The Spine Journal - (2013) -

Clinical Study

Anterior or posterior approach of thoracic disc herniation? A comparativecohort of mini-transthoracic versus transpedicular discectomies

Mark P. Arts, MD, PhDa,*, Ronald H.M.A. Bartels, MD, PhDb

aDepartment of Neurosurgery, Medical Center Haaglanden, PO Box 432, 2501 CK, The Hague, The NetherlandsbDepartment of Neurosurgery, Radboud University Medical Center Nijmegen, PO Box 9101, 6500HB, Nijmegen, The Netherlands

Received 29 March 2013; revised 23 September 2013; accepted 27 September 2013

Abstract BACKGROUND CONTEXT: The optimal su

FDA device/drug

Author disclosures

sored by Amedica C

Paid directly to institu

The disclosure key

TheSpineJournalOnlin

1529-9430/$ - see fro

http://dx.doi.org/10.10

rgical treatment of thoracic disc herniations remainscontroversial and dependson the consistency of the herniation and its location related to the spinal cord.PURPOSE: To compare the outcomes of patients with symptomatic thoracic disc herniations trea-ted with anterolateral mini-transthoracic approach (TTA) versus posterior transpedicular discectomy.STUDY DESIGN: This is a prospective comparative cohort study.PATIENT SAMPLE: One hundred consecutive patients with symptomatic thoracic herniateddiscs were operated by mini-TTA (56 patients) or transpedicular discectomy (44 patients).OUTCOME MEASURES: Neurologic assessment by American Spinal Injury Association(ASIA) Impairment Scale and patients’ self reported perceived recovery and complications.METHODS: The consistency and location of the herniated disc in relation to the spinal cord wasevaluated by preoperative computed tomography and magnetic resonance imaging. Patients wereassessed neurologically before surgery and at regular outpatient controls at 2 months or later.Long-term follow-up was achieved by questionnaires sent by mail.RESULTS: In both groups, most patients had symptoms of myelopathy and radicular pain; pa-tients who underwent mini-TTA, more frequently suffered from spasticity. Fifty-eight percent ofthe herniated discs were calcified and 77% were larger than one-third of the spinal canal. All pa-tients presented with ASIA Grade C or D (64%) or ASIA Grade E (36%). Postoperatively, 50% ofthe patients treated with mini-TTA and 37% of the transpedicular group improved at least one gradeon the ASIA scale (p5.19). The duration of surgery, blood loss, hospital stay, and complication ratewere significantly higher in patients treated with mini-TTA and were mainly related to the magni-tude and consistency of the herniated disc. At long-term follow-up, 72% of the mini-TTA patientsreported good outcome versus 76% of the transpedicular discectomy group (p5.80).CONCLUSIONS: Surgical treatment of a symptomatic herniated disc contributed to a clinical im-provement in most cases. The approach is dependent on the location, the magnitude, and the con-sistency of the herniated thoracic disc. Medially located large calcified discs should be operatedthrough an anterolateral approach, whereas noncalcified or lateral herniated discs can be treatedfrom a posterior approach as well. For optimal treatment of this rare entity, the treatment shouldbe performed in selected centers. � 2013 Elsevier Inc. All rights reserved.

Keywords: Thoracic disc; Herniation; Mini-TTA; Posterolateral; Calcification; Spine; Surgery

Introduction

The first case of thoracic disc herniation with spinal cordcompression was reported by Key in 1838 [1]. The earliestrecord of thoracic disc surgery was carried out by Adson

status: Not applicable.

: MPA: Grants: Cascade trial (F, Cascade trial spon-

orporation). RHMAB: Grants: NutsOhra Fonds (F,

tion/employer).

can be found on the Table of Contents and at www.

e.com.

nt matter � 2013 Elsevier Inc. All rights reserved.

16/j.spinee.2013.09.053

in 1922, who performed a laminectomy and disc removal[2]. Thoracic disc herniations are relatively rare, and surgi-cal treatment comprises less than 1% of all intervertebraldisc surgeries [3–5]. Whenever symptomatic, patients may

* Corresponding author. Department of Neurosurgery, Medical Center

Haaglanden, PO Box 432, 2501 CK, The Hague, The Netherlands. Tel.:

070-3302054; fax: 070-3809459.

E-mail address: m.arts@mchaaglanden.nl (M.P. Arts)

2 M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

present with progressive myelopathy, localized thoracicpain, and/or radicular pain.

Although the entity of thoracic disc herniation is wellknown, the appropriate indications and choice of surgicalstrategy are controversial. Formerly, dorsal decompressionwith laminectomy was the standard procedure in the treat-ment of thoracic disc herniations. Because of poor outcomeand neurologic deterioration, laminectomy was abandonedand alternative approaches were studied [6]. Patterson andArbit [7] introduced the posterolateral transpedicular dis-cectomy without laminectomy that has been shown as safeand effective. This technique has been modified into atransfacet pedicle-sparing approach [8–10]. However, sub-stantial paravertebral muscle dissection is needed to gainaccess to centrally located disc herniations, but even then,adequate ventral dura decompression can be challenging[11,12]. Anterior transthoracic discectomy was advocatedto improve direct ventral access and avoid spinal cord ma-nipulation, although pulmonary complications and intercos-tal neuralgia may occur [6,13,14].

In recent years, extensive open thoracotomy was modifiedinto less invasive techniques, suchas themini-transthoracic ap-proach (mini-TTA) and thoracoscopic discectomy [15–18]. Arecent comparative study has shown benefits in favor of mini-TTA, especially in patients with paramedially located discherniations [19]. To our knowledge, no comparative studybetweenmini-TTA and posterolateral transpedicular approachof thoracic disc herniations has been performed. In the presentcohort, we report the results of 100 consecutive patients withthoracic disc herniationswho underwentmini-TTAor transpe-dicular discectomy.

Materials and methods

Patient population

Between March 2005 and January 2013, 100 consecutivepatients with 106 symptomatic thoracic disc herniations weretreated in two Dutch neurosurgic centers (66 patients in TheHague and 34 patients in Nijmegen). Preoperative and postop-erative neurologic examinations were available of all patients.Symptomswere assessedusing theAmericanSpinal InjuryAs-sociation (ASIA) Impairment Scale. Patients were examinedneurologically during theoutpatient control 2months after sur-gery, ormore oftenwhennecessary (follow-upmoment 1).Thelong-term results were investigated by means of a question-naire sent by mail (follow-up moment 2) that included self-reported recovery, visual analog pain score, and neurologiccomplaints related to thoracic disc herniation. Patients’ selfperceived recovery was measured by the seven-point Likertscale; ‘‘complete recovery,’’ ‘‘almost complete recovery,’’and ‘‘some recovery’’ were determined as good outcome [20].

Radiographic imaging

Preoperative magnetic resonance imaging (MRI) wasperformed in all patients to document the localization of

the herniated disc, relation of the herniated disc to the spi-nal cord, and the size of the herniated disc; less than one-third of the spinal canal diameter, one-third to two-thirdsof the spinal canal diameter, or larger than two-thirds ofthe spinal canal diameter (Figs. 1 and 2, A and B). All pa-tients underwent additional computed tomography (CT) forthe documentation of calcified (Fig. 1, C and D) or noncal-cified herniated discs (Fig. 2, C and D). Postoperative imag-ing of the spinal cord was not used routinely, but only onindication and when additional fusion was performed.

Surgical procedures

Mini-TTAUnder general anesthesia and two-lumen tube ventila-

tion, the patient is placed in lateral decubitus position withthe table tilted. Spinal cord monitoring was not used. In allcases, a right side approach is performed, except for thosepatients with disc herniation at the thoracolumbar junctionwhere the diaphragm on the right side prevents adequateexposure. The involved thoracic disc and adjacent vertebralbodies are drawn on the skin under fluoroscopic control. A6 to 8 cm long skin incision is made parallel to the orienta-tion of the rib. After splitting the fibres of the latissimusdorsi and serratus anterior muscles, the underlying rib isexposed and the mini-TTA spreader (Braun-Aesculap, Mel-sungen, Germany) is positioned between the ribs. Whennecessary, the rib can be removed partially for extensive ex-posure of the thoracic spine. After deflating the lung, thevisceral pleura is opened and the lung is retracted with aninflatable lung retractor. The affected disc is verified fluoro-scopically. Under microscopic magnification, the rib head,cranial part of the pedicle, the inferolateral part of thecranial vertebral body, and the superolateral part of the cau-dal vertebral body are removed with a high-speed drill.After opening the longitudinal ligament, the herniated disccan be mobilized anteriorly away from the spinal cord(Fig. 3) In case of a large calcified herniated disc, substan-tial bony removal may be necessary, and additional instru-mented fusion is performed in the same procedure when anestimated 25% or more of the craniocaudal extent of thevertebral body has been removed. Finally, a chest tube isplaced and the wound is closed in layers [15].

Transpedicular discectomyThe patient is placed in a prone position under general

anesthesia, and no spinal cord monitoring was used. The in-volved thoracic disc is verified fluoroscopically and a mid-line skin incision of approximately 10 cm length is made.The ipsilateral paravertebral muscles are subperiostally dis-sected and retracted laterally to expose the facet joint andpedicle. After fluoroscopic verification of the affected disclevel, the medial part of the inferior and superior facets andthe cranial part of the pedicle are drilled off with a high-speed burr. The disc space lateral to the dural sac is incised,and the disc fragments are removed under microscopic

Fig. 1. Example of a calcified herniated disc. (A) Sagittal and (B) transversal T2 magnetic resonance images of a paramedian herniated disc T9–T10, larger

than two-third of the spinal canal diameter. (C and D) The herniated disc is hypointense on T2 imaging. Computed tomography confirms complete calci-

fication of the herniated disc.

3M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

magnification (Fig. 4) In case of a calcified herniated disc, atrough is drilled in the underlying vertebral body to mobi-lize the herniated disc dorsally. When necessary, partialcostotranversectomy is performed to gain access ventrallyto the dura in case of central disc extrusion. Additional in-strumented fusion was not performed. After adequate spinalcord decompression, the wound is closed in layers over asuction drain.

Statistical analysis

Student paired two-tailed t tests were used to compareparametric data; otherwise Pearson chi-square tests, Fisherexact test, or Mann-Whitney U tests were used. Results arepresented as means6standard deviations. A probability val-ue less than .05 was considered statistically significant.Statistical analysis was performed with SPSS software (ver-sion 20; SPSS Inc., Chicago, IL, USA).

Results

Fifty-six patients underwent mini-TTA and 44 patientswere treated with transpedicular discectomy. Baseline char-acteristics of the patients are presented in Table 1. The levelof the herniated disc ranged from T2–T3 to T12–L1, andthe most common affected disc levels were in the lowerthoracic region. Nearly 80% of the herniated discs werelarger than one-third of the spinal canal and almost 60%of the herniated discs showed calcifications on CT.

Surgical findings and postoperative course

Surgical findings and postoperative course are depictedin Table 2. The mean duration of a mini-TTA procedurewas 229 minutes versus 98 minutes for transpedicular dis-cectomy (p!.001), with substantially more blood loss dur-ing mini-TTA procedure (1,157 mL vs. 213 mL, p!.001).

Fig. 2. Example of a noncalcified herniated disc. (A) Sagittal and (B) transversal T2 magnetic resonance images of paramedian noncalcified herniated disc

T8–T9, smaller than one-third of the spinal canal diameter. (C and D) The herniated disc is hypointense on T2 imaging. Computed tomography documents no

calcification.

4 M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

In the mini-TTA group, 17 patients presented with calcifiedherniated disc adherent to dura versus 3 patients treated bytranspedicular approach (p5.005). In nine patients treatedby mini-TTA, more than 25% of the vertebral body wasdrilled away to remove the herniated disc safely, and addi-tional instrumented fusion was performed.

The mean hospital stay in patients treated with mini-TTAprocedure was significantly longer (10.1 days vs. 4.9 days;p5.03). Pulmonary complications, such as pneumonia andpleural effusions (18%), cerebrospinal fluid leakage requir-ing external drainage and bed rest for a few days (18%),were the main complications. Two patients of the transpe-dicular group were reoperated; one patient developed para-plegia caused by rebleeding and the other patient wasoperated on the wrong level. In the mini-TTA group, threepatients were reoperated; one patient developed a cerebro-spinal fluid fistula to the pleura that needed transthoracic re-intervention and closing the duradefect followed by external

cerebrospinal fluid drainage; one patient had insufficient du-ra decompression requiring repeated mini-TTA; and one pa-tient underwent a mini-TTA for an assumed herniated disc.During surgery an epidural abnormality was not found, andan immediate postoperativeMRI showed an intradural mass.At a second operation through a posterolateral approachwith intradural inspection, a meningioma was identifiedand completely removed.

Clinical outcome

Data was available for 96 patients at follow-up moment 1and 77 patients returned the long-term questionnaires atfollow-up moment 2. The mean (6standard deviation)follow-ups from surgery to moment 1 and moment 2 were34631 weeks and 4.162.1 years, respectively. In themini-TTA group, 28 patients (50%) improved one or moregrades on the ASIA scale versus 15 patients (37%) in the

Fig. 3. Intraoperative microscopic view of right-sided mini-transthoracic approach of a large paramedian calcified herniated disc at T9–T10 (same patient as

in Fig. 1). Parts of the inferolateral cranial and superolateral caudal vertebral bodies are removed with a high-speed drill. (Top Left) A dissector is placed in

the formed cavity. (Top Right) The posterior longitudinal ligament is opened and the calcified herniated disc is mobilized anteriorly away from the spinal

cord. (Bottom) After removal of the calcified herniated disc, the dura is adequately decompressed.

5M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

transpedicular group (p5.19) (Table 3). After the mini-TTAprocedure, three patients worsened one or more grades onthe ASIA scale; the calcified thoracic disc appeared to beossification of the posterior longitudinal ligament in two pa-tients and a meningioma in the other patient. In the transpe-dicular group, one patient worsened neurologically becauseof rebleeding.

On follow-up moment 2, 72% of the mini-TTA patientsreported good outcome versus 76% of the transpedicular dis-cectomy group (p5.80). Visual Analog Score (0–100mm)thoracic back pain in the mini-TTA group and posterolateral

Fig. 4. Intraoperative microscopic view of posterior transpedicular discectomy o

The paravertebral muscles are dissected on the right side and the wound spreader

speed drill to gain access to lateral and ventral of the dura. (Right) The asterisk

group was 30628 mm versus 36632 mm, respectively(p5.38). Other long-term symptoms are shown in Table 4.

Discussion

Initial attempts to decompress the dura via laminectomywere disappointing because of vascular insufficiency andmicrocontusions secondary to spinal cord manipulation.Therefore, variations in surgical techniques were developedto allow a more direct approach to the intervertebral disc,

f a right-sided noncalcified mediolateral herniated disc at T10–T11. (Left)

is placed. The facet, pedicle, and adjacent laminae are reduced with a high-

marks the disc space after removal of the herniated disc.

Table 1

Demographic data of 100 patients with 106 thoracic disc herniations, who underwent mini-TTA or transpedicular discectomy

Baseline characteristics Mini-TTA (N556) Transpedicular (N544) p Value

Female gender 33 (59) 26 (59) .99

Mean age in yrs (range) 52.4 (29–82) 56.9 (28–86) .08

Body Mass Index (range) 27.8 (20–46) 26.6 (20–34) .23

Comorbidities .38

Diabetes 3 (5) 2 (5)

Pulmonary disease 1 (2) 2 (5)

Morbid obesitas 3 (5) 0

Symptoms

Localised thoracic pain 16 (29) 20 (45) .10

Radicular pain 14 (25) 17 (39) .20

Sensory deficit 46 (82) 35 (80) .80

Motor deficit 32 (57) 22 (50) .55

Spasticity/hyperreflexia 42 (75) 24 (55) .04

Miction disturbance 23 (41) 17 (39) .84

Indications for surgery .34

Radiculopathy 13 (23) 14 (32)

Myelopathy 43 (77) 30 (68)

Mean duration of symptoms (wks) (range) 93.4 (1–520) 68.8 (0–312) .25

Level of herniated disc .02

T2T3 0 1 (2)

T4T5 0 4 (9)

T5T6 3 (5) 2 (5)

T6T7 7 (13) 2 (5)

T7T8 7 (13)* 3 (7)x

T8T9 8 (14)*,y 3 (7)x

T9T10 14 (25)*,y 6 (14)

T10T11 10 (18)z 4 (9)

T11T12 5 (9)z 12 (27)x

T12L1 2 (4) 7 (16)

Disc calcification 47 (94) 11 (25) !.001

Axial localization !.001

Paramedian 35 (63) 10 (23)

Mediolateral 19 (34) 22 (50)

Lateral 2 (4) 12 (27)

Size of herniated disc .04

!1/3 of spinal canal 10 (18) 13 (30)

1/3–2/3 of spinal canal 29 (52) 29 (66)

O2/3 of spinal canal 17 (30) 2 (5)

TTA, transthoracic approach.

Numbers in parentheses are percentages.

* One patient had a herniated disc at three levels.y One patient had a herniated disc at two levels.z One patient had a herniated disc at two levels.x One patient had a herniated disc at three levels.

6 M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

such as ventral transthoracic and posterolateral transpedic-ular. The best surgical treatment of thoracic disc herniationsremains controversial. Comparative studies on conventionaltransthoracic access versus posterior decompression havebeen published previously [10]. However, to our knowl-edge, our study is the first comparative cohort of patientswith symptomatic thoracic disc herniations treated withmini-TTA versus transpedicular discectomy.

The clinical outcome of mini-TTA and posterolateral dis-cectomy is favorable in both groups, and more than 70% ofthe patients reported good outcome on short and long terms.This study confirmed that the choice of approach is depend-ent on the consistency of the herniated disc, its size, and itslocation related to the spinal cord. Especially calcified, larger

and medially located herniated discs were approachedthrough a mini-TTA. However, the complication rate is sub-stantial and many of these are directly related to the TTA.The complication rate of 38% in the mini-TTA group inour study is high, but in accordance to other recently reportedpapers on thoracic disc surgery [8,10,13,17]. Cerebrospinalfluid leakage was the most frequently reported complicationin patients treated with mini-TTA that is directly related tothe calcifications of the herniated disc adherent to the dura.The significantly less frequent complication rate in the post-erolateral group can be explained by smaller and less oftencalcified disc herniations. Pulmonary related morbidity isthe other main complication of the mini-TTA procedure thatis related to opening the pleura and retraction on the deflated

Table 2

Operative characteristics

Surgical characteristics Mini-TTA (N556) Transpedicular (N544) p Value

Duration of surgery (min) (range) 229 (90–510) 98 (30–220) !.001

Blood loss (mL) (range) 1,157 (100–12,000) 213 (10–600) !.001

Instrumented fusion 9 (16) 0 .004

Disc adherent to dura 17 (30) 3 (7) .005

Intradural disc extension 4 (7) 1 (2) .38

Duration of chest tube (d) (range) 2.2 (0–11) NA !.001

Duration of stay ICU (d) (range) 1.0 (0–5) NA !.001

Hospital stay (d) (range) 10.1 (3–70) 4.9 (2–30) .03

Total complications 21 (38)* 2 (5) !.001

Neurologic deterioration 2 (4) 1 (2)

Cerebrospinal fluid leakage 10 (18) 0

Pneumonia/pleural effusion 10 (18) 0

Wrong level surgery 0 1 (2)

Repeated surgery 3 (5) 2 (5) .85

Wrong level surgery 0 1

Insufficient dura decompression 2 0

Rebleeding 0 1

Cerebrospinal fluid fistula 1 0

ICU, intensive care unit; TTA, transthoracic approach; NA, not applicable; SD, standard deviation.

Numbers presented are means (with ranges), means6SD, and percentages (parentheses).

* Two patients had multiple complications.

Table 4

7M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

lung. Possibly, this complication could be reduced by analternative extrapleural approach of the spinal column inwhich the pleura is dissected from the rib and remains intact,and therefore, no chest tube is required [21]. However, theuse of a chest tube in our most recent cases was abandoned,as our pulmonary surgeons convinced us that it was not nec-essary because the lung tissue was not violated. These pa-tients were also directly transferred to the ward with anintermittent stay at the intensive care unit. The higher com-plication rate with mini-TTA could explain the longer stay inthe hospital for these patients in addition to the preopera-tively more severe symptoms in the group.

The long-term results of patients with surgically treateddisc herniations are favourable. Although most patients pre-sented with spinal cord symptoms, which are often regardedas permanent disability, more than 70% of the patients inboth groups reported good outcome, ranging from complete

Table 3

Neurologic status measured by the ASIA classification.

ASIA grade

Mini-TTA Transpedicular

Preoperative

(%)

Postoperative

(%)

Preoperative

(%)

Postoperative

(%)

A 0 0 0 0

B 0 2 (4) 0 0

C 11 (20) 1 (2) 11 (25) 3 (7)

D 28 (50) 12 (21) 14 (32) 10 (23)

E 17 (30) 41 (73) 19 (43) 28 (64)

ASIA, American Spinal Injury Association; TTA, transthoracic ap-

proach.

In the mini-TTA group, 28 patients (50%) improved one or more grades

on the ASIA scale versus 15 patients (37%) in the transpedicular group

(p5.19).

recovery to some recovery. However, most patients reportedpersistent sensory and motor deficits with thoracic backpain. These long-term results together with potential com-plications should be discussed in detail with patients whileobtaining consent.

The main limitation of the present study is selection biasand lack of randomization. Based on the presence of disccalcification and axial localization, patients were selectedfor mini-TTA or transpedicular discectomy. Large parame-dian calcified discs were nearly always operated throughmini-TTA, and patients with smaller lateral herniated discswere mostly operated by a posterior approach. Possibly,two different patient populations were studied. Ideally, fu-ture randomized controlled studies seemed to be warranted.However, in our opinion, a golden standard will be difficult

Long-term symptoms of 77 patients, reported on the questionnaire sent by

mail.

Outcome

Mini-TTA

(N544)

Transpedicular

(N533) p Value

Mean FU in (mo) (range) 29.1 (3–64) 26.6 (4–64) .51

VAS thoracic pain (mm)

(range)

29.9 (0–100) 36.1 (0–92) .38

Good outcome 32 (72) 25 (76) .80

Symptoms

Local thoracic back pain 9 (20) 13 (39)

Radicular pain 18 (41) 9 (27)

Sensory deficit legs 29 (66) 16 (48)

Motor deficit legs 19 (43) 12 (36)

Pulmonary discomfort 6 (14) 0

TTA, transthoracic approach; FU, follow-up; VAS, Visual Analog

Score.

Likert good recovery is defined as ‘‘complete recovery,’’ ‘‘almost com-

plete recovery,’’ and ‘‘some recovery.’’

Fig. 5. Postoperative (Left) X-ray, (Middle) computed tomography, and (Right) magnetic resonance image of mini-transthoracic approach removal of a large

calcified herniated disc T9–T10, with supplemental instrumentation (same patient as in Figs. 1 and 3). (Left) Whenever more than 25% of the craniocaudal

extent of the vertebral body had to be removed, additional anterior bicortical screw and rod fixation was performed to prevent iatrogenic scoliosis. (Middle

and Right) The herniated disc was removed completely and the spinal cord was adequately decompressed.

8 M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

to obtain through randomized controlled trials because theapproach will be defined not only by the surgeon’s prefer-ence, but, more importantly, by the consistency of the her-niated disc and its relation to the spinal cord. We areconfident that a randomized trial on posterior approach ver-sus anterolateral approach for large calcified and mediallylocated herniated thoracic discs will not be approved.The optimal approach for laterally located disc herniationon the other hand, is still open for discussion. However,considering the risk of complications and duration of hos-pital stay, there seems to be a slight preference for a post-erolateral approach based on the results of the presentstudy. This is in agreement with the data of Bransfordet al. [10].

The surgical treatment of patients with symptomaticthoracic disc herniations can be challenging and may be as-sociated with varying risks. Therefore, these complex spinalprocedures should be concentrated in supraregional ex-perienced spine centers with high-volume patient care.The treatment algorithm for the optimal surgical approachshould consist of MRI, CT, and conventional radiographof the lumbar spine. The latter is mandatory for documenta-tion of possible transition spine, which is one of the pitfallsof wrong level surgery. Additional CT focusing on axial lo-calization and presence of calcifications is also essential be-cause MRIs of calcified and noncalcified herniated discs canbe similar (Figs. 1 and 2). Based on our experience, patientswith large calcified paramedian herniated discs should betreated with mini-TTA. All other patients can be treatedby posterolateral approach.

There is no consensus whether routine instrumentationis needed in patients with thoracic herniated discs[10,13,17,18]. We have instrumented 9 of 56 patients ofthe mini-TTA group and none of the posterior group. When-ever more than 25% of the craniocaudal extent of the adja-cent vertebral body was reduced to remove the herniateddisc safely, we performed an additional fusion with place-ment of bicortical screws and rod fixation (Fig. 5). In thisway, iatrogenic scoliosis will be prevented. In our opinion,supplemental fusion is not needed in patients with posteriorremoval of thoracic disc herniation because the unilateralpedicle and facet joints are only partially reduced and insta-bility is not likely. Also, bony removal of the adjacent verte-bral bodies is limited in case of noncalcified or lateralherniated discs.

In conclusion, the long-term surgical outcome of thora-cic disc herniation was favourable in most cases treatedwith mini-TTA and posterolateral discectomy. The compli-cation rate of transthoracic procedures was higher andmainly related to pulmonary morbidity, and the neurologiccomplications were not different between both approaches.Patients with calcified herniated discs should be informedabout the additional risk of cerebrospinal fluid leakage.Based on this experience, it is our opinion that large calci-fied paramedian herniated discs are best treated by an ante-rior approach, whereas noncalcified or lateral herniateddiscs can be treated from posterior as well. Because the en-tity is rare, its treatment should be concentrated in experi-enced centers to optimize the surgical strategy for eachspecific thoracic herniated disc.

9M.P. Arts and R.H.A. Bartels / The Spine Journal - (2013) -

References

[1] Patwardhan RV, Hadley MN. History of surgery for ruptured disk.

Neurosurg Clin N Am 2001;12:173–9.

[2] Love JG, Schorn VG. Thoracic disk protrusions. Rheumatism

1967;23:2–10.

[3] Dietze DD Jr, Fessler RG. Thoracic disc herniations. Neurosurg Clin

N Am 1993;4:75–90.

[4] McInerney J, Ball PA. The pathophysiology of thoracic disc disease.

Neurosurg Focus 2000;9:e1.

[5] Stillerman CB, Chen TC, Couldwell WT, et al. Experience in the sur-

gical management of 82 symptomatic herniated thoracic discs and re-

view of the literature. J Neurosurg 1998;88:623–33.

[6] McCormick WE, Will SF, Benzel EC. Surgery for thoracic disc dis-

ease. Complication avoidance: overview and management. Neurosurg

Focus 2000;9:e13.

[7] Patterson RH Jr, Arbit E. A surgical approach through the pedicle to

protruded thoracic discs. J Neurosurg 1978;48:768–72.

[8] Bransford R, Zhang F, Bellabarba C, et al. Early experience treating

thoracic disc herniations using a modified transfacet pedicle-sparing

decompression and fusion. J Neurosurg Spine 2010;12:221–31.

[9] Stillerman CB, Chen TC, Day JD, et al. The transfacet pedicle-sparing

approach for thoracic disc removal: cadaveric morphometric analysis

and preliminary clinical experience. J Neurosurg 1995;83:971–6.

[10] Bransford RJ, Zhang F, Bellabarba C, Lee MJ. Treating thoracic-disc

herniations: do we always have to go anteriorly? Evid Based Spine

Care J 2010;1:21–8.

[11] Bilsky MH. Transpedicular approach for thoracic disc herniations.

Neurosurg Focus 2000;9:e3.

[12] Le Roux PD, Haglund MM, Harris AB. Thoracic disc disease: expe-

rience with the transpedicular approach in twenty consecutive pa-

tients. Neurosurgery 1993;33:58–66.

[13] Ayhan S, Nelson C, Gok B, et al. Transthoracic surgical treatment for

centrally located thoracic disc herniations presenting with myelop-

athy: a 5-year institutional experience. J Spinal Disord Tech 2010;23:

79–88.

[14] Vollmer DG, Simmons NE. Transthoracic approaches to thoracic disc

herniations. Neurosurg Focus 2000;9:e8.

[15] Mayer HM. Microsurgical anterior approach to T5-T10 (Mini-TTA).

Minimally invasive spine surgery. Springer, 2006:129–37.

[16] Quint U, Bordon G, Preissl I, et al. Thoracoscopic treatment for single

level symptomatic thoracic disc herniation: a prospective followed co-

hort study in a group of 167 consecutive cases. Eur Spine J 2012;21:

637–45.

[17] Russo A, Balamurali G, Nowicki R, Boszczyk BM. Anterior thoracic

foraminotomy through mini-thoracotomy for the treatment of giant

thoracic disc herniations. Eur Spine J 2012;21:S212–20.

[18] Uribe JS, Smith WD, Pimenta L, et al. Minimally invasive lateral ap-

proach for symptomatic thoracic disc herniation: initial multicenter

clinical experience. J Neurosurg Spine 2012;16:264–79.

[19] Bartels RH, Peul WC. Mini-thoracotomy or thoracoscopic treatment

for medially located thoracic herniated disc? Spine 2007;32:E581–4.

[20] Bombardier C. Outcome assessments in the evaluation of treatment

of spinal disorders: summary and general recommendations. Spine

2000;25:3100–3.

[21] Moran C, Ali Z, McEvoy L, Bolger C. Mini-open retropleural trans-

thoracic approach for the treatment of giant thoracic disc herniation.

Spine 2012;37:E1079–84.