Abstract
Rationale LQD was an open-label randomised trial comparing lithium versus quetiapine augmentation of antidepressants in treatment-resistant depression (TRD). In the primary analysis, quetiapine was more effective at reducing depressive symptoms. Objectives Conduct a secondary analysis of LQD to examine anxiolysis of Quetiapine compared to Lithium and examine baseline anxiety as a moderator of antidepressant effects. Methods Using linear mixed-models we examined the difference between lithium and quetiapine in anxiety (GAD-7) at 8, 26 and 52 weeks. We modified the main trial’s linear mixed-model with an interaction term for baseline anxiety to estimate the differences in area under the curve (AUC) between treatments for three baseline anxiety levels across the 52 weeks. Results There was little difference between the anxiolysis of quetiapine compared to lithium at 8 (-0.07, p = 0.92), 26 (-0.71, p = 0.30) or 52 weeks (-0.05, p = 0.95). For those with severe baseline anxiety, participants taking quetiapine showed a greater reduction in depressive symptoms than lithium (AUC − 111.6 (95% CI -5.6 to -217.6), p = 0.039). For those with moderate (AUC − 79.7 (95% CI 34.8 to -194.2), p = 0.17) and mild/absent baseline anxiety (AUC 29.7 (95% CI 73.4 to -132.8), p = 0.57), the difference was not statistically significant. Conclusions We did not identify a difference between quetiapine and lithium in anxiolytic effects. The superior antidepressant effects of quetiapine were more apparent in those with higher baseline anxiety and may not be present in those with mild or absent baseline anxiety. Further exploration by pooled analyses of prior trial data is warranted.