Does oral beta-blocker use cause depression or depressed mood?
The purported association between beta-blockers and depression is equivocal and heavily influenced by study design and case definitions, suggesting these drugs may have been unjustly avoided for this reason.
OBJECTIVE: To review the literature regarding the purported association between oral ingestion of beta-blocker drugs and depressed mood. DATA SOURCE: MEDLINE was searched for published articles using the key words propranolol, atenolol, metoprolol, nadolol, timolol, beta-blocker, beta-adrenergic antagonist, or beta-adrenergic blocker in combination with the key words depression, depressive symptomatology, major depressive disorder, or depressed mood from January 1966 through December 1996. DATA SYNTHESIS: Findings regarding the association are equivocal. Plausible explanations include study design, case definition, and confounding disease states. Most of the evidence supporting an association has used case series and case reports. Findings from cross-sectional observational studies and case-control studies are equivocal. Case definition and measurement instruments may partially explain these inconsistencies. Studies using a diagnosis of depression generally do not support the relationship. Trials using depressive symptoms are about evenly split, but they have generally enrolled a small number of patients and have questionable statistical power. Studies defining antidepressant prescriptions dispensed as a marker for depression generally support the association. Evidence exists both for and against the hypothesis that lipophilic beta-blockers cause more depression than do hydrophilic beta-blockers. CONCLUSIONS: beta-Blockers may have been unjustly associated with depression and their use avoided for that reason. Future studies into the association between depression and beta-blocker use should evaluate whether the association is affected by case definition and study design characteristics, including disease, dose-response, bias, measurement error, or ability to precisely measure the length of the exposure.
Ried et al. (1998) studied this question.