1. ¹Department of Psychiatry and Behavioral Neurosciences, McMaster University, Hamilton, ON, Canada
2. ²Department of Psychiatry, Dalhousie University, Halifax, NS, Canada
3. ³Prague Psychiatric Centre, 3rd Medical School, Charles University, Prague, Czech Republic
4. ⁴Department of Psychiatry, McGill University, Montreal, QC, Canada

Correspondence: Dr M Alda, MD, FRCPC, Department of Psychiatry, Dalhousie University, 5909 Veterans' Memorial Lane, Halifax, Nova Scotia, Canada B3H 2E2. E-mail: malda@dal.ca

Received 9 November 2004; Revised 5 May 2005; Accepted 11 May 2005; Published online 21 June 2005.

Abstract

The search for susceptibility genes for bipolar disorder (BD) depends on appropriate definitions of the phenotype. In this paper, we review data on diagnosis and clinical features of BD that could be used in genetic studies to better characterize patients or to define homogeneous subgroups. Clinical symptoms, long-term course, comorbid conditions, and response to prophylactic treatment may define groups associated with more or less specific loci. One such group is characterized by symptoms of psychosis and linkage to 13q and 22q. A second group includes mainly bipolar II patients with comorbid panic disorder, rapid mood switching, and evidence of chromosome 18 linkage. A third group comprises typical BD with an episodic course and favourable response to lithium prophylaxis. Reproducibility of cognitive deficits across studies raises the possibility of using cognitive profiles as endophenotypes of BD, with deficits in verbal explicit memory and executive function commonly reported. Brain imaging provides a more ambiguous data set consistent with heterogeneity of the illness.