18 - Plasma levels

Plasma levels

290 The Maudsley® Prescribing Guidelines in Psychiatry CHAPTER 2 less effective but better tolerated than lithium.15 Overall, data relating to efficacy in mania are less convincing for valproate than for lithium and a range of antipsychotics.16,17 A meta-­analysis of four small RCTs concluded that valproate is effective in bipolar depression with a small to medium effect size.18 A 2020 meta-­analysis placed divalproex fifth out of 21 treatments for bipolar depression.19 Valproate has limited utility in rapid-cycling bipolar disorder.20 Although open-­label studies suggest that valproate is effective in the prophylaxis of bipolar affective disorder,21 RCT data are limited.22,23 Bowden et al.24 found no difference between lithium, valproate and placebo in the primary outcome measure, time to any mood episode, although valproate was superior to lithium and placebo on some secondary outcome measures. In another RCT22 there was no difference in relapse rates between valproate and olanzapine. A post-­hoc analysis of data from this study found that patients with rapid-cycling illness had a better very early response to valproate than to olanzapine but that this advantage was not maintained.23 Outcomes with respect to manic symptoms for those who did not have a rapid-cycling illness were better at 1 year with olanzapine than with valproate. In a further 20-­month RCT of lithium versus valproate in patients with rapid-cycling illness, both the relapse and attrition rates were high, and no difference in efficacy between valproate and lithium was apparent.25 The independent BALANCE study found lithium to be numerically superior to valproate, and the combination of lithium and valproate statistically superior to valproate alone.26 Aripiprazole in combination with valproate is superior to valproate alone.27 Overall, data suggest that adjunctive valproate provides additional protection against relapse.28 In the UK, the National Institute for Health and Care Excellence (NICE) recommends valproate as a first-­line option for the treatment of acute episodes of mania, in combination with an antidepressant for the treatment of acute episodes of depression, and for prophylaxis,29 but importantly NOT in women of child-­bearing potential.29,30 A Cochrane review concluded that the evidence supporting the use of valproate as prophylaxis is limited.31 Valproate is sometimes used to treat aggressive behaviours of variable aetiology.32 One RCT (n = 16) failed to detect any advantage for risperidone augmented with valproate over risperidone alone in reducing hostility in patients with schizophrenia.33 A mirror-­image study found that, in patients with schizophrenia or bipolar disorder in a secure setting, valproate decreased agitation.34 There is a small positive placebo-­controlled RCT of valproate in generalised anxiety disorder.35 Valproate may also have preventive benefits against COVID-­19.36 Plasma levels The pharmacokinetics of valproate are complex, following a three-­compartmental model and showing protein-­binding saturation. Plasma level monitoring is supposedly of more limited use than with lithium or carbamazepine.37 There may be a linear association between valproate serum levels and response in acute mania, with serum levels <55mg/L being no more effective than placebo, and levels >94mg/L being associated with the most robust response.38 Optimal serum levels during the maintenance phase are unknown, but are likely to be at least 50mg/L.39 Achieving therapeutic plasma levels rapidly using a loading dose regimen is generally well tolerated. Plasma levels can also be used to detect non-­compliance or toxicity. Using total valproate concentration (the standard method) is no less useful than free valproate levels in most situations.40