New trial should clarify lithium use in bipolar disorder

Editor—We agree with Dinan's statement in his editorial that there is considerable evidence that lithium is an effective maintenance treatment in bipolar disorder.¹ Our recent Cochrane review found that lithium reduced the relative risk of relapse in bipolar disorder by 42% (95% confidence interval 30% to 52%).² We also accept that the widespread switch away from lithium—especially in the United States—is based on marketing and opinion rather than compelling evidence. The absence of good evidence for valproate does not mean, of course, that lithium is more efficacious or more acceptable than valproate. It is also possible that the combination of lithium plus valproate is more effective than either drug alone. We are therefore less confident in accepting the unequivocal recommendation that lithium should remain the first line treatment. There is genuine clinical uncertainty about this issue—and such wide international variations in clinical practice—that an overwhelming case can be made for a clinical trial comparing lithium and valproate.³,⁴

We are currently conducting a large randomised trial in the United Kingdom comparing valproate monotherapy, lithium monotherapy, and combination therapy with valproate plus lithium.⁵ This large collaborative trial (bipolar affective disorder lithium anticonvulsant evaluation, BALANCE) funded by the Stanley Foundation, a mental health charity in the United States, and the trial drugs have been generously donated by Sanofi-Synthelabo. BALANCE has been designed and conducted entirely independently of the pharmaceutical industry. Clinicians and patients are welcome to participate in this pivotal clinical trial. Further information is available at www.psychiatry.ox.ac.uk/balance.