Duration and stability of the rapid-cycling course: a long-term personal follow-up of 109 patients.

BACKGROUND: Recognition by the DSM-IV of rapid cyclicity as a course specifier has raised the question of the stability and long-term outcome of rapid-cycling (RC) patients. Data on this topic is sparse and often inconsistent. To our knowledge, these are the first personally followed patients over the long term, dealing directly with the issue of the duration of the RC course. METHODS: We examined the evolution of the course of 109 RC patients (68 women and 41 men) followed for a minimum of 2 years and up to 36 years, beginning with the index episode when the RC course was diagnosed by the authors (A.K., G.P.M., P.G., L.P., D.R.). Patients were included in the study if they met criteria for RC as defined by>or=4 affective episodes per year (Dunner and Fieve, 1974). The follow-up period varied from 2-5 years for 25 patients, 6-10 years for 24 patients, 11-15 years for 24 patients, 16-20 years for 19 patients, 21-25 years for 13 patients, 30-36 years for four patients. RESULTS: In 13 patients (12%), RC emerged spontaneously and in 96 patients (88%), it was associated with antidepressant and other treatments. In 19 women (28% of all women) RC course started in perimenopausal age (45-54 years). The mean duration of RC during the follow-up period was 7.86 years (range 1-32) and its total duration (including RC course prior to the follow-up period) was 11 years (range 1-40). The total duration of the affective disorder, from the first episode to the end of the follow-up, was 21.78 years (range 1-70). At the end of the follow-up, 36 patients (33%) had complete remission for at least the past year, 44 (40%) stayed rapid cycling with severe episodes (six of this group committed suicide), while 15 (14%) were rapid cycling but with attenuated episodes. The other 14 patients (13%) became long cyclers, eight with severe episodes and six with milder ones. The main distinguishing features between those who remitted from and those who persisted in the RC course were: (1). the initial cycle pattern: patients with Depression-Hypomania(mania)-Free interval cycles (53 patients) had a worse outcome: 26.4% remitted and 52.8% persisted in the RC course through to the end of the follow up period. The Mania/Hypomania-Depression-Free interval cycles (22 patients) had a significantly better outcome, with 50% remitted and 27.2% persisting RC; and (2). the occurrence of the switch process from depression to hypomania/mania and the occurrence of agitated depressions made the prognosis worse. Continuous treatment was more effective against mania/hypomania than against depression, yet in all persisting RC cases the mania/hypomania remitted only partially. LIMITATIONS: These data derive from clinics known for their expertise in mood disorders, and they may have attracted and retained patients with a more severe course. Treatment was uncontrolled and consisted more of lithium than divalproex, lamotrigene and olanzapine, recently shown to be beneficial in subgroups of patients with rapid-cycling. CONCLUSIONS: Our findings suggest that rapid cyclicity, spontaneous or induced, once established, becomes for many years a stable rhythm in a substantial proportion of patients, linked to endogenous and environmental factors. The suggestion is made to consider as rapid-cyclers, at least for research purposes, those patients who have had a rapid cycling course for at least 2 years, borrowing the duration criterion currently employed for other chronic disorders such as Dysthymia and Cyclothymia. That our patients had poorer prognosis than some other cohorts in the literature is probably due to the shorter duration of "rapid-cycling" at entry in the latter cohorts. A true understanding of the nature of rapid-cycling will require a rigorous definition of not only duration, but also pole-switching and course patterns at entry into study.