The single most important research study in the past year in the area of clinical therapeutics of bipolar disorder was conducted by Geddes and colleagues at Oxford University in England[1]. Using a randomized open-label design (no placebo control group and subjects knew which medications they were receiving), the BALANCE study sorted 330 subjects with bipolar disorder type I into three treatment groups: lithium alone, valproate (Depakote) alone, or combination treatment with both lithium and valproate. The outcome measures were time to recurrence of a major mood episode, either mania or depression. The study design allowed for an extended, two year follow-up on these subjects. This time frame allows for meaningful assessment of genuine prophylactic effects. The results found that combination therapy was most effective, marginally more so that lithium alone, but significantly greater than valproate monotherapy. The interpretation of the data supports the unique efficacy of lithium as the single-most effective mood stabilizer available.
In recognition of the singular importance of this study, the journal Bipolar Disorders devoted an issue for commentaries from major luminaries in the field including Ross Baldessarini[2], Rasmus Licht[3], and S. Nassir Ghaemi[4] and others. The commentary by Ghaemi, a researcher and analytic thinker whose work I respect enormously, was forthright, pointed, and compelling. In this commentary, he challenges the pharmaceutically-inspired practice habits of American psychiatrists and their seduction by the next, newest, sexiest drug brought to market. His writing deserves to be read by patients and clinicians alike. I include portions of his commentary below:
“Clinical conclusions about lithium”
“Clinicians can, and should, draw some conclusions, if we have the courage. We need to avoid being mugwumps, refusing to commit to using lithium out of vague fears, despite clear benefits that outweigh the real risks. With the results of BALANCE, in the setting of forty years of lithium research, it seems to me that one clinical conclusion is hard to avoid: Lithium is, by far, the first-line treatment for bipolar disorder. There should be very good reasons not to give lithium to the majority of patients with bipolar disorder as initial treatment. Patient preference, by itself, is not a good enough reason to avoid lithium; the hassles of being a doctor (checking blood levels, assessing kidney function long term) are not good enough reasons either. Patients need to be educated about the many benefits of lithium, including two other major areas, besides mood prophylaxis: mortality reduction, both by suicide and by cardiovascular death, and neuroprotective effects, especially probable reduction of dementia risk and potential protection against the cognitive impairment that is a long-term consequence of multiple mood episodes (1). The drawbacks of lithium are well known, though exaggerated: long-term chronic renal insufficiency, in the best prospective studies with decades of follow-up, is not more than 5% (8); other kidney effects, like decreased urinary concentration capacity, are more common but reversible and not medically dangerous; hypothyroidism is more common but treatable and reversible; nuisance side effects are less frequent than many believe; weight gain is less than with valproate and much less than most neuroleptics; cognitive side effects are problematic in some, but not most persons, and counteracted by long-term cognitive benefits; toxicity in overdose is a risk but this is the only drug that is proven to prevent suicide by a huge effect size (estimated to be nine-fold decreased risk) (1).
Some people cannot take lithium. But everyone should be offered it, most should try it, and a minority can then stop it if it is intolerable. If we take this approach, we find that many persons tolerate it, do well, and do not need the common current rigamarole of antidepressants plus neuroleptics, which leaves patients partially treated at best, and hardly treated at worst.
Lithium is unique because it actually treats an entire disease—manic-depressive illness. It is not merely a treatment for a symptom—like neuroleptics for mania or antidepressants for depression. Studies indicate that about one-third of patients get completely well long term with lithium monotherapy (10). This figure is not minor, and compares favorably with the long-term remission seen with antidepressants in major depressive disorder in the STAR*D study (11). BALANCE is another source of evidence for the notable benefit of lithium monotherapy in a substantial minority of persons with bipolar disorder. The makers of our DSM-IV nosology have assumed that all our pills are mere symptom treatments; they do not think our diagnoses reflect diseases in any meaningful way, and they do not believe that any of our treatments cure diseases. Hence, the claim that we should only make pragmatic judgments, compensating for the follies of practitioners and the manipulations of pharmaceutical companies. Lithium is their refutation, and BALANCE is a modern proof that, though they are hard to conduct and require a great deal of labor, we can clarify clinical controversies with rigorous studies. We researchers need to do the studies, perpend on their importance, teach clinicians to implement the results in practice, and educate our nosologist colleagues, especially now as DSM-5 is in process, as to what they mean.”
We listen in full agreement to this elegant and rigorous analysis. Lithium really is the closest thing we have to a wonder drug. It is our first choice for the great majority of bipolar patients that we treat in our practice.