Lithium is still recommended as a first-choice treatment for acute bipolar mania, especially in pure euphoric mania of mild to moderate severity. Despite the large quantity of evidence supporting the efficacy of lithium, in clinical practice its use has often been limited because of management issues related to its narrow therapeutic index. International guidelines suggest combining lithium with a second mood stabilizer (anticonvulsant or atypical antipsychotic) for treatment of mixed states, rapid cycling and severe forms of mania with atypical features, which are classically considered to be poorly responsive to lithium alone. To date, however, the specific modalities of these associations on the basis of different clinical presentations have been poorly investigated in clinical trials. In this study, we aimed to evaluate the modalities of use of lithium in a naturalistic setting of manic and mixed bipolar patients, and to investigate the effects of its combination with valproate on the clinical course.
Materials and methods
Seventy-five bipolar I patients (DSM-IV-TR) in a manic (14.7%) or mixed (85.3%) phase, treated with lithium alone or in association with valproate, were recruited at the day hospital of the Psychiatric Unit of the Department of Clinical and Experimental Medicine, University of Pisa, and followed-up in a naturalistic trial for an average period of 6 months. Diagnosis was confirmed using SCID-I. All subjects recruited underwent at least two standardized evaluations of clinical course, assessed by the CGI-BP, at baseline and at each subsequent check of serum lithium levels. Variables concerning clinical features of patients and clinical course of episodes were analyzed by comparison between the two treatment groups (lithium monotherapy vs. lithium plus valproate).
The group of subjects treated with the combination of lithium and valproate (n = 41, 54.7%) was composed mainly of men, had a higher percentage of rapid cyclers and a higher severity of psychotic symptoms at baseline. The two treatment groups did not differ in the other demographic and clinical features analyzed. Patients treated with lithium plus valproate had a higher remission rate at endpoint than subjects treated with lithium monotherapy. The association of valproate significantly reduced the severity of specific symptomatological dimensions, such as mixed, anxiety and psychotic features.
Our data confirm the use of combination therapy in more severe forms of mania. Prospective data on the clinical course have shown that the combination of lithium with valproate is associated with a greater improvement of specific symptomatological dimensions, which are poorly responsive to lithium monotherapy.