Objectives: Reliable predictors of response to lithium are still lacking in bipolar disorders (BDs). However, childhood trauma has been hypothesized to be associated with poor response to lithium.
Methods: We included 148 patients with BD, euthymic when retrospectively and clinically assessed for response to lithium and childhood trauma using reliable scales.
Results: According to the 'Alda scale', the sample consisted in 20.3% of excellent responders, 49.3% of partial responders and 30.4% of non-responders to lithium. A higher level of physical abuse significantly correlated with a lower level of response to lithium (P = 0.009). As compared to patients not exposed to any abuse, patients with at least two trauma abuses (emotional, physical or sexual) were more at risk of belonging to the non-responders group (OR = 4.91 95% CI (1.01-27.02)). Among investigated clinical variables, lifetime presence of mixed episodes and alcohol misuse were associated with non-response to lithium. Multivariate analyses demonstrated that physical abuse and mixed episodes were independently associated with poor response to lithium (P = 0.005 and P = 0.013 respectively).
Conclusions: Childhood physical abuse might be involved in a poor future response to lithium prophylaxis, this effect being independent of the association between clinical expression of BD and poor response to lithium.
Keywords: alcohol misuse; bipolar disorder; childhood trauma; lithium response; mixed episodes; physical abuse.
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.