Downloaded from http://jnnp.bmj.com/ on October 22, 2017 - Published by group.bmj.

com

560 EDITORIAL

Gluten sensitivity may not be the sole protagonist in this

................................................................................... disease.

Gluten sensitivity as a neurological THE NEUROLOGY OF COELIAC

DISEASE
illness In 1966 Cooke published a landmark
paper on 16 patients with neurological
disorders associated with adult CD.8 This
M Hadjivassiliou, R A Grnewald, G A B Davies-Jones was the first systematic review of the
................................................................................... subject after the introduction of diagnos-
tic criteria for CD. Ten of these patients
From gut to brain had a severe progressive neuropathy. All
patients had gait ataxia and some had
limb ataxia. Neuropathological data

I
t has taken nearly 2000 years to appre- this disease was the gut. The first report from postmortem examinations showed
ciate that a common dietary protein of neurological manifestations associ- extensive perivascular inflammatory
introduced to the human diet rela- ated with CD was by Carnegie Brown in changes affecting both the central and
tively late in evolutionary terms (some 1908.3 In his book entitled Sprue and its peripheral nervous systems. A striking
10 000 years ago), can produce human treatment he mentioned two of his pa- feature was the loss of Purkinje cells
disease not only of the gut but also the tients who developed peripheral neuri- with atrophy and gliosis of the cerebel-
skin and the nervous system. The pro- tis. Elders reported the association
lum. All 16 patients had evidence of
tean neurological manifestations of glu- between sprue and ataxia in 1925.4 The
severe malabsorption as evidenced by
ten sensitivity can occur without gut validity of these and other such reports
anaemia and vitamin deficiencies as well
involvement and neurologists must before 1960 remains doubtful given that
as profound weight loss.
therefore become familiar with the com- a precise diagnosis of CD was not possi-
Several case reports followed, prima-
mon neurological presentations and ble before the introduction of small
rily based on patients with established
means of diagnosis of this disease. bowel biopsies.
The treatment of CD remained empiri- CD, often with persisting troublesome
cal until 194050 when the Dutch paedi- gastrointestinal symptoms followed by
COELIAC DISEASE THROUGH THE neurological dysfunction. Data from pa-
atritian Willem Dicke noted the deleteri-
AGES: FROM GUT TO SKIN ous effect of wheat flour on children with tients with CD presenting with gastro-
CD.5 Removal of dietary products contain- intestinal symptoms followed up in a
. . .the stomach being the ing wheat was shown to result in com- gastrointestinal clinic suggest that
otherwise unexplained neurological dys-
digestive organ, labours in plete resolution of the gastrointestinal
function is a complication in 6% to 10%
symptoms and a resumption of normal
digestion, when diarrhoea health. The introduction of the small of cases.9
seizes the patient . . .and if bowel biopsy in 195060 confirmed the
in addition, the patients gut as a target organ. The characteristic
features of villous atrophy, crypt hyperpla- Table 1 Neurology of coeliac
general system be sia and increase in intraepithelial lym- disease (based on a review of 35
debilitated by atrophy of phocytes with improvement while on papers of single or multiple case
gluten-free diet, became the mainstays of reports from 1964 to 2000)
the body, the coeliac the diagnosis of CD. In 1961 Taylor
disease of a chronic nature published an immunological study of
Total number of patients
Male to female ratio
83
44:39
is formed.1 CD.6 In his paper he commented that Mean age 48
. . .an obstacle to the acceptance of the Neurological diagnosis
immunological theory of causation has Ataxia 29
This extract is from the book on been the lack of satisfactory demonstra- Peripheral neuropathy 29
chronic diseases by Aretaeus the Cappa- Myopathy 13
tion of antibodies to the protein con-
docian, one of the most distinguished Ataxia with myoclonus 9
cerned. He went on to demonstrate the Myelopathy 4
ancient Greek doctors of the first century presence of circulating antibodies against Dementia (usually with 6
AD. This chapter, entitled on the coeliac gliadin (antigliadin antibodies), the pro- additional features)
diathesis, was the first description of tein responsible for CD. This provided fur-
coeliac disease (from the greek word ther evidence that CD was immunologi-
meaning abdominal). Are- cally mediated and that the immune
taeus books were first published in Latin response is not confined to the mucosa of
in 1500 and the new Latin word coeliac the small bowel. Antigliadin antibodies A review of all such reports (with
was used to translate . Coeliac became a useful screening tool for the biopsy proved CD) from 1964 to date
disease (CD) remained obscure until diagnosis of CD. shows that ataxia and peripheral neu-
1887 when Samuel Gee gave a lecture In 1966, Marks et al demonstrated an ropathy are the commonest neurological
entitled On the coeliac affection2 at the Hos- enteropathy in nine of 12 patients with manifestations seen in patients with
pital for Sick Children, Great Ormond dermatitis herpetiformis,7 an itchy ve- established CD (table 1). Less common
Street, London. In it he acknowledged sicular skin rash mainly occurring over manifestations include inflammatory
Areteaus contribution and went on to the extensor aspect of the elbows and myopathies10 and myoclonic ataxia.11 Iso-
give an accurate description of CD based knees. The enteropathy had a striking lated dementia is uncommon and most
on his own clinical observations. similarity to that seen in CD. It was later cases tend to have additional neurologi-
With clinical manifestations primarily shown that the enteropathy and the skin cal features (for example, ataxia or neu-
confined to the gastrointestinal tract or rash were gluten dependent but skin ropathy). Patients with epilepsy associ-
attributable to malabsorption, it was involvement could occur even without ated with occipital calcifications on CT
logical to assume that the target organ histological evidence of gut involvement. and CD have been described,12 mainly in
and hence the key to the pathogenesis of This was the first evidence that the gut Italy. Most present with epilepsy in

www.jnnp.com
 Downloaded from http://jnnp.bmj.com/ on October 22, 2017 - Published by group.bmj.com

EDITORIAL 561

with neurological dysfunction of un-

Table 2 Neurology of gluten sensitivity known aetiology. Our original study con-
Patients seen in the gluten sensitivity/neurology clinic at the Royal Hallamshire 131 cluded that gluten sensitivity played an
Hospital, Sheffield over the past 8 years important part in neurological illness.17
Ataxia (four patients with myoclonus) 56 The evidence was statistical: Patients
Sensorimotor axonal neuropathy 26
with neurological disease of unknown
Mononeuropathy multiplex 15
Motor neuropathy (three MND- like picture on NCS/EMG) 10 aetiology were found to have a much
Small fibre neuropathy 4 higher prevalence of circulating antiglia-
Mixed demyelinating/axonal neuropathy 2 din antibodies (57%) in their blood than
Myopathies 8 either healthy control subjects (12%) or
Abnormal white matter on MRI (with episodic headache) 19 those with neurological disorders of
Stiff-man syndrome 4
known aetiology (5%). Since then we
Neuromyotonia 1
have identified 131 patients with gluten
sensitivity and neurological disorders of
unknown aetiology. Table 2 shows the
neurological diagnoses we have encoun-
childhood. Such cases are rare in the finding is said to be a marker of potential tered. Perhaps not surprisingly the com-
United Kingdom. CD.16 This procedure is only available in a monest manifestations are ataxia (also
very few pathology laboratories, render- known as gluten ataxia18) and peripheral
RECENT ADVANCES: ing its use limited. neuropathy.19
PREVALENCE, SMALL BOWEL Finally, CD has a very strong associ-
GLUTEN ATAXIA
HISTOLOGY AND GENETIC ation with the human lymphocyte anti-
gen (HLA) of the major histocompatibil- Systematic screening of 143 patients
SUSCEPTIBILITY with so-called idiopathic sporadic
Some studies looking at normal popula- ity complex. Ninety per cent of patients
ataxia showed that 41% had gluten
tions have shown that the prevalence of with CD have the HLA DQ2; the rest have
sensitivity as defined by the presence of
CD is much higher than previously DQ8.
circulating antigliadin antibodies20 (IgG
thought13 14 (approximating to 1 in 100). These advances suggest that gastro-
with or without IgA). The prevalence of
Most of such patients have no gastro- intestinal symptoms are absent in most
antigliadin antibodies in 51 patients
intestinal symptoms. In addition, experi- patients with CD, that the definition of
with familial ataxia did not differ from
mental data in patients with gluten sen- gluten sensitivity can no longer be solely
that found in normal healthy control
sitivity suggest that there is a range of based on the presence of an enteropathy subjects (13%). The mean age of onset of
mucosal abnormalities affecting the and that genetic susceptibility may be an the ataxia was 54 but we have recently
small bowel ranging from preinfiltrative important additional marker for gluten seen three patients with early onset
(histologically normal) to infiltrative, to sensitivity. Given the knowledge of these (under 20 years of age) sporadic idio-
hyperplastic to flat destructive (seen in advances and approaching gluten sensi- pathic ataxia and gluten sensitivity.
CD), and finally to the irreversible hypo- tivity from a neurological perspective we Recently four patients have been de-
plastic atrophic lesions.15 Increasing the set up to address the following question: scribed with CD presenting as gait
gluten load may result in progression of Does cryptic gluten sensitivity play a part disturbance and ataxia in infancy.21
the severity of the lesion. In those in neurological illness? Alhough the ataxia tends to be slowly
patients where the histology is normal, progressive, in some cases it can take a
staining of the T cell subpopulations of very rapid course with the development
the intraepithelium of the small bowel THE NEUROLOGY OF GLUTEN of cerebellar atrophy within a year of the
biopsies shows alteration of T cell sub- SENSITIVITY onset of the illness (fig 1). Ataxia and
populations of the intraepithelial lym- Over the past 8 years we have used myoclonus is a much less common pres-
phocytes (increase of the / T cells). This antigliadin antibodies to screen patients entation (only four patients in these

Figure 1 Brain MRI of a patient with gluten ataxia showing rapid onset of cerebellar atrophy over a period of 15 months before the
diagnosis of gluten ataxia.

www.jnnp.com
 Downloaded from http://jnnp.bmj.com/ on October 22, 2017 - Published by group.bmj.com

562 EDITORIAL

series). We have encountered two pa- association of CD with other auto- What do I do with a patient with
tients who in addition to ataxia had evi- immune diseases23 (for example, positive anti-gliadin antibody test
dence of chorea but normal genetic test- diabetes, thyroid disease). This may but normal duodenal biopsy
ing for Huntingtons disease. Gluten account for the finding of the presence of Only one third of the patients with
ataxia primarily affects the lower limbs antigliadin antibodies in all four of our neurological disorders associated with
and gait. Extrapyramidal or autonomic patients with stiff-person syndrome. gluten sensitivity have villous atrophy on
features are rarely apparent and these Some researchers think that pro- duodenal biopsy. Even some with bio-
features distinguish it from the cerebel- longed exposure to gluten in a gluten chemical markers of malabsorption such
lar variant of multisystem atrophy sensitive person may be the trigger for as low serum vitamin B12, low red cell
(MSA). Screening of patients with clini- the development of other autoimmune folate, or vitamin D concentrations had
cally probable MSA (cerebellar variant) disease.23 normal conventional duodenal
for the presence of antigliadin antibodies histology.17 These cases may illustrate the
showed the prevalence to be similar to CONTENTIOUS ISSUES patchy nature of bowel involvement in
the normal population. Brain MRI usu- But antigliadin antibodies lack coeliac disease and the inaccurate
ally shows cerebellar atrophy; sometimes specificity interpretation of duodenal biopsies by
with evidence of white matter abnor- IgG anti-gliadin antibodies have been inexperienced histopathologists. Pre-
malities. Up to 40% of patients also have the best diagnostic marker in the neuro- liminary data based on staining of the
a sensorimotor axonal peripheral neu- logical population we have studied. IgG subpopulation of T cells in the small
ropathy that can often be subclinical. In anti-gliadin antibodies have a very high bowel epithelium suggests that these
a few cases oligoclonal bands are present sensitivity for CD but they are said to patients have potential CD.24 There are,
in the CSF. lack specificity. In the context of a range however, patients where the immuno-
of mucosal abnormalities and the con- logical disorder is primarily directed at
cept of potential CD, they may be the the nervous system with little or no
PERIPHERAL NEUROPATHY damage to the gut. Our practice is to offer
Peripheral neuropathy is the second only available immunological marker for
the whole range of gluten sensitivity of a gluten-free diet to these patients unless
commonest manifestation of gluten sen- the HLA genotype is not consistent with
sitivity. Prospective screening of 101 which CD is only a part. Further support
for our contention comes from our HLA susceptibility to gluten intolerance (that
patients with idiopathic peripheral neu- is, other than HLA DQ2, DQ8, or DQ1).
ropathy has shown the prevalence of studies. Within the group of patients
All patients are followed up and any
gluten sensitivity to be 40% (unpub- with neurological disease and gluten
clinical response is documented.
lished data). The commonest type of sensitivity (defined by the presence of
peripheral neuropathy we encountered is anti-gliadin antibodies) we have found a But my patient has not responded
sensorimotor axonal (26) followed by similar HLA association to that seen in to a gluten free diet
mononeuropathy multiplex (15), pure patients with CD: 70% of patients have Reports in the literature of the effect of
motor neuropathy (10), small fibre neu- the HLA DQ2 (30% in the general popu- the gluten-free diet on neurological dys-
ropathy (four) and mixed axonal and lation), 9% have the HLA DQ8, and the function are conflicting. Almost all pa-
demyelinating (two). The neuropathy is remainder have HLA DQ1. The finding of tients reported in the literature have the
usually chronic and of gradual progres- an additional HLA marker (DQ1) seen in diagnosis of CD before the development
sion. Patients with a pure motor neu- the remaining 20% of our patients may of neurological dysfunction. They may
ropathy may progress to involvement of represent an important difference be- represent a different group of patients
sensory fibres. tween the genetic susceptibility of pa- from those presenting with neurological
tients with neurological presentation to dysfunction without bowel involvement.
those with gastrointestinal presentation Additionally, improvement of gastro-
LESS COMMON NEUROLOGICAL within the range of gluten sensitivity. intestinal symptoms and improvement
MANIFESTATIONS of the histological abnormalities on
We encountered eight patients with But antigliadin antibodies have repeat small bowel biopsy often were the
myopathy and gluten sensitivity. Three been superseded by measures used to assess the response to
had an additional neuropathy. Muscle anti-endomysial and the diet. Serological evidence of response
biopsy showed inflammatory changes in transglutaminase antibodies (for example, sustained elimination of
five. One patient had evidence of inclu- The introduction of more CD specific antigliadin antibodies) has rarely been
sion body myositis. One patient had low serological markers such as anti- used as confirmation of strict adherence
concentrations of vitamin D and a endomysium and more recently trans- to the gluten-free diet. Incomplete elimi-
predominantly proximal myopathy. glutaminase antibodies may have helped nation of gluten from the diet may be
Myelopathy is rare in our series and in diagnosing CD but their sensitivity as enough to abolish gastrointestinal symp-
was only seen in two patients. markers of other manifestations of glu- toms with recovery of the small bowel
We have recently identified a subgroup ten sensitivity (where the bowel is not mucosa but is insufficient to arrest the
of patients with gluten sensitivity who affected) is low. This certainly reflects state of heightened immunological re-
complained of episodic severe headache our experience with patients with gluten sponsiveness resulting in neuronal in-
often with transient neurological deficit sensitivity who present with neurologi- jury. There is a group of patients with CD
and extensive white matter abnormali- cal dysfunction. Endomysium and trans- resistant to gluten-free diet. This may
ties on MRI.22 Some of them also had glutaminase antibodies are only positive reflect hypersensitivity to the minute
ataxia or neuropathy. Their headache in the majority but not in all patients amounts of gluten present in most
resolved with the introduction of a who have an enteropathy. Patients with gluten-free products. An analogous
gluten free diet though the MRI abnor- an enteropathy represent only a third of situation may exist in cases of gluten
malities persisted at least for the short patients with neurological manifesta- ataxia or gluten related neuropathy. The
follow up period. We have also found a tions and gluten sensitivity. Antigliadin monitoring of neurological improvement
higher incidence of gluten sensitivity in antibodies unlike endomysium and in such cases is made difficult by the
patients with systemic vasculitis and transglutaminase antibodies are not au- slow and sometimes incomplete regen-
neurological involvement, perhaps re- toantibodies. They are antibodies against eration of the nervous system. In cases of
flecting the autoimmune nature of glu- the protein responsible for gluten sensi- gluten ataxia where the underlying
ten sensitivity. There is a well known tivity. pathology is loss of Purkinje cells, one

www.jnnp.com
 Downloaded from http://jnnp.bmj.com/ on October 22, 2017 - Published by group.bmj.com

EDITORIAL 563

may only expect the stabilisation of the diet. Early diagnosis and removal of the 10 Henriksson K G, Hallert C, Norrby K, et al.
Polymyositis and adult coeliac disease. Acta
disease without any definite clinical trigger factor by the introduction of Neurol Scand 1982;65:30119.
improvement. This is in marked contrast gluten-free diet is a promising therapeu- 11 Lu C S, Thompson PD, Quin NP, et al.
to the response seen in patients with tic intervention. IgG antigliadin antibod- Ramsay Hunt syndrome and coeliac disease:
a new association. Mov Disord
florid gastrointestinal symptoms who ies should be part of the routine investi- 1986;1:20919.
notice almost immediate improvement gation of all patients with neurological 12 Gobbi G, Bouquet F, Greco L, et al. Coeliac
after the introduction of a gluten-free dysfunction of obscure aetiology, par- disease, epilepsy, and cerebral calcifications.
Lancet 1992;340:43943.
diet. ticularly patients with ataxia and periph- 13 Grodzinsky E, Franzen L, Hed J, et al. High
eral neuropathy. prevalence of celiac disease in healthy adults
Isnt the neurological damage revealed by antigliadin antibodies. Annals of
Allergy 1992;69:669.
nutritional? ACKNOWLEDGEMENTS 14 Catassi C, Ratsch IM, Fabiani E, et al.
Nutrient deficiencies (B12, folate, vita- We are currently conducting a trial of the Coeliac disease in the year 2000: exploring
min D, vitamin E) are rare in this neuro- effect of gluten-free diet in patients with glu- the iceberg. Lancet 1994;343:2003.
15 Marsh MN. The natural history of gluten
logical population. Given that two thirds ten ataxia and would welcome referrals of sensitivity: defining, refining, and re-defining.
of these patients have no enteropathy patients with sporadic idiopathic ataxia. We Q J Med 1995;85:913.
this is hardly surprising. The concept of have received funds from SHS International 16 Maki M, Holm K, Collin P, et al. Increase in
and Ultrapharm Ltd towards a pilot study of gamma/delta T-cell receptor bearing
the neurological manifestations being lymphocytes in normal small bowel mucosa in
the effect of gluten-free diet on patients with
nutritional in origin is now outmoded. latent coeliac disease. Gut
neurological dysfunction and gluten sensitiv- 1991;32:141214.
Intestinal mucosal damage in coeliac ity and a grant from The Friedreichs Ataxia 17 Hadjivassiliou M, Gibson A, Davies-Jones G
disease is the result of both humoral and Group for research into the immunological A B, et al. Is cryptic gluten sensitivity an
T cell mediated inflammation. Such mechanisms of the pathogenesis of gluten important cause of neurological illness? Lancet
ataxia. 1996;347:36971.
inflammation is not, however, confined 18 Hadjivassiliou M, Grnewald RA,
to the gut, as activated HLA restricted J Neurol Neurosurg Psychiatry Chattopadhyay AK, et al. Clinical,
gliadin specific T cells25 and antigliadin 2002;72:560563 radiological, neurophysiological and
neuropathological characteristics of gluten
antibodies are found systemically. An- ataxia. Lancet 1998;352:15825.
tigliadin antibodies are also found in the ..................... 19 Hadjivassiliou M, Chattopadhyay AK,
CSF.26 Postmortem findings from two of Davies-Jones GAB, et al. Neuromuscular
Authors affiliations disorder as a presenting feature of coeliac
our patients with gluten ataxia has M Hadjivassiliou, R A Grnewald, G A B disease. J Neurol Neurosurg Psychiatry
shown perivascular cuffing with both Davies-Jones, Department of Neurology, The 1997;63:7705.
Royal Hallamshire Hospital, Glossop Road, 20 Hadjivassiliou M, Grnewald RA,
CD4 and CD8 cells. This inflammation Sheffield, S10 2JF, UK Davies-Jones GAB. Causes of cerebellar
was primarily seen in the white matter of degeneration: gluten ataxia in perspective. J
the cerebellum. There was also marked Correspondence to: Dr M Hadjivassiliou, Neurol Sci 2001;187(suppl 1):S520.
Department of Neurology, The Royal 21 Hahn JS, Sum JM, Crowley RS, et al. Coeliac
but patchy Purkinje cell loss. We have disease presenting as gait disturbance and
Hallamshire Hospital, Glossop Road, Sheffield,
also found antibodies against Purkinje S10 2JF, UK; [email protected] ataxia in infancy. J Child Neurol
cells in patients with gluten ataxia. Our 1998;13:3513.
22 Hadjivassiliou M, Grnewald RAG, Lawden
research suggests that IgG antigliadin REFERENCES M, et al. Headache and CNS white matter
antibodies cross react with epitopes on 1 Aretaeus. Liber IV e e. abnormalities associated with gluten
Purkinje cells from human cerebellum.27 In: Corpus Medicorum Graecorum. Berlin: sensitivity. Neurology 2001;56:3858.
Akademie-Verlag GmbH 1956:74. 23 Ventura A. Coeliac disease and
Characterisation of the anti-Purkinje cell autoimmunity. In: Lohiniemi S, Collin P, Maki
2 Gee S. On the coeliac affection. St
antibodies by immunoblotting may pro- Bartholomews Hospital reports M, eds. Changing features of Coeliac
vide a useful marker for the diagnosis of 1888;24:1720. disease. Tampere: The Finnish Coeliac Society
3 Brown C W. Sprue and its treatment. London: 1998:6772.
gluten ataxia in a manner analogous to 24 Hadjivassiliou M, Grnewald RA, Davies
J Bale, sons, and Danielson, 1908.
the use of antiendomysium antibodies as 4 Elders C. Tropical sprue and pernicious Jones A, et al. Gluten ataxia insights into the
a marker for coeliac disease or the anaemia, aetiology and treatment. Lancet pathophysiology. J Neurol Neurosurg
Psychiatry 1999;67:262.
anti-Yo antibody in paraneoplastic cer- 1925;i:757.
25 Jensen K, Sollid LM, Scott H, et al.
5 Dicke W K, Weijers HA, Van De Kamer JH.
ebellar degeneration. Gliadin-specific T cell responses in peripheral
Coeliac disease II; the presence in wheat of a
blood of healthy individuals involve T cell
factor having a deleterious effect in cases of clones restricted by the coeliac disease
FROM GUT TO BRAIN coeliac disease. Acta Paediatrica associated DQ heterodimer. Scand J Immunol
1953;42:3442. 1995;42:16670.
Gluten sensitivity is best defined as a 6 Taylor KB, Truelove SC, Thomson DL, et al. 26 Chinnery PF, Reading PJ, Milne D, et al. CSF
state of heightened immunological re- An immunological study of coeliac disease antigliadin antibodies and the Ramsay Hunt
sponsiveness in genetically susceptible and idiopathic steatorrhoea. BMJ syndrome. Neurology 1997;49:11313.
1961;2:172731. 27 Hadjivassiliou M, Grnewald RA,
people.15 This definition does not imply 7 Marks J, Shuster S, Watson AJ. Small bowel Davies-Jones GAB, et al. The humoral
bowel involvement. That gluten sensitiv- changes in dermatitis herpetiformis. Lancet response in the pathogenesis of gluten ataxia.
ity is regarded as principally a disease of 1966;ii:12802. Ann Neurol 2001;50(suppl 1):S37.
8 Cooke WT, Thomas-Smith W. Neurological 28 Hadjivassiliou M. On the neurological
the small bowel is a historical disorders associated with adult coeliac manifestations of gluten sensitivity [MD thesis].
misconception.28 Gluten sensitivity can disease. Brain 1966;89:683722. Sheffield: Department of Health Sciences,
be primarily and at times exclusively a 9 Holmes GKT. Neurological and psychiatric University of Sheffield, 1999.
neurological disease.29 The absence of an complications in coeliac disease. In: Gobbi G, 29 Hadjivassiliou M, Grnewald RA,
Anderman F, Naccarato S, et al, eds. Davies-Jones GAB. Gluten sensitivity: a
enteropathy should not preclude pa- Epilepsy and other neurological disorders in many-headed hydra [editorial]. BMJ
tients from treatment with a gluten-free coeliac disease. London: John Libbey, 1997. 1999;318:171011.

www.jnnp.com
 Downloaded from http://jnnp.bmj.com/ on October 22, 2017 - Published by group.bmj.com

Gluten sensitivity as a neurological illness

M Hadjivassiliou, R A Grnewald and G A B Davies-Jones

J Neurol Neurosurg Psychiatry 2002 72: 560-563

doi: 10.1136/jnnp.72.5.560

Updated information and services can be found at:

http://jnnp.bmj.com/content/72/5/560

These include:

References This article cites 22 articles, 5 of which you can access for free at:
http://jnnp.bmj.com/content/72/5/560#BIBL

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

Notes

To request permissions go to:

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

To order reprints go to:

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

To subscribe to BMJ go to:

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