Comparative efficacy of mitochondrial agents for bipolar disorder during depressive episodes: a network meta-analysis using frequentist and Bayesian approaches.

Abstract

Mitochondrial dysfunctions have emerged as new biological hypothesis and therapeutic target for bipolar disorder. This network meta-analysis has been done to evaluate the comparative efficacy of mitochondrial agents in bipolar depression. After a comprehensive literature search on PubMed/MEDLINE, Cochrane databases, and International Trials Registry Platform, efficacy data were extracted from 15 randomized controlled trials. Random-effects meta-analysis was done following both frequentist and Bayesian approaches to pool the effects across the interventions. A network graph was built, relative effects of interventions in respect to one another and placebo were calculated, and treatments were ranked as per P- and SUCRA scores. Change in depression rating score was the primary outcome. Data was entered in contrast level and arm level for frequentist and Bayesian approaches, respectively. Amongst mitochondrial agents, N-acetylcysteine (NAC) was shown to have the highest probability of being the best treatment, followed by coenzyme Q10 and combination therapy of alpha-lipoic acid (ALA) and acetyl-L-carnitine (ALCAR) as depicted by P- and SUCRA scores. In the Bayesian approach, none of the treatments had better efficacy than placebo, but in the frequentist approach, NAC (effect estimate: - 1.18 (95% CI: - 2.05; - 0.31)) was significantly better than placebo. Methodically, there may be a difference of magnitude in frequentist and Bayesian approaches, but the direction of effect and ranking probabilities do not differ. We conclude that none of the existing mitochondrial agents showed better efficacy than placebo in bipolar depression regarding depression rating scores.