Proceedings Society ofBritish Neurological Surgeons the ... · CRANIOPHARYNGIOMA RHayward, UAndar,...

journal of Neurology, Neurosurgery, and Psychiatry 1992:55:513-521

Proceedings of the Society of British Neurological Surgeonswith the New England Neurosurgical Society, London,September 1991

RADICAL EXCISION OF CRANIOPHARYNGIOMA BY

THE TEMPORAL ROUTE

L Symon. National Hospital, London

Craniopharyngioma remains a difficult man-agement problem. The central situation anddifficult texture ofthe tumour, its adhesion toneighbouring blood vessels and basal cranialnerves, and its capacity to wander throughmore than one intracranial compartment hasoften resulted in years of progressivelyimpaired health in a patient punctuated byattempts at partial removal, each one fre-quently leading to further deficit. Attemptedradical excision of craniopharyngioma havebeen made increasingly possible by progressin neuroradiological imaging and the use ofmicroscopic technique.Between 1977-91 fifty four patients in the

National Hospital for Neurology and Neuro-surgery have had radical surgical excision ofcraniopharyngioma by the temporal routewhere the configuration of the tumourseemed appropriate for this approach. Surgi-cal mortality was 4%. Forty two patients(78%) remain well on an average follow up ofthirty months. Twelve per cent had majorcomplications: hypothalamic damage, sub-dural haematoma, scalp collections requiringshunt drainage, and there have been fourrecurrences in patients where the initialoperations have been considered to be micro-scopically complete.

THE GOOD, THE BAD, AND THE UGLY: TREATMENTSTRATEGIES IN THE MANAGEMENT OF

CRANIOPHARYNGIOMA

R Hayward, U Andar, W Harkness. GreatOrmond Street, London

There is little argument about the idealtreatment of craniopharyngioma. Thetumour is benign and non-invasive and thebest prognosis should therefore follow a

complete removal. The technical difficulties,however, involved in carrying out this stra-tegy make management policies controver-sial, as can be seen by the number of articlesdevoted to the subject in the neurosurgicalliterature, with estimates of complete remo-val ranging from 7% (for transphenoidalsurgery) upwards. By analysing a variety ofrisk factors, it is possible to prescribe for eachpatient a treatment policy, both surgical andnon-surgical, which should provide themwith the lowest immediate mortality andmorbidity as well as allowing them the mostfavourable longterm outlook.

Analysis of all the 30 new cases of cranio-pharyngioma seen over a ten year perioddemonstrates how it is now possible, on thebasis of both clinical and imaging informa-tion (particularly MRI) to determine pre-operatively: the scale of the operation that itis safe to perform, the operative approach(such as, subfrontal, temporal, cranio-facial)and the likely requirement for post-operativeradiotherapy and the most suitable methodfor its delivery (conventional, stereotactic or

interstitial).

THE TREATMENT OF ACTH SECRETING PITUITARYADENOMATA BY TRANSPHENOIDAL SURGERYGS Cruickshank, GM Teasdale. Glasgow

A recent British study' on the care ofCushings Disease, cast doubt on the relation-ship between the type of operation, thepathological findings and the outcome. Atotal of 84 consecutive patients werereviewed with Cushings disease or Nelsonssyndrome. The aims were to determine theaccuracy of pre-operative investigations, theoperative findings and to relate those with thespecific surgical technique (complete selec-tive adenomectomy, incomplete selectiveadenomectomy, radical subtotal resection,total hypophysectomy) and outcome.

Seventy two patients with pituitary drivenCushings disease, and 12 patients with Nel-sons syndrome had transphenoidal pituitaryexploration. A discrete macro or micro-adenoma was located in 60 patients and was

judged to be completely removed in 47. Ineach case the surgical diagnosis was con-firmed by histology and subsequent remis-sion.

Patients without a discrete tumour hadeither radical subtotal (19) or total hypophy-sectomy (6). Four of these had a micro-adenoma. In total 64 patients had an adeno-ma confirmed by histology, five glandularhyperplasia and in 15 the gland appearednormal. Of those patients with a discretelesion seen at operation, only 27/60 (45%)were correctly predicted by CT scanning andonly 6/12 by MRI. Petrosal venous samplingwas accurate at confirming a pituitary sourcefor ACTH over-production, but poor atidentifying the pituitary localisation of theadenoma. Every patient who had a selectiveadenomectomy went into remission, thesepatients recovered their pituitary reservemore rapidly and frequently, than after otheroperations. Nineteen patients had radicalsubtotal excision, 11 remitted, and of sixpatients who had a total hypophysectomythree remitted. Apparent reasons for failureto remit included a missed diagnosis ofectopic ACTH production in two patientsand obviously invasive adenoma in a numberof patients.

Radiological studies were unable to reli-ably identify the precise site of a pituitarymicro-adenoma. When the operative findingsindicated that selective but complete micro-adenomectomy was feasible, there was a veryhigh prospect of remission and of return orretention of pituitary function and that thisremains the optimum treatment for ACTHproducing adenoma.

1 Burke CW, Adams CBT, Esiri MM, Morris C,Bevan JS. Clinical endocrinology 1990;33:525-37.

CURRENT STATUS OF BORON NEUTRON CAPTURE

THERAPY (BNCT) FOR INTRACRANIAL TUMORS

H Hatanaka, R Fairchild, D Joel, D Slatkin, JCoderre,WH Sweet. Boston, USA

Hatanaka's group in Japan have four patients,given BNCT using B,2HI ,SH, who are

working productively as of this year, onehaving survived for 19 years with a glio-blastoma grade IV, despite all eight of themajor prognostic criteria being in the adversecategory. The authors were unaware of anyother such case in the world literature. Theother three Japanese patients with gliomasGrades II and III are active after BNCT for14, 13 and 10 years.Further progress includes completion of

reactor portals for a high flux of epithermalneutrons to achieve adequate depth doses(Brookhaven National Laboratory [BNL]and Massachusetts Institute of Technology[MIT]). Definitive design of two other reac-tors for BNCT at the Idaho EngineeringLaboratory [INEL] and Petten, Hollandinclude a striking reduction in undesirableradiation not selective for tumour. Threemuch improved methods have been devel-oped for quantitative analysis of tissue sam-ples for boron and for precise identificationof the organelles within and around the cellsof all types.The securing of massive amounts of phar-

macokinetic data from each patient will bepossible with the greatly modified GeneralElectric magnetic resonance scanner whichdepicts "B concentrations in small areasthroughout the intracranial cavity.

Perhaps the most important progress hasbeen the achievement at Brookhaven of longsurvival of frontal gliosarcoma in six of 10rats given B12H,,SSH,,B 2 as the boroncarrier for the BNCT. Long survival was alsoachieved in seven of 16 rats using bor-onophenylalanine. These are the first criti-cally controlled studies to produce highpercentages of probable cure of gliomas inany species and are perhaps the most crucialfavourable biological information relevant toBNCT obtained to date.

TOTAL THORACIC VERTEBRECTOMY FOR PRIMARYMAUGNANT DISEASE BY POSTERIOR APPROACH-ACASE REPORTG Findlay, P Eldridge. Liverpool

Considerable controversy continues to existregarding the ideal operative approach formany spinal problems, including tumour. Inthe high thoracic region, an anteriorapproach is usually felt necessary to providesafe access for anterior disease. Subsequentlya further posterior procedure may be neces-sary for stabilisation.An unusual case was presented of primary

bone tumour (probable chondrosarcoma)which presented as an exophytic mass in theparavertebral region: involvement of anteriorand posterior elements of two vertebrae andribs on one side.

Resection of tumour gave adequate accessto allow total vertebrectomy grafting andHartshill rectangle fusion at one procedure,achieved from a posterior approach. This wasachieved at a high thoracic level where thegreat vessels made a trans-thoracic proceduremore hazardous.An excellent recovery occurred although

unfortunately the malignant nature of thetumour makes a good prognosis unlikely. A

513 on M

ay 23, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.6.513 on 1 June 1992. Dow

nloaded from

http://jnnp.bmj.com/

514 Proceedings of the Association of British Neurologists, London, September 1991

posterior approach may be possible by indi-vidual anatomy and operative approachestailored to individual cases rather than dicta-ted by convention.

RECENT EXPERIENCE WITH HEMISPHRECTOMYCE Polkey. Maudsley Hospital, London

A personal series was reviewed of 28 hemi-spherectomies for intractable epilepsy, car-ried out since 1978. Four were subtotalhemispherectomies; the remaining 24 weretotal hemispherectomies using Adam's modi-fication.' Follow up was available for 17patients operated on at least two years before(range 2-14), and using Engels's outcomescale.2 Thirteen patients (76%) were seizurefree or virtually so, two were improved, onewas not and one patient died. Four of thepatients undergoing total hemispherectomyrequired CSF shunts from the remainingventricle, two in the immediate post-oper-ative period. There were no instances of latedelayed bleeding although many patientsshowed asymptomatic dilatation of theremaining ventricle and midline shift. In twocases there were unexpected neurologicalconsequences.The pathology in the specimens showed 14

examples of Rasmussen's encephalitis, fourcases with cortical neuronal migration defect(hemimegalencephaly) and 10 with otherpathologies.

1 Adams CBT. J Neurol Neurosurg Psychiatry1983;46:617-9.2 Engel J. Outcome with respect to epileptic

seizures. In: Engel J, ed. Surgical treatment ofthe epilepsies. New York, Raven Press,1987:553-71.

HIPPOCAMPAL COMMISSUROTOMY AT THE TIME OFCORPUS CALLOSUM SECTION FOR INTRACTABLEEPILEPSYDW Roberts, AG Reeves, RE Nordgren, MSGazzaniga.

The hippocampal commissure, a structure ofunclear significance in the human, is conven-tionally divided at the time of section of theoverlying posterior corpus callosum. Thebenefit in terms of improved seizure controland the risk in terms of cognitive impairmentare not well understood. Should the com-missure's division accompany corpus callo-sum section?Three sets of data were studied for this

purpose. Firstly, 98 patients had undergonecorpus callosum section. Of these, 27patients had combined posterior corpus cal-losum and hippocampal commissure sectiononly when seizure control was suboptimalfollowing anterior corpus callosum section.The percentage of these patients obtaining80-100% reduction in generalised seizuresapproximately doubled following the secondsurgery (29% versus 62%), but correlation ofresponse to extent of initial section was onlymoderate and distinction between effect ofcorpus callosum versus hippocampal com-missure section was not possible. The seconddata set was that of EEG cross-correlationand autoregression studies that suggested adiminished role of the hippocampal com-missure, compared with the corpus callosum,in seizure propagation. The third area ofinterest was that of neuropsychological stud-ies that demonstrated impairment of recallbut not recognition memory following divi-sion of the posterior corpus callosum andhippocampal commissure.

While these lines of evidence are notconclusive, they suggested that in selectedpatients, particularly those without complexpartial seizure manifestations and those inwhom protection of memory is a high prior-ity, sparing of the hippocampal commissureat the time of posterior corpus callosumsection may be warranted.

HINDBRAIN HERNIA: PRESENTATION, MANAGEMENTAND OUTCOMEPO Byrne, BN Williams. Midland Centre,WarleyA review of 100 patients with symptomatichindbrain hernia, also known as Chiari TypeI malformation, was presented. These caseswere selected for being adults with no obvi-ous cause of the hindbrain hernia such asintracranial tumour and also because theyhad been treated by craniovertebral decom-pression.Symptoms and signs on presentation were

analysed as "motor", "sensory", "bulbar"and "other" groups. Serious difficulties wereencountered in the assessment of the resultsof management in this condition. Post-oper-atively, 76% of the patients were clinicallyimproved, 18% were stable and 6% wereworse following craniovertebral decompres-sion. There was little correlation betweenoutcome clinically and post-operative imag-ing. Four patients required additional proce-dures aimed at improving residual syrinxcavities. The details of the operative approachwere presented.

BRAINSTEM DECOMPENSATION IN PATIENTS WITHCHIARI DEFORMITIES AND THE REIATIONSHIP TOCSF VOLUMEEG Fischer. Boston

A patient was reported with a shunt-inducedChiari I deformity and who developed symp-toms of pronounced brainstem dysfunctionwhen a large bifrontal craniectomy defect wascreated. The symptoms were relieved whenthe shunt, originally placed because of apostoperative CSF rhinorrhea, was ligated.The author suggested that if CSF volume

affects the rostro-caudal position of the cere-bellar tonsils in relation to the foramenmagnum, buoyancy of the brain and reduc-tion in CSF volume may be important factorsin the development of a Chiari deformityfollowing insertion of a lumbar shunt. Bothfactors may be responsible for the caudalmigration of the hindbrain seen followingventricular shunts which has been ascribed tocephalo-cranial disproportion. If the hind-brain can be displaced caudally at the fora-men magnum when atmospheric pressure istransmitted to the frontal lobes through acraniectomy defect, brainstem decompensa-tion in infants with Chiari II malformationmay be explained by the effect of atmos-pheric pressure transmitted to the frontallobes through the open anterior fontanelle.

THE NATURAL HISTORY OF DIASTEMATOMYELLAUB Andar, RD Hayward, WFJ Harkness.Great Ormond Street, London

Diastematomyelia (57 cases over 12 years)accounted for about a quarter of the caseswith spinal disorders seen by the authors.

To decipher its natural history, cases seenearly and treated conservatively were com-pared with patients who had undergone earlysurgery. Many children had no neurologicalsigns at birth or at time of referral. Some hadspinal or limb deformities with associateddysraphic features. After long follow up fewwere left without any neurological deficits.The problems ofprediction of natural historyand timing of surgical intervention provokeda vigorous debate.

HARVEY CUSHING'S BRIGHAM HOSPITAL-1991J Shillito. Boston, USA

In 1932 Dr Harvey Cushing retired as Chiefof Surgery at the Peter Bent Brigham Hospi-tal, Boston. The retirement age of 63 yearshad been established by Dr Cushing and hiscounterpart in medicine, Dr Henry Chris-tian, when the hospital was opened in1913.

It was apparently quite a shock to DrCushing that this time arrived so quickly. Hereturned from a trip to Europe to find that hissuccessor, Dr Eliot Cutler, whom he hadtrained, was already physically established inthe Chief s office, and Dr Cushing's posses-sions had been removed elsewhere. TheDepartment of Surgery became a very gen-eral surgical department, and the number ofneurosurgical operations per year droppedprecipitously. Dr Cutler maintained that hecould operate equally well "in the bodycavity" and he assumed responsibility notonly for all the intracranial operations but thecardiac operations as well. Dr Horrax, Cush-ing's longstanding associate, went to thenearby Deaconness Hospital with the LaheyClinic. Neurosurgery at the Children's Hos-pital, across the street from the Brigham,continued after Cushing's department underDr Franc Ingraham, a pupil of Cushing'swho had been assigned to set up a pediatricneurosurgical department in 1929.When World War II took many surgeons

away from the Brigham, including EliotCutler, Dr Ingraham also covered the neu-rosurgical activities at the Peter Bent Brig-ham Hospital as well as the Children'sHospital. This arrangement continues to thepresent day. With the help of several nowprominent fellows, and his ultimate associate,Dr Donald Matson, neurosurgery at theBrigham was returned to the hands of neu-rosurgeons.Dr Ingraham retired as Chief of Neuro-

surgery in 1964 and died of a myocardialinfarction in 1965. Dr Matson continued asChief but succumbed in 1969 to Jacob-Creutzfeldt Disease. Neurosurgery wasadministered temporarily by the author withthe help of several other former residents andassociates until 1971, at which time it cameunder the direction of Dr Keasley Welch.After his retirement Dr Peter Black becamethe Ingraham Professor. Neurosurgical activ-ity now exceeds that of Harvey Cushing'sday.Dr Cushing not only left a remarkable

heritage, but he also left some close asso-ciates. Adolph Watska, his operating roomorderly, continued to work at the Brighamuntil his death in 1953. Mildred Codding,Cushing's medical artist, continued to workwell past her official retirement, until 1980.She lives on Cape Cod, enjoying her memo-ries and her water colours of the NewEngland scene.

on May 23, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.55.6.513 on 1 June 1992. D

ownloaded from


Proceedings of the Association of British Neurologists, London, September 1991

NEUROSURGICAL MANPOWER AND ACTIVIT SURVEY*.

1988-1990AA Kemeny, PJ Hamlyn. Sheffield, and St.Bartholomew's Hospital, London

The activity and manpower of all neuro-surgical units in the UK and the Republic ofIreland for 1988 and 1989 were reported.The 132 consultant posts provided 1179sessions, of which 88 were academic. Theirage was given by 112 consultants: only 13were over 60 years of age. There were 33-5senior registrars and 20 were due for accre-ditation by mid 1991. There were 106 regis-trars: of the 69 clinical posts 33 were occu-pied by overseas trainees. Whilst half theunits have three consultants, 25% have less.36% have no senior registrar and 14% ofunits have no UK clinical trainee at any level.This contrasts with a fifth of units havingmore than two UK clinical trainees.There was no significant difference in

performance between large and small units,although the range was more varied in thelatter. The following activity data representthe means and ranges. Admissions per unit,1225 (469-3200); operations per unit, 800(157-1600); operations per admission, 0-67(0-36-1-3); operations per consultant, 242(86-508); bed numbers, 40 3 (14-100); bedsper consultant, 12-8 (4-7-22); admissionsper bed per year, 30 4 (17-8-70-2); opera-tions per bed per year, 19-8 (11-4-45-3).One third of units had no designated beds

for ventilation and one quarter had bedsclosed. Forty per cent refused admission ofventilated cases and one quarter did so

monthly. Half of these units were outsideLondon. In spite of a 2-5% nationwidereduction in bed numbers, there was a 6-5%increase in the total operations from 1988 to1989.

ALLEGED PROFESSIONAL NEGLIGENCE INVOLVING

NEUROLOGICAL AND NEUROSURGICAL PATIENTS: A

REVIEW OF 1I15 CASES

PJEM Wilson. Swansea

Medical negligence litigation is a growthindustry for attorneys, while for doctors it isperhaps the most rigorous form of audit.Lawyers will increasingly require us to scruti-nise the practices of our peers and thus,indirectly, our own; a task seldom other thaninstructive.

In the last 115 cases in which the author, asa neuro-clinician, was asked to provide"plaintiff" expert reports, the largest"defendant" group, unsurprisingly, com-

prised neurosurgeons (42). Orthopaedists(17), casualty officers (13), and generalphysicians (9) followed.

In the neurosurgical group, spinal disastersexceeded cranial by 2 to 1. Commonest were,tetraplegia after cervical discectomy or

decompressive laminectomy; mishaps oflumbar disc surgery; and major wound sep-sis. Cranial problems related mainly to badoutcome with benign tumours. In 15 cases

the nexus of causation was very doubtful.Only two cases seemed in essence indefen-sible. Importantly, only a handful of neu-

rosurgical accidents occurred at the forefrontof "high technology." Most were in thesetting of commonplace routine.Among referring specialties there predomi-

nated the age-old and seemingly ineradicabletragedies of neglected extradural haemato-ma; aneurysmal subarachnoid haemorrhagemistaken for migraine; benign spinal

tumours labelled as multiple sclerosis; andtardy perception of cord and cauda equinacompression.

STEREOTACTIC BIOPSY OF CEREBRAL LESIONS INPATIENTS WITH AIDSA Sofat, MF Pell, DGTThomas, IRWhittle.London and Edinburgh

The CNS is affected in up to 50% of patientswith AIDS. CT scanning can identify cere-bral lesions but cannot distinguish toxo-plasmosis from other AIDS-related lesions.Between May 1987 and November 1990

11 patients had stereotactic biopsy. Six ofthese patients had previously been empiri-cally treated for toxoplasmosis. Histologicaldiagnosis of the biopsy specimens showedmultifocal leucoencephalopathy in six cases,toxoplasmosis in three cases, lymphoma inone case and non-specific changes withhypercellular cerebral white matter andincreased astrocytic cells in one case. Onepatient died and another had a transientworsening of his previous hemiparesis.CT guided sterotaxy is a reliable and

accurate method of obtaining a tissue diag-nosis, but the experience of the authorssuggests that it is associated with a highermortality and morbidity than in non-AIDSpatients. It is recommended that when neu-rological signs and symptoms develop inpatients with AIDS and a lesion is identifiedon CT or MRI scanning the patient shouldreceive at least two weeks of anti-toxoplasmatherapy. Biopsy should be reserved for thosepatients with atypical presentation, negativeserology or progressive clinical deteriora-tion.

MENINGIOMAS-HORMONE THERAPY IS AN

ALTERNATIVE TO SURGERYC Davis. Preston

A review of surgical mortality and morbidityfor most benign and malignant intracranialtumours over the last 15 years reveals amarked improvement, except in the case ofmeningiomas where operative morbidity andmortality was 20% with a 25% recurrencerate.A series of eight patients (representing a

quarter of patients referred from 1988-1990)was presented. These patients were treatedfor 3 to 38 months (average 24 months) withGestrinone, an anti-progesterone syntheticsteroid. This selected group of patients weremonitored clinically, neurophysiologicallyand with accurate tumour volume estimationon CT. The results showed no significanttumour growth in any of these cases in themedium term. It was suggested that hormonetherapy may be an alternative to surgery formeningiomas and was worthy of a rigorouscontrolled trial.

A QUANTITATIVE ASSESSMENT OF BRAIN TUMOURULTRASONIC ENHANCEMENT ULTRASTABLE

LIPID-COATED MICROBUBBLES AS A CONTRASTAGENTRS Simon, S-Y Ho, CR Perkins, JS D'Arrigo.Farmington, USA

The authors developed lipid-coated micro-bubbles that were stable for months in vitroand had a narrow range of diameters, from

0-8 to 4-5 um. These ultrastable micro-bubbles were used as an ultrasonic contrastagent in the brain, causing increased intensityof echoes persisting for many hours. Intra-venously administered lipid-coated micro-bubbles accumulated selectively in rat braingliomas, exhibiting enhancement of echoge-nicity for up to one hour. With this contrast,growing lesions could be imaged two daysearlier than without contrast. Using a lipid-specific stain, the distribution of the bubblesin the tumour and throughout the brain wascharacterised.On the scan, the enhancement of the

tumour was accompanied by a change in thesignal-to-noise ratio of the echoes from thetumour. Using spectral analysis characteristictextural changes were associated with con-trast-enhanced tumour. Lipid-coated micro-bubbles could target other brain tumours inaddition to gliomas; specifically Walkertumour and melanoma metastases in the ratbrian were identified on the scan using lipid-coated microbubble contrast. lipid-coatedmicrobubbles, when injected intravenously,intensified the ultrasonic echoes from atumour lodged in subcutaneous tissue (Wal-ker-256 carcinosarcoma). The enhancementof the tumour remained for 15-30 minutes.Oil-Red-O lipid stain identified the micro-bubbles in the tumour area but not in theadjacent soft tissue. There was a significantchange of signal-to-noise ratio associatedwith the echo enhancement.

Finally, the bubbles could be made toresonate, causing a bubble-tissue energyexchange (cavitation). Cavitation causes tis-sue destruction in other biological systems.In this context, we will present our prelimi-nary data for therapeutic applications of thelipid-coated microbubbles.

IMAGING OF AXONAL TRANSPORT: IS THE

AXOPLASMIC FLOW CLINICALLY RELEVANT?AG Filler, BA Bell, FA Howe, JR Griffiths,M Flowers, H Sharma, HR Winn, TWDeacon. London, Seattle and Boston, USA

Many neurosurgical and neurological dis-eases cause disturbances in the natural trans-port of molecules and vesicles along axons.However, because these disturbances oftransport have not been directly detectable inliving humans or intact animals, little isknown about their severity. Because thedynamics of disturbed axoplasmic flow innerve compression and neuropathy has beenunknown, it has never figured prominently inmodels of clinical problems.A series of advances in the laboratory has

now led to preliminary success in developingaxonal transport tracer compounds whichcan be detected by MRI and by PET. Afterintramuscular injection of a ferrite-basedtracer, MRI show a marked contrast changein the nerve supplying that muscle reflectinga 50% reduction in T2.For this project, a new kind of PET tracer

had been developed with transition metalpositron emitting nuclei included in spinelferrite crystals. These "spinel moderatedpositron emitters" reduce tissue ionizations,improve image resolution, and increase theintensity of the PET signal. In initial tests anerve gel phantom 1 mm in diameter wasdetected using a low resolution multiwireproportional PET camera.

It is expected that these new agents will beuseful in locating and in confirming sus-pected sites of nerve injury, testing for

515 on M


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from


516Proceedings of the Association of British Neurologists, London, September 1991

transport across sites of spinal cord compres-sion, distinguishing nerve crushes from nervelacerations, and in the demonstration of avariety of other pathologies previously inac-cessible to neuroradiological detection.

RESULTS OF MICROVASCULAR DECOMPRESSION,OPEN RHIZOTOMY, RADIOFREQUENCY LESION ANDNEURECTOMY IN 332 PATIENTS wITH TICDOULOUREUX: ARE THESE DENERVATIONOPERATIONS?EC Tarlov. Burlington, MA, USA

The mechanisms by which relief of ticdouloureux occurs after surgery have beenmuch discussed. Cogent criticisms of theJanetta vascular decompression theory havebeen raised by Adams. The practical matter isthat tic pain can be relieved, often lastingly,by a variety of operations. Whether they sharea common mechanism has not been estab-lished.Three hundred and thirty two patients

with medically intractable tic douloureuxwere treated surgically between June 1982and June 1988. Ninety one had micro-vascular decompression; 58 patients in whomno vascular compression was found weretreated by trigeminal rhizotomy of 30-100%of the sensory root. The operative findings atopen operation were categorized in five subgroups: no vascular contact at trigeminalnerve root entry zone, artery touching root,artery indenting root, artery wedged betweenroot entry zone and brainstem, and venouscontact. Seven patients had infraorbital neu-rectomy, five had treatment of mass lesionsand 148 had radiofrequency lesions. Of thistotal group, during 1982 to 1989 there wereseven recurrences among the microvasculardecompression group (7/91), none recurredamong the neurectomy group (0/7), tworecurred among the rhizotomy group (2/58),none recurred among the mass lesion group(0/5),and 18 recurred among the 148 radio-frequency lesion group (18/148).The entire group was followed up by

questionnaire two to nine years postoper-atively to evaluate long term relief of tickpain, late recurrence of tic pain, annoyingsensory loss, anesthesia or hypesthesia dolor-osa and other symptoms. The author sup-ports the hypothesis that denervation of thetrigeminal system accounts for the relief ofpain afforded by all of these procedures.

REACTIVITY OF THE RAT MIDDLE CEREBRAL ARTERYIN CONTROLS AND FOLLOWING ISCHAEMIA-APRELIMINARY REPORTRJ Laing, A Stawowy, J Warnke, J Jaku-bowski. Sheffield

Little is known about the reactivity of cere-bral blood vessels after a period of acuteischaemia. Ischaemia might damage themuscle cells or the endothelium or both.Longa's model' was used to produce a focalischaemia in the MCA territory. The severityof ischaemia was evaluated by measuringcerebral blood flow (CBF) using the hydro-gen clearance technique. Following fourhours of ischaemia the animal was killed andthe MCAs prepared for myography. Thecontractility of the arteries to potassium andprostaglandin F2A was determined. Thecontrol group comprised 21 animals withmean CBF 87+/- 10 ml/100 gm/min in theMCA territory. The preparations remainedstable for up to eight hours. In ten animalssevere ischaemia was produced with mean

CBF 17+/- 7 in the MCA territory and wasmaintained for four hours. In MCAs fromthe ischaemic area the preliminary resultssuggested that contraction in response topotassium was preserved (1 -8 +/- 0-36 mN/mm in control and 1-7 +/- 1 0 mN/mm inischaemic) but that to PGF2a was impaired(1-0 +/- 0-25 mN/mm in control and 0-65+/- 0 37 in ischaemic).These early results were sufficiently

encouraging to warrant further studies withparticular emphasis on endothelial functionwhich may be more susceptible to ischaemicdamage.

1 Longa EL, et al. Stroke 1989;20:84-91.

THE EFFECTS OF HYPOTENSION ON INFARCT SIZE,CEREBRAL OEDEMA AND CEREBRAL BLOOD FLOWAFTER EXPERIMENTAL MIDDLE CEREBRAL ARTERYOCCLUSIONRD Strachan, PJ Kane, JR Chambers, SCook, AD Mendelow. Middlesbrough andNewcastle

Hypotension after a cerebral ischaemic insultaffects clinical outcome in patients withocclusive cerebrovascular disease, subarach-noid haemorrhage or head injury. This studywas designed to establish the effect of a shortperiod of modest hypotension on infarct size,cerebral oedema and cerebral blood flowafter MCA occlusion in the rat. A newmethod of closed-loop blood pressure con-trol was presented which allowed a precisereduction in blood pressure by hypovolaemiaalone.

Forty four male Wistar rats were studied insix groups. In the hypotensive groups, theblood pressure was reduced to 70 mmHg for30 minutes immediately after MCA occlu-sion. Results showed that in the hypotensiveanimals: Infarct size was larger(190-3 + 8-8 mm3 and 230-6 ± 10 1 mm3;p < 0 01); cerebral oedema was greater in thecortex (specific gravity = 10455 ± 0-0013and 1-0493 ± 0 0011; p < 0 05) and post-occlusion cerebral cerebral blood flow wasless in the lesioned hemisphere (39-1 ± 2-1and 27-2 ± 2-4 ml/100 g/min; p < 0-001)and blood flow failed to show the samerecovery characteristics after reperfusion.These studies confirm that modest hypo-

tension after infarction has significant delete-rious effects on the extent of ischaemicinjury.

CALCITONIN GENE RELATED PEPTIDE INCREASESCEREBRAL BLOOD FLOW IN A MODEL OF FOCALCEREBRAL ISCHAEMIA?WAS Taylor, SGC Sydserff, BA Bell. Atkin-son Morley's Hospital, London

Calcitonin gene related peptide (CGPR) isan endogenous neuropeptide presentthroughout the perivascular neural networkwhich is released following intracranialhaemorrhage. It may have a protective rolevia its potent vasodilatory action, and be ofuse in vasospasm following subarachnoidhaemorrhage, were early clinical results havebeen encouraging.'

In this study, anaesthetised male Vistar ratswere ventilated at normocapnia, and cerebralblood flow (CBF) was measured by hydrogenclearance. A dose response curve for CGPR Iwas established by infusing 40, 80, 100 and120 ng/kg/min, and no change was found inCBF until a dose of 120 ng/kg/min wasreached, when mean arterial pressure fell

from 96-4 ± 3-0mmHg to 85-1 ± 2-4(t = 3 00, p < 0 01), and CBF from915 ± 3 0 ml/lOOg/min to 81-9 ± 1-8 ml/lOOg/min (t = 2.92, p < 0-01). Middle cere-bral artery (MCA) occlusion was producedusing an intraluminal suture technique,2 anda dose of 100 ng/kg/min was selected for ratsundergoing MCA occlusion. Groups of tenrats were infused with either CGPR ornormal saline, and in the CGPR group CBFwas significantly higher (85-9 ± 2-9 ml/1OOg/min) when compared to the controlgroup (63-7 ± 2-1 mI/lOOg/min; t = 6-26,p < 0-001). This increase was maintainedthroughout the period of ischaemia in boththe ischaemic and non ischaemic hemi-spheres.These findings suggest that CGPR, in a

dose below that which produces hypotension,only affects CBF when an insult to thecerebral circulation has occurred, and mayexplain its possible efficacy in cerebral vaso-spasm secondary to subarachnoid haemor-rhage.

1 Johnston FG, Bell BA, Robertson IJA, et al.Lancet 1990;335:869-72.2 Longa EZ, Weinstein PR, Carlson S, et al. Stroke

1989;20:84-9 1.

TRANSCRANIAL DOPPLER VELOCITY AND CEREBRALBLOOD FLOW CHANGES. A SIMULTANEOUS STUDYUSING THREE VASODILATORSJL Martin, S Perry, JD Pickard. South-ampton

Uncertainty surrounds the prediction ofCBF changes from the analysis of Dopplerultrasound flow velocity changes in basalcerebral arteries. The simultaneous effects ofa-CGRP, nimodipine, and glyceryl trinitrate(GTN) on CBF and basilar artery meanDoppler flow velocity in the anaesthetisedrabbit model' were presented.

Paired measurements of cortical CBF(hydrogen clearance method) and mean basi-lar artery Doppler flow velocity ("Transcan"EME, Uberlingen, Germany) were recordedduring infusions of: a-CGRP iv (1, 5, 10, 50,100 ng/kg/min.), nimodipine iv (0 1, 0 5, 1,5 ,g/kg/min), and glyceryl trinitrate iv (1, 4,8, 16, 32, 64 ,ug/kg/min). Each dose wasinfused for fifteen minutes.a-CGRP caused a dose-dependent

increase in TMV (from 23 ± 1 SE to30 ± 2 cm/s, p < 0-05), whilst CBFremained unchanged. GTN caused a dope-dependent increase in CBF (from 36 ± 4 to65 ± 10 mllOOg/min, p < 0 05), whilstTMV remained unaltered. With NimodipineTMV increased in a dose dependent manner(from 38 ± 3 to 84 ± 7 cm/sec, p < 0-01),and CBF also increased in a dose dependentmanner (from 38 ± 3 to 95 ± 6 ml/lOOg/min,p < 0-01).CBF changes cannot be predicted from

basal cerebral artery flow velocity changesalone. Increases in flow velocity in one arterysupplying the circle of Willis, even if thisrepresents a true increase in volume flow,equally may represent an increase in intra-cranial to extracranial shunt flow. CGRP-induced flow velocity changes may occur bythis mechanism. GTN increased CBF butwithout any change in flow velocity perhapsreflecting dilatation of the basilar artery aswell as dilatation of intracerebral arteries.With nimodipine the changes in CBF paral-leled the changes in flow velocity.1 Nelson RJ, Perry S, Hames A, et al

J Neurosurg1990;73:601-10.



j.com/

J Neurol N


ownloaded from



TREATMENT OF EXPERIMENTAL CEREBRALVASOSPASM USING THE lpS PULSED DYE LASER

R Macfarlane, A Teramura, NT Zervas.Cambridge, UK and Boston, USA

Laser energy at a wavelength of 480 nm wasapplied in 1 pus pulses of 3-10 mJ to twomodels of vasospasm. Rabbit common car-otid arteries (n = 40) constricted chronicallyby the application of human blood within asilicone sheath, and dog basilar arteries(n = 5) subjected to a "two-bleed" sub-arachnoid haemorrhage protocol, were dila-ted by endovascular laser therapy from asingle quartz fibre. Vessels were restored tonear normal diameter, and dilatation per-sisted for the duration of study. No instancesof perforation, arterial thrombosis, or ofaneurysm formation were observed for up to60 days after treatment. Prophylactic pre-treatment of normal vessels was able toattenuate the development of vasospasm.This study suggests that the 480 nm pulseddye laser may have an application for thetreatment or prophylaxis of cerebral vaso-spasm.

EFFECT OF CALCITONIN GENE REIATED PEPTIDE ONOUTCOME AFTER ANEURYSMAL SUBARACHNOIDHAEMORRHAGE

European CGRP in subarachnoid haemor-rhage study group

Calcitonin gene related peptide (CGRP) isan endogenous neuropeptide presentthroughout the perivascular neural networkand is released following subarachnoid haem-orrhage (SAH).' It is a potent vasodilator,and a pilot study in aneurysmal SAH sug-gested that it can reverse ischaemic deficits insome patients.2 A randomised multicentreopen comparison of CGRP against standardbest managment in patients suffering ischae-mic deficits after surgery for a rupturedintracranial aneurysm has been conducted.Patients aged 18-70 years who developed afocal neurological deficit, or had a reductionof two or more points on the Glasgow comascale (GCS) after surgery, were entered aftera CT scan had excluded non-ischaemiccauses for their neurological deficit.Informed consent was obtained and preg-nancy, uncorrected hypovolaemia, and ser-ious concomitant illness were exclusion cri-teria. After randomisation, 62 patients wereallocated to receive an infusion of 0-6 ug/minof CGRP for a minimum of 4 hours up to amaximum of 10 days, and 55 patientsreceived standard best management. GCSand haemodynamic parameters wereassessed during the patient's hospital stay,and all patients were followed up at threemonths by an independent investigator, whowas blind to their treatment. Outcome wasmeasured on the Glasgow Outcome Scale.

Sixteen neurosurgical units contributedpatients, twelve centres in the UK andIreland, and four in Europe. Of the 117patients randomised, 99 met the full inclu-sion criteria of the study protocol. At threemonth follow up 66% of patients treated withCGRP and 59% of patients receiving bestmanagement, had a good recovery, but thedifference between the two groups was notsignificant. Hypotension was a common sideeffect of the CGRP infusion. Although asignificant beneficial effect has not beendemonstrated by this trial, a clinically useful

benefit has not been exluded.

1 Juul R, Edvinsson L, Gisvold SE, et al. Br JfNeurosurg 1990;4:171-80.

2 Johnston FG, Bell BA, Robertson IJA, et al.Lancet 1990;335:869-72.

DISORDERED SALT AND WATER REGULATION INNEUROSURGICAL PRACTICE: MODERN MANAGEMENTPERSPECTIVES OBTAINED FROM RECENT STUDIESY Lolin, A Jackowski, L Symon. QueenSquare, London

In the 1950s hyponatraemia in intracranialdisease was felt to represent a derangement inan unknown salt regulating factor with anexcessive urinary excretion of sodium. Theterm cerebral salt-wasting syndrome wascoined and patients treated with salt andwater replacement therapy. Then in 1957Schwartz postulated inappropriate secretionof antidiuretic hormone (SIADH) resultingin dilutional hyponatraemia as causing thehyponatraemia observed in two patients withbronchogenic carcinoma; SIADH came to beaccepted as the aetiology of the majority ifnot all cases ofhyponatraemia in neurologicaland neurosurgical practice and fluid restric-tion became the standard treatment.The distinction between which of these

two mechanisms is responsible for the hypo-natraemia so commonly encountered is notjust academic as their management differs somarkedly. The authors now routinely meas-ure the fractional clearances of sodium andwater, and assay plasma renin, aldosterone,and ADH as necessary in hyponatraemicpatients suffering from SAH and other intra-cranial disorders to correctly identify theunderlying causation. The results demon-strated that in the majority of patients,hyponatraemia is due to salt-wasting with anincrease in free water clearance whilst only aminority are due to SIADH where the freewater clearance is decreased. This hasenabled us to choose appropriate treatmentregimes for individual hyponatraemicpatients without the risk of instituting ther-apy that is at best inappropriate or at worstpostively harmful.

ANEURYSMAL SUBARACHNOID HAEMORRHAGE: AN

AUDIT OF CLINICAL FEATURES, OUTCOMEASSESSMENT & COMMUNICATION WITH GPSKM Morris, MDM Shaw, PM Foy. Liver-pool

During 1990 all patients admitted to theMersey Regional Neurosciences Unit afterintracranial haemorrhage were prospectivelystudied to determine clinical course andoutcome at six months. A total of 228patients were screened; 142 had provenaneurysmal subarachnoid haemorrhage(SAH) of whom 128 were medically andneurolgocially suitable for surgery. The man-agement outcome of this group of patientswas assessed using the Glasgow OutcomeScale. They attained the following scores atsix months: 1-64-8%, II-12-7%, III-7-7%,IV-0%, V-14-8%.Management outcome was assessed by

General Practitioners and compared withcase note reports and a single neurosurgicalobserver. Surprisingly 10% of GPs had neverreceived a typed discharge summary. Ofthose GPs receiving a summary 96% felt thatthese were adequate but 32% suggestedalterations in emphasis and contents to allowthem to more easily continue the manage-ment of their patients.

INFLUENCE OF SURFACE COATING ON ADHERENCEOF BACTERIA TO SHUNTSPR Eldridge, J Bridget, J Punt, SP Denyer.Nottingham

Infection remains a significant cause of shuntfailure. Effects of such infection, and theneed for shunt revision contribute to a highmorbidity for this complication. The initialadherence of bacteria to the shunt may beimportant in the establishment of both colo-nisation and infection as has been found withintravascular catheters. In these cases, it wasnoted that coatings designed and applied toavoid thrombosis also decreased bacterialadherence. One such coating is "hydrom"which is a hydrogel material.An in vitro assessment of this material

applied to shunts was presented. Specimensof tubing, coated and uncoated wereimmersed in bacterial suspensions of varyingconcentrations for various times. Adherencewas assessed by biofluorescence staining withautomated counting. Several strains of bacte-ria were used, principally Staphylococcusepidermis, including isolates from shuntinfections. In both kinetic and concentrationstudies a tenfold advantage of the coatedmaterial was demonstrated. Unfortunately,poor bonding between hydromer and shuntmake the method currently impractical.

SHUNT IMPLANATATION: TOWARDS ZERO INFECTIONDA Lang, G Lena, L Genitori, E Empime,M Choux. Marseille, France

Reduction of shunt infection has importantimplications for late neurological and intel-lectual development in children with hydro-cephalus. However, infection is the mostimportant complication of shunt implanta-tion in children, and even in the most recentseries infection rates range from 2-5%.

In Marseille shunt practice was audited inorder to study the shunt infection rate and todetermine the risk factors associated withshunt infection. Two groups of patients werestudied. From 1975 to 1982, 1068 proce-dures were performed in 517 children. In thisgroup the infection rate was 5-60% (10-55%per patient). A new protocol for shuntimplantation was then introduced, compris-ing step by step modifications to the pre,intra and post operative management of thechildren and between January 1983 andDecember 1990 have carried out 1043 pro-cedures. In this group of 617 patients therewere two infections. One of these occurred ina new patient and the other after a shuntrevision. Overall after introduction of thenew protocol the infection rate was 0-19%(0-32% per patient).Shunt infection is an avoidable operative

complication and that a 0% infection rate is arealistic goal. Shunt implantation is a com-plex procedure and must be done by anexperienced neurosurgeon.

A PROSPECTIVE STUDY OF IDIOPATHIC NORMALPRESSURE HYDROCEPHALUS-GUIDELINES FOROUTPATIENT INVESTIGATION

JD Pickard, H Newton, A Greene, HKRichards, AH Lovick, DA Lang, RD Hoare.Cambridge, Southampton and London

Elderly patients with possible NPH are frag-ile and their investigation and managementas inpatients may prove protracted, expensiveand of uncertain overall benefit. The value ofCT, MRI, Cerebral Blood Flow (SPECT;

517 on M


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from


518 Proceedings of the Association of British Neurologists, London, September 1991

Predictive Factor (n) Sensitivity Specificity AccurayNo vascular history (24) 57% 47% 50%CT-gyral atrophy (25) 43% 44% 44%-absent PVL (25) 57% 89% 80%MR-absent deep white

matter lesions (21) 71% 86% 81%IPC-B waves (>30%) (14) 80% 88% 79%- t (<0*09) (22) 100% 24% 41%

133Xenon) and ICP monitoring with CSFoutflow conductance measurements in pre-dicting the response to ventriculo-peritonealshunting was assessed in 25 patients (meanage 73 years; 19 M; 6 F) with possibleidiopathic hydrocephalus. Detailed assess-ment included psychometric and functionalscores (table).

In this age group, patients with idiopathichydrocepahlus, gait apraxia and slowness ofmentation should be considered for shuntingwhere there are no focal deep white matterlesions on MRI (71% (5/7)) improved formore than one year; revision: 1; postoperativefits: 1;). Such white matter lesions do notpreclude considerable benefit from shuntingin a minority (2 excellent results out of 14patients-14%) but further support by 24hour ICP monitoring and test removal ofCSF should be used to justify the risk ofsurgical intervention in this vulnerable sub-group (postoperatiuve cardiac arrest-resus-citated: 1; postoperative hemiparesis: 1;infection: 2; fractured femur: 1; subduralhaematoma: 1; revision: 4; late strokes (3)and late deaths (5) not necessarily related tooperation).Supported by the Sir Jules Thorn Chari-

table Trust.

THE PROBLEM OF PARADOXICAL AIR EMBOLISM ANDITS PREVENTION DURING NEUROSURGERY: THEIMPORTANCE OF PREOPERATIVE CONTRASTECHOCARDIOGRAPHYMP Roberts, LD Gillam, JS Mikan. Hart-ford, Con, USA

Air may enter the vascular system duringalmost any surgical procedure. Air embolisa-tion during neurosurgery is most likely tooccur with the patient in the sitting position.With the exception of massive emboli, the airenters the pulmonary vasculature, passesacross the alveolar membrane, and is harm-lessly expired. Paradoxical embolism takesplace when air is shunted from the right tothe left heart, typically through a patentforamen ovale. Disastrous injury or deathmay then result if even a small quantity of airenters the cerebral circulation or coronaryarteries.Two cases of paradoxical air embolism

during cervical disc surgery in the sittingposition were discussed, one in which thepatient was left decerebrate and a secondcase in which the patient suffered a chronicParkinsonian syndrome. The techniques usedto prevent air embolism and the importanceof preoperative contrast echocardiographywere detailed.

epilepsy due to arteriovenous malformations(AVM) as proven by head CT scan andcerebral angiography. They were studied pro-spectively with dynamic cerebral blood flowand brain SPET, at the same session, usingnon-diffusible human serum albumin (HSA)labelled withTc-99m (600 MBq). They werefollowed up for a mean of9 5 ± 1 months andhad 2-4 ± 0-2 studies.Two types of cerebral AVM could be

identified: 1) Fast Flow AVM (65%) withaffected to non-affected hemisphere meantransit time ratio of 0-87 ± 0-02; and a flowratio of 1 X31 ± 0 04; and 2) Slow Flow AVM(35%) with a mean transit time ratio of 1-10± 0-03 and a flow ratio of 0-92 ± 0-06, whichtended to be mainly venous AVM. Bothdifferences were significant for transit timeand flow (p < 0-01). Studies with HMPAOshowed significant perfusion defects, in con-trast to HSA-SPET, indicating seriousregional malfunction.

In the fast flow AVM group reduction inflow was detected between three and sixmonths after radiosurgery. This contrastswith the minimum twelve month period forangiography.' This assessment may prove ofgood prognostic value.

1 Kemeny, A, et al. Neurol Neurosurg Psychiatry1989;52:554.

SURGICAL OUTCOME WITH ARTERIOVENOUSMALFORMATIONSAR Aspoas, AD Mendelow, J Arrotegui, AGholkar. Newcastle

Despite advances with interventional neuro-radiology and stereoradiosurgery, surgicalexcision of arteriovenous malformations(AVM) remains an effective form of treat-ment, often used in conjunction with thesenew developments. To balance the hazards ofsurgery against the natural history ofAVM, itis necessary to be able to predict the outcomefrom surgery (table).

Spetzler' proposed a grading system basedon angiographic findings which predicts theoperative morbidity and mortality of surgicalexcision. The results in forty four patientswere compared with those of Spetzler toassess the validity of the grading system.

Spetzler grading is useful in quantifyingsurgical risk.

1 Spetzler RF, Martin NA. Neurosurg 1986;65:476-83.

MANAGEMENT OF UFE-THREATENING ACUTEINTRACEREBRAL HAEMATOMAS DUE TO VASCULARLESIONS.D Gentleman, R Bullock. Glasgow

The management ofunconscious or deterior-ating patients with life-threatening clots andunderlying vascular anomalies is still con-troversial. Even among experienced surgeonsattitudes range from therapeutic nihilism tovigorous intervention.The authors reported their recent experi-

ence with 30 patients (60% male, 90% agedunder 50) presented in coma or deterioratedafter admission, in whom CT scan showed amassive clot from rupture of a cerebralaneurysm (15 cases) or arteriovenous malfor-mation (AVM) (15 cases). The clots were inthe left cerebral hemisphere (14 cases), righthemisphere (15 cases), or cerebellum (onecase). All patients had emergency clot evac-uation, 90% within 24 hours of the onset ofcoma or deterioration, with pre-operativecerebral angiography in 19 cases. At the sameoperation, 14 of 15 aneurysms were clipped,and six of 15 AVMs were excised. At sixmonths, outcome showed good recovery ormoderate disability in 20 cases, five wereseverely disabled, and five died. Outcomewas unaffected by pre-operative coma scoreor by the hemisphere affected, but patientswith AVMs did better (13 of 15 good ormoderate) than those with aneurysms (sevenof 15 good or moderate).These results show the value of emergency

clot evacuation, with or without definitivesurgery, in such desperately ill patients,especially if there is evidence that the under-lying lesion is an AVM.

SKULL BASE CHORDOMASLD Watkins, ESS Khudados, M Kaleoglu, HAlan Crockard. Queen Square, London

The presentation and results of treatmentwere reviewed for 45 patients with skull basechordoma presenting between 1958-88.With few exceptions,' previous studies havecombined results for clival and sacral chordo-mas, or for chordomas and other similartumours such as chondrosarcoma. Analysisof the survival data for our patients suggeststhat there are two subgroups with distinctsurvival patterns: one group with high mor-tality within the first four years, and a secondgroup with an indolent disease process andnear normal life expectancy. Survival did notcorrelate with the year of operation, suggest-ing that changes in treatment over the periodof the study did not distort the results. Seriesin which conventional radiotherapy has beengiven to all patients after surgery have notshown greatly improved survival comparedwith results published 57 years ago for

HAEMODYNAMIC FUNCTIONAL CLASSIFICATION OFCEREBRAL ARTERIOVENOUS MALFORMATIONS USINGSINGLE PHOTON EMISSION TOMOGRAPHY (SPET)AHS Huneidi, MJ Carroll, J Bomaji, CNimmon, J Jenkins, DS Montefiori, PNPlowman, P Hamlyn, KE Britton, F Afshar.St Bartholomew's Hospital, London

Forty five patients, age 35 ± 2 yrs (23 M),presented with cerebral haemorrhage and/or

Table Independent = normal/good recovery/minimal or moderate disability; Dependent = severe disability/majordeficit/vegetativeNo Independent (Si) Dependent (%i)

Grade P S P S P SI 6 23 100 100 0 0II 8 21 100 100 0 0III 15 25 93 96 7 4IV 9 15 100 93 0 7V 6 16 67 88 33 12Total: 44 100P: Present study. S: Spetzler study.



j.com/

J Neurol N


ownloaded from



patients treated by surgery alone. However,proton radiation therapy, has shown verypromising results.2 A treatment protocol issuggested initially using subtotal excisionalone, supplemented with proton radiation inthose patients who have early recurrent dis-ease. This treatment protocol could be com-pared with a matched group for whomproton therapy is used immediately followinginitial surgery.

1 O'Neill P, Bell BA, Miller JD, et al. Neurosurgery1985;16: 166-70.

2 Austin-Seymour M, Munzenrider J, Goitein M,et al. Neurosurg 1989;70:13-17.

IMMEDIATE REVERSAL OF "CRUCIATE PARALYSIS"FOLLOWING TRANSORAL REMOVAL OF ODONTOID IN

RHEUMATOID ARTHRMS WITH BASILARINVAGINATION AND CHAIRI I MALFORMATIONG Mohr, R Schondorf, M Black, M Maleki.Montreal, Quebec, Canada

"Cruciate paralysis" described by HS Bell'consists of disproportionate motor deficitsinvolving both upper limbs with relativesparing of the lower limbs and is usuallydiagnosed initially as central cord syndrome.The currently accepted, although still con-troversial patho-physiologic mechanisms ofthis "alternate brachial diplegia" relates tothe midline involvement of the more rostallydecussating upper limb fibres ofboth cortico-spinal tracts at the medullo-cervical junction.This rare entity has been described in uppercervical spine trauma, mainly odontoid frac-ture-dislocation, rheumatoid arthritis andbasilar invagination with cerebellar ectopia.

This 47 year old woman with severerheumatoid arthritis developed rapidly pro-gressive tetraparesis with complete flaccidparalysis of both upper limbs and less severeinvolvement of lower limbs as well as severelower cranial nerve deficits requiring tra-cheostomy and feeding jejunostomy. Burningdysesthesia and neuralgic pain in trunk andupper extremities were also present. Becauseof significant medullary compression fromodontoid impaction associated with basilarinvagination and Chiari I malformation, pos-terior fossa decompression and lateral occi-put-C2 bony fusion were first performedwithout improvement.

In a second sitting, via transoral transphar-yngeal approach, the odontoid process andpart of the C2-body were removed using ahigh-speed diamond drill and the transverseligament of the axis was divided longitudinal-ly, resulting in bulging of the dura throughthe bony decompression. Immediately post-operatively, return of motor function of botharms was noted, followed by a rapid normal-isation of her gait. The tracheostomy andjejunostomy were removed ten days laterafter complete restitution of swallowing andspeech capacities.Rapid and complete motor recuperation

has consistently been observed in cruciateparalysis and indicates minimal parenchymaldamage but important physiological impair-ment. Another patient with similar symptomsfrom rheumatoid dens-subluxation and rapidreversal of branchial diplegia after C1-C2posterior fusion was also presented.

1 Bell HS. J Neurosurg 1970;33:376-90.THE VALUE OF MONITORING TRANSCRANIAL

DOPPLER AND JUGULAR BULB VENOUS OXYGEN

SATURATION DURING INTRACRANIAL PRESSURE

THERAPY AFIER SEVERE BRAIN INJURY

KH Chan, NM Dearden, S Midgeley, JDMiller. Edinburgh

The role of transcranial doppler (TCD) andjugular bulb venous oxygen saturation(SJ02) in monitoring intracranial pressure(ICP) therapy was examined in 22 patientswith severe brain injury undergoing con-tinuous recording of cerebral perfusion pres-sure (CPP), arterial oxygen saturation andSJ02, from which the arterial-jugular venousoxygen content difference (AVDO2) couldbe derived. Continuous recording of middlecerebral artery blood flow velocity and pulsa-tility index (PI) was performed during chang-ing CPP. Treatment was initiated when ICPremained 22 mmHg or CPP fell below 60mmHg. Treatment was considered successfulonly if ICP fell below 20 mmHg and CPPwas increased or preserved above 60mmHg.

Forty nine treatment responses were recor-

ded. Therapy always reduced ICP to someextent. When pretreatment CPP was 60mmHg and treatment was successful, CPPchange (post-treatment minus pretreatmentvalue) correlated with changes in SJ02 (r =0-678, p < 0-01), AVD02 (r = -0-69, p <0-01) and PI (r = -0-76, p < 0-001),indicating global improvement in cerebraloxygen delivery. In contrast, with treatmentsuccess at pretreatment CPP 60 mmHg,there was no change in SJ02, AVDO2 or PIbut CPP increase correlated with increases inmean velocity (r = 0-501, p < 0 05), suggest-ing intact autoregulation. In seven (all hyp-notic therapy) out of 10 treatment failures,CPP decreased. This was associated with fallsin SJ02 and mean velocity and rises in PI andAVDO2.

In conclusion, CPP is the most crucialparameter to monitor during ICP therapy.TCD and SJ02 recordings may complementCPP monitoring in identifying a lowerthreshold value for adequate CPP duringtherapy.

CAN DOCUMENTATION OF HEAD INJURIES IN

ACCIDENT AND EMERGENCY DEPARTMENTS BE

IMPROVED? RESULTS OF PILOT STUDIES USING A

NEW HEAD INJURY PROFORMA

RW Gullan, PO Byme, S Wallace. BrookHospital, London and Brighton

Head injury accounts for approximately 10%of all accident and emergency attendances. Acomprehensive audit of "free hand" doc-umentation of 158 head injuries in one of thethree hospitals used in the survey demon-strated a poor level, or even a completeabsence, of reference to what could bedescribed as fundamental clinical details. Aspecial proforma for documenting thesepatients has been designed to allow allaspects, including nursing notes, neurolog-ical observations, a simple diagnostic triageand management/treatmnent section, to bekept together on a single form. Adequatespace for recording other injuries or opinionsis available, thus avoiding separate records.Every effort has been made to make the inputof data as easy as possible and mostlyYes/Noor tick boxes are used, so that subsequentcomputerised data storage and easy auditwould be possible in the future.Two separate pilot studies (684 and 576

consecutive patients in each) have been

undertaken and after the first, various recom-mendations incorporated, including the pro-duction of a form for children less than sixyears old. Some dramatic improvements indocumentation (compared with "free hand"casualty notes) have been obtained: forinstance, post traumatic amnesia recorded in98-5% of cases compared with 34% before;evidence of whether alcohol had recentlybeen consumed or not recorded in 97% ofcases as opposed to 43%; and evidence ofwhether or not there was a compound injuryin 89% as opposed to 10% previously recor-ded. At the completion of the second study,out of 22 Casualty Officers involved, onlyone was not in favour of the proforma beingused on a permanent basis. The structure,design, problems and advantages of this formwere discussed with reference to its use in1260 patients in these two pilot studies.

IS HELMET USE BENEFICIAL TO MOTORCYCLISTS?VG Wagle, C Perkins, A Vallera. Hartford,CT, USA

This study included 83 motorcyclists, hel-meted and nonhelmeted, involved in acci-dents. The majority of these patients wereflown into a major trauma centre in theNortheast directly from the scene by ahelicopter ambulance service (Life Star). Thestudy showed that the majority of thesepatients were male, and under the age of 30.Twenty seven had alcohol levels above thelegal limit. Sixty nine per cent were non-helmeted with a statistically significant great-er number in this category with a GlasgowComa Scale of 8 or under. Cervical spineinjuries were discovered in only six patients,with none of the injuries occurring in theupper cervical region. Nine of the tenpatients declared dead on arrival (DOA)were nonhelmeted.

In conclusion, helmets provided protectionand did not increase the incidence of cervicalspine injuries.

AN ANALYSIS OF BRITISH NEUROSURGICAL PRACTICE

IN THE MANAGEMENT OF CHRONIC SUBDURAL

HAEMATOMA

PO Byrne, JR Bartlett. Brook Hospital,London

Chronic subdural haematoma is a commoncondition which seems to be managed in avariety of ways with a relatively favourableprognosis. To get an idea of the frequencies ofthe various different approaches to manage-ment, a postal questionnaire was sent toevery consultant neurosurgeon in the UnitedKingdom in the early part of 1991. It wasdesigned to examine the operative techniqueand postoperative management in patientswith symptomatic chronic subdural haema-toma warranting intervention. At attemptwas also made to discover if there was an agelimit or clinical state that would precludetreatment at all.One hundred and twelve replies were

received. Most respondents used one ormore, usually two, burr holes and generallyirrigated the cavity, often using a catheter; aquarter would sometimes leave in a drain.One used craniotomy as the main treatment.Postoperative bed rest was employed by halfof respondents, usually for forty eight hoursand not infrequently longer, with less than a

519 on M


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from



fifth allowing immediate mobilisation. Mostused clinical criteria to assess recurrence,though a quarter used CT scanning andthree used needle aspiration. A quarter ofrespondents used steroids, fifteen using themregularly.

STEREOTACTIC RADIOFREQUENCY UPPER THORACIC

SYMPATHECTOMY-AN ELEVEN YEAR EXPERIENCE

HA Wilkinson. Worcester, Mass, USA

Between April 1979 and March 1991 theauthor performed 121 sympathectomies on205 limbs in 91 patients using a percuta-neous stereotactic radiofrequency technique,usually on an outpatient surgery basis.Patient ages ranged from 10 to 81 years, with41 males and 50 females. Four had hadprevious open surgical sympathectomyunsuccessfully. Patients suffered from hyper-hidrosis, vascular occlusion, Raynaud's orother chronic vasculopathies, painful cau-salgia or reflex sympathetic dystrophy orPrinzmetals' angina.The sympathectomy technique has evolved

and is currently in its third phase. Changeswere based on anatomical and clinical/radio-graphic correlations as well as careful patientfollow up assessments. The modificationshave reduced the frequency of early failuresand seem to be reducing long term failures.The present technique relies on neurolept-analgesia with local anesthesia of superficialtissues only. Two 18 gauge radiofrequency"tic" needle electrodes are employed, with10 mm bared tips for adults and 7 mm tipsfor children. A series of three lesions at 90degrees for 180 seconds are made rostro-caudally at each of the ganglion sites selected.Lesion sites are based on C-arm fluoroscopyand an electrical output to produce a thres-hold greater than 1-0 volt. Lesion effective-ness is monitored by finger plethysmographyand hand skin temperature measurement.Complications have generally not been severeexcept for symptomatic pneumothorax insome patients. Re-operation is easily per-formed when necessary.

Epilepsy Surgery Symposium

LONG TERM CONTROL OF EPILEPSY FOLLOWING

TEMPORAL LOBECTOMY IN PATIENTS WITH LOW

GRADE TEMPORAL TUMOURSPJ Kirkpatrick, M Honvar, CE Polkey.Maudsley Hospital, London

Thirty seven patients (age range three-53yrs) with tumour related temporal lobe seiz-ures were presented. All had suffered chronicdrug resistant temporal lobe seizures (meanage of onset 8 1 yrs, range 0-30 yrs; meanduration 11 1 yrs, range 2-39 yrs) andrepresented 20-5% of all cases undergoingtemporal lobe resection for epilepsy over a 14year period. Preoperative EEG tracings indi-cated localised epileptic foci in 89% ofpatients and of the 35 CT scans performed89% showed temporal lobe abnormalitiesincluding calcificaiton (47%), low densitylesions (31 %) and isolated temporal horndistortion (11%). Electrocorticography(ECoG) was performed during 34 of theoperations; 21 (61%) showed an improvedpost-resection tracing. Tumour remvoal wasmicroscopically incomplete in 70% andreview of pathological specimens revealed ahigh incidence of dysembryoblastic neuro-epithelial tumours. Postoperative mortality

was zero, but one patient developed a perma-nent hemiplegia. Psychological assessmentshowed no measurable change in verbal andperformance intelligence quotient followingsurgery, but a mild impairment of memorywas detected in 51%. Long term follow up(mean duration 5-6 yrs) indicated that 78%of patients were completely free of epilepsy.Relief of seizures could not be predicted bythe intraoperative ECoG tracings nor by theearly postoperative EEG findings whichshowed epileptiform phenomenon in 68%.Outcome was also independent of tumourpathology and completeness of resection.

UNILATERAL TEMPORAL NEOCORTICECTOMY IN THE

MANAGEMENT OF INTRACTABLE TEMPORAL LOBE

EPILEPSY-LONG TERM OUTCOME AND

NEUROPSYCHOLOGICAL SEQUELAED McMackin, J Phillips, T Burke, H Staun-ton. Dublin

Temporal neocorticectomy had been carriedout on over 70 patients at the RichmondInstitute for Neurology and Neurosurgery,Dublin, since 1975. This is a unique proce-dure among centres offering surgery forepilepsy in that excision is limited to thelateral neocortex whilst sparing all mesialstructures.

Factors which predicted a good resultwere: 1) stereotyped onset of the seizure (p <0 05); 2) unilateral antero-mid temporalspike (p < 0-01); 3) greater volume of tissueremoved at surgery (p < 0 05). On the basisof these predictive factors, a sub-category ofpatients was shown to have a 91% chance ofhaving a favourable outcome from neocorti-cectomy. Overall outcome in the unselectedseries, show 60% of patients to be effectivelyseizure free.'The results of pre/post-operative neuro-

psychological evaluation suggested that theexcision does not result in a deleterious effecton memory or cognition as measured by theWeschler Scales. Furthermore, althoughpatients exhibit mild memory deficits relatedto the side of surgery, these are less severethan in patients from Montreal where deeperstructures have been excised.The authors recommended neocorticect-

omy as a conservative approach which canhave maximum benefit in a carefully selectedpopulation.

1 McMackin D, Staunton H. Journal of the IrishColleges of Physicians and Surgeons 199 1;20:193-7.

MULTIPLE SUBPIAL TRANSECTION IN THE

TREATMENT OF FOCAL MEDICALLY RESISTANT

EPILEPSY

GP Kratimenos, DGT Thomas, SD Shor-von, D Fish. National Hospital, London

Based on experimental evidence suggestingthat epileptogenic discharges require hori-zontal interaction of cortical neurons whilecortical function depends on vertical fibreconnections of the columnar units, the opera-tion of multiple subpial transection wasintroduced some 20 years ago, as an attemptto relieve the medically intractable cases offocal epilepsy with epileptogenic foci locatedin functionally important cortical territories.The purpose of the transection of the hori-zontal intracortical fibres with preservation ofthe vertically oriented neural elements as wellas the penetrating blood vessels, was to

reduce the occurrence of synchronised celldischarge without affecting the functionalproperties of the cortical tissue.Four cases of focal epilepsy had been

treated with multiple subpial transectionduring the last five years. In all cases theepileptogenic foci were located under or incortical areas of major functional importanceand therefore non-resectable. There were 2male and 2 female patients with ages rangingfrom 21 to 33 years and with a history ofmedically intractable focal epilepsy of a meanduration of 12 years. All lesions were locatedunder the motor cortex, two in the right andtwo under the left. Three of the patients hada radiologically identifiable structural lesion.Two of the lesions were biopsied sterotac-tically before the subpial transection confirm-ing the preoperative diagnoses of low gradetumour. In the third case there was a previoushistory of cerebral abscess treated withaspiration 14 years before the currentoperation.The functional postoperative results were

excellent, although one of the cases wascomplicated by a subdural collection. Theoperation successfully controlled the epilepsyin only one case (the only one withoutstructural abnormality), while from theremaining three cases only in one was anappreciable reduction of seizures detected.From our limited experience conclusions

suggest that the operation of multiple subpialtransections may have a place in the surgicalmanagement of cases of focal epilepsy asso-ciated with non-resectable cortex and with-out an identifiable structural lesion. In thosecases where lesions are present, the proce-dure could be combined with a stereotacticvolumetric excision, as our cases indicate thatin this context, subpial transection alone onlypoorly controls cortical spreading of epilepticdischarges.

INTRAOPERATIVE TESTING OF MEMORY FUNCTION

DURING HIPPOCAMPAL COOLINGKW Lindsay, R Duncan, R Gillham, DCossar, D Sutton. Glasgow

Many centres perform the WADA test rou-tinely before temporal lobectomy for intract-able epilepsy in an attempt to identify theextent to which the hemisphere under testcontributes to language and memory func-tion. The production of complete amnesiawhen Na amytal is injected on the side of theproposed resection creates a major concern,but does not exclude the possibility of resec-tive surgery as WADA testing carries a highincidence of false positives. In one suchpatient intraoperative memory testing wasperformed during hippocampal cooling usinga specially designed probe. The patient wasanaesthetised with propofol and alfentanilinfusion and a laryngeal mask was used forairway control. Local anaesthetic wasinstilled around the craniotomy wound andat the sites of the headrest fixation. Aftercompleting the first stage of a right temporalresection and exposing the hippocampus, thepropofol was discontinued and the laryngealmask was removed. Memory testingremained intact before and after cooling thehippocampus to 23 degrees centrigrade, indi-cating the feasibility of proceeding to mesialresection. This was performed after reintro-ducing the propofol. Neuropsychologicalassessment in the immediate postoperativeperiod detected no evidence of memoryimpairment.

520 on M


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from



THE PLACE OF FORAMEN OVALE TELEMETRY IN

PRESURGICAL EVALUATION OF PATIENTS WITH DRUGRESISTANT EPILEPSY

CE Polkey, CD Binnie. Maudsley Hospital,London

Videotelemetry with foramen ovale electro-des was first introduced byWieser as a meansofdemonstrating the onset ofpartial complexseizures in the mesial temporal structures.Since 1987 over 130 patients had beeninvestigated using this method of recording.The electrodes were inserted under general

anaesthesia and with radiographic control. Inthe first eighty or so patients single contactwires were inserted through needles used forthermocoagulation introduced through theforamen ovale by the usual method. In theremaining patients multi-contact electrodeswere inserted through splittable cannualae.All patients received antibiotic cover whilstthe electrodes were in place. Recordings werecarried out for periods of up to ten days withanticonvulsant withdrawal, sleep deprevationand close supervision.The method proved reasonably safe, the

youngest patient investigated was five yearsold. There was no mortality and no majorneurological morbidity. Three patients hadproven meningitis, one from an unsuccessfulinsertion; all were rapidly controlled withintravenous antibiotics. A number of patientshad numbness in the divisions of the fifthnerve, in proportion to the difficulty ofinsertion, but all recovered. In one patientthe electrodes had clearly entered brainsubstance and were removed with 24 hoursand the patient sucessfully reimplanted threemonths later.

HIPPOCAMPAL VOLUMETRIC STUDIES AND SURGICALTREATMENT OF TEMPORAL LOBE EPILEPSYMJ Cook, DR Fish, SD Shorvon, JM Ste-vens. Queen Square and St Mary's Hospitals,London

Determination of structural lesions on imag-ing studies correlates strongly with post-operative seizure prognosis in epilepsy sur-gery. Temporal lobe pathology is mostfrequently Ammons Horn sclerosis (AHS),and several different imaging techniques havebeen utilised in detection of this with varyingsuccess. MRI using a 3D volume techniquehad been used which allowed total hippo-campal volume to be measured and sym-metry evaluated. Hippoccampal surface areawas calculated in sequential 1-5 mm thick

contiguous images, using a IGE IC work-station. Total volumes were calculated andsurface area versus hippocampal length plot-ted, permitting morphometric analysis of thehippocampus. Regional involvement andextent of volume loss were then assessed.Previous studies have all been significantlylimited by thick slices and large inter slicegap, and have not provided this detailedinformation. Patients with hippocampal andextra-hippocampal pathology were studied,and volumes compared with normal valuesderived from 10 neurologically normal con-trols. Hippocampal volume loss was demon-strated in patients with clinically typicaltemporal lobe epilepsy, who did not haveextra-hippocampal lesions. Hippocampalvolume loss did not occur in patients withextra-temporal or temporal neocorticallesions.

Volumetric analysis of the hippocampuscan demonstrate asymmetry, regionalinvolvement, and distinguish between hippo-campal and extra-hippocampal pathologies.

PHASE II PRESURGICAL EVALUATION OF COMPLEXPARTIAL SEIZURES WITH DEPTH ELECTRODESW Harkness, M Levesque. Los Angeles,California, USA

Thirty one patients with complex partialseizures had phase II presurgical evaluationwith stereotactically implanted electrodes(SEEG) between January 1988 and Decem-ber 1990 at UCLA medical center. Therewere 20 males and 11 females in this groupwith a mean age of 30 years (range 16-48).Mean age at seizure onset was 14 years andthe mean delay to phase II evaluation was 16years. Five patients had both depth electrodeimplantation and insertion of subdural stripsin the same procedure (16%). There werefour clinical complications of electrodeimplantation (12-9%) and no mortality. Onepatient required early electrode deplantbefore seizure activity could be recorded onSEEG. As a result of the phase II data 17patients had surgery to remove the identifiedepileptogenic focus. En bloc temporal lobec-tomy was carried out in 14 cases, a selectiveamygdalohippocampectomy in 2 cases andan extratemporal resection in 1. Fourteenpatients failed to proceed to surgery follow-ing phase II. Three cases had seizures loc-alised to a mesial temporal focus but did nothave surgery for other reasons, while five hadbitemporal seizure onset and five had extra-temporal, diffuse or nonlocalised seizure

onset. The phase I data for all cases wasreviewed and analysed in an attempt toidentify factors that would suggest previousto phase II that the patients would not besurgical candidates, thus preventing unnec-cessary invasive recordings.

APPLICATION OF THE G-T LOCALISER FOR DEPTH

ELECTRODE INSERTION IN THE INVESTIGATION OFDRUG RESISTANT EPILEPSY

GP Kratimenos, DGT Thomas, SD Shor-von, DR Fish. National Hospital, London

Intracerebral insertion of electrodes for elec-troencephalographic recording provides anaccurate spatiotemporal localisation of theepileptic activity during the pre-operativeevaluation of patients with intractable epi-lepsy not otherwise localised. The localisingvalue of the method depends on the preciseand predetermined placement of the electro-des and for this, a stereotactic technique ofinsertion based on information derived fromcomputerised imaging (CT and MRI) andsterotactic angiography is highly desirable.

Various methods of multimodal stereo-tactic image integration have been described.A common limitation is the serial perform-ance of the imaging studies following thestereotactic frame application or if the frameis to be removed between studies, localisingpins penetrating the outer skull table mustremain in place. Computerised analysis andimage integration is an inherent part of thedescribed methods and the operation takesplace immediately following the target defi-nition.An alternative method of stereotactic aqui-

sition of multimodal image information waspresented which uses the G-T stereotacticrelocatable localiser. The temporal freedomprovided during the investigation-operationperiod could potentially be advantageousoffering an unhurried multi-image integra-tion and targetting in any individual casecombined with less discomfort for thepatient. The accuracy of the system is wellproven and the localising value of the tech-nique is excellent as the presenting casessuggest. There were no complications in anyof the patients included in the study althoughthe recording period extended in some casesas long as 18 days. The accuracy of theelectrode placement was easily verified dur-ing the post-operative period using standardradiographic and MRI studies and this infor-mation added to the localising value of thetechnique.

521



j.com/

J Neurol N


ownloaded from


Proceedings Society ofBritish Neurological Surgeons the ... · CRANIOPHARYNGIOMA RHayward, UAndar,...

Documents

Transcript of Proceedings Society ofBritish Neurological Surgeons the ... · CRANIOPHARYNGIOMA RHayward, UAndar,...