To the Editor: We read with interest the article by Franca Centorrino, M.D., et al. (1). For similar reasons—enhanced patient compliance resulting from once-daily dosing and the potential for greater tolerability because of less peak-trough blood-level fluctuation—we also performed a pilot switching study that we presented at the 2001 APA annual meeting and would like to share with your readers.

Ten patients with bipolar I or bipolar II disorder (some with other axis I or axis II comorbidity) who exhibited side effects that limited their compliance or tolerability to delayed-release divalproex sodium were switched to the once-daily extended-release formulation. The patients were switched based on bioavailable dose data and available tablet strength. In two multiple-dose studies, the average bioavailability of extended-release divalproex given once daily was 81%–89% relative to delayed-release divalproex tablets given b.i.d. (2). The patients were evaluated over 12 weeks to monitor their clinical status (Clinical Global Impression and Global Assessment of Functioning scales), and side effects were measured with a 7-point Likert rating scale. No additions, deletions, or dose changes of concomitant medications occurred in any patients during the 12-week observation period. Laboratory assessments included baseline and follow-up divalproex blood levels, liver function tests, and CBCs. Our case series was unblinded, open label, and naturalistic. All patients were maintained in their usual outpatient treatment settings, with no alteration in the type or frequency of their clinic appointments.

Six male and four female patients ages 27 to 52 who were currently taking divalproex for a DSM-IV diagnosis of bipolar I, bipolar II, or schizoaffective disorder were followed. This study was approved by the Aurora Healthcare Institutional Review Board, and all subjects provided appropriate informed consent. All patients had liver function tests with results within normal limits and had no history of hepatitis, pancreatitis, or hematological abnormalities. All patients had at least one side effect attributed to divalproex (namely, gastrointestinal discomfort, sedation, weight gain, or tremor) that prompted their desire to switch to a potentially more tolerable extended-release formulation.

Our 12-week follow-up results substantiated that extended-release divalproex was effective (nine of 10 patients showed equal or mildly improved clinical status), well tolerated (five of 10 had reductions in side effects), and led to enhanced medication compliance in certain individuals. Pharmacokinetic data from extended-release divalproex studies suggest that its peak and trough blood levels do not fluctuate significantly compared to the conventional divalproex formulation (2). Peak levels may be associated with increased side effects, and adequate trough levels are thought to affect the efficacy of divalproex. Thus, the extended-release formulation of divalproex (which has a more level steady-state curve) may confer the dual advantage of lesser side effects (tolerability) and more sustained efficacy.

Studies have suggested that noncompliance with pharmacological treatments for bipolar disorders is as high as 50%, and compliance increases as the number of daily doses decreases (3). Thus, once-daily dosing should improve patient compliance and, consequently, may improve treatment outcomes. This may be especially important in the treatment of individuals with bipolar disorder, for whom noncompliance often not only precipitates acute decompensation but also leads to the development of a more intractable disease pattern or loss of responsiveness to previous regimens (e.g., lithium). Our open pilot findings add to the growing literature (4) that supports the safety and efficacy of extended-release divalproex sodium in psychiatric patient populations, and we hope that it stimulates more rigorous controlled clinical trials.