Clinical and psychopathological features associated with treatment-emergent mania in bipolar-II depressed outpatients exposed to antidepressants

Background: Treatment-emergent affective switch (TEAS), including treatment-emergent mania (TEM), carry significant burden in the clinical management of bipolar depression, whereas the use of antidepressants raises both efficacy, safety and tolerability concerns. The present study assesses the prevalence and clinical correlates of TEM in selected sample of Bipolar Disorder (BD) Type-II (BD-II) acute depression outpatients. Methods: Post-hoc analysis of the clinical and psychopathological features associated with TEM among 91 BD-II depressed outpatients exposed to antidepressants. Results: Second-generation antipsychotics (SGA) (p =.005), lithium (≤.001), cyclothymic/irritable/hyperthymic temperaments (p = ≤.001; p =.001; p =.003, respectively), rapid-cycling (p =.005) and depressive mixed features (p =.003) differed between TEM+ cases vs. TEM− controls. Upon multinomial logistic regression, the accounted psychopathological features correctly classified as much as 88.6% of TEM+ cases (35/91 overall sample, or 38.46% of the sample), yet not statistically significantly [Exp(B) =.032; p = ns]. Specifically, lithium [B = − 2.385; p =.001], SGAs [B = − 2.354; p =.002] predicted lower rates of TEM+ in contrast to the number of lifetime previous psychiatric hospitalizations [B = 2.380; p =.002], whereas mixed features did not [B = 1.267; p = ns]. Limitations: Post-hoc analysis. Lack of systematic pharmacological history record; chance of recall bias and Berkson's biases. Permissive operational criterion for TEM. Relatively small sample size. Conclusions: Cyclothymic temperament and mixed depression discriminated TEM+ between TEM− cases, although only lithium and the SGAs reliably predicted TEM+/− grouping. Larger-sampled/powered longitudinal replication studies are warranted to allow firm conclusions on the matter, ideally contributing to the identification of clear-cut sub-phenotypes of BD towards patient-tailored-pharmacotherapy. © 2018 Elsevier B.V.