 |
 |
|
|
|
Although the nosologic distinction between bipolar disorder (BD) and schizophrenia was a fundamental perspective in psychiatry for nearly a century, current theory and research supports a revised view emphasizing potential areas of overlap between these conditions (for example, 1). Especially prominent have been propositions that there may be shared etiologic risk factors, primarily, genetic susceptibility. Family and twin studies have shown coaggregation for symptoms and disorders (2–4), such as evidence for psychotic symptoms in families selected for the presence of BD (5). In addition, although the precise genetic locations conveying vulnerability for BD and schizophrenia are unknown, current genomic search strategies have identified several potential locations for susceptibility genes conveying risk for both conditions (2–4,6,7). The interest in overlapping genetic susceptibility to BD and schizophrenia has raised the possibility of more homogeneous subtypes of BD that may reflect more clearly the etiologic connections between these disorders. One possibility is that the etiologic overlap is evident for the most severe and chronic manifestations of BD. For example, Valles and others reported increased rates of schizophrenia in families of BD probands with severe illness course, especially among women (8). Similarly, Potash and others found elevated rates of psychotic symptoms in families of BD probands suffering from psychosis, compared with BD probands without psychosis, and suggested a possible BD subtype with psychosis that may be informative for genetic studies (5). A more recent replication study provided perhaps more modest evidence in favour of such a subtype (9). Indeed, molecular studies using high-density pedigrees have found evidence of familial aggregation for both schizophrenia and BD (7). Given the heightened interest in exploring the blurred etiologic boundaries between BD and schizophrenia, we present findings from a family study of BD wherein we focused on the familial aggregation of psychotic symptoms as well as schizophrenia. Although most family studies to date have examined clinical samples from treatment facilities, we focused on BD probands recruited from community consumer groups, which increased our potential to detect familial aggregation of psychotic symptoms and schizophrenia in less severe and chronic manifestations of BD. When recruited, the participants with BD all had at least one child between the ages of 5 and 12 years; 11% of the men and 23% of the women had comorbid lifetime substance use disorders. All participants were French Canadian, a geographic feature of interest because of the group’s genetic homogeneity arising from a founder effect of approximately 350 years (10). In this paper, we focus on 3 critical issues raised by current family studies: 1. We attempt to confirm that schizophrenia and psychotic symptoms aggregate in families of BD probands recruited from community sources rather than treatment facilities, especially in a sample of known genetic homogeneity. 2. We explore the hypothesis raised in family studies that the first-degree relatives of female BD probands have elevated rates of schizophrenia and psychotic symptoms. 3. We also examine whether there is evidence for a clinical subtype of BD associated with familial aggregation for psychotic symptoms. MethodsParticipants Probands. This study is part of a prospective, longitudinal investigation comparing the development of the offspring of parents with BD (BD probands) and those of parents with no mental disorder (NMD probands). We recruited individuals with BD and NMD if they had at least one child between the ages of 5 and 12 years. When the diagnostic status of the parent (either BD or NMD) was confirmed by an experienced clinician trained to administer the Structured Clinical Interview according to the DSM-IV criteria (11,12), we invited the other biological parent of the child to participate in the study. BD probands were recruited from consumer groups, and NMD probands were recruited from parent associations, pediatric services, and newspaper articles. All probands have an IQ above 70 and no chronic physical disease or handicap. We collected all records of psychiatric care and used them to verify and complete information obtained by interview. All study participants provided informed consent. The final sample included 49 BD and 99 NMD probands. Table 1 presents basic descriptive information. As Table 1 shows, the BD group had, on average, fewer years of education, together with increased rates of alcohol abuse or dependence, drug abuse or dependence, and personality disorders.
Informants. All probands reported on their own family. As well, we recruited and interviewed 168 first-degree relatives who knew the family well and agreed to participate. Instruments and Procedure Clinicians blind to group status interviewed all informants individually, either in their home or at the university, using the Family Interview for Genetic Studies (FIGS, 13). The FIGS is currently the most commonly used instrument designed to collect information on past and current mental disorders among family members; it is considered to yield the highest possible reliability and validity for a family history measure (14–16). (Detailed information on the FIGS, including the structured symptom checklists and instructions for administration, may be found at http://zork.wustl.edu/nimh/digs.) The FIGS has, however, been found to underestimate the prevalence of disorders when compared with face-to-face diagnostic interviews (17–19). Informants (7 probands and 50 informants) who could not be directly interviewed were contacted by phone. The interview began with the drawing of a family tree that included first names, birth dates, and cause and date of death (if applicable) for all first-degree relatives and as many second- and third-degree relatives as possible. Next, the interviewers asked general screening questions about the presence of relatives with mental retardation; epilepsy; emotional or nervous disturbances; affective disorders; schizophrenia; antisocial, schizoid, schizotypic, or paranoid traits; problems with alcohol, drugs, or the law; or who had attempted or committed suicide. When the interviewers identified a relative as possibly presenting any of the above disorders or behaviours, they asked the proband or informant to provide more details about the relative’s symptoms and psychosocial functioning, along with information on the frequency and quality of his or her relationship with the relative in question. When possible, a symptom checklist was also completed. Examples of questions from the FIGS Psychosis checklist include the following: “Did he or she ever believe people were following him or her, or that someone was trying to hurt or poison him or her?” “Did he or she ever believe someone was reading his or her mind?” “Did he or she ever believe he or she was under the control of some outside person or power or force?” The information collected from each proband and informant was kept strictly confidential. The senior clinician reviewed the questionnaires and symptom checklists describing each relative and removed any information that could identify the diagnostic status of the probands. Next, a research assistant reviewed all questionnaires for each relative. When a relative was reported to have been diagnosed and (or) treated for a mental disorder, the diagnosis was entered. If one or more checklists were completed but there was no mention of a diagnosis and (or) treatment for a particular disorder, they were reviewed by a psychiatrist who established the diagnoses according to DSM-IV criteria. Statistics We performed standard contingency table analyses to determine group differences in rates of symptoms and disorders. We executed these using the Crosstabs procedure in version 10.1 of the Statistical Package for Social Sciences (SPSS) (20). ResultsRates of Psychotic Symptoms in BD and NMD Probands
Table 2 presents the rates of schizophrenia and psychotic symptoms in the first-degree relatives of BD and NMD probands. The rates of both schizophrenia (c21,1096 = 6.05, P = 0.014) and psychotic symptoms (c21,1096 = 14.67, P < 0.001) were significantly higher in the relatives of BD probands, compared with relatives of NMD probands.
Sex and Rates of Psychotic Symptoms We next explored the hypothesis that the first-degree relatives of female BD probands may have especially elevated rates of psychotic symptoms. Table 2 also presents the rates of schizophrenia and psychotic symptoms in the first-degree relatives of BD probands stratified by proband sex. The first-degree relatives of female BD probands had elevated rates of both schizophrenia (c21,550 = 4.74, P = 0.03) and psychotic symptoms (c21,550 = 11.91, P = 0.001), compared with rates in first-degree relatives of female NMD probands. In contrast, first-degree relatives of male BD probands had elevated rates of psychotic symptoms (c21,544 = 4.07, P = 0.044) but not schizophrenia (c21,544 = 1.73, P = 0.189), compared with relatives of male NMD probands. The raw rates of both schizophrenia and psychotic symptoms were higher in relatives of female BD probands (1.2% and 3.0%, respectively), compared with male BD probands (0.5% and 2.0%, respectively). We next compared rates of schizophrenia and psychotic symptoms in first-degree relatives stratified by relatives’ sex as well as proband sex (also presented in Table 2). The pattern of results for male relatives was similar to the pattern reported above. Male relatives of female BD probands had elevated rates of both schizophrenia (c21,274 = 4.55, P = 0.033) and psychotic symptoms (c21,274 = 9.17, P = 0.002), compared with male relatives of female NMD probands. Male relatives of male BD probands did not have elevated rates of schizophrenia (c2 1,277 = 1.63, P = 0.20) but did have elevated rates of psychotic symptoms (c21,277 = 4.92, P = 0.027). Again, in male relatives of female BD probands, the rates of both schizophrenia (2.4%) and psychotic symptoms (4.7%) were higher than the rates in male relatives of male BD probands (0.9% and 2.8%, respectively) Clinical Characteristics of BD Probands With and Without a Familial Loading for Psychotic Symptoms Our next concern was whether and to what extent the elevated familial loading of schizophrenia and psychotic symptoms described above might reflect a homogeneous subtype of BD. We first wanted to ensure that these elevated rates were not due to a single “heavily loaded” family. Over 6% (3/49) of the BD probands had a first-degree relative with schizophrenia, and over 16% (8/49) of the BD probands had a relative with psychotic symptoms, suggesting that the results were not unduly skewed by an outlier family and that it would thus be worthwhile to explore whether familial loading was attributable to detectable clinical characteristics. We first examined the presence of psychotic symptoms in the BD probands stratified by presence or absence of psychotic symptoms in first-degree relatives. Of the 8 BD probands who had first-degree relatives with psychotic symptoms, 2 (25%) had a personal history of psychotic symptoms. Over 60% of the BD probands without familial loading for psychotic symptoms had a personal history of psychotic symptoms, suggesting a lack of specificity of aggregation in families. We also explored several other clinical characteristics of the BD probands, again stratified by presence or absence of psychotic symptoms in their first-degree relatives. There were no detectable differences in any of these clinical indicators associated with familial aggregation for psychotic symptoms. Although the BD probands with first-degree relatives suffering from psychotic symptoms had a younger age of hospitalization, a higher number of hospitalizations, and a higher rate of suicide attempts, none of the comparisons approached statistical significance, given the sample size of BD probands in this study. DiscussionIn this study, we examined a large sample of first-degree relatives of individuals with BD. We note that this sample is especially interesting because we recruited the BD probands from community sources rather than from treatment facilities. Although they did not constitute a representative community sample, the BD probands studied here were functioning in the community and therefore offered an opportunity to determine familial loading for schizophrenia and psychotic symptoms without introducing the confounds of studying a more homogeneous population such as inpatients (who could overrepresent more severe forms of BD). Further, the participants were all French Canadian, an ethnic group of interest owing to a significant founder effect leading to genetic homogeneity that increases the likelihood of successful genomic search strategies. Another strength of the study is that all NMD participants completed a diagnostic interview covering both Axis I and II disorders. Using our sampling frame, we found strong evidence for familial loading for both schizophrenia and psychotic symptoms in the first-degree relatives of BD probands, compared with a control group with no lifetime history of mental disorders. Our work is congruent with recent family studies showing increased risk of schizophrenia in families of inpatients with BD (8) and aggregation of psychotic symptoms in BD pedigrees (5,9). Moreover, it extends these findings to a sample of probands recruited from nontreatment sources. Taken together, it appears that strong evidence is accumulating from several family studies that an etiologic and familial association indeed exists between BD, schizophrenia, and psychotic symptoms. Our study yielded further evidence that sex may be a critical marker for genetic studies. Familial loading for schizophrenia and psychotic symptoms was elevated in female BD probands, compared with male BD probands. Further, familial aggregation of schizophrenia and psychotic symptoms was observed primarily in male relatives of BD probands, especially in male relatives of female BD probands. These findings support the report by Valles and others, who found a descriptive trend for a higher familial loading for women with BD (8). Taken together, current family studies imply that genetic strategies may be well served by a focus on male relatives of female BD probands in the search for susceptibility loci that confer overlapping risk for BD and schizophrenia. However, we were not able to differentiate BD probands with and without familial loading for psychotic symptoms according to several clinical indicators (including presence of psychotic symptoms in the probands themselves). This contrasts with the results reported by Potash and others (5,9), who suggested the existence of a subtype BD with psychosis suitable for genetic studies. For genomic studies, the search for homogeneous subtypes of disorders is an extremely important topic, and for current research on BD, we suggest simply that more family studies may be needed to confirm the presence of robust subtypes, especially those focused on psychotic symptoms. Larger samples of BD probands may be necessary to detect the descriptive differences that we report as indicators of clinical severity, such as age and number of hospitalizations and suicide attempts. Alternatively, differences among BD patients with and without a family history of psychosis may emerge in the disorder’s later stages, that is, when patients are older than those in this study. Finally, the familial association between BD and psychotic symptoms may be more detectable in samples recruited from inpatient sources rather than from community sources. Overall, our results suggest that sex (that is, female BD probands and their male relatives) is the most robust index to identify more homogeneous subtypes of families for genetic studies, whereas clinically based family studies have provided stronger evidence for subtyping BD according to the presence or absence of psychotic symptoms. Several limitations of this study should be considered. First, although our sample size is not small by typical family-study criteria, it may not be large enough to detect some effects, because we focused on phenotypic symptoms that are relatively rare in the population. The raw frequencies of disorders and symptoms are consequently modest and must be interpreted accordingly. Our study is subject to the limitations inherent in most family studies: diagnostic interviewing was retrospective to gather information on the lifetime history of disorders and symptoms; a prospective design might yield more conclusive findings. This combination of sample size and reliance on retrospective report may underestimate the magnitude of familial associations. Given these limitations, we conclude that our results are in general quite consistent with the current perspectives on the etiologic overlap between BD and schizophrenia, with an emphasis on psychotic symptoms as a key phenotypic indicator that aggregates in families of BD probands. We believe that, because of our study’s community sampling frame, our results offer a necessary complement to prior studies. There have been numerous suggestions that psychotic symptoms may reflect a single portion of a continuum of genetic liability stretching between schizophrenia and BD. Our results suggest a more complicated picture only in the sense that we did not detect a link between the severity of BD and familial aggregation of psychotic symptoms. One possibility is that the postulated genetic liability continuum reflects several susceptibility genes that are nonspecific to either disorder. As suggested by Bramon and Sham (4), genomic searches may require dimensional perspectives on the phenotypic indicators that may reflect the behavioural manifestations of genetic susceptibility to BD and schizophrenia; these could include psychotic symptoms as one key clinical indicator of common genetic vulnerability. Funding and SupportThis work was supported by a Mentored Scientist Research Award (MH15590) to Dr Rende and by a grant from the combined program of the Conseil Québecois de la recherche sociale and the Fonds de la recherche en santé du Québec awarded to Dr Hodgins and Dr Palmour. References1. Torrey EF, Knable MB. Are schizophrenia and bipolar disorder one disease or two? Introduction to the symposium. Schizophr Res 1999;39:93–4. 2. Berrettini WH. Are schizophrenia and bipolar disorders related? A review of family and molecular studies. Biol Psychiatry 2000;48:531–8. 3. Berrettini WH. Genetics of psychiatric disease. Ann Rev Med 2000;51:465–79. 4. Bramon E, Sham PC. The common genetic liability between schizophrenia and bipolar disorder: a review. Curr Psychiatr Rep 2001;3:332–7. 5. Potash JB, Willour VL, Chiu YF, Simpson SG, Mackinnon DF, Pearlson GD, and others. The familial aggregation of psychotic symptoms in bipolar disorder pedigrees. Am J Psychiatry 2001;158:1258–64. 6. Baron M. Genetics of schizophrenia and the new millennium: progress and pitfalls. Am J Hum Genet 2001;68:299–312. 7. Maziade M, Roy MA, Rouillard E, Bissonnette L, Fournier JP, Roy A, and others. A search for specific and common susceptibility loci for schizophrenia and bipolar disorder: a linkage study in 13 target chromosomes. Mol Psychiatry 2001;6:684–93. 8. Valles V, Van Os J, Guillamat R, Gutierrez B, Campillo M, Gento P, and others. Increased morbid risk for schizophrenia in families of in-patients with bipolar illness. Schizophr Res 2000;42:83–90. 9. Potash JB, Chiu Y-F, MacKinnon DF, Miller EB, Simpson SG, McMahon FJ, and others. Familial aggregation of psychotic symptoms in a replication set of 69 bipolar disorder pedigrees. Am J Med Genet (Neuropsychiatr Genetics) 2003;116B:90–7. 10. Maziade M, Fournier A, Phaneuf D, Cliche D, Fournier JP, Roy MA, and others. Chromosome 1q12-q22 linkage results in eastern Quebec families affected by schizophrenia. Am J Med Genet (Neuropsychiatr Genet) 2002;114:51–5. 11. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID). I: history, rationale, and description. Arch Gen Psychiatry 1992;49:624–9. 12. Williams JB, Gibbon M, First MB, Spitzer, RL, Davies M, Borus J, and others. The Structured Clinical Interview for DSM-III-R (SCID). II: multisite test–retest reliability. Arch Gen Psychiatry 1992;49:630–6. 13. Maxwell ME, Manual for the FIGS. Bethesda (MD): Clinical Neurogenetics Branch, National Institute of Mental Health; 1992. 14. Gershon ES, Hamovit J, Guroff JJ, Dibble E, Leckman JF, Sceery W, and others. A family study of schizoaffective, bipolar I, bipolar II, unipolar and normal control probands. Arch Gen Psychiatry 1982;39:1157–67. 15. Grigoroiu-Serbanescu M, Nothen M, Propping P, Poutska F, Magureanu S, Vasilescu R, and others. Clinial evidence for genomic imprinting in bipolar I disorder. Acta Psychiatr Scand 1995;92:365–70. 16. Blehar MC, DePaulo JR, Gershon ES, Reich T, Simpson SG, Nurnberger JI. Women with bipolar disorder: findings from the NIMH genetics initiative sample. Pharmacol Bull 1998;34:239–43. 17. Roy M-A, Walsh D, Kendler KS. Accuracies and inaccuracies of the family history method: a multivariate approach. Acta Psychiatr Scand 1996;93:224–34. 18. Thompson WD, Orvaschel H, Prusoff BA, Kidd KK. An evaluation of the family history method for ascertaining psychiatric disorders. Arch Gen Psychiatry 1982;39:53–8. 19. Andreasen NC, Rice J, Endicott J, Reich T, Coryell W. The family history approach to diagnosis: how useful is it? Arch Gen Psychiatry 1986;43:421–9. 20. SPSS Inc. Statistical package for the social sciences. Version 10.1. Chicago (IL): SPSS Inc; 2001. Author(s)Manuscript received February 2004, revised, and accepted April 2004. 1. Associate Professor, Department of Psychiatry and Human Behavior and Centers for Behavioral and Preventive Medicine, Brown Medical School, Providence, Rhode Island. 2. Professor, Department of Forensic Mental Health Science, Institute of Psychiatry, King’s College, University of London, London, UK. 3. Professor, Department of Psychiatry, McGill University, Montreal, Quebec. 4. Doctoral Student, Department of Psychologie, Université de Montréal, Montreal, Quebec. 5. Statistician, Centre de Recherche Institut Philippe Pinel de Montréal, Montreal, Quebec. Address for correspondence: Dr R Rende, Department of Psychiatry and Human Behavior and Centers for Behavioral and Preventive Medicine, Brown Medical School, Coro Building, Suite 5000, 1 Hoppin Street, Providence RI 02903 e-mail: Richard_Rende@Brown.edu
1 | 2
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||