doi: 10.1111/j.1365-2869.2008.00651.x Does REM sleep contribute to subjective wake time in primary insomnia? A comparison of polysomnographic and subjective sleep in 100 patients BERND FEIGE 1 , ANAM AL-SHAJLAWI 1 , CHRISTOPH NISSEN 1 , ULRICH VODERHOLZER 1 , MAGDOLNA HORNYAK 1,2 , KAI SPIEGELHALDER 1 , CORINNA KLOEPFER 1 , MICHAEL PERLIS 3 and DIETER RIEMANN 1 1 Department of Psychiatry and Psychotherapy, 2 Interdisciplinary Pain Center, University Medical Center, Freiburg, Germany and 3 Department of Psychiatry, University of Rochester, New York, USA Accepted in revised form 31 January 2008; received 8 October 2007 SUMMARY Primary insomnia (PI) is characterized by low subjective sleep quality which cannot always be verified using polysomnography (PSG). To shed light on this discrepancy, subjective estimates of sleep and PSG variables were compared in patients with PI and good sleeper controls (GSC). 100 patients with PI (age: 42.57 ± 12.50 years, medication free for at least 14 days) and 100 GSC (41.12 ± 13.99 years) with a sex distribution of 46 men and 54 women in each group were included. Both PSG and questionnaire variables showed clear impairments of sleep quality in PI compared with GSC. The arousal index within total sleep time was increased, which was mainly because of a strong increase within rapid eye movement (REM) sleep. Subjectively, more PI than GSC subjects estimated wake times longer than obtained from PSG. Linear modeling analysis of subjective wake time in terms of PSG parameters revealed that in addition to PSG defined wake time, REM sleep time contributed significantly to subjective wake time. This REM sleep contribution was larger for PI than for GSC subjects. The findings suggest that REM sleep-related processes might contribute to subjectively disturbed sleep and the perception of waking time in patients with PI. keywords polysomnography, primary insomnia, sleep perception INTRODUCTION Primary insomnia (PI) is a prevalent health complaint of yet unknown etiology (Ohayon, 2002). The disorder is character- ized by subjectively non-restorative sleep associated with daytime complaints such as impaired social or professional functioning, lasting for at least 4 weeks, in the absence of any possibly underlying mental or physical disease or substance intake. The disorder frequently takes a chronic course, causing severe limitation of quality of life and performance over several years (Katz and McHorney, 2002; Leger et al., 2002; Ozminkowski et al., 2007). In addition, recent studies indicate that PI is a risk factor for the development of psychiatric disorders (Breslau et al., 1996; Ford and Kamerow, 1989; Riemann and Voderholzer, 2003). Polysomnographic (PSG) recordings, allowing conventional sleep staging according to the widely accepted standard criteria of Rechtschaffen and Kales (1968) are not required for the diagnosis of PI according to the current diagnostic criteria (DSM-IV, American Psychiatric Association., 1994, ICD-10, World Health Organisation, 1992 or ICSD-2, International Classification of Sleep Disorders Diagnostic and Coding Manual, American Academy of Sleep Medicine, 2005). Yet, if PSG sleep recordings are performed, a subset of patients with PI display undisturbed PSG sleep patterns in contrast to their subjective sleep complaint (Carskadon et al., 1976; Edinger and Fins, 1995; Frankel et al., 1976; Means et al., 2003). This phenomenon has been labeled Ôsleep state Correspondence: Bernd Feige, Department of Psychiatry and Psycho- therapy, University Medical Center, Hauptstrasse 5, 79104 Freiburg, Germany. Tel.: 49-761-270-6824; fax: 49-761-270-6619; e-mail: Bernd.Feige@uniklinik-freiburg.de J. Sleep Res. (2008) 17, 180–190 Insomnia 180 Ó 2008 European Sleep Research Society