Different profile of substance abuse in relation to predominant polarity in bipolar disorder: The Vitoria long-term follow-up study.

BACKGROUND: There is a need for comparisons of long-term outcomes in bipolar disorder patients with predominantly manic symptoms vs. predominantly depressive symptoms, especially the course of comorbid alcohol/substance abuse.
METHOD: A naturalistic sample of bipolar I patients (n=120) was followed prospectively for up to 10years. At baseline, number and polarity of past episodes were used to classify patients as predominantly manic or predominantly depressive if there were more manic or more depressive episodes, respectively. 25 patients were excluded from the analyses. Outcomes including episodes, hospitalisations and suicide attempts were recorded at bimonthly visits. Mixed effects models compared the course of alcohol and other substance abuse in predominantly manic vs. depressive patients.
RESULTS: Of the 95 patients analyzed, 44 (46.3%) had predominantly manic episodes and 51 (53.7%) had predominantly depressive episodes. At baseline, the predominantly depressive group had more history of suicide attempts (45.1% vs. 20.5%; p=0.021) and more family history of affective disorders (64.7% vs. 38.6%; p=0.020), but they had fewer previous hospitalisations than the manic group (mean 0.38 vs. 0.50; p=0.025). During the 10-year follow-up, the predominantly depressive group was associated with more episodes (p=0.001), more hospitalisations (p=0.004) and more suicide attempts (p=0.002). At baseline, there were no differences between the manic and depressive groups in the frequency of alcohol abuse (43.2% and 35.3%, p=0.565) or other substance abuse (13.6% and 9.8%, p=0.794). During the 10-year follow-up, the frequency of alcohol and other substance abuse decreased significantly in the manic group only, after controlling by age at onset and civil (marital) status.
CONCLUSION: Long-term clinical outcomes differ between predominantly manic vs. depressive bipolar patients, with the predominantly depressive group having a worse prognosis and maintained alcohol and other substance abuse. These differences should be considered when designing treatment approaches for bipolar patients with comorbid alcohol/substance abuse. Copyright 2009 Elsevier B.V. All rights reserved.