7ournal ofNeurology, Neurosurgery, and Psychiatry 1994;57:1433-1435

SHORT REPORT

Anterior spinal hernia: an increasingly recognisedcause of thoracic cord dysfunction

Barrie D White, John L Firth

Department ofNeurosurgery,University Hospital,Queen's MedicalCentre, Nottingham,UKB D WhiteJ L FirthCorrespondence to:Dr Barrie D White,Department ofNeurosurgery, UniversityHospital, Queen's MedicalCentre, NottinghamNG7 2UH, UK.Received 7 December 1993and in revised form7 April 1994.Accepted 3 May 1994

AbstractTwo cases of anterior spinal hernia arepresented. The medical literature isreviewed, the syndrome characterised,and its cause and treatment discussed.The patient is typically middle aged witha history of stepwise slowly progressivemid-thoracic anterior hemicord syn-drome manifesting as hemianalgesiabelow the affected segment, followed bycontralateral lower limb spasticity thatdevelops into an asymmetric paraparesiswith sparing of dorsal column sensation.Radiological investigation demonstratesan enlarged dorsal arachnoid space inassociation with an apparently focallynarrowed thoracic cord, kinked towardsthe anterior dura. At operation the cord

Figure1 Left; patient 1, preoperative Tl weighted spin echo sagittalMRI showingfocalanterior cord tethering at D4. Right; patient 2, preoperative Tl weighted spin echo sagittalMRI showingfocal anterior cord tethering at D8.

is found to be prolapsed into an antero-lateral dural diverticulum. The mostlikely cause of this syndrome is anteriorspinal artery segmental branchischaemia, in a cord chronically incar-cerated in a congenital anteriormeningocele. This readily treatable con-dition should be considered in all cases ofthoracic cord dysfunction and surgicalrepair effected early to prevent stepwiseprogression to paraplegia.

(i Neurol Neurosurg Psychiatry 1994;57: 1433-1435)

Thoracic spinal cord hernia is a rare condi-tion presenting in middle age with a typicalhistory of anterior hemicord dysfunction pre-ceding progressive paraparesis with preserva-tion of posterior column sensation. Theradiological findings, although subtle, aretypical and surgical repair is easily effected.This condition is probably more commonthan currently appreciated and should beconsidered in all patients presenting withprogressive thoracic cord dysfunction.

Case reportsPATIENT 1In 1987 a 61 year old woman, suddenlydeveloped a "tender warm tingling)) sensationin the lateral aspect of her left leg. In 1988,after a fall caused by her legs transiently "giv-ing way", she was found to have reduced painand temperature sensation on the left with anupper level of T6, but otherwise normal neu-rology. During the next six months, shenoticed her right leg becoming stiff; withimpairment of walking to the point that asingle stick was needed. Neurological followup over the same period confirmed increasinglower limb spasticity, then motor deteriora-tion to MRC4 in the right leg without tactileor proprioceptive impairment.

Magnetic resonance imaging showed thespinal cord to be acutely kinked and closelyapplied to the D4 vertebral body in associa-tion with dorsally placed arachnoid diverticu-lae (fig 1 left).

At operation, the arachnoid appearedthickened in the region of the abnormalityand on filling with CSF during expiration thesubarachnoid space was seen to contain mul-tilocular dorsal arachnoid diverticulae. On

1433 on F

ebruary 1, 2022 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1433 on 1 Novem

ber 1994. Dow

nloaded from

White, Firth

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~p

Figure 2 Top; patient 1, photograph at operation showing cord hernia deliveredfromanterior sharp edged oval dural diverticulum. Edge just visible to left of hernia. Bottom;patient 2, operative photograph at operation showing cord hernia deliveredfrom anteriorsharp edged oval dural diverticulum. Edge just visible to right of hernia.

opening the arachnoid the anterior cord wasfound to have prolapsed into, and bedeformed and tightly held by a sharp edged,wide necked oval dural diverticulum slightlyto the right of midline. The cord was gentlydelivered (fig 2 top) and to prevent recurrentprolapse a sling of fascia taken from thetrapezius was passed anterior to the cord andfixed with interrupted sutures to obliterate

Figure 3 Left; patient 1, postoperative Tl weighted spin echo sagittal MRI showing cordrealignment but persistent focal cord abnormality. Right; patient 2, postoperative Tlweighted spin echo sagittal MRI showing cord realignment but persistantfocal cordabnormality.

the dural opening while simultaneously actingas a hammock for the abnormal cord.

Postoperatively, the patient remainsunchanged, analgesic below T6 on the leftwith an asymmetric spastic paraparesis affect-ing the right leg more than the left, but withproprioception preserved. Magnetic reso-nance imaging confirmed that the cord hadbeen realigned although it remained focallyabnormal (fig 3 left).

PATIENT 2A thirty nine year old man presented in 1991,having noticed an inability to appreciate withhis left leg the heat of a radiator against whichhe was leaning. Neurological assessment con-firmed an isolated left sided spinothalamicloss with a level at T10. Stepwise progressionoccurred over a period of 18 months to a stiffclumsy right leg that prevented him running.Clinically during this time he developed aright anterior hemicord syndrome withpreserved proprioception despite spinothala-mic loss on the left below T10, right lowerlimb hypertonia and hyper-reflexia, ankleclonus, and an upgoing plantar (but no motorweakness).

Examination by MRI showed the abruptlykinked cord distorted towards the right ante-rior border of the thoracic canal at D8 and anenlarged dorsal arachnoid space (fig 1 right).Exploration confirmed the nature of theabnormality and allowed delivery and repairof the cord hernia as before (fig 2 bottom).Postoperatively there has been no change inneurological status. Magnetic resonanceimaging confirmed realignment of the cordbut showed a syrinx at the level of the originalprolapse (fig 3 right).

Review ofthe medical literatureFour additional cases of anterior thoracicspinal cord hernia have been published since1990 but the clinical syndrome has not previ-ously been characterised or the cause of thelesion critically analysed.

In 1990, Oe et al presented a 61 year oldman with a 10 year history of progressive T5hemicord syndrome leading to an asymmetricparaparesis.' Full investigation showed thatthe thoracic cord was kinked anteriorly andclosely applied to the vertebral body of D4. Apreoperative diagnosis of dorsal arachnoidcyst was substantiated at operation, duringwhich an anterior dural cleft with cord herniawas also found. Both were repaired but withno improvement in the patient's longstandingdisability.

In 1991, Isu et al published two cases ofanterior cord hernia.2 One was a 43 year oldwoman with a one year history of altered painsensation below T5 on the right and a hyper-reflexic left leg. The other was a 45 year oldwoman with a 20 month history of right legweakness progressing to paraparesis. Investi-gation showed anterior cord displacementwith enlarged dorsal subarachnoid spaces inboth cases.

Operation confirmed the presence of

1434 on F

ebruary 1, 2022 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1433 on 1 Novem

ber 1994. Dow

nloaded from

Anterior spinal hernia: an increasingly recognised cause of thoracic cord dysfunction

dorsally situated arachnoid cysts and ventro-lateral dural defects with cord herniation. Inboth cases, surgery was confined to removalof the dorsal arachnoid cysts without duralrepair. The authors postulate that the ventraldural dehiscence was the result of pressuretransmitted through the cord from the dorsalintradural arachnoid cyst.

Later the same year, Tronnier et al pre-sented a 45 year old woman with a four yearhistory of right sided loss of pain and temper-ature sensation below T5.3 Magnetic reso-nance imaging showed the cord narrowedand closely applied to the vertebral body atD3/4, interpreted as a "...dorsally located cys-tic lesion displacing the spinal cord anteri-orly...." At operation a left anterolateral duraldehiscence with incarcerated cord hernia wasfound. The hernia was reduced and thedefect repaired with a dural cuff held by fibringlue. The authors attributed the dural split toforgotten trauma.

DiscussionSix cases of this syndrome now exist in themedical literature, each with identical clinical,radiological, and surgical features. The aetiol-ogy of the dural lesion remains in doubt withprevious authors either offering no explana-tion or invoking forgotten trauma or pressureerosion due to dorsal intradural arachnoidcysts.

Dural defects may follow trauma, but thelikelihood of this passing unnoticed in thethoracic region is small because of the sever-ity of injury required and almost invariableoccurrence of neurological damage.

Similarly, it seems inconceivable that pres-sure sufficient to erode the dura would not, ifplausible, be more commonly associated withsolid intradural tumours. Furthermore, unlikeextramedullary cord compression, whichcauses motor then global sensory change, thepresenting syndrome in each of these casesrelates specifically to the contents of the her-nia with (at least in the initial stages) dis-turbed contralateral spinothalamic sensation,ipsilateral motor loss, and preserved dorsalcolumn sensation.The association of anterior hernia with

dorsal arachnoid thickening requires explana-tion, but it would seem more reasonable tosuggest that the arachnoid changes may besecondary to inflammation resulting fromincarceration of the cord.On the evidence available, it seems most

likely that this condition is a consequence ofcord prolapse into a congenital dural defect.The absolute rarity of true anterior spinabifida, the absence of somitic vertebral abnor-malities in these cases, and the clinical pre-sentation of cord herniation in middle age,however, argues against even this as a cause.One possibility for the late presentation of

this syndrome, its stepwise progression, andfailure to improve after cord reduction is thatthe hernia is longstanding but only becomessymptomatic as a result of superaddedischaemic events affecting small distorted seg-

mental branches of the anterior spinal arteryin middle age.

In support of a congenital basis for thislesion is the original description by Wortzmanet al, who in 1974 described a similar butmore severe case involving thoracic cordprolapse into a corticated cavity likely to be aneurenteric canal remnant within the D7vertebral body.4The only other report cited is that of

Matsuzawa et al.5 This case differs from theothers in its clinical presentation, radiology,and surgical findings and seems to be aunique report of a distinct clinical entity ofspontaneous cord herniation into a conven-tionally sited extradural arachnoid diverticu-lum.

ConclusionsAlthough the pathogenesis of the dural lesionhas yet to be settled, the syndrome, whichdiffers subtly from other causes of thoraciccord dysfunction, is well characterised bythese cases.

Affecting both men and women in theirmiddle years, it commences as a unilateralspinothalamic tract disturbance with anupper level between T4 and T10, and pro-gresses slowly but stepwise over many monthsto an asymmetric paraparesis with preserveddorsal column sensation.

Imaging shows a narrowed, acutely kinkedcord, closely applied to the vertebral body, inassociation with dorsal arachnoid abnormali-ties.

Surgery to deliver the cord and repair thedural defect results in stabilisation of anotherwise stepwise progressive cord vascularsyndrome and should be performed as soonas the condition is suspected despite the rela-tively innocuous early symptoms and signs.Of major importance is the appreciation

that low pressure dorsal arachnoid abnormali-ties may be a marker of, or associated withventral spinal herniation and to differentiatethis clinically, radiographically, and surgicallyfrom a truly symptomatic high pressurearachnoid cyst. Failure to do so may result ininadequate surgical treatment and persistenceof the (quite literally) underlying causativelesion.Our sincere thanks are extended to Dr Alan M Whitely, DrDavid Jefferson (consultant neurologists) Dr Ian Holland, DrTim Jaspan, and Dr Geoff Narborough (consultant neurora-diologists) for their invaluable help with the care, diagnosis,and rehabilitation of these patients.

1 Oe T, Hoshino Y, Kurokawa T. A case of idiopathic her-niation of the spinal cord associated with duplicateddura mater and with an arachnoid cyst. J Jpn OrthopAssoc 1990;64:43-9.

2 Isu T, lizuka T, Iwasaki Y, Nagashima M, Akino M, AbeH. Spinal cord hemiation associated with an intraduralspinal arachnoid cyst diagnosed by magnetic resonanceimaging. Neurosurgery 199 l;29: 137-9.

3 Tronnier VM, Steinmetz A, Albert FK, ScharfJ, Kunze S.Hemia of the spinal cord: case report and review of theliterature. Neurosurgery 1991;29:916-9.

4 Wortzman G, Tasker RR, Rewcastle B, Richardson JC,Pearson FG. Spontaneous incarcerated hemiation of thespinal cord into a vertebral body: a unique cause ofparaplegia. J Neurosurg 1974;41:631-5.

5 Masuzawa H, Nakayama H, Shitara N, Suzuki T. Spinalcord herniation into a congenital extradural arachnoidcyst causing Brown-Sequard syndrome. Y Neurosurg1981 ;55:983-6.

1435

on February 1, 2022 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.57.11.1433 on 1 N

ovember 1994. D

ownloaded from