10.1192/bjp.162.5.593Access the most recent version at doi:
1993 162: 593-596 The British Journal of Psychiatry
V Eapen, DL Pauls and MM Robertson
syndrome. United Kingdom cohort study
Evidence for autosomal dominant transmission in Tourette's
Article cited in:
To obtain reprints or permission to reproduce material from this paper, please write
to this article at
You can respond
top right corner of the article or
Receive free email alerts when new articles cite this article - sign up in the box at the
The Royal College of PsychiatristsPublished by
on July 14, 2011 bjp.rcpsych.org
go to: The British Journal of PsychiatryTo subscribe to
British Journal of Psychiatry (1993), 162, 593â€”596
Tourette's syndrome (TS)ischaracterised by multiple
motor and one or more vocal tics of more than a
year's duration and an age of onset before 21 years
(American Psychiatric Association,
family studies suggested that a single major gene
confers susceptibility for TS (Baron eta!, 1981;Kidd
& Pauls, 1982; Comings et a!, 1984; Devor et a!,
1984;Price eta!, 1984;Curtis eta!, 1992).However,
fmdings from these studies were inconsistent as to the
precise mode of inheritance. Some suggested auto
somal dominant inheritance with reduced penetrance,
others were consistent with polygenic inheritance,
while others suggested that the genetic model was
additive. Pauls& Leckman
complex segregation analysis on a sample of 27
families with TS sufferers in which all available
relatives were personally interviewed. These investi
gators reported that an autosomal dominant model
best described the transmission. Subsequent analyses
of large kindreds with TS (Pauls et a!, 1990; Curtis
eta!, 1992) also supported the hypothesis of a major
dominant gene with high penetrance.
However, studies of large kindreds need to be
interpretedwith the knowledge
introduced by the ascertainment of such multiplex
families cannotbe easily incorporated
analyses. A much better test of genetic transmission
can be accomplished with a larger sample of smaller
families that have been ascertained without regard to
familial loading. The present study was undertaken
to examine the inheritance pattern of TS in a UK
cohort ascertained through consecutive admissions
to a TS clinic.
that the bias
Familiesincludedin this study wereascertainedthrough
40 consecutive new cases of TS registered at the National
Hospital for Neurology and Neurosurgery (NHNN), Queen
Square,London, over eight
patients were seen in this time, but their families were not
included in this study as information was not available on
the biological relatives. In all 40 families, direct clinical
interviews were conducted with the index case (by MMR)
andall living relatives(VE). Datawereobtainedabout all
first-degree relatives from each proband
relatives) duringtheinitial interviewattheclinic.Thesedata
were included in the final diagnosticestimatesfor each
relative.All subjectswereinterviewedto assess â€˜¿?caseness'
usingasemistructured interviewbyoneof theauthors(yE)
who iswell acquaintedwith theinterviewschedulewidely
used(Robertsoneta!, 1988;Robertson& Gourdie, 1990).
Complexsegregation analyses werecompletedusingthe
unified model asimplementedin the computer program
POINTER(Lalouelet a!, 1983).
POINTERisthe nuclear family.
families were broken down into smaller units. The 40
(i.e. affected probands)to nuclearfamilies indicate how
each family is related to the proband's nuclear family
(Lalouel & Morton, 1981). The unified model as incorporated
in POINTER hasfive main parameters:q, the frequencyof
theputativemajor gene;d, thedegreeof dominanceof the
putative major gene;h2,the heritability of the polygenic
componentimportant for the expressionof the disorder;
t, the major-gene effect measuredasthe distancebetween
two homozygotes;and tau, the probability that the risk
allele will be transmitted by the heterozygousgenotype.
Segregationanalyseswerecarriedout for five different
(a) TS only
(b) TS or chronic multiple tics (CMT)
(c) TS, CMT or transient tic disorder (TTD)
(d) TS or obsessiveâ€”compulsive
(e) TS, CMT, TTD or OCB.
OCBwasincludedin thediagnosticscheme asprevious
studieshavesuggested an associationbetweenit and TS
eta!, 1982;Neeeta!, 1982;Frankeleta!, 1986;Paulset
United Kingdom Cohort Study
V. EAPEN, D. L. PAULS and M. M. ROBERTSON
Complex segregationanalyseswere performed on families ascertainedthrough 40 unselected
consecutivepatients with Tourette's syndrometo examinethe hypothesisthat its transmission
is consistent with genetic inheritance. Analyses were done using several diagnostic
classifications. All results were consistent with an autosomal dominant gene with high
penetrance. The penetrances rangedfrom 0.882 to 1.000 for males and 0.452 to 0.980 for
females, depending upon the specific classification scheme incorporated into the analyses.
no. (%)TSITi no.cs/OCB
EAPEN ET AL
Ratesof TS, tics and OCBamongfirst-degree relatives of probands
a!, 1986; Robertson
age and sex. To incorporate these differences into the
analyses, separate estimates of prevalence were made. For
the first three diagnostic schemes, four fairly narrow age
classes (0â€”5,6â€”10,11â€”15, and over 15 years) were
incorporated. For the analysis that included OCB, four
somewhat broader age classes (0â€”15, 16â€”25, 26â€”35, and
over 35 years) were used. Furthermore,
carried out using a wide range
from 0.00032 to 0.001 for TS only; from 0.005 to 0.030 for
TS or CMT; and from 0.008 to 0.05 forTS or CMT/TTD;
when OCB was included in the analyses, the overall preva
lences ranged from 0.003 to 0.05. The penetrance results
are reported only for the accepted prevalence rate (0.5 per
1000;Bruun, 1984),because parameter estimates with the
other valuesproved verysimilar. Becauseascertainment was
through consecutive cases, it is likely that the probability
of any person with TS in the UK being a proband was quite
small. Thus, an ascertainment probability of i = 0.01 was
incorporated into the analyses.
et a!, 1988; Robertson& Gourdie,
of TS, CMT and OCB differ with both
and sex of age-specific
Of 168relatives included in the study (Table 1),30(17.9Â°lo)
had TS, 21 (12.5%) CMT, and 10(6%) OCB. Consistent
with other reports (Pauls eta!, 1991),malesweremore likely
to have TS or CMT, while females were more likely to
receive a diagnosis of OCB.
A wide range of genetic models was examined in a
hierarchical fashion. Firstly, the model of no transmission
postulates that there is a gene of major effect against a
polygenic background that contributes to the manifestation
of the disorder. Since there was evidence for vertical
transmission in these families (i.e. the model of no trans
mission could be rejected), additional analyses were done
to test specific genetichypotheses
the five diagnostic schemes there was evidence that the
transmission was consistent with a hypothesis of single-locus
transmission. Furthermore, the generalisedsingle-locusmodel
converged to the dominant model for all diagnostic hier
archies. The mixed model converged at the boundary with
polygenic heritability (h2) being zero, and the parameter
estimates were essentially identical to the best-fitting
Mendeian major-locus model (d= 1, t = 5.36, q = 00002,
h2= 0; and 0% phenocopies
Furthermore, the mixed model resulted in a significantly
better likelihood when compared
hypothesis, which therefore could be rejected. There
was no evidence to suggest non-Mendelian transmission
probabilities. The most parsimonious Mendeian single-locus
model was the autosomal dominant model; the estimates
of penetrance for this model were 0.966 males and 0.452
The analyses with the other diagnostic schemes also
suggest autosomal dominant transmission. The analysis
when subjects with TS or CMT were included gave higher
penetrance figures for males (0.999) and females (0.554).
When the defmition of â€˜¿?affected'
TSor OCB,the penetranceestimatedwas0.882for males
and females. When relatives with TS, tics (CMT/TTD)
sexes (Table 3).
with the mixed model.The mixed model
for malesand females).
with the polygenic
status included those with
Complexsegregationanalysison 49 nuclearfamilies (Tourette's syndromeonly)
Consistent indicates that model of transmission cannot be rejected at P< 0.05.
AUTOSOMAL DOMINANT TRANSMISSION OF TOURETFE'S SYNDROME
Geneticmodelestimatesfor maleand female subjectsaccordingto diagnostic schemes
p2,p1 andp0 denotethepenetrance forgenotype withtwo susceptibility alleles(as),onesusceptibility allele(As),andnosusceptibility
allele (AA), respectively, and q the frequency of the susceptibilityallele â€˜¿?a'.
1. Values from Bruun (1984).
Goodness-of-fittests(xi) werecarriedout for all solutions
obtained for all diagnostic schemes. First-degree relatives
weregroupedaccordingto sexof theproband, sexof the
relative, and relationship to the proband (parents v.
siblings).Expectedrisksof beingaffectedwith TS, ticsor
OCB were calculated using the parameters of the autosomal
dominant model (the best-fitting model) and compared
with the observed rates for all first-degree relatives.
For the TS-only scheme,the predictedand the observed
frequencieswere not significantly different (x2= 12.4,
d.f.=6, 0.05<P<0.l0). For the TS or CMT group, the
predicted rates were significantly different from the
observed (x@ = 20.33, d.f. = 6, P<0.005).
(x2=21.60,d.f. =6, P<0.005), indicating a poor fit for
the data. In the TS/OCB scheme,x2for goodness-of-fit
was not statistically significant (x2= 3.7934, d.f. = 6,
0.990<P<0.9995), suggesting that OCBisanintegralpart
of thespectrumof expression of TS.Finally, thegoodness
of-fittest for TS, CMT/TTD or OCB gavestatistically
significantvalues(@= 119.465, d.f. =6, P<0.0005), again
indicating a poor fit for the data. These results suggestthat,
within thesefamilies,motor tics(chronicor transient)may
not alwaysbeaetiologically or geneticallyrelatedto TS.
We found 17.9Â°lo of first-degree relatives to have TS.
This is higher than previously reported
may be due in part to referral bias, in that NHNN
is a tertiary centre, and families with more than one
affected member may be more likely to be referred.
It may also be a reflection of a change in trend in the
diagnosis of TS overtime. It should be noted, however,
that eventhough the rates among first-degree relatives
are higher, the distribution of diagnoses is similar
to that in other reports (Pauls eta!, 1991):males are
more likely to exhibit TS or tics, and females
are more likely to manifest OCB.
The results of the study support the hypothesis that
TS is inherited as an autosomal
with high penetrance. Furthermore,
are consistent with the hypothesis that OCB is
part of the spectrum of the syndrome.
fit for the data was obtained when OCB was also
In order to allow comparison with previous studies,
data were analysed using the higher prevalence rates
as assumed by Comings et a! (1984) and Devor
(1984). This did not alter the inferences; all results
remained consistent with autosomal dominant trans
mission. Although segregation analysis suggested an
autosomal dominant mode of transmission for all
diagnostic schemes, the estimated values did not
correspond with the observed rates in the relatives
for those schemes including tic disorders.
difference was most marked in the rates for mothers
of male probands, where the observed rate was much
higher than that expected. The rate for fathers showed
a more slight but similar trend. This should be viewed
in the context of the fact that the sexratio distribution
of affected relatives in our data isdifferent from that
in other published studies, with more females being
affected than expected. Thus the rates do not conform
with the sex ratio of population prevalences used
in the analyses, which may be contributing to the
goodness-of-fit results. However, a more recent
epidemiological study (Apter eta!, 1992)has shown
a male: female ratio of approximately 1.6: 1, which
is similar to our findings.
EAPEN ET AL
It is interesting to note that the estimated values
compared best with the observed values for the TS
and OCB group. This is consistent with a report by
Pauls & Leckman (1986) on an independent
sample suggesting that obsessive-compulsive disorder
(OCD) is aetiologically and genetically related to TS.
On the other hand, our results suggest that not all
chronic tics may be related to TS. When goodness
of-fit werecalculated,theanalyses thatincluded
CMT category resulted in the most significant
differences between observed and expected. This
suggests that either CMT is related to TS and the
genetic model is wrong, or that some individuals with
CMT do not have a disorder that is related to TS,
and that, within these families,
and transient) are phenocopies.
To help understand the relationship between OCB,
CMT and TS, phenomenological
personalinterview to assess different expressions
of the syndrome are indicated. Italsoemerges
that sound epidemiological studies are of crucial
importance at this juncture to examine some of the
issues raised here, particularly
of the prevalence of TS in the general population,
sex ratio, and sex-dependent
expression of the disorder.
motor tics (chronic
that oftrue estimates
Gilles de IaTourette's syndrome. British Journal of Psychiatry,
DEvOR,E. J. (1984) Complex segregation analysis of Gilles de la
Tourette syndrome: further evidence for a major locus of
transmission. American Journal of Human Genetics,36, 704-709.
Fsar@mo, S. 3. M. (1967)Gillesde Iaburette's
on four cases and a review of published case reports. British
Journal of Psychiatry, 113, 607-617.
FRANKEL, M.,CUMMINGS, 3. L.,
Obsessions and compulsions in Gilles de IaTourette's syndrome.
Neurology, 36, 378â€”382
KIDD, K. K. & PAULS, D. L. (1982) Genetic hypothesis for Toureue
syndrome. In Gilles de la Tourette
& A. 3. Friedhoff),pp.243â€”249.
LALOUEL,3. M. & MORTON,N. E. (1981) Complex
analysis with pointers. Human Heredity,
syndrome: a report
ROBERTSON, M.M., et al (1986)
NewYork: Raven Press.
(eds T. N. Chase
â€”¿?, RAO,D. C., MORTON,N. E.,etal (1983)A unifiedmodel
for segregation analysis.AmericanJournal of Human Genetics,
MONTGOMERY, M. A., CLAYTON,P. J. & FRIEDHOFF,A. 3. (1982)
Psychiatric illness in Toureue syndrome patients and first degree
relatives. In Gilles de Ia Tourette Syndrome (eds T. N. Chase
& A. 3. Friedhoff), pp. 335â€”339.New York: Raven Press.
NEE, L. E., POLINSKY,R. J. & EBERT,M. H. (1982) Tourette
syndrome: clinical and family studies. In Gilles de la Tourette
syndrome (eds T. N. Chase & A. J. Friedhoff),
New York: Raven Press.
PAULS,D. L. & LECKMAN, 3. F. (1986)
de Ia Tourerte syndrome and associated behaviors: evidence for
autosomal dominant transmission.
Medicine, 315, 993â€”997.
The inheritanceof Gilles
â€”¿?, TOWBIN,K.E.&LECKMAN, 3.F. (1986)Gilles deIa
Tourette syndrome and obsessive-compulsive disorder: evidence
supportinga genetic relationship.
Psychiatry, 43, 1180â€”1182.
â€”¿?,PAKsTIS,A. 3.,KURLAN, R.,
and related disorders.
linkage analysis of Tourette's
Journal of the American Academy of Child Psychiatry, 29,
RAYMOND, C. L.,STEVENSON, 3. M.,
study of Gilles de la Tourette syndrome. American Journal of
Human Genetics, 48, 154â€”163.
PRICE,R. A.,PAULS,D.L. & CAINE,
segregationanalysis of clinically defined subgroups
syndrome. AmericanJournal of Human
Roeleusou, M. M., Tjustm.E,M. R. & LEES, A. J. (1988)The psycho
pathology of the Gilles de LaTourette
logical analysis. British Journal of Psychiatry, 152, 383-390.
et al(l99l) A family
E.D. (1984) Pedigree
Genetics, 36 (suppl. 4),
â€”¿?&GOURDIE, A. (1990)Familial Tourette's syndrome
large British pedigree. Associated
potential for linkage analysis. British Journal
YARYURA TOBIAs, 3. A.,
Clinicalaspectsof Gillesde IaTourette syndrome. Orthomolecular
Psychiatry, 10, 263â€”268.
NEZIROGLU, F., HOWARD, S., et a! (1981)
Manual of Mental Disorders (3rd edn, revised) (DSMâ€”IIIâ€”R).
Washington, DC: APA.
APTER, A., PAUL5, D.L., BLEICH, A..
based epidemiological study of Tourette's
adolescents in Israel.In Gilesde la Tourettesyndrome(2ndedn)
(eds T. N. Chase, A. J. Friedhoff&
New York: Raven Press.
BARON,M., SHAPIRO,E., SHAPIRO,A., et a! (1981) Genetic analysis
of Tourette syndrome suggesting major gene effect. American
Journal of Human Genetics, 33, 767â€”775.
BRUUN,R. D. (1984) Gilles de Ia Tourette
of clinical experience. Journal of the American Academy of
Child Psychiatry, 23, 126â€”133.
COMINGS, D.E., COMINGS, B.0.,
Detection of major gene for Gilles de Ia burette
American Journal of Human Genetics, 36, 586â€”600.
Autosomal dominant gene transmission in a large kindred with
AssocwrIoN (1987) Diagnosticand Statistical
et al (1992) A population
D. J. Cohen), pp. 61-65.
DEVOR,E.J., eta! (1984)
Valsamma Eapen, MBBS, MRCPsych, Honorary Research Assistant, The National Hospital for Neurology and
Neurosurgery, QueenSquare,London WCJN 3BG, and Lecturer, Academic Department of Psychiatry,
UniversityCollegeLondon Medica!School,MiddlesexHospital, Mortimer Street,London WiN 8AA; David
Pauls, PhD,Associate Professor of Human Genetics, Yale Child Study Centre, 230 South Frontage Road,
P0 Box 3333,New Haven, Connecticut 06510-8009,USA; â€˜¿?Mary
Honorary SeniorLecturerandHonorary ConsultantPsychiatrist, TheNational Hospitalfor Neurologyand
Neurosurgery,QueenSquare,London WCJN3BG, andSeniorLecturer,AcademicDepartmentof Psychiatry,
University CollegeLondon Medical School, Middlesex Hospital, Mortimer Street, London WiN 8 AA
M. Robertson, MBChB, MD,FRCPsych,