Journal of Neurology, Neurosurgery, and Psychiatry 1987;50:375-382

Hysterical paraplegiaJ H E BAKER,* J R SILVERt

From the Rookwood Hospital,* Cardif, Wales, and National Spinal Injuries Centre,t Stoke MandevilleHospital, Aylesbury, Bucks, UK

SUMMARY Between 1944 and 1984 20 patients were admitted to a spinal injuries centre with a

diagnosis of traumatic paraplegia. They subsequently walked out and the diagnosis was revised tohysterical paraplegia. A further 23 patients with incomplete traumatic injuries, who also walkedfrom the centre, have been compared with them as controls. The features that enabled a diagnosis ofhysterical paraplegia to be arrived at were: (1) They were predominantly paraplegic, (2) There was a

high incidence of previous psychiatric illness and employment in the Health Service or alliedprofessions, (3) Many were actively seeking compensation. The physical findings were a

disproportionate motor paralysis, non anatomical sensory loss, the presence of downgoing plantarresponses, normal tone and reflexes. They made a rapid total recovery. In contrast, the controltraumatic cases showed an incomplete recovery and a persistent residual neurological deficit.Investigations apart from plain radiographs of the spinal column were not warranted, and thediagnosis should be possible on clinical grounds alone.

The diagnosis of non organic paraplegia is not easy.Problems arise with this diagnosis because ofdifficulties with psychiatric terminology and uncer-tainty as to outcome. Cases considered on psychiatricgrounds to be classical cases of hysterical paraplegiahave subsequently turned out to have organic disease.Thus the outcome of a complete recovery must beused as one of the criteria for establishing the diagno-sis confidently as being non organic. There is a sep-arate and more common phenomenon of elaborationof real deficit which we will not analyse in this paper.The practical question is "how does the physician

diagnose the patient likely to have a feigned or nonorganic paraplegia?" Between 1944 and 1981, 20patients with non organic paralysis were referred tothe National Spinal Injuries Centre (NSIC) fromreferring hospitals having been completelyinvestigated and considered to have traumatic spinalcord injuries. These patients subsequently recoveredcompletely and walked out of the National SpinalInjuries Centre, after they had been diagnosed as hav-ing hysterical paraplegia. We have therefore studiedthese 20 patients and in addition three others whowere consecutively referred to Spinal Injuries Centresassociated with one of the authors, JHB. This total

Address for reprint requests: Dr J R Silver, Stoke MandevilleHospital, Aylesbury, Bucks HP21 8AL, UK.

Received 14 February 1984 and in revised form 24 July 1986.Accepted 26 July 1986

group is termed the "non organic" group. We havecompared the clinical findings, investigations andnatural history with a further 23 control patientsalso seen by JHB, in whom there was no doubt thatthere was a traumatic spinal cord injury. These alsomade a recovery to the extent of being able to walk.We have attempted to isolate the features of his-

tory, examination and investigation which differbe-tween the two groups which might be useful in pre-dicting non organic paraplegia. Whether these fea-tures conform to the accepted criteria for thediagnosis of Hysteria is open to discussion. Indeed, inpractice, whether these patients have an isolated con-version reaction, are malingerers, or are merelyimmature personalities reacting to a life situation witha feigned neurological complaint is often impossibleto deduce.

Material and incidence

Twenty patients were admitted or seen as outpatients at theNSIC. They were initially thought to be cases of traumaticparaplegia but subsequently the diagnosis of non organicparaplegia was made at Stoke Mandeville between 1944 and1981. A further three patients were similarly seen atOswestry in 1984.The 20 Stoke Mandeville cases were seen among a total of

7000 cases ofwhich 5000 were traumatic and 2000 diagnosedas medical cases of paraplegia, for example multiple sclerosisor tumour. The three cases at Oswestry were seen over a two

year period; the referral rate being 80 traumatic and 20 med-ical cases per annum.

375

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

376

The control patients were provided by 23 patients whowalked out of a centre having recovered from a traumaticspinal cord injury. In these patients we were confident thatan organic lesion had been sustained by virtue of a vertebralfracture or an appropriate injury associated with bruising,tenderness, soft tissue injury and associated injuries com-bined with the clinical picture.

Sex distributionThe distribution of the non organic group was 14:9(male:female). The distribution for traumatic paraplegia was20:3 (male:female). We have been unable to match the sexdistribution in our control patients.

InjuryAll patients in both series identified an injury to which theyrelated the start of their symptoms. No difference in declaredseverity was noted. Some of the patients in the non organicgroup had falls of a minor nature which could support adiagnosis of an extension injury but this was only in retro-spect following radiography. It was difficult to reconcile theinjury with the severity of the paralysis. In contrast in theorganic group, the controls, it was always possible to recon-cile the severity with the mechanism of the injury and withthe clinical picture.

Associated injuryThere was no associated injury in the non organic group. Incontrol cases, there was the appropriate associated traumaof a head injury associated with a cervical injury, fracturedribs seen in thoracic injuries and fractured long bones. Onepatient had an associated brachial plexus lesion and anothera bulbar palsy. The brachial plexus injury was seen in amotor cyclist and arose from impacting on his shoulderwhen decanted from his machine. The bulbar lesion relatedto a very incomplete high cervical lesion and remitted com-pletely.

Time ofpresentationAll non organic cases were referred to us in the immediateperiod (up to four months) following injury. One patient wasadmitted on two occasions with a paraplegia following sep-arate accidents. The first time he was admitted within daysand the second time after an interval of 4 months followinga shipboard incident in Dubai. All our control series werereferred in the immediate period also. We also had availablethe case records of the primary referring hospitals and couldreview the signs presenting in casualty and on their wards aswell as when we first saw them.

Level ofparalysisAll non organic patients presented initially with a paraplegiaand there were no primary tetraplegics. Two presented sec-

ondarily; one woman was admitted from the Stoke Man-deville Games, having sustained a trivial injury whensomeone banged the back of her wheelchair. Her upper limbsymptoms remitted within 36 hours and she returned to

competition; paraplegic as before. When we informed her"parent" spinal centre, they replied that they considered herparaplegia to be functional also. The second case was a mer-

chant seaman who had presented with a paraplegia after hisfirst accident, made recovery and went back to sea. There,

Baker, Silver

following a shipboard fall, he became tetraplegic. This wasalso non organic and once more he made complete recovery.

In the control group, there were nine cervical injuries and14 dorsal and lumbar injuries.

Employment: maleFour non organic patients were serving in the ArmedForces. One was a merchant seaman and one was a police-man. Two patients were prisoners. One was working in theHealth Service as a porter. The distribution in the controlgroup showed no pattern other than that dictated by age(25% were schoolboys/students). There were no armed orpublic service employees. None were in legal custody andonly two were unemployed.

Employment: femaleFour of the nine females were involved in the health system.All were Registered Nurses. One was a servicewoman. Nohealth workers were in the control female group; this wastoo small to draw any direct conclusions.The high incidence of five males and females employed

within the health service out of a sample of 23 contrasts withthe absence of any in the control group.

PAST MEDICAL HISTORYMedical In the non organic group, one patient sufferedfrom Addison's disease and had had tuberculosis in the past.The control group was little different; there was one case oftuberculosis and an asymptomatic ventriculoseptal defect.Orthopaedic One non organic patient without underlyingbone disease had sustained nine separate accidents produc-ing broken bones. Another male had had four operations onthe same knee. Two control patients had had one fractureeach, both occurred in association with previous motor cycleaccidents.

Gynaecological In the non organic group one woman hadhad a previous hysterectomy: another was demanding a hys-terectomy from her medical attendants; she had been rapedat the age of eight and had given birth to several malformedinfants. There had been two other gynaecological pro-cedures. None had been sterilised. No gynaecological oper-ations took place on any of the women in the control group.As the numbers were not sex matched, no significance can bedrawn from these figures.

PSYCHIATRIC ILLNESSSeven non organic patients had previously been admitted tomental hospitals. The details are in table 1. The controlgroup had two patients with psychiatric disturbance. Onepatient had had ECT 25 years earlier and had been wellsince. The other was suicidal from a depressive psychosisand her spinal cord injury was due to self precipitation froma window in a suicide attempt.

Table I Psychiatric hospital admission

Admitted for intruding foreign bodies IPrisoner ILSD abuse IDepressive illness 2Admission for narcosis therapy IPersecution delusions I

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

Hysterical paraplegia

Table 2 Previousfunctional disorder

Non neurologicalCardiac neurosis 2Functional dyspnoea IHyperventilation syndrome INeurologicalFoot drop IParalysis IField defect ISeizures IBizarre ankle posturing IDysphonia 2Dysphagia IDysgeusia IAnosmia I

Table 3 Modellingfactors

Previous spinal injury with full recoveryPrevious stab wound of back: no neurological signsPrevious traction for muscular back injuryParaplegic mother in lawCohabiting catamite with paraplegia from DS

PREVIOUS FUNCTIONAL DISTURBANCEPresentations with functional disorder had occurred beforein 14 non organic patients and in 10 of these the symptomshad related to the central nervous system. The details are

shown in table 2. There were no previous functional dis-turbances in the control group.

SPINAL-FAMILY HISTORYFive non organic patients had had previous injuries to thespine (non paralytic) resulting in full recovery or there hadbeen spinal cord injury or non traumatic paralysis in theirfamilies or associates. The details appear in table 3. Onewoman had "sprained her neck" in a previous hunting acci-dent. There had been no bony, ligamentous or neurologicaldamage and she had regarded this as "trivial". It was

recalled only because it had been recent. No control patientshad a past history of spinal trauma.

SECONDARY GAINIn the non organic males, three servicemen were in theprocess of attempting to secure their discharge from thearmed forces; two were prisoners; one was a victim ofcriminal violence; one was a policeman injured on duty; one

man was avoiding arrest; one man was suing his employerfor negligence; one was suing the driver of the car in whichhe was injured. In the remainder, no obvious cause ofsecondary gain was evident. Altogether 10 out of 14 caseshad some potential for secondary gain.

In the women, three had reached an active stage in theirmatrimonial litigation; one was attempting to coerce theService authorities to return her husband from the SouthAtlantic (a previous functional disorder had succeeded insecuring his return from an unaccompanied overseas

posting); one girl had been forced by her parents into a

musical career for which she had no aptitude. Thus potentialfor gain was present in five out of nine cases.

In the control group, only four civil actions were pending;two were car passengers suing the driver, one for theCriminal Injuries Compensation Board and one IndustrialInjury. In no case was liability in doubt. There were no preexisting criminal proceedings. Minor motoring proceedings

377Table 4 Motor and reflexfeatures

Jerky and erratic movements on formal testing 3Agonist/antagonist contractions clinically 4Inconsistency of groups in motor assessment 6Bizarre functional posturing 7Incompatible formal testing with observed acts 5Normal motor tone and reflexes 22Normal plantar responses 23

arose from two cases, both of which were eligible fordisposal in absentia by mail. None of our control cases wereinvolved in matrimonial proceedings, nor were anyprecipitated; nor did matrimonial disharmony present aproblem in management.The overall incidence of potential gain in the non organic

group was 15 out of 23; in the organic group it was six out of23.SELF DISCHARGEFive non organic patients took their own discharge and twoothers effectively did so by engineering events. Theimportant feature was that none of them took their owndischarge while still overtly paralysed: all made recoverybefore discharge and walked out of the unit. This was neverseen in organic paraplegia and no patients in the controlgroup did this.MULTIPLE FACTORSOf the last five features psychiatric illness, previous func-tional disturbance, spinal/family history, and secondarygain, all the non organic patients had at least two of these intheir histories. In the control group, only seven had any fea-tures at all and none more than one.

PHYSICAL EXAMINATION: MOTOR FEATURESAll non organic patients had a complete motor paralysiswhen first seen after the provoking episode. However, threecases improved during transfer to us and had no motor signson arrival. The features which we found helpful in identi-fying the cases of non organic paraplegia are given in table 4.None of these were present in our control group.

Apart from those motor features seen during' formalexamination, in five cases there was marked incompatibilitybetween the examination and observed or self declared activ-ity. One patient was observed skipping in the physiotherapydepartment. One, with an ostensible triparesis, assisted thedoctor taking blood with her allegedly paralysed arm.Another got herself back normally into a wheelchair havingfallen out of it. Another, who had had to be supportedbetween two nurses to move across the consulting room wasobserved walking without assistance to her car. Anotherhaving virtually to be carried across the consulting room wasliving satisfactorily, by her own admission, in a primitivecottage with an outside privy.

In the control group, four patients appeared to have com-plete loss of motor power and sensory appreciation in casu-alty, but some evidence of preservation of sensory functionwithout the return of motor function had appeared by thetime they arrived in the wards of the admitting hospital. Noincompatibility between examination and function was seenin the organic control series.Reflex activityIn the non organic group, the deep tendon reflexes werenormal in 22 cases. One held the limbs rigid with

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

378Table 5 Clustering ofmotorfeatures

No motor symptoms 3No marker features NilOne marker feature NilTwo marker features 7Three marker features 5Four marker features 3Five marker features 5

(marker features are those listed in table 4)agonist/antagonist contractions and reflexes were not elic-itable. There were no patients with enhanced or pathologicalreflexes. The plantar responses were normal (downgoing) inall cases. In the control group the most important diagnosticfeature was the presence of an extensor plantar response.This was seen in 13 cases. In five it was absent, and in two itwas downgoing. The tendon reflex was frequently absent atthe level of the injury. Below the level they showed anorderly progression through the stage of spinal shock tobecome pathologically increased.

ClusteringAll the non organic patients who were still paralysed at thetime they were seen at the centre had two or more of themarker features. See table 5.

PHYSICAL EXAMINATION: SENSORY FEATURESNon organic The useful features seen on sensory exam-ination were inconsistency and non reproducibility of sen-sory signs. Non-dermatomal circumferential truncal sensoryloss was seen in 14 cases; three other cases had non-anatomical sensory disturbances; and in two cases therewere consistently wrong responses on testing of joint posi-tion sense. Several of the features occurred groupedtogether. In three cases, sensation was reported as normal.In two cases none of these markers were found; in six casesone was found; in nine cases, two, and in three cases, three.

In the control group, all patients had some preservation ofsensory function by the time they were examined in the wardof the receiving hospital. In 16 our of 23 cases, some poste-rior column function was seen; it was intact in eight. Sacralsparing was seen in five cases. All examination findings wereconsistent and in an anatomical distribution and, in the caseof the central cord lesions, were compatible and appropriateto the level of the bony injury as shown by bruising,tenderness and the radiological findings.Wasting and pressure soresWasting not seen in either group. No patient in our nonorganic series developed a pressure sore or pressure mark.The control group had three patients who developed soresby the time of admission to a spinal centre.AUTONOMIC FEATURESBladder disturbance No non organic patients had a pri-mary disturbance of bladder function and all continued topass urine normally after the precipitating incident. Four,who were not catheterised, despite complaints that they werein retention, voided normally in the Accident Department ofthe receiving hospital before transfer. Three patients arrivedwith catheters already inserted; these were removed onarrival and they were able to void and to initiate and sup-press flow on command. Three patients developed secondaryurological complaints having been placed in the spinal cen-tre where routine bladder training was being carried out.

Baker, SilverThese cases show inconsistencies with other aspects of auto-nomic function such as wearing a condom and leg bag whileable to initiate and suppress flow to command and with nor-mal urodynamic studies. One woman, with recurrent epi-sodic retention, was further investigated with myelographyand this was negative. This secondary disturbance we regardas an example of acquired behaviour.

Incontinence was seen subsequently in two patients. Onehad ability to initiate and suppress flow to command and theother was a prisoner who had been ignored on a previouscomplaint of retention and had ultimately leapt out of bedand rushed to the lavatory to void, following which he hadfallen to the ground at the base of the urinal, reputedlyunable to move. Left there, he had surreptitiously crawledback to bed. The next time his bladder filled, he was inconti-nent in the bed. We regard both these as volitional.

All patients in the control group except two initiallyshowed retention of urine and required catheterisation. Inthe two with preserved voiding, there was complete sacralsparing. In two other cases, who needed catheterisation,there was some bladder sensation without motor functionand in these cases, there was some sacral sensation but shortof normal. A final case which did have apparent normalsacral sensation was in retention but the conus reflexes weredisturbed and urodynamic studies were abnormal.Bowel disturbance In the non organic group, this wasextremely rare and again showed inconsistencies. One youthperformed manual evacuations despite all other sacral func-tions being normal; the other, despite being allegedly totallyconstipated, had no faecal masses on plain radiographs andat barium enema was able to expel the medium to command.In the control group, the two patients with intact bladderfunction also had intact bowel function. One patient withsacral sparing could feel rectal distention. All other patientsrequired manual evacuations and laxatives with or withoutsuppositories in the first instance.Sexualfunction This was only rarely reported. One patientreported impotence in the presence of preserved normalbladder and bowel function. The other recovered hispotency pari passu with sensorimotor function following aplacebo injection of sterile water. The control group showedappropriate loss of erection and ejaculation consistent withtheir level of injury.

INVESTIGATIONS: PLAIN RADIOGRAPHSPlain films of the spine were obtained in all cases. Only threeof the non organic group were abnormal. One had thechanges of osteomyelitis; one had a spina bifida and thethird had evidence of a previous gunshot wound. The firstcase was investigated with myelography (negative). Therewas no evidence of acute injury to either bony or soft tissue.The second was not investigated further (there was recoverywithin 24 hours and the patient was a drug abuser). The lastwas incidental.

This was markedly different from the control group. Abony injury compatible with the last normal segment wasfound in all save seven cases. In six of these seven, there wasevidence of a hyperextension injury and these were accom-panied by the picture of a central cord lesion. In the last (achild) no definite injury could be seen on first films but onsubsequent films, bony bridging at the appropriate level wasevident.

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

Hysterical paraplegiaINVESTIGATIONS: STUDIES OF CEREBROSPINAL FLUID

Undertaken in six non organic cases, all were normal. Allwere examined before immune phoretic methods were avail-able. This was not performed in the control cases.

INVESTIGATIONS: ELECTROPHYSIOLOGICALEvoked responses were undertaken in three non organicpatients in connection with medicolegal reports. All werenormal. One of the patients, who knew we felt his lesion wasfunctional, defaulted from his first appointment, the nightbefore being admitted to his local hospital with "seizures"which the attending neurologist felt were functional also.Electromyography also showed agonist/antagonist con-tractions. These were not performed in the control group.Sphincter EMGs were however performed in selected uro-dynamic studies. All were abnormal.

INVESTIGATIONS: MYELOGRAPHYMyelography was undertaken in six non organic cases, fourbefore arrival in the spinal centre and two after. All werenormal. We undertook myelography in one patient becauseof significant abnormality on plain films, (osteomyelitis);and in the other because of episodic recurrent retention atOP follow up. Myelography was undertaken in other hospi-tals in five control cases and CT scanning in one prior tosurgery. All showed either a block to contrast flow or canalnarrowing by bone fragments.

SURGERYThis was not undertaken as either a therapeutic or diagnos-tic tool. Two patients had presented with paralysis, however,following negative laminectomies for neurological symp-toms without, according to the submitted record, any neu-rological signs. In the control series, three patients werefinally operated on: two spinal fusions and one decom-pression. One patient deteriorated after surgery transiently.

FARADISMTreatmentThis produced only one cure and this was thought to be dueto the pain produced. Used in two other cases, it producedbizarre movements and posturings but had no therapeuticeffect.

Suggestion, abreaction and placeboSuggestion was used specifically in five cases, producinggood results in all. Intravenous pentothal producedimprovement in one case and made no difference in theother. A placebo injection of sterile water was used success-fully in one man, a teenage soldier from rural Libya. Manip-ulation was used successfully in one case, and one manresponded to the arbitrary injunction to "Get up". In themajority of patients, no specific treatment was used. Thepatient was given a diagnosis of "cord concussion" and thiswas coupled with a firm confident prediction that recoverywould shortly take place.

In the control group, this treatment clearly was notmatched or controlled. They were admitted to the spinalcentre and received the appropriate treatment for theirfractures.

379Table 6 Timefor complete recovery in non organic cases

On arrival/within one day 41 day 32 days 85 days 26 days 17 days 116 days 217 days 118 days I

NATURAL HISTORYThe pattern of recovery in the non organic cases is shown intable 6. All had recovered completely within 18 days; thir-teen within the first 48 hours and all apart from four withina week. Improvement in the sensorimotor function occurredtogether and often in either global or random pattern.

In the control group, four patients deteriorated comparedwith the findings at the accident site following the first neu-rological examination. Two patients walked about at theroadside after injury and one deteriorated after being stood24 hours after injury for fitting of a doll's collar. These caseshad started to improve within hours of the precipitant eventand finished up better than the original level. The patientsimproved in slower time than the non organic cohort, andsensory improvement occurred first, with motor second andautonomic last. Proximal improvement preceded distal, andlower limb in cervical cases, preceded upper (probably dueto the high incidence of central cord injuries).

FOLLOW UP: NON ORGANIC CASESIt is not our policy to give formal follow up appointmentswithin the spinal centre to patients in whom a diagnosis ofnon organic paraplegia has been made because of the risksof developing acquired behaviour. The general practitionersand referring hospitals were contacted by post and followedup any untoward events. Sixteen of our cases have eitherbeen seen or reports have been received. Two patients devel-oped functional seizures. One developed epilepsy, confirmedby electroencephalography, no cause being found. Onepatient, after return to her mental institution, relapsed. Shewas seen by a consultant neurosurgeon who reported a sen-sory level "going up and down like a yoyo". Two patientsdeveloped secondary episodic retention of urine; their man-agement is reported in the section on autonomic dis-turbance, but no organic lesion was found. One patient,living close to Stoke Mandeville, has become a cardiac neu-rotic and has had multiple admissions to our Coronary Unitwith chest pain but no evidence of either cardiac ischaemiasor myocardial infarction.Syndrome shift has occurred in three others, developing

either chest pain in one or asthma in two. Three have beenseen for performance of spinal evoked responses formedicolegal purposes. One reported to the physiotherapydepartment demanding a wheelchair and a disabled parkingsticker while still walking normally and with no signs. Onepatient telephoned to announce she had a pressure sore; avisit by the general practitioner revealed her to be tellingfalsehoods. The last was seen by one of us (JHB) out canoe-ing with junior servicemen some 9 months later; she had, shesaid, celebrated her discharge from the Spinal Unit by goingrafting with the service apprentices the same day down the

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

380River Severn! No new autonomic disturbances have beenreported in this group.

FOLLOW UP: CONTROL CASESThe control group were all followed up for a period betweenone and 3 years. All patients walked without any aids. Powerdeficits could be elicited on formal testing in some and fivepatients had increased tone and spasticity. All patients hadpermanently abnormal deep tendon or superficial reflexesand/or extensor plantar responses. Residual sensory dis-turbance in one modality (blunting to pin prick) was seen infour. Intractible pain was a feature in three cases. Persistentdisturbance of bladder bowel and sensory function were seen

in the majority of cases. Of the 15 adult males, five statedthat their sexual function (sensation, erection and ejacu-lation) is entirely normal. None has admitted to siring chil-dren as yet. Despite their disability, only three patients havenot returned to previous occupations.

Discussion

Our numbers would confirm that non organic para-plegia is rare. When comparison is made with theoverall admission rate, there is an excess of females.The features in the history of particular value were

the difficulties in reconciling the severity of the paraly-sis with the vagueness of the initially presented inci-dent and the absence of appropriate associatedtrauma. The findings on physical examination thatwere helpful have all been reported previously' - 9 andno new features were found. Of greatest value were

preservation of normal tone and reflexes and flexorplantar responses despite apparent weakness. Thiswas accompanied by bizarre posturing and simulta-neous contraction of the agonist and antagonist mus-cles. Within the sensory examination, variable,nonanatomical and inconsistent sensory man-

ifestations was of next value. The most difficult phys-ical finding to interpret was inconsistency between thedemonstrated signs and observed function or behav-iour, especially in the motor and autonomic fields,since this was dependent on the knowledge and skillof the examining doctor. The non organic patients didnot just have one anomalous feature alone; usually atleast two were present in each system examined.Owing to the subjective nature of sensory testing, sen-

sory findings were not as marked as those in themotor system. Care has to be taken that repeatedexaminations, especially by different or inexperienceddoctors, does not reinforce disability.9-1The hallmarks of the control group were either

preservation of some sensory modality or recovery ofsensation preceding motor recovery. The sensorycomponent had the anatomical features consistentwith spinal cord injury, that is, dissociation betweenthe different columns and sacral sparing. Difficultiesalso arise where there is definite evidence of neu-rological disease but disproportionate disabil-

Baker, Silverity. 12 - 17 We have therefore kept all such cases out ofthe series.We evaluated the role of investigations in assess-

ment, especially as new investigations have appearedsince the war series of non organic paraplegia. Plainfilms are essential. The majority of non organiclesions have normal films and abnormalities if foundare of long standing and not associated with trauma.The occasional case where an abnormality whichmight be relevant is found needs further investigation.Where strong clinical evidence for a non organiclesion existed, the investigation was negative. In con-trast, the control series all had an obvious abnormal-ity in either the bones, ligaments or other soft tissues,eg prevertebral soft tissue swelling. Spinal evokedresponses in the three cases we have performed havebeen conclusive insofar as they have been normal, andthe parallel EMG studies have shown agonist/antagonist contractions. CSF examination has beenunhelpful but no tests have been performed sinceassays of oligoclonal bands have become available.Myelography gave no new information in any casewhere it was undertaken. The indications for the stud-ies performed outside our units was unclear (otherthan as part of routine assessment). In the two casesin which we undertook myelography, while justifiablebecause of potentially sinister bony abnormalities andrecurrent autonomic disturbance, we would todayhave resorted to CT scanning and urodynamic stud-ies. It must remain a reserve tool for cases where theonly support for a non organic diagnosis is a lack ofpositive abnormal neurological findings or where sus-picion is raised because of abnormal investigations orevents on follow up in units where otherinvestigations are not available. However, the pro-cedure is not without hazard: patients have developedneurological signs subsequently and adhesive arach-noiditis may occur; this would be disastrous if noorganic lesion were present. Indeed, we have seenseven such cases which will be reported separately.

It has been argued that a non organic diagnosisshould have positive features in the psychiatric orpsychological history or examination and we wouldsupport this. All the non organic patients had at leasttwo such features in their histories. The features wefound useful, in descending order, were self dischargeby walking out, previous functional dis-turbance, - 3 7 9 1l 18 19 a model for paralysis in eithertheir own or their family/friend's histories, and thepresence of secondary gain.7 10 13 17 20 21 The first ofthese is pathognemonic. Previous functional dis-turbance occurred in 60% of cases and in just under50% involved the central nervous system. Thirty percent had a "model" on which to base their illness.Both males and females had evidence of secondarygain but this may not be significant where the gain is

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

Hysterical paraplegia

a civil action, for negligence, because of a similar highincidence in organic paraplegia. In our control series,however, only four civil actions were pending. Theassociation with crime (other than motoring offences)is higher with an overrepresentation among criminalsand their victims (5 out of 13 male cases). In the con-

trols, there were only two motoring offences,sufficiently minor to be disposed of by post.The association with active matrimonial pro-

ceedings in women is significant; but the overalldivorce rate and matrimonial disharmony is no

higher between groups and may be no higher than inthe community at large. We may have been fortunatein our controls as there were no problems extant (andnone arose during treatment) from marital dishar-mony. The association between non organic paraple-gia and "health" occupations in women is previouslyunreported. Although the NHS is the largestemployer in the country, 5 out of 23 is a dis-proportionate representation. No such associationwas seen in the control group, even when both sexes

were considered. Although there is a 30% incidenceof gynaecological procedures in women22 and 25% ofour patients had been psychiatric inpatients at some

stage,'9 these aspects may not be helpful and could bemisleading, even though their incidence was higherthan in our control group where only two psychiatricpast histories were obtained (one of which was longinactive and in the other the psychiatric conditioninduced the cord injury).We found any relationship with intelligence, the

presence of "belle indifference" or circumscribedmemory deficit, and subjective comments by doctorson the quality of the patient's history totallyunhelpful.3 10 11 23 - 25 Sexual history was unhelpful;many patients in both groups had problems withpotency of libido, albeit transient, before injury. Fewpatients, in either group, reported impotence as a fea-ture. The recovery in incomplete lesions also laggedbehind improvement in other autonomic parameters.The response to treatment was rapid, with com-

plete recovery from symptoms, usually within days,and improvement commenced within hours. The lat-ter can also be seen in organic lesions, some patientsproduced complete functional recovery in similartimes, but overall, the non organic recovered faster.The pattern of recovery (global improvement simulta-neously occurring in motor, sensory and autonomicsystems) in the non organic group was different fromthat in the organic group. In the organic group sen-

sory recovery appeared to occur before motor. Fur-ther, there was always some residual objective sign ofneurological disturbance in organic lesions, and a

majority had some persistent autonomic disturbance,although the pattern was always one of improvement(not deterioration) during follow up.

381

The most successful management of the nonorganic lesion seemed to be that of a firm diagnosis, aconfident prediction of improvement and "sympathy,interest and common sense" advocated by BarhamCarter. We have found that showing patients theirnormal spinal evoked responses will assist in recov-ery. Any florid theatrical performances should beignored. Patients should be seen, if possible, as out-patients rather than as inpatients. If admitted (as isusually the case where the referring hospital has nosuspicion of a non organic lesion), once the suspicionhas been raised, the patient should be nursed in a sideward away from ward routine, which may induceacquired behaviour. They must not be forgotten anda firm directed course of exercises seekingimprovement with each session imposed.Of the series, excepting those with major

psychiatric disease, only four went on to develop fur-ther functional disturbances through two indulged inattention seeking behaviour. In these patientspsychiatric intervention may be needed sub-sequently.7 15 19 In the control group, without pre-existing psychiatric disease, psychiatric interventionwas never required and in the one pre-existingpsychiatric case, these features became better and notworse. It has not been our practice to routinely seek apsychiatric opinion in non organic cases.

Over two thirds of the non organic group have beenseen or heard of on follow up, and none have beenfound to harbour organic neurological disease. Wefeel, therefore, that a diagnosis of non organic para-plegia can be arrived at using clinical criteria coupledwith simple investigations. Positive abnormalities willbe found on physical examination in addition to amere absence of conventional neurological abnormalsigns, as will evidence of psychological features. Morethan one abnormal feature will be found. The diagno-sis involves both neurological and psychiatric crite-ria.4 26 - 29Difficulties will occur if attempts are madeto use only one or the other.'2 14 30-32 With suchfeatures, the diagnosis can be confidently made33 andany residual doubt can be removed by evokedresponses34 and urodynamic studies and, rarely, CTscanning. Myelography is not required for assess-ment, today and above all not as a routine to excludeneurological disease. Those with organic lesions willhave abnormal neurological signs and abnormal plainradiological findings. Without markers for a nonorganic lesion and without psychological features, thediagnosis of a functional disorder is incorrect.

References

1 Carden NL, Schamel DJ. Observations of concersionreactions seen in troops involved in Vietnam conflict.Am J Psychiatry 1966;123:21-3 1.

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from

3822 Ziegler D. Neurological disease and hysteria: the

differential diagnosis. International J Neuropsychiatry1967;3:388-96.

3 Green MA. Neurological manifestations of conversionhysteria. Trans Am Neurol Assoc 1955;86: 196-8.

4 Ziegler FJ. Hysterical conversion reactions. PostgradMed 1970;47:174-8.

5 Weiser HI. Motor sensory dysfunction of the upper limbdue to conversion syndrome. Arch Phys Med Rehab1976;57: 17-9.

6 Rogoff JB, Chrystal M. Hysterical paralysis diagnosedas poliomyelitis. Physical Therapy Reviews1952:331-7.

7 Friedman SB. Conversion symptoms in adolescents.Pediatr Clin North Am 1973;20:873-82.

8 Worden RE, Johnson EW, Burk RD. Diagnosis of hys-terical paralysis. Arch Phys Med Rehab 196 1;42: 122-3.

9 Lightburn JL, Cattell RB, Stephenson WF. Differentialdiagnosis ofconversion reactions in a general hospital.Postgrad Med 1958;23:140-7.

10 O'Connor PJ. Hysterical reactions. Practitioner1973;210:58-64.

11 Stevens H. Conversion hysteria: a neurologic emergency.Mayo Clin Proc 1968;43:54-64.

12 Althaus J. On Epilepsy, Hysteria and Ataxy. London:John Churchill and Sons, 1866.

13 Slater E. Shorvon Memorial Lecture. Diagnosis of "hys-teria". Br Med J 1965;1395-9.

14 Slater E. The 35th Maudsley Lecture. "Hysteria 311".J Men Sci 1961;107:359.

15 Skey FC. Hysteria. London: Longmans & Co, 1867.16 Babinski J. Definition de l'hysterie. Soc Neurol Paris. 7

Novembre 1901.17 Gowers WR. A Manual of Diseases of the Nervous Sys-

tem. I & II. London: J & A Churchill, 1886-88.18 Stevens H. Conversion hysteria: a therapeutic oppor-

tunity for the neurologist. Trans Am Neurol Assoc1971;96:313-4.

Baker, Silver19 Kirschner LA, Kaplan N. Conversion as a manifestation

of crisis in the life situation. Compr Psychiatry1970;11:260-88.

20 Meyer-Gross W, Slater E, Roth M. Clinical Psychiatry.3 ed London: Cassell & Co, 1969.

21 Rabkin R. Conversion hysteria as social maladaption.Psychiatry 1964;27:349-63.

22 Cohen ME, Robins E, Purtell JJ, Altman MW. Excessivesurgery in hysteria: a study of surgical procedures in50 women with hysteria and 190 controls. JAMA1953;151:977-86.

23 Chodoff P, Lyons H. Hysteria, the hysterical personalityand hysterical conversion. Am J Psychiatry1958;114:734-40.

24 Ljungberg L. Hysteria. Acta Psych Scand 1957;Suppl112.

25 Aring CD. Observations on multiple sclerosis and con-version hysteria. Brain 1965;88:663-74.

26 Merskey H. The Analysis of Hysteria. London: BailiereTindall, 1979.

27 Head H, Riddoch G. In: Head H, ed. Studies in Neu-rology. London: Hodder, 1920:466-530. Originallypublished in: Brain 1917;XL:188-263.

28 Head H. An address on the diagnosis of hysteria. Br MedJ 1922;i:827-9.

29 Arietti S. Adult Clinical Psychiatry. American Handbookof Psychiatry. 2nd ed Vol 3. New York: Basic Books,1974:166-8.

30 Sydenham T. In: Veith I, ed. Hysteria: a History of theDisease. Chicago: Chicago University Press, 1965.

31 Wilson SAK. Neurology 1940;1:156.32 Sebastianpillai FJY, Wijesinghe CP. Hysterical paraly-

sis. Ceylon Medical Journal 1972;7:75-9.33 Barham Carter A. The prognosis of certain hysterical

symptoms. Br Med J 1949;1: 1253-4.34 Levy R, Mushin J. Somatosensory evoked response in

patients with hysterical anaesthesia. J Psychosom Res1973;17:81-4.

Protected by copyright.

on February 4, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.50.4.375 on 1 April 1987. D

ownloaded from