Our aim was to analyse existing data on the efficacy and tolerability of valproate
for the treatment of acute bipolar depression.
Randomized controlled trials comparing valproate with placebo were identified using
searches of electronic databases in October 2008. Outcomes investigated were depression,
anxiety, hypomania, attrition, and adverse events. Trial quality was assessed, and
data were summarized using meta-analyses.
Four randomized, controlled, doubleblind trials of 142 participants were included.
Trial quality was good, although individual study sample sizes were small. Study duration
was six weeks (2 studies) and eight weeks (2 studies). Meta-analysis showed a significant
difference in favour of valproate for reduction in depressive symptoms, both on depression
symptom scales (standardized mean difference (SMD) -0.35 (95% confidence interval,
-0.69, -0.02)), and participants with at least 50% improvement in symptoms - relative
risk (RR) 2.00 (1.13, 3.53). Effects on anxiety symptoms were small, SMD -0.32 (-0.72,
0.08) and inconclusive (p=0.12). No evidence of a difference in mania symptoms, withdrawal
for any reason, lack of effectiveness or adverse events was detected. Nausea occurred
more frequently with valproate compared with placebo though the difference was not
significant, RR 2.01 (0.98, 4.11). Other adverse events occurring more frequently
with valproate (somnolence, fatigue/muscle weakness, headache, diarrhoea and dry mouth)
did not differ significantly between treatment groups.
Sample sizes were small warranting a larger study to confirm or disprove these findings.
Valproate is effective for the reduction of depressive symptoms of acute bipolar depression,
and was well tolerated.
Copyright 2009 Elsevier B.V. All rights reserved.