C-reactive protein Levels and Sleep Disturbances: Observations Based on The Northern Finland 1966 Birth Cohort Study TIMO LIUKKONEN, MD, PIRKKO R¨ AS ¨ ANEN, MD, PHD, AIMO RUOKONEN, MD, PHD, JAANA LAITINEN,PHD, JARI JOKELAINEN, MSC,MAIJA LEINONEN,PHD, V. BENNO MEYER-ROCHOW,PHD, DSC, AND MARKKU TIMONEN, MD, PHD Objective: To investigate whether sleep disturbances are associated with C-reactive protein (CRP) levels at the population level. Elevated CRP levels have been found to accompany sleep disturbances, but evidence so far comes only from limited clinical and experimental studies; epidemiological studies are lacking. Methods: We utilized the Northern Finland 1966 Birth Cohort, whose participants have been followed up to the age of 31 years. The hs-enzyme immunoassay method was used to measure highly sensitive-CRP (hs-CRP) concentrations (4011 participants). Self-reported sleep disturbances were ranked from 1 (no problem) to 5 (severe disturbances). Results: Multivariate analyses, after adjusting for confounders, revealed that hs-CRP levels in men in the sleep disturbance category “moderate, considerable and severe” (i.e., self-reported sleep disturbances rated 3, 4, or 5), were 18% (18.2%, 95% Confidence Interval 3.0% to 36.3%) higher than those in men with “no” sleep disturbance. In women, hs-CRP levels did not significantly differ between different sleep disturbance categories. Conclusions: Our results support the hypothesis that moderate-to-severe sleep disturbances in men are associated with slightly increased CRP levels at the epidemiological level. Further investigations are called for to see whether our results can be replicated in other databases. Key words: C-reactive protein, inflammation, sleep, sleep disturbances. CRP = C-reactive protein; IL-6 = interleukin-6; EIA = enzyme immunoassay; Hs-CRP = highly sensitive C-reactive protein; HSCL-25 = Hopkins Symptom Checklist-25. INTRODUCTION T here is a growing body of evidence, based on animal models (1) and also observations in humans (2–5), that suggests that sleep disturbance and impaired physical well- being are connected. Reduced sleep durations have been as- sociated with increased risks of developing cardiovascular disorders (2,6,7) and Type 2 diabetes (2,8,9), whereas low- grade inflammation has been assumed to be an underlying mechanism for these disorders (10 –12). Several studies have reported associations between sleep latency, sleep restriction or impaired night time sleep, and elevated levels of interleukin-6 (IL-6) (13–17), which is thought to regulate C-reactive protein (CRP) production. When compared with IL-6, CRP-level is, however, known to be more stable across 24 hours and for that reason may reflect more accurately the state of inflammation (18). In two differ- ent experiments, based on 10 participants each, it was shown that partial (4.2 hours of night time sleep) as well as total (continuous 88 hours) sleep restriction caused an elevation of CRP concentrations (19). Larkin et al. also found CRP levels to be associated negatively with sleep duration in a study with 143 adolescent patients suffering from sleep disordered breathing (20). Very recently, McDade et al. found sleep latency to be independently associated with elevated CRP in among 188 participants, aged 52 to 72 years (21). To the best of our knowledge, there is no large population- based study that investigates, at epidemiological level, whether sleep disturbance is associated with CRP levels. In the present study, we used the highly sensitive (hs) enzyme immunoassay (EIA) method and applied it to the measurement of CRP concentrations to investigate whether sleep distur- bances, with regard to the general population birth cohort database, are associated with increased levels of hs-CRP. METHODS Study Design We used data from the large and genetically homogeneous Northern Finland 1966 Birth Cohort. In the two northernmost Finnish provinces, i.e., Oulu and Lapland, 96% of all women (n = 12,068), with an expected date of delivery falling between 1st January and 31st December in 1966, were evaluated. They gave birth to 12,058 live infants. Many biological, socioeco- nomic, and health conditions as well as living habits and family characteristics of the cohort members were collected from prenatal stages through the age of 31 years. A detailed description of the data has been published earlier (22,23). This study was approved by the Ethics Committee of the Faculty of Medicine, University of Oulu (Finland), and the informed consent was signed by all participants. Instruments and Measurements In 1997, as part of the 31-year follow-up survey, 11,541 of the original 12,058 cohort members were available and were sent postal questionnaires. All those who were living in northern Finland or in the capital area were invited to undergo a clinical examination (n = 8463), and 6025 (71.2%) attended (24,25). The question regarding sleep disturbances was addressed by question 5 in 15-D- questionnaire (26), which was included in the postal questionnaires. Sleep dis- turbance was scaled from 1 to 5, where 1 indicated no problem, 2 slight, 3 moderate, 4 considerable, and 5 severe sleep disturbances. For the purpose of this study, sleep disturbances were assigned to three categories: 1) “no,” 2) “slight,” and 3) “moderate, considerable, and severe.” Hs-CRP-levels were measured from the serum samples, collected during clinical examinations, and assayed immu- noenzymometrically (MedixBiochemica, Helsinki, Finland). The sensitivity of the assay was 0.08 mg/l (27). Initially, hs-CRP concentration was analyzed as continuous variable. We then dichotomized the hs-CRP variable by using a cut-off point of 1.0 mg/L, which had earlier been shown to be associated with an elevated risk of another inflammatory associated disorder, namely, cardiovascular disease (10). Con- From the Department of Psychiatry (T.L), Savonlinna Central Hospital, Savonlinna, Finland; Department of Public Health Science and General Prac- tice (T.L., J.J., M.T.), University of Oulu, Oulu, Finland; Department of Psychiatry (P.R.), University of Oulu, Oulu, Finland; Department of Psychi- atry (P.R.), Oulu University Hospital, Oulu, Finland; Department of Clinical Chemistry (A.R.), University of Oulu, Oulu, Finland; Finnish Institute of Occupational Health (J.L.), Promotion of Work Ability and Health Team, Oulu, Finland; Unit of General Practice (J.J.), Oulu University Hospital, Oulu, Finland; National Public Health Institute (M.L.), Oulu, Finland; School of Engineering and Science (V.B.M.-R.), Jacobs University (formerly known as International University Bremen), Bremen, Germany; Department of Physi- ology (V.B.M.-R.), University of Oulu, Oulu, Finland. Address correspondence and reprint requests to Timo Liukkonen, Depart- ment of Psychiatry, Savonlinna Central Hospital Keskussairaalantie 6, FIN- 57120 Savonlinna, Finland. E-mail: liukkonen.timo@dnainternet.net Received for publication December 18, 2006; revision received May 21, 2007. This study was supported by grants from Finnish Psychiatric Association (T.L.). DOI: 10.1097/PSY.0b013e318157cb96 756 Psychosomatic Medicine 69:756 –761 (2007) 0033-3174/07/6908-0756 Copyright © 2007 by the American Psychosomatic Society