Sleep patterns in borderline personality disorder.

Sleep patterns of borderline patients with and without a history of affective disorder were compared to each other and to normal reference data. The three groups could not be distinguished in terms of REM latency because a wide spread of values was seen within each group. Borderlines were different from normal controls in other aspects of sleep architecture; they had less total sleep, more stage 1 sleep, and less stage 4 sleep. If one assumes that REM latency is a biological marker for mood disorder, then our results do not support the hypothesis that borderline personality disorder is a variant of affective illness. However, other data suggest that REM latency should not be used to validate the presence of affective illness.