Controlled trial of pyridoxine in the premenstrual syndrome.

A total of 617 patients diagnosed by their general practitioner, according to set criteria, as having premenstrual symptoms were treated in general practice for three menstrual cycles with either pyridoxine or placebo. Treatment was randomized and administered blind. In the 434 patients analyzed, an improvement was found in 7 of the 9 symptoms assessed for both treatments, but the differences between treatments did not reach conventional significance levels. However, improvement as measured by global assessment after three cycles was significantly greater in the patients treated with pyridoxine (p less than 0.02).