and DC (Table 4). Two patients died, vegetative status was observed after 30 days in 1 pa-tient. Severe disability (GCS 2—3) and death were caused by ischemia of several lobes of the brain. Ischemic disor-ders of 1 lobe were diagnosed in 2 patients with relatively favorable outcomes (GCS 4 and 5).

Five patients with ischemia of several lobes of the brain had delayed craniectomies. There were similar out-comes in this subgroup: 2 patients died, vegetative state was in 1 patient.

Impaired liquor flow due to blood tamponade of arachnoid cisterns and cerebral ventricles can also cause ICH besides ischemia and intracerebral hematomas. It is difficult to analyze effectiveness of DC for this type of ICH in our study, since only 2 patients with preventive craniectomy were referred to this group. In both cases X-ray criteria confirmed advisability and efficacy of crani-ectomy. One of these patients died from infectious com-plications.

Postoperative mortality rate among 28 ICH patients was 25% (n=7). Mortality among all patients with DC was 15.2% (7 out of 46 cases).

DiscussionIn order to determine X-ray criteria of advisability of

DC we used previous data [9] and our own ideas that ce-rebral edema is followed by brain protrusion through trephination defect. At the same time, in our opinion ef-

Fig. 8. Justified ineffective craniectomy. Large ischemic focus in right hemisphere followed by brain prolapse through defect on the right for about 15.7 mm and left-sided dislocation by 7.1 mm.

Table 2. Criteria of advisability and effectiveness of decompressive craniectomy

Postoperative type of DC Prolapse (>5 mm)

Lateral dislocation (>5 mm)

Refractory ICH (>20 mm Hg)

Impaired consciousness up to coma (GCS<9)

Unjustified No No No NoJustified, effective Yes No No No/YesJustified, ineffective No/Yes Yes Yes Yes


fective craniectomy means increased brain prolapse pro-portional to decreased dislocation that is followed by re-duced ICH and risk of severe neurological complications. Craniectomy should be considered ineffective in severe dislocation followed by strangulation-associated death. Absent dislocation and prolapse may be considered as signs of unjustified procedure since there was no cerebral edema or it was mild.

Data confirmed that DC is advisable and effective within the first days after subarachnoid hemorrhage when signs of ICH are intraoperatively identified.

Our study also showed that preoperative risk factors of advanced postoperative cerebral edema are patient's condition Hunt-Hess grade V (early after hemorrhage), intracerebral hematoma over 30 mL and dislocation over 5 mm. Validity of these factors as indications for preven-tive DC is also confirmed by literature data [3, 7, 9].

It is difficult to predict refractory ICH associated with cerebral ischemia and to determine indications for preventive DC in such patients.

According to our data, indication for preventive cra-niectomy in patients with aneurysmal subarachnoid hemorrhage may be severe vasospasm (over 240 cm/s) or signs of acute cerebral ischemia.

Obviously, DC is indicated in patients with clinical and/or X-ray signs of brainstem strangulation (unilateral mydriasis, severe compression of ambient cistern in CT-data). However, surgery is ineffective in these patients as a rule with unfavorable outcome in most of them [10, 11].

According to the literature [12], ineffective craniec-tomy may be caused by insufficient dimension of defect and absence or small size of dura mater repair. In our opinion, cerebral protrusion may be also discouraged by hematomas between dura mater and skin-aponeurotic flap. It is also clear that DC cannot always prevent brain-stem compression in severe ischemic lesion, large hema-tomas and cerebral ventricles tamponade with blood.

Some reports [9, 13] consider preventive craniectomy more appropriate over delayed procedure since outcomes in these patients are retrospectively better. At the same time, we did not find any literature data about inexpedi-ency of this procedure in patients with aneurysmal sub-arachnoid hemorrhage. Perhaps, it is partly related to incorrect analysis of the outcomes. For example, in our data overall mortality in patients with DC was 15.2%, in patients with reliable ICH (preventive justified and de-layed procedures) — 25%.

It is reasonably to ask what is the cause of unjustified preventive craniectomies in our group? It is more likely that our own indications developed in our center were not sensitive enough to determine the risk of advanced ICH. However, it is worth to note that previous trial [7] devoted to DC for aneurysmal subarachnoid hemorrhage was based on the analysis of patients for the period 1995—2005. At the same time, ICP sensors were not rou-tinely used and CT within intensive care unit was un-available. Accordingly, decision about emergency proce-

Tabl

e 3.

Pre

oper

ativ

e fa

ctor

s of j

ustif

ied

and

unju

stifi

ed p

reve

ntiv

e cr

anie

ctom

y

Varia

ble

Age,

ye

ars

Tim

e fro

m S

AH,

days

Rec

urre

nt

SAH

Hun

t—H

ess g

rade

Velo

city

ove

r 24

0 сm

/sLo

caliz

atio

n of

an

eury

sms

Intra

cere

bral

he

mat

oma,

ml

Loca

lizat

ion

of IH

(lo

be)

Late

ral d

isloc

atio

n, m

m

40>4

0<3

4 ‒

77‒

14N

oYe

sII

III

IVV

No

Yes

ACA

ICA

MC

A30

>30

No

IHFr

onta

lTe

mpo

ral

0 5

> 5

No

data

Unj

ustif

ied

cra-

niec

tom

y (1

8 pa

-tie

nts)

612

96

315

34

67

116

26

48

32

132

311

50

2Ju

stifie

d cr

anie

c-to

my

(20

pa-

tient

s)6

1414

51

128

16

310

128

73

102

126

59

72

74

Betw

een-

grou

p di

ffere

nces

NS

NS

NS

Sign

ifica

nt (u

-tes

t; p=

0,02

)

Sign

ifica

nt

(u-t

est;

p=0,

002)

NS

Sign

ifica

nt (u

-tes

t; p=

0,01

).N

SSi

gnifi

cant

(u-t

est;

p=0,

01)


ORIGINAL ARTICLES

dure has been accepted later. Thus, indications for DC were extended in order to prevent mortality from ad-vanced ICH.

It is difficult to determine indications for preventive craniectomy because even preoperative ICH is not oblig-atory followed by postoperative hypertension. This is due to the fact that removal of hematoma, elimination of blood clots from basal cisterns, external ventricular drain-age may be effective for ICH management. We also admit that in patients with advisable preventive craniectomy ac-cording to our criteria ICH may be prevented and/or stopped by other measures (elevated head position, os-motherapy, deep sedation, hypothermia, etc.).

Following data are not in favor of preventive DC in patients without ICH and acute cerebral ischemia:

— advanced ischemic areas in hemisphere contralat-eral to craniectomy that makes this procedure ineffective;

— similar outcomes in patients with advanced isch-emic cerebral disorders who underwent preventive and delayed decompressive craniectomies.

There are following assumptions besides above-men-tioned ones:

— advanced surgery consisting of craniotomy en-largement is inevitably followed by more severe blood loss, while anemia with vasospasm aggravate ischemic cerebral disorders [14];

— if cranial defect will be not filled with swollen brain within the first few days after craniectomy then it is likely to be that in future blood and liquor between dura mater and skin-aponeurotic tissues will prevent protru-sion of swollen brain.

Potential complications of craniectomy including in-fections, hemorrhagic events, etc. should be considered for DC. Moreover, this procedure is followed by in-creased cost of treatment due to prolonged surgery, need for additional materials for dura mater repair and subse-quent cranioplasty in most of survivors.

Thus, we consider preventive DC unjustified in pa-tients with aneurysmal subarachnoid hemorrhage with-out severe vasospasm, acute ischemic disturbances and signs of hematoma followed by ICH during clipping of aneurysm. Patients with high risk of vasospasm and isch-emia (massive basal hemorrhage, recurrent hemorrhag-es, advanced age, etc.) should be followed-up at ICU af-ter microsurgical repair of aneurysm. Invasive measure-ment of ICP is required in case of impaired conscious-ness or elective sedation. Prevention of severe course of vasospasm is the main goal of treatment in these patients (correction of hypercoagulation, transfusion, hemody-namic therapy regarding central perfusion pressure, sys-temic and selective endovascular administration of cal-cium channel blockers, etc.). Urgent delayed craniecto-my is indicated for the first signs of unresponsive to

Table 4. Efficacy of DC and postoperative outcomes in patients with ICH

Cause of ICH DC type Number of patients DC efficacy

GCS in 30 days after SAHVI—V III—II Death

IH Preventive DC 12 11 2 8 2Delayed DC 3 3 2 1 ‒

Ischemia Preventive DC 6 4 2 2 2Delayed DC 5 3 ‒ 3 2

Other disturbances Preventive DC 2 2 1 ‒ 1Total 28 23 7 14 7

Fig. 9. Ineffective preventive DC in patient with two IHs. CT-scan in 1 day after clipping of left middle cerebral artery aneurysm, left-sided brain prolapse through DC by 4.3 mm and dislocation to the right by 9.4 mm. Explanations in the text.


therapy ICH. Hemisphere with more severe edema or bilateral approach for lesion of both hemispheres should be preferred.

It is worth to note that intraoperative factors also af-fect probability of postoperative ICH.

It is impossible to determine objective intraoperative risk factors of ICH due to their variability. Therefore,

need for simultaneous craniectomy during aneurysm clipping will depend on subjective assessment of situation by neurosurgeon: brain trauma severity, coagulation of small arteries and veins within surgical approach, features of hematoma removal and others.

In our opinion, analysis of correlation between dura-tion of cerebral vessels clipping and hemorrhage severity on the one hand and postoperative ICH on the other hand is perspective.

Concepts of advisability and effectiveness of DC for aneurysmal subarachnoid hemorrhage may be used for further study of ICH and its correction in prospective samples. The main purposes of our further research are to reduce the incidence of unjustified craniectomies, to in-crease effectiveness of advisable procedures and to im-prove overall outcomes in patients with aneurysmal sub-arachnoid hemorrhage.

ConclusionSign of effective craniectomy is brain protrusion into

trephination defect over 5 mm followed by decreased dis-location less than 5 mm and no unresponsive to therapy ICH.

In most of cases preventive DC in aneurysms clip-ping is indicated in extremely ill patient (Hunt-Hess grade V), lateral dislocation over 5 mm, intracerebral he-matoma over 30 mL and signs of acute cerebral ischemia (severe cerebral vasospasm and ischemic foci formation). In other cases, it is advisable to apply intensive therapy aimed at prevention and treatment of vasospasm and de-layed craniectomy for the first signs of refractory ICH.


Fig. 10. Ineffective preventive right-sided DC in patient with advanced zone of left hemispheric ischemia. CT-scan in 3 days after clipping of right internal carotid artery aneurysm, right-sided brain protrusion through DC by 19.6 mm and right-sided dislocation by 19 mm. Explanations in the text.

REFERENCES1. Buschmann U, Yonekawa Y, Fortunati M, Cesnulis E, Keller E. Decom-

pressive hemicraniectomy in patients with subarachnoid hemorrhage and intractable intracranial hypertension. Acta Neurochir (Wien). 2007 Jan;149(1):59-65. Epub 2006 Dec 28. PubMed PMID: 17180307.

https://doi.org/10.1007/s00701-006-1069-x2. Güresir E, Raabe A, Setzer M, Vatter H, Gerlach R, Seifert V, Beck J. De-

compressive hemicraniectomy in subarachnoid haemorrhage: the influence of infarction, haemorrhage and brain swelling. J Neurol Neurosurg Psychia-try. 2009 Jul;80(7):799-801. PubMed PMID: 19531687.

https://doi.org/10.1136/jnnp.2008.1556303. Otani N, Takasato Y, Masaoka H, Hayakawa T, Yoshino Y, Yatsushige H,

Miyawaki H, Sumiyoshi K, Chikashi A, Takeuchi S, Suzuki G. Surgical outcome following decompressive craniectomy for poor-grade aneurysmal subarachnoid hemorrhage in patients with associated massive intracerebral or Sylvian hematomas. Cerebrovasc Dis. 2008;26(6):612-617. Epub 2008 Oct 23. PubMed PMID:18946217. https://doi.org/10.1159/000165115

4. Konovalov AN, Belousova OB, Pilipenko YuV, Eliava ShSh. Dekompres-sivnaya trepanaciya cherepa u bol’nych s vnutricherepnym krovoizliyaniem anevrimaticheskogo geneza. Voprosy neirochirurgii. 2016;5 (80):144-149. (In Russ.). https://doi.org/10.17116/neiro2016805144-150

5. Heuer GG, Smith MJ, Elliott JP, Winn HR, LeRoux PD. Relationship be-tween intracranial pressure and other clinical variables in patients with an-eurysmal subarachnoid hemorrhage. J Neurosurg. 2004 Sep;101(3):408-416. PubMed PMID: 15352597

6. Lv Y, Wang D, Lei J, Tan G. Clinical observation of the time course of raised intracranial pressure after subarachnoid hemorrhage. Neurol Sci. 2015 Jul;36(7):1203-1210. Epub 2015 Jan 22. PubMed PMID: 25604576.

https://doi.org/10.1007/s10072-015-2073-9

7. Sazonov IA, Filatov YuM, Eliava ShSh, Zolotuhin SP, Cheireddin AS, Be-lousova OB, Shekhtman OD, Zarzur H, Sazonova OB, Ogurcova AN, Dausheva AA, Torbotryas N. Naruzhnaya dekompressiya v hirurgii arterial’nyh anevrizm v ostrom periode subarahnoidal’nogo krovoizliyaniya. Mat. IV s»ezda nejrohirurgov Rossii. 2006;286. (In Russ.).

8. Dausheva AA. Subarahnoidal’nye krovoizliyaniya, oslozhnennye arterial’nym spazmom (kliniko-dopplerograficheskoe issledovanie): Dis. … kand. med. nauk. M. 1994;26. (In Russ.).

9. Schirmer CM, Hoit DA, Malek AM. Decompressive hemicraniectomy for the treatment of intractable intracranial hypertension after aneurysmal sub-arachnoid hemorrhage. Stroke. 2007 Mar;38(3):987-992. Epub 2007 Feb 1. PubMed PMID: 17272765.

https://doi.org/10.1161/01.STR.0000257962.58269.e2

10. Hwang US, Shin HS, Lee SH, Koh JS. Decompressive surgery in patients with poor-grade aneurysmal subarachnoid hemorrhage: clipping with si-multaneous decompression versus coil embolization followed by decom-pression. J Cerebrovasc Endovasc Neurosurg. 2014 Sep;16(3):254-261. Epub 2014 Sep 30. PubMed PMID: 25340028; PubMed Central PMCID: PMC4205252. https://doi.org/10.7461/jcen.2014.16.3.254

11. Dorfer C. Frick hemicraniectomy after aneurysmal subarachnoid hemor-rhage. World Neurosurg. 2010 Oct-Nov;74(4-5):465-471. Epub 2011 Jan 12. PubMed PMID: 21492596. https://doi.org/10.1016/j.wneu.2010.08.001

12. Timofeev I, Santarius T, Kolias AG, Hutchinson PJ. Decompressive crani-ectomy ‒ operative technique and perioperative care. Adv Tech Stand Neu-rosurg. 2012;38:115-136. PubMed PMID: 22592414.

https://doi.org/10.1007/978-3-7091-0676-1_6


ORIGINAL ARTICLES

13. Wang HJ, Ye YF, Shen Y, Zhu R, Yao DX, Zhao HY. Surgical treatment of poor grade middle cerebral artery aneurysms associated with large sylvian hematomas following prophylactic hinged craniectomy. J Huazhong Univ Sci Technolog Med Sci. 2014 Oct;34(5):716-721. Epub 2014 Oct 16. PubMed PMID: 25318882. https://doi.org/10.1007/s11596-014-1341-x

14. Le Roux PD. Participants in the international multi-disciplinary consensus conference on the critical care management of subarachnoid hemorrhage. Anemia and transfusion after subarachnoid hemorrhage. Neurocrit Care. 2011 Sep;15(2):342-353. Review. PubMed PMID: 21769459.

https://doi.org/10.1007/s12028-011-9582-zReceived: 12.06.17

Decompressive craniectomy (DC) is one of the most an-cient methods for intracranial hypertension. Currently, DC is one of the last steps to control ICH when all other intensive care and minimally invasive approaches are inefficient to decrease intracranial pressure (ICP) [1]. However, DC is often delayed and does not provide proper clinical effect [2]. DCs are often performed after irreversible changes occurred especially in se-vere patients with aneurysmal SAH with impaired postopera-tive consciousness or need for deep postoperative sedation fol-lowed by difficult permanent assessment of neurological status.

In our data, angiospasm occurred in 94% of patients with advanced SAH including critical or severe circulatory disorders in 64% of them. In these patients preventive DC and early inva-sive ICP monitoring may be followed by for more effective in-tensive care, improved cerebral perfusion pressure and reduced risk of edema-associated brain strangulation. According to our data [3] preventive DC reduces mortality by 32% in patients with high risk of angiospasm and is important for treatment of its complications.

However, DC may be followed by some complications: syndrome of trephined, impaired outflow of cerebrospinal fluid and hydrocephalus, convulsive syndrome, infectious complica-tions, etc. [4]. Early repair of skull defect is useful to prevent some unfavorable events [5].

Authors’ main objective was to clarify indications for pre-ventive DC in patients with SAH after microsurgical clipping of aneurysms and to decrease incidence of unjustified procedures. Criteria of "justified" and "unjustified" preventive DC have been developed.

In authors' opinion, unjustified DC is considered in post-operative brain dislocation less than 5 mm. Choice of this vari-

able as criterion of unjustified DC is not entirely clear, since dislocation may be absent in edema and/or ischemia unilateral to trephination defect. This situation arose in the 2nd group where advanced brain protrusion without dislocation were ob-served in 17 patients. In our opinion, deployment of ICP sensor at preoperative level would be able to assess more objectively ICH and to determine "justified" and "unjustified" DC (only Decompressive craniectomy (DC) is one of the most ancient methods for intracranial hypertension. Currently, DC is one of the last steps to control ICH when all other intensive care and minimally invasive approaches are inefficient to decrease intra-cranial pressure (ICP) [1]. However, DC is often delayed and does not provide proper clinical effect [2]. DCs are often per-formed after irreversible changes occurred especially in severe patients with aneurysmal SAH with impaired postoperative consciousness or need for deep postoperative sedation followed by difficult permanent assessment of neurological status.

In our data, angiospasm occurred in 94% of patients with advanced SAH including critical or severe circulatory disorders in 64% of them. In these patients preventive DC and early inva-sive ICP monitoring may be followed by for more effective in-tensive care, improved cerebral perfusion pressure and reduced risk of edema-associated brain strangulation. According to our data [3] preventive DC reduces mortality by 32% in patients with high risk of angiospasm and is important for treatment of its complications.

However, DC may be followed by some complications: syndrome of trephined, impaired outflow of cerebrospinal fluid and hydrocephalus, convulsive syndrome, infectious complica-tions, etc. [4]. Early repair of skull defect is useful to prevent some unfavorable events [5].

Commentary

Indications for DC in patients with aneurysmal SAH is still unclear problem. Despite this fact there are only few publica-tions in world literature. Outcomes in decompensated patients are still unfavorable. The main pathophysiological features en-countered by neurosurgeon regarding DC are consequences of primary hemorrhage, secondary ischemia due to cerebral vas-cular spasm, cerebral edema and ICH.

In this report authors retrospectively analyzed sufficient number of patients (n=46) in order to develop criteria of advis-ability and effectiveness of DC, to clarify indications for pre-ventive DC in patients with SAH after microsurgical clipping of aneurysms.

Most of patients (n=38) underwent preventive DC that in-dicates their severity and risk of advanced postoperative ICH.

The authors formulated the main criteria for preventive DC in certain cases.

In our practice we also adhere to similar recommenda-tions. It is important that besides ICH control, DC eliminates the need for total sanation of distal Sylvian fissure on back-ground of ruptured aneurysms of middle cerebral artery. The last is followed by reduced trauma of cerebral tissue.

We prefer bifrontal decompression instead of bilateral DC for axial dislocation.

This work is undoubtedly relevant and concretizes con-cepts for further researches.

R.S. Dzhindzhikhadze (Moscow, Russia)


Authors’ main objective was to clarify indications for pre-ventive DC in patients with SAH after microsurgical clipping of aneurysms and to decrease incidence of unjustified procedures. Criteria of "justified" and "unjustified" preventive DC have been developed.

In authors' opinion, unjustified DC is considered in post-operative brain dislocation less than 5 mm. Choice of this vari-able as criterion of unjustified DC is not entirely clear, since dislocation may be absent in edema and/or ischemia unilateral to trephination defect. This situation arose in the 2nd group where advanced brain protrusion without dislocation were ob-served in 17 patients. In our opinion, deployment of ICP sensor at preoperative level would be able to assess more objectively ICH and to determine "justified" and "unjustified" DC (only 21 out of 38 patients underwent invasive ICP monitoring in this trial).

Due to some limitations (retrospective analysis, small sam-ple size) authors were unable to significantly establish the role of some important prognostic factors of unfavorable outcomes including age, terms of surgery, redo ruptured aneurysm, SAH severity [6]. Nevertheless, the question of advisability and effi-cacy of craniectomy is very important. Obviously, surgeon de-termines type of craniectomy in view of both preoperative data and intraoperative subjective considerations in some patients. Objective verification of these subjective factors and prevention of craniectomies without proper causes is the main goal of this work. Undoubtedly, more frequent invasive ICP monitoring would only strengthen evidence base for the study.

The article is a new step towards further research of preven-tive DC in patients with high risk of ICH.

V.G. Dashyan (Moscow, Russia)

REFERENCES

1. Rangel-Castillo L, Gopinath S, Robertson CS. Management of intracranial hypertension. Neurol Clin. 2008;26(2):521-541.

2. Krylov VV, Petrikov SS, Solodov AA, Talypov AE, Puras YuV, Khamido-va LT, Kutrovskaya NYu. Vnutricherepnaya gipertenziya u bol'nykh s vnu-tricherepnymi krovoizliyaniyami. Diagnostika i lechenie. Zhurnal im NV Sklifosovskogo Neotlozhnaya meditsinskaya pomoshch'. 2012;4:44-50. (In Russ.).

3. Dash'yan VG, Levchenko OV, Airapetyan AA, Luk'yanchikov VA, Tok-arev AS, Shatokhin TA, Kalinkin AA, Sharifulin FA, Krylov VV. Dekom-pressivnaya kraniotomiya v khirurgii razorvavshikhsyaanevrizm golovnogo mozga. Rossiiskii neirokhirurgicheskii zhurnal im prof AL Polenova. 2015;4:17-24. (In Russ.).

4. Kurland DB, Khaladj-Ghom A, Stokum JA, Carusillo B, Karimy JK, Gerza-nich V, Sahuquillo J, Simard JM. Complications associated with decompres-sive craniectomy: a systematic review. Neurocrit Care. 2015;23(2):292-304.

5. Halani SH, Chu JK, Malcolm JG, Rindler RS, Allen JW, Grossberg JA, Pradilla G, Ahmad FU. Effects of cranioplasty on cerebral blood flow fol-lowing decompressive craniectomy: a systematic review of the literature. Neurosurgery. 2017;81:204-216.

6. Krylov VV, Dash'yan VG, Shatokhin TA, Sharifullin FA, Solodov AA, Prirodov AV, Levchenko OV, Tokarev AS, Khamidova LT, Kuksova NS, Airapetyan AA, Kalinkin AA. Vybor srokov otkrytogo khirurgicheskogo lecheniya bol'nykh s razryvom tserebral'nykh anevrizm, oslozhnennykh massivnym bazal'nym subarakhnoidal'nym krovoizliyaniem (Fisher 3). Nei-rokhirurgiya. 2015;3:11-20. (In Russ.).


ORIGINAL ARTICLES

Neuroparalytic keratitis (NPK) is caused by dysfunc-tion of trigeminal and facial nerves and includes two clin-ical components: corneal injury — neurotrophic keratitis and paralytic lagophthalmos.

In our practice, NPK is the most common after sur-gery for cochlear nerve schwannoma, cerebellum tumors and strokes in some cases.

Impaired function of trigeminal nerve is associated with aggravation of corneal sensitivity and tissue metabo-lism. There is decreased level of trophic factors and neu-rotransmitters (insulin-like growth factor-1, substance P, nerve growth factor, etc.) which are key to maintain ana-tomical integrity and function of ocular surface [1]. It is accompanied by neurotrophic keratitis with recurrent in-flammation, recurrent or persistent corneal erosions and impaired regeneration [2—4]. Dysfunction of facial nerve additionally leads to lagophthalmos that aggravates the course of neurotrophic keratitis.

NPK is poorly responsive to medication because it is characterized by pronounced tear deficiency, impaired or absent corneal sensitivity. These factors aggravate healing of corneal epithelial defects and lead to "clean" corneal ulcers and melting of corneal stroma. It is especially true in concomitant lagophthalmos.

The situation is complicated by contraindications to medicines stimulating corneal epithelialization (solcos-eryl, actovegin, etc.), vitamins and physiotherapeutic procedures in NPK patients after surgery for brain tu-mors.

All above-mentioned factors increase risk of second-ary infection, progressive suppurative corneal ulcers (SCU), perforation, endophthalmitis followed by loss of the eye [5—7].

Keratoplasty is often preferred by ophthalmologists due to inefficient medication and unfavorable prognosis of corneal ulcers [8].



Surgical Treatment of Suppurative Corneal Ulcers Due to Neuroparalytic KeratitisEvg.A. KASPAROVA*, O.I. SOBKOVA, E.A. KASPAROVA, A.A. KASPAROV

Research Institute of Eye Diseases, str. Rossolimo, 11 A, Moscow, Russia,119021

Objective — the study objective was to develop a method for surgical treatment of advanced purulent corneal ulcer in the setting of neuroparalytic keratitis developed after acoustic neuroma removal and strokes. Material and methods. The study included 12 patients (13 eyes). All patients underwent single-step keratoplasty, conjunctival autografting, and partial permanent tarsorrhaphy. Results. The eye as an organ was spared in 100% of cases; vision was spared in 50% of cases and improved, from 0.09±0.05 to 0.21±0.13, in 29% of cases. Lagophthalmos was reduced from 5.86±1.35 to 3.01±0.75 mm.Conclusion. Patients with intracranial lesions complicated by neuroparalytic keratitis often develop progressive purulent corneal ulcers. The simultaneous use of surgery, including keratoplasty, conjunctival autografting, and partial (external) tarsorrhaphy, is an effective treatment that can preserve the visual function of these eyes.

Keywords: purulent ulcer, cornea, neuroparalytic, keratitis, lagophthalmos, conjunctival autografting, keratoplasty, tarsorrhaphy.

Various methods of lagophthalmos repair are known [9—11]. However, they are insufficient per se for NPK correction. Similarly, corneal repair alone is also ineffective without treatment of lagophthalmos.

We did not find any optimal surgical algorithm for NPK-associated SCUs in available literature.

The aim of our study was to develop surgical treat-ment of NPK-associated SCUs. Keratitis is a result of intracranial pathological processes in these cases.

Material and methodsThere were 12 patients (8 women and 4 men aged

28—65 years, mean 43±5 years, 13 eyes) with NPK-asso-ciated SCUs. Mean anamnesis of SCUs was from 2 weeks to 1.5 months prior to admission to Research Institute of Eye Diseases. Outpatient antibiotic therapy was adminis-tered in all patients without positive effect.

Mean visual acuity was 0.09±0.05. Schirmer’s test on affected eye revealed moderate and severe impairment of tearing in all patients (1—10 mm), on contralateral eye – mild or moderate grade (5—15 mm, norm 15—20 mm). Corneal sensitivity was assessed by using of Radzik-hovsky’s algesimeter with a weight of 10 g, it was absent in all quadrants in all patients. Palpebral fissure width was 8.6±0.16 mm, lagophthalmos — 5.86±1.35 mm (Table).

Staphylococcus aureus, epidermal staphylococcus and pseudomonas aeruginosa were revealed in conjuncti-val cavity in 8, 4 and 1 case, respectively.

Besides lagophthalmos, clinical signs include mild conjunctival hyperemia, copious mucopurulent dis-charge from conjunctival cavity. Central SCUs with ad-vanced corneal melting were predominant. Patients were divided into 2 groups depending on severity of disease: group 1 with moderate severity of SCUs and group 2 with severe grade. In group 1 (4 patients, 4 eyes) SCUs occu-pied ½ of corneal stroma depth, ulcers’ diameter was


5—7 mm. These patients underwent lamellar keratoplas-ty (LK). There were advanced SCUs in the 2nd group (8 patients, 9 eyes): in 5 eyes deep ulcers 5—8 mm in di-ameter occupied over 2/3 of corneal depth; in 4 eyes SCUs with diameter of 7—10 mm extended entire cor-neal layers: descemetocele was in 2 cases and corneal per-foration – in other 2 patients. In this group penetrating keratoplasty (PK) was performed.

In all patients flicker fusion rate was over 25 Hz, la-bility — 20 uA. There were no signs of endophthalmitis according to beta-scanning.

Two aspects we indications for surgery: 1) advanced suppuraive process without positive dynamics despite ad-equate medication; 2) threat of perforation or perforation of cornea.

In view of severe state of patients with NPK-associat-ed SCUs surgical treatment was aimed at corneal sana-tion and correction of neurotrophic disorders.

All patients underwent keratoplasty followed by total auto-conjunctival covering of the graft (auto-conjunctiva was fixed to the limbus by interrupted seams) and partial permanent tarsorrhaphy (patent of the Russian Federa-tion "Method of treatment of neurotrophic keratitis with lesion of central corneal optical zone in lagophthalmos" No. 2299048 dated May 20, 2007).

LK and urgent K were applied in groups 1 and 2, re-spectively. In 5 cases sinus trabeculectomy was performed simultaneously with KP du to decompensated intraocu-lar pressure (IOP). Silicone drainage tube into lower outer quadrant was required in 2 patients with abundant purulent exudation and hypertension ("Method of treat-ment of complicated purulent corneal ulcer" No. 2309710 dated February 21, 2006). Silicone drainage tube was useful already in the first hours after keratoplas-ty for permanent outflow of abundant fibrinous exudate from anterior chamber of the eye and IOP normalization. 2 patients with cataract and no profuse purulent exuda-tion required extracapsular cataract extraction followed by intraocular lens deployment simultaneously with PK.

Active postoperative medication was administered in all patients: frequent or forced instillations of antibiotics and antiseptics (fluconazole and chlorhexidine in sus-pected fungal infection); nonsteroidal anti-inflammatory drugs, preservative-free artificial tears drugs, parabulbar injections of antibiotics, oral intake of antibiotic and an-

tifungal agents. Instillations of antiseptics have been per-formed for a long time (12—15 months), preservative-free artificial tears drugs — for life.

Follow-up period was 1—8 years. Median of catam-nesis — 39.4 months.

ResultsIn 1 eye conjunctival flap adhered to lower third of

cornea. In other cases, partially dislocation downwards with exposure of epithelialized central zone of corneal transplant was observed after 2.7±0.4 weeks.

In the 1st group engraftment was pellucid in all 4 cases without recurrent suppurative keratitis within follow-up (5 years). In the 2nd group PK was followed by pellucid engraftment in 5 eyes and translucent in 4 eyes, mean vi-sual acuity increased from 0.09±0.05 to 0.21±0.13 (in 29% — by 0.55), lagophthalmos decreased from 5.86±1.35 to 3.01±0.75 mm (Table, Figs. 1, 2).

Recurrent SCUs on the transplant (Pseudomonas aeruginosa in 2 cases, staphylococcal infection — 1, in 2 cases causative agent was unknown) occurred in 5 pa-tients (5 eyes) of the 2nd group within 1.5±0.5 years on the average. Therefore, redo keratoplasty was performed. After that muddy engraftment was observed in all pa-tients. In view of severe comorbidities this outcome was considered favorable since eye was preserved as organ. Moreover, there was preserved visual acuity from correct light projection to 0.02.

In 1 patient (1 eye) of the 2nd group "clean" melting of corneal stroma 4×5 mm occurred on penetrating graft. Amnioplasty and biocoverage with sclerocorneal flap was carried out that was followed by stable transplant epithe-lization within further follow-up (6 years).

Histological examination of excised corneal discs re-vealed necrosis and incomplete regeneration that is typi-cal for neurodystrophic keratitis as a symptoms of ad-vanced corneal lesion.

Increased tearing in affected eye occurred in 2 pa-tients after 5 years (Schirmer’s test value increased from 0 to 5 mm in one patient and from 0 to 20 mm in the other). In our opinion, it is due to surgical facial nerve repair in 3 years after keratoplasty. In other cases, tearing was permanently impaired even in long-term period.

Pre- and postoperative data (n=13) (M±m)

Variable Visual acuity Lagophthalmos, mm

Palpebral fissure, mm

Schirmer’s test of affected eye, mm

Schirmer’s test of contralateral eye,

mm

Corneal sensitivity, from 0 to 2 scores

Prior to surgery

0.09±0.05 (correct light projection 0.4) 5.86±1.35 (2—10) 8.6±0.16 (8—12) 5.5±2.86 (1—10) 8.75±2.07 (5—15) 0.57±0.28 (0—2)

1 year after keratoplasty

0.21±0.13 (correct light projection 0.8) 3.01±0.75 (1—5) 5.11±0.84 (3—8) 5.5±2.5 (1—10) 8.9±2.1 (5—15) 0.65±0.3 (0—2)

5 year after keratoplasty

0.21±0.14 (correct light projection 0.7) 3.3±0.7 (1—6) 5.26±1 (3—8) 6.86±2.43 (1—15) 8.8±1.8 (5—15) 0.71±0.3 (0—2)


ORIGINAL ARTICLES

DiscussionKeratoplasty, autoconjunctival repair and partial

permanent tarsorrhaphy are optimal for unresponsive to treatment NPK-associated SCUs.

Palpebral fissure repair and total autoconjunctival coverage of the graft are followed by complete epitheliza-tion of transplant. Trophic support of the graft is achieved through the contact of "denervated" cornea with con-junctival bloodstream, mechanical protection of the graft

from damage and drying, and epithelial cells migration from conjunctival flap to corneal surface. This mecha-nism was experimentally confirmed by V.A. Vasilieva and some foreign authors [11—13]. Autoconjunctival flap as a permanent source of epithelial cells reduces the risk of persistent erosions of the graft. In our opinion, growth of conjunctival vessels and adhesion of autoconjunctival

а

b

c

Fig. 1. Right eye of patient P., 69 years old. State after previous stroke.a — lagophthalmos, suppurative corneal ulcer; b — 1 month after PK, total autoconjunctival coverage, partial external tarsorrhaphy; c — 12 months postoperatively, complete epithelization, pellucid engraftment.

Fig. 2. Right eye of patient R., 35 years old. State after cochlear schwannoma removal.a — lagophthalmos, advanced suppurative corneal ulcer with local thinning; b — 2 weeks after PK, total autoconjunctival coverage, partial external tarsorrhaphy; c — 10 months after PK, complete epithelization, pellucid engraftment.

а

b

c


flap to corneal surface are considered favorable factors due to permanent protection and nutrition of the cornea.

Patients with NPK should be carefully followed-up due to high risk of recurrent purulent infection on the eye with impaired innervation: early detection and treatment of corneal erosions and ulcers and prevention of second-ary infection. Patients with NPK should be informed by ophthalmologists about high risk of complications, need for regular observation and use of local antiseptics and permanent administration of preservative-free tear eye drops.

Keratoplasty, autoconjunctival repair and tarsorrha-phy helps are able to save eye as an organ and to achieve favorable outcomes in most of cases. Good results after LK are probably associated with less severe preoperative corneal lesion. We managed to keep the eye as an organ in all 13 cases, eyesight in 6 patients and improve visual acu-ity in 4 patients.

Conclusions1. Optimal treatment of NPK-associated advanced

suppurative corneal ulcer is corneal transplantation with simultaneous autoconjunctival covering and partial per-manent tarsorrhaphy.

2. There is high risk of recurrent corneal erosion in patients with neuroparalytic keratitis and suppurative corneal ulcers due to permanent decrease of tearing and impaired innervation. Preservative-free tear eye drops are indicated in all patients with neuroparalytic keratitis.

3. Regular lifelong follow-up is required after kerato-plasty and autoconjunctival repair for neurotrophic ul-cerative keratitis and paralytic lagophthalmos followed by corneal ulcers because of high risk of recurrent infection.


REFERENCES1. Mastropasqua L, Massaro-Giordano G, Nubile M, Sacchetti M. Under-

standing the pathogenesis of neurotrophic keratitis: the role of corneal nerves. J Cell Physiol. 2017;232(4):717-724.

https://doi.org/10.1002/jcp.256232. Chikama T, Fukuda K, Morishige N, Nishida T. Treatment of neurotroph-

ic keratopathy with substance-P-derived peptide (FGLM) and insulin growth factor 1. Lancet. 1998;351:1783-1784.

https://doi.org/10.1016/S0140-6736(98)24024-43. Bonini S, Lambiase A, Rama P, Caprioglio G, Aloe L. Topical treatment

with nerve growth factor for neurotrophic keratitis. Ophthalmology. 2000;107:1347-1351, discussion 1351-1352.

https://doi.org/10.1016/S0161-6420(00)00163-94. Lambiase A, Rama P, Bonini S, Caprioglio G, Aloe L. Topical treatment

with nerve growth factor for corneal neurotrophic ulcers. N Engl J Med. 1998;338:1174-1180. https://doi.org/10.1056/NEJM199804233381702

5. Kopaeva VG. Glaznye bolezni. M.: Medicina, 2012. (In Russ.).6. Ryoji Y, Teruo N, Tai-Ichiro Ch, Naoyuki M, Naoyuki Y, Koh-Hei Sono-

da. Potential new modes of treatment of neurotrophic keratopathy. Cornea. 2015;34(11):121-127. https://doi.org/10.1097/ICO.0000000000000587

7. Kasparova EA. Gnoinye yazvy rogovitci: etiologiya, patogenez, klassifi-caciya. Vestnik oftalmologii. 2015;131(5):87-97. (In Russ.).

https://doi.org/10.17116/oftalma2015131587-97

8. Kasparova EA. Purulent corneal ulcers: etiology, pathogenesis, classification. Vestnik oftalmologii. 2016;132(5):125-135. https://doi.org/10.17116/oftal-ma20161325125-135. (In Russ.).

9. Grusha YaO, Fedorov AA, Ivanchenko YuF. Experimental assessment of various external tarsorrhaphy techniques. Vestnik oftalmologii. 2010;126(3):15-18. (In Russ.).

10. Agafonova EI, Grusha YaO, Iskusnych NS. Pervyj otechestvennyj implantat dlya korrektsii lagoftalma. II Natsional’nyj congress «Plasticheskaya hirur-giya». M.: Krokus Expo, 2012;83. (In Russ.).

11. Vasil’eva VA. O zamene epiteliya rogovicy epiteliem kon’unktivy v experimental’nyh usloviyah. Sbornik nauchnyh rabot kafedry gistologii I embriologii Leningradskogo sanitarno-gigienicheskogo medicinskogo in-stitute «Reaktivnost’ I plastichnost’ tkanej. T.16». L.: Gosudarstvennoe izdatel’stvo medicinskoj literatury. 1953;142-143. (In Russ.).

12. Huang AJ, Tseng SC. Corneal epithelial wound healing in the absence of limbal epithelium. Invest Ophthalmol Vis Sci. 1991 Jan;32(1):96-105.

13. Dua HS, Forrester JV. The corneoscleral limbus in human corneal epithe-lial wound healing. Am J Ophthalmol. 1990 Dec 15;110(6):646-656.

https://doi.org/10.1016/S0002-9394(14)77062-X

Received: 07.06.17

Problem of NPK-associated suppurative corneal ulcers is of interest for neurosurgeons and neurologists despite its pre-dominant ophthalmological field.

It is known that in neurosurgical practice neurotrophic keratopathy complicated by lagophthalmos is caused by tumors of posterior cranial fossa (trigeminal and cochlear schwanno-ma, meningioma), traumatic brain injury, brainstem stroke, etc.

Current approaches to treatment, intraoperative monitor-ing, advanced medication, development of various methods for prevention and treatment of keratopathy significantly reduced incidence of severe neurotrophic keratopathy and especially corneal ulcers. Nevertheless, this serious complication can oc-cur in some circumstances (medical, social). Thus, review of this issue in neurosurgical literature is actual.

Commentary

Treatment of 12 patients with suppurative corneal ulcers is presented in the article. Surgery included elimination of unre-sponsive to therapy focal inflammation. Authors proposed combination of corneal repair (lamellar or penetrating depend-ing on grade of corneal lesion), autoconjunctival coverage of the graft and partial blepharorrhaphy. There are favorable out-comes due to effective management of inflammation, improved visual acuity in some cases and preserved eye as an organ in all cases. It is important that it was done in patients with impaired corneal regeneration and high risk of graft rejection. The last is confirmed by redo keratoplasty almost in half of cases after pre-vious penetrating keratoplasty.

N.K. Serova (Moscow, Russia)


CASE REPORTS

Tumors of subcortical nuclei are rare pathology ac-counting for about 1—5% of all intracerebral tumors in children and adults. Incidence of correct diagnosis of subcortical tumors was low (about 10%) and postopera-tive mortality was 50—70% before introduction of cur-rent neuroimaging methods [1—3]. CT and MRI are useful to correctly determine intracerebral tumors topog-raphy while neurosurgeons are able to define the most optimal surgical approach with lower risk of postopera-tive complications [4—10]. Nevertheless, precise topical diagnosis of subcortical nuclei tumors may be still diffi-cult, while surgical procedure may be followed by high risk of injury of deep cerebral structures and accordingly severe disability of patient.

MR-tractography makes it possible to determine normal course of motor conductive pathways within in-ner capsule and brainstem and their dislocation in tu-mors of subcortical nuclei [11—13]. This method gives more accurate information about tumor topography and optimal approach without involvement of functionally important zones may be determined.

The objective of the study was to demonstrate the choice of surgical approach and surgery for deep local tu-mors by using of preoperative MR-tractography, MRI and neurological status in two cases.

Normal MR-anatomy of the main subcortical struc-tures and inner capsule is presented in Fig. 1.

Clinical case №1

Patient A., 8 years old, awkwardness in right hand has been occurred 3 months before admission to the Bur-denko Neurosurgery Center. Therefore, advanced weak-ness in the arm followed by weakness in the right leg was noted. In contrast-enhanced MRI left-sided cystic tumor of lenticular nuclei and lateral thalamus was observed (Fig. 2a, b). Tumor with clear borders deformed adjacent



Mr-Tractography in Diagnosis and Choice of Neurosurgical Approach for Subcortical Nuclei TumorsSH.U. KADYROV*, A.N. KONOVALOV, I.N. PRONIN


We describe two cases of surgical treatment of well-circumscribed basal ganglia tumors. The choice of a neurosurgical approach to a deep tumor was based on the MR tractography data and depended on the course and dislocation extent of the corticospinal tract. MR tractography provides information on the course and dislocation or destruction extent of the corticospinal tract running in the internal capsule and brainstem and clarifies the exact location of a tumor within the basal ganglia. This information promotes the choice of an optimal approach for radical resection of well-circumscribed tumor, leading to improvement in neurological symptoms and patient’s quality of life.

Keywords: MR tractography, basal ganglia tumors, neurosurgical approach.

thalamic parts, inner capsule and bottom of lateral ven-tricle. Contrast agent accumulation was noted within solid node.

There was severe right-sided hemiparesis with im-paired muscular strength up to 2 and 3 scores in arm and leg, respectively. Extrapyramidal muscular disorders were diagnosed. Visual fields were intact.

MR-tractography was used to determine tumor to-pography and optimal surgical approach (Fig. 2c—e). Reconstruction of motor pathways of inner capsule re-vealed their deformation and dorsal and medial displace-ment on the left. So, tumor of lenticular nucleus was sup-posed. There was predominant dorsomedial displace-ment of inner capsule and thalamus and lateral disloca-tion of insular cortex.

Transinsular approach seemed to be optimal in view of dorsomedial and medial displacement of motor path-ways of inner capsule.

Surgery. Osteoplastic frontotemporal craniotomy followed by dissection of posterior lateral fissure was made. Deformed insular cortex was found. Cortex was dissected within the most severe deformation and swell-ing. Cystic tumor lied 5 mm deep to the cortex, large volume of xanthochromic output was drained. Dense gray tumor node was visualized. Tumor was excised by using of bipolar coagulation, suction, ultrasonic suction. There were clear borders of tumor as a rule. Direct stimu-lation of motor pathways has been simultaneously per-formed. Response from inner capsule fibers was ob-tained. We completely excised tumor up to visually intact brain tissue. Tumor removal was followed by formation of large cavity.

Histological diagnosis: pilocytic astrocytoma (WHO I).

Early postoperative improvement of right-sided hemiparesis was noted.


Control MRI: picture of total tumor removal, no de-formation of subcortical structures. There is left-sided postoperative defect in posterior lenticular nucleus (Fig. 2f, g). Thalamus and inner capsule are intact. No deformation of left corticospinal tract in MR-tractogra-phy (Fig. 2h, i).

Further rehabilitation was followed by complete re-covery of right limbs’ function.

Clinical case № 2

Patient S., 15 years old with periodic headaches and transient numbness and weakness in right limbs for 1.5 years before hospitalization. Therefore, advanced weakness and impaired walking were observed.

MRI-imaging revealed left-sided anterobasal tha-lamic tumor (Fig. 3a, b). Trivial deformation of lateral ventricle’s bottom and dorsal thalamus was noted.

MR-tractography was used to determine tumor to-pography and optimal surgical approach (Fig. 3c, d). Re-construction of corticospinal tracts confirmed postero-lateral dislocation of inner capsule’s pathways on the left.

Right-sided hemiparesis with impaired muscular strength (4 scores) was diagnosed. Hemianopsia was ab-sent.

In view of MRI and MR-tractography data transcal-losal approach was preferred despite no hydrocephalus, minimal deformation of lateral ventricle and need to dis-sect anterior thalamus. It seemed to be the least traumat-ic due to posterior displacement of corticospinal tract.

Surgery. It was performed linear incision of soft tis-sues in frontal area. Right-sided osteoplastic frontal cra-

niotomy was followed by exposure of sagittal sinus mar-gin. Dura mater was dissected arcuately with base turned to sagittal sinus. Right hemisphere was withdrawn later-ally to visualize corpus callosum, pericallosal cistern has been opened, anterior cerebral arteries were diverged to the sides. Сorpus callosum was dissected for 1.5 cm in order to enter left lateral ventricle. Its bottom was some-what deformed. There were vascular plexus, interventric-ular orifice and thalamo-striate vein in ventricular lu-men. Ependyma was dissected within the most severe deformation of lateral ventricle’s bottom between vascu-lar plexus and thalamo-striate vein at the level of Monro foramen. Gray-yellow tumor lied at the depth of 5 mm. Specimen was excised for histological examination. Tu-mor removal was carried out by using of conventional and ultrasound suction, bipolar coagulation, dissector and fenestrated forceps. Resection of central tumor fol-lowed by subsequent excision of peripheral parts were ap-plied. Basal penetrating vessels were partially coagulated and intersected. It should be noted that borders were un-clear in some zones. Entire tumor was gradually excised up to intact cerebral tissue. Regression of lateral ventricle deformation was observed after tumor removal. Hemo-stasis was achieved with bipolar coagulation and gauze.

Histological diagnosis: pilocytic astrocytoma (WHO I).

Hemiparesis remained the same in postoperative pe-riod.

After 1 year MRI-data of postoperative left-sided de-fect of anterior thalamus without residual tumor were obtained (Fig. 3e, f). MR-tractography did not show de-

Fig. 1. Normal MRI-scan. T1-mode, axial (a) and frontal (b) plane: 1 — head of caudate nucleus, 2 — putamen, 3 — globus pallidus, 4 — claustrum, 5 – inner capsule, 6 — thalamus, 7 – external capsule, 8 — insula, 9 – nucleus ruber.

а b


CASE REPORTS

а b c

d e

f g

Fig. 2. Case report №1.a, b – MRI prior to surgery. Contrast-enhanced T1-mode in axial (a) and frontal (b) planes. Explanations in the text; c—e — MR-tractography prior to surgery. Thinning, dorsal and medial displacement of corticospinal tract on the left (arrows); f, g — contrast-enhanced MRI after surgery in axial (f) and frontal (g) planes.

See continue of the Figure on the next page


h i

Fig. 2. Case report №1 (continued).Explanations in the text; h, i — MR-tractography in 6 months after surgery. Course of corticospinal tract on the left, no displacement. Asymmetry of pyramidal pathways (S<D).

formation and displacement of inner capsule on the left (Fig. 3g—i).

Complete muscular recovery of right arm and leg was noted.

DiscussionIt is often difficult to define optimal surgical access

for subcortical surgery due unclear data about tumor placement. There are similar problems in surgery for tha-lamic tumors. The most advisable approach is deter-mined depending on localization of the tumor within these structures, its dimensions, dislocation of adjacent structures (inner capsule, rostral midbrain, fornix, lateral and third ventricles), presence or absence of hydrocepha-lus. Surgical access should be able to avoid injury of func-tional structures and disability of the patient.

Until now, surgeons have adhered to medication for subcortical tumors due to high incidence of postoperative complications and mortality [6, 14]. High-field MRI, comprehensive awareness for neurosurgical anatomy, ac-cumulation of microsurgical experience and develop-ment of minimally invasive procedures are followed by advanced indications for subcortical surgery. At the same time, postoperative mortality and morbidity are signifi-cantly improved. There are more and more publications highlighting outcomes of surgery for subcortical nuclei tumors in not only single cases but also in series of pa-tients [4, 9, 15]. Authors [6, 9, 11, 16] classify deep tu-mors depending on their topography, describe neurosur-gical approaches to subcortical tumors depending on lo-calization, highlight their advantages and disadvantages.

In current neurosurgery the main objectives of intra-cerebral tumors management are their radical removal and minimization of postoperative disability. At the pres-

ent time it is undoubtedly to excise subcortical focal tu-mor (pilocytic astrocytoma) because radical surgery may be followed by improved neurologic symptoms, pro-longed disease-free period or recovery. Neurosurgical treatment of subcortical tumors is particularly important due to high incidence of pilocytic astrocytoma in chil-dren. Infiltrative tumors with compact part (in MRI) may be also successfully resected with improvement of patient's state. It is a factor of better recurrence-free and overall survival.

Standard MRI-scans including high anatomical quality don’t assess conductive pathways, their deforma-tion and dislocation or destruction in focal and infiltra-tive brain tumors. In this regard MR-tractography is very important for deep tumors located near corticospinal and visual tracts.

Diffusion-tensor MR-tractography is based on mea-surement of water molecules movement along nerve fi-bers. It is followed by reconstruction of normal conduc-tive pathways and those in different pathologies of central nervous system such as metabolic cerebral lesions, dis-seminated sclerosis, vascular encephalopathy, severe cra-niocerebral trauma, brain and spinal cord tumors [12, 13].

Burdenko Neurosurgery Center accumulates an ex-perience of tractography for deep tumors. This report de-scribes two cases of subcortical tumors with certain topo-graphic features. Surgical approach was determined via analysis of MR-tractography data. It was confirmed that subcortical tumors depending on dimensions and loca-tion can deform and displace corticospinal tract within inner capsule and brainstem. This information is valuable to choose optimal approach. Thus, transcallosal access is advisable for posterolateral displacement of motor path-ways (case report №2). Transinsular or temporal tran-schoroidal approaches are less traumatic for anterome-


CASE REPORTS

а b

c d

e f

Fig. 3. Case report №2.a, b — MRI prior to surgery. Contrast-enhanced T1-mode in axial (a) and frontal (b) planes. Explanations in the text; c, d — MR-tractography prior to surgery. Posterolateral displacement of corticospinal tract on the left (arrows); e, f – MRI in 1 year after surgery. Contrast-enhanced T1-mode in axial (e) and frontal (f) planes. Explanations in the text. Trajectory of contralateral transcallosal approach is determined in frontal projection.

See continue of the Figure on the next page


Fig. 3. Case report №2 (continued).g—i — MR-tractography in 1 year after surgery. No deformation of inner capsule and left corticospinal tract located behind postoperative defect.

g h

i

dial and medial dislocation of inner capsule (case report №1). Complete excision was performed in both cases that was followed by improved neurological symptoms and quality of life.

ConclusionsMR-tractography is useful to determine features of

dislocation or destruction of corticospinal tract within

inner capsule and brainstem and to detail tumor location within subcortical nuclei. Therefore, it is followed by choice of optimal approach, radical resection of focal tu-mor with regression of neurological symptoms and im-proved quality of life.


REFERENCES1. Arseni C. Tumors of the basal ganglia: their surgical treatment. AMA Ar-

chives of Neurology & Psychiatry. 1958;80(1):18-24. https://doi.org/10.1001/archneurpsyc.1958.02340070036003

2. Cheek WR, Taveras JM. Thalamic tumors. Journal of neurosurgery. 1966;24(2):505-513. https://doi.org/10.3171/jns.1966.24.2.0505

3. Tovi D, Schisano G, Liljeqvist B. Primary tumors of the region of the thala-mus. Journal of neurosurgery. 1961;18(6):730-740.

https://doi.org/10.3171/jns.1961.18.6.0730

4. Konovalov AN, Kadyrov ShU. Surgical approaches to thalamic tumors. Zh Vopr Neirokhir Im NN Burdenko. 2011;75(1):4-11. (In Russ.).

5. Konovalov AN, Kadyrov ShU. Temporal transchoroidal approach for tu-mors of the midbrain and thalamus. Zh Vopr Neirokhir Im NN Burdenko. 2013;77(4):16-25. (In Russ.).

6. Albright AL. Feasibility and advisability of resections of thalamic tumors in pediatric patients. Journal of Neurosurgery: Pediatrics. 2004;100(5): 468-472. https://doi.org/10.3171/ped.2004.100.5.0468

7. Hamlat A, Morandi X, Riffaud L, Carsin-Nicol B, Haegelen C, Helal H, Brassier G. Transtemporal-transchoroidal approach and its transamygdala extension to the posterior chiasmatic cistern and diencephalo-mesence-phalic lesions. Acta neurochirurgica. 2008;150(4):317-328.

https://doi.org/10.1007/s00701-007-1460-28. Özek MM, Türe U. Surgical approach to thalamic tumors. Child’s Nervous

System. 2002;18(8):450-456. https://doi.org/10.1007/s00381-002-0608-x9. Rangel-Castilla L, Spetzler RF. The 6 thalamic regions: surgical approaches

to thalamic cavernous malformations, operative results, and clinical out-comes. Journal of neurosurgery. 2015;123(3):676-685.

https://doi.org/10.3171/2014.11.JNS14381


CASE REPORTS

Besides brainstem deep cerebral structures are also somewhat difficult for surgery. Choice of approach is the key for outcomes after these procedures. Complications due to trauma of functionally important adjacent deep structures such as inner capsule, midbrain can lead to se-vere disability of patient. Constant desire to reduce inva-siveness in both medical and diagnostic medicine has been reflected in this work. It is natural that clinicians strive to obtain as much information as possible about both anatomical and functional individual characteristics of particular patient.

Radiological diagnosis has come out beyond X-ray procedures alone. Advanced development of MRI (new applications, contrast agents, quantitative and visual post-processing) influenced quality of neurosurgery. One of these applications (tractography) is analyzed in the article. MR-tractography is able to give information about normal course of motor pathways within inner capsule and brainstem and their dislocation in case of

Commentary

subcortical tumors. Method is not absolutely new, how-ever advanced quality significantly improves diagnostic assessment of cerebral and spinal conductive pathways, their relationship with affected zones and vessels.

Technique is actual because dislocation and destruc-tion of conductive pathways may be estimated by using of diagnostic data. There is an opportunity to choose the most optimal surgical approach besides accurate infor-mation about tumor’s boundaries and invasion into nerve fibers.

There are various reports in foreign literature about this issue in contrast to national databases. Relatively low prevalence of this technique is due to difficult post-pro-cessing of data although methodology is not technically complex in general.

The authors rightly note that until now surgeons pre-ferred medication for subcortical tumors due to high in-cidence of postoperative complications.

M.B. Dolgushin (Moscow, Russia)

10. Zheng X, Xu X, Zhang H, Wang Q, Ma X, Chen X, Sun G, Zhang J, Jiang J, Xu B, Zhang J. A preliminary experience with use of intraoperative mag-netic resonance imaging in thalamic glioma surgery: a case series of 38 pa-tients. World neurosurgery. 2016;89:434-441.

https://doi.org/10.1016/j.wneu.2016.01.09211. Kadyrov ShU, Konovalov AN, Ozerova VI, Korniyenko VN, Pronin IN.

X-ray neurodiagnosis of thalamic tumors. Vopr Neirokhir Im NN Burdenko. 2007;3:3-11. (In Russ.).

12. Pronin IN, Fadeeva LM, Zakharova NE, Dolgushin MB, Podoprigora AE, Kornienko VN. Diffusion tensor magnetic resonance imaging and tractogra-phy. 2008;2(1):32-40. (In Russ.).

13. Kis D, Máté A, Kincses ZT, Vörös E, Barzó P. The role of probabilistic tractography in the surgical treatment of thalamic gliomas. Operative Neu-rosurgery. 2014;10(2):262-272.

https://doi.org/10.1227/NEU.0000000000000333

14. Kadyrov ShU. Surgical treatment of thalamic tumors. Revue of literature. Neirokhirurgiia i nevrologiia detskogo vozrasta. 2007; 13(2):71-77. (In Russ.).

15. Bilginer B, Narin F, Işıkay I, Oguz K, Söylemezoglu F, Akalan N. Tha-lamic tumors in children. Child’s Nervous System. 2014;30(9):1493-1498.

https://doi.org/10.1007/s00381-014-2420-916. Steinbok P, Gopalakrishnan CV, Hengel AR, Vitali AM, Poskitt K,

Hawkins C, Drake J, Lamberti-Pasculli M, Ajani O, Hader W, Mehta V, McNeely PD, McDonald PJ, Ranger A, Vassilyadi M, Atkinson J, Ryall S, Eisenstat DD, Hukin J. Pediatric thalamic tumors in the MRI era: a Cana-dian perspective. Child’s Nervous System. 2016;32(2):269-280.

https://doi.org/10.1007/s00381-015-2968-z

Received: 11.07.17


Nasal liquorrhea is one of the serious problems in skull base tumors surgery and associated with high risk of suppurative-septic complications. Liquoric fistula repair after endoscopic and microsurgical surgery for basal tu-mors usually results closure of the defect with various au-tologous materials (fat, fascia, pedicled periosteal-apo-neurotic or mucoperiosteal flap) and occlusion with fi-brin-thrombin glue [1, 2].

Endoscopic transnasal approaches to ACF base be-come more and more popular because minimally inva-sive technique is able to avoid traction of brain and cra-nial nerves [3]. However, only small tumors located along midline are suitable for such procedures. Disadvantage of endoscopic procedures is high incidence of nasal liquor-rhea (5—31.6%) [4, 5].

Intracranial microsurgical accesses (subfrontal/su-praorbital) are indicated for large meningiomas of ACF base, as well as in patients with intact smelling in order to prevent postoperative anosmia [6].

Nasal liquorrhea is found in 3.5% after microsurgical transcranial removal of ACF tumors with intra- and ex-tracranial proliferation. Liquorrhea is usually caused by bone defect after frontal sinus opening or excision of skull base tumor [1].

Incidence of liquorrhea after frontal sinus dissection (supraorbital approach) is 0—9% [7, 8]. F. Thaher et al. [9] reported nasal liquorrhea after supraorbital surgery in 2.3% of patients with meningitis in 1 patient (12.5% of those with nasal liquorrhea). According to R. Komotar et



Platelet Gel for Repair of Skull Base Liquoric Fistula (Case Report and Literature Review)O.I. SHARIPOV, M.A. KUTIN, A.V. BAYUKLIN, A.A. IMAEV, A.A. ABDILATIPOV, A.B. KURNOSOV, D.V. FOMICHEV, N.I. MIKHAYLOV, P.L. KALININ*

Burdenko Neurosurgical Institute, 4-ya Tverskaya-Yamskaya Str., 16, Moscow, Russia, 125047

Nasal liquorrhea is a serious problem in surgery of skull base tumors, which is associated with a high risk of purulent-septic complications. This paper presents a case of successful repair of a cerebrospinal fluid fistula with an autologous platelet gel in the postoperative period after removal of meningioma of the anterior cranial fossa base, which was accompanied by a purulent-inflammatory complication in the surgical wound area.

Keywords: nasal liquorrhea, platelet gel.

Abbreviations: DM — dura materACF — anterior cranial fossaCT — computed tomographyMV — mechanical ventilationPRP — platelet rich plasmaPPP — platelet poor plasma

al. [10], postoperative liquorrhea after surgery for olfac-tory meningioma occurred in 6% of cases, osteomyelitis and wound inflammation were noted in 4% of patients.

Frontal sinus repair is traditionally performed with pedicled periosteal-aponeurotic flap which is transferred through the sinus, sutured to basal frontal DM and sealed with fibrin-thrombin glue [11, 12]. In several patients af-ter subfrontal surgery for pituitary adenoma liquoric fis-tula was successfully closed by transcutaneous injection of fibrin-thrombin glue into frontal sinus as it was de-scribed by B.A. Kadashov in 2007 [13].

Platelet gel (thrombogel) is a two-component glue consisting of concentrated fibrinogen solution with plate-let growth factor and thrombin solution. Both compo-nents are obtained from autologous PRP that is followed by completely autologous gel and prevents risk of infec-tions [14].

In 2011 G. Paternoster et al. [15] described case re-port of successful DM defect closure by percutaneous platelet gel injection in patient with pseudomeningocele after craniovertebral decompression for Chiari type I malformation. In their opinion, this technique is effec-tive and minimally invasive.

J. Thorn et al. [14] used thrombogel combined with bone crumb for reconstructive maxillofacial surgery.

We present successful closure of frontal sinus fistula after removal of large ACF meningioma via right-sided subfrontal approach by transcutaneous platelet gel injec-tion into the sinus.


CASE REPORTS

Case report

Patient G., 60 years old, with large basal ACF menin-gioma (Fig. 1) followed by emotional (reduced criticism, memory for current events) and visual impairment (vi-sual acuity: OD=1.0; OS=0,4. Visual field: OD — nor-mal, OS – impaired central vision, narrowed in temporal half).

Tumor was excised via right-sided subfrontal ap-proach without any peculiarities and technical difficul-ties. Intraoperative dissection of frontal sinus was fol-lowed by its repair with pedicled periosteal-aponeurotic flap and fixation to dura mater (Fig. 2). Tumor was com-pletely removed. There were no postoperative neurologic disorders and nasal liquorrhea.

Severe stunning and hiccough appeared after 3 days. There were CT-signs of advanced diffuse edema of fron-tal lobes which was prior to surgery, diencephalic edema and narrowing of ambient cistern (Fig. 3).

Patient underwent bilateral decompressive trepana-tion next day (primary bone flap and external parts of sphenoid wing resection on the right; frontotemporal trepanation on the left). Right-sided frontal subcutane-ous and epidural accumulation of pus was intraoperative-ly observed (Enterobacter asburiae). Frontal sinus repair with aponeurosis was visually intact – frontal defect was covered by periosteal-aponeurotic flap all over and su-tured to DM.

Normal consciousness and no hiccough were re-vealed after surgery. Meningitis was excluded after lum-bar puncture. Antibacterial therapy was administered due to pyoinflammatory changes of the wound (subcutane-ous and epidural).

After 2 days nasal liquorrhea occurred. Pneumo-cephalus progression was revealed on CT (Fig. 4). There were complaints of severe headache, consciousness ag-gravated up to moderate stunning.

In view of subcutaneous growth of Enterobacter as-buriae in right frontal area it was considered that redo repair of frontal sinus was associated with high risk of in-tracranial infection. External lumbar drainage under ad-vanced cerebral edema could lead to dislocation syn-drome.

Therefore, frontal sinus obliteration with autologous platelet gel was preferred. Surgery was performed under sedation with propofol 4 mg/kg/h and respiratory pro-tection with laryngeal mask, SIMV ventilation followed by saturation 98—100%. Patient's venous blood was con-ventionally withdrawn into container (CPDA1 stabilizer 63 mL, “Macopharma”, France). Autologous blood 450 mL (+ 63 mL of stabilizer) was separated into red cells 233 mL, PPP 240 mL and PRP 40 mL in Xtra auto-transfusion system ("Sorin group", Germany) according to standard protocol for PRP. Red cells and PPP were intravenously reinfused immediately after separation (36 ml/min). 6 out of 40 mL of PRP were used to prepare thrombin. PRP was mixed with PPP 8 mL and calcium chloride 10% 1 mL. Therefore, mixture was divided into

4 Petri dishes until formation of clot (40 min). 8 mL of thrombin rich liquid were obtained from this mixture af-ter clot retraction (Fig. 5a). In vitro control was performed before thrombogel injection into frontal sinus. Thrombin 1 mL and PRP 4 mL were mixed in a syringe 10 mL fol-lowed by stopwatch switching-on. Thrombogel was formed after 15 s (Fig. 5b).

Aseptic percutaneous puncture of frontal sinus was performed within anterior margin of the defect by using of needle 14G. Sinus was irrigated with saline. PRP mixed with thrombin was injected by 5 mL step-by-step (PRP 4 mL + thrombin 1 mL) into frontal sinus where final platelet gel was formed. Entire procedure was per-

а

b

Fig. 1. Contrast-enhanced MRI of the brain.a – preoperative sagittal plane; b — frontal plane: giant meningioma of ACF base.


а b

Fig. 4. CT-data of advanced pneumocephalus.a — 8 days after the 2nd procedure; b — 12 days after the 2nd procedure

formed within 15 min; 30 mL of platelet gel were injected (Fig. 6).

There was no recurrent postoperative liquorrhea. CT confirmed significant regression of pneumocephalus (Fig. 7).

Antibacterial therapy was very effective for suppura-tive inflammation of postoperative wound. Patient was discharged without severe neurologic disorders and nasal liquorrhea.

Two-month follow-up did not reveal nasal liquor-rhea and paranasal sinuses inflammation.

DiscussionFrontal sinus repair by temporal muscle or fat from

anterior abdominal wall may be associated with risk of infection. In this situation pedicled periosteal flap should be preferred for sinus reconstruction [16].

External lumbar drainage for 5—7 days is usually de-ployed for nasal liquorrhea after skull base repair. Redo repair of sinus is applied if drainage is ineffective [13].

Lumbar drainage was refused in our clinical case due to risk of dislocation followed by brainstem strangulation

Fig. 2. Frontal sinus repair by pedicled frontoparietal periosteal-aponeurotic flap passed through frontal sinus and fixed to dura mater. Fixation to DM is indicated by asterisk; yellow line — DM; red line — periosteum; green line — frontal sinus.

Fig. 3. CT in 3 days after surgery. Severe cerebral edema, mainly in frontal lobes, diencephalon, ambient cistern constriction, small intracerebral hematoma in right frontal lobe.


CASE REPORTS

in conditions of severe edema of frontal lobes. Moreover, redo reconstruction of frontal sinus could be accompa-nied by intracranial infection, meningitis, cerebral ab-scesses and epidural empyema due to epidural infection (Enterobacter asburiae).

In our opinion, nasal liquorrhea was due to cerebro-spinal fluid outflow through sutures fixing aponeurosis to the edge of DM. Conventional adhesive compositions (fibrin-thrombin glue) is not advisable for frontal sinus filling and DM sealing because frontal sinus volume is too large for standard amount of fibrin-thrombin glue used for skull base repair. In this situation the use of au-tologous blood is advisable to prepare enough volume of autologous platelet gel for filling of entire frontal sinus

Fig. 5. Type of glue.a — mixture of PRP with PPP 8 mL and calcium chloride 10% 1 mL as a thin layer in Petri dish; b — thrombogel appearance.

а

b

Fig. 6. Percutaneous filling of right frontal sinus by thrombogel.

Fig. 7. CT in 9 days after the 3rd surgery.Improvement of pneumocephalus.


REFERENCES1. Bekyashev AKh. Nizkiye subfrontal’nyye dostupy k osnovaniyu cherepa.

Kliniko-anatomicheskoye obosnovaniye: Dis. … kand. med. nauk. M. 2003. (In Russ.).

2. Kutin MA, Kalinin PL, Fomichev DV, Kadashev BA, Shilin AD, Nersesy-an MV, Fomochkina LA, Astafieva LI. Experience of skull base defect re-construction using local pedicled grafts in endoscopic transsphenoidal sur-gery. Voprosy Neirokhirurgii. N.N. Burdenko. 2012;2:42-49. (In Russ.).

3. Borg A, Kirkman MA, Choi D. Endoscopic endonasal anterior skull base surgery: a systematic review of complications during the past 65 years. World Neurosurg. 2016 Nov;95:383-391. Epub 2016 Mar 4.

https://doi.org/10.1016/j.wneu.2015.12.105.4. Gardner PA, Kassam AB, Thomas A, Snyderman CH, Carrau RL, Mintz AH,

Prevedello DM. Endoscopic endonasal resection of anterior cranial base me-ningiomas. Neurosurgery. 2008 Jul;63(1):36-52; discussion 52-54.

https://doi.org/10.1227/01.NEU.0000335069.30319.1E5. Komotar RJ, Starke RM, Raper DM, Anand VK, Schwartz TH. Endoscop-

ic endonasal versus open transcranial resection of anterior midline skull base meningiomas. World Neurosurg. 2012 May-Jun;77(5-6):713-724. Epub 2011 Nov 7. https://doi.org/10.1016/j.wneu.2011.08.025

6. Abbassy M, Woodard TD, Sindwani R, Recinos PF. An overview of ante-rior skull base meningiomas and the endoscopic endonasal approach. Oto-laryngol Clin North Am. 2016 Feb;49(1):141-152.

https://doi.org/10.1016/j.otc.2015.08.0027. Paiva-Neto MA, Tella OIJr. Supra-orbital keyhole removal of anterior fossa

and parasellar meningiomas. Arq Neuropsiquiatr. 2010;68(3):418-423.8. Ormond DR, Hadjipanayis CG. The supraorbital keyhole craniotomy

through an eyebrow incision: its origins and evolution. Minim Invasive Surg. 2013;2013:296469. Epub 2013 Jul 10.

https://doi.org/10.1155/2013/2964699. Thaher F, Hopf N, Hickmann AK, Kurucz P, Bittl M, Henkes H, Feigl G.

Supraorbital keyhole approach to the skull base: evaluation of complications related to csf fistulas and opened frontal sinus. J Neurol Surg A Cent Eur Neurosurg. 2015 Nov;76(6):433-437. Epub 2015 Jul 27.

https://doi.org/10.1055/s-0034-1389368

10. Komotar RJ, Starke RM, Raper DM, Anand VK, Schwartz TH. Endoscop-ic endonasal versus open transcranial resection of anterior midline skull base meningiomas. World Neurosurg. 2012 May-Jun;77(5-6):713-724. Epub 2011 Nov 7. https://doi.org/10.1016/j.wneu.2011.08.025

11. Konovalov AN, Konovalov AN, Makhmudov UB, Ivanov SL, Kada-shev BA, Cherekayev VA, Trunin YuK, Vinokurov AG, Tanyashin SV, Mukhametzhanov DZh, Spallone A., Usachev DYu, Shamanskiy VN, Shkarubo AN. Khirurgiya osnovaniya cherepa. Voprosy neurokhir. 1998;4:3-9. (In Russ.).

12. Korniyenko VN, Konovalov AN, Cherekayev VA, Belov AI, Vinokurov AG, Gulyayeva OA, Bekyashev AKh. Rasshirennyy subfrontal’nyy dostup k opukholyam osnovaniya cherepa. Voprosy neyrokhirurgii. im. N.N. Burden-ko. 2003;2:10-16. (In Russ.).

13. Adenoma gipofiza. Klinika, diagnostika, lecheniye. Pod red. Kadasheva BA. M. 2007;278-279. (In Russ.).

14. Thorn JJ, Sørensen H, Weis-Fogh U, Andersen M. Autologous fibrin glue with growth factors in reconstructive maxillofacial surgery. Int J Oral Maxil-lofac Surg. 2004 Jan;33(1):95-100.

15. Paternoster G, Massimi L, Capone G, Tamburrini G, Caldarelli M, Di Rocco C. Subcutaneous blood patch for iatrogenic suboccipital pseudo-meningocele following decompressive suboccipital craniectomy and enlarg-ing duroplasty for the treatment of Chiari I malformation. Technical note. Childs Nerv Syst. 2012 Feb;28(2):287-290. Epub 2011 Dec 8.

https://doi.org/10.1007/s00381-011-1641-416. Kushel’ YuV, Semin VYe. Kraniotomiya. Khirurgicheskaya tekhnika. M.

1998. (In Russ.).17. Patel MR, Caruso PA, Yousuf N, Rachlin J. CT-guided percutaneous fibrin

glue therapy of cerebrospinal fluid leaks in the spine after surgery. AJR Am J Roentgenol. 2000 Aug;175(2):443-446.

https://doi.org/10.2214/ajr.175.2.1750443Received: 03.04.17

cavity. Fibrin filaments are able to firmly fix entire throm-bogel to bone defect followed by hermetic closure of DM.

Viscosity of platelet gel is similar to that of packed red cells, so injection should be done through the needle 16—14G. Gel formation time in particular patient should be assessed in vitro prior to injection. The last is necessary to determine optimal delivery rate and ensure further re-liable fixation of platelet gel within the sinus. PRP ob-tained in Xtra centrifuge ("Sorin group", Germany) ad-ditionally consists of macrophages and granulocytes which stimulate local immunity besides proliferation and reduce risk of infectious complications.

Normal level of red blood cells and platelets is neces-sary for successful preparation of platelet gel (platelets near upper rather lower limit is desirable).

Aseptic meningitis as complication of platelet glue injection was described by M. Patel et al. [17]. There were no any complications in our practice.

ConclusionPercutaneous obliteration of frontal sinus by throm-

bogel proved to be effective and minimally invasive pro-cedure for liquoric fistula repair. In our example we have demonstrated efficacy and safety of platelet gel for skull base repair in case of infectious wound complications. We consider thrombogel is only effective sealant with lo-cal disinfectant effect rather main material for repair. Main disadvantages of this technique for transsphenoidal surgery are dense consistency of the gel that requires large needle and uncontrolled rate of polymerization.



CASE REPORTS

The authors presented treatment of liquoric fistula of fron-tal sinus after transcranial removal of anterior cranial fossa me-ningioma.

Prolonged liquorrhea after transsphenoidal surgery or cra-niotomy followed by frontal sinus dissection often requires ad-ditional treatment or redo surgery. Liquorrhea may be followed by inflammation and need for long antibacterial therapy in 5—10% of cases. So, adequate closure of the defect is prerequi-site for successful treatment.

Focal traumatic fistula is usually closed spontaneously and does not require surgery in acute period while iatrogenic fistula often needs for additional intervention. Obviously, prevention of postoperative fistula is the most optimal. The use of temporal muscle and adipose tissue combined with pedicled periosteal flap is standard for fistula repair. In our clinic fibrin glue (TIS-SEEL, Baxter, Deerfield, IL, USA) is standard method besides above-mentioned approach. Adhesive is used for filling of epi-dural space and sinus defect.

One of the first reports about fibrin glue for fistula repair belongs to Fujii et al. (Simple management of CSF rhinorrhea after pituitary surgery, Surg Neurol, 1986; 26: 345—348).

Commentary

However, there is risk of adverse reactions if these products are applied. These reactions are described in the literature de-spite their small incidence due to use of autologous plasma. High cost of commercial products is another limitation.

Authors applied autologous blood for fibrin glue. This technique almost eliminates the risk of adverse reactions asso-ciated with commercial products. Procedure may be recom-mended if access to such products is limited or undesirable or large volume of fibrin glue is required. The authors used percu-taneous administration of the glue. It should be noted that lo-calization of the clot after injection is unclear despite complete closure of the fistula. Either navigation or transsphenoidal en-doscopy-assisted administration seem to be more appropriate in this situation. However, these comments do not derogate in-terest to the article.

A.Grigoryan (Macon, USA)


General anesthesia was introduced into neurosurgi-cal practice much later than in other surgical fields — on-ly in the 50s of the last century [1—3]. This is logically explaned: since H. Cushing neurosurgical interventions have been followed by diathermocoagulation and this is always risk of explosion in operating theatre if inflam-mable anesthetics are applied. Thus, in analysis of 230 cases of explosions and fires in operating theatre 91% of these incidents were associated with use of such anes-thetics [4]. This problem was solved only after introduc-tion into clinical practice of the first nonflammable anes-thetic Halothane. Obviously, neurosurgical interventions were also performed earlier under local anesthesia of the scalp because its sensory innervation has been compre-hensively studied and there are no pain receptors in the brain [5—8]. Undoubtedly, local anesthesia has a lot of inconveniences, especially in critically ill patients with severe mental disorders and children, while basal pro-cesses impede adequate analgesia. Local anesthetics were also imperfect with cardio- and neurotoxic properties and short analgesic effect. Therefore, neurosurgeons and their patients quickly evaluated the comfort of neurosur-gery under general anesthesia. As a result, local anesthe-sia was almost abandoned (except for neurosurgery in drug-resistant epilepsy where local anesthesia of the scalp is still widely used [8—11]). Anesthetics and anes-

e-mail: [email protected]© A.Yu. Lubnin, 2018


Awake Neurosurgery: Forward to the PastA.YU. LUBNIN


We present an analytical review of various neurosurgical interventions in conscious patients. An analysis of the literature indicates growing interest in this problem. Craniotomy in conscious patients has been extensively used in resection of space-occupying cerebral lesions in the eloquent hemispheric areas and in epilepsy surgery. In recent years, there have been a number of reports on interventions in conscious patients with other neurosurgical pathologies, which may be regarded as a new emerging tendency in neurosurgery and neuroanesthesiology. Neurosurgery in conscious patients provides a special advantage because it enables highly functional neuromonitoring without use of complex devices.

Keywords: neurosurgery, craniotomy in conscious patients, neuromonitoring.

thesia modes have been improved for the decades after passed since halothane introduction. However, the main thing remained unchanged — patient was unconscious during neurosurgical procedure. However, some changes occurred over time. The first sign of the new trend was idea of intraoperative awakening to control higher corti-cal functions (primarily speech) in surgery for hemi-spheric processes (tumors, arteriovenous malforma-tions). One of the first statements about this absolutely adequate idea belongs to Dr. J. Girvin [12]. But its further development is indisputable merit of two American neu-rosurgeons — G. Ojemann and M. Berger [13—16]. They suggested new neurosurgical approach (awake cranioto-my, AC) in contrast to general anesthesia. Twenty years ago there were less than 10 neurosurgical clinics in the world where AC was actually used. It seems to be now there is no hospital where this technique is not applied and advanced interest to this approach is present [17]. There are over 300 reports for this issue in PubMed data-base. The reason for interest is understandable. Proce-dure is able to improve functional outcomes of neurosur-gery and type of tumor resection according to recent meta-analysis [18].

Hemispheric tumors surgery is now a classic example of AC application. But it is interesting that several reports have recently appeared in literature describing the use of

Abbreviations:ICA — internal carotid arteryEP — evoked potentialsDNM — dynamic neurological monitoringAС — awake craniotomyCSR — chiasmatic-sellar regionCCT — craniocerebral traumaEEG — electroencephalographyECoG — electrocorticographyDBS — deep brain stimulation

REVIEWS


REVIEWS

AC in surgery for cerebral aneurysms, extracranial-intra-cranial bypass, transsphenoidal surgery for chiasmatic-sellar tumors, carotid endarterectomy, endovascular and functional neurosurgery, spinal neurosurgery and even in surgery for cochlear neuroma. This work is an attempt to understand this trend. Why do neurosurgeons around the world prefer general anesthesia despite certain difficulties inevitably associated with awake neurosurgery?

Hemispheric tumors surgery

This is perhaps the most known field for AC in mod-ern neurosurgery. We are talking about tumors or less of-ten arteriovenous malformations of posterolateral, pari-etal and temporal lobes in dominant hemisphere which may be associated with risk of postoperative aphasia. The most impressive results of AC are obtained in these cases. There are a lot of reports about this issue including our works devoted to various aspects of the technique: neuro-surgical and anesthesiological features, unfavorable events and personal patient’s feelings [19—36]. So, we suggest to consider those publications rather detailed de-scription here. I note only the most representative in my opinion report. It is a meta-analysis published in the Journal of Clinical Oncology in 2012 and devoted to out-comes of neurosurgery for hemispheric tumors [18]. Analysis included 90 publications enrolling 8091 pa-tients. Two important aspects were considered: postop-erative neurological deficit and type of resection. Results of neurological deficit were generally expected: 3.4% (2.3—4.8%) in group of intraoperative mapping vs. 8.2% (5.7—11.4%) without mapping (significant difference). Interestingly, incidence of radical surgery was also differ-ent between groups: 75% (66—82%) in intraoperative mapping vs. 58% (48—69%) in control group. The au-thors logically explain these data: mapping is followed by more active neurosurgeon’s actions without fear of unfa-vorable functional outcomes.

Surgery for drug-resistant epilepsy

This is a classic direction for AC without any changes since its introduction. In fact, it is a special variant of functional neurosurgery where functional outcome is one of the main purposes [37—39]. Large literature data are devoted to this procedures including articles and handbooks [40]. AC technology in epilepsy surgery is similar to that in neurooncology, but the main surgical goal is area of epileptic signals generation.

Functional neurosurgical procedures

Some functional neurosurgical interventions are principally impossible in deeply anesthetized patient. Firstly, it is deployment of deep electrodes for DBS in Parkinson's disease, essential tremor, dystonia, chronic pain and epilepsy [11, 41—45]. Correct placement of electrode and selection of effective stimulation mode re-quire interaction with patient during test loads. Intraop-erative DBS testing in 34-year-old professional pianist with severe dystonia in right hand is presented in figure

(photo is given by Dr. E.M. Salova). Intraoperative play-ing on synthesizer was applied to evaluate effect of DBS and select stimulation parameters. Undoubtedly, such procedure is absolutely impossible under general anes-thesia.

Carotid arteries repair

It is one of the most interesting and at the same time controversial field for AC. Background is understandable here: ICA cross-clamping and complete blood flow ces-sation (15—30 min on the average) are followed by the risk of ipsilateral hemispheric ischemia due to inadequate collateral circulation. In case of general anesthesia these disorders may be diagnosed only by various modes of neuromonitoring including transcranial sonography, ce-rebral oximetry, scalp EEG and EP, pressure in distal carotid stump [46—49]. This approach is associated with serious cost (equipment, specialists), but more important – certain percentage of false-positive and, that is espe-cially unpleasant, false-negative results [47—49]. Rea-sonable clinical alternative to complex and not always reliable hardware neuromonitoring is simple DNM which may be easily applied in awake neurosur-gery [50—52]. Acute hemiparesis contralateral to proce-dure’s side or aphasia in right-handed people during left-sided surgery after ICA cross-clamping clearly demon-strates inadequate collateral circulation and need for ur-gent correction (temporary bypass and/or increase of blood pressure). Both approaches (general or locoregion-al anesthesia) have their supporters and opponents, and debate is still continuing now [53—61]. Certain expecta-tions were for GALA European trial comparing immedi-ate and long-term results of carotid endarterectomy un-der general or local anesthesia. First of all, cardiac (myo-cardial infarction) and cerebral (stroke) events and mor-tality were assessed. This study involved 95 centers from 24 countries and 3526 patients. Results were published in November 2008 and confirmed similar outcomes and morbidity [62]. And this is despite various reports have earlier demonstrated advantages of local anesthe-sia [63—65]. In fact, these studies justified GALA trial. You can talk a lot about the reasons for such results. Probably, someday we will understand the reasons of negative outcomes in not only this but also other large trials. However, simplicity of neuromonitoring during carotid endarterectomy under DNM and local anesthesia is absolutely obvious and does not require other special evidence despite GALA trial data.

Endovascular neurosurgery

The majority of endovascular neurosurgical proce-dures are not associated with severe pain and do not re-quire deep anesthesia. Therefore, anesthesiologist can use only consciousness-sparing sedation that is followed by certain advantages [66—69]. Obviously, DNM may be easily realized in such situation, as it was rightfully writ-ten in the above-mentioned guideline for neuromonitor-


ing: "It is obvious that endovascular neurosurgical inter-ventions under consciousness-sparing sedation may be followed by intraoperative DNM as the most effective neuromonitoring mode" [70].

Extracranial-intracranial bypass

Case report of extracranial-intracranial bypass under local anesthesia for ICA occlusion has been recently pub-lished in Burdenko Journal of Neurosurgery. Surgery was performed in advanced age patient with severe comor-bidities according to clinical indications [71]. Procedure was successful, intraoperative DNM was useful to control patient's state. It should be noted that it is not the first report. Japanese authors were first [72], but whoever was first it is obviously another field for AC.

Miscellaneous

Here I gathered information about various AC-asso-ciated neurosurgical interventions described within sin-gle case reports.

Transsphenoidal surgery for chiasmatic-sellar tumors. Unlike multitudinous ACs in neurooncology or ICAs re-pair transsphenoidal surgery of chiasmatic-sellar tumors under local anesthesia is an undeniable casuistry. The first of the known case reports of surgery for CSR tumor is quite old and mysterious. It is supposedly excision of Turkish saddle tubercle meningioma under local anes-thesia in famous Hungarian pianist C. Haskil [73] by French neurosurgeon M. David. There are no reliable detailed data about this procedure since it was performed in Marseille in 1942 and all materials are still in private archives and not available. However, successful surgery for CSR tumor under local anesthesia is fact. Patient continued her concert activity after surgery and was rec-ognized as an unsurpassed performer of Mozart's compo-sitions. She died 18 years later from domestic craniocere-bral trauma.

About 10 years ago I reviewed article of Japanese au-thors describing another interesting clinical observa-tion [74]. Chiasmatic symptoms (rapid bilateral impair-ment of visual acuity) occurred in 88-year-old patient with severe cardiac comorbidities. MRI confirmed large pituitary adenoma that was followed by transsphenoidal surgery under local anesthesia. Partial resection was scheduled and accompanied by visual tracts decompres-sion. Vision was rescued.

Emergency surgery of brain abscess. This is a very re-cent publication of Swiss colleagues describing rare clini-cal case: 39-year-old patient with complex cyanotic con-genital heart disease and brain abscess underwent suc-cessful surgery in AC fashion. This approach was pre-ferred due to advanced perioperative morbidity and mor-tality in patients with cyanotic congenital heart disease undergoing surgery under general anesthesia [75].

Intracranial aneurysms repair

The first two reports were published in 2005. There were 2 short reports in the appendix to Neurosurgery

journal. In the first one P. Chen et al. [76] described in-traoperative awakening to control visual functions after ophthalmic artery aneurysm clipping. In the second pub-lication J. Luders et al. [77] reported 3 successful AC pro-cedures for mycotic aneurysms. Clinical observation of 46-year-old patient with headache and left MCA aneu-rysm was published in Surgical Neurology International in 2013. Aneurysm was found to be unsuitable for endo-vascular occlusion and AC followed by DNM was pre-ferred due to high risk of ischemic complications. Neuro-logical status was intact during temporary MCA occlu-sion and subsequent trapping [78].

And here is one of the latest publications analyzing 30 patients. Article was published at the end of 2016 in the Journal of Neurosurgery and worthy for discussion [79]. AC approach was used in 30 patients with intracranial aneurysms for transcranial clipping. Four patients were excluded: 2 patients refused themselves, 2 of them had anesthetic risks (sleep apnea in one and difficult previous tracheal intubation in another patient). Besides conven-tional control (blood pressure, heart rate, ECG, body temperature, capnography and pulse oximetry) intraop-erative monitoring included following neuromonitoring modalities: scalp EEG, motor and sensory EP, DNM. In-traoperative monitoring of visual function was applied in another 4 patients with carotid-ophthalmic aneurysms. Neurological deficit after temporary clip deployment oc-curred in 3 patients and disappeared after its removal. Another patient had neurophysiological symptoms of ischemia and neurological deficits after permanent clip deployment. Clip reposition was not followed by regres-sion of neurological deficit and neurophysiological signs. This patient was discharged with ischemic focus. But what is interesting that in 3 other patients DNM revealed neurological deficit after temporary clipping but neuro-physiological monitoring was uneventful! Therefore, au-thors believe that cerebral aneurysms repair in AC mode

Intraoperative monitoring of deep brain stimulation.


REVIEWS

and dynamic neurological monitoring have certain ad-vantages. Neurological deficit was revealed in 3 patients despite normal data of intraoperative neuromonitoring (false-negative result). And this is 10%! Obviously, these data need to be thought over despite there were patients with unruptured aneurysms in the sample.

Spinal neurosurgery

Wake-up test in scoliosis surgery is a classic example of intraoperative consciousness recovery in spinal cord surgery. It is still "gold standard" of intraoperative neuro-logical injury diagnosing along with neurophysiological techniques [80—83]. However, I have recently met edi-tor’s very interesting commentary (M. Lund-Johansen) to one of the articles of Acta Neurochirurgica journal. This Norwegian neurosurgeon witnessed old and experi-enced neurosurgeons sought to perform spinal interven-tions under local anesthesia for DNM when he trained for spinal neurosurgery [84].

I can add from my experience that we often used in-travenous infusion of propofol for superficial sedation when local (epidural) anesthesia has been introduced into lumbar spine surgery. Patient immediately woke up and informed about his sensations and neurosurgeon’s dangerous actions in case of any traumatic manipulations (spinal compression as a rule). It was followed by signifi-cantly lower postoperative morbidity in epidural anesthe-sia group [85]. Interestingly, recent analysis of local and general anesthesia in lumbar spine surgery confirmed advantages of local anesthesia in these patients [86].

Cochlear neuroma surgery

Japanese neurosurgeons’ report has been recently published in Acta Neurochirurgica journal where out-comes of cochlear neuroma surgery under local anesthe-sia were analyzed [87]. Sample size was only 8 patients who underwent less radical tumor excision. However, functional results were better including hearing preserva-tion. Perhaps, it is only the first sign in this field.

Outpatient awake craniotomy (OAC)

This part of modern neurosurgery is still quite prob-lematic because of its unusual nature. We are talking about AC for hemispheric tumors within "one-day sur-gery"! Canadian neurosurgeon M. Bernstein is the main adherent of this direction [88], therefore we will discuss his works first of all. He had mastered AC approach be-fore he made it clinically routine in surgery for supraten-torial tumors even in cases of minimal risk of verbal dis-turbances. Further analysis of patients’ state after AC, their sensations and postoperative rehabilitation con-firmed a simple conclusion: there is no need for postop-erative hospital-stay after uneventful surgery in stable patients. So, they may be discharged to indisputably more comfortable home conditions that was confirmed in subsequent work of Bernstein and el. [89—94]. In au-thors’ opinion, natural and quite actual question about possible postoperative complications seems to be com-

pletely solvable in this approach. There were following statistical data in one of the largest sample size: 228 (36%) out of 1,003 prospectively selected patients underwent OAC. 92.8% of patients were discharged in the evening after successful procedure, 5.2% stayed in the clinic for various reasons, but 2% of discharge followed by repeated admission are the most interesting. Severe headache, nausea and vomiting were the main causes of hospitaliza-tion. One patient with intracerebral hemorrhage was hos-pitalized in 12 hours after surgery, medication was effec-tive [95]. Well, everything looks quite reasonable at first sight. However, this apparently does not correspond to realities of our country and risk of even one severe com-plication at home can easily cross out entire economic attractiveness of this approach considering possible legal consequences. Nevertheless, authors [35] insist on expe-diency of this approach if financial capacity of health care is limited in the country.

Functional disorders after neurosurgical procedures and the role of intraoperative neuromonitoring in prevention of neurological deficit de novo

Probably this is the most important problem and real explanation of advanced interest to AC at present time. Postoperative aggravation of previous neurological defi-cit or disorders de novo is one of the key problems of neu-rosurgery. This is a real problem followed by emotional stress and financial risks. Intraoperative neuromonitor-ing is one of the reasonable approaches for this serious problem. Numerous publications and even guidelines are devoted to this issue. In my opinion, "Intraoperative Neuromonitoring" by C. Loftus, J. Biler and E. Baron is one of the most successful one [96]. This and other man-uals describe various intraoperative neuromonitoring modalities aimed at early diagnosis of neuronal injury and prevention of permanent neurological deficit via changing of surgical tactics. These postoperative neuro-logical disorders may be only revealed as soon as patient woke up after general anesthesia. There are motor, sen-sory and visual evoked potentials, myography in various modes, facial nerve’s function, scalp EEG and ECoG, cerebral and jugular oximetry, transcranial sonography and pressure in vascular stump, retractive pressure and others. Currently, large number of intraoperative neuro-monitoring modalities have been developed, but let’s try to understand the essence of this process. First, let's try to formulate our wishes regarding ideal method of intraop-erative neuromonitoring. So, this method should: 1) re-flect neurons’ functional state in high risk area; 2) be simple, cheap, reliable and preferably non-invasive; 3) be resistant to effect of anesthetics, temperature changes and electrical interferences. It is easy to understand that there is no such method at present time and it is unlikely to be that it will appear in foreseeable future.

Let's try other approach. We rank intraoperative neu-romonitoring modalities according to informative value. Undoubtedly, monitoring of neurons’ functional state in


area of the greatest risk of injury should be the highest in this hierarchy. There is surrogate assessment of neuronal activity (EEG or ECoG and EP) on the 2nd place fol-lowed by secondary modalities including blood flow, ox-ygenation, metabolism, etc. It is interesting that only dy-namic neurological monitoring may be placed on top of this pyramid. There are no any other modalities compa-rably by their informative value and simplicity at the same time. However, DNM should be followed by AC approach. According to above-mentioned data, this ap-proach is possible for various neurosurgical pathology and may be accompanied by favorable outcomes.

ConclusionDirection of modern neurosurgery progress is under-

standable – maximally reduced risk of postoperative ag-

gravation of previous neurological deficit or disorders de novo. Different modes of intraoperative neuromonitoring are apparently able to solve this problem. However, ad-vanced costs are necessary while absolute guarantee is absent. AC approach followed by DNM seems to be rea-sonable alternative. Certainly, this requires changed an-esthetic concept and it is currently difficult problem due to technical and pharmacological causes. Modern anes-thesiology is able to provide comfortable intraoperative state of patient (controlled superficial sedation followed by easy awaking at the right time, no pain and other un-pleasant sensations). Perhaps, it is prospect of our devel-opment in near future.


REFERENCES1. Frost EAM. History of neuroanesthesia. In: Albin VS. (ed.). Textbook of

Neuroanesthesia. 1997. McGraw ‒ Hill Co. 1-20.2. Lubnin AYu, Salalykin VI. Neuroanesthesiology. Past, present and future.

Anest. Rean. 1999;6:41-47 (In Russ.).3. Surbeck W, Hildebrandt G, Duffau H. The evolution of brain surgery on

awake patients. Acta Neurochir. 2015;157:77-84. https://doi.org/10.1007/s00701-014-2249-84. Neufeld GR. Fires and Explosions. In: Gravenstein N, Kirby RR. (eds.).

Complications in Anesthesiology. 1996. Lippincott&Raven. Phil.:73-78.5. Salalykin VI. Local anesthesia. In: Neuroanesthesiology. M.: Meditsina.

1977;52-60. (In Russ.). 6. Penfield W. Combined regional and general anesthesia for craniotomy and

cortical exploration. Part I. Neurosurgical considerations. Int Anesth Clin. 1986;24:1-11 (reprint article 1953 year).

7. Pasquet A. Combined regional and general anesthesia for craniotomy and cortical exploration. Part II. Anesthetic considerations. Int Anesth Clin. 1986;24:12-20 ((reprint article 1953 year).

8. Girvin JP. Neurosurgical considerations and general methods for cranioto-my under local anesthesia. Int Anesth Clin. 1986;24:89-115.

9. Ojeman GA. Mapping of neuropsyhological language parameters at sur-gery. Int Anesth Clin. 1986;24:115-131.

10. Trop D. Conscious-Sedation analgesia during the neurosurgical treatment of epilepsies — Practice at the Montreal Neurological Institute. Int Anesth Clin. 1986;24:175-184.

11. Erickson KM, Cole DJ. Anesthetic considerations for awake craniotomy for epilepsy and functional neurosurgery. Anesth Clin. 2012;30:241-268.

https://doi.org/10.1016/j.anclin.2012.05.00212. Girvin JP. Resection of intracranial lesions under local anesthesia. Int Anes-

th Clin. 1986;24:133-155.13. Berger MS, Ojeman GA. Intraoperative brain mapping techniques in neu-

ro-oncology. Stereotact Funct Neurosurg. 1992;58:153-161.14. Berger MS. Lesions in functional («eloquent») cortex and subcortical white

matter. Clin Neurosurg. 1994;41:444-463.15. Ojemann GA, Creutzfeldt O, Lettich E, Haglund MM. Neuronal activity in

human lateral temporal cortex related to short-term verbal memory, naming and reading. Brain. 1988;111:1383-1403.

16. Ojeman G, Ojeman J, Lettich E, Berger M. Cortical language organization in left, dominant hemisphere. An electrical stimulation mapping investiga-tion in 117 patients. J Neurosurg. 1989;71:316-326.

17. July J, Manninen P, Lai J, Yao Z, Bernstein M. The history of awake crani-otomy for brain tumor and its spread into Asia. Surg Neurol. 2009;71:621-625. https://doi.org/10.1016/J.Surneu.2007.12.022

18. De Witt Hammer PC, Robles SG, Zwinderman AH. Impact of intraopera-tive stimulation brain mapping on glioma surgery outcome: a metaanalysis. J Clin Oncol. 2012;30:2559-2565. https://doi.org/10.1200/JCO.2011.38.4818

19. Manchella S, Khurana VG, Duke D, Brussel T, French J, Zuccherelli L. The experience of patients undergoing awake craniotomy for intracranial masses: expectations, recall, satisfaction and functional outcome. Br J Neu-rosurg. 2011;25:391-400. https://doi.org/10.3190/02688697.2011.568640

20. Bilotta F, Rosa G. «Anesthesia» for awake neurosurgery. Curr Opin Anaesth. 2009;22:560-565. https://doi.org/10.1097/ACO.0b013e3283302339

21. Joy Khu K, Doglietto F, Radovanovic I, Taleb F, Mendelsohn D, Zadeh G, Bernstein M. Patients’ perceptions of awake and outpatient craniotomy for brain tumor: a qualitative study. J Neurosurg. 2010;112:1056-1060.

https://doi.org/10.3171/2009.6.JNS0971622. Rughani AI, Rintel T, Desai R, Cushing DA, Florman JE. Development of

a safe and pragmatic awake craniotomy program at Maine medical center. J Neurosurg Anesth. 2011;23:18-24.

https://doi.org/10.1097/ANA.0b013e3181ebf05023. Seemann M, Zech N, Graf E, Hansen E. Anasthesiologisches management

zur Wachkraniotomie. Anaesthesist. 2014;64:128-136. https://doi.org/10.1007/s00101-014-2396-624. Milian M, Tatagiba M, Feigl GC. Patient response to awake craniotomy—a

summary overview. Acta Neurochir. 2014;156:1063-1070. https://doi.org/10.1007/s00701-014-2038-425. Harvey-Jumper SL, Li J, Lau D, Molinaro AM, Perry DW, Meng L, Berg-

er MS. Awake craniotomy to maximize glioma resection: methods and tech-nical nuances over a 27-year period. J Neurosurg. 2015;123:325-339.

https://doi.org/10.3171/2014.10.JNS14152026. Kulikov AS, Sel’kov DA, Kobyakov GL, Lubnin AYu. Awake craniotomy:

in search of optimal sedation. Anest Rean. 2015;4:4-8. (In Russ.).27. Kobyakov GL, Lubnin AYu, Kulikov AS. Awake craniotomy. Journal Vo-

prosy Neirochir. 2016;1:107-116. (In Russ.). https://doi.org/10.17116/neu-ro2016801107-116

28. Kulikov AS, Kobyakov GL, Gavrilov AG, Lubnin AYu. Awake craniotomy: analysis of complicated observations. Journal Voprosy Neirochir. 2015;6:15-21. (In Russ.).

https://doi.org/10.17116/neuro201579615-2129. Meng L, McDonagh DL, Berger MS, Gelb AW. Anesthesia for awake cra-

niotomy: a how-to guide for occasional practitioner. Can J Anesth. 2017;64:517-529. https://doi.org.10.1007/s12630-017-0840-1

30. Meng L, Berger MS, Gelb AW. The potential benefits of awake craniotomy for brain tumor resection: An anesthesiologist’s perspective. J Neurosurg Anesth. 2015;27:310-317. https://doi.org/10.1097/ANA.0000000000000179

31. Nossek E, Matott I, Shahar T, Barzilai O, Rapoport Y, Gonen T, Sela G, Korn A, Hayat D, Ram Z. Failed awake craniotomy: a retrospective analysis in 424 patients undergoing craniotomy for brain tumor. J Neurosurg. 2013;118:243-249. https://doi.org/10.3171/2012.10JNS12511

32. Nossek E, Matot I, Shahar T, et al. Intraoperative seizures during awake craniotomy: Incidence and consequences: analysis of 477 patients. Neuro-surgery. 2013;73:135-140.


REVIEWS

https://doi.org/10.1227/01.neu.0000429847.91707.9733. Brown T, Shah AH, Bregy A, Shah NH, Thambuswamy M, Barbarite E,

Fuhrman T, Komotar RJ. Awake craniotomy for brain tumor resection: The rule rather than the exception? J Neurosurg Anesth. 2013;25:240-247.

https://doi.org/10.1097/ANA.0b13e318290c23034. Potters JW, Klimek M. Awake craniotomy: improving the patient’s experi-

ence. Curr Opin Anesth. 2015;28:511-516. https://doi.org/10.1097/ACO.000000000000023135. Ibrahim GM, Bernstein M. Awake craniotomy for supratentorial gliomas:

why, when and how? CNS Oncol. 2012;1:71-83. https://doi.org/10.2217/cbs.12.136. Szelenyi A, Bello L, Duffau H, Fava E, Feigl GC, Galanda M, Neuloh G,

Signorelli F, Sala F. Workgroup for intraoperative management in low-grade glioma surgery within the European low-grade glioma network. Intra-operative electrical stimulation in awake craniotomy: methodological as-pects of current practice. Neurosurg Focus. 2010;28:E7(1-8).

https://doi.org/10.3171/2009.12.FOCUS0923737. Kulikov AS, Stepanenko AYu, Lubnin AYu. Surgery of epilepsy — what’s

need from anesthesiologist? Anest Rean. 2011;4:4-10 (In Russ.).38. Archer DP, McCenna JM, Morin L, Ravussin P. Conscious sedation during

craniotomy for intractable epilepsy: a review of 354 consecutive cases. Can J Anaesth. 1988;35: 244-338.

39. Trop D, Oliver A, Dubeau F, Jones-Gotman M. Seizure surgery. Anesthet-ic, neurologic, neurosurgical and neurobehavioral considerations. In: Al-bin MS (ed.). Textbook of Neuroanesthesia with neurosurgical and neurosci-ence perspectives. 1997. McGraw‒Hill Co. NY etc.:643-696.

40. Miller JW, Silbergeld DL. (eds). Epilepsy surgery: Principles and controver-sies. Taylor & Francis. NY — London. 2006;838.

41. Gildenberg P, Neal MB, Distefano I. Stereotactic surgery and ablative proce-dures for pain. In: Albin MS (ed.). Textbook of Neuroanesthesia with neurosurgi-cal and neuroscience perspectives. 1997. McGraw — Hill Co. NY etc.:697-706.

42. Venkatraghavan L, Manninen P. Anesthesia for deep brain stimulation. Curr Opin Anesthesiol. 2011;24:495-499.

https://doi.org/10.1097/ACO.0b13e32834a894c43. Poon CCM, Irwin MG. Anaesthesia for deep brain stimulation and in pa-

tients with implanted neurostimulator devices. Br J Anaesth. 2009;103:152-165. https://doi.org/10.1093/bja/aep179

44. Mulroy E, Robertson N, MacDonald L, Bok A, Simpson M. Patients’ peri-operative experience of awake deep-brain stimulation for Parkinson disease. World Neurosurg. 2017;105:526-528.

https://doi.org/10.1016/j.wneu.2017.05.13245. Chakrabati R, Ghazanwy M, Tewari A. Anesthetic challenges for deep brain

stimulation: a systematic approach. N Am J Med Sci. 2014;6:359-369. https://doi.org/10.4103/1947-2714.13928146. Shmigelsky AV, Usachev DYu, Lukshin VA, Lubnin AYu. Multimodal neu-

romonitoring in early diagnosis of cerebral ischemia during reconstructive carotid artery surgery. Anest Ran. 2008;2:16-22. (In Russ.).

47. Horsch S, Ktenidis K. (eds.). Perioperative monitoring in carotid surgery. Methods, limits and results. Long-term results in carotid surgery. Springer & Steinkopff. 1998;198.

48. Bookland M, Loftus CM. Monitoring considerations during extracranial carotid reconstruction. In: Loftus CM, Biller J, Baron EM. (eds.). Intraop-erative Neuromonitoring. 2015. McGraw Hill. NY etc.:11-22.

49. Li J, Shalabi A, Ji F, Meng L. Monitoring cerebral ischemia during carotid endarterectomy. J Biomed Res. 2016;31.

https://doi.org/10.7555/JBR.31.2015017150. Hickey R. Carotid artery disease. Neuroanesthetic management. In: Al-

bin MS (ed.). Textbook of Neuroanesthesia with neurosurgical and neuro-science perspectives. 1997. McGraw—Hill Co. NY etc.;913-930.

51. Shmigelsky AV, Lubnin AYu. Anesthesia for carotid endarterectomy. Anest Rean. 2008;2:47-57. (In Russ.).

52. Stoneham MD, Stamou D, Mason J. Regional anaesthesia for carotid end-arterectomy. Br J Anaesth. 2015;114:372-383.

https://doi.org/10.1093/baj/aen30453. Markovic D, Vlatkovic G, Sindjelic R, Markovic D, Ladjevic N, Kalezic N.

Cervical plexus block versus general anesthesia in carotid surgery: single center experience. Arch Med Sci. 2012;8:1035-1040.

https://doi.org/10.5114/aoms.2012.3241154. Jaques F, Elkouri S, Bracco D, Hemmerling T, Daniel V, Beaudoin N, Bru-

neau L, Blair JF. Regional anesthesia for carotid surgery: less intraoperative hypotension and vasopressor requirement. Ann Vasc Surg. 2009;23:324-329.

https://doi.org/10.1016/javsg.2008.05.1555. Lee J, Up Huh, Song S, Chung SW, Sung SM, Cho HJ. Regional anesthe-

sia with dexmedetomedine infusion: A feasible method for the awake test during carotid endarterectomy. Ann Vasc Dis. 2016;9:295-299.

https://doi.org/10.3400/avd.0a.16-0004956. Liu J, Martinez-Wilson H, Neuman MD, Elkassabany N, Ochroch EA.

Outcome of carotid endarterectomy after regional anesthesia versus general anesthesia — a retrospective study using two independent data bases. Trans Periop Pain Med. 2014;1:14-21. PMC 4444227

57. Knappich C, Kuehnl A, Tsautibas P, Schmid S, Breitkreuz T, Kallmayer M, Zimmermann A, Eckstein HH. Intraoperative completion studies local an-esthesia, and antiplatelet medication are associated with lower risk in ca-rotid endarterectomy. Stroke. 2017;48:955-962.

https://doi.org/10.1161/STROKEAHA.116.01486958. Kfoury E, Dort J, Trickey A, Crosby M, Donovan J, Hashemi H, Mukher-

jee D. Carotid endarterectomy under local and/or regional anesthesia has less risk of myocardial infarction compared to general anesthesia: an analy-sis of national surgical quality improvement program data base. Vascular. 2015;23:113-119. https://doi.org/10.1177/1708538114537489

59. Codergreen P, Swiatek F, Nielsen HB. Local anaesthesia for carotid endar-terectomy. Pro: protect the brain. Eur J Anaesth. 2016;33:236-237.

https://doi.org/10.1097/EJA.000000000000037060. Unic-Stojanovic D, Jovic M. Local anaesthesia for cfrotid endarterectomy.

Con: decrease the stress for all. Eur J Anaesth. 2016;33:238-240. https://doi.org/10.1097/EJA.000000000000037761. Licker M. Regional or general anaesthesia for carotid endarterectomy. Does

it matter? Eur J Anaesth. 2016;33:236-237. https://doi.org/10.1097/EJA.000000000000037662. GALA trial collaboration group. General anaesthesia versus local anaesthe-

sia for carotid surgery (GALA): a multicentre, randomized controlled trial. Lancet. 2008;372:2132-2142. https://doi.org/10.1016/S0140-6736(08)61699

63. Sternbach Y, Jllig K, Zhang R, Shortell CK, Rhodes JM, Davies MG, Lyden SP, Green RM. Hemodynamic benefits of regional ane-sthesia for carotid endarterectomy. J Vasc Surg. 2002;35:333-339.

64. Hakl M, Michalek P, Seveik P, Pavlíková J, Stern M. Regional anaesthesia for carotid endarterectomy: an audit over 10 years. Br J Anaesth. 2007;99:415-420. https://doi.org/10.1093/bja/aen171

65. McCleary AJ, Maritati G, Gough MJ. Carotid endarterectomy; Local or General Anaesthesia? Eur J Vasc Endovasc Surg. 2001;22:1-12.

https://doi.org/10.1053/ejvs.2001.138266. Young WL, Pile-Spellman J. Intervention neuroradiology. In: Al-

bin MS (ed.). Textbook of Neuroanesthesia with neurosurgical and neuro-science perspectives. 1997. McGraw — Hill Co. NY etc.;807-843.

67. Lee CZ, Yong WL. Anesthesia for endovascular neurosurgery and interven-tional neuroradiology. Anesth Clin. 2012;30:127-147.

https://doi.org/10.1016/j.anclin.2012.05.00968. Lee CZ, Gelb AW. Anesthesia management for endovascular treatment.

Curr Opin Anaesthesiol. 2014;27:484-488. https://doi.org/10.1097/ACO.000000000000010369. Brinjikji W, Murad MH, Rabinstein AA, Cloft HJ, Lanzino G, Kallmes DF.

Conscious sedation versus general anesthesia during endovascular acute ischemic stroke treatment: a systematic review and metaanalysis. Am J Neu-roradiol. 2015;36:525-529. https://doi.org/10.3174/ajnr.A4159

70. Taussky P, Tauk RG, Miller DA, Hanel RA. Intraoperative monitoring strat-egies for endovascular procedures. In: Loftus CM, Biller J, Baron EM. (eds.). Intraoperative Neuromonitoring. 2015. McGraw Hill. NY etc.;141-147.

71. Usachev DYu, Lukshin VA, Shmigesky AV. Extra-intacranial microsurgery bypass grafting under regional anesthesia. Klinical observation and litera-ture review. Journal Voprosy Neirohir. 2017;2:88-94. https://doi.org/10.3171/201781288-94 (In Russ.).

72. Yasuhiko K, Yamashita K, Kokuzawa J, Kanou K, Tsujimoto M. Superficial temporal artery — middle cerebral artery bypass using local anesthesia and a sedative without endotracheal general anesthesia. J Neurosurg. 2012;117:288-294. https://doi.org/10.3171/2012.4.JNS111958

73. Gasenzer ER, Ranat A, Neugebauer EAM. First report of awake cranioto-my of a famous musician: Suprasellar tumor surgery of pianist Clara Haskil in 1942. J Neurol Surg. 2016;78:260-268.

https://doi.org/10.1055/s-0036-159789574. Arai T, Okada K, Yamaguchi T, Sakuma A. Endonasal transsphenoidal sur-

gery under local anesthesia for elderly patient with pituitary tumor: case re-port. No Shinkei Geka. 2000;28:991-995.


75. D’Antico C, Hofer A, Fassl J, Tobler D, Zumofen D, Steiner LA, Goet-tel N. Case report: Emergency awake craniotomy for cerebral abscess in a patient with unrepaired cyanotic congenital heart disease. F1000. Research. 2017;5:2521. https://doi.org/10.12688/f1000research.9722.2

76. Chen P, Dunn IF, Aglio LS. Intraoperative awakening for vision examina-tion during ophthalmic artery aneurysm clipping: technical case report. Neurosurgery. 2005;56(Suppl. 2):E440.

77. Luders JC, Steinmetz MP, Mayberg MR. Awake craniotomy for microsur-gical obliteration of mycotic aneurysms: technical report of three cases. Neurosurgery. 2005;56(Suppl.):E201.

78. Passacantilli E, Anichini G, Cannizzaro D, Fusco F, Pedace F, Lenzi J., Santoro A. Awake craniotomy for trapping a giant fusiform aneurysm of the middle cerebral artery. Surg Neurol Internat. 2013;4:39.

https://doi.org/10.4103/2152-7806.10965279. Abdulrauf SI, Vuong P, Patel R. «Awake» clipping of cerebral aneurysms:

report of initial series. J Neurosurg. 2016 Oct.21;1-8. https://doi.org/10.3171/2015.12.JNS15214080. Hagberg C, Welch WC, Bowman-Howard M. Anesthesia and surgery for

spinal cord procedures. In: Albin MS (ed.). Textbook of Neuroanesthesia with neurosurgical and neuroscience perspectives. 1997. McGraw—Hill Co. NY etc.;1039-1081.

81. Gambrall M.A. Anesthetic implications for surgical correction of scoliosis. AANA J. 2007;75:277-285.

82. Malhorta NR, Shaffrey CI. Intraoperative electrophysiological monitoring in spine surgery. Spine. 2010;35:2167-2179.

https://doi.org/10.1097/BRS.0d013e3181f60d083. Strike SA, Hassanzadeh H, Jain A. Intraoperative neuromonitoring in pe-

diatric and adult spine deformity surgery. Clin Spine Surg. 2016. May 27. https://doi.org/10.1097/BSD.000000000000038884. Johansen M. Awake craniotomy for vestibular schwannoma. Invited edito-

rial. Acta Neurochir. 2017. https://doi.org/10.1007/s00701-017-3238-585. Solenkova AV, Lubnin AYu, Tenedieva VD, et al. Epidural anesthesia in

surgery of spinal cord and vertebral column. Part I. Comparative analysis of efficacy of anesthetic protection under epidural anesthesia and neurolept-analgesia in neurosurgical intervention on spinal cord and vertebral column. Anest Rean. 2000;4:27-32. (In Russ.).

86. De Rojas JO, Syre P, Welch WC. Regional anesthesia versus general anes-thesia for surgery on the lumbar spine: a review of the modern literature. Clin Neurol Neurosurg. 2014;119:39-43.

https://doi.org/10.1016/j.clineuro.2014.01.01687. Shinoura N, Midorikawa A, Hiromitsu K. Preservation of hearing following

awake surgery via the retrosigmoid approach for vestibular schwannomas in 8 consecutive patients. Acta Neurochir. 2017 July 3.

https://doi.org/10.1007/s00701-017-3235-888. Taylor MD, Bernstein M. Awake craniotomy with brain mapping as a rou-

tine surgical approach to treating patients with supratentorial intraaxial tu-mors: a prospective trial of 200 patients. J Neurosurg. 1999;90:35-41.

89. Bernstein M. Outpatient craniotomy for brain tumor: a pilot feasibility study in 46 patients. Can J Neurol Sci. 2001;28:120-124.

90. Bernstein M, Bampoe J. Surgical innovation or surgical evolution: an ethi-cal and practical guide to handling novel neurosurgical procedures. J Neu-rosurg. 2004;100:2-7. https://doi.org/10.3171/jns.2004.100.1.0002

91. Boulton M, Bernstein M. Outpatient brain tumor surgery: innovation in surgical neurooncology. J Neurosurg. 2008;108:649-654.

https://doi.org/10.3171/jns/2008/108/4/064992. Carrabba G, Venkatraghavan L, Bernstein M. Day surgery awake cranioto-

my for removing brain tumors: technical note describing a simple protocol. Minim Invasive Neurosurg. 2008;51:208-210.

https://doi.org/10.1055/s-2008-107313293. Khu KJ, Doglietto F, Radovanovic I, Taleb F, Mendelsohn D, Zadeh G,

Bernstein M. Patient’s perceptions of awake and outpatient craniotomy for brain tumor: a qualitative study. J Neurosurg. 2010;112:1056-1060.

https://doi.org/10.3171/2009.6.JNS0971694. Kirsch B, Bernstein M. Ethical challenges with awake craniotomy for tu-

mor. Can J Neurol Sci. 2012;39:78-82.95. Purzner T, Purzner J, Massicotte EM, Bernstein M. Outpatient brain tumor

surgery and spinal decompression: a prospective study of 1003 patients. Neurosurgery. 2011;69:119-126.

https://doi.org/10.1227/NEU.0b013e318215a27096. Loftus CM, Biller J, Baron EM. (eds.). Intraoperative Neuromonitoring.

2015. McGraw Hill. NY etc.;565.

Received: 20.06.17

Commentary

There is no need to look at the title of the article to under-stand who is author of this distinctive and recognizable report. Reading the article you understand that only specialist with great experience and encyclopaedic knowledge about certain is-sue is able to recognize awake craniotomy as anesthesiological top in neurosurgery. So, we have an article in scientific essay genre here with the elements of literature analysis and emotions of one's own experience. How to evaluate article’s content? Principal difference of anesthesiology in neurosurgery from general anesthetic practice is peculiarity of brain surgical trau-ma. Bottom-up proprioceptive pain impulses as the main object of blockade in extracerebral procedures is absent in neurosur-gery. Consequently, there is also no need to blockade certain brain departments for prevention of vegetative responses. Main risk is intraoperative injury of associative and motor zones that is followed by focal pyramidal insufficiency (top-down). Intra-operative cerebral injury may be caused by secondary microvas-

cular and manipulative lesion. Brain as an object of anesthesia is characterized by minimum of shock-responsible receptors and needs for only hypnogenic and amnestic medication. In other words, bottom-up proprioception blockade in neurosur-gery may be realized by local anesthesia alone that is postulated by the author. Advantage of AC approach is also evident, since neurosurgical zone of interest may be accurately determined by focal symptoms. The method seems to be even more advisable if we consider that even high-quality anesthetics are associated with risk of cerebral toxic effects and persistent cognitive im-pairment especially in children and elderly patients. What are the counter-arguments?

1. Emergency neurosurgery (craniocerebral trauma and strokes) cannot be limited by local anesthesia due to uncertain zone of cerebral injury followed by mosaic clinical picture and need to create "metabolic quiet" for prevention of sympathetic "storm".


REVIEWS

2. Multimodal neuromonitoring allows to control any dis-orders on-line and to predict certain delayed violations. So, it is mandatory component of intraoperative management if con-sciousness is undesirable intraoperatively, for example in intra-cerebral aneurysms repair.

3. I would refer AC to the highest anesthetic level in neuro-surgery, in other words we can hardly talk about wide introduc-tion of this approach among anaesthesiologists of general prac-tice.

A.Yu. Lubnin’s article as polemical and very interesting re-port completely corresponds to journal’s profile and status.

A.A. Belkin (Yekaterinburg, Russia)


There is anatomical complex related to internal and external skull base. It is under surgical interest of neuro-surgery and rhinosurgery and may be designated as "mid-line structures of anterior skull base". Anterior cranial fossa (ACF) base, anterior parasellar area, ethmoid, pa-ranasal sinuses and optic canal should be mentioned among them.

Surgery for anterior skull base disease is based on CT and MRI data (relationship of tumor with bones, dura mater, brain tissue, orbit, great vessels, cranial nerves, etc.), as well as evaluation of individual anatomical fea-tures of patient including structural variability and age-related anatomic differences of skull base. This is neces-sary to determine optimal surgical access, staging and type of resection and to predict possible skull base inju-ries and their repair.

The purpose of this work is literature data analysis about individual variability and age-related anatomic features of midline structures of anterior skull base (orig-inal images from the authors’ personal archive were used for illustrations).

Variability of anterior skull base anatomy

Craniobasal surgery requires good knowledge of in-tra- and extracranial skull base anatomy [1]. Variability is anatomic peculiarity of middle structures of anterior and middle skull base. It is predominantly actual for ethmoid since ethmoid labyrinths are characterized by difference cellular structure while asymmetry is norm [2]. We are also talking about variable anatomy of key structures such as olfactory fossa, ethmoid arteries, etc. Variable anato-my of midline skull base structures throughout frontal to sphenoid sinus is discussed in the article.

Frontal sinus. Frontal sinus growth begins at the 13th month of life and lasts up to 12 years. After that horizon-

*e-mail: [email protected]© D.A. Gol’bin, V.A. Cherekaev, 2018


Variability and Age-Related Anatomic Features of Midline Structures of Anterior Skull Base D.A. GOL’BIN*, V.A. CHEREKAEV


The article presents the literature data on the structural variability and age-related features of the midline anatomical structures of the anterior skull base (frontal sinus, ethmoid bone, anterior parasellar region, and medial orbital wall). This is the area of surgical interests of neurosurgeons and rhinosurgeons. The study objective is to analyze the literature data on the individual variability and age-related anatomy of these structures. The work is illustrated with original images from the authors’ personal archive. The individual anatomical features of eloquent structures in the surgical area (structures within the surgical corridor, key anatomical landmarks, optic tract, internal carotid and ethmoidal arteries, etc.) should be considered in planning surgery in patients of all age groups because they can limit the view and the amount of safe manipulations or increase the risk of complications. The presented data may be useful for neurosurgeons and otolaryngologists whose surgical interests are focused on the midline structures of the anterior skull base.

Keywords: skull base surgery, variability, age-related features, ethmoid bone, ethmoidal arteries, parasellar region.

tal enlargement occurs in girls and vertical in boys [3]. Moreover, length and width of left frontal sinus usually exceed those of right one due to unexplained causes [4]. Frontal sinuses are finally formed by 18 years [5]. Left frontal sinus is more developed compared with right one in 57.4% of women and in 54.05% of men. Anteroposte-rior dimension is 12.019±3.07 and 10.16±2.12 mm in men and women, respectively; height is 31.72±6.479 mm in men and 28.57±7.36 mm in women; mean width 56.33±13.43 mm in men and 51.05±17.74 mm in women; mean volume is 129.96±82.06 mL in men and 80.28±60.81 mL in women [3]. Transfrontal approach followed by intracranial and extracranial structures expo-sure is advisable in well-developed sinuses (height over 3 cm) [6] (Fig. 1).

Ethmoid bone. Ethmoid bone is sagittally formed by ethmoid lamina, crista galli above and perpendicular lamina below it. Ethmoid labyrinths are placed from each side lateral to ethmoid lamina. External wall of the laby-rinth is involved into structure of posterior two-thirds of medial orbital wall [2].

Crista galli is ethmoid bone’s structure with the larg-est thickness and pyramidal shape. It is necessary for processus falciformis major attachment, there is dura mater invagination between two folds bilateral to crista galli. Mean vertical dimension of size of crista galli is 12.7±2.4 mm (range 8.8—16.8), craniocaudal dimen-sion — 12.9±2.5 mm (range 9.1—16.3) [1]. Crista galli pneumatization occurs in 13% of cases [7]. Foramen ce-cum consisting of dura mater processus is in front to cris-ta galli, it is open in 1.4% of cases [8].

Olfactory fossa is formed by ethmoid lamina and its lateral lamella as extension of basal lamina of middle na-sal concha and its height determines olfactory fossa’s depth [9]. In embryogenesis lateral lamella arises as inde-


REVIEWS

pendent part of ethmoid bone and begins to be ossified after birth followed by final development by 10 years [10]. This structure is the thinnest in ACF [11]. In 1962 P. Ke-ros analyzed 450 observations and identified 3 main types of olfactory fossa depending on lateral lamella’s height: type I — 1—3 mm, type II — 4—7 mm and type III — 8—16 mm [12] (Fig. 2). Their incidence in popula-tion is 30%, 49% and 21%, respectively [8].

Ethmoidal labyrinths roof separates ethmoid cells from ACF. It is part of frontal bone with slant posteriorly under 15 ° on the average [13]. Convexity of ethmoidal labyrinth roof is more pronounced at the level of anterior ethmoidal cells; dehiscences are often here (areas with-out bone tissue) [2].

Anterior and posterior ethmoidal arteries are one of the key anatomical landmarks within ACF base from the side of both ethmoidal labyrinths and orbits. Both ethmoidal arteries originate from ophthalmic artery within orbit and pass between superior oblique and internal rectus mus-cles along with self-titled nerves. Anterior ethmoidal ar-tery (AEA) is larger than posterior artery (PEA). Arteries enter anterior and posterior ethmoidal orifices and eth-moidal labyrinth therefore. AEA is almost horizontal here and passes close to crista galli at the level of anterior margin of ethmoidal lamina. There are two variants of AEA course through the labyrinth: outside the labyrinth roof in 43% of cases on the right and in 49% on the left, passage through skull base in 57% and 51%, respective-ly [9] (Fig. 3). PEA reaches lateral margin of ethmoidal lamina and passes through bone channel, then it is be-hind ethmoidal lamina and anterior to optic nerve before entering cranial cavity. Mean distance from nasal septum to left AEA is 64.81±3.83 mm in both genders (range 60.37—71.86), 63.45±2.27 mm (range 60.37—66.06) in women and 66.17±4.81 mm (range 60.94—71.86) in men. The same values on the right are 64.18±3.74 mm (range 55.69—68.94), 62.95±4.38 mm (range 55.69—68.94) and 65,41±2.92 mm (range 61.51—68.44), respectively. Mean distance from nasal septum to left PEA is

71.01±3.99 mm in both genders (range 65.83—77.57 mm), 68.48±2.12 mm (range 65.83—71.66) in women and 73.53±3.94 mm (range 69.02—77.57) in men. The same values on the right are 72.27±3.97 mm (range 66.39—77.79), 69.15±2.24 mm (range 66.39—72.58) and 75,39±2.47 mm (range 72.05—77.79), respectively [13].

Anterior and posterior ethmoidal orifices in orbit’s me-dial wall are also characterized by variable topography. Intraorbital ligation of ethmoidal arteries has been suc-cessfully applied in practice [14]. There are anterior and posterior ethmoidal orifices as a rule, but additional ones are also found. Frontoethmoidal suture is important ana-tomical landmark to determine ethmoidal orifices. Ante-rior ethmoidal orifice is placed within suture in 77.7% of cases, outside — in 23.3% (mean distance 0.5±2.1 mm); posterior ethmoidal orifice is located within suture in 77.25% of observations, outside – in 4.25%. Additional orifices are revealed in 25.5% of cases and placed outside the suture in all cases. Ethmoidal orifices are the weakest points of orbit’s medial wall with the least mechanical re-sistance [15]. Anterior lacrimal crest is permanent ana-tomical reference point in transorbital approach to eth-moidal arteries. Distance from anterior lacrimal crest to anterior ethmoidal orifice is 27.6±2.8 mm (range 21.6—25.1), to posterior ethmoidal orifice — 36.6±4 mm (range 24.1—46,1). Mean distance between anterior and posterior ethmoidal orifices is 10.6±3.3 mm (range 4.3—19.3). Posterior ethmoidal orifice close to optic ca-nal should be taken into account: distance between them is 7.4±2.9 mm (range 2.4—17.6). Finally, distance from anterior lacrimal crest to optic canal is 41.4±3.8 mm (range 32.9—60.8) [16].

Anterior parasellar area. These are some key anatom-ical intracranial structures (Fig. 4a).

Sphenoid bone area is located between small wings. Its anterior margin reaches ethmoidal lamina and fronto-ethmoidal suture. Mean anteroposterior dimension of sphenoid area is 17.5 mm (range 15.2—21.3) [17]. Poste-rior margin borders with prechiasmatic sulcus and ante-

Fig. 1. Hypertrophic frontal sinus. There is a pronounced pneumatization of orbits’ upper walls and frontal bone.


rior clinoid process on both sides. These two structures form superior wall of optic canal [13].

Optic canal is located lateral to sphenoid bone, below to small wing and medial to optic strut [18]. Optic nerves and ophthalmic arteries pass through the canal. Mean distance between both canals at the level of inlets is 15.13±1.69 mm (range 11.62—17.82), 13.82±1.28 mm (range 11.62—16.37) in women and 16.44±0.79 mm (range 15.20—17.82) in men. Distance between intracra-nial orifices of optic canals is 15.13±1.69 mm (range 12.37—17.82) [13]. Optic strut pneumatization during endoscopic transsphenoidal surgery is followed by lateral opticocarotid recess exposure as a key reference point to detect projection of optic nerve and internal carotid ar-tery (Fig. 5). K. Abhinav et al. [19] studied relationship between these two structures: distance from posterome-dial edge of lateral opticocarotid recess to intracranial orifice of optic canal is 0.65±0.75 mm, distance between anterolateral edge of lateral opticocarotid recess and in-traocular orifice of the canal – 0.40±0.50 mm. Length of optic canal is up to 6.05±0.72 mm, distance between ophthalmic artery ostium and proximal optic ca-nal — 2.50±1.24 mm, length of falciform liga-ment — 1.70±0.47 mm. Optic nerve and internal carotid artery have no bone wall from the side of sphenoid sinus in some persons [20]. Ophthalmic artery is located infe-rior and medial to optic nerve at the entrance to optic canal, then nerve is circumflexed laterally. This artery oc-

cupies inferolateral position in 81% of cases or placed un-der the nerve in 19% of cases [21]. It must be taken into account in transsphenoidal approach to optic nerve.

Prechiasmatic sulcus separates sphenoid bone from sella turcica. Its width is the distance between postero-medial edges of optic strut on both sides (mean 19.3 mm, range 14—25); length — distance from the limb to sella turcica tubercle (7.45 mm, range 5—10), mean angle be-tween sphenoid bone plane and the line between limb and sella turcica tubercle is 31° (range 5—64°) [ 22] (Fig. 4b).

Optic chiasm also has variable structure. In 70% of cases it is located above sellar diaphragm, anterior and posterior placement is found in 15% and 15% of cases, respectively [23].

Sphenoid sinus. Sphenoid sinus pneumatization visi-ble at the 6th month of life begins within apertures and spreads in inferior, posterior and lateral directions [24, 25]. Complete pneumatization of recesses continues after puberty period. There are 3 types of pneumatization by Hamberger classification: sellar (86%), presellar (11%) and conchal (3%) [26]. A. Rahmati et al. [27] report postsellar type additionally (Fig. 6).

Superior wall of sphenoid sinus is formed by sphe-noid bone area anteriorly and sella turcica posteriorly. Sellar type of sinus is characterized by pneumatization anteriorly under sphenoid bone area. The thinnest sinus

Fig. 2. Variability of olfactory fossa. * — Keros type I (olfactory fossa’s depth 13 mm); ** — type II (depth 4—7 mm); *** — type III (depth 8—16 mm). Olfactory fossa’s depth is determined by lateral lamella’s height of ethmoidal lamina (arrows).

Fig. 3. Variability of anterior ethmoidal artery course through ethmoidal labyrinth. Left: arterial channels are located directly in labyrinth’s roof. Right: channels are outside the labyrinth’s roof.


REVIEWS

wall is within anterior sella turcica and sphenoid bone area [24].

Pneumatization of left-sided anterior clinoid process is observed in 10.7% of cases, right-sided — 7.8%, bilat-eral pneumatization — in 14.6% [27] (Fig. 7). If small sphenoid wing is medially pneumatized, optic nerve is surrounded by sphenoid sinus. However, optic nerve po-sition does not depend on sphenoid sinus pneumatiza-tion grade [17]. There was a single report of optic nerve course through inferior wall of sphenoid sinus [28]. Mean

distance between optic canals at the level of anterior edge of sphenoid bone area is 15.6 mm (range 14.7—16.2), at the level of medial edge of canals’ intracranial aper-tures — 13.7 mm (range 12.9—14.2) [17].

Special variant of sphenoid sinus structure is ob-served in presence of Onodi cells (sphenoethmoidal cells) which occupy upper position. In this situation approach to optic canals is through Onodi cell rather sphenoid si-nus [20, 29, 30]. Sphenoethmoidal air cells are found in 8% of cases [31] (Fig. 8).

General feature of midline structures of anterior and middle skull base is maximum contact of intracranial ele-ments with pneumatic cavities (frontal sinus, nasal cavity, ethmoidal cells, sphenoid sinus). "Pneumosinus dilat-ans" phenomenon should be considered. It is abnormally enlarged paranasal sinuses without local bone destruc-tion, hyperostosis or mucosal thickening. "Pneumosinus dilatans" is characteristic of ACF meningiomas including small ones [32] (Fig. 9).

Age-related features

Smaller anatomical structures, incomplete pneuma-tization of paranasal sinuses, incomplete development of dental system and other aspects should be considered for skull base surgery in children and adolescents [33]. Nor-mal craniofacial development includes growth, configu-ration changes, ossification and pneumatization of bone structures.

Ossification. Interpretation of skull base imaging da-ta may be difficult in children since most of bone struc-tures are in cartilage model stage and ossification occurs in numerous points. Ossification is poorly expressed at birth, but it quickly develops within the first 6 months. Anterior skull base ossification rate is greatly variable, but this process is completed in 100% by 3 years 10 months.

а b

Fig. 4. Anterior parts of internal skull base.a — anatomical marks of anterior internal skull base (GW – great wing of sphenoid bone, STT – sella turcica tubercle, SOF – superior orbital fissure, OP — orbital process of frontal bone, OC — optic canal, SBA – sphenoid bone area, ELR – ethmoidal labyrinth roof, L — limb, SW — small wing of sphenoid bone, OF — olfactory fossa, PG – prechiasmatic groove, CG – crista galli, ACP – anterior clinoid process, FC – foramen cecum, * — optic strut; b — α angle between sphenoid bone area plane and line drawn through the limb (1) and sella turcica tubercle (2). Explanations in the text.

Fig. 5. Anatomical marks of sella turcica cavity, left side (anatomical specimen, endoscope 0°). ICA — internal carotid artery, ON — optic nerve, SBA – sphenoid bone area, LR — lateral recess of sphenoid sinus, OCR – optic-carotid recess, C — clivus, ST — sella turcica. Optic-carotid recess corresponds to optic strut.


Impaired bone tissue metabolism or dysplasia should be suspected if definitive ossification is absent in children older than 4 years. Some children have persistent cecum orifice (remainder of dural diverticulum extending from anterior margin of crista galli to nasal subcutaneous ar-ea). It may be confused with bone defect, but it differs by clear edges formed by cortical bone [34].

Pneumatization. J. Spaeth et al. [35] analyzed pneu-matization of paranasal sinuses after birth till 25 years and received valuable data. According to CT-data there is equable paranasal sinuses grow almost in all directions in children of different age. Dimensions of left and right si-nuses are similar excepting frontal sinuses which are al-most always asymmetrical in favor of left one.

Frontal sinuses are detected in CT-scans in 1.4—1.5% of newborns of both genders and are distinguishable in only 10.7% of 4-year-old and 50% of 8-year-old chil-dren. They are found in 90% of people over 15 years old. Further development of frontal sinuses is absent after 18 years in boys and 15 years in girls. Advanced growth in pubertal period is absent. Frontal sinuses aplasia was more common in girls (18.2%, mean incidence after complete development — 9.4%) compared with boys (10 and 4.9%, respectively). According to M. Gulisano et al. [36], frontal sinus is absent in 5% of people, hypopla-sia is noted in 4.8% (Fig. 10). J. Pondé et al. [3] reported aplasia in 6.86% of persons (9.75% of men and 5.26% of women). Septum between both frontal sinuses was not found in 5% of cases.

Ethmoidal cells are observed in 94.4% of newborn boys and in 93.9% of newborn girls. Advanced growth of ethmoidal labyrinths in length rather width is observed in

4—5-year-old boys. Labyrinths reach their final dimen-sions after 13 years. There is no further enlargement up to 25 years. Final width and length of ethmoidal labyrinths in girls are observed after 10 and 13 years, respectively. Complete development of ethmoidal sinuses is noted af-ter 12—13 years in both genders, but it may be near 2 years earlier in girls.

Sphenoid sinus is found in 6.3% of boys and 6.7% of girls within the first year of life, in 2-year children – in 51.1 and 57.7%, respectively. CT-scans visualize sphenoid sinus after 8 years, craniograms – after 6 years. Sphenoid sinus is seen in more than 90% of children older than 8 years. Sphenoid sinuses are characterized by stable de-velopment. Rapid development is completed before

Fig. 6. Variants of sphenoid sinus pneumatization. a – conchal type (sphenoid bone body is almost completely formed by spongy tissue; sinus is reduced); b – pre-sellar type (posterior wall of sinus corresponds to the level of anterior wall of pituitary fossa); c – sellar type (sinus cavity is also projected onto pituitary fossa); d – retro-sellar type (sinus cavity extends to sella turcica; * — retro-sellar pneumatization).

Fig. 7. Bilateral pneumatization of anterior clinoid processes (arrows).


REVIEWS

5 years old. It is obviously that sphenoid sinuses growth is stopped after 16 years in boys and 14—15 years in girls [35].

Anatomical features of ethmoid bone. C. Güldner et al. [37] compared courses of AEA and incidence of vari-ous types of olfactory fossa in persons younger and older than 18 years. There were 3 AEA courses: 1) through skull base directly; 2) through ethmoidal labyrinth within 3 mm from skull base; 3) through ethmoidal labyrinth at a distance over 3 mm from skull base. AEA was predomi-nantly located in skull base in younger patients (48.1% vs. 44.1%, p<0.05), AEA type II was more typical for older group (11.2% vs. 18.5%, p<0.05), type III was more com-mon in young people (40.1% vs. 37.4%, p<0.05). Olfac-tory fossa Keros type I was significantly more frequent in "young" group (27.6% vs. 15.7%, p<0.001), while types II

and III were more common in persons older 18 years (50.9% vs 59.8%, p<0.001 and 21.5% vs. 24.5%, p<0.001, respectively). It is noteworthy that other researchers re-vealed similar structure of central ACF regardless age.

Influence of anterior approaches on facial skeleton and dental system development in children

Facial skull growth and development is one of the factors limiting extracranial approaches in pediatric pa-tients with intra- and extracranial skull base tumors [38].

In children younger 6 years ACF is smaller and fron-tal sinuses are not completely formed. Therefore, tran-scranial accesses in children are technically simpler com-pared with adults [39]. Moreover, small cerebral skull and thin cranial vault require special surgical techniques of craniotomy [40]. There are no conventional anatomical reference points in children younger 8 years such as su-praorbital orifice or incisura [41, 42]. Frontonasal suture as active growth point of facial skeleton up to 16 years old should be kept in mind in choosing transcranial ap-proach. Nevertheless, some effect of subcranial approach (≤10%) on vertical and horizontal growth of facial skele-ton in middle zone was reported [43]. Another feature of transcranial surgery in children is possible difficult brain relaxation; ventricular puncture and liquor evacuation from basal cisterns are less effective than in adults [44].

Sublabial approaches are followed by wide exposure but dangerous for permanent teeth development in chil-dren younger 6—10 years [39, 45]. Endoscopy-assisted "midfacial degloving" approach was used in children of various ages [46]. Complications of sublabial approaches included teeth denervation and anesthesia of infraorbital nerve’s innervation zone [45], oro-antral fistulae and epiphora [47].

Endoscopic endonasal approaches including transs-phenoidal access are widely used in children [48, 49]; un-like traditional transfacial approaches they do not affect dental system development and facial skeleton growth. However, advanced excision of bone and cartilaginous tissue from nasal cavity walls can lead to impaired growth later [49]. It should be also taken into account that sphe-

Fig. 8. Left-side sphenoethmoidal cell (Onodi). Optic canals are indicated by triangles. Right optic canal is bordered by right sphenoid sinus, left optic canal is bordered by Onodi cell (*) (** — sphenoid sinus).

Fig. 9. Advanced left-sided pneumatization of ethmoidal cells in hyperostotic cranio-orbital meningioma.


REFERENCES1. Lee JM, Ransom E, Lee JYK, Palmer JN, Chiu AG. Endoscopic anterior

skull base surgery: intraoperative considerations of the crista galli. Skull Base. 2011;21(2):83-86. https://doi.org/10.1055/s-0030-1263283

2. Terrier F, Weber W, Ruefenacht, Porcellini B. Anatomy of the ethmoid: CT, endoscopic, and macroscopic. AJR Am J Roentgenol. 1985;144(3):493-500.

https://doi.org/10.2214/ajr.144.3.4933. Pondé JM, Metzger P, Amaral G, Machado M, Prandini M. Anatomic

variations of the frontal sinus. Minim Invasive Neurosurg. 2003;46(1):29-32. https://doi.org/10.1055/s-2003-379564. Pobornikova S. An x-ray investigation of the development of the frontal si-

nuses in children. Folia Med (Plovdiv). 1974;16(4):213-220.5. Szilvássy J. Development of the frontal sinuses. Anthropol Anzeiger; Bericht

uber die Biol Lit. 1981;39(2):138-149.6. Konovalov AN. (ed). Skull base tumor surgery. M.: Antidor; 2004. (In

Russ.).7. Som PM, Lawson W. The frontal intersinus septal air cell: a new hypothesis

ofiIts origin. Am J Neuroradiol. 2008;29(6):1215-1217. https://doi.org/10.3174/ajnr.A10578. Lund VJ, Stammberger H, Fokkens WJ, Beale T, Bernal-Sprekelsen M,

Eloy P, Georgalas C, Gerstenberger C, Hellings P, Herman P, Hose-mann WG, Jankowski R, Jones N, Jorissen M, Leunig A, Onerci M, Rim-mer J, Rombaux, Simmen D, Tomazic PV, Tschabitscher M, Werge-Lues-sen A. European position paper on the anatomical terminology of the inter-nal nose and paranasal sinuses. Rhinol Suppl. 2014;50(24):1-34.

9. Güldner C, Diogo I, Windfuhr J, Bien S, Teymoortash A, Werner JA, Bremke M. Analysis of the fossa olfactoria using cone beam tomography (CBT). Acta Otolaryngol. 2011;131(1):72-78.

https://doi.org/10.3109/00016489.2010.50665310. Krmpotić-Nemanić J, Padovan I, Vinter I, Jalsovec D. Development of the

cribriform plate and of the lamina mediana. Ann Anat. 1998;180(6):555-559. https://doi.org/10.1016/S0940-9602(98)80065-4

11. Kainz J, Stammberger H. The roof of the anterior ethmoid: a locus minoris resistentiae in the skull base. Laryngol Rhinol Otol (Stuttg). 1988;67(4):142-149.

12. Keros P. On the practical value of differences in the level of the lamina cribrosa of the ethmoid. Zeitschrift fur Laryngol Rhinol Otol und ihre Gren-zgebiete 1962;41:809-813.

13. Song M, Zong X, Wang X, Pei A, Zhao P, Gui S, Yan Y, Zhang Y. Ana-tomic study of the anterior skull base via an endoscopic transnasal ap-proach. Clin Neurol Neurosurg 2011;113(4):281-284.

https://doi.org/10.1016/j.clineuro.2010.11.01914. Caliot P, Plessis JL, Midy D, Poirier M, Ha JC. The intraorbital arrange-

ment of the anterior and posterior ethmoidal foramina. Surg Radiol Anat. 1995;17:29-33. https://doi.org/10.1007/BF01629496

15. Kazak Z, Celik S, Ozer MA, Govsa F. Three-dimensional evaluation of the danger zone of ethmoidal foramens on the frontoethmoidal suture line on the medial orbital wall. Surg Radiol Anat. 2015;37(8):935-940.

https://doi.org/10.1007/s00276-015-1429-416. Celik S, Ozer MA, Kazak Z, Govsa F. Computer-assisted analysis of anatom-

ical relationships of the ethmoidal foramina and optic canal along the medial orbital wall. Eur Arch Oto-Rhino-Laryngology 2014;272(11):3483-3490.

https://doi.org/10.1007/s00405-014-3378-717. Wang J, Bidari S, Inoue K, Yang H, Rhoton A. Extensions of the sphenoid

sinus: a new classification. Neurosurgery. 2010;66(4):797-816. https://doi.org/10.1227/01.NEU.0000367619.24800.B118. Rhoton ALJr. The orbit. Neurosurgery. 2002;51(4 Suppl):S303-S334. https://doi.org/10.1227/01.NEU.0000028329.38986.4719. Abhinav K, Acosta Y, Wang W-H, Bonilla LR, Koutourousiou M, Wang E,

Snyderman C, Gardner P, Fernandez-Miranda JC. Endoscopic endonasal approach to the optic canal. Neurosurgery. 2015;11(3):431-446.

https://doi.org/10.1227/NEU.000000000000090020. Berhouma M, Jacquesson T, Abouaf L, Vighetto A, Jouanneau E. Endo-

scopic endonasal optic nerve and orbital apex decompression for nontrau-

noid sinus pneumatization is rare in children younger 3 years [33]. There is a point of view that endoscopic ap-proaches are limited in children due to small nasal cavity and need for very thin endoscopes and instruments [50]. Nevertheless, according to literature endoscopy is possi-ble in children of any age (even 1.5 months) despite small nasal cavity. So, F. Di Rocco et al. [49] reported no cor-relation between age and outcomes of endoscopic endo-nasal repair of anterior skull base defects.

ConclusionIndividual characteristics, variability of structures

and age-related features of ACF should be considered if surgery for anterior skull base tumors is planned. It is true for both children and adults. Bones growth and ossifica-tion, pneumatization of sinuses, dental system develop-ment determine choice of optimal and safe surgical ap-proach in young patients with incomplete development of craniofacial area. Individual anatomic features of criti-cal structures within surgical field (surgical course, key anatomical marks, optic tracts, internal carotid and eth-moidal arteries, etc.) should be taken into account in any patients. These aspects can impair exposition, safe ma-nipulations, or increase risk of, for example, vascular in-jury or impaired skull base impermeability (in transnasal accesses), etc. Age-related and individual variability of midline structures of anterior skull base is summarized in this article. These data may be useful for neurosurgeons and otorhinolaryngologists with interests in this anatom-ical area.

The authors are grateful to K.E. Klimenko, M.Z. Dzha-farova and G.B. Bebchuk for help in illustrations selection.


Fig. 10. Aplasia of right frontal sinus (sinus cavity is completely absent), left-sided hypoplasia (* — very small cavity in frontal bone is identified).


REVIEWS

matic optic neuropathy: surgical nuances and review of the literature. Neu-rosurg Focus. 2014;37(4):E19.

https://doi.org/10.3171/2014.7.FOCUS1430321. Li J, Wang J, Jing X, Zhang W, Zhang X, Qiu Y. Transsphenoidal optic

nerve decompression: an endoscopic anatomic study. J Craniofac Surg. 2008;19(6):1670-1674. https://doi.org/10.1097/SCS.0b013e31818b4316

22. Guthikonda B, Tobler WD, Froelich SC, Leach JL, Zimmer LA, Theo-dosopoulos PV, Tew JM, Keller JT. Anatomic study of the prechiasmatic sulcus and its surgical implications. Clin Anat. 2010;23(6):622-628. https://doi.org/10.1002/ca.21002

23. Rhoton AJ. Cranial Anatomy and Surgical Approaches. Philadelphia: Lip-pincott Willams & Wilkins;2003.

24. Fujioka M, Young LW. The sphenoidal sinuses: radiographic patterns of normal development and abnormal findings in infants and children. Radiol-ogy. 1978;129(1):133. https://doi.org/10.1148/129.1.133

25. Tan HKK, Ong YK, Teo MSK, Fook-Chong SMC. The development of sphenoid sinus in Asian children. Int J Pediatr Otorhinolaryngol. 2003;67(12):1295-1302. https://doi.org/10.1016/j.ijporl.2003.07.012

26. Hamberger CA., Hammer G, Norlen G, Sjogren B. Transantrosphenoidal hypophysectomy. Arch Otolaryngol 1961;74:2-8.

27. Rahmati A, Ghafari R, Anjomshoa M. Normal variations of sphenoid sinus and the adjacent structures detected in cone beam computed tomography. J Dent Shiraz Univ Med Sci. 2016;17(1):32-37.

28. Girguis-Bucher A, Schlegel-Wagner C. Anatomical variation of the extra-cranial course of the optic nerve in the floor of the sphenoid sinus: first re-ported case. J Laryngol Otol. 2013;127:822-824.

https://doi.org/10.1017/S002221511300101129. Jacquesson T, Abouaf L, Berhouma M, Jouanneau E. How I do it: the en-

doscopic endonasal optic nerve and orbital apex decompression. Acta Neu-rochir (Wien). 2014;156(10):1891-1896.

https://doi.org/10.1007/s00701-014-2199-130. Pletcher SD, Sindwani R, Metson R. Endoscopic orbital and optic nerve

decompression. Otolaryngol Clin North Am. 2006;39(5):943-958. https://doi.org/10.1016/j.otc.2006.06.00331. Al-Abri R, Bhargava D, Al-Bassam W, Al-Badaai Y, Sawhney S. Clinically

significant anatomical variants of the paranasal sinuses. Oman Med J. 2014;29(2):110-113. https://doi.org/10.5001/omj.2014.27

32. Parizel PM, Carpentier K, Van Marck V, Venstermans C, De Belder F, Van Goethem J, Van Den Hauwe L, Menovsky T. Pneumosinus dilatans in an-terior skull base meningiomas. Neuroradiology. 2013;55(3):307-311.

https://doi.org/10.1007/s00234-012-1106-933. Gump WC. Meningiomas of the pediatric skull base: a review. J Neurol Sur-

gery, Part B Skull Base. 2015;76(1):66-73. https://doi.org/10.1055/s-0034-139001234. Hughes DC, Kaduthodil MJ, Connolly DJA, Griffiths PD. Dimensions

and ossification of the normal anterior cranial fossa in children. Am J Neu-roradiol. 2010;31(7):1268-1272. https://doi.org/10.3174/ajnr.A2107

35. Spaeth J, Krügelstein U, Schlöndorff G. The paranasal sinuses in CT-imag-ing: development from birth to age 25. Int J Pediatr Otorhinolaryngol. 1997;39(1):25-40. https://doi.org/10.1016/S0165-5876(96)01458-9

36. Gulisano M, Pacini P, Orlandini GE. Frontal sinus dimensions in relation to the cranial index: anatomo-radiologic findings. Boll della Soc Ital di Biol Sper. 1978;54(1):66-69.

37. Güldner C, Zimmermann AP, Diogo I, Werner JA, Teymoortash A. Age-dependent differences of the anterior skull base. Int J Pediatr Otorhinolaryn-gol. 2012;76(6):822-828.

https://doi.org/10.1016/j.ijporl.2012.02.05038. Tsai EC, Santoreneos S, Rutka JT. Tumors of the skull base in children:

review of tumor types and management strategies. Neurosurg Focus. 2002;12(5):e1. https://doi.org/10.3171/foc.2002.12.5.2

39. Lang DA, Neil-Dwyer G, Evans BT, Honeybul S. Craniofacial access in children. Acta Neurochir (Wien). 1998;140(1):33-40.

https://doi.org/10.1007/s00701005005440. Gross N, Ganly I, Patel S, Bilsky M, Shah J, Kraus D. Results of anterior

skull base surgery in pediatric and young adult patients. Skull Base. 2010;20(2):075-081. https://doi.org/10.1055/s-0029-1238215

41. Brockmeyer D, Gruber DP, Haller J, Shelton C, Walker mL. Pediatric skull base surgery. 2. Experience and outcomes in 55 patients. Pediatr Neurosurg. 2003;38(1):9-15. https://doi.org/10.1159/000067563

42. Chivukula S, Koutourousiou M, Snyderman CH, Fernandez-Miranda JC, Gardner PA, Tyler-Kabara EC. Endoscopic endonasal skull base surgery in the pediatric population. J Neurosurg Pediatr. 2013;11(3):227-241.

https://doi.org/10.3171/2012.10.PEDS1216043. Shlomi B, Chaushu S, Gil Z, Chaushu G, Fliss DM. Effects of the subcra-

nial approach on facial growth and development. Otolaryngol Head Neck Surg. 2007;136(1):27-32.

https://doi.org/10.1016/j.otohns.2006.07.01944. Gil Z, Constantini S, Spektor S, Abergel A, Khsfif A, Beni-Adani L, Le-

onor TL, Derowe A, Fliss DM. Skull base approaches in the pediatric population. Head Neck. 2005;27(8):682-689.

https://doi.org/10.1002/hed.2022645. Lewark TM, Allen GC, Chowdhury K, Chan KH. Le Fort I osteotomy and

skull base tumors: a pediatric experience. Arch Otolaryngol Head Neck Surg. 2000;126(8):1004-1008. https://doi.org/10.1001/archotol.126.8.1004

46. Uretzky ID, Mair EA, Schoem SR. Endoscopically guided midfacial de-gloving in infants for removal of congenital and acquired midfacial masses. Int J Pediatr Otorhinolaryngol. 1998;46(3):145-158.

https://doi.org/10.1016/S0165-5676(98)00087-147. Howard DJ, Lund VJ. The role of midfacial degloving in modern rhino-

logical practice. J Laryngol Otol. 1999;113(10):885-887.48. Kassam A, Thomas AJ, Snyderman C, Carrau R, Gardner P, Mintz A,

Kanaan H, Horowitz M, Pollack IF. Fully endoscopic expanded endonasal approach treating skull base lesions in pediatric patients. J Neurosurg. 2007;106(2 Suppl):75-86. https://doi.org/10.3171/ped.2007.106.2.75

49. Di Rocco F, Couloigner V, Dastoli P, Sainte-Rose C, Zerah M, Roger G. Treatment of anterior skull base defects by a transnasal endoscopic ap-proach in children. J Neurosurg Pediatr. 2010;6(5):459-463.

https://doi.org/10.3171/2010.8.PEDS0932550. Ciechomski J, Aufgang R, Villanueva L, Demarchi V. Subcranial approach

in pediatric craniofacial surgery. Craniomaxillofac Trauma Reconstr. 2010;3(4):231-236. https://doi.org/10.1055/s-0030-1268521

Received: 07.06.17


The report is devoted to actual issue of topographic anato-my variability of anatomically complex area per se. Interdisci-plinary approach to anterior skull base is of additional interest: internal base is historically field of neurosurgery, while transna-sal approaches are being studied together with otolaryngolo-gists. Thus, article is integration of both specialties. Moreover, age-related topographic features determining optimal access are discussed. The last forces the reader to think about advanced surgical approaches and to step back from accepted stereotypes in favor of the most anatomically advisable procedure in certain patient. The authors rightly point at need for CT- and MRI-data for preoperative planning of both optimal surgical ap-proach and skull base repair. For example, it is known that tran-scranial approach is preferred over transnasal transcribriform access in pediatric practice due to small area of nasoseptal flap.

The authors rightly emphasize considerable anatomic vari-ability of ethmoidal bone, that has been recently studied in nu-merous publications along with endoscopic rhinosurgery devel-opment. However, it is necessary to unify some terms to gener-ally accepted concepts according to European Position Paper on the Anatomical Terminology of the Internal Nose and Para-nasal Sinuses (Rhinology Journal Supplement 24, European Society of Rhinologists) and their official Russian translation.

In total, the most important "points of variability" are consid-ered which must be assessed to determine endoscopic approach to median structures of anterior skull base: olfactory fossa’s depth by Keros, variable placement of ethmoidal arteries within skull base, presence of Onodi cells, which should be taken into account in advanced transethmoidal sphenotomy and optic channels decompression, sphenoid sinus and clinoid processes pneumatization which are the key in sellar and parasellar expo-sure.

Age is important factor emphasized by the authors (para-nasal sinuses pneumatization, presence of teeth rudiments, ad-vanced indications for endoscopic transnasal approach in chil-dren if it is surgically justified). In early 2000s we got acquainted publications about dangerous endoscopic surgery of paranasal sinuses in children due to injury of growth zones of facial skel-eton. This paradigm has been completely changed over the last decade, when large trials have shown that endoscopic proce-dures do not affect growth rate of facial skull. It should be noted that endoscopic rhinosurgery in pediatric practice is being rap-idly developed as independent direction — PESS (Pediatric Endoscopic Sinus Surgery). Thus, relevance of this issue is confirmed once again.

G.A. Polev (Moscow, Russia)

Commentary


REVIEWS

Preservation of facial nerve function is important problem in treatment of tumors of posterior cranial fossa and cerebellopontine angle first of all (cochlear and facial nerves neuromas, meningioma of posterior cranial fossa base, trigeminal nerve neuroma). This pathology is often associated with involvement of intracranial or intracanal segments of facial nerve (Figs. 1, 2) [1—4]. Facial nerve injury leads to severe functional disturbances and psy-chological trauma in view of important aesthetic compo-nent in face lesion. So, treatment of cerebellopontine angle tumors requires coordinated actions of multidisci-plinary team including neurosurgeons, neurophysiolo-gists and anaesthesiologists in order to reduce likelihood of facial nerve injury and paralysis of facial muscles.

At present time there is low incidence of facial nerve injury followed by postoperative dysfunction in various medical centers [5, 6]. Nevertheless, this unfavorable event (dysfunction House-Brackmann grade III—IV) occurs in 6—20% of cases after surgery especially in large tumors (over 3 cm) [7—9]. Anatomically intact facial nerve is not always accompanied by good long-term functional recovery.

Paresis or paralysis of facial muscles occur due to fa-cial nerve compression, its destruction by tumor (facial nerve neuroma) and surgical injury. Tumor dissection may be followed by postoperative paresis or paralysis of facial muscles even in anatomically intact facial nerve. In case of large tumors of cerebellopontine angle facial nerve may be adherent to tumor capsule as thin mem-brane under operating microscope. In these cases, radi-cal excision of tumor is inevitably associated with nerve trauma and advanced probability of postoperative facial muscles dysfunction. In this regard, most authors suggest



Neurosurgery Followed by Facial Nerve Injury: Rehabilitation Potential of Botulinum TherapyM.A. AKULOV1*, O.R. ORLOVA2, Т.V. TABASHNIKOVA1, V.V. KARNAUKHOV1, A.S. ORLOVA2

1Burdenko Neurosurgical Institute, str. 4-ya Tverskaya-Yamskaya, 16, Moscow, Russia, 125047; 2Sechenov First Moscow State Medical University, str. Trubetskaya, 8, bld 2, Moscow, Russia, 119991

Surgical treatment of posterior cranial fossa and cerebellopontine angle tumors is associated with a risk of facial nerve dysfunction. The causes for facial muscle paresis include nerve compression by the tumor, destruction of the nerve structure by the tumor growing from nerve fibers, nerve injury during surgical removal of the tumor, etc. The first 3 months after facial nerve injury are a potential therapeutic window for the use of botulinum toxin type A (BTA). During this period, the drug is introduced both in the healthy side to improve the facial symmetry at rest and during mimetic movements and in the affected side to induce drug-induced ptosis. Post-paralytic syndrome develops 4—6 months after facial nerve injury. At this stage, administration of BTA is also an effective procedure; in this case, drug injections are performed on the affected side at small doses and symmetrically on the healthy side at doses doubling those for the affected side. BTA injections are mandatory in complex treatment of facial muscle paralysis.

Keywords: posterior cranial fossa tumors, cerebellopontine angle tumors, skull base tumors, facial nerve injury, drug-induced ptosis, synkinesis, botulinum toxin type A.

subtotal resection and subsequent follow-up [10] or com-bined treatment (resection + radiosurgery) of large co-chlear neuroma [11].

Postoperative function of facial nerve depends on various factors. Thus, large tumors are associated with less favorable outcomes compared with small and middle neoplasms [12—14]. Relationships of facial nerve and tu-mor also affect the outcome. Usually, nerve is located within anteroinferior pole of tumor. Rarely (about 1%) nerve may be placed on posterior or inferior fragments of capsule (Fig. 3) [15]. Advanced intraoperative manipula-tions and dislocations of facial nerve are also associated with less favorable outcomes [16].

Morphological features of tumor can affect surgical outcome especially in its tuberous structure (cerebello-pontine angle neurofibroma in neurofibromatosis type II) or if nerve passes through the tumor (meningio-ma overgrowing cochlear-facial nerves, cystic cochlear neuroma). There are literature data [17, 18] confirming correlation between cystic tumor and adverse outcomes, while other reports revealed similar postoperative func-tion of facial nerve regardless cystic or solid tumors.

Surgeon’s experience and arachnoid membrane dis-section are significant favorable factors for postoperative function of facial nerve [19]. In case of cochlear neuro-ma, the greatest adhesion of tumor and facial nerve is usually observed within internal acoustic canal. There-fore, trepanation of its wall followed by subsequent exci-sion of intracanal neuroma is necessary after intracapsu-lar tumor extraction. Intraoperative neurophysiological monitoring of facial nerve function is useful to locate fa-cial nerve intraoperatively and to minimize its trauma [20, 21].


Acute injury of facial nerve is manifested by unilat-eral paralysis or paresis of facial muscles due to impaired anatomic integrity or function of nerve. Facial muscles dysfunction aggravates eyelids closure, eating, articula-tion, and subsequently quality of life [22—24].

House-Brackmann (Table 1) and Yanagihara (Table 2) scales are the most famous and widely used classification of facial nerve lesion [25].

House-Brackmann 6-score scale is used to assess motor dysfunction of facial muscles. It includes eyebrow and mouth angle movements and their comparison with those on the opposite side. Special scale with divisions of 0.25 cm is used to estimate facial asymmetry. Maximum possible value is 8 (4 or 1 cm for mouth and 4 or 1 cm for eyebrow). Yanagihara scale evaluates 10 motor reactions of facial muscles from 0 (severe palsy) to 4 scores (no dis-orders) for each movement. Both these scales are suffi-ciently demonstrative and choice should be made by spe-cialist [26, 27].

The first weeks and months of postoperative reha-bilitation are the most difficult for patients [28]. Corneal

protection is one of the most important measures at this stage. Ineffective conventional methods including oint-ments and protective dressings to prevent corneal chang-es should be emphasized. It is true even if treatment was started from the 1st postoperative day [29]. Currently, temporary blepharoptosis by botulinum toxin injection followed by corneal protection is possible for postopera-tive paresis of orbicular muscle of the eye. Thus, conven-tional invasive approaches such as blepharoplasty are not so actual at present time. Early period after facial nerve injury (within 3 months) is considered as potential thera-peutic window for botulinum toxin type A (BTA) admin-istration to achieve unilateral blepharoptosis [2] and to improve facial symmetry at rest and in mimic movements on contralateral side [1, 30, 31].

BTA injection into contralateral facial muscles early after facial nerve lesion is performed to reduce muscular hypertonicity [32]. Hyperactivity of these muscles is as-sociated with advanced facial asymmetry. It is followed by paretic muscles affection and their impaired recov-ery [32, 33].

BTA injection into m. levator palpebrae superioris fol-lowed by blepharoptosis is adequate care for acute tro-phic keratopathy, corneal erosions or ulcers. This proce-dure prevents severe corneal disorders and allows to avoid unjustifiably aggressive correction of lagophthalmos [1].

BTA-induced ptosis is achieved by injection of toxin 10—25 units into m. levator palpebrae superioris through superior orbital-palpebral fissure (Lantox, Botox, Xeo-min). Initial effect is observed after 2 days. Complete BTA-induced ptosis occurs in 3 days (length of ef-fect — 25—35 days). This period is followed by residual ptosis until the end of BTA effect (30—40 days after toxin injection) (Figs. 4, 5) [25].

а

b

Fig. 1. Cerebellopontine angle exposure through retrosigmoid suboccipital approach in prone position. a — intraoperative photo; b — schematic image: 1 — superior petrosal sinus, 2 — petrosal vein, 3 — trigeminal nerve, 4 — subarcuate artery, 5 — facial nerve, 6 — anterior inferior cerebellar artery, 7 – vestibulocochlear nerve, 8 — glossopharyngeal nerve, 9 — vagus nerve, 10 — additional spinal nerve, 11 — posterior inferior cerebellar artery, 12 — spatula, 13 — cerebellum hemisphere.

Fig. 2. Cochlear neuroma (intraoperative photo). View through retrosigmoid suboccipital aproach. 1 – petrous part, internal auditory meatus area, 2 – cochlear neuroma, 3 – cochlear-facial nerves, 4 – spatulas pushing cerebellum, 5 — cerebellum covered by wadding.


REVIEWS

Ptosis is useful for mechanical corneal protection until facial nerve function will be restored. Surgical repair is considered if this approach is ineffective.

Indications for BTA-induced ptosis:— lagophthalmos and/or unresponsive to therapy

early postoperative severe trophic keratopathy;— aggravation of trophic keratopathy in long-term

postoperative period despite adequate protective therapy;— delayed repair of lagophthalmos in facial nerve

dysfunction.Some reports confirmed BTA effectiveness for post-

operative paresis of orbicular muscle of the eye. J. Prell et al. [4] reported percutaneous and transconjunctival BTA injections (25 units) for postoperative paresis of facial muscles House-Brackmann grade IV and over, keratopa-thy, corneal erosions or ulcers. Temporary eyelids closure for 2—6 months followed by sufficient recovery of facial nerve function was achieved in all cases. The authors concluded that BTA-induced ptosis is adequate low risk alternative to other invasive approaches for severe post-operative lagophthalmos and anatomically intact nerve.

Post-paralytic syndrome including contractures and synkinesis of facial muscles, fasciculations, lacrimation and salivation besides mimic insufficiency develops in 4—6 months after facial nerve injury [34]. Synkinesis is involuntary movement accompanying voluntary move-ment (eye screwing up in orbicularis oris muscle move-ments) or contraction of entire unilateral mimic muscles during voluntary work of certain group of muscles. Sun-

nybrook Facial Grading Scale is used for assessing synki-nesis (Table 3) [35].

This scale is useful for monitoring of postoperative rehabilitation in long-term period after facial nerve injury and consists of 3 points (resting symmetry, voluntary movements symmetry and synkinesis).

Fig. 3. Facial nerve on cochlear neuroma’s capsule (schematic image, facial nerve is indicated by arrow). a – on superior surface; b — on posterior surface; c — on anterior surface; d — on inferior surface [15].

Table 1. House-Brackmann scale

I No dysfuction 8/8 100%II Mild dysfunction 7/8 80%III Moderate dysfunction 5/8—6/8 60%IV Moderate-to-severe

dysfunction 3/8—4/8 40%V Severe dysfunction 1/8—2/8 20%VI Total palsy 0/8 0%

Table 2. Yanagihara scale

At rest 0 1 2 3 4Wrinkle forebead 0 1 2 3 4Blink 0 1 2 3 4Slight closure of both eyes 0 1 2 3 4Closure of one eye 0 1 2 3 4Depress lower lip 0 1 2 3 4Swell cheeks 0 1 2 3 4Whistle 0 1 2 3 4Wrinkle nose 0 1 2 3 4Grin 0 1 2 3 4


BTA injection is effective for synkinesis. Small doses of toxin are used on affected side and 2 times higher dos-es on the contralateral side symmetrically [36—38].

Rehabilitologists of Burdenko Neurosurgery Center revealed that BTA administration reduces severity of post-paralytic syndrome after facial nerve injury [1, 28, 33]. Six-month and 1-year incidence of synkinesis after facial nerve injury was significantly lower in patients after BTA injections compared with conventional approaches. Facial Disability Index and SFGS scores were also sig-nificantly better [1].

Other trials have also shown effectiveness of BTA for synkinesis management. Thus, E. Toffola et al. [39] re-ported 55 injections of BTA (2.5 units per each area) in 30 patients with facial muscles palsy followed by synkine-sis. Orbicular muscle of the eye, temporal muscle, de-pressor labii inferioris muscle, platysma, frontal muscle and corrugator supercilii muscle were target. Improved synkinesis was subsecuently observed in all patients. The authors concluded that BTA injections effectively reduce facial synkinesis severity and improve resting and volun-tary movements symmetry.

K. Choi et al. [40] analyzed 42 patients with partial recovery after facial nerve injury. There were 1.5—2.5 units of BTA per each unilateral area and total dose near 10—26 units. Dose injected into contralateral hypertonic muscles was 2.5—5 units. Reduced synkinesis and im-proved facial symmetry were noted in all patients after bilateral BTA injections. SFGS score before and after in-jection was 38.8±10.68 and 58.4±12.46, respectively. The authors concluded that BTA administration is associated

with significantly reduced synkinesis and improved facial symmetry. Therefore, improved quality of life, social in-teractions, appearance and eating may be observed.

ConclusionAdequate administration of botulinum toxin im-

proves rehabilitation in patients with postoperative dys-function of facial nerve. BTA therapy should be per-

Fig. 5. Lagophthalmos before and after BTA administration. a – prior to BTA-induced ptosis: open and closed eyes; b – the 1st day after BTA injection: open and closed eyes; c — one year after BTA injection: open and closed eyes.

Fig. 4. BTA injection points in facial nerve lesion followed by acute paresis of facial muscles. (arrow — point of injection into m. levator palpebrae superioris) [1].

а

b

c


REVIEWS

formed both in early (for BTA-induced ptosis on affected side and to improve contralateral facial symmetry) and long-term (post-paralytic syndrome correction) period after facial nerve damage. This approach is followed by improved rehabilitation and reduced incidence of long-

term complications. Facial muscles palsy due to complete intraoperative intersection of the nerve requires consulta-tion of maxillofacial surgeons to define terms of repair. BTA injection is advisable for facial asymmetry correc-tion in these cases.


Table 3. Sunnybrook Facial Grading Scale


REFERENCES1. Orlova OR, Akulov MA, Usachev DYu, Taniashin SV, Zakharov VO,

Saksonova EV, Mingazova LR, Surovykh SV. The use of botulinum toxin type a in the acute phase of facial nerve injury after neurosurgical surgery. Voprosy neirokhirurgii. 2014;78(6):50-54. (In Russ.).

https://doi.org/10.17116/neiro201478650-542. Tabashnikova TV, Serova NK, Shimanskiĭ VN, Grigor’eva NN. Botulo-

toxin type A (lantox) for inducing temporal ptosis in neurosurgical patients. Voprosy neirokhirurgii. 2012;4:43-48. (In Russ.).

3. Chan HH, Asadi H, Dowling R, Hardy TG, Mitchell PJ. Facial nerveinjury as a complication of endovascular treatment for cavernous duralarteriove-nous fistula. Orbit. 2014;33(6):462-464.

https://doi.org/10.3109/01676830.2014.9502914. Prell J, Rampp S, Rachinger J, Scheller C, Alfieri A, Marquardt L,

Strauss C, Bau V. Botulinum toxin for temporary corneal protection after surgery for vestibular schwannoma. J Neurosurg. 2011;114(2):426-431.

https://doi.org/10.3171/2010.4.JNS101045. Arts HA, Telian SA, El-Kashlan H, Thompson BG. Hearing preservation

and facial nerve outcomes in vestibular schwannoma surgery: results using the middle cranial fossa approach. Otol Neurotol. 2006;27:234-241.

https://doi.org/10.1097/01.mao.0000185153.54457.166. Brackmann DE, Cullen RD, Fisher LM. Facial nerve function after trans-

labyrinthine vestibular schwannoma surgery. Otolaryngol Head Neck Surg. 2007;136:773-777. https://doi.org/10.1016/j.otohns.2006.10.009

7. Anaizi AN, Gantwerker EA, Pensak mL, Theodosopoulos PV. Facial nerve preservation surgery for koos grade 3 and 4 vestibular schwannomas. Neurosur-gery. 2014;75:671-677. https://doi.org/10.1227/NEU.0000000000000547

8. Jeltema HR, Bakker NA, Bijl HP, Wagemakers M, Metzemaekers JDM, van Dijk JMC. Near total extirpation of vestibular schwannoma with salvage radiosurgery. Laryngoscope. 2015;125:1703-1707.

https://doi.org/10.1002/lary.251159. Ryzenman JM, Pensak mL, Tew JM Jr. Facial paralysis and surgical reha-

bilitation: a quality of life analysis in a cohort of 1,595 patients after acoustic neuroma surgery. Otol Neurotol. 2005;26:516-521.

https://doi.org/10.1097/01.mao.0000169786.22707.1210. Schwartz MS, Kari E, Strickland BM, Berliner K, Brackmann DE,

House JW, Friedman RA. Evaluation of the increased use of partial resec-tion of large vestibular schwannomas: facial nerve outcomes and recur-rence/regrowth rates. Otol Neurotol. 2013;34:1456-1464.

https://doi.org/10.1097/MAO.0b013e318297655211. Brokinkel B, Sauerland C, Holling M, Ewelt C, Horstmann G, van Eck AT,

Stummer W. Gamma Knife radiosurgery following subtotal resection of vestibular schwannoma. J Clin Neurosci. 2014;21:2077-2082.

https://doi.org/10.1016/j.jocn.2014.03.03712. Ben Ammar M, Piccirillo E, Topsakal V, Taibah A, Sanna M. Surgical re-

sults and technical refinements in translabyrinthine excision of vestibular schwannomas: the GruppoOtologico experience. Neurosurgery. 2012;70: 1481-1491. https://doi.org/10.1227/NEU.0b013e31824c010f

13. Falcioni M, Fois P, Taibah A, Sanna M. Facial nerve function after vestibu-lar schwannoma surgery. J Neurosurg. 2011;115:820-826.

https://doi.org/10.3171/2011.5.JNS10159714. Moffat DA, Parker RA, Hardy DG, Macfarlane R. Factors affecting final

facial nerve outcome following vestibular schwannoma surgery. J Laryngol Otol. 2014;128:406-415. https://doi.org/10.1017/S0022215114000541

15. Sanna М, Mancini F, Russo A, Taibah A, Falcion M, iDi Trapani G. Atlas of Acoustic Neurinoma Microsurgery. 2nd ed. Thieme Stuttgard. New York. 2011;125.

16. Esquia-Medina GN, Grayeli AB, Ferrary E, Tubach F, Bernat I, Zhang Z, Bianchi C, Kalamarides M, Sterkers O. Do facial nerve displacement pat-tern and tumor adhesion influence the facial nerve outcome in vestibular schwannoma surgery? Otol Neurotol. 2009;30:392-397.

https://doi.org/10.1097/MAO.0b013e318196787417. Zhang Z, Wang Z, Huang Q, Yang J, Wu H. Removal of large or giant spo-

radic vestibular schwannomas via translabyrinthine approach: a report of 115 cases. ORL J Otorhinolaryngol Relat Spec. 2012;74:271-277.

https://doi.org/10.1159/00034379118. Tang IP, Freeman SR, Rutherford SA, King AT, Ramsden RT, Lloyd SKW.

Surgical outcomes in cystic vestibular schwannoma versus solid vestibular schwannoma. Otol Neurotol. 2014;35:1266-1270.

https://doi.org/10.1097/MAO.0000000000000435

19. Seo JH, Jun BC, Jeon EJ, Chang KH. Predictive factors influencing facial nerve outcomes in surgery for small-sized vestibular schwannoma. Acta Oto-laryngol (Stockh). 2013;133:722-727.

https://doi.org/10.3109/00016489.2013.77617820. Liu SW, Jiang W, Zhang HQ, Li XP, Wan XY, Emmanuel B, Shu K,

Chen JC, Chen J, Lei T. Intraoperative neuromonitoring for removal of large vestibular schwannoma: Facial nerve outcome and predictive factors. Clin Neurol Neurosurg. 2015;133:83-89.

https://doi.org/10.1016/j.clineuro.2015.03.01621. Moffat DA, Parker RA, Hardy DG, Macfarlane R. Factors affecting final

facial nerve outcome following vestibular schwannoma surgery. J Laryngol Otol. 2014;128(5):406-415. https://doi.org/10.1017/S0022215114000541

22. Tastanbekov MM, Oljushin VE, Bersnev VP, Fadeeva TN, Ruslyakova IA, Goman PG, Burchenya JuV, Pustovoy SV. Surgical treatment of large and giant acoustic neurinomas: features of surgical strategy and treatment re-sults. Nejrohirurgija. 2010;3:25-29. (In Russ.).

23. Kleiss IJ, Beurskens CH, Stalmeier PF, Ingels KJ, Marres HA. Quality of life assessment in facial palsy: validation of the Dutch Facial Clinimetric Evaluation Scale. Eur Arch Otorhinolaryngol. 2015;272(8):2055-2061.

https://doi.org/10.1007/s00405-015-3508-x24. Sun D, Shi Z, Li P, Shi S, Cai Y. Psychological status and quality of life in

acoustic neuroma patients with facial palsy after microsurgery: a 1-year postoperative follow-up study. Acta Neurol Belg. 2015;115(3):311-316.

https://doi.org/10.1007/s13760-014-0382-z25. Shirshov I, Dreval’ O, Lihterman L, Gorodanin A. Facial nerve palses. M.

2011;192. (In Russ.).26. Stodulski D, Skorek A, Mikaszewski B, Wiśniewski P, Stankiewicz C. Facial

nerve grading after parotidectomy. Eur Arch Otorhinolaryngol. 2015;272(9):2445-2450. https://doi.org/10.1007/s00405-014-3196-y

27. Katsumi S, Esaki S, Hattori K, Yamano K, Umezaki T, Murakami S.Quan-titative analysis of facial palsy using a three-dimensional facial motion mea-surement system. AurisNasus Larynx. 2015;42(4):275-283.

https://doi.org/10.1016/j.anl.2015.01.00228. Akulov M, Orlova O, Orlova A, Usachev D, Tanjashin S, Zakharov V. Inco-

botulinumtoxina (Xeomin) treatment for the correction of mimic muscle function after facial nerve injury. Toxicon. 2016;123:2.

https://doi.org/10.1016/j.toxicon.2016.11.01129. Yücel OE, Artürk N. Botulinum toxin-A-induced protective ptosis in the

treatment of lagophthalmos associated with facial paralysis. Ophthal Plast Reconstr Surg. 2012;28(4):256-260.

https://doi.org/10.1097/IOP.0b013e31824ee70230. Ho AL, Scott AM, Klassen AF, Cano SJ, Pusic AL, Van Laeken N. Mea-

suringquality of life and patientsatisfaction in facialparalysispatients: a sys-tematic review of patient-reported outcome measures. Plast Reconstr Surg. 2012;130(1):91-99. https://doi.org/10.1097/PRS.0b013e318254b08d

31. Mehdizadeh OB, Diels J, White WM. Botulinum toxin in the treatment of facial paralysis. Facial Plast Surg Clin North Am. 2016;24(1):11-20.

https://doi.org/10.1016/j.fsc.2015.09.00832. Saksonova EV, Orlova OR, Kurenkov AL. The functional asymmetry of the

facial neuromotor apparatus in patients with facial nerve neuropathy and its treatment with botulinum toxin type A lantox. Zhurnal nevrologii i psihiatrii im. S.S. Korsakova. 2013;113(10-1):29-35. (In Russ.).

33. Akulov MA, Orlova OR, Orlova AS. Correction of mimic muscle function after facial nerve trauma in acute and chronic period. RMZh. 2016;24(14):902-906. (In Russ.).

34. Bran GM, Börjesson PK, Boahene KD, Gosepath J, Lohuis PJ. Effect of endoscopic brow lift on contractures and synkinesis of the facial muscles in patients with a regenerated postparalytic facial nerve syndrome. Plast Re-constr Surg. 2014;133(1):121-129.

https://doi.org/10.1097/01.prs.0000436834.19066.7c35. Neely JG, Cherian NG, Dickerson CB, Nedzelski JM. Sunnybrookfacial-

gradingsystem: reliability and criteria for grading. Laryngoscope. 2010; 120(5):1038-1045. https://doi.org/10.1002/lary.20868

36. Bennis Y, Duquennoy-Martinot V, Guerreschi P. Epidemiologic overview of synkinesis in 353 patients with longstanding facial paralysis under treat-ment with botulinum toxin for 11 years. Plast Reconstr Surg. 2016;138(2):376-378. https://doi.org/10.1097/PRS.0000000000002348

37. Dall’Angelo A, Mandrini S, Sala V, Pavese C, Carlisi E, Comelli M, Tof-fola ED. Platysmasynkinesis in facial palsy and botulinum toxin type A. Laryngoscope. 2014;124(11):2513-2517. https://doi.org/10.1002/lary.24732


REVIEWS

This research is actual due to high incidence of intraopera-tive facial nerve injury especially in tumors of posterior cranial fossa. Problem of postoperative facial nerve dysfunction is caused by anatomical proximity of tumor and cochlear-facial nerves followed by increased risk of prosoplegia and impaired rehabilitation. This lesion requires longer and more intensive recovery compared with idiopathic paresis of facial nerve.

Botulinum toxin injection is one of the most current and effective approach for facial neuropathy. Advantages of this method are local prolonged effect and no systemic undesirable reactions. Main effect of BTA is inhibition of presynaptic re-lease of acetylcholine.

Authors reviewed literature data about botulinum toxin ad-ministration in early and long-term period after facial nerve damage. Botulinum toxin is used for different purposes. Its ad-ministration on intact side is useful to improve facial symmetry.

Commentary

According to our experience and other researchers it is followed by significantly improved rehabilitation and reduced long-term morbidity including contractures and synkinesis. Botulinum toxin is widely used for induced unilateral ptosis preventing such formidable complication of facial nerve dysfunction as lag-ophthalmos. This unfavorable event leads to advanced corneal disorders while medication alone is often ineffective. It seems to be that botulinum toxin is also effective for synkinesis in long-term period. This problem has always been considered difficult for rehabilitators and extremely undesirable for patients.

There are numerous references in the review, it is easy and interesting to read this report. Description of anatomical fea-tures of cochlear-facial nerves and tumor is supplemented by images that makes it possible to present this pathology in more details.

V.L. Golubev (Moscow, Russia)

38. Filipo R, Spahiu I, Covelli E, Nicastri M, Bertoli GA. Botulinum toxin in the treatment of facial synkinesis and hyperkinesis. Laryngoscope. 2012;122(2):266-270. https://doi.org/10.1002/lary.22404

39. Toffola ED, Furini F, Redaelli C, Prestifilippo E, Bejor M. Evaluation and treatment of synkinesis with botulinum toxin following facial nerve palsy. Disabil Rehabil. 2010;32(17):1414-1418.

https://doi.org/10.3109/09638280903514697

40. Choi KH, Rho SH, Lee JM, Jeon JH, Park SY, Kim J. Botulinum toxin injection of both sides of the face to treat post-paralytic facial synkinesis. J Plast Reconstr Aesthet Surg. 2013;66(8):1058-1063.

https://doi.org/10.1016/j.bjps.2013.04.012Received: 27.06.17

burdenko of neurosurgery · issn 0042-8817 (russian ed. print) issn 2313-8254 (english ed. online)...

Documents