comparison of mouse bioassay and immunoprecipitation assay for botulinum toxin antibodies

Comparison of mouse bioassay andimmunoprecipitation assay for botulinum toxinantibodies

Philip A Hanna, Joseph Jankovic, Angela Vincent

AbstractObjective—To compare a recently devel-oped immunoprecipitation assay (IPA) tothe mouse protection bioassay (MPB),currently considered the “gold standard”,for detecting antibodies against botuli-num toxin A (BTX-A) and to correlatethese assay results with clinical responsesto BTX-A injections.Methods—MPB and IPA assays were per-formed on serum samples from 83 pa-tients (38 non-responders, 45 responders)who received BTX-A injections. Six non-responders had serum tested on two sepa-rate occasions. Some patients alsoreceived a “test” injection into either theright eyebrow (n=29) or right frontalis(n=19).Results—All patients antibody positive(Ab+) by MPB were also Ab+ by IPA,whereas an additional 19 patients (17 withreduced or no clinical response) who wereMPB Ab− were Ab+, with low titres, byIPA. Two of these 19 patients (non-responders) were initially MPB Ab− butlater became MPB Ab+. Similar to previ-ous studies, the sensitivity for the MPBwas low; 50% for clinical, 38% for eyebrow,and 30% for frontalis responses whereasthe IPA sensitivity was much higher at84% for clinical (p<0.001), 77% for eye-brow (p=0.111, NS) and 90% for frontalisresponses (p<0.02). The IPA specificitywas 89% for clinical, 81% for eyebrow, and89% for frontalis responses, whereas theMPB specificity was 100% for all threeresponse types, which were all non-significant diVerences.Conclusions—Both assays had high spe-cificity although the sensitivity of the IPAwas higher than the MPB. In addition, theIPA seems to display positivity earlierthan the MPB, and as such, it mayprognosticate future non-responsiveness.Eyebrow and frontalis “test” injectionscorrelated well with clinical and immuno-logical results and are useful in the assess-ment of BTX non-responders.(J Neurol Neurosurg Psychiatry 1999;66:612–616)

Keywords: dystonia; botulinum toxin; antibodies; im-munoresistance

An ever increasing number of disordersincluding dystonia, tremors, tics, hemifacialspasm, spasticity, sphincter dyssynergia, andachalasia are now being treated eVectively with

botulinum toxin type A (BTX-A).1–10 Inaddition, BTX-A is also being used for variousnon-neurological (for example, cosmetic) indi-cations. As the range of uses for BTX-Acontinues to expand, there is a growingconcern regarding the development of immu-noresistance secondary to blocking antibodies(Ab).2 11 The reported frequency of suchantibodies has ranged from 3% to 57%depending on the assay method used.12 13 Thestandard assay for detecting BTX Ab is the invivo mouse protection bioassay (MPB), whichevaluates the ability of increasing dilutions of apatient’s serum to protect mice from lethaldoses of BTX-A.14 In vitro assays, including thesphere linked immunodiagnostic assay(SLIDA),13 enzyme linked immunosorbentassay (ELISA),15 16 a monoclonal antibodybased immunoassay,17 and western blottechnique18 have also been reported to detectsuch antibodies. These assays, however, do notcorrelate well with clinical responses becausethey do not detect specific blocking Ab.

The MPB has been shown to have high spe-cificity, but its sensitivity is relatively low.18 Theprimary aim of this study was to compare theMPB with a more recent immunoprecipitationassay (IPA) developed by Palace et al19 and tocorrelate the presence of antibodies detectedby these two assays to the patients’ clinicalresponse to BTX-A injections. The resultsdescribed by Palace et al19 needed to beconfirmed using a larger number of patients, aswell as incorporating more clinical detailsincluding correlation with facial (eyebrow andfrontalis) “test” injections. Additionally, weevaluated the utility of eyebrow or frontalisinjections18 as clinical “tests” for immunoresist-ance.

MethodsEighty three patients (17 men and 66 women)with a mean age of 56 (SD 12.2) years: range19 to 81) were selected for this study. Most ofthe patients were treated primarily for dystonia;cervical (n=62; 32 non-responders), cranialand cervical (n=10; four non-responders), andcranial (n=7, all responders). Other conditionsincluded spastic hemiplegia (n=1; responder),hemifacial spasm (n=1; responder), focal legdystonia (n=1; non-responder), and segmentalmyoclonus (n=1; non-responder). Clinicalresponse to BTX-A (Botox®, Allergan Phar-maceuticals, Irvine, CA, USA) injections wasrated on a 0 to 4 “peak eVect” scale (0=noeVect; 1=mild eVect, no functional improve-ment; 2=moderate improvement, no change in

J Neurol Neurosurg Psychiatry 1999;66:612–616612

Parkinson’s DiseaseCenter and MovementDisorders Clinic,Department ofNeurology, BaylorCollege of Medicine,Houston, Texas, USAP A HannaJ Jankovic

Neurosciences Group,Department of ClinicalNeurology, Institute ofMolecular Medicine,John RadcliVeHospital, Oxford UKA Vincent

Correspondence to:Dr Joseph Jankovic,Department of Neurology,Director of Parkinson’sDisease Center, andMovement Disorders Clinic,Baylor College of Medicine,6550 Fannin St No 1801,Houston, Texas 77030, USA.Telephone 001 713 7985998; fax 001 713 798 6808.

Received 27 August 1998and in revised form16 November 1998Accepted 26 November 1998

group.bmj.com on July 16, 2011 - Published by jnnp.bmj.comDownloaded from

http://jnnp.bmj.com/

http://group.bmj.com/

functional disability; 3=moderate change inseverity and function; 4=marked improvementin severity and function).20 There were 38 non-responders (0 or 1 “peak eVect” responserating after their last injection), and 45responders randomly chosen from the botuli-num toxin clinic population. Six patients (allnon-responders) had samples drawn on twooccasions (with a minimal latency betweensample collections of 4 months; mean 4.3months). Thus, there were 89 total serum sam-ples on 83 patients included in this study. Thislow ratio of responders/non-responders doesnot represent the actual patient response rate inthe BTX clinic as we did not collect samples onall patients seen in the clinic.

The blood collected was separated and sentto Northview Pacific Laboratories, Berkeley,California, USA for the MPB assay and to theInstitute of Molecular Medicine, John Rad-cliVe Hospital, Oxford University, UK for theIPA assay. The individual laboratories were“blinded” to the clinical information beforesample testing to maintain objectivity.

The MPB is a qualitative test reported aseither positive (Ab+) indicating that thepatient’s serum neutralises the eVects ofBTX-A injected intraperitoneally with survivalof 3/4 mice.14 In a negative result, two or moreof the mice die, presumably indicating the lackof blocking Ab in the patient’s serum.

The IPA method was performed as de-scribed by Palace et al19 with slight modifica-tions. After the iodination reaction,19 the125I-BTX was microfiltered and stored at 4°C inphosphate buVered saline (PBS). When re-quired, it was diluted in PBS and centrifuged toremove any aggregates immediately before use.Supernatant (25 µl) containing 30 000–50 000cpm, was incubated with 2.5 µl of each serumin a total volume of 50 µl PTX buVer PBS(0.02M phosphate, pH 7.4 m 0.1% tri-ton×100). After 2 hours at room temperatureor overnight at 4°C excess goat antihuman Igwas added. When a precipitate had formed,600 µl PTX was added before centrifuging.The pellets were washed twice briefly in PTXand counted on a Cobra Packard gammacounter. Results were expressed as pM (pmolesof 125I-BTX precipitated/l serum) after subtrac-tion of the mean results (<1000 cpm) fromhealthy control serum samples run in parallel.

Twenty nine patients were also injected with atest dose of 15 units (n=26) to 20 units (n=3) ofBTX-A into the medial aspect of their right eye-brow (one site), and 19 patients received 15units (in two divided doses) into the frontalismuscle on the right side.18 For the eyebrowinjections, a positive (good) response is indi-cated by the presence of asymmetry on frowningafter unilateral BTX-A injections, whereas anegative (no) response (immunoresistance) ispresent when there is no asymmetry with frown-ing. A positive response to unilateral frontalisinjections is strongly suggested by asymmetry ofeyebrow elevation or forehead wrinkling on rais-ing of eyebrows whereas a symmetric contrac-tion of forehead muscles indicates a negativeresponse (immunoresistance).18

Sensitivity, specificity, and positive predictivevalue of the two assays was determined as fol-lows:

Sensitivity=A/(A+C); specificity=D/(D+B);positive predictive value=(PPV) A/(A+B);negative predictive value=(NPV): D/(D+C)

where A=true positive (Ab+ with negativeresponse to injection), B=false positive (Ab+with positive response to injection), C=falsenegative (Ab− with negative response to injec-tion), D=true negative (Ab− with positiveresponse to injection).

Comparisons of the above parameters of thetwo assays were performed using the Fisher’sexact test.

ResultsThe distribution of results of the first sampleson the 83 patients is shown in fig 1. Thethreshold for positivity, 50 pM of 125I-BTXbinding sites precipitated/l of serum, was lowerthan that reported previously19 due to slightimprovements in the assay that reducednon-specific precipitation by control serumsamples.

There was a clear correlation between theresults of the IPA and MPB assays (fig 2). Allserum samples which were Ab+ by MPB wereAb+ by IPA, and all Ab− samples by IPA wereAb−by MPB. However, 20 serum samples(from 19 patients) were Ab−by MPB but Ab+by IPA. The antibody titres in this group, witha mean of 183.2 pM (SD 111.8): range 51 to459 pM) were, however, significantly lower(p<0.0001, Kruskal-Wallis test) than those inthe MPB Ab+ group, in which the mean was1378.1 (SD 921.5): range 101 to 3663 pM). Ofthe 19 IPA Ab+/MPB Ab− patients, 14 werenon-responders and two of these non-responders became Ab+ by MPB on repeattesting as shown in figure 2. The remaining fivewere considered false positive as they contin-

Figure 1 Frequency distribution of IPA results (pM of125I-BTX precipitated /l serum) on the initial samples fromthe 83 patients, divided on the basis of clinical response.Results from healthy control serum samples were subtractedfrom all test values.

40

30

20

10

0

–25–

0

1–25

26–5

0

51–1

00

101–

200

201–

400

401–

800

801–

1600

1601

–320

0

IPA (pM)

Clinical responseReduced or no clinical response

Ob

serv

atio

ns

(n)

Mouse bioassay versus immunoprecipitation assay for botulinum toxin antibodies 613




ued to respond to BTX-A despite low, butpositive, titres (112–353). Three of these fivepatients had a reduced, peak eVect score 2,response (fig 3).

In a previous report, we showed that lack ofresponse to a test injection into the facial mus-cles is a more sensitive measure of non-responsiveness than the MPB.18 In the presentstudy, 29 and 19 patients respectively weregiven eyebrow or frontalis “test” injections, andthe IPA titres corresponded well with responsesto the facial “test” injections. Four patientsshowed no response to the eyebrow testinjections despite continuing clinical response.However, three of these patients were border-line (reduced) clinical responders (peak eVectscore 2), who previously had a more robustresponse to BTX-A, and two of these patientswere IPA Ab+ suggesting that the eyebrow and

IPA may both be early predictors of immunore-sistance.

Of the 10 clinical non-responders who alsohad eyebrow injections, only one had a goodeyebrow response. This patient was MPB Ab−but IPA Ab+ (titre of 409 pM). Seven patientswho were responders had a frontalis injection,and all seven had a good frontalis response. Ofthe 12 patients who were clinical non-responders and who received a frontalisinjection, two had a good frontalis response.Both patients were MPB Ab− whereas one wasIPA Ab+ (with a low titre, 82 pM) (see fig 4 forcorrelation of clinical responses with responsesto “test” injections).

The specificity of both assays was relativelyhigh, although the sensitivity of the IPA wassubstantially higher than the MPB (tables 1and 2). Specificity of the MPB was 100% on allthree parameters (clinical, eyebrow, and fronta-lis) whereas the IPA specificity was 89% forclinical (p=0.056, NS, Fisher’s exact test), 81%for eyebrow (p=0.226, NS), and 89% for fron-talis responses (p=0.99, NS). Sensitivity for theMPB was low; 50% for clinical, 38% foreyebrow and 30% for frontalis whereas the IPAsensitivity was much higher at 84% for clinical(p<0.001), 77% for eyebrow (p=0.111, NS)and 90% for frontalis responses (p<0.02).

The PPV of the MPB was 100% for clinical,eyebrow, and frontalis responses, whereas theNPV was 67% for clinical responses, 66% foreyebrow, and 56% for frontalis responses. ThePPV of the IPA was 88% for clinical, 77% foreyebrow, and 90% for frontalis responses,whereas the NPV was 85% for clinical, 81% foreyebrow, and 89% for frontalis responses.

Sensitivity, specificity, PPV, and NPV of theindividual test injections were determined inrelation to clinical responses. False positives inthis determination were a positive test injectionresponse with a negative clinical response.False negatives were a negative test responsewith a positive clinical response. Thus, for theeyebrow injections, sensitivity was 79%, spe-cificity was 90%, PPV was 94%, and NPV was69%. For the frontalis injections, sensitivitywas 100%, specificity was 83%, PPV was 78%,and NPV was 100%. For the test injectionscombined, sensitivity was 85%, specificity was86%, PPV was 88%, and NPV was 83%.

Figure 2 IPA results on the 89 samples (from 83 patients)depicted on the basis of whether the sample tested negative(open circles) or positive (shaded triangles) by MPB. Thelines join the values from the two patients whose MPB Abstatus changed. IPA titres >50 pM are considered positive.

4000

3000

2000

1000

50

25

0

–25Ab– Ab+

MPB result

IPA

(p

M)

Figure 3 IPA results separated on the basis of the patient’sclinical response to BTX injections. Grade 0 or 1 responseindicates non-responders (shaded triangles), grade 2response (open triangles) indicates reduced response, andgrade 3 or 4 (open circles) are responders.

4000

3000

2000

1000

50

25

0

–253/4 2 0/1

Clinical response to BTX-A

IPA

(p

M)

Figure 4 Correlation of clinical response (grade 0 or 1response indicates non-responders, grade 2 responseindicates reduced response, and grade 3 or 4 are responders)with response to test injections.

20

0


0/1 2 3/4

No response to test injectionGood response to test injection

Ob

serv

atio

ns

(n)

10

614 Hanna, Jankovic, Vincent




DiscussionAs the number of patients treated with BTX-Acontinues to grow, the prevention and accuratedetection of immunoresistance have becomehigh priorities. The MPB, originally describedby Hatheway and Dang,14 has been consideredby many to be the “gold standard” assay for thedetection of BTX-A Ab. Here we show that anassay based on immunoprecipitation of radio-labelled BTX-A is a highly reliable test which isslightly less specific, but considerably moresensitive than the MPB. Six non-respondingpatients were tested twice by both assays, typi-cally secondary to patient request or for verifi-cation purposes. Two of these were initiallyMPB Ab− but became Ab+ by MPB on repeattesting; the IPA values were positive on firsttesting and the titres rose over the 4 monthsbetween the samples (fig 2) suggesting the earlydetection of immunoresistance by IPA. Fur-thermore, there were five false positives (clini-cal responders with Ab+ result by IPA), butthree of these patients have had decliningresponse to BTX as well as relatively low titres

by IPA, which is a quantitative test. Thus, posi-tivity by the IPA may be a useful predictor offuture non-responsiveness.

The IPA correlated well, not only with theoverall clinical responses, but also with the eye-brow and frontalis “test” injections, with a spe-cificity of 81% and 89% respectively to theseupper face injections. Additionally, the strongcorrelation of these “test” injections with clini-cal response ratings provides a strong supportfor using these simple biological tests to evalu-ate patients for immunoresistance. Overall, weprefer the eyebrow injections as these are morecosmetically acceptable in that the asymmetricresponses are present only during voluntarycontractions whereas unilateral disappearanceof frontal wrinkles may not be desirable.

The only commercially available in vitro testutilises a western blot assay. Although this testoVers potential advantages over MPB in that itis less cumbersome and does not require theuse of experimental animals, our previousstudy18 showed that this in vitro test does notcorrelate as well as the MPB with clinicalresponses.

Based on the results or our study, we oVerthe following guidelines for evaluation ofpatients who fail to respond to BTX injections(secondary non-responders) (fig 5). Whensuch a patient returns to the clinic after obtain-ing a poor or no response to the previous injec-tion, the clinician may re-inject with the sameor higher dose and/or an alteration of the siteand at the same time inject 15–20 units of BTXinto the right eyebrow or right frontalis. If thepatient responds to either the clinical (for theprimary condition—that is, dystonia) or testinjection, the clinician may continue injections,possibly adjusting the dose or site of injection.If the patient shows no response to both (clini-cal and test) injections, the use of serologicalassays, such as IPA or MPB may be considered,before preceding to the next step of using otherBTX serotypes,21–23 plasma exchange, immu-noadsorption, or surgery. Based on the results

Table 1 Clinical-immunological correlation

Response

Mouse bioassay (MPB) Immunoprecipitation assay (IPA)

Ab+ (n=22) Ab− (n=67) Ab+ (n=42) Ab− (n=47)

+ − + − + − + −

Clinical (n=83 subjects,89 samples) 0 22 45 22 5 37 40 7

Eyebrow (n=29 subjects) 0 5 16 8 3 10 13 3Frontalis (n=19 subjects) 0 3 9 7 1 9 8 1Total responses 0 30 70 37 9 56 61 11

+=Responder; −= non-responder.

Table 2 Mouse bioassay - immunoprecipitation assay comparison

Response

Mouse bioassay (MPB) Immunoprecipitation assay (IPA)

Sensitivity(%)

Specificity(%)

PPV/NPV(%)

Sensitivity(%)

Specif.icity(%)

PPV/NPV(%)

Clinical 50 100 100/67 84 89 88/85Eyebrow 38 100 100/66 77 81 77/81Frontalis 30 100 100/56 90 89 90/89

PPV=Positive predictive value; NPV=negative predictive value.

Figure 5 Decision tree for the evaluation and subsequent treatment of patients based on response to BTX injections.

Clinical response to BTX

+ (good) Response — (no) Response

1) Reinject; adjust dose and/or site2) Administer "test" injection (eyebrow or frontalis)3) Collect serum for assays

— (no) Response toboth clinical and testinjection (symmetriccontraction)

+ (good) Response to clinical and/or testinjection (asymmetriccontraction)

Reinject with the smallestdose needed to achieveoptimal response

IPA or MPB assay

+—

Other BTX serotypes, plasmaexchange and immuno-adsorption, or surgery

Reinject; adjustdose and/or site

Mouse bioassay versus immunoprecipitation assay for botulinum toxin antibodies 615




of our study, we recommend the IPA assay(given the high sensitivity and specificity) as theassay of choice to confirm immunoresistance.Eight of nine patients who were clinical andtest (eyebrow) non-responders were IPA Ab+,and nine of 10 patients who were clinical andfrontalis non-responders were IPA Ab+. As itcan be predicted with relative certainty that ifboth the clinical and test injections result in noresponse, the IPA will be positive, there may beno need to test for antibodies by the IPA in thiscategory of patients. Given the low sensitivityof the MPB, this assay has a limited value com-pared with the IPA. Furthermore, the IPA doesnot require the use of experimental animalsand it quantitatively assesses the degree ofimmunoresistance by providing antibody titreswhich can be measured serially.

It is important to recognise some possibleshortcomings of our study. Although the “0–4peak eVect” scale is an established method ofassessing response to BTX injections, it maynot always reliably diVerentiate respondersfrom non-responders. Patients were considerednon-responders if they described no eVect oronly mild eVect with no functional improve-ment from their most recent injection. Thesepatients may have had suboptimal benefit fromtheir recent injection secondary to technique,injection of inappropriate muscles, low potencyof the BTX batch, or inadequate dose, and assuch, the reported sensitivities of the two assaysmay be artificially low. A wide range of doseswas given per visit at diVerent intervals makinga correlative analysis diYcult. A furtherpossible shortcoming is the definition of sensi-tivity and specificity used. “True positive”assumed that the Ab+ patient must be anon-responder, which is supported by our pre-vious finding that all 20 MPB Ab+ patients hadno response to BTX-A injections on at leasttwo consecutive treatment sessions.20 “Falsenegatives” refer to those patients who do notrespond to BTX injections despite an Ab− test.

In conclusion, our study shows that bothassays have a high specificity, but because theIPA is more sensitive than the MPB and becausethe IPA is an in vitro assay, it may have relativeadvantage over the MPB. A further advantage ofthe IPA is that this is a quantitative assay whichmay be useful for serial evaluations and mayhave a predictive value in determining impend-ing or future unresponsiveness. Eyebrow andfrontalis “test” injections correlated well withthe clinical and immunological results and canbe used as reliable screening tests in patients

who have either no response or an equivocalresponse to BTX injections.

The study was supported by grants from Allergan Pharmaceuti-cals and the Medical Research Council of Great Britain. We hadcomplete control over the collection and analysis of the data.

1 Jankovic J, Brin MF. Therapeutic uses of botulinum toxin. NEngl J Med 1991;324:1186–94.

2 Jankovic J. Use of botulinum toxin in neurology. In:Kennard C, ed. Recent advances in clinical neurology. Vol 8.Edinburgh: Churchill Livingstone, 1995;89–110.

3 American Academy of Neurology. Assessment: the clinicalusefulness of botulinum toxin-A in treating neurologicaldisorders; report of the Therapeutics and TechnologyAssessment Subcommittee of the American Academy ofNeurology. Neurology 1990;40:1332–6.

4 American Academy of Ophthalmology. Botulinum toxintherapy of eye muscle disorders. Safety and eVectiveness.Ophthalmology 1989;96(suppl):37–41.

5 Davidson BJ, Ludlow CL. Long-term eVects of botulinumtoxin injections in spasmodic dysphonia. Ann Otol RhinolLaryngol 1996;105:33–42.

6 National Institutes of Health. Consensus development con-ference statement on clinical use of botulinum toxin. ArchNeurol 1991;48:1294–8.

7 Clarke CE. Therapeutic potential of botulinum toxin inneurologic disorders. Q J Med 1992;82:197–205.

8 Jankovic J, Schwartz K, Donovan DT. Botulinum toxintreatment of cranial-cervical dystonia, spasmodic dyspho-nia, other focal dystonias and hemifacial spasm. J NeurolNeurosurg Psychiatry 1990;53:633–9.

9 Berardelli A, Formica A, Mercuri B, et al. Botulinum toxintreatment in patients with focal dystonia and hemifacialspasm. A multicenter study of the Italian Movement Disor-der Group. Ital J Neurol Sci 1993;14:361–7.

10 Geller BD, Hallett M, Ravits J. Botulinum toxin therapy inhemifacial spasm: clinical and electrophysiological studies.Muscle Nerve 1989;12:716–22.

11 Scott AB. Foreword. In: Jankovic J, Hallett M, eds. Therapywith botulinum toxin. New York: Marcel Dekker, 1994:vii-viii.

12 Zuber M, Sebald M, Bathien N, et al. Botulinum antibodiesin dystonic patients treated with type A botulinum toxin:frequency and significance. Neurology 1993;43:1715–18.

13 Siatkowski RM, Tyutyunikov A, Biglan AW, et al. Serumantibody production to botulinum A toxin. Ophthalmology1993;100:1861–6.

14 Hatheway CH, Dang C. Immunogenicity of the neurotoxinsof Clostridium botulinum. In: Jankovic J, Hallett M, eds.Therapy with botulinum toxin. New York: Marcel Dekker,1994:93–107.

15 Notermans S, Dufrenne J, Van Schothorst M. Enzyme-linked immunosorbent assay for detection of Clostridiumbotulinum toxin type A. Japan J Med Sci Biol 1978;31:81–5.

16 Notermans S, Nagel J. Assays for botulinum and tetanustoxins. In: Simpson LL, eds Botulinum neurotoxin andtetanus toxin. San Diego: Academic Press, 1989;319–331.

17 Shone C, Wilton-Smith P, Appleton N, et al. Monoclonalantibody-based immunoassay for type A Clostridium botu-linum toxin is comparable to the mouse bioassay. ApplEnviron Microbiol 1985;50:63–7.

18 Hanna PA, Jankovic J. Mouse bioassay versus western blotassay for botulinum toxin antibodies: correlation with clini-cal response. Neurology 1998;50:1624–9.

19 Palace J, Nairne A, Hyman N, et al. A radioimmunoprecipi-tation assay for antibodies to botulinum A. Neurology 1998;50:1463–6.

20 Jankovic J, Schwartz, KS. Response and immunoresistanceto botulinum toxin injections. Neurology 1995;45:1743–6.

21 Eleopra R, Tugnoli V, Rossetto O, et al. Botulinumneurotoxin serotype C: a novel eVective botulinum toxintherapy in human. Neurosci Lett 1997;224:91–4.

22 Houser MK, Sheean GL, Lees AJ. Further studies usinghigher doses of botulinum toxin type F for torticollis resist-ant to botulinum toxin type A. J Neurol Neurosurg Psychia-try 1998;64:577–80.

23 Lew MF, Adomato BT, Duanne DD, et al. Botulinum toxin B(BotB): a double-blind, placebo-controlled, safety andeYcacy study in cervical dystonia. Neurology 1997;49:701–7.

616 Hanna, Jankovic, Vincent




LETTERS TOTHE EDITOR

Cerebral metabolism during vegetativestate and after recovery toconsciousness

One way to approach the study of conscious-ness is to explore lesional cases in whichimpairment of consciousness is the promi-nent clinical sign. Vegetative state is such acondition wherein awareness is abolishedwhereas arousal persists. It can be diagnosedclinically soon after a brain injury and may bereversible (as in the following case report) orprogress to a persistent vegetative state ordeath. The distinction between vegetativestate and persistent vegetative state is that thesecond is defined as a vegetative state that hascontinued or endured for at least 1 month.1

We present a patient who developed a vegeta-tive state after carbon monoxide poisoningand in whom we had the opportunity tomeasure brain glucose metabolism distribu-tion during the vegetative state and afterrecovery to consciousness. Using [18F]fluoro-deoxyglucose (FDG) PET and statisticalparametric mapping (SPM) we comparedboth patient’s sets to a normal control popu-lation. Our findings oVer an insight into theneural correlates of “awareness”, pointing toa critical role for posterior associative corticesin consciousness.

A 40 year old right handed womanattempted suicide through CO intoxicationand was found unconscious. She was treatedwith hyperbaric oxygen but evolved to a veg-etative state diagnosed according to thefollowing criteria:1 (1) spontaneous eye open-ing without evidence of awareness of theenvironment; (2) no evidence of reproduciblevoluntary behavioural responses to anystimuli; (3) no evidence of language compre-hension or expression; (4) intermittent wake-fulness and behaviourally assessed sleep-wake cycles; (5) normal cardiorespiratoryfunction and blood pressure control; (6) pre-served pupillary, oculocephalic, corneal, andvestibulo-ocular reflexes. Brain MRI per-formed 14 days after admission was normal.Electroencephalography showed a 6 Hz basalactivity with more pronounced slowing on theleft parietal regions. Auditory evoked poten-tials were normal. Somaesthetic evokedpotentials of the median nerve showed

normal latency and amplitude of P14 andN20 potentials without any late cortical com-ponents. After remaining in a vegetative statefor 19 days the patient regained conscious-ness. Her sequelae consisted of a bilateralspastic paresis of upper and lower limbs.Neuropsychological testing 1 month afteradmission showed an attention deficit withmoderate impairment of short term memory.One year after the accident she showed aspastic gait with altered fine motor function,most prominent on the right, a slurredspeech, and minor short term memorydisturbances. FDG-PET was performed dur-ing the vegetative state (day 15 after admis-sion) and after recovery to consciousness(day 37).

The control population consisted of 48drug free, healthy volunteers, aged from 18 to76 years (mean: 42 (SD 21) years).

The study was approved by the ethicscommittee of the University of Liège. In-formed consent was obtained by the husbandof the patient and for all control subjects. Fiveto 10 mCi FDG was injected intravenously;PET data were obtained on a Siemens CTI951 R 16/31 scanner in bidimensional mode.Arterial blood samples were drawn during thewhole procedure and cerebral metabolic glu-cose rates (CMRGlu) were calculated for allsubjects. PET data were analysed using SPMsoftware (SPM96 version; Welcome Depart-ment of Cognitive Neurology, Institute ofNeurology, London, UK).2 The use of SPMto assess between subject (rather than withinsubject) variability is unlikely to alter the rel-evance of our results given their high degreeof significance. Data from each subject werenormalised to a standard stereotactic spaceand then smoothed with a 16 mm full widthhalf maximum isotropic kernel. The analysisidentified brain regions where glucose me-tabolism was significantly lower in eachpatient scan compared with the controlgroup. The resulting foci were characterisedin terms of peak height over the entire volumeanalysed at a threshold of corrected p<0.05.2

During the vegetative state, average greymatter glucose metabolism was 38% lowerthan in controls (4.5 v 7.3 (SD 1.4) mg/100g/min). No substantial change in meanCMRGlu was found after recovery (4.7mg/100 g/min). During the vegetative state,significant regional CMRGlu decreases werefound in the left and right superior parietallobule; the left inferior parietal lobule; theprecuneus; the left superior occipital, supe-rior and middle temporal gyri; and thepremotor and postcentral and precentral cor-tex (figure, yellow colour). After recovery,metabolic impairment was confined to the

left and right precentral and postcentral gyriand premotor cortices (figure, blue colour).

This case report oVers an insight into theneural correlates of human consciousness (atleast, external awareness as it can be assessedat the patient’s bedside). Given that globalglucose utilisation levels remained essentiallythe same, the recovery of consciousnessseems related to a modification of theregional distribution of brain function ratherthan to the global resumption of cerebralmetabolism. The main decreases in metabo-lism seen during the vegetative state but notafter recovery were found in parietal areas,including the precuneus. This is in agreementwith postmortem findings in persistent veg-etative state, in which involvement of theassociation cortices is reported as a criticalneuroanatomical substrate3 and with PETstudies in postanoxic syndrome, in which theparieto-occipital cortex showed the mostconsistent impairment.4 The functions ofthese areas are manifold: lateral parietal areasare involved in spatial perception and atten-tion, working memory, mental imagery, andlanguage, whereas the precuneus is activatedin episodic memory retrieval, modulation ofvisual perception by mental imagery, andattention.2 Our data point to a critical role forthese posterior associative cortices in theemergence of conscious experience.

STEVEN LAUREYSCHRISTIAN LEMAIRE

PIERRE MAQUETCyclotron Research Centre, University of Liège, Sart

Tilman, 4000 Liège, Belgium

CHRISTOPHE PHILLIPSInstitute of Cognitive Neurology, University College

London, Alexandra House, 17 Queens Square, LondonWC1N3AR, England, UK

GEORGE FRANCKDepartment of Neurology, CHU Liège Sart, Tilman

B-35, 4000 Liège, Belgium

Correspondence to: Dr Pierre Maquet, CyclotronResearch Centre (B30), University of Liège, SartTilman, 4000 Liège, Belgium Telephone 0032 4366 36 87; fax 0032 43 66 29 46; [email protected]

1 The Multi-Society Task Force on PVS. Medicalaspects of the persistent vegetative state (1). NEngl J Med 1994;330:1499–508.

2 Frackowiak RSJ, Friston KJ, Frith CD, et al.Human brain function. San Diego: AcademicPress, 1997.

3 Kinney HC, Samuels MA. Neuropathology ofthe persistent vegetative state: a review. J Neu-ropathol Exp Neurol 1994;53:548–58.

4 De Volder AG, GoYnet AM, Bol A, et al. Brainglucose metabolism in postanoxic syndrome.Positron emission tomographic study. ArchNeurol 1990;47:197–204.

Localisation of voxels in which cerebral glucose metabolism was impaired during vegetative state (in yellow) and after recovery to consciousness (in blue),compared with the control population. SPM{Z} threshold was set at voxel level corrected p<0.05 and projected on the patient’s coregistered MRI,normalised to the stereotaxic space of Talairach.

J Neurol Neurosurg Psychiatry 1999;67:121–133 121




Electrical inexcitability of nerves andmuscles in severe infantile spinalmuscular atrophy

Spinal muscular atrophy (SMA) is one of themost common fatal autosomal recessivedisorders, characterised by progressive de-generation of anterior horn cells. Before theadvent of genetic testing, the diagnosis ofSMA was based on clinical, histopathologi-cal, and electrophysiological features. In1992, the International SMA Consortiumdefined diagnostic criteria of proximal SMAbased on clinical findings.1 In SMA type I(severe; Werdnig-HoVmann disease), af-fected persons have onset of symptoms before6 months of age and are never able to sitwithout support. Electromyography demon-strates denervation features. In early 1995,the candidate gene, the survival motorneuron (SMN) gene, was identified, makingthe confirmation of SMA by DNA analysispossible.2

With the availability of a genetic test forSMA, many investigators are refining thediagnostic criteria published by the Consor-tium. Studies involving hundreds of patientswith proximal SMA have disclosed a subset ofpatients who fulfill at least one exclusion cri-terion defined by the Consortium.3 We iden-tified an infant with severe SMA who fulfilledtwo exclusion criteria and also showedinexcitability of all nerves as well as muscles.This report will further delineate the widerange of phenotypes for this particular genemutation.

A 2945 g male infant was born at term.First fetal movements were noted at 13 weeksof gestation. Chorionic villus sampling at 10weeks of gestation disclosed normal chromo-somes. Decreased fetal movement and poly-hydramnios were noted at about 34 weeks ofgestation. At delivery, the infant was cyanoticwith no respiratory eVort and was subse-quently intubated. On physical examination,the infant had no spontaneous movements.He opened his eyes with brief fixation but nofollowing. Tongue fasciculations werepresent. Other cranial nerves seemed intact.Mild flexion contractures of both elbows,knees, and ankles were noted. Tone was flac-cid in both upper and lower limbs, and therewas no movement response to painful stimu-lus. Deep tendon reflexes were absent.

Brain MRI disclosed mild diVuse corticaland deep atrophy. His EMG was severelyabnormal, with widespread fibrillations andabsent voluntary motor units except in thegenioglossus, where mildly neurogenic motorunits with decreased recruitment were seen.Stimulation of the median, ulnar, tibial, andperoneal nerves with a maximal stimulusresulted in no clinical or electrical response.The biceps brachii and rectus femorismuscles were electrically inexcitable by directneedle stimulation. Median, ulnar, and suralsensory potentials were not obtainable. DNAtesting showed a homozygous deletion ofexons 7 and 8 in the telomeric SMN gene,confirming the diagnosis of SMA. The infantexpired at 3 weeks of age, and the parentsdeclined postmortem examination.

Typical EMG studies in those with SMAshow fibrillations and fasciculations at restand an increased mean duration and ampli-tude of motor units. Motor nerve conductionvelocities may be slowed but are usually nor-mal. Korinthenberg et al reported inexcitabil-ity of motor nerves in three siblings, each ofwhom died from SMA before 1 month ofage.4 In addition to a homozygous deletion of

exons 7 and 8 of the telomeric SMN gene, allthree siblings showed a large deletion in theregion that includes all alleles of the multi-copy markers Ag1-CA and C212, localised atthe 5' end of the two SMN gene copies. It hasbeen postulated that the severity of diseasemay be correlated to the extent of a deletioninvolving the SMN gene and the multicopymarkers.3–5 The infant in our report withSMA type I showed electrical inexcitability ofmotor nerves as well as the characteristicalteration of the SMN gene.

Although it has been known for some timefrom histological studies that sensory systemsare involved in SMA, electrophysiologicalsensory findings have been previously re-ported only once.4 Sensory nerve conductionvelocity was tested in an infant with severeSMA and showed no recordable potential,but the infant in our report also exhibiteduniversal absence of sensory potentials. Inboth cases, DNA analysis disclosed the 5qdeletion. It is unclear whether this findingrepresents a distinct entity or merely thesevere end of classic Werdnig-HoVmann dis-ease. The diagnostic criteria produced by theInternational SMA Consortium currentlylists “abnormal sensory nerve action poten-tials” as an exclusion criterion.1 Our findingof absent sensory potentials in a 5q deletionestablished case of SMA indicates furtherneed for revision of the Consortium criteria.

Studies involving large series of patientswith SMA have identified cases of SMAvariants.3 These patients were diagnosed asinfantile SMA by the presence of proximalweakness and atrophy, hypotonia, and evi-dence of neurogenic alterations in EMG andmuscle biopsy. In addition, these patients alsoexhibited one of the exclusion criteria definedby the Consortium—for example, diaphrag-matic weakness, involvement of the CNS, orarthrogryposis. Although these patients didnot show the typical SMN deletion and weretherefore probably not linked to chromosome5q, they could have had point mutations. Theinfant in our report showed no respiratoryeVort after birth, indicating diaphragmaticweakness. He did, however, possess the char-acteristic SMN gene alterations. This findingsuggests that diaphragmatic weakness shouldbe reconsidered as an exclusion criterion bythe Consortium.

Review of the literature disclosed no previ-ous reports of electrically inexcitable musclesin SMA. This phenomenon is known to occurin a few other neuromuscular conditions suchas periodic paralysis and critical illnesspolyneuropathy. Fibrillations, as seen in theinfant in our report, are commonly seen inacute denervation and are thought to becaused by perturbation of the sarcolemmalmembrane, rendering it unstable. One possi-bility may be that the severe denervation inSMA type I can result in abnormal functionof the membrane to make it electrically inex-citable. Further electrophysiological studiesat the cellular level are required to delineatethis interesting finding.

ALICE A KUODepartment of Pediatrics

STEFAN-M PULSTDAWN S ELIASHIV

CAMERON R ADAMSDivision of Neurophysiology, Cedars-Sinai Medical

Center, Los Angeles, CA, USA

Correspondence to: Dr Cameron R Adams, De-partment of Neurophysiology, Cedars-Sinai Medi-cal Center, 8631 West Third Street, Room 1145,East Tower, Los Angeles, CA 90048, USA.

1 The International SMA consortium. Meetingreport. Neuromusc Disord 1992;2:423–8.

2 Lefebvre S, Burglen L, Reboullet S, et al. Identi-fication and characterization of spinal muscu-lar atrophy-determining gene. Cell 1995;80:155–65.

3 Zerres K, Wirth B, and Rudnik-Schoneborn S.Spinal muscular atrophy-clinical and geneticcorrelations. Neuromusc Disord 1997;7:202–7.

4 Korinthenberg R, Sauer M, Ketelsen UP, et al.Congenital axonal neuropathy caused by dele-tions in the spinal muscular atrophy region.Ann Neurol 1997;42:364–8.

5 Wirth B, Hahnen E, Morgan K, et al. Allelicassociation and deletions in autosomal reces-sive proximal spinal muscular atrophy: associ-ation of marker genotype with disease severityand candidate cDNAs. Hum Mol Gen 1995;4:1273–84.

Acute overdosage and intoxication withcarbidopa/levodopa can be detected inthe subacute stage by measurement of3-o-methyldopa

Although the eVects of a chronic overdosagewith levodopa are well known, few cases ofacute intoxication have been described.1 2 Aparticular problem in establishing a diagnosisof levodopa overdosage is the relatively shorthalf life in the circulation of levodopa.3 4 Ifthere is a delay in bringing an acutely intoxi-cated patient to hospital, perhaps due to latediscovery, the blood concentration of levo-dopa could already be normal (correspond-ing to the peak levodopa concentration inParkinson’s disease therapy) after 6–8 hours.Depending on the extent of the overdosage,the time could be even shorter. This reportdescribes the clinical eVects and the plasmaconcentrations of levodopa and specificmetabolites over a period of 132.5 hours afteringestion of 30 tablets of carbidopa/levodopa(50 mg/200 mg tablets).

A 76 year old patient had a pre-existingmild akinetic rigid Parkinson’s syndrome,which had been treated for the past 1.5 yearswith 3×1 tablets of carbidopa/levodopa (50mg/200 mg) a day without a substantialresponse. The weight of the patient was 74kg. A known chronic obstructive airwaydisease was treated with a home oxygenappliance. At about 8.30 pm, the patient hadattempted suicide by taking 30 tablets ofcarbidopa/levodopa. About 9.00 pm he ap-peared psychically altered, crying withoutreason, anxious, and depressed. After about30 minutes he was increasingly inadequate,agitated, and subeuphoric, and was experi-encing visual hallucinations; he was restless,tossing and turning, and getting out of bed.He did not represent peak dose dyskinesia orother extrapyramidal clinical features. At10.00 pm he showed bilaterally maximallydilated pupils. The muscle stretch reflexeswere lively, there were no pyramidal tractsigns, and he did not show any signs ofParkinson’s syndrome or dyskinesia. Arterialhypertonus and sinus tachycardia could beregistered.

After an empty box of Striaton (carbidopa/levodopa, 50 mg/200 mg) was found in thepatient’s flat, 1 g of carbon was given bystomach tube after gastric lavage. Cranial CTwas carried out before the diagnosis ofintoxication had been made; it showed a pro-nounced subcortical arteriosclerotic en-cephalopathy with reduced brain volume.The patient was moved to the medical inten-sive care unit and observed for 24 hours. TheECG showed a P pulmonale, but no otherunusual features. Echocardiography showednormal right and left ventricular functionwith suspicion of right ventricular hypertro-

122 Letters, Correspondence, Book reviews, Correction




phy. Blood was taken for the measurement oflevodopa and metabolites 2.5 hours afteringestion, and again after 9, 14.5, 34, and132.5 hours. Using high performance liquidchromatography with electrochemical detec-tion, we measured the plasma concentrationsof levodopa, 3-o-methyldopa, dihydroxyphe-nylacetic acid, homovanillic acid, noradren-aline, adrenaline, and dopamine. The timecourse of the concentrations of levodopa and3-o-methyldopa are shown in the figure.

After 24 hours the patient was moved fromthe intensive care unit to a normal medicalward. At this point no neuropsychiatric signsof levodopa intoxication could be detected.Clinically, the most prominent symtoms oflevodopa overdosage are confusion, agitation,sleeplessness, and excessive motor activity.The initial levodopa concentration in ourpatient was 66 763 ng/ml, the concentrationsof DOPAC, homovanillic acid, noradren-aline, adrenaline, and dopamine were raised2.5 hours after ingestion and rapidly returnedto normal. A very noticeable feature of thiscase was the maximal bilateral mydriasis, withabsent light reaction, at the time of the maxi-mal intoxication with a 30-fold increase inplasma levodopa concentration. To ourknowledge, this association of a levodopaintoxication with maximally dilated, lightunresponsive pupils and without signs of dys-kinesia has not previously been reported. Itcan be assumed that the eVect is caused bythe peripheral conversion of levodopa intonoradrenaline, which stimulates á-adrenergicreceptors in the dilatator iridis. There is noindication from animal experiments of a spe-cific activation of dopamine receptors.5 Thearterial hypertonus measured initially canalso be attributed to the high systemicconcentrations of noradrenaline, and thetachycardia to the raised concentrations ofadrenaline and dopamine. As seen in the fig-ure, the only indicator which can show alevodopa intoxication in the subacute stage isthe concentration of 3-o-methyldopa. Themetabolite 3-o-methyldopa results from theo-methylation of levodopa, which explains thedelayed peak of the 3-o-methyldopa concen-tration. The half-life of 3-o-methyldopa inplasma was calculated at 16.7 hours in thispatient. On the other hand, the plasma halflife of levodopa was 111 minutes; this isslightly longer than normal, and can beexplained by assuming a rate limited metabo-lism of levodopa when the substrate concen-tration for the enzymes metabolising it israised.

Distribution into muscles rather thenmetabolism may largely determine theplasma half life of levodopa and explain whythis was only slightly altered with overdose.The measured peak concentration of 66 763ng/ml is about 30 times higher than the peakconcentration to be expected after taking onetablet of carbidopa/levodopa (50 mg/200mg). It is apparent that the 30 tablets did notinterfere with absorption or lead to a gastro-intestinal paralysis due to the high dose oflevodopa; the relation between amount in-gested and plasma concentration seems to belinear, at least in this dose range.

We conclude from these findings that incases of suspected levodopa intoxicationsome hours previously, it could be importantto measure the concentration of 3-o-methyldopa, so as not to overlook anoverdosage with levodopa, which may be dueto a suicide attempt. In addition to the diag-nostic uncertainty in relation to the immedi-ate treatment of the patient, this would alsohave an eVect on further psychiatric and psy-chological therapy.

H J STUERENBURGB G H SCHOSER

Neurological Department, University HospitalHamburg-Eppendorf, Hamburg, Germany

Correspondence to: Dr Hans Joerg Stuerenburg,Neurological Department, Universitètskranken-haus Eppendorf Martinistrasse 52, 20246 Ham-burg, Germany. Telephone 0049 40 4717 4832; fax0049 40 4717 5086; email [email protected]

1 Hoehn MM, Rutledge CO. Acute overdose withlevodopa. Clinical and biochemical conse-quences. Neurology 1975;25:792–794.

2 Sporer KA. Carbidopa-levodopa overdose. AmJ Emerg Med 1991;9:47–8.

3 Contin M, Riva R, Martinelli P, et al. Longitudi-nal monitoring of the levodopa concentration-eVect relationship in Parkinson’s disease. Neu-rology 1994;44:1287–92.

4 Bredberg E, Lennernas H, Paalzow L. Pharma-cokinetics of levodopa and carbidopa in ratsfollowing diVerent routes of administration.Pharm Res 1994;11:549–55.

5 Korczyn AD, Keren O. The eVect of L-dopa onpupillary diameter in mice. Experimentia 1982;38:481–2.

The use of olanzapine for movementdisorder in Huntington’s disease: a firstcase report

Movement disorder is a prominent feature ofHuntington’s disease and consists of involun-tary and voluntary components as well asassociated bradykinesia. Pharmacologicaltreatment is problematic because of the sideeVects of the drugs used, which may furthercompromise cognitive functioning and mo-bility. Patients are often not subjectivelyaware of their movements but can be consid-erably disabled by them and carers are oftendistressed and enquire about treatment op-tions. If drug treatment is considered it isimportant to achieve the maximum improve-ment in movements with the minimum ofnegative side eVects. This paper describes theeVect of olanzapine on movements whenother treatment options had been ineVectiveor limited by side eVects.

Huntington’s disease is a hereditary pro-gressive neurodegenerative disorder. It con-sists of a triad of symptoms comprisingmotor, psychological, and cognitive abnor-malities. The motor component consists ofinvoluntary choreiform movements and in-creasing diYculties with voluntary move-ment. The degree of the involuntary move-ments is variable but in some patients can be

very marked. Progression over time of themovement disorder in Huntington’s diseasecan be monitored using the quantitativeneurological examination (QNE). This meas-ure has three subscales, an eye movementscale, a motor impairment scale (MIS) quan-tifying voluntary movement, and a choreascale measuring involuntary movement.1 2

Pharmacological control of the symptomshas been shown to be eVective with dopamineantagonists,3 4 but their use is limited becauseof the side eVects. Clinically the mostproblematic of these are sedation, cognitiveslowing, increased mobility problems, andhypotension. The inability of traditionaldopamine antagonists to improve functionalcapacity, despite ameliorating chorea, is pos-sibly due to suppression of voluntary motoractivity.4 5 Tardive dyskinesia has occasionallybeen reported in patients with Huntington’sdisease treated with these drugs.5 6 The atypi-cal antipsychotic clozapine has been shown tobe eVective in improving the movementdisorder. However, in a double blind ran-domised trial of clozapine which includedpatients who were already receiving tra-ditional antipsychotic medication and agroup who had not received drug treatmentsfor their movement disorder, chorea wasreduced in those who were antipsychoticnaive only and the authors concluded thatclozapine was of little additional benefit inHuntington’s disease.7 Olanzapine is a newatypical antipsychotic drug. It is a thienod-ibenzodiazepine structurally very similar toclozapine. Unlike clozapine it is not associ-ated with the potentially serious side eVect ofagranulocytosis and therefore frequent bloodmonitoring is not necessary.

This report describes the progress of a manwho has Huntington’s disease. He developeda marked movement disorder and was unableto tolerate both sulpiride and risperidone buthad symptomatic improvement when treatedwith olanzapine.

He is a man in his early 50s who had a con-firmatory genetic test for Huntington’s dis-ease in 1994, after the development ofclinically obvious motor symptoms. It is likelythat the onset of symptoms had occurred afew years previously as he had experienceddiYculties in concentration and attention atwork, attributed at the time to stress, leadingto the loss of employment. In addition hisfamily, watching family videos of a few yearsearlier, thought that there was evidence ofearly signs of his movement disorder. How-ever there was no known family history ofHuntington’s disease which might have led toan earlier diagnosis. By May 1995 hisinvoluntary movements were becoming morenoticeable, although control of voluntarymovement was good. A trial of sulpiridecommencing at 200 mg twice daily andincreasing over 1 week to 800 mg daily wasundertaken with a subsequent decrease in thefrequency and extent of involuntary move-ment recorded in case notes; unfortunatelythe QNE was not repeated at this time. How-ever, the patient experienced a subjectiveslowing of his cognitive processes, concur-rently became depressed, and decided to stopthe treatment within 3 weeks. Paroxetine, aselective serotonin reuptake inhibitor antide-pressant, was started at a dose of 20 mg a day,which led to an improvement in his lowmood. His involuntary movements continuedto cause diYculties in his daily living. He wasunable to sit comfortably in a chair and whenout he felt that he was disturbing others byknocking into them. He agreed to a trial of

Time course of the measured concentrations oflevodopa (half filled circles) and3-o-methyldopa (full circles) after an overdose oflevodopa (6000 mg). The peak of 3-OMDappears after a delay, and the concentrationenfalls very slowly, corresponding to a half life inplasma of 16.7 hours.

70 000

60 000

50 000

40 000

30 000

20 000

10 000

0100 140120806040

Time (h)

ng

/ml

200

Letters, Correspondence, Book reviews, Correction 123




risperidone. This was started at a dose of 1mgtwice daily, increasing to a dose of 1mg fourtimes a day over a period of 2 weeks, stoppedafter a brief period. He developed hypoten-sion (blood pressure 100/60 mg Hg), com-plaining of dizziness after the initial dose. Hisblood pressure remained stable, althoughlow, after this and as there was improvementin his movements the drug was continued.However, he decided to stop the risperidoneafter 4 months because of his subjectiveexperience of slowed thinking and occasionaldizziness. A repeated trial of sulpiride wascarried out in March 1997. Sulpiride wasstarted at a dose of 200 mg twice a day andincreased to a total daily dose of 1000 mgover 2 weeks. He was on sulpiride for 4 weekswith no improvement in his movements,so itwas discontinued. The patient continued toexperience low mood and after the discon-tinuation of sulpiride, his antidepressant drugwas changed to lofepramine commencing at70 mg once a day and increasing after a fewdays to 140 mg daily. There were no changesnoted in his movements during this change.

Although the patient was subjectively una-ware of the extent of his movements his eve-ryday life continued to be aVected. The socialvenues he felt able to attend were becomingmore limited and activities he wanted to pur-sue such as travelling abroad by air wereproblematic. A trial of olanzapine was theninstituted. He was started on 5 mg a day inthe morning. There was a marked improve-ment in his involuntary movements within 1week but once again he experienced slowedthinking. However, adjusting the time ofmedication to the evening led to an improve-ment in this. Six months later the improve-ment in his involuntary movements is main-tained. Serial quantitative neurologicalexamination scores are illustrated in figure 1.

In the absence of a cure for Huntington’sdisease, it is very important that any interven-

tions considered enhance the quality of life ofthe patient and improve overall functioning.It may not always be in the best interests ofthe patient to use drug treatments for themovement disorder. In those patients whohave severe movements, however, a trial oftreatment may be appropriate and continuedif a clear benefit has been achieved. Neuro-logical monitoring and the patient’s own per-ception of the eVect of the drug must be takeninto account.

The mechanism by which olanzapine mayhave beneficial eVects is unclear. Olanzapinehas been shown to have high aYnity for alarge number of receptors including D1, D2,D4, 5HT2A, 5HT2C, 5 HT3, á-1-adrenergic, histamine H1, and 5 muscarinicreceptors. This binding profile is similar toclozapine, another atypical antipsychoticdrug, but substantially diVerent to theconventional antipsychotic haloperidol.7

Preferential loss of D2 projection neuronswhich are involved in a feedback loopnormally active in the suppression of involun-tary movements is thought to be the patho-physiological basis of chorea in patients withHuntington’s disease.8 The D2 antagonistproperties of olanzapine may explain its pos-sible benefits in the improvement of chorea.However, the eVect at other receptors such asD4 may also be important, as D4 receptordensity has been shown to be raised in Hunt-ington’s disease, therefore the D4/D2 ratio ofactivity may also be relevant. DiVerences inbinding profile across a range of receptorsmay explain clinical diVerences in outcomewhen comparing diVerent antipsychoticdrugs.

This case report indicates that olanzapinemay be a useful addition to the treatments formovement disorder, for some patients, andcontrolled trials of its use in Huntington’sdisease would be welcome.

HEATHER C DIPPLEDepartment of Psychiatry, Leicestershire Mental

Health Service Trust, Huntington’s Disease Service,Mill Lodge, Mill Lane, Kegworth, Derby DE74 2EJ,

UK. Telephone 0044 1509 670774.

1 Folstein SE, Jensen B, Leigh RJ, et al. Themeasurement of abnormal movements: meth-ods developed for Huntington’s disease. Neu-robehav Toxicol Teratol 1983;5:605–9.

2 Folstein S. Huntington’s disease: a disorder offamilies. Baltimore: John Hopkins UniversityPress 1989.

3 Quinn N, Marsden CD. A double blind trial ofsulpiride in Huntington’s disease and tardivedyskinesia. J Neurol Neurosurg Psychiatry 1984;47:844–7.

4 Girotti F, Carella F, Scigliano G, et al. EVect ofneuroleptic treatment on involuntary move-ments and motor performances in Hunting-ton’s Disease. J Neurol Neurosurg Psychiatry1984;47:848–52.

5 Schoulson I. Care of patients and families withHuntington’s disease. In: Marsden CD, Fahn,eds.Movement disorders. London: Buttersworth,1982:277–90.

6 Schott K, Ried S, Stevens I, et al. Neurolepti-cally induced dystonia in Huntington’s disease:a case report. Eur Neurol 1989;29:39–40.

7 Bymaster FP,Calligaro DO, Falcone JF, et al.Radiorecptor binding profile of the atypicalantipsychotic olanzapine. Neuropsychopharma-cology 1996;14:87–96.

8 van Vugt JP, Siesling S, Vergeer M, et al. Cloza-pine versus placebo in Huntington’s disease: adouble blind study. J Neurol Neurosurg Psychia-try 1997;63:35–9.

Transient hiccups after posteroventralpallidotomy for Parkinson’s disease

Hiccup is defined as an abrupt intermittent,involuntary, contraction of the diaphragmaticand external (inspiratory) intercostal mus-cles, with inhibition of expiratory intercostalactivity. This results in a sudden inspiration,abruptly opposed by closure of the glottis.1

Hiccup may result from various structural orfunctional disorders of the medulla, the aVer-ent or eVerent nerves to the respiratory mus-cles, and the gastrointestinal tract.2 3 NewsonDavis performed a study of hiccup with elec-trophysiological techniques and concludedthat hiccup is served by a supraspinal mech-anism distinct from that generating rhythmicbreathing.3 The principal site of interaction ofthe hiccup discharge with other descendingdrives to the respiratory motoneuron is at thespinal level. Neurogenic hiccup is particularlyassociated with structural lesions of themedulla oblongata.

Since 1994 we have performed 66 palli-dotomies for Parkinson’s disease in 60patients. So far, we have seen transienthiccups in seven patients after the operation(table). Our target coordinates for the poster-oventral globus pallidus at the border of themedial and lateral segments are 2–3 mmanterior to the midcommissural point, 5 mmbelow the intercommisural line, and 22 mmlateral to the midline of the third ventricle.Ventriculography was performed for target

Quantitative neurological examination scores showing the progress of the movement disorder. 06/95:before trial sulpiride, no medication; 05/96: before risperidone, 20 mg paroxetine daily; 07/96: 1 mgrisperidone four times daily and 20 mg paroxetine daily; 03/97: before retrial sulpiride, 20 mgparoxetine daily; 04/97: 400 mg sulpiride in the mornings 600 mg at night and 20 mg paroxetine;04/97: before olanzapine, 140 mg lofepramine daily; 06/97: 5 mg olanzapine at night, 140 mglofepramine daily.

14

12

10

4

2

0

Sco

re

06/95

6

8

MISChorea

05/96 07/96 03/97 04/97 04/97 06/97

Patient characteristics

PatientAge atsurgery Sex

YearswithPD

H and Y*stagingOn/oV

UPDRS oV†prepallidotomy/postpallidotomy

Pallidotomyside Transient side eVects Medication additional to levodopa

1 66 M 8 2/5 57/NP‡ R Slight facial paresis, swallowingproblems, drooling

Tryptizol, temazepam, alprazolam,apomorphine

2 43 F 7 2/2.5 22/7 L Slight dysarthria Trihexifenidyl3 49 M 15 2/3 55/15 L Facial paresis Pergolide, amantadine4 50 M 12 2/2 45/22 L Slight dysarthria Selegeline, biperideen5 53 M 14 2.5/4 69/36 R Facial paresis, hypophonia Pergolide, selegeline6 55 M 13 2.5/3 48/27 L Facial paresis, aphasia Selegeline, biperideen7 61 F 15 2.5/4 55/NP R Clozapine, temazepam, cisapride

*H and Y=Hoehn and Yahr; †UPDRS oV=unified Parkinson’s disease rating scale part 3 (motor examination), in a standardised oV state, 12 hours without antipar-kinson medication; ‡NP=not performed.





localisation. Patients started with a shortschedule of corticosteroids (5 days) the nightbefore surgery.

The hiccups started immediately after theoperation or the next day, were intermittent,and the bouts of hiccup of six patients, with aduration of hours, resolved within 3 days afterthe procedure. One patient complained ofyawning more often and frequent bouts ofhiccup for 6 months.

Five patients were men. All patients wereright handed. The mean age at surgery was54 years and the mean duration of Parkin-son’s disease was 12 years. All patients weretaking levodopa. In four patients the hiccupsappeared after a left sided pallidotomy.Patient 2 had a right sided thalamotomy 4years before the pallidotomy. Patient 5underwent a left sided pallidotomy 10months before the right sided pallidotomywhich caused the hiccups. The pallidotomiesimproved parkinsonism in the “oV” state(table), contralateral dyskinesias, and painaccompanying Parkinson’s disease. Six pa-tients had transient adverse events: fourpatients had a transient facial paresis postop-eratively and two a slight transient dysarthria(table). Two patients had choreatic move-ments after the pallidotomy at the contralat-eral side which resolved spontaneously within2 hours and is associated with a favourablesurgical outcome.4

Postoperative MR scans were obtained inthe first six patients, and showed that in fivepatients the lesions were located in the poste-rior part of the globus pallidus pars externa(GPe) and interna (figure). In patient 5 thelesion was situated slightly more anterior inthe GPe and putamen. In patient 3 there wasa small separate lesion more dorsal, probablyan infarct.

We never encountered hiccups in 150 otherstereotactic procedures for Parkinson’s dis-ease, such as thalamotomies or deep brainstimulation electrode implantation in thethalamus and therefore it is unlikely thatmedication or positive contrast medium ven-triculography with Iohexol evoked the hic-cups.

A possible cause for the transient hiccupscould be the lesion in the ventral medial seg-

ment of the globus pallidus or pressure, dueto oedema, on an adjacent structure like theinternal capsule or putamen. We could notfind other reports of hiccups as an adverseevent after functional stereotactic surgicalinterventions, nor after lesions of otheraetiology involving the striatium.5 Based onour experience we hypothesise that the globuspallidus or a neighbouring structure may beinvolved in a supramedullary system involvedin triggering hiccups.

R M A DE BIEJ D SPEELMAN

Department of Neurology

P R SCHUURMAND A BOSCH

Department of Neurosurgery, Academic MedicalCenter, University of Amsterdam, The Netherlands

Correspondence to: Dr R M A de Bie, Departmentof Neurology, Academic Medical Center, PO Box22700, 1100 DE Amsterdam, The Netherlands.Telephone 0031 20 566 3856; fax 0031 20 6791438; email [email protected]

1 Newson Davis J. An experimental study ofhiccup. Brain 1970;93:851–72.

2 Howard RS. Persistent hiccups. British MedicalJournal 1992;305:1237–8.

3 Newson Davis J. Pathological interoseptiveresponses in respiratory muscles and the mech-anism of hiccup. In: Desmedt J, ed. New devel-opments in electromyography and clinical neuro-physiology. Vol 3. Basel: Karger, 1973:751–60.

4 Merrello M, Cammarota A, Betti O, et al. Invol-untary movements during thermolesion pre-dict a better outcome after microelectrodeguided posteroventral pallidotomy. J NeurolNeurosurg Psychiatry 1997;63:210–13

5 Bathia KP, Marsden CD. The behavioral andmotor consequences of focal lesions of thebasal ganglia in man. Brain 1994;117:859–76.

Psychological adjustment and selfreported coping in stroke survivors withand without emotionalism

Emotionalism after stroke is common, occur-ring in 10%–20% of a community sample.1

Psychological factors in its cause or mainte-nance have not been studied; research hastended to concentrate instead on location ofthe stroke lesion. We suspect that one reasonfor this neglect of psychological aspects of

emotionalism is that most people do notmake a distinction between emotionalism,and pathological crying and laughing. As aresult all disorders of emotionality afterstroke are stereotyped as being related tobrain damage and therefore psychologicallymeaningless.

None the less, many patients with emotion-alism describe their crying as provoked byemotionally congruent experiences, whichmakes the tearfulness seem understandable.1

In two previous studies1 2 we have shown thatstroke patients with emotionalism have moresymptoms of psychological disorder than dopatients without emotionalism. In the presentstudy, we explored further the psychologicalcharacteristics of stroke patients with emo-tionalism. Our aim was to determine whetherthey diVered from patients without emotion-alism in their psychological reactions tostroke, or in the coping strategies theyreported.

Post-traumatic stress disorder is also char-acterised by recurrent episodes of intrusiveand uncontrollable emotion, and we weretherefore interested in whether patients withemotionalism also experienced the intrusivethoughts typical of post-traumatic stressdisorder. Because emotionalism is oftendescribed as uncontrollable, we were inter-ested in the possibility that patients weremore generally helpless, passive, or avoidantin their responses to stroke. Again, because ofthe reported uncontrollability of emotional-ism, we postulated that patients with emo-tionalism would report a more external locusof control3 than those without emotionalism

Participants were adults admitted to localgeneral hospitals after stroke, and were inter-viewed within 1 month of admission. Exclu-sions were due to poor physical health, cogni-tive impairment, communication diYculties,or lack of consent. Approval for the study wasobtained from the local research ethics com-mittees.

All participants completed a standardisedmeasure of distress—the general health ques-tionnaire, GHQ-124; a widely used measureof intrusive thoughts of the sort encounteredin post-traumatic stress disorder—the impactof events rating scale5; a measure of cognitivecoping—the mental adjustment to strokescale (O’Rourke S, Dennis M, MacHale S,Slattery J. The development of the mentaladjustment to stroke scale: reliability, patientoutcome and associations with mood andsocial activity, manuscript in preparation);and a measure of beliefs about responsibilityfor recovery from illness—the recovery locusof control scale.3 All the measures are selfreport questionnaires.

A total of 177 stroke patients werescreened, of whom 112 were excluded. The65 participants (29 men, 36 women) had amean age of 71.8 years (range 43 to 88 years).Nineteen (29.2%) patients met our criterionfor emotionalism,1 a rate similar to that foundin other studies. Their scores on the studymeasures are compared with the scores ofpatients without emotionalism in the table.

It might be that these associations withemotionalism were accounted for by thegreater general levels of distress experiencedby those with emotionalism. We thereforeundertook analysis of covariance with GHQ-12 and presence of emotionalism as the cov-ariates, and each of the other test items inturn as the independent variable. The resultsshowed an association, after adjustment forGHQ-12 score, between emotionalism andthe impact of events subscales intrusion

Five months after left sided pallidotomy, MRI of patient 6: (A) transversal slice at the level of theanterior commisure and (B) 6 mm more ventral.





(F=15.33, p<0.001), and avoidance(F=11.84, p=0.001); the mental adjustmentto stroke scale subscales helplessness/hopelessness (F=11.7 1, p=0.00 1) and anx-ious preoccupation (F=8.05, p=0.006). Theassociations with fatalism (F=14.79,p=0.052) and avoidance (F=0.06, p=0.80)on the mental adjustment to stroke scale wereno longer significant after adjustment forGHQ-12 score.

This study confirms earlier work by show-ing that stroke survivors with emotionalismhave more other mood symptoms (here ratedby the GHQ-12) than do those without emo-tionalism. It goes further however, in showingthat they also have intrusive thoughts abouttheir stroke, of a sort similar to those reportedby people with post-traumatic stress disorder.This unpleasant remembering is probablyresponsible for their higher ratings on anxiouspreoccupation. It is compatible with ourfinding in a previous study2 that irritability isassociated with emotionalism, as irritability isa common response to threatening intrusivememories of the sort encountered in post-traumatic stress disorder. It may not be thatemotionalism is a direct manifestation ofpost-traumatic stress disorder, although thatcondition has been described after stroke,6

but the analogy raises the possibility that anabnormality in processing emotionally im-portant stimuli may be one of the causes ofemotionalism. If correct it suggests possibletreatment strategies along the lines of thoseused in post-traumatic stress disorder.

A corollary is our finding of increased feel-ings of helplessness and hopelessness, cou-pled with avoidance—at least as a cognitivecoping strategy reported on one of our meas-ures. Avoidant coping may perpetuate thesymptom of emotionalism, by preventinghabituation to the social stimuli whichprovoke it. Alternatively it may lead to areduction in social support, exacerbatingcoexistent mood disturbance. Thus, it may bethat avoidant coping is not an integral part ofemotionalism, but rather that it is animportant maintaining factor.

We predicted that patients with emotional-ism would have more “external” scores on thelocus of control measure, reflecting theirsense of lack of personal control over crying.They did not, perhaps because the emotionalexpression, although not apparently control-lable by internal resources, is none the lessperceived as having psychological meaning,so that responsibility for it it cannot readily bedevolved to others.

Our study used a relatively weak between-groups design, the number of patients wasnot large, and we cannot be sure that all con-founders were dealt with. None the less, ourresults suggest that future research into emo-tionalism could profitably concentrate notjust on seeking its biological correlates, but

should also explore the psychological factorswhich might contribute to its cause orcontinuation.

We thank those patients who participated in thestudy and the staV of local hospitals and the LeedsStroke Database for their invaluable help. We alsothank Dr Louise Dye for her statistical advice. Thisstudy was completed as part of work for the degreeof DClinPsychol at Leeds University (SE).

STEVEN ECCLESALLAN HOUSE

Division of Psychiatry and Behavioural Sciences inRelation to Medicine, University of Leeds, Leeds, UK

PETER KNAPPStroke Outcome Study, Research School of Medicine,

Leeds, UK

Correspondence to: Dr Allan House, Division ofPsychiatry and Behavioural Sciences in Relation toMedicine, University of Leeds, 15 Hyde Terrace,Leeds LS2 9LT, UK.

1 House A, Dennis M, Molneaux A, et al.Emotionalism after stroke BMJ 1989;298:991–4.

2 Calvert T, Knapp P, House A. Psychologicalassociations with emotionalism after stroke. JNeurol Neurosurg Psychiatry 1998;65:928–9

3 Partridge C, Johnston M. Perceived control ofrecovery from physical disability: measurementand prediction. Br J Clin Psychol 1989;28:5359.

4 Goldberg D. General health questionnaire(GHQ–12). In: Compendium of Health Psychol-ogy Assessments. Windsor: NFER-Nelson,1992.

5 Horowitz MJ, Wilner N, Alvarez W. Impact ofevents scale as a measure of subjective stress.Psychosom Med 1979;41:209–18.

6 Sembi S, Tarrier N, O’Neill P, et al. Does post-traumatic stress disorder occur after stroke: apreliminary study. International Journal of Geri-atric Psychiatry 1998;13:315–22.

Paraneoplastic stiV limb syndrome

StiV man syndrome (SMS) is a rare, severeprogressive motor disorder characterised bypainful spasms, symmetric axial musclerigidity, and uncontrollable contractionsleading to distorted posturing. The disorderhas been associated with the autoantigens,glutamic acid decarboxylase (GAD), andamphiphysin, which are cytoplasmic proteinsin neurons of the CNS. A large series ofpatients with SMS found that most haveautoantibodies against GAD,1 whereas am-phiphysin is presumably the predominantautoantigen in paraneoplastic SMS.2 Re-cently, Brown et al3 presented four patientswith a stiV leg syndrome marked by progres-sive rigidity and spasms of the lower extremi-ties. This group of patients tested negative foranti-GAD antibody by immunoprecipitationand demonstrated distinct electrophysiologi-cal features. By contrast, another reportdescribed two patients with stiV leg syndromewho tested positive for anti-GAD antibody.4

Finally, in presenting a group of 13 patients,Barker et al5 proposed that the nomenclature

“stiV limb syndrome” refers to the focal formof SMS when one or more distal limbs areinvolved; two of their patients were also anti-GAD antibody positive, but none were testedfor antibodies to amphiphysin or identified ashaving an underlying neoplasm. We present apatient clinically consistent with the stiV limbsyndrome who was found to have autoanti-body to GAD and breast cancer.

A 68 year old woman presented with a 1month history of painful spasms in her legs.Cramps were associated with tactile stimuliand emotional upset. Within weeks, inversionbegan at the left and then right ankle, makingambulation diYcult. Her medical history wassignificant for Graves’ disease treated withthyroidectomy and radiation therapy, andhyperlipidaemia. She was a chronic smoker.General examination was noteworthy forlymphadenopathy in the right axilla. Hermental status was worse during periods oflower extremity spasms, during which shebecame anxious, diaphoretic, and tachy-cardic. Cranial nerve and motor evaluationswere unremarkable, but assessment of the leftleg, due to painful spasms elicited by lighttouch, was diYcult. Inversion and plantar-flexion were essentially fixed at the left anklebut could be overcome on the right. Deeptendon reflexes were 3+ in the upper andlower extermities, with sustained clonus atthe right ankle. Sensory examination, withthe exception of hyperaesthesia in the distallower extremities, and coordination testingwere grossly normal. No hyperlordosis ormyoclonus was noted. Gait was limited dueto ankle posturing.

The laboratory evaluation was noteworthyfor a CSF with increased IgG indices (2.5,3.4; normal, 0.2–0.8) and oligoclonal bands(5, 5) but no pleocytosis. Serological testingfor anti-Hu, anti-Yo, and anti-Ri antibodieswas unremarkable, and the haemoglobinA1C was 6.6 (5.6–7.7)%. Skin biopsy at threesites on the patient’s leg showed diminishedepidermal nerve fibre density and terminalaxonal swelling distally, consistent with asmall fibre sensory neuropathy.6 The patientwould not tolerate EMG. Magnetic reso-nance images of the brain and the entire spi-nal cord were normal. Fine needle aspirationof a soft tissue right axillary mass showedmetastatic adenocarcinoma. On an open sur-gical procedure, infiltrating duct carcinomaof the breast was identified. Anti-GADautoantibodies were positive by immunocyto-chemical assay and immunoprecipitation, butantibodies to amphiphysin were not detectedby immunocytochemistry, immunoprecipita-tion, or western blotting (Dr P De Camilli,Yale University).

Ongoing therapy with clonazepam and atrial of oral dexamethasone did not improvethe lower extremity symptoms. The patient’sankle posturing continued a slow progressionto marked inversion, with spontaneous exten-sion of hallucis longus. The patient died 18months after symptom onset. Gross necropsyattributed the cause of death to aspirationpneumonia. Neuropathological evaluationshowed a grossly normal brain and spinalcord. Microscopically, the lumbar cord hadmild reactive gliosis in the anterior horns butno evidence of inflammation. Sections of thefrontal cortex, pons, and medulla showedmild diVuse reactive astrocytosis.

StiV man syndrome is increasingly recog-nised as a heterogeneous disorder.7 Othercase reports have documented patients with“focal” disease involving either lower3–5 orupper extremity posturing,8 which contrast

Comparison of stroke survivors with and without emotionalism, assessed in hospital 1 month afterstroke

No emotionalism(n=45) Mean (SD)

Emotionalism (n=19)Mean (SD)

GHQ- 12* 3.2 (2.4) 5.3 (3.5)Recovery locus of control scale 33.2 (3.3) 34.2 (3.7)Impact of events scale intrusion subscale** 2.9 (4.6) 9.2 (6.6)Impact of events scale avoidance subscale* 4.7 (4.6) 9.9 (6.1)MASS Fighting spirit subscale 49.1 (4.0) 48.8 (5.2)MASS Anxious preoccupation subscale** 22.2 (2.8) 25.2 (4.0)MASS Fatalism subscale* 20.0 (1.9) 21.3 (2.2)MASS Avoidance subscale 1.7 (0.8) 1.9 (0.8)MASS Helplessness/hopelessness subscale** 10.9 (2.5) 14.1 (3.5)

MASS = Mental adjustment to stroke scale.*p<0.05; **p<0.01, t tests.





with the “diVuse” axial and subsequentproximal muscle distribution of the classicdisorder.9 Our patient diVers from thosereported with stiV leg syndrome in that anoccult malignancy was present. Unfortu-nately, we were unable to obtain electrophysi-ological studies for comparison. The searchfor a paraneoplastic process was based on thefindings of axillary lymphadenopathy and anabnormal CSF. Our patient is only the secondreported patient with paraneoplastic SMSassociated with anti-GAD antibody; the otherhad upper limb rigidity in the setting of breastcancer and additionally mounted an immuneresponse to amphiphysin.8

Paraneoplastic processes can aVect anycomponent of the nervous system and, occa-sionally, multiple levels, as in the syndrome ofsensory neuronopathy-encephalomyelitis. Ourpatient’s findings were not entirely consistentwith criteria for classic SMS9 in that anapparent encephalopathy and a small fibreneuropathy were identified—for example,her dysautonomia (tachycardia and relativehypertension) during spasms may have beena manifestation of involvement of smallfibres. The role of autoantibodies in thepathogenesis of SMS and cancer isunclear.2 7 8 Via its probable function inendocytosis,10 amphiphysin has been postu-lated to play a part in the regulation ofgrowth factor internalisation; however, theabsence of an autoimmune response to thisautoantigen in our patient suggests thatother mechanisms of oncogenesis in SMSexist. Given anecdotal evidence of improve-ment in paraneoplastic SMS after treatingthe underlying malignancy,8 we suggest thatall patients with SMS, diVuse or focal, bescreened for occult cancer.

ISAAC E SILVERMANDepartment of Neurology, Johns Hopkins University,

Baltimore, USA

Correspondence to: Dr I E Silverman, Johns Hop-kins Hospital, Pathology 509, 600 North WolfeStreet, Baltimore, MD 21287, USA. Telephone 001

410 955 6626; fax 001 410 614 1008; [email protected]

1 Solimena M, Folli F, Aparisi R, et al. Autoanti-bodies to gaba-ergic neurons and pancreatic âcells in stiV-man syndrome. N Engl J Med1990;322:1555–60.

2 Folli F, Solimena M, Cofiell R, et al. Autoanti-bodies to a 128-kd synaptic protein in threewomen with the stiV-man syndrome and breastcancer. N Engl J Med 1993;328:546–51.

3 Brown P, Rothwell JC, Marsden CD. The stiVleg syndrome. J Neurol Neurosurg Psychiatry1997;62:31–7.

4 Saiz A, Graus F, Valldeoriola F, et al. StiV-legsyndrome: a focal form of stiV-man syndrome.Ann Neurol 1998;43:400–3.

5 Barker RA, Revesz T, Thom M, et al. Review of23 patients aVected by the stiV man syndrome:clinical subdivision into stiV trunk (man)syndrome, stiV limb syndrome, and progressiveencephalomyelitis with rigidity. J Neurol Neuro-surg Psychiatry 1998;65:633–40.

6 McCarthy BG, Hsieh S-T, Stocks A, et al. Cuta-neous innervation in sensory neuropathies:evaluation by skin biopsy. Neurology 1995;45:1848–55.

7 Kissel JT, Elble RJ. StiV-person syndrome: stiVopposition to a simple explanation. Neurology1998;51:11–14.

8 Rosin L, De Camilli P, Butler M, et al. Stiff-mansyndrome in a woman with breast cancer: anuncommon central nervous system paraneo-plastic syndrome. Neurology 1998;50:94–8.

9 Lorisch TR, Thorsteinsson G, Howard FM.StiV-man syndrome updated. Mayo Clin Proc1989;64:629–36.

10 David C, McPherson PS, Mundigl O, et al. Arole of amphiphysin in synaptic vesicle endocy-tosis suggested by its binding to dynamin innerve terminals. Proc Natl Acad Sci USA 1996;93:331–5.

Tetrodotoxin intoxication in a uraemicpatient

Tetrodotoxin intoxication results from ingest-ing puVer fish or other animals containing thetoxin. Clinical presentation is mainly acutemotor weakness and respiratory paralysis.Death is common in the worst aVectedvictims. Although the severity of the symp-toms generally depends on the amount oftoxin ingested, it may be influenced by the

victim’s medical condition, as described inthis report. The patient was a 52 year olduraemic woman. The uraemia was of unde-fined aetiology. Over the past 3 years she hasreceived regular haemodialysis. One day bothshe and her husband, a healthy 55 year oldman, ate a fish soup. About 4 hours after themeal she developed a headache and a lingualand circumoral tingling sensation and numb-ness at the distal parts of all four limbs. Shewas dizzy and unsteady, had diYculty inswallowing, and became very weak. She wastaken to the emergency service and wasplaced on machine assisted ventilation as res-piratory distress and cyanosis developed. Herhusband remained asymptomatic throughoutthis time.

The patient’s condition kept on deteriorat-ing, developing eventually into a comatous-like state with no spontaneous or reflexive eyeopening or limb movement within 30 min-utes of intubation. On neurological examina-tion, the pupillary light reflex was absent andoculocephalic manoeuvre elicited no ocularmovements. All four limbs were areflexic andBabinski’s signs were absent. Brain CT andlaboratory studies of arterial blood gas (underassisted ventilation), electrolytes, liver func-tion, blood glucose, and CSF study wereunremarkable. An examination of renal func-tion indicated chronic renal insuYciency withmild azotaemia (urea nitrogen 70 mg/dl, cre-atinine 9.1 mg/dl). An EEG, recorded 18hours after the onset of symptoms when theneurological condition was unchanged,showed posterior dominant alpha wavesintermixing with trains of short duration, dif-fuse theta waves. When brief noxious stimuliwere applied to the sternum, they werereplaced transiently by beta activities. Thefindings suggested that the profound neuro-logical dysfunction might be peripheral inorigin. The patient was given a course ofhaemodialysis according to the set schedulefor uraemia at 21 hours after onset of thesymptoms. Her condition improved dramati-

Changes in the symptoms of poisoning in relation to each course of haemodialysis. Scales in the vertical axis represent the arbitrary measurements ofseverity of each symptom; the numbers indicating day(s) after onset; ↓= haemodialysis).

Depressed tendon reflex

Papillary areflexia

Ophthalmoparesis andptosis

Bulbar weakness

Distal weakness

Proximal weakness

Respiratory paralysis

Orolingual numbness

Acroparaesthesia

1 2 3 4 5 6 7 8 9 10 11





cally within an hour. She could open her eyesand she communicated and answered ques-tions correctly by blinking. Pupillary reflexrecovered and voluntary eye movements werelimited only at the extreme lateral gaze. Mus-cle power was grade 3 and 4 in the proximaland distal parts of the four limbs. Tendonreflexes were still absent. She was taken oVmechanical ventilation the next day. Herclinical condition continued to improve andher symptoms subsided in a stepwise pattern,in response to each course of haemodialysis(figure). When recalling, she could remembercertain events such as the recording of theEEG, but was “too weak to move” at thattime. She regained her initial strength by thetime she was discharged on day 16.

When analysing the remains of the cookedfish (identified as Yongeichthys nebulosus), tet-rodotoxin was demonstrated by thin layerchromatography, high performance liquidchromatography, and cellulose acetate mem-brane electrophoresis. Toxicity was assayedby using Institute of Cancer Research strainadult male mice and the toxicity score was 25mouse units (MU)/g in fish muscle (1MU=0.178 µg in the ICR strain mouse).1

Tetrodotoxin exerts its eVect throughbinding with and blocking the voltage de-pendent sodium channel.2 The voltage clampexperiments showed that tetradotoxin dimin-ished the early sodium inward currentresponsible for the depolarisation of excita-tory membrane. The gating properties of thesodium channel, such as the activation andinactivation mechanism, are not altered—that is, the sodium channel is not perma-nently damaged and its function recoverswhen the bound toxin is released. In uraemia,ion conductance through the sodium channelis also impaired. Sodium permeabilitythrough excitatory membranes is reducedand small inward sodium current and re-duced action potential amplitudes are notedin experimental uraemic neuropathy.3 Bycontrast with the eVects of tetrodotoxin,uraemia changes the basic property of thesodium channel by an increased inactivationand an impaired activation mechanism. Theexcitability of peripheral nerves will be moresignificantly depressed when these two condi-tions combine. The synergistic eVect of urae-mia and tetrodotoxin is obvious in thisincident in which the patient and herhusband ingested roughly an equal amount oftetrodotoxin (about 200 µg, calculated fromtoxic score times the weight of ingested fish).The amount is about 10% of the estimatedlethal dose in humans—2200 µg/60 kg bodyweight4 (body weights of the patient and herhusband were 54.5 and 62 kg respectively)—and caused no clinical evidence of poisoningin the healthy person. It was of interest thatthe CNS was relatively spared from the toxic-ity as the EEG showed a posterior dominant,promptly reactive alpha rhythm and thepatient retained consciousness when thesymptoms were at their most severe.

One of the most striking clinical features inour patient was the response to haemodialy-sis. Despite the small amount of toxiningested, the dramatic improvement of herclinical condition was most likely attributedto the rapid elimination of absorbed toxin inthe course of haemodialysis, rather thanspontaneous recovery. The physical andchemical properties of tetrodotoxin are alsosupportive to this hypothesis.5 It has a lowmolecular weight (C11H17N3O8), is water solu-ble, and is not significantly bound to protein—all these features are often found in toxins

amenable to haemodialysis. Traditionally, themanagement of tetrodotoxin intoxication ismainly supportive, such as gastric lavage toremove unabsorbed toxin and machine as-sisted ventilation when respiration is severelyaVected. We suggest that haemodialysis maybe an eVective method in the treatment oftetrodotoxin intoxication.

MIN-YU LANSHUNG-LON LAI

SHUN-SHENG CHENDepartment of Neurology, Kaohsiung Medical College,

Kaohsiung City, Taiwan

DENG-FWU HWANGDepartment of Food Science, National Taiwan Ocean

University, Keelung City,Taiwan

Correspondence to: Dr Shun-Sheng Chen,Department of Neurology, Kaohsiung MedicalCollege Hospital, 100 Shih-Chung 1st Road,Kaohsiung City 807, Taiwan. Telephone 00886 73121101 ext 6771; fax 00886 7 3234237; [email protected]

1 Hwang DF, Jeng SS. Bioassay of tetrodotoxinusing ICR mouse strain. J Chinese Biochem Soc1991;20:80–6.

2 Takata M, Moore JW, Kao CY, et al. Blockage ofsodium conductance increase in lobster giantaxon by tarichatoxin (tetrodotoxin). J GenPhysiol 1966;49:977–88.

3 Brismar T, Tegnèr R. Experimental uremic neu-ropathy. Part 2. Sodium permeability decreaseand inactivation in potential clamped nervefibers. J Neurol Sci 1984;65:37–45.

4 Tani I. Toxicological studies of puVers in Japan.Teikoku-Tosho (Tokyo) 1945:103.

5 Kao CY. Tetrodotoxin, saxitoxin and theirsignificance in the study of excitation phenom-ena. Pharmacol Rev 1966;18:997–1049.

Relation between critical illnesspolyneuropathy and axonalGuillain-Barré syndrome

The clinical entity critical illness polyneu-ropathy occurs almost exclusively in patientsin critical care units and has been character-ised as a complication of sepsis and multipleorgan failure.1 2 Critical illness polyneuropa-thy may be a common cause of the diYcultyin weaning patients from the ventilator,particularly those who show intractableventilator dependence. All the measures usedto prevent or treat sepsis and multiple organfailure are the main methods now used todeal with critical illness polyneuropathy.Knowledge of this type of polyneuropathy isof help in making decisions about respiratortechniques, nursing care, prognosis, andoverall management. Moreover, recognitionof critical illness polyneuropathy indicates theneed for physiotherapy, rehabilitation, andother supportive measures as the patientrecovers. Bolton et al1 have made an impor-tant positive contribution to the care ofpatients with critical illness polyneuropathy.The actual aetiology, however, has yet to bedetermined. The pathogenesis needs to beclarified to treat patients more eVectively.

Critical illness polyneuropathy invariablyoccurs at the peak of critical illness andsepsis, but in Guillain-Barré syndrome thereis a brief period of recovery after a relativelyminor illness or inoculation. Except fordiVerences in the predisposing causes, asBolton et al1 reported, it is diYcult to distin-guish critical illness polyneuropathy fromGuillain-Barré syndrome on purely clinicalgrounds. In both, polyneuropathy runs amonophasic course, the onset being relativelyacute but with subsequent improvement inmost instances. The clinical features also aresimilar; evidence of muscle weakness in all

four limbs, occasional involvement of facialmuscles and frequent involvement of themuscles of respiration, the depression orabsence of deep tendon reflexes, and someevidence of distal sensory impairment.

The first step by Bolton et al1 in determin-ing exact aetiology was to diVerentiate criticalillness polyneuropathy from Guillain-Barrésyndrome. In reviewing the patients withcritical illness polyneuropathy and Guillain-Barré syndrome who were studied in theirEMG laboratory, they found marked diVer-ences between the two types of polyneuropa-thy. Patients with Guillain-Barré syndromehad greater slowing of the speed of impulseconduction, and, in the initial stages, abnor-mal spontaneous activity in the muscle wasabsent, indicative of a predominantly demy-elinating polyneuropathy. The CSF was onlymildly increased in patients with criticalillness polyneuropathy, but it was muchincreased in patients with Guillain-Barrésyndrome. Comprehensive studies done atnecropsy and nerve biopsies of patients withcritical illness polyneuropathy showed thepresence of primary axonal degeneration ofthe motor and sensory fibres, mainly distally,with no evidence of inflammation.2 Zochodneet al (including Bolton) therefore concludedthat the two types of polyneuropathies mostprobably are separate entities.

Guillain and colleagues enumerated theclinical and spinal fluid features of one formof acute flaccid paralysis without regard forthe underlying physiology or pathology. Clas-sic pathological studies of Guillain-Barrésyndrome, however, have identified promi-nent demyelination and inflammatory infil-trates in the spinal roots and nerves. Guillain-Barré syndrome often has been considered tobe synonymous with the pathological desig-nation of acute inflammatory demyelinatingpolyneuropathy, and physiological abnor-malities consistent with demyelination havebeen taken as supportive evidence for thediagnosis of Guillain-Barré syndrome.Feasby et al (with Bolton)3 first calledattention to patients who were clinically con-sidered as having Guillain-Barré syndrome,but who were characterised electrophysi-ologically as having early axonal degenerationof the motor and sensory nerve fibres. Theevidence included a rapid fall in compoundmuscle action potentials and sensory nerveaction potentials, and no evidence of demyeli-nation. Such patients often had severeparalysis and made a slow recovery, presum-ably reflecting the need to regenerate axonsrather than remyelination. Pathological find-ings are consistent with axonal degenerationwithout demyelination. Feasby et al3 termedthis pattern axonal Guillain-Barré syndromeand suggested that there is a fundamentaldiVerence in the underlying pathophysiology,resulting in primary axonal damage ratherthan demyelination. GriYn et al4 then con-firmed the existence of the acute motor-sensory axonal neuropathy (AMSAN) pat-tern of Guillain-Barré syndrome described byFeasby et al.3

Infection caused by the gram negative bac-terium Campylobacter jejuni, a leading cause

Critical illnesspolyneuropathy

AxonalGuillain-Barré

syndrome





of acute diarrhoea, commonly precedes thedevelopment of Guillain-Barré syndrome.5

There is a close association between axonalGuillain-Barré syndrome and antecedent Cjejuni infection.5 The antecedent infectioussymptom was diarrhoea in three of fivepatients with axonal Guillain-Barré syn-drome described by Feasby et al.3 Observa-tions by GriYn et al4 confirmed that AMSANfollows C jejuni infection. Serum samplesfrom patients with axonal Guillain-Barrésyndrome subsequent to C jejuni enteritisoften have IgG class autoantibodies togangliosides GM1, GM1b, GD1a, orGalNAc-GD1a in the acute phase of theillness,6 and there is molecular mimicrybetween these gangliosides and the lipopoly-saccharides of C jejuni isolates from patientswith Guillain-Barré syndrome.6 This ganglio-side mimicry may trigger high production ofthe IgG antiganglioside antibodies, and theseautoantibodies may cause motor nerve dys-function in patients with GBS.

Interestingly, Hagensee et al7 reported acase of “C jejuni bacteremia and subsequentGuillain-Barré syndrome” that occurred in apatient with chronic graft versus host diseaseafter allogenic bone marrow transplantation.Because there was acute flaccid paralysisassociated with sepsis, some physicians mighthave diagnosed critical illness polyneuropa-thy. Conversely, the existence of this casestrongly suggests that some diagnoses ofcritical illness polyneuropathy should actuallybe axonal Guillain-Barré syndrome orAMSAN. Our hypothesis of the nosologicalrelation between critical illness polyneuropa-thy and axonal Guillain-Barré syndrome isshown in the figure. Serum IgG antibodiesagainst GM1, GM1b, GD1a, or GalNAc-GD1a could be used as immunologicalmarkers for axonal Guillain-Barrésyndrome.6 To examine the aetiology of criti-cal illness polyneuropathy and its nosologicalrelation to axonal Guillain-Barré syndrome, itis necessary to investigate whether patientswith critical illness polyneuropathy have anti-ganglioside antibodies during the acute phaseof the illness.

NOBUHIRO YUKIKOICHI HIRATA

Department of Neurology,Dokkyo University School of Medicine, Japan

Correspondence to: Dr Nobuhiro Yuki, Depart-ment of Neurology, Dokkyo University School ofMedicine, Kitakobayashi 880, Mibu, Shimotsuga,Tochigi 321–0293, Japan.

1 Bolton CF, Laverty DA, Brown JD, et al.Critically ill polyneuropathy. Electrophysiologi-cal studies and diVerentiation from Guillain-Barré syndrome. J Neurol Neurosurg Psychiatry1986;49:563–7.

2 Zochodne DW, Bolton CF, Wells GA, et al.Critical illness polyneuropathy. A complicationof sepsis and multiple organ failure. Brain1987;110:819–42.

3 Feasby TE, Gilbert JJ, Brown WF, et al. Anacute axonal form of Guillain-Barré polyneu-ropathy. Brain 1986;109:1115–26.

4 GriYn JW, Li CY, Ho TW, et al. Pathology ofmotor-sensory axonal Guillain-Barré syn-drome. Ann Neurol 1996;39:17–28.

5 Rees JH, Soudain SE, Gregson NA, et al.Campylobacter jejuni infection and Guillain-Barré syndrome. N Engl J Med 1995;333:1374–9.

6 Yuki N. Anti-ganglioside antibody and neu-ropathy. Review of our research. J Periph NervSyst 1998;3:3–18.

7 Hagensee ME, Benyunes M, Miller JA, et al.Campylobacter jejuni bacteremia and Guillain-Barré syndrome in a patient with GVHD afterallogenic BMT. Bone Marrow Transplant 1994;13:349–51.

Repetitive transcranial magneticstimulation in the treatment of chronicnegative schizophrenia: a pilot study

Recently, a new technology known as repeti-tive transcranial magnetic stimulation(RTMS) has been developed.1 In 1994, theuse of magnetic stimulation in clinicalpsychiatry was suggested.2 Since then, it hasbeen used in the study or treatment ofobsessive-compulsive disorder, conversiondisorder, schizophrenia, and particularly,depression.3

Our pilot study aimed to assess the possibleadverse eVects of this treatment in chronicschizophrenic patients with severe negativesymptoms; to evaluate if direct RTMS of theprefrontal cortex might improve negativesymptoms or cognitive impairments4 in pa-tients with chronic schizophrenia; andthirdly, to note if RTMS might modify thedeficit in prefrontal cortical activity, oftenreferred to as hypofrontality, long establishedin schizophrenia,5 specially under conditionsof task activation.

Six right handed patients with chronicschizophrenia were identified at the outpa-tient psychiatric service of the Hospital Clínicof Barcelona. There were two men and fourwomen (mean age 39).

Exclusion criteria included alcohol or sub-stance abuse dependence disorder in the past5 years, focal neurological findings, systemicneurological illness, taking cerebral metabolicactivator or vasodilator medications, electro-convulsive therapy within 6 months, and sig-nificant abnormal findings on laboratoryexamination.

All patients were taking neuroleptic drugs,but a stable dose for at least 3 months wasrequired. All patients were studied oV benzo-diazepines for at least 1 week before begin-ning the treatment. During the RTMS,psychotropic medications were continued atthe initial dosage.

All patients were admitted to hospital.Inpatients underwent the UKU side eVectsscale,6 the positive and negative syndromescale (PANSS), and a neuropsychologicalbattery, the day before beginning the treat-ment and at the end of the treatment. TheUKU scale was also administered after eachsession.

An equivalent neuropsychological batterywas used on both occasions, which consistedof the block design subtest of the Wechsleradult intelligence scale, the trail making testsA and B, the FAS verbal fluency test, and twosubtests of the Wechsler memory scale (thevisual memory reproduction and the verbalpaired associates subtests).

A brain SPECT study was performedusing a rotating dual head gamma camera,fitted with high resolution fanbeam collima-tors. Two 99mTc-HMPAO SPECT scans withcognitive activation, such as the Wisconsincard sorting test (WCST), were performedon each patient (24 hours before thebeginning of the treatment and 24 hours afterthe last session).

RTMS was given with a Mag Pro magneticstimulator, 5 days a week, during 2 weeks, ata dosage of 20 Hz for 2 seconds, once perminute for 20 minutes at 80% motor thresh-old. The motor threshold was determined byvisualisation of finger movement. A butterflymagnetic coil was placed tangential to theorbital area, on the C3 and C4 EEG point.

An important finding of this study was thatRTMS may be given to stable schizophrenicpatients without exacerbating their psycho-

ses. All patients tolerated the RTMS well,with minimal side eVects (mild headache andtinnitus).

Initial SPECT of one patient was reportedto be normal, showing no evidence of hypo-frontality. The remainder of the patientsshowed hypofrontality on the initial neuroim-aging. The results after RTMS indicated nochange in the hypofrontality.

Negative symptoms showed a generaldecrease for all patients (table). Significance(p<0.02) was noted on the PANSS negativesymptoms subscale. These patients seemedto be more sociable than when originallyseen. Nevertheless, clinical eVects of theRTMS were subtle and diYcult to distinguishfrom those derived from the supportive envi-ronment of the psychiatric ward.

With regard to the neuropsychological bat-tery, we found a general improvement in allpost-treatment scores (table), but only de-layed visual memory achieved significance(p<0.05). This feature might be basicallyexplained by improvement of attention,specifically of the maintenance of attention,which allows the correct function of theworking memory. Thus, although there aremethodological limitations regarding thepower of our conclusions, it is certain thatthere has been an improvement in the atten-tional capability.

We found that all patients (except one, whowas always within the normal range) dimin-ished their number of perseverative answersand errors on WCST (items characteristicallyaltered in schizophrenia) after the RTMS.However, significance was not achieved onany WCST scores.

Two patients who initially did not performany categories on WCST, after the treatment,achieved one category, a possible indicationof improvement of their abstract thinking.This change leads us to consider a researchstrategy previously reported, in which theWCST is used as a screening test for selectingschizophrenic patients. Those initially achiev-ing low category scores would be comparedto higher category scorers in an eVort toidentify a subgroup most likely to benefitfrom RTMS.

Taking into account these mild improve-ments together, and the lack of changes in

Table Neuropsychological tests and PANSSscores

Test Mean (SD)

Block design Pre 49 (11.95) NSPost 50 (8.69)

Trail making testA

Pre 38.3 (9.83) NSPost 42.6 (14.1)

Trail making testB

Pre 38.3 (4.5) NSPost 41 (10.03)

Inmediate visualreproduction

Pre 50.5 (4.82) NSPost 54.8 (11.2)

Delayed visualreproduction

Pre 46.19 (8.23) p<0.05Post 53.8 (12.64)

Inmediate verbalpairedassociates

Pre 54 (7.46) NS

Post 59.5 (10.03)

Delayed verbalpairedassociates

Pre 8.8 (1.1) NS.

Post 8.8 (1.17)

PANSS-PG Pre 37.67 (11.15) NSPost 36.5 (11.47)

PANSS-N Pre 31.67 (8.26) p<0.02Post 27.83 (8.47)

PANSS-P Pre 16.83 (7.28) NSPost 15.33 /7.55)

Pre=pretreatment; Post=post-treatment; PANSS=positive and negative syndrome scale; PG=generalpsychopathology scale; N=negative scale;P=positive scale.





hypofrontality after treatment, we are consid-ering extending the treatment course to 20sessions, each at 30 Hz for 1 second, at 90%of motor threshold. It was also suggested thatother positions of the coil and other kinds ofcoils might give better results.

The clinical change in our cohort after theRTMS could be attributed to both the treat-ment and the supportive environment of thepsychiatric ward, and even to enhancecompliance to medication during hospitaladmission. We are aware that the small sam-ple size and lack of controls compel a verycareful interpretation of the results.Nevertheless, in the light of these, we suggestfurther controlled studies of the eYcacy ofRTMS in negative symptoms of schizophre-nia, not only as an add on technique but alsoas a sole therapeutic procedure. Research onRTMS also requires some controlled studiesaimed to the complexity of the methodology(dosage, duration, and localisation), as thisform of intervention may prove to be an eco-nomical and convenient therapy in treatingseveral psychiatric disorders.

E COHENM BERNARDO

J MASANAF J ARRUFATV NAVARRO

Department of Psychiatry

J VALLS-SOLÉDepartment of Neurophysiology

T BOGETN BARRANTESS CATARINEU

M FONTDepartment of Psychology

F J LOMEÑADepartment of Nuclear Medicine,

Institut d’ Investigacions Biomédiques August Pi iSunyer, Hospital Clínic i Provincial. Universitat de

Barcelona, Spain

Correspondence to: Dr M Bernardo, Servicio dePsiquiatría, Hospital Clínic i Provincial, Villarroel170, 08036 Barcelona, Spain. Telephone 003432275400, ext 2405; fax 00 343 2275477; [email protected]

1 Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of the humanmotor cortex. Lancet 1985;i:1106–7.

2 George MS, Wassermann EM. Rapid-rate tran-scranial magnetic stimulation and ECT. Con-vuls Ther 1994;10:251–4.

3 Pascual-Leone A, Catala MD, Pascual-LeoneAP. Lateralized eVect of RTMS on the prefron-tal cortex on mood. Neurology 1996;46:499–502.

4 Elliot R, Sahakian B. The neuropsychology ofschizophrenia: relations with clinical and neu-robiological dimensions. Psychol Med 1995;25:581–94.

5 Wolkin A, Sanfilipo M, Wolf AP, et al. Negativesymptoms and hypofrontality in chronicschizophrenia. Arch Gen Psychiatry 1992;49:959–65.

6 Lingjaerde O, Ahlfors UG, Bech P, et al. TheUKU side eVects scale. Acta Psychiatr Scand1987;76(suppl 334):1–100.

CORRESPONDENCE

Sensory alien hand syndrome

The case report by Ay et al1 of alien hand syn-drome and review of the literature neglectedthe intriguing issue of why in every case so farreported the patient seems to be terrified of

the alien limb. Not believing that you are anymore in control of a limb is not likely to be apleasant experience.

Those with alien hand syndrome seem tojump to extremely negative conclusions con-cerning the intent of the limb. Typically, as inthe report of Ay et al, the common belief isthat the limb has deeply malevolent inten-tions towards the victim.

It is this aspect of alien hand syndrome thatI suggest also needs incorporating into itsneurological explanations, and which pro-vides a clue as to why our everydayexperience of being in charge of our bodies,and so initiating all personal action, itself hasa neurological basis. In other words, while thebrain is the seat of all our actions and experi-ences, there is also a part of our nervous sys-tem which is responsible for our belief that wehave free will over our behaviour. Patientswith alien hand syndrome think that they areno longer in control of a limb because thepart of the brain that gives us the sensation ofcontrol over our bodies has been damaged.When that happens, our limbs seem to actindependently of us.

Research2 conducted in the 1980s hasfound that the same electrical brain wavechanges that characteristically precede alllimb movements, occur several 100 ms beforewe seem to consciously decide to move alimb. If our conscious decision to act ispreceded by brain changes that anticipateaction, then our “decision” to choose how tobehave or “freedom”, as in free will, is in factillusory. Our choices have in a sense beendecided beforehand by our brains.

Spence3 asserts that evidence such as this,combined with phenomena such as alienhand syndrome, means that philosophershave to reconsider whether we have free will.He argues that these data suggest that oursense of agency is illusory and it follows thatmost of us share in common the useful delu-sion that we have free will. Patients with alienhand syndrome have lost this experience inrelation to a particular limb. There is a sensethen that those who experience the syndromeare closer to the reality of how much we areresponsible for our actions than the rest of us.This is because they have lost the function ofthe part of the brain that normally works tomake us think that we have consciousfreedom of will. They develop the experience,therefore, of becoming mere remote specta-tors to the actions of their bodies.

Defenders of human “free will” argue whathappens before the brain itself decides to actis still unknown, and there may be a role forour own autonomy there. But even these freewill guardians concede the neurologicalresearch indicates that whatever happensbefore the brain is roused, must occur belowour conscious awareness.

Yet in alien hand syndrome the patientthinks that the hand has hostile motivations;it is invariably the case that the patient notonly thinks that the limb is “not self” butfinds that the limb behaves towards the self ina destructive and aggressive manner. Thiscould be explained by the suggestion that ifwe lose our conscious sense of voluntary con-trol over our bodies, our minds have to comeup with an explanation for the location ofaction of our movements. We decide that ifourselves are not in control, then someone orsomething else must be; therefore, we nolonger have a sense of the limb belonging tous.

Because to lose control over our bodies isone of the most terrifying experiences, our

attempt to explain this finding occurs in thecontext of fear. It may be that our apprehen-sion leads us to misinterpret innocent reflex-ive acts of our hands, such as scratching orrubbing, as malevolently inspired. Plus itcould be that our interpretation of spitefulpossession in turn inspires the hand itself,only this is beyond our conscious awareness.

It may therefore be that we need to believein our own free will and personal control overour acts, because if we did not, we might findthe experience of our bodies acting as if wemerely came along for the ride, too frighten-ing. Also, we may no longer believe that ourbodies or its relevant parts belong to us. Allneurologists who have reported alien handsyndrome remark on how psychologicallydisturbing the symptom is for the patient.Psychiatrists would be interested in theparallels between alien hand syndrome andthe “passivity phenomena” of schizophrenia,plus the fact that the two diseases may sharecorpus callosum pathology,4 could go someway to explaining why schizophrenic symp-toms are frightening to the patient. So itseems we know that our limbs belong to usbecause they obey us. When they seem tostop responding to our wills, we conclude thatour limbs are no longer our own, and try tofend them oV. Hence it would seem that oneof the prices we had to pay for consciousawareness of ourselves to evolve as a functionof the brain, is the delusion that we areresponsible for all our actions. If we had con-scious awareness of ourselves, but no sense offree will, our bodies would feel alien to us.The philosophical importance of alien handsyndrome is that it shows emphatically vianeurology that it is possible to drive a wedgebetween consciousness and the experience offree will. The brain had to develop the sensa-tion of free will after developing conscious-ness, because being without the sensation offree will produces extremely negative emo-tional experiences. So the fact that every caseso far reported of alien hand syndromeimputes negative intent to the alien limbmight not be an incidental finding, but a coreaspect of the disorder.

R PERSAUDThe Maudsley Hospital, Croydon Mental Health

Services, Westways Rehabilitation Unit, 49 St James’sRoad, West Croydon, Surrey CR9 2RR, UK. Telephone

0044 181 700 8512; fax 0044 181 700 8504; [email protected]

1 Ay H, Buonanno FS, Price BH, et al. Sensoryalien hand syndrome: case report and review ofthe literature. J Neurol Neursurg Psychiatry1998;65:366–9.

2 Libet B, Gleason CA, Wright EW. Time of con-scious intention to act in relations to onset ofcerebral activity: readiness-potential. Brain1983;106:623–42.

3 Spence SA. Free will in the light of neuropsy-chiatry. Philosophy Psychol Psychiatry 1996;3:75–90.

4 Crow TJ. Schizophrenia as a transcallosalmisconnection syndrome. Schizophr Res 1998;30:111–4.

The authors reply:We appreciate Persaud’s comments regardingthe alien hand syndrome, “the perceivedmalevolence of the aVected limb towards itsvictim, and the question of whether with lossof the conscious sense of voluntary controlover our bodies, our minds... decide that ifourselves are not in control then someone orsomething else must be”. We would oVer thatthe value of our particular case is that it wasdue to a central deaVerentation—thereforethe term “sensory alien hand syndrome”. As





opposed to the idea that “we know our limbsbelong to us because they obey us”, we knowthat our limbs belong to us because they pro-vide us with sensory input that is recognisedas self. Many patients with movementdisorders or paralysis lose control of theirlimbs but still have no diYculty in realisingthem as self. Indeed even in “phantom limb”there is sense of self due to central processesin the absence of a limb. Our patient, as doothers with anosognosia and primary abnor-malities of central sensory systems, showsperhaps that it is central sensory processesthat are the key to identifying “self”. Weknow our limbs not because they obey us butbecause there is a pattern of sensory activa-tion that accompanies our own limb move-ments. When this pattern never reachesspecific cortical regions, then the limb is notperceived as self; called “amorphosynthesis”by Denny-Brown and Banker.1 Self stimula-tion by the centrally deaVerented limb in“sensory” or “posterior” alien hand syn-drome, or kinaesthetic stimuli due to move-ment of the limb as in the “anterior” or“motor” alien hand syndrome, is perceived asdue to another person or thing without criti-cal questioning. This raises the most interest-ing question of what brain region is deaVer-ented in the anterior alien hand syndromewhere the sensory processing is intact.

It is not our clinical experience nor theconclusions based on published reports thatall patients suVering with alien hand syn-drome are terrified by the eVected limb. Inone author’s experience (BHP), two patientswith alien hand syndrome and relatedintermanual conflict were irritated by but notterrified by their opposing limbs simultane-ously vying for a cigarette or book. Anotherpatient was amused but rather indiVerent tohis aVected left side. The most terrifying situ-ation we have heard is when the patient iden-tified his aVected left side as belonging to hismother in law! A patient reported byHeilman’s group2 with persistent alien handsyndrome referred to it as “my little sister”.Similar to our experience, they suggest that apredisposing personality type may be neces-sary given that most patients with collosalinfarcts or tumours do not emphasise thiscomplaint.

Unlike our case of limited duration, thepersistence of alien hand syndrome seemsdependent on mesial frontal dysfunction.These patients rarely deny that the aVectedlimb belongs to them. Instead, they under-stand it in terms of their “anarchic hand”.Hence, although the initial syndrome mayresult in disjointed and terrifying perceptions,it seems that the brain quickly re-establishesits control by presently unknown adaptivecapacities. Furthermore, why it almost exclu-sively involves the left body side in righthanded people remains unknown. Studyingthis syndrome in greater detail may yieldadditional insights into the pathophysiologyof denial and misidentification.

HAKAN AYFERDINANDO S BUONANNO

DEAN A LEWALTER J KOROSHETZ

Department of Neurology, Stroke Service,Massachusetts General Hospital, Harvard Medical

School, 32 Fruit Street, Boston MA 02114, USA

BRUCE H PRICEDepartment of Neurology, McLean Hospital, 115 Mill

Street, Belmont MA 02178–9106, USA

1 Denny-Brown D, Banker B. Amorphosynthesisfrom left parietal lesion. Arch Neurol Psychiatry1954;71:302.

2 Heilman KM, Valenstein E, eds. Clinical neu-ropsychology 3rd ed. New York: Oxford Univer-sity Press 1993:361.

Vasomotor reactivity is exhausted intransient ischaemic attacks with limbshaking

The article of Baumgartner and Baumgartnerentitled “Vasomotor reactivity is exhausted intransient ischaemic attacks with limbshaking”1 provides interesting new infor-mation regarding the nature of involuntarylimb movements contralateral to haemody-namic failure from severe carotid arteryocclusive disease. The authors evoke an“exhausted cerebral vasoreactivity in thehemispheres opposite the involuntary limbmovements”. In their report, involuntarymovements aVected only the limbs, anddisplayed no tonic contraction, tonic-clonicjerking, or Jacksonian march and no epilepticactivity during or between the attacks. Thesefindings led the authors to strongly argueagainst seizures as the cause of limb shakingin these transient ischaemic events.

In contradistinction, a 72 year old righthanded man was admitted to our hospitalwith a 3 month history of episodic weaknessand numbness of the right arm. The patientthen had six discrete stereotypic episodes ofright arm weakness and clumsiness that werealso associated with diYculty in speaking.Several episodes of dysarthria, numbness andweakness of the right arm and leg (MRCgrade 4/5) were seen, unrelated to posture,some of which occurred when the patient wassupine. Most episodes were characterised byslight tremulousness and asterixis-like move-ments of the outstretched right arm. Therewas a return to baseline functioning betweenevents.2 Video/EEG monitoring, however,showed low voltage spikes in the left central-parietal head regions contralateral to thefacial twitching and the right arm and rightleg weakness. Although ongoing clinical andEEG seizure activity stopped after 2 mgintravenous lorazepam, they reoccurred afterloading with phenytoin. Because angiographydisclosed a greater than 95% stenosis of theleft internal carotid artery (while the patientwas treated with phenytoin at a concentrationof 16.5 mg/l), the patient was anticoagulatedwith heparin, but episodes continued. It wasonly after a left carotid endarterectomy thatall episodes of weakness, tremulousness, andEEG epileptiform activity stopped. Theyhave not recurred over the past 5 years.

The literature includes several cases offocal motor inhibitory seizures causingweakness.3 4 Although it is impossible toprove a negative, it could be argued thatalthough no epileptiform or other evidence ofseizure activity is present in a particular case,the abolition of ongoing clinical and EEGevidence of inhibitory motor activity byintravenous diazepam argues in favour, atleast in part, of an ictal contribution. The factthat in virtually all reported cases, abnormalmovements are more definitively resolved bycarotid endarterectomy, argues for an under-lying ischaemic aetiology that induces focalseizures. There are few reports that clearlydelineate the interaction and association ofinhibitory focal motor seizures and transientischaemic attacks, as there are few sequentialtrials of antiseizure drugs or anticoagulation(under EEG monitoring) and finally carotidendarterectomy. Several authors support theconcept of an inhibition of motor function inparietal and secondary somatosensory re-

gions by seizure activity which then inter-rupts the sensory feedback loop to motorintegration with inhibition of subcortical andcortical areas.5

PETER W KAPLANJohns Hopkins Bayview Medical Center, 4940 Eastern

Avenue, Baltimore, MD 21224, USA

1 Baumgartner RW, Baumgartner I. Vasomotorreactivity is exhausted in transient ischaemicattacks with limb shaking. J Neurol NeurosurgPsychiatry 1998;65:561–4.

2 Kaplan PW. Focal seizures resembling transientischemic attacks due to subclinical ischemia.Cerebrovasc Dis 1993;3:241–3.

3 Lee H, Lerner A. Transient inhibitory seizuresmimicking crescendo TIAs. Neurology 1990;40:165–6.

4 Tinuper P, Aguglia U, Laudadio S, et al.Prolonged ictal paralysis: electroencephalo-graphic confirmation of its epileptic nature.Clin Electroenceph 1987:18:12–14.

5 Luders H, Lesser RP, Dinner DS, et al. The sec-ondary sensory area in humans: evoked poten-tial and electrical stimulation studies. AnnNeurol 1985;18:177–84.

Baumgartner and Baumgartner reply:We are grateful for the response of Kaplan toour short report. We agree that somaticinhibitory seizures may mimick transientischaemic attacks (TIAs). Such TIAs areassociated with negative symptoms such assensorimotor deficits and diYculty withspeaking, EEG evidence of seizure activity,and ceasing of the TIAs after the administra-tion of an anticonvulsant drug.1 2 Limb shak-ing TIAs, however, diVer from TIAs relatedto inhibitory seizures in several ways. (1)They are associated with positive phenomena(limb shaking), and the involuntary move-ments do not aVect the facial muscles. (2)Patients with attacks of shaking movementsof the limbs have no EEG evidence of epilep-tic activity, and involuntary movements donot stop after administration of anticonvul-sive therapy. (3) Although the patient pre-sented by Kaplan had a 95% stenosis of theleft internal carotid artery, it is unclearwhether haemodynamic failure was presentor not, because no studies evaluating thehaemodynamic reserve of the homolateralhemisphere were presented. This is inaccordance with the finding that the involun-tary movements as well as the sensorimotordeficits of Kaplans’ patient were not relatedto posture. (4) The pathogenesis is thought tobe due to disinhibition of subcortical controlmechanisms as a result of ischaemia.

In our opinion, it is not clear whether theasterixis-like movements of the outstretchedright arm of Kaplan’s patient are due to epi-leptic seizures, because unilateral asterixis ofthe outstretched arm has been reported withcontralateral vascular lesions aVecting almostall cerebral structures involved in motioncontrol including ischaemia in the territory ofthe middle cerebral artery.3

RALF W BAUMGARTNERDepartment of Neurology, University Hospital of

Zürich , Switzerland

IRIS BAUMGARTNERDivision of Angiology, University Hospital of Bern,

Switzerland

Correspondence to: Dr Ralf W Baumgartner, Neu-rologische Klinik, Frauenklinikstrase 26, CH-8091Zürich, Switzerland. Telephone 0041 1 255 56 86;fax 0041 1 255 43 80; [email protected]

1 Kaplan PW. Focal seizures resembling transientischemic attacks due to subclinical ischemia.Cerebrovasc Dis 1993;3:241–3.





2 Lee H, Lerner A. Transient inhibitory seizuresmimicking crescendo TIAs. Neurology 1990;40:165–6.

3 Melo TP, Bogousslavsky J. Specific neurologicmanifestations of stroke. In: Ginsberg MD,Bogousslavsky J, eds. Cerebrovascular disease.Pathophysiology, diagnosis, and management. Vol2. Malden, MA: Blackwell Science, 1998:961–85.

BOOK REVIEWS

Neurology for the MRCP Part II. ByDAVID SMITH. (Pp164, £15.99). Published byArnold, London 1998. ISBN 034069175.

To the MRCP candidate neurology is one ofthe more daunting specialties. The unfamiliarnerve conduction study and the frankly mys-terious EEG can distress an otherwise wellrounded senior house oYcer. Despite the factthat much of neurology is commonly seen ona general medical ward—strokes, dementiasand so forth—the general perception is of anunimaginable list of eponymous syndromesand obscure signs. Rather than dwell on thelast, in this book Dr Smith tries to address thecommoner complaints as examination stylequestions each with a “simple clinical les-son”. The “grey case” section, for instance,includes questions on multiple sclerosis,cluster headache, and HSV encephalitis,while broadening the topics to includepostinfective demyelination, chronic hemi-crania, and acute haemorrhagic encephalo-myelitis. There is, however, a tendency for thediscussion after each question to be ratherbrief. A fuller explanation, with more allow-ance for the reader’s ignorance, would havebeen appreciated. The data interpretationsection is somewhat better, covering CSF,EEG, and other data extremely well. Perhapsa little too well; would an MRCP candidatereally be expected to recognise the character-istic EEG of Creutzfeldt-Jakob disease? Isurely hope not. Finally, the slide tests aredisappointing. If anything, neurology lendsitself best to this section of the written exam-ination but it is let down by the poor qualityof some of the images in this book. This isespecially unfortunate, as other images in thesame section are remarkably impressive. TheSturge-Weber skull radiograph and centralpontine myelinolysis MRI are beautiful. Insummary, this is a creditable first edition. Ilook forward to the second.

STEFAN MARCINIAK

An Atlas of Multiple Sclerosis. Edited byCharles M Poser. (Pp 134, £48.00/$85.00).Published by Parthenon Publishing,Carnforth 1998. ISBN 1-85070-946-7.

This book, after a short introduction to someof the fundamental features of the diseasegoes on to provide some 117 illustrations ofaspects of the disease from Cruveihier’s platesto histopathological specimens and also aheavy leaning to imaging particularly mag-netic resonance scanning, as might beexpected. There is no doubting the aestheticimpact of this short book. In addition, the factthat these illustrations emanate from a wellestablished figure in the multiple sclerosisworld and are likely to be a representative set

of personal teaching slides from a successfulacademic career all vouch for the provenanceand informative nature of the atlas. Howeverthe place of such a book within a neurologist’slibrary has to be questioned. There are aplethora of high quality textbooks devoted toall aspects of multiple sclerosis, all well illus-trated and most in colour. They provide indepth analysis of all aspects of the disease andalthough their illustrations tend to be smallerthis is where I would choose to put mymoney. It may be that the circulation of thisbook will be higher than expected as it islikely to be a popular choice for somepharmaceutical companies.

NEIL ROBERTSON

Calvarial and Dural Reconstruction.Edited by SETTI R RENGACHARY and EDWARD C

BENZEL. (Pp 200). Published by AANS,Illinois 1998. ISBN 1-879284-63-4.

This monograph is the latest to be producedby the American Association of NeurologicalSurgeons as part of its Neurosurgical Topicsseries. It begins by tracing the history of cal-varial reconstruction from ancient times.There follows a discussion of the diVerentautologous donor sites and synthetic materi-als currently available for repair of calvarialand facial defects. The merits, disadvantages,and contraindications of each are considered.Dural substitutes are then dealt with in simi-lar fashion. Specific problems, such as scalpreconstruction, the management of commi-nuted frontal sinus fractures, and reconstruc-tion of the anterior skull base are the subjectof separate chapters. The final part of thebook is devoted to craniosynostosis. A reviewof current knowledge on pathogenesis isfollowed by a good account of some of themore common techniques used to treat singlesuture synostosis. Understandably, in a bookof this type there is space only for an overviewof the treatment and complications of multi-suture involvement, but the chapter provideswell chosen references for further reading.

The reconstruction of traumatic and post-surgical calvarial defects occupies the bulk ofthis volume, and is dealt with very eVectively.Operative techniques and the relative meritsof various materials are covered in a clear andconsise manner. By contrast, the section onaural substitutes is a little disappointingbecause it does not provide the reader withreasoned argument on how to select the mostappropriate graft from the sometimes bewil-dering variety of autologous, synthetic, andxenograft materials which are available whenvascularised pericranial tissue is not anoption.

Craniosynostosis is a topic which is cov-ered very well in standard paediatric neuro-surgical texts and it is not worth buying thisbook for that section alone. However, theaccount of techniques for repair of calvarialdefects is excellent and merits the inclusion ofthis text in a departmental library.

ROBERT MACFARLANE

Echoenhancers and Transcranial ColorDuplex Sonography. Edited by ULRICH

BOGDAHN, GEORG BECKER, and FELIX

SCHLACHETZKI. (Pp 420). Published byBlackwell Science, Oxford 1998. ISBN0-632-04856-5.

Transcranial colour duplex sonography is anultrasound technique which is becoming

increasingly available for the non-invasiveimaging of intracranial structures, particu-larly the basal cerebral arteries. There arenow four principal components to thetechnique: B mode ultrasound which can beused to image the brain parenchyma; colourcoded Doppler which provides a colourimage of the basal vessels; spectral analysis ofpulsed wave Doppler which is used to deriveblood flow velocities; and latterly “power”Doppler which is also used for vascular imag-ing following analysis of the amplitude ratherthan the frequency of the reflected ultrasoundbeam. In addition, echocontrast agents arenow available which can increase the signal tonoise ratio and thus help overcome some ofthe detrimental acoustic eVects of the skull.

This volume of 400 pages and liberalcolour diagrams and prints is edited by threeexponents of the technique. Thirty one chap-ters written by a further 20 contributors covertopics from the history of transcranial ultra-sound, through the physics of Doppler ultra-sound to potential clinical applications. Thebook is helpfully split into two sections withthe theoretical aspects described in the firsthalf and clinical aspects in the second.

This is certainly a specialised book and willreally only appeal to those interested in, orwishing to develop, expertise in transcranialcolour coded ultrasound. For such readersthe technical chapters on instrumentation,signal processing, echocontrast agents, har-monic imaging etc will certainly provide acomplete understanding of the priciplesbehind the technique. I think that some of theerrors made in the interpretation of vascularultrasound examinations are due to anincomplete understanding of the physics ofDoppler ultrasound, hence the attention paidto this area is commendable.

The clinical section covers the examinationtechnique, normal reference values, the maincategories of cerebrovascular disease, andalso contains chapters on areas which may beless immediately suitable for ultrasoundstudy. For example, the findings in headtrauma, tumours, psychiatric disorders, andmovement disorders are the subject ofseparate chapters. Although I have no prob-lem with enthusiasm for this technique a littlepragmatism would not go amiss. A more bal-anced discussion of the limitations as well aspotentials of the technique could have beenapplied.

As with any book with multiple authorsthere is some variation in style and overlap,particularly in the introductions and conclu-sions of the chapters. Nevertheless, it is acomprehensive current review of transcranialcolour coded sonography. Although thereader must decide exactly how this tech-nique fits into clinical practice the book willcertainly stimulate some ideas.

PETER MARTIN

Spinal Cord Diseases-Diagnosis andTreatment. Edited by GORDON L ENGLER,JONATHAN COLE, and W LOUIS MERTON. (Pp675, US$195). Published by Marcel Dekker,New York, 1998. ISBN 0-8247-9489-3.

This is volume 47 of a series entitledNeurological Disease and Therapy, series editorW C Koller. This volume is edited by anAmerican surgeon and two British neuro-physiologists. Most of the 45 contributors areAmerican or British, almost half of whom,including Dr Cole, are from Southampton.The book begins with a pathophysiological





introduction setting the scene for the fivemain disease sections covering develop-mental/genetic disease, spinal injury, infec-tion, tumour, and the eVect of neurologicaland systemic disease on the spinal cord. Thischapter covers a wide area from multiplesclerosis to motor neuron disease to vasculardisease to metabolic diseases. Then follows asection on investigation considering imaging,neurophysiology, and urodynamics. Finally,there is a miscellaneous section coveringclinically important entities such as pain,sexual problems, and terminal care associatedwith spinal cord disease but also including ahighly specialised chapter on the role ofomental transfer in spinal cord injury.

This is an ambitious attempt at being com-prehensive. The editors themselves worrythat the emphasis favours surgical conditions.Although this might be the case, many surgi-cal cases present to the neurologist orrheumatologist, care for spinal disease oftenfalling between several specialties. Therefore,it is of benefit to the clinician to have allaspects of spinal disease in one volume. Thestandard and style of the individual chaptersvaries, that on motor neuron disease being upto date and topical, malignancies beingcovered in depth. That on sexual problemsassociated with spinal cord disease is excel-lent, particularly practical and a must forboth doctors dealing with spinal disease andfor patients themselves who are often unin-formed (our fault, not theirs). The chapter ondecompression illness will be food forthought for many doctors who enjoy recrea-tional diving, for although studies have notyet shown adverse aVects on the quality of lifein those who dive frequently but withoutincident, the evidence for cumulative neuro-logical damage from neurophysiological, im-aging, and pathological studies is compelling.

The quality of illustration is high. Perhapsnot surprisingly, this is particularly evident inthe imaging section (where there is a ratherspectacular sagittal T2 weighted MRI of aintramedullary arteriovenous malformation).In addition to imaging many of the chaptersalso make good use of schematic diagramsand line drawings to enhance the text.

Drs Engler, Cole, and Merton end theirpreface by commenting that “Our main hope,however, is that the chapters will read as aseries of views on the spinal cord and its dis-ease, so that a surgeon may learn aboutcurrent practice as well as the wide range ofconditions aVecting the cord that are outsidethe field of surgery”. While I agree that edu-cating surgeons is an admirable aim, I thinkthat the authors rather undersell themselvesand that this book’s main strength, as I havesaid above, is that it will appeal to alldisciplines that deal with spinal cord disease,bringing together neurological, rheumato-logical, and surgical disease that is often cov-ered in separate textbooks.

GILLIAN HALL

Neurologic Disease in Women. Edited byPeter Kaplan. (Pp 480 $150). Published byDemos, New York 1998. ISBN1-888799-153.

This is the second time that I have been askedto review a book on this topic. The first timeI approached the task with some scepticism—were neurological diseases in women really sodiVerent from those in men that theywarranted their own text book? But I rapidlybecame a convert to the cause, beingreminded that there are issues specific tofemales that influence both disease, investiga-tion, and treatment (pregnancy, breast feed-ing, menopause, to name the most obvious)and that not all neurological diseases attackthe sexes equally. There are also wider socio-economic and legal issues that play a part inthe complete disease picture which many ofus neglect too often but which this book iscareful to address (see below). Leavingcontent aside for a moment, this is abeautifully presented book; clearly headedand with wide use of well constructed tables.It encourages one to read on. It seems up todate and well referenced.

The contributors (40 in total) are exclu-sively American, and east coast American atthat with only occasional forays westward.The text is divided into three sections. Thefirst, entitled General Issues in Women includesan anatomical chapter considering the sexdiVerences of regional brain structure andfunction. More novel for this type of text, itcontains two thoughtful chapters consideringwomen’s health within the context of theirlifestyles and women’s health and its relationwith the law. This chapter considers issuessuch as coercive approaches to preventingfoetal harm, those relating to informedconsent to medical treatment, and diYcultchoices with neurological implications. Thelaw and the case examples are exclusivelyAmerican but the issues are universal. Thisopening section leaves no doubt that this is abook that has taken female issues extremelyseriously.

The second section looks at neurologicaldiseases as they aVect females at diVerent lifestages, from birth through menarche, preg-nancy, and menopause, to the elderly woman.As well as considering genetic diseases thatstrike at a particular age, these chaptersconsider the influence of changing physiologyand hormonal balance on neurological dis-ease. The third section is the most conven-tional. Each chapter considers a neurologicaldisease representing these diseases with em-phasis on their eVect on women and there is,by necessity, some overlap between this andthe previous section. As a non-American, Iwould feel more comfortable to believe thatthe high number of female patients withperipheral nerve injuries secondary to physicalbeatings, knife wounds, or gunshot woundsreflected the country of origin of this book!

If pushed to criticise, the indexing could bemore complete and certain conditions con-sidered in more detail, in particular, paraneo-plastic conditions associated with breast andgynaecological malignancies. However, thataside, I think this a rather special book andnot only a good addition to any neurologicallibrary but a useful purchase for anyoneinterested in female medical issues.

GILLIAN HALL

The reader may be interested in thefollowing:

Neuro-Oncology for Nursing. Edited byDOUGLAS GUERRERO. (Pp 330, £24.50).Published by Whurr Publishers Ltd, London1998. ISBN 1-86156-087-7.

A Research Portfolio on ChronicFatigue. Edited by ROBIN FOX. (Pp 72,£14.95). Published by The Royal Society ofMedicine Press Ltd, London 1998. ISBN 185315 367 2.

CORRECTION

Hanna PA, Jankovic J, Vincent A.Comparison of mouse bioassay andimmunoprecipation assay for botuli-num toxin antibodies. J Neurol Neuro-surg Psychiatry 1999;66:612–16.During the editorial process the de-scriptions of the histograms in figure 4(p 614) were wrongly ascribed. Thecorrected figure is reproduced below.

20

0


0/1 2 3/4

No response to test injectionGood response to test injection

Ob

serv

atio

ns

(n)

10

Figure 4 Correlation of clinical response(grade 0 or 1 response indicates non-responders,grade 2 response indicates reduced response, andgrade 3 or 4 are responders) with response to testinjections.





doi: 10.1136/jnnp.66.5.612 1999 66: 612-616J Neurol Neurosurg Psychiatry

Philip A Hanna, Joseph Jankovic and Angela Vincent antibodies

toxinimmunoprecipitation assay for botulinum Comparison of mouse bioassay and

http://jnnp.bmj.com/content/66/5/612.full.htmlUpdated information and services can be found at:

These include:

References

http://jnnp.bmj.com/content/66/5/612.full.html#related-urlsArticle cited in:

http://jnnp.bmj.com/content/66/5/612.full.html#ref-list-1This article cites 19 articles, 11 of which can be accessed free at:

serviceEmail alerting

box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in the

CollectionsTopic

(8624 articles)Neuromuscular disease � (8854 articles)Drugs: CNS (not psychiatric) �

Articles on similar topics can be found in the following collections

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://jnnp.bmj.com/content/66/5/612.full.html

http://jnnp.bmj.com/content/66/5/612.full.html#ref-list-1

http://jnnp.bmj.com/content/66/5/612.full.html#related-urls

http://jnnp.bmj.com/cgi/collection/drugs_cns_not_psychiatric

http://jnnp.bmj.com/cgi/collection/neuromuscular_disease

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/



comparison of mouse bioassay and immunoprecipitation assay for botulinum toxin antibodies

Documents