Depression in borderline patients: a prospective EEG sleep study.

The relationship between borderline personality disorder and primary major depression was studied prospectively using Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) interviews and electroencephalographic (EEG) sleep studies. Ten consecutively admitted borderline patients (a prospective sample), defined by Gunderson's Diagnostic Interview for Borderlines (DIB), underwent EEG sleep studies on two consecutive nights and were compared to previously reported samples of nonborderline depressed patients (defined by Research Diagnostic Criteria; RDC), normal controls, and DIB-defined borderline patients who had been referred "to rule out major depression" (a retrospective sample). EEG sleep data were analyzed visually and by automated techniques. Rapid eye movement (REM) latency values were similar in depressed and both borderline groups but significantly different from controls. Eighty-five percent of REM latency values in RDC major depressives were less than or equal to 65 minutes, compared to similar rates of 75% in the prospective sample of borderline patients and 65% in the retrospective sample, versus 35% for controls (chi 2 = 10.7, p less than 0.005). The REM latency in borderline patients did not vary with the severity of depression as measured by the Hamilton Rating Scale for Depression. In the prospective borderline sample, the major SADS-L diagnoses were chronic intermittent depression (five), current major depression (four) (two unipolar, two bipolar II), and labile personality (one). A convergence of nosologic and EEG sleep data is suggested, and supports the concept of a close relationship between criteria-defined borderline personality disorder and affective illness.