Low-dose valproate: a new treatment for cyclothymia, mild rapid cycling disorders, and premenstrual syndrome.

Background: Valproate has proved useful in the treatment of manic-depressive and schizoaffective disorders, usually in daily doses above 500 mg with corresponding blood levels in the range established for treatment of epilepsy (50-100 micrograms/mL). Since milder bipolar disorders may be more prevalent than bipolar I disorder, a prospective study was undertaken to determine whether lower doses of valproate might be useful for stabilization of mood cycling in patients having primary diagnoses of cyclothymia or rapid cycling bipolar II disorder. Additionally, open trials of low-dose valproate were conducted in a small number of women complaining of premenstrual syndrome.
Method: Over a 3-year period, outpatients with non-menstrually-related rapid cycling who had fulfilled DSM-III-R criteria for cyclothymia or bipolar II disorder were started on open trials of valproate at daily doses of 125 or 250 mg. Doses were adjusted upward on approximately a monthly basis depending upon clinical response, and valproate blood levels were obtained.
Results: Twenty-six (79%) of 33 patients (15 cyclothymics, 11 bipolar II) reported sustained partial or complete stabilization of mood cycling with valproate doses ranging from 125 to 500 mg (mean = 351.0 mg) corresponding to serum valproate levels (mean = 32.5 micrograms/mL) substantially below the current recommended range. Cyclothymics required significantly lower doses and blood levels of valproate than patients with bipolar II disorder for stabilization of mood. Five patients (all bipolar II) failed to respond fully to low doses of valproate but improved with higher doses corresponding to blood levels in the 50 to 100 micrograms/mL range. Two patients had poor responses to valproate or intolerable side effects. In contrast to bipolar spectrum patients, only three (38%) of eight women with menstrually related cycling of mood reported good responses to low doses of valproate, while five reported no response to valproate.
Conclusion: The findings suggest that (1) low-dose valproate may be useful in the treatment of cyclothymia and milder rapid cycling bipolar disorders and (2) there may be a correlation between the severity of bipolar disorder and the blood level of valproate required for stabilization such that milder forms of bipolar cycling require lower doses of valproate.