Postgraduate Medical Journal (August 1972) 48, 496--500.

Myelomeningocele

H. B. ECKSTEINM.A., M.D., M.Chir., F.R.C.S.

The Hospital.for Sick Children, Great Ormond Street, London, W.C. 1, andQueen Mary's Hospitalfor Children, Carshalton, Surrey

IntroductionMyelomeningocele has, in the last 10 years, become

one of the most important and commonest problemsthat a paediatric surgeon has to deal with. Theabnormality has greatly affected the entire hospitalservice in relation to children and has made a greatimpact on the social services. There is no goodevidence that this particular abnormality has becomemore common but there are a number of perfectlyvalid reasons why children with myelomeningoceleshould present a major problem in 1970 when theydid not do so some 20 years ago. The simple answeris that many more of the infants born with this par-ticular abnormality are now surviving than in thepast. It is generally estimated that without treatment,about 10% of myelomeningocele infants will survivefor at least a year (Laurence & Tew, 1971), and asmall but significant number survive much longer;grossly deformed and often mentally retarded.The reasons for the increased survival rate include

the following:(a) The overall advances in paediatric surgery and

especially in neonatal anaesthesia.(b) The introduction and development of anti-

biotics which can treat or prevent meningitis whichused to be a common cause of death in this group ofpatients.

(c) The development and introduction of ventri-culo-atrial drainage using shunts by Holter (Macnab,1963) and Pudenz (1966).

(d) The concept introduced originally by Sharrardand his colleagues in 1962 (Sharrard et al., 1963) thatemergency closure of the spinal defect within 24 hrof birth produced not only a higher survival ratebut also a better quality of survivor. They claimedthat early closure prevented ascending meningitis,reduced the degree of paralysis and had no adverseinfluence on the development of hydrocephalus.

Largely as the result of the Sheffield school ofteaching, emergency closure of myelomeningocelebecame an accepted and standard practice and themortality has dropped to around 300 (Eckstein &Macnab, 1966). It is noted that of those childrenwho die, 900 will die before their second birthdayand indeed the large majority of these will die withinthe first 6 months.

The whole problem of myelomeningocele patientshas been highlighted by a recent publication ofLorber (1971) analysing the results of so-called earlyclosure and the fact that the large majority of thesurvivors of this treatment are grossly handicapped.A not insignificant proportion is mentally retarded.Lorber, like others before him, therefore suggestsselection for treatment and gives criteria for such aselective process. The idea of selection, while soundat first sight, presents a great many problems. Whilesimple facts such as hydrocephalus at birth, com-plete paraplegia and obvious kypho-scoliosis areregarded as contra-indications to surgery none ofthese criteria can be regarded as absolute. In myexperience, quite a number of infants who had ob-vious hydrocephalus at birth and were treated with aventriculo-atrial shunt at the same time as theirmyelomeningocele was repaired, have done remark-ably well. Complete paraplegia by itself cannot reallybe regarded as an absolute contra-indication totreatment as the totally paralysed infant or childmay be far better off than his partially paralysedcounterpart who is likely to have multiple lower limbdeformities caused by muscle imbalance. Those whoadvocate selection fail to recognize the simple factthat a great many of the so-called untreated myelo-meningocele patients will survive for many monthsor even years and cause a great deal of hardship totheir parents and their families. If selection is to bepractised no hard-and-fast rules can be made andthe individual infant must be assessed in relation tohis individual family environment. A myelomeningo-cele baby of a 16-year-old unmarried mother maywell be handled differently from the clinically identi-cal infant of a couple who have tried for many yearsto have a child.

MyelomeningoceleThe actual closure of the myelomeningocele is

relatively straightforward. The neural plaque shouldbe isolated and preserved but all membranous tissueshould be excised so as to leave a clean skin edge.Primary closure of the dura is usually possible, andfollowing wide mobilization to the flanks the skincan usually be closed producing a vertical sutureline (Eckstein, 1971). Rotation flaps should be

copyright. on A

pril 21, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.48.562.496 on 1 A

ugust 1972. Dow

nloaded from

Myelomeningocele 497

avoided as these tend to slough and in any eventsubsequent scars will interfere with the wearing ofcalipers and urinary appliances. Muscle flaps arecontra-indicated as these will tend to increase kypho-scoliosis later on, but in the occasional patient, aspinal osteotomy at the initial operation is justifiableto correct gross kyphosis and to allow primary skinclosure (Eckstein & Vora, 1971), (Figs. 1 and 2).

HydrocephalusBetween 80 and 90% of myelomeningocele patients

will develop progressive hydrocephalus. The signsand symptoms of this complication or associatedmalformation include widening of the sutures, atense fontanelle, rapidly increasing skull circum-ference, vomiting, drowsiness or a general failure torespond to stimuli. The diagnosis of hydrocephaluscan obviously be confirmed by ventriculography butthis procedure is not strictly necessary as a confident

..'i'''

-i

i·

··*

·. ·: ···iii:i

·s'I..

FIG. 1. Myelomeningocele with gross kyphosis.

e

FIG. 2. Same patient after spinal osteotomy and primaryskin closure; note that shunt has also been inserted atthe same time.

diagnosis can usually be made on clinical grounds.The one essential pre-operative investigation beforeshunt therapy is an analysis of a specimen of cerebro-spinal fluid with particular reference to its proteincontent on the one hand and to its sterility on theother. While a high protein content (1000 mg/100ml) is no absolute contra-indication to shunt therapy,it is widely recognized that ventriculo-atrial shuntswill not function satisfactorily in the presence of ahigh protein content.We have found it useful in this situation to set up

an experimental drainage system (using a discardedbut functioning valve) with the patient's CSF andto assess the pressure at which the valve will func-tion when the protein content of the CSF is raised.The effectiveness of the shunt is not directly corre-lated to the protein content of the CSF. Sterility ofthe CSF is obviously essential before a ventriculo-vascular shunt can be performed. Progressive hydro-cephalus associated with active ventriculitis has tobe managed by daily ventricular taps or by conti-nuous drainage; neither procedure is without riskand a severe electrolyte depletion may result fromprolonged external drainage of CSF.The standard treatment of hydrocephalus today

is the use of ventriculo-atrial shunts usually employ-ing either the Holter or Pudenz system. Both shuntsystems are similar in their working principlesalthough their external appearance differs consider-ably. There is probably little to choose between thesetypes of shunt although the writer has had con-siderably greater success using the Holter system.Both in the U.K. and on the Continent, paediatricsurgeons tend to use the Holter system while neuro-surgeons use the Pudenz system; the complicationand failure-rate of large reported series is almostidentical in both situations. The insertion of aventriculo-atrial shunt is relatively simple and carriesa negligible mortality but the complication rate isalarmingly high (Tsingoglu & Forrest, 1971).Approximately 12% of shunt-treated patients willdevelop bacteraemia due to colonization of the valveusually with Staphylococcus albus. Such bacteraemiais best treated by immediate shunt-replacement underfull antibiotic cover (Nicholas, Kamal & Eckstein,1970). Blockage of the shunt is common and mayoccur either at the proximal or distal end. There isa tendency for the choroid plexus to grow into theventricular catheter and to block this and the useof the Rickham reservoir is said to reduce suchblockage (Rickham, 1964). In any event, blockage ofa ventricular catheter is relatively simple to correctsurgically. The distal end of the catheter is likely tobecome blocked when, as a result of growth, it comesout of the right atrium into the superior vena cava.Clot formation then takes place around the catheterwhich seals it off from the circulation. Prophylactic

copyright. on A

pril 21, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.48.562.496 on 1 A

ugust 1972. Dow

nloaded from

498 H. B. Eckstein

lengthening of the distal catheter has been recom-mended to avoid blockage of the distal end of theshunt. The position of the distal end of the shuntcan be checked at all times, especially with the Holtersystem, as the tubing is radio-opaque. Other com-plications of shunt therapy such as perforationof theheart are uncommon and from a purely surgicalpoint of view the technical problems start when allfour jugular veins have been used (Dickson et al.,1969). In such an event, the CSF can be shuntedeither into the peritoneal cavity, the pleural cavity orthe heart, using either the azygos vein as a means ofaccess or a direct cardiotomy. At least half the shunt-treated patients will require further surgery for theirshunt.

Orthopaedic problemsThe large majority of myelomeningocele patients

will have a greater or lesser degree of paralysis in thelegs, often associated with a considerable amount ofmuscle imbalance and subsequent deformity. Thechild with total paraplegia usually has straight legsand can be fitted with calipers relatively easily.Children with partial paralysis usually require mul-tiple orthopaedic operative, procedures to correctdeformities such as talipes equinovarus, calcaneusdeformity, genurecurvatum, fixed flexion of theknees or dislocation of the hip. In children with dis-location of the hip-joint, it is particularly importantto remember that such dislocation is paralytic andnot truly congenital and that conservative methods.by manipulation or splintage are unlikely to succeed(Walker, 1968). While many myelomeningocelepatients will ultimately be confined to wheelchairsit is a generally accepted policy that all children withmyelomeningocele should be given the opportunityto walk in calipers provided -their intelligence isadequate, and multiple orthopaedic surgical pro-cedures are justifiable to make a particular patientsuitable for ambulation. Research into the develop-ment and design of calipers is urgently needed.

Urinary continenceAlthough a previous study of mine had suggested

that over 30/0 of myelomeningocele patients woulddevelop normal urinary control (Eckstein, 1968), itis now clear that that particular series was selectedand it is now generally agreed that at the most, 10%/of myelomeningocele children will develop normalcontrol. The problems associated with the neuro-genic bladder are therefore important aspects of thiscongenital abnormality. Intravenous pyelographyshould be performed soon after birth and thereafterat 2-yearly intervals to detect upper tract dilatation.Regular urinalysis is essential to diagnose urinarytract infection so that this can be treatedwithsuitableantibiotics. There is some evidence that such infec-

tion may in many of these particular patients belocalized to the bladder only and in this situation,antibiotic therapy is indeed not necessary. On theother hand, if there is any clinical evidence of uppertract infection or radiological evidence of upper tractdeterioration, then prolonged antibiotic therapy isessential. Upper urinary tract dilatation in infancycanoften be arrested or improved by a transurethralresection of the external sphincter (Figs. 3 and 4).It should be noted that the internal sphincter orbladder neck in this particular group of patients isusually widely open and a bladder neck resection istherefore not only unnecessary but positively contra-indicated. The majority of male children with incon-tinence can ultimately be managed with a penileappliance, provided the penis is of adequate size.Since no suitable appliance exists for the female,incontinent females have to have a urinary diversionto the skin using either a loop of ileum or colon toprovide a suitable conduit. While the present methodsof urinary diversion, whether cutaneous ureter-ostomy or ileal conduit, carry a significant propor-tion of early complications, such diversions appearto function well for at least 5 years and possibly upto 10 years. There is, to date, insufficient informationto allow us to predict the function of such conduitsfor periods exceeding 10 years.

I-:.·

·.·: ·i

i·:::··:.

:::·

4:(":

.I:.

:'P...idl· ·,

···I:i·:··..i.i

.:ii:·:i,·:::: ·· ·.i.. ;:::ii::

.,iiii'··I·: .:

·i i:

i·l

FIG. 3. Excretion cystogram showing upper tract dilata-tion.

copyright. on A

pril 21, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.48.562.496 on 1 A

ugust 1972. Dow

nloaded from

Myelomeningocele 499

..:,....r. i· ..····;:-s··--?ei;-: ''· ,,,,,u:-i·;.r*,,,,,.. ;;:..I.-;I::: ·;... ";;;-·";:·*;":

··::';;· ·· ·I ·· " ; :I;"''

:;il··i'l

i·:·.:i.H

i"*i:llt:'

i.'c:·

;.i.:iiiiil.:i

···.;I·· ..·:: ····;sr···.r..ii;-sr:r ·-::r.*-"I··I···*.·:. ·N·:i ":

··i

:·i.·:'·· --r·:,.;·r: .;

···:::r,; .··i··"i::liilSiili!iiiilii .·,.·II :·::;;·;;I·:;;·· ;;··:·*:.··

iZi.ii::;i i:i .i.··: '·'·. . i·· :·· .. :i.

..: ···:·:.ilp.ia:·ii :.·I·-I.·.;;;;;;';;·:;':::'···r:· ·;·:

::,:;.·::' "''''',l.i·i'i.

·:·.·:.·:i '·l':'id.i8

tesaLi%LB

ii*::·.

ii·iii%ai:

···i:i.

·*-

:. ' ..··

·aii

Fig. 4. Same patient, 5 months after transurethral re-section of external sphincter.

Undoubtedly one of the most important recentdevelopments in relation to urinary diversion to theskin is the introduction of plastic disposable appli-ances. The old rubber bags were difficult to cleanand quite impossible to sterilize and should no longerbe used. Stoma complications have become mostunusual since the plastic disposable bags ('Car-shalton' by Eschman's or 'Mitcham' by DownBros.) were introduced. Other essential features ofthe new disposable appliances include translucencyso that the stoma can be inspected without removingthe appliance and the use of a non-return flap valveto avoid the escape of large quantities of urine if thebag should accidentally come off. One ofmy patientswith an ileal conduit diversion practises judo andmany others play soccer or even rugger so that sucha urinary diversion need not necessarily restrict theactivities of these children.

Contrary to expectation, stone formation follow-ing congenital neuropathic bladder is relatively un-common in contrast to this complication in patientswith traumatic paraplegia. While it may well be thatrenal failure will play an important cause in themortality in myelomeningocele patients in the yearsto come, there is to date no satisfactory evidence thatthis will indeed happen.

Bowel continenceJust as the majority of myelomeningocele patients

have a neuropathic bladder, so a similar proportionhave a neuropathic bowel. In practice, however,faecal incontinence is much less of a problem, largelybecause this group ofpatients tends to be constipated.At the most, only 50%/ of myelomeningocele patientshave rectal incontinence as judged by social criteriaand in a recent study, we were unable to correlaterectal function to the level of the neurological lesion,the degree of limb paralysis or indeed to bladderfunction. It appears that good rectal training wasthe most important factor in achieving social con-tinence (Scobie, Eckstein & Long, 1970). Conscien-tious parents were the only common factor whichimproved bowel continence in this group of children.

EducationThe education of myelomeningocele patients is

giving rise to considerable concern. On the whole,myelomeningocele patients are of normal intelligenceand their admission to normal schools is governedby the degree of paralysis and of incontinence onthe one hand, while on the other hand the localfacilities for schooling are of paramount importance.Approximately halfof ourmyelomeningocelepatientsare attending normal schools at primary school levelbut there is likely to be an alarming drop when thesesame children reach the stage of secondary educa-tion. In this particular case, the problems are notrelated so much to the children as to the physicalcharacteristics of the schools. While the majority ofprimary schools are relatively modern and one-levelbuildings, most secondary schools are old and requirethe negotiation of stairs to get pupils from one class-room to another. Such simple architectural problemsare likely to interfere with schooling of these multiplyhandicapped children. There is no doubt whateverthat the facilities for post-school vocational trainingfor handicapped people are totally inadequate andit is only hoped that suitable facilities will be pro-vided in time as the flood of myelomeningocelepatients and multiply handicapped adolescents in-creases at a steady and alarming rate.The fate of these unfortunate children born with

myelomeningocele depends not only on whether wecan provide the necessary hospital facilities andsurgical expertise but also on whether the State willprovide, in time, suitable facilities for vocationaltraining and sufficient sheltered employment tojustify the whole treatment of spina bifida and all itsassociated problems.ReferencesDICKSON, J., ECKSTEIN, H.B., GLASSON, M. & KAPILA, L.

(1969) Shunt therapy for hydrocephalus when both jugularveins have been used. Zeitschrift fur Kinderchirurgie undGrenzgebiete, 7, 176.

copyright. on A

pril 21, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.48.562.496 on 1 A

ugust 1972. Dow

nloaded from

500 H. B. Eckstein

ECKSTEIN, H.B. (1968) Urinary control in children withmyelomeningocele. British Journal of Urology, 40, 191.

ECKSTEIN, H.B. (1971) Problems of primary closure ofmyelomeningocele. Proceedings of the Royal Societ' ofMedicine, 64, 1143.

ECKSTEIN, H.B. & MACNAB, G.H. (1966) Myelomeningoceleand hydrocephalus; the impact of modern treatment.Lancet, i, 842.

ECKSTEIN, H.B. & VORA, R.S. (1971) Spinal osteotomy inchildren with myelomeningocele. Journal of Bone and JointSurgery (in press).

LAURENCE, K.M. & TEW, B.J. (1971) Natural history of spinabifida cystica and cranium bifidum cysticum. Annals ofDiseases of Childhood, 46, 127.

LORBER, J. (1971) Results of treatment of myelomeningocele.Developmental Medicine and Chikl Neurology, 13, 279.

MACNAB, G.H. (1963) Recent Advances in Paediatric Sturgery'(Ed. by A. W. Wilkinson), p. 261. Churchill, London.

NICHOLAS, J., KAMAL, I. & ECKSTEIN, H.B. (1970) Immediate

shunt replacement in the treatment of bacterial colonisationof Holter valves. Developmental Medicine and ChildNeurology, 12, Suppl. 22, 110.

PUDENZ, R.H. (1966) The ventriculo-atrial shunt. Journal ofNeurosurgery, 25, 602.

RICKHAM, P.P. (1964) A ventriculostomy reservoir. BritishMedical Journal, 2, 173.

SCOBIE, W., ECKSTEIN, H.B.& LONG, J. (1970) Bowel controlin children with myelomeningocele. Developmental Medi-cine and Child Neurology, 12, Suppl. 22, 150.

SHARRARD, J.W., ZACHARY, R.B., LORBER, J. & BRUCE,A.M. (1963) A controlled trial of immediate and delayedclosure of spina bifida cystica. Archives of Diseases ofChildhood, 38, 18.

TsINGOGLU, S. & FORREST, D.M. (1971) A technique for theinsertion of Holter ventriculo-atrial shunt for infantilehydrocephalus. British Journal of Surgery, 58, 367.

WALKER, G.F. (1968) The orthopaedics of myelomeningocelein infancy. Hospital Medicine, 2, 900.

copyright. on A

pril 21, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.48.562.496 on 1 A

ugust 1972. Dow

nloaded from