KEYWORDS bipolar disorder, bipolar II disorder, depression, lithium, treatment, venlafaxine In a recent article, Amsterdam et al. 1 conducted secondary analyses of study data assessing the efficacy of lithium versus venlafaxine monotherapy over a 12-week period for 129 outpatients meeting criteria for the depressed phase of bipolar II disorder, 2 as well as relapse prevention therapy in a 6-month continuation study of 59 of those subjects. 3 The primary study concluded that short-term venlafaxine may provide effective antidepressant treatment for bipolar II depression without an increase in hypomanic symptoms relative to lithium. While commenting about the limited sample size of the lithium continuation group and lack of power in the continuation study, the authors note that the study did not demonstrate differences in relapse rates between the venlafaxine and lithium prophylactic therapy, nor a difference in treatment-emergent hypomanic episodes. In the current secondary analysis, Amsterdam et al. explored the relationship between the number of prior antidepressant treatment trials, determined retrospectively, and the step-wise increase in loss of prospectively observed response. Conducting regression analysis, they demonstrated that treatment with venlafaxine is predictive of both response or remission than treatment with lithium, consistent with findings in the initial report. Other variables included in the regression model were race, inter-episode recovery, baseline Hamilton Depression Rating Scale score, and number of prior episodes. None of these were predictive of response.When the number of past antidepressant exposures was included, however, a negative relationship also emerged between the number of past antidepressant trials and both response and remission. Although based on naturalistic treatment history ascertained by interview and supplemented with available medical and pharmacy records, the observation none the less has potential clinical relevance and utility as a focus for consideration. Figure 1 in their paper suggests that patients who have had increasing numbers of antidepressant medication trials are less likely to respond to antidepressant treatment, in this case treatment with venlafaxine or lithium. That observation is in concordance with findings from the prospective sequenced treatment alternatives to relieve depression (STAR*D) treatment studies, 4 conducted in contrast in subjects diagnosed with unipolar depression. Those studies reported decreasing rates of remission with increasing number of medication trials as patients prospectively moved further into the study algorithm, specifically 36.8%, 30.6%, 13.7%, and 13.0% with each succeeding trial. Amsterdam et al.'s Figure 1 likewise does not appear linear and suggests that at some stage there may also be a point of diminishing returns when subsequent medication trials are less likely to be effective. The reader is referred back to the published figure for his or her own inspection and interpretation. Amsterdam et al. 1 provide possi...