Background: EEG sleep measures in child and adolescent subjects with depression have shown considerable variability regarding group differences between depressed and control subjects. This investigation was designed to assess whether some of the observed variability is related to undifferentiated unipolar and bipolar disorders in a sample that was reported previously.
Methods: Twenty-eight adolescents who met criteria for unipolar major depression and 35 controls with no lifetime psychiatric disorder participated in a cross-sectional sleep polysomnography study. Approximately 7 years later, follow-up clinical evaluations were conducted in 94% of the original cohort. Clinical course during the interval period was assessed without knowledge of subjects' initial diagnostic and psychobiological status. Re-analysis of the original sleep data were performed with the added information of longitudinal clinical course.
Results: Depressed subjects who had a unipolar course showed reduced REM latency, higher REM density, and more REM sleep (specifically in the early part of the night) compared with depressed adolescents who converted to bipolar disorder and controls who remained free from psychopathology at follow-up. In contrast to the unipolar group, depressed subjects who would later switch to bipolar disorder had demonstrated more stage 1 sleep and diminished stage 4 sleep.
Conclusions: These preliminary results indicate that some of the observed variability in EEG sleep measures in adolescent depression appear to be confounded by latent bipolar illness. The findings also suggest that sleep regulatory changes associated with unipolar versus bipolar mood disorders may be different.
Copright 2002 Elsevier Science BV.