Chronobiology International, 2014; 31(5): 698–704 ! Informa Healthcare USA, Inc. ISSN: 0742-0528 print / 1525-6073 online DOI: 10.3109/07420528.2014.895742
ORIGINAL ARTICLE
Correlations between objective and subjective sleep and circadian markers in remitted patients with bipolar disorder
1
INSERM, U955, Cre´teil, France, 2AP-HP, Hoˆpital H. Mondor – A. Chenevier, Poˆle de Psychiatrie, Cre´teil, France, Fondation FondaMental, Cre´teil, France, 4Poˆle de psychiatrie, Univ Lille Nord de France, CHRU de Lille, Lille, France, 5 AP-HP, Hoˆpital Fernand Widal, Poˆle Addictologie-Toxicologie-Psychiatrie and Universite´ Paris-7, Paris, France, 6 Faculte´ de me´decine, Universite´ Paris Est, Cre´teil, France, 7Institut de Psychologie, Universite´ Paris Descartes, Paris, France, and 8Body Rhythms and Shiftwork Centre, Swansea University, Swansea, Wales, United Kingdom 3
Bipolar disorder (BD) is a chronic psychiatric condition characterized by recurrences of depressive and (hypo)manic episodes. Patients in remission report a wide range of sleep and circadian disturbances that correlate with several outcomes measures such as functioning or physical health. The most appropriate way to measure these abnormalities in clinical practice requires further investigation since the external validity of self-reports, as compared to more physiological measures (such as polysomnography or actigraphy), has been questioned. Despite the fact that questionnaires are inexpensive, fast and easy to use, they need to be validated against objective measures. This study aims to validate three sleep and circadian questionnaires, namely the Pittsburgh Sleep Quality Index (PSQI), the Composite Scale of Morningness (CSM) and the Circadian Type Inventory (CTI) – against actigraphy in BD patients in remission. Twenty-six carefully assessed BD patients in remission completed the PSQI, the CTI and the CSM, and wore an actigraph (AW7, Camntech) for 21 consecutive days. Phase preference assessed by the CSM strongly correlated with actigraphic phase markers (M10 onset ¼ 0.69 and L5 onset ¼ 0.63). Sleep duration and sleep latency assessed by the PSQI and by actigraphy were also highly correlated ( ¼ 0.76; ¼ 0.50). Moderate correlation coefficients were observed between questionnaires and actigraphy for markers that explored the stability of rhythms, sleep quality, sleep latency and sleep disturbances (jj40.40) although these were not significant after correcting for multiple testing. No correlation was observed between markers for the amplitude of rhythms. While the external validity of the CTI clearly requires further investigation, this study supported the external validity of the CSM and the PSQI for phase preference, sleep duration and latency. We conclude that the CSM and the PSQI could be useful in routine practice and research when actigraphy is not easily available. Keywords: Actigraphy, bipolar disorder, questionnaires, validation
INTRODUCTION
reactivity, cognitive functioning, inflammatory processes and metabolic disturbances (Giglio et al., 2010; McKenna & Eyler, 2012; Soreca et al., 2009). Thus, BP is now conceptualized as a multisystemic disorder with sleep and circadian disturbances as a central physiopathological pivot (Harvey et al., 2005; Leboyer & Kupfer, 2010). In view of this, Kapczinski et al. have recently highlighted the urgent need for the objective assessment of circadian rhythms and sleep in BD patients (Kapczinski et al., 2011). A wide variety of tools are likely to be used to assess sleep or circadian biomarkers, ranging from questionnaires, through actigraphy and polysomnography (PSG) to plasmatic or salivary profiles of neurohormones
20 14
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Carole Boudebesse1,2,3, Pierre Alexis Geoffroy1,3,4, Frank Bellivier3,5, Chantal Henry1,2,3,6, Simon Folkard7,8, Marion Leboyer1,2,3,6, and Bruno Etain1,2,3
Bipolar disorder (BP) is a chronic psychiatric condition commonly conceptualized as cyclically alternating depressive and (hypo)manic episodes separated by periods of symptomatic remission. Although the precise determinants of the disorder remain unclear, part of the pathophysiology is thought to be linked to sleep and circadian disturbances (Leboyer & Kupfer, 2010). Indeed BD patients display disturbances of subjective and objective sleep/circadian markers; especially sleep duration and regularity, cortisol and melatonin secretion profiles and phase preference. All these circadian abnormalities are thought to impact on emotional
Submitted December 20, 2013, Returned for revision February 6, 2014, Accepted February 14, 2014
Correspondence: Carole Boudebesse, AP-HP, Hoˆpital H. Mondor – A. Chenevier, Poˆle de Psychiatrie, Cre´teil, France. E-mail: [email protected]
698
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder (like melatonin or cortisol). However, clinicians may experience difficulties in choosing between all the available assessment tools because of considerations of objective versus subjective assessments, cost constraints in daily practice, availability, invasiveness, acceptability to patients and the external validity of questionnaires. Most biochemical methods, such as cortisol or melatonin plasmatic or salivary secretion profiles, are clearly primarily used for research purposes. PSG – the gold standard for exploring sleep abnormalities – is not readily available in many countries and is mainly used to diagnose sleep disorders (e.g. Obstructive Sleep Apnea). In everyday clinical practice sleep and circadian rhythm assessments may be made by the use of questionnaires (subjective, inexpensive, and easy to interpret) and by actigraphy (objective, more expensive, and more difficult to interpret). These tools are meant to measure sleep parameters, or some characteristics of circadian rhythms (i.e. phase, amplitude or stability), and have already produced promising results in studies of BD patients (Etain et al., 2011; Milhiet et al., 2011). Indeed, actigraphic studies of BD patients have mainly reported abnormalities in terms of sleep duration sleep efficiency, sleep onset latency and wake after sleep onset (Gershon et al., 2012; Harvey et al., 2005; Jones et al., 2005; Kaplan et al., 2012; Millar et al., 2004; Mullin et al., 2011; Ritter et al., 2012; Salvatore et al., 2008; St-Amand et al., 2012). Actigraphy has been validated against PSG in BD patients for several sleep parameters (sleep onset latency, awakenings after sleep onset, total sleep time, and sleep efficiency) (Kaplan et al., 2012). More subjective assessments can use sleep/ circadian questionnaires such as the Pittsburgh Sleep Quality Index (PSQI) which was first developed to assess sleep quality in patients with mood disorders and in patients with sleep disorders (Buysse et al., 1989). When used in samples of BD patients in remission, it showed a high frequency of abnormalities, and in particular with respect to sleep latency and sleep efficiency (Rocha et al., 2013). Two circadian questionnaires can also be used to describe rhythms in terms of their three main characteristics, namely phase, amplitude and stability. The Composite Scale of Morningness (CSM) indicates preferences associated with morning or evening activities (Smith et al., 1989) and is thus considered as a marker of circadian phase. The Circadian Type Inventory (Folkard et al., 1979). CTI describes circadian rhythms as flexible/rigid (reflecting rhythms stability) and as languid/vigorous (reflecting rhythms amplitude). We have recently demonstrated that BD patients in remission assessed with these two questionnaires show an evening preference, and are more languid-type, than healthy controls (Boudebesse et al., 2013a). Before these questionnaires can be used in everyday clinical practice for BD patients more information is !
Informa Healthcare USA, Inc.
699
required about their external validity. The main goal of the present study was thus to correlate selected objective actigraphic measures with scores obtained using subjective questionnaires (CSM, CTI and PSQI) in BD patients in remission.
MATERIALS AND METHODS Population Twenty-six BD outpatients in remission were included after obtaining their written informed consent. The appropriate French ethics committee approved this study. The experimental protocol was conformed to international ethical standards (Portaluppi et al., 2010). The patients were all from the psychiatric department affiliated to the Paris XII University. Primary diagnosis of BD was made according to DSM-IV criteria using a structured interview, namely the Diagnostic Interview for Genetic Studies (DIGS) (Nurnberger et al., 1994). Current psychotropic treatments were systematically recorded. All patients were in remission as defined by both the Mongomery and Asberg Depression Rating Scale (MADRS) (Montgomery & Asberg, 1979) and the Young Manic Rating Scale (YMRS) (Young et al., 1978) (with scores below 8 and the absence of major mood episodes according to DSM-IV criteria within the 3 months prior to the study). The main exclusion criteria for patients were hospitalization or any change in psychiatric medications within the 3 months prior to the study. To ensure that the patients had close to normal sleep–wake patterns, other exclusion criteria included shift work, recent travel with more than 3 h of jet-lag, pregnancy, child birth and recent bereavement (within 2 months prior to the study). Sleep and circadian questionnaires All participants were also assessed with the PSQI, the CSM and the CTI. The PSQI is a 19-item self-assessment questionnaire that examines subjective sleep quality (Buysse et al., 1989). It generates a total score and seven component scores, namely: subjective sleep quality, sleep latency, sleep duration, sleep disturbances, use of sleeping medication and daytime dysfunction. We used a validated French version (Blais et al., 1997). The PSQI had already been validated against actigraphy for patients with lung cancer (Grutsch et al., 2011), but not for BD patients. The Composite Scale of Morningness (CSM) is a 13-item self-assessment questionnaire that examines preferences associated with morning or evening activities (Smith et al., 1989). It is considered as a marker of circadian phase. We used a validated French version of the CSM (Caci et al., 1999). The CTI is an 11-item self-assessment questionnaire that describes circadian rhythms as flexible/rigid (a marker of rhythm stability) and as languid/vigorous
700
C. Boudebesse et al.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
(a marker of rhythm amplitude). For example, rigid types claim to be less able to sleep at unusual hours and languid types claim to be more lethargic following reduced sleep. We have recently validated a French version of this questionnaire (Boudebesse et al., 2013b). The CSM had been previously associated with melatonin plasmatic level in healthy volunteers (MoreraFumero et al., 2013). However, we are unaware of any studies that have validated the CTI and the CSM against actigraphy in any population.
Actigraphy Sleep was objectively assessed by means of actigraphy. An actigraph is a watch-like tool containing an accelerometer that detects, scores, and stores information about the intensity and timing of wrist movements over several consecutive 24-h periods. Actigraphy allows the prospective and objective monitoring of sleep (inactivity) and activity patterns, and can capture the variability in sleep–wake patterns over several consecutive days (Morgenthaler et al., 2007). Participants wore the actigraph device (AW-7 CamNtech) continuously on the non-dominant wrist for 21 consecutives days. Participants were instructed to press the event-marker when they went to bed to sleep and when they got out of bed to start the day. They also completed an adapted version of the Pittsburgh Sleep Diary (Monk et al., 1994). The data were sampled in 1-min epochs and analyzed with the sleep detection algorithm provided by Actiwatch software (Actiwatch Activity & Sleep Analysis Ltd CamNtech 7.28). The following sleep scoring procedure was used: (1) information provided by the event-marker was given priority, (2) if participants forgot to press the event-marker, the missing information concerning bedtime and/or rise time was taken from the diary, (3) visual inspection was used to detect and correct any incongruence between the times provided by the event marker, or the diary, and the actigraph recordings (Boudebesse et al., 2013b). Statistics Data analysis was conducted using JMPÕ Pro 9.0.0 from the SAS Institute. We used Spearman tests for the correlations since most measures were not normally distributed. The correlations were computed between the actigraphy measures and the various scores from the sleep and circadian questionnaires, and the Bonferroni correction for multiple testing was used. We preselected the actigraphic variables that were most likely to measure the same construct as the questionnaires in order to limit the number of correlations computed. All subjective sleep and circadian parameters and actigraphic variables are defined in Table 1. The PSQI sub-components, ‘‘use of sleeping medication’’ and ‘‘day dysfunction’’, were not used since they had no actigraphic equivalents.
RESULTS Population The patients’ mean age was 53.50 years (SD ¼ 11.49) and there were 17 women and 9 men. There were 16 patients with type 1 BD, 8 with type 2 BD and two with BD not otherwise specified. The mean scores on the mood symptoms scales were very low, reflecting euthymia, with a mean MADRS score of 1.85 (SD ¼ 2.80) and a mean YMRS score of 0.65 (1.38). With respect to treatment, all the patients but one took at least one psychotropic medication (lithium n ¼ 12; anticonvulsivants n ¼ 15, atypical antipsychotics n ¼ 7, benzodiazepines n ¼ 4; typical antipsychotics n ¼ 4; antidepressants ¼ 5). Polymedication was fairly common with nine patients taking a combination of two mood-stabilizers. Medication is resumed in Table 2. Correlation between sleep and circadian questionnaires and actigraphy parameters The results are summarized in Table 3. Correlation of phase markers assessed by actigraphy and by the CSM Phase markers assessed by actigraphy (M10 onset and L5 onset) and by the CSM score were strongly correlated (respectively ¼ 0.69 and ¼ 0.63). The tests remained significant (p50.0001) after correcting for multiple testing. Correlation of stability and amplitude markers assessed by actigraphy and by the CTI Although not highly significant, we observed a moderate correlation (jj40.40) between stability measured using actigraphy (inter-day stability and the SD (standard deviation) of get up time calculated over 21 d) and rigidity (CTI F/R score). Flexible BD patients showed more variability in their get up times and lower interday stability, meaning higher variability of their rhythms. There was no correlation between amplitude assessed by actigraphy (relative amplitude or absolute amplitude) and the CTI L/V score. Correlation of sleep markers assessed by actigraphy and by the PSQI subcomponents Subjective and objective estimate measures of sleep duration correlated strongly ( ¼ 0.76) and this remained significant (p50.0001) after correcting for multiple testing. Other correlations between the objective actigraphic measures and the subjective markers measured using the PSQI were of moderate to strong magnitude (jj40.40), although they were not significant after correcting for multiple testing. These findings provide some degree of external validity for the PSQI total score and for the sleep quality, sleep latency and sleep disturbances component scores. Chronobiology International
!
Informa Healthcare USA, Inc.
Sleep disturbances (PSQI component)
Sleep duration (PSQI component) Habitual sleep efficiency (PSQI component)
Estimated duration from bedtime to sleep start Estimated sleep duration Ratio of time spent asleep (total sleep time) to the amount of time spent in bed Subjective sleep disturbances
Sleep latency (PSQI component)
Global indicator of sleep quality
Sleep markers PSQI total score
Subjective sleep quality assessed from good to bad
Indicator of rhythms’ stability
Stability markers CTI F/R score
Sleep quality (PSQI component)
Indicator of rhythms’ amplitude
Eveningness: preference for activity in the evening
Definition
Amplitude markers CTI L/V score
Phase markers CSM total score
Questionnaires
TABLE 1. Definitions of questionnaires and actigraphy markers.
Mean activity of sleep epochs Wake after sleep onset (WASO)
Sleep duration Sleep efficiency
Nighttime motor activity during sleep Wake after sleep onset duration
Duration from sleep start to sleep end Percentage of sleep between bedtime and get up time
Nighttime motor activity during sleep Duration from bedtime to sleep start
Mean activity of sleep active epochs Sleep latency
Wake after sleep onset (WASO)
Nighttime motor activity during sleep Percentage of sleep between bedtime and get up time Wake after sleep onset duration
Strength of coupling of the sleep–wake rhythm to the 24-h regularity in the environment Variability of the get up time between days
Quantifies the difference between daytime (M10) and nighttime (L5) activity levels Difference between the minimum and maximum of the motor activity
Beginning of the 10 most active hours of the 24-h cycle Beginning of the five less active hour of the 24-h cycle
Definition
Mean activity of sleep active epochs Sleep efficiency
standard deviation (SD) of get up time
Inter-day stability
Amplitude
Relative amplitude
L5 onset
M10 onset
Actigraphy markers
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder 701
702
C. Boudebesse et al.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
DISCUSSION We found varying levels of concordance between subjective and objective circadian and sleep assessment. Strong correlations were observed between actigraphic and questionnaires-based measures of circadian phase, sleep latency and sleep duration. Moderate correlations were also observed for some other parameters such as the stability of rhythms, sleep quality, sleep efficiency and sleep disturbances. While these findings support the use of such questionnaires in clinical practice for assessing some sleep/circadian markers in BD patients, further validation studies are clearly needed in larger and independent samples. The use of questionnaires to assess sleep or rhythmicity in clinical practice has long been questioned. In comparison to laboratory measures, such as PSG, they are thought to be too subjective despite their obvious practicality and cost-effectiveness. For example, it has been argued that compared to actigraphy questionnaires are likely to underestimate WASO or to overestimate sleep onset latency. Among in-patients with alcohol dependence, objective and subjective
TABLE 2. Patients’ medication. Treatment
Percentage of patients (%)
Lithium Anticonvulsivants Atypical antipsychotics Typical antipsychotics Benzodiazepines Antidepressants
46 58 27 15 15 19
measures of sleep efficiency correlate, but measures of sleep duration do not (Brooks et al., 2012). Similar misestimates have also been observed in PTSD (Kobayashi et al., 2012), insomnia (Gooneratne et al., 2011) and among women with fibromyalgia (Stuifbergen et al., 2010). However, such discrepancies remain controversial, for example in insomnia sleep-disordered patients (Kushida et al., 2001), leading to the recommendation that subjective data should be used as an adjunct to actigraphic data in estimating total sleep time and sleep efficiency. Little is known about such potential discrepancies in BD patients. Harvey et al. (2005) observed that BD patients, when assessed both objectively and subjectively, tend to overestimate sleep onset latency and underestimate sleep duration. Gonzalez et al. (2013) investigated the correlation between objective (7 d actigraphy) and subjective (estimated via sleep diaries) measures of total sleep time in a sample of 39 patients with bipolar type I disorder and found a significant correlation (r ¼ 0.51). In BD patients, actigraphy and PSG are highly correlated for sleep onset latency, wake after sleep onset, number of awakenings, total sleep time, and sleep efficiency (Kaplan et al., 2012). Thus it is, as yet, difficult to be sure about the external validity of sleep/circadian questionnaires and hence to definitively recommend their use in clinical practice. Nevertheless, our study provides evidence for substantial correlations for phase preference, sleep latency and sleep duration. Further investigations in larger samples are required for other variables such as the stability and amplitude of rhythms, sleep quality, sleep efficiency and sleep disturbances.
TABLE 3. Correlations between actigraphy and questionnaires for circadian and sleep markers in BD patients. Statistics Questionnaires Phase markers CSM total score Amplitude markers CTI L/V score Stability markers CTI F/R score Sleep markers PSQI total score PSQI components scores Sleep quality (PSQI) Sleep latency (PSQI) Sleep duration (PSQI) Habitual sleep efficiency (PSQI) Sleep disturbances (PSQI)
Actigraphy
p Value
M10 onset L5 onset
0.69 0.63
50.0001** 0.0005**
Relative amplitude Amplitude
0.09 0.15
0.64 0.46
Inter-day stability SD of get up time
0.42 0.47
0.03* 0.02*
Sleep efficiency Mean activity of sleep active epochs
0.46 0.38
0.02* 0.05*
Mean activity of sleep active epochs WASO Sleep latency Sleep duration Sleep efficiency WASO Mean activity of sleep epochs
0.46 0.23 0.50 0.76 0.38 0.35 0.48
0.02* 0.27 0.009* 50.0001** 0.05* 0.08 0.01*
**if p is significant after Bonferonni correction for multiple testing. *if p5 or ¼0.05 with jj40.35 but no longer significant after Bonferonni correction. Chronobiology International
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder Our methodological approach deserves some comments. Although the sample size is in the classical range for actigraphic studies, it may have reduced our ability to achieve significance for some of the parameters. Moreover, statistical considerations regarding the number of tests performed led us to restrict the range of variables examined, leaving some potentially important variables unexamined. The fact that our patients were normothymic may also have altered some results by reducing variance due to the absence of symptoms. However, this was done to avoid the risk of the discrepancies between subjective and objective assessments that have been suggested to be associated with depression (Gonzalez et al., 2013). We should also highlight the heterogeneity of our sample in terms of gender and BD subtypes distributions. Such heterogeneity may reduce the generalizability of our results: for example, sex difference can impact circadian rhythms and sleep parameters (Mong et al., 2011). Finally, some subjective measures do not have a ‘‘strict’’ actigraphic equivalent. For example, we specifically chose the absolute amplitude (based on minimum and maximum motor activity) and the relative amplitude (difference between daytime and night time activity) assessed by actigraphy as CTI L/V proxy measures. Since the languidity scale was designed to measure the amplitude of the underlying oscillator, other objective markers such as melatonin secretion may have been more appropriate. A more systematic assessment of sleep and circadian rhythms with valid and easy to use questionnaires in BD patients could enable us to better evaluate these parameters and follow these disturbances across time. For example, more severe sleep disturbances assessed by self-questionnaire during inter-critical periods were associated with a higher risk of depressive relapse (Franzen & Buysse, 2008). Sleep disturbances have also been associated with cardiovascular risk in BD patients (Soreca et al., 2009). Furthermore, biological rhythm dysfunction has been suggested as a potent predictor of functioning in interepisodic BD patients (Giglio et al., 2010). This non-exhaustive list indicates the crucial need to systematically assess sleep and circadian features in regular practice. This should enable the provision of personalized treatment care plans that integrate sleep/circadian perturbations and that may, in turn, help to improve other components of BD such as proneness to relapses, functioning and physical health. In this context, prospective studies using circadian questionnaires are required to explore the validity of such questionnaires as prognosis markers. BD patients display sleep and circadian disturbances even during remission periods that could impact their prognosis and their global functioning. This external validation study suggested that phase preference and sleep duration can be satisfactorily assessed using the CSM and the PSQI. The external validity of the CTI requires further investigation. Sleep and circadian !
Informa Healthcare USA, Inc.
703
research in BD could benefit from a more systematic use of the PSQI and the CSM in clinical assessments since they are more easily implementable in large-scale cohorts than actigraphy (Henry et al., 2011).
ACKNOWLEDGMENTS We thank the patients who agreed to participate to the study. We thank E. Abadie for her contribution.
DECLARATION OF INTEREST This work was supported by INSERM (Institut National de la Sante´ et de la Recherche Me´dicale), AP-HP (Assistance Publique des Hoˆpitaux de Paris) and the Fondamental foundation (RTRS Sante´ Mentale). The authors 1, 2, 3, 4, 6 and 7 are members of the FondaMental Foundation and as such this work was supported in part by the Investissements d’Avenir, program managed by the ANR under reference ANR-11-IDEX-0004-02. Laboratoires Servier provided funding to purchase actiwatches, but Servier had no role in the design of the protocol, or the acquisition, analysis and/or interpretation of data. The authors report no conflicts of interest related to the topic of this publication. The authors alone are responsible for the content and writing of the paper. The French version of the CTI is available on request: [email protected].
REFERENCES Blais FC, Gendron L, Mimeault V, Morin CM. (1997). Evaluation of insomnia: Validity of 3 questionnaires. Encephale. 23:447–53. Boudebesse C, Lajnef M, Geoffroy PA, et al. (2013a). Chronotypes of bipolar patients in remission: Validation of the French version of the circadian type inventory in the FACE-BD sample. Chronobiol Int. 30:1042–9. Boudebesse C, Leboyer M, Begley A, et al. (2013b). Comparison of five actigraphy scoring methods with bipolar disorder. Behav Sleep Med. 11:275–82. Brooks AT, Krumlauf MC, Whiting BP, et al. (2012). Are you sleeping? Pilot comparison of self-reported and objective measures of sleep quality and duration in an inpatient alcoholism treatment program. Subst Abuse. 6:135–9. Buysse DJ, Reynolds III CF, Monk TH, et al. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Res. 28:193–213. Caci H, Nadalet L, Staccini P, et al. (1999). Psychometric properties of the French version of the composite scale of morningness in adults. Eur Psychiatr. 14:284–90. Etain B, Milhiet V, Bellivier F, Leboyer M. (2011). Genetics of circadian rhythms and mood spectrum disorders. Eur Neuropsychopharmacol. 21:S676–82. Folkard S, Monk TH, Lobban M. (1979). Towards a predictive test of adjustment to shift work. Ergonomics. 22:79–91. Franzen PL, Buysse DJ. (2008). Sleep disturbances and depression: Risk relationships for subsequent depression and therapeutic implications. Dialogues Clin Neurosci. 10:473–81. Gershon A, Thompson WK, Eidelman P, et al. (2012). Restless pillow, ruffled mind: Sleep and affect coupling in interepisode bipolar disorder. J Abnorm Psychol. 121:863–73.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
704
C. Boudebesse et al.
Giglio LM, Magalha˜es PVS, Kapczinski NS, et al. (2010). Functional impact of biological rhythm disturbance in bipolar disorder. J Psychiatr Res. 44:220–3. Gonzalez R, Tamminga C, Tohen M, Suppes T. (2013). Comparison of objective and subjective assessments of sleep time in subjects with bipolar disorder. J Affect Disord. 149: 363–6. Gooneratne NS, Bellamy SL, Pack F, et al. (2011). Case-control study of subjective and objective differences in sleep patterns in older adults with insomnia symptoms. J Sleep Res. 20: 434–44. Grutsch JF, Wood PA, Du-Quiton J, et al. (2011). Validation of actigraphy to assess circadian organization and sleep quality in patients with advanced lung cancer. J Circadian Rhythms. 9:4. Harvey AG, Schmid DA, Scarna` A, et al. (2005). Sleep-related functioning in euthymic patients with bipolar disorder, patients with insomnia, and subjects without sleep problems. Am J Psychiatr. 162:50–7. Henry C, Etain B, Mathieu F, et al. (2011). A French network of bipolar expert centres: A model to close the gap between evidence-based medicine and routine practice. J Affect Disord. 131:358–63. Jones SH, Hare DJ, Evershed K. (2005). Actigraphic assessment of circadian activity and sleep patterns in bipolar disorder. Bipolar Disord. 7:176–86. Kapczinski F, Frey BN, Vieta E. (2011). Sleep and circadian rhythm disturbances in bipolar disorder: An urgent need for objective assessment and systematic follow-up. J Clin Psychiatr. 72:724. Kaplan KA, Talbot LS, Gruber J, Harvey AG. (2012). Evaluating sleep in bipolar disorder: Comparison between actigraphy, polysomnography, and sleep diary. Bipolar Disord. 14:870–9. Kobayashi I, Huntley E, Lavela J, Mellman TA. (2012). Subjectively and objectively measured sleep with and without posttraumatic stress disorder and trauma exposure. Sleep. 35:957–65. Kushida CA, Chang A, Gadkary C, et al. (2001). Comparison of actigraphic, polysomnographic, and subjective assessment of sleep parameters in sleep-disordered patients. Sleep Med. 2: 389–96. Leboyer M, Kupfer DJ. (2010). Bipolar disorder: New perspectives in health care and prevention. J Clin Psychiatr. 71:1689–95. McKenna BS, Eyler LT. (2012). Overlapping prefrontal systems involved in cognitive and emotional processing in euthymic bipolar disorder and following sleep deprivation: A review of functional neuroimaging studies. Clin Psychol Rev. 32:650–63. Milhiet V, Etain B, Boudebesse C, Bellivier F. (2011). Circadian biomarkers, circadian genes and bipolar disorders. J Physiol Paris. 105:183–9.
Millar A, Espie CA, Scott J. (2004). The sleep of remitted bipolar outpatients: A controlled naturalistic study using actigraphy. J Affect Disord. 80:145–53. Mong JA, Baker FC, Mahoney MM, et al. (2011). Sleep, rhythms, and the endocrine brain: Influence of sex and gonadal hormones. J Neurosci. 31:16107–16. Monk R, Kupfer D, Buysse DJ, et al. (1994). The Pittsburgh sleep diary. J Sleep Res. 3:111–20. Montgomery SA, Asberg M. (1979). A new depression scale designed to be sensitive to change. Br J Psychiatr. 134:382–9. Morera-Fumero AL, Abreu-Gonza´lez P, Henry-Benı´tez M, et al. (2013). Chronotype as modulator of morning serum melatonin levels. Actas Esp Psiquiatr. 41:149–53. Morgenthaler T, Alessi C, Friedman L, et al. (2007). Practice parameters for the use of actigraphy in the assessment of sleep and sleep disorders: An update for 2007. Sleep. 30:519–29. Mullin BC, Harvey AG, Hinshaw SP. (2011). A preliminary study of sleep in adolescents with bipolar disorder, ADHD, and non-patient controls. Bipolar Disord. 13:425–32. Nurnberger Jr JI, Blehar MC, Kaufmann CA, et al. (1994). Diagnostic interview for genetic studies. Rationale, unique features, and training. NIMH Genetics Initiative. Arch Gen Psychiatr. 51:849–59; discussion 863–4. Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int. 27:1911–29. Ritter PS, Marx C, Lewtschenko N, et al. (2012). The characteristics of sleep in patients with manifest bipolar disorder, subjects at high risk of developing the disease and healthy controls. J Neural Transm. 119:1173–84. Rocha PMB, Neves FS, Correˆa H. (2013). Significant sleep disturbances in euthymic bipolar patients. Compr Psychiatr. 54: 1003–8. Salvatore P, Ghidini S, Zita G, et al. (2008). Circadian activity rhythm abnormalities in ill and recovered bipolar I disorder patients. Bipolar Disord. 10:256–65. Smith CS, Reilly C, Midkiff K. (1989). Evaluation of three circadian rhythm questionnaires with suggestions for an improved measure of morningness. J Appl Psychol. 74:728–38. Soreca I, Fagiolini A, Frank E, et al. (2009). Chronotype and body composition in bipolar disorder. Chronobiol Int. 26:780–8. St-Amand J, Provencher MD, Be´langer L, Morin CM. (2012). Sleep disturbances in bipolar disorder during remission. J Affect Disord. 146:112–19. Stuifbergen AK, Phillips L, Carter P, et al. (2010). Subjective and objective sleep difficulties in women with fibromyalgia syndrome. J Am Acad Nurse Pract. 22:548–56. Young RC, Biggs JT, Ziegler VE, Meyer DA. (1978). A rating scale for mania: Reliability, validity and sensitivity. Br J Psychiatr. 133:429–35.
Chronobiology International
ORIGINAL ARTICLE
Correlations between objective and subjective sleep and circadian markers in remitted patients with bipolar disorder
1
INSERM, U955, Cre´teil, France, 2AP-HP, Hoˆpital H. Mondor – A. Chenevier, Poˆle de Psychiatrie, Cre´teil, France, Fondation FondaMental, Cre´teil, France, 4Poˆle de psychiatrie, Univ Lille Nord de France, CHRU de Lille, Lille, France, 5 AP-HP, Hoˆpital Fernand Widal, Poˆle Addictologie-Toxicologie-Psychiatrie and Universite´ Paris-7, Paris, France, 6 Faculte´ de me´decine, Universite´ Paris Est, Cre´teil, France, 7Institut de Psychologie, Universite´ Paris Descartes, Paris, France, and 8Body Rhythms and Shiftwork Centre, Swansea University, Swansea, Wales, United Kingdom 3
Bipolar disorder (BD) is a chronic psychiatric condition characterized by recurrences of depressive and (hypo)manic episodes. Patients in remission report a wide range of sleep and circadian disturbances that correlate with several outcomes measures such as functioning or physical health. The most appropriate way to measure these abnormalities in clinical practice requires further investigation since the external validity of self-reports, as compared to more physiological measures (such as polysomnography or actigraphy), has been questioned. Despite the fact that questionnaires are inexpensive, fast and easy to use, they need to be validated against objective measures. This study aims to validate three sleep and circadian questionnaires, namely the Pittsburgh Sleep Quality Index (PSQI), the Composite Scale of Morningness (CSM) and the Circadian Type Inventory (CTI) – against actigraphy in BD patients in remission. Twenty-six carefully assessed BD patients in remission completed the PSQI, the CTI and the CSM, and wore an actigraph (AW7, Camntech) for 21 consecutive days. Phase preference assessed by the CSM strongly correlated with actigraphic phase markers (M10 onset ¼ 0.69 and L5 onset ¼ 0.63). Sleep duration and sleep latency assessed by the PSQI and by actigraphy were also highly correlated ( ¼ 0.76; ¼ 0.50). Moderate correlation coefficients were observed between questionnaires and actigraphy for markers that explored the stability of rhythms, sleep quality, sleep latency and sleep disturbances (jj40.40) although these were not significant after correcting for multiple testing. No correlation was observed between markers for the amplitude of rhythms. While the external validity of the CTI clearly requires further investigation, this study supported the external validity of the CSM and the PSQI for phase preference, sleep duration and latency. We conclude that the CSM and the PSQI could be useful in routine practice and research when actigraphy is not easily available. Keywords: Actigraphy, bipolar disorder, questionnaires, validation
INTRODUCTION
reactivity, cognitive functioning, inflammatory processes and metabolic disturbances (Giglio et al., 2010; McKenna & Eyler, 2012; Soreca et al., 2009). Thus, BP is now conceptualized as a multisystemic disorder with sleep and circadian disturbances as a central physiopathological pivot (Harvey et al., 2005; Leboyer & Kupfer, 2010). In view of this, Kapczinski et al. have recently highlighted the urgent need for the objective assessment of circadian rhythms and sleep in BD patients (Kapczinski et al., 2011). A wide variety of tools are likely to be used to assess sleep or circadian biomarkers, ranging from questionnaires, through actigraphy and polysomnography (PSG) to plasmatic or salivary profiles of neurohormones
20 14
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Carole Boudebesse1,2,3, Pierre Alexis Geoffroy1,3,4, Frank Bellivier3,5, Chantal Henry1,2,3,6, Simon Folkard7,8, Marion Leboyer1,2,3,6, and Bruno Etain1,2,3
Bipolar disorder (BP) is a chronic psychiatric condition commonly conceptualized as cyclically alternating depressive and (hypo)manic episodes separated by periods of symptomatic remission. Although the precise determinants of the disorder remain unclear, part of the pathophysiology is thought to be linked to sleep and circadian disturbances (Leboyer & Kupfer, 2010). Indeed BD patients display disturbances of subjective and objective sleep/circadian markers; especially sleep duration and regularity, cortisol and melatonin secretion profiles and phase preference. All these circadian abnormalities are thought to impact on emotional
Submitted December 20, 2013, Returned for revision February 6, 2014, Accepted February 14, 2014
Correspondence: Carole Boudebesse, AP-HP, Hoˆpital H. Mondor – A. Chenevier, Poˆle de Psychiatrie, Cre´teil, France. E-mail: [email protected]
698
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder (like melatonin or cortisol). However, clinicians may experience difficulties in choosing between all the available assessment tools because of considerations of objective versus subjective assessments, cost constraints in daily practice, availability, invasiveness, acceptability to patients and the external validity of questionnaires. Most biochemical methods, such as cortisol or melatonin plasmatic or salivary secretion profiles, are clearly primarily used for research purposes. PSG – the gold standard for exploring sleep abnormalities – is not readily available in many countries and is mainly used to diagnose sleep disorders (e.g. Obstructive Sleep Apnea). In everyday clinical practice sleep and circadian rhythm assessments may be made by the use of questionnaires (subjective, inexpensive, and easy to interpret) and by actigraphy (objective, more expensive, and more difficult to interpret). These tools are meant to measure sleep parameters, or some characteristics of circadian rhythms (i.e. phase, amplitude or stability), and have already produced promising results in studies of BD patients (Etain et al., 2011; Milhiet et al., 2011). Indeed, actigraphic studies of BD patients have mainly reported abnormalities in terms of sleep duration sleep efficiency, sleep onset latency and wake after sleep onset (Gershon et al., 2012; Harvey et al., 2005; Jones et al., 2005; Kaplan et al., 2012; Millar et al., 2004; Mullin et al., 2011; Ritter et al., 2012; Salvatore et al., 2008; St-Amand et al., 2012). Actigraphy has been validated against PSG in BD patients for several sleep parameters (sleep onset latency, awakenings after sleep onset, total sleep time, and sleep efficiency) (Kaplan et al., 2012). More subjective assessments can use sleep/ circadian questionnaires such as the Pittsburgh Sleep Quality Index (PSQI) which was first developed to assess sleep quality in patients with mood disorders and in patients with sleep disorders (Buysse et al., 1989). When used in samples of BD patients in remission, it showed a high frequency of abnormalities, and in particular with respect to sleep latency and sleep efficiency (Rocha et al., 2013). Two circadian questionnaires can also be used to describe rhythms in terms of their three main characteristics, namely phase, amplitude and stability. The Composite Scale of Morningness (CSM) indicates preferences associated with morning or evening activities (Smith et al., 1989) and is thus considered as a marker of circadian phase. The Circadian Type Inventory (Folkard et al., 1979). CTI describes circadian rhythms as flexible/rigid (reflecting rhythms stability) and as languid/vigorous (reflecting rhythms amplitude). We have recently demonstrated that BD patients in remission assessed with these two questionnaires show an evening preference, and are more languid-type, than healthy controls (Boudebesse et al., 2013a). Before these questionnaires can be used in everyday clinical practice for BD patients more information is !
Informa Healthcare USA, Inc.
699
required about their external validity. The main goal of the present study was thus to correlate selected objective actigraphic measures with scores obtained using subjective questionnaires (CSM, CTI and PSQI) in BD patients in remission.
MATERIALS AND METHODS Population Twenty-six BD outpatients in remission were included after obtaining their written informed consent. The appropriate French ethics committee approved this study. The experimental protocol was conformed to international ethical standards (Portaluppi et al., 2010). The patients were all from the psychiatric department affiliated to the Paris XII University. Primary diagnosis of BD was made according to DSM-IV criteria using a structured interview, namely the Diagnostic Interview for Genetic Studies (DIGS) (Nurnberger et al., 1994). Current psychotropic treatments were systematically recorded. All patients were in remission as defined by both the Mongomery and Asberg Depression Rating Scale (MADRS) (Montgomery & Asberg, 1979) and the Young Manic Rating Scale (YMRS) (Young et al., 1978) (with scores below 8 and the absence of major mood episodes according to DSM-IV criteria within the 3 months prior to the study). The main exclusion criteria for patients were hospitalization or any change in psychiatric medications within the 3 months prior to the study. To ensure that the patients had close to normal sleep–wake patterns, other exclusion criteria included shift work, recent travel with more than 3 h of jet-lag, pregnancy, child birth and recent bereavement (within 2 months prior to the study). Sleep and circadian questionnaires All participants were also assessed with the PSQI, the CSM and the CTI. The PSQI is a 19-item self-assessment questionnaire that examines subjective sleep quality (Buysse et al., 1989). It generates a total score and seven component scores, namely: subjective sleep quality, sleep latency, sleep duration, sleep disturbances, use of sleeping medication and daytime dysfunction. We used a validated French version (Blais et al., 1997). The PSQI had already been validated against actigraphy for patients with lung cancer (Grutsch et al., 2011), but not for BD patients. The Composite Scale of Morningness (CSM) is a 13-item self-assessment questionnaire that examines preferences associated with morning or evening activities (Smith et al., 1989). It is considered as a marker of circadian phase. We used a validated French version of the CSM (Caci et al., 1999). The CTI is an 11-item self-assessment questionnaire that describes circadian rhythms as flexible/rigid (a marker of rhythm stability) and as languid/vigorous
700
C. Boudebesse et al.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
(a marker of rhythm amplitude). For example, rigid types claim to be less able to sleep at unusual hours and languid types claim to be more lethargic following reduced sleep. We have recently validated a French version of this questionnaire (Boudebesse et al., 2013b). The CSM had been previously associated with melatonin plasmatic level in healthy volunteers (MoreraFumero et al., 2013). However, we are unaware of any studies that have validated the CTI and the CSM against actigraphy in any population.
Actigraphy Sleep was objectively assessed by means of actigraphy. An actigraph is a watch-like tool containing an accelerometer that detects, scores, and stores information about the intensity and timing of wrist movements over several consecutive 24-h periods. Actigraphy allows the prospective and objective monitoring of sleep (inactivity) and activity patterns, and can capture the variability in sleep–wake patterns over several consecutive days (Morgenthaler et al., 2007). Participants wore the actigraph device (AW-7 CamNtech) continuously on the non-dominant wrist for 21 consecutives days. Participants were instructed to press the event-marker when they went to bed to sleep and when they got out of bed to start the day. They also completed an adapted version of the Pittsburgh Sleep Diary (Monk et al., 1994). The data were sampled in 1-min epochs and analyzed with the sleep detection algorithm provided by Actiwatch software (Actiwatch Activity & Sleep Analysis Ltd CamNtech 7.28). The following sleep scoring procedure was used: (1) information provided by the event-marker was given priority, (2) if participants forgot to press the event-marker, the missing information concerning bedtime and/or rise time was taken from the diary, (3) visual inspection was used to detect and correct any incongruence between the times provided by the event marker, or the diary, and the actigraph recordings (Boudebesse et al., 2013b). Statistics Data analysis was conducted using JMPÕ Pro 9.0.0 from the SAS Institute. We used Spearman tests for the correlations since most measures were not normally distributed. The correlations were computed between the actigraphy measures and the various scores from the sleep and circadian questionnaires, and the Bonferroni correction for multiple testing was used. We preselected the actigraphic variables that were most likely to measure the same construct as the questionnaires in order to limit the number of correlations computed. All subjective sleep and circadian parameters and actigraphic variables are defined in Table 1. The PSQI sub-components, ‘‘use of sleeping medication’’ and ‘‘day dysfunction’’, were not used since they had no actigraphic equivalents.
RESULTS Population The patients’ mean age was 53.50 years (SD ¼ 11.49) and there were 17 women and 9 men. There were 16 patients with type 1 BD, 8 with type 2 BD and two with BD not otherwise specified. The mean scores on the mood symptoms scales were very low, reflecting euthymia, with a mean MADRS score of 1.85 (SD ¼ 2.80) and a mean YMRS score of 0.65 (1.38). With respect to treatment, all the patients but one took at least one psychotropic medication (lithium n ¼ 12; anticonvulsivants n ¼ 15, atypical antipsychotics n ¼ 7, benzodiazepines n ¼ 4; typical antipsychotics n ¼ 4; antidepressants ¼ 5). Polymedication was fairly common with nine patients taking a combination of two mood-stabilizers. Medication is resumed in Table 2. Correlation between sleep and circadian questionnaires and actigraphy parameters The results are summarized in Table 3. Correlation of phase markers assessed by actigraphy and by the CSM Phase markers assessed by actigraphy (M10 onset and L5 onset) and by the CSM score were strongly correlated (respectively ¼ 0.69 and ¼ 0.63). The tests remained significant (p50.0001) after correcting for multiple testing. Correlation of stability and amplitude markers assessed by actigraphy and by the CTI Although not highly significant, we observed a moderate correlation (jj40.40) between stability measured using actigraphy (inter-day stability and the SD (standard deviation) of get up time calculated over 21 d) and rigidity (CTI F/R score). Flexible BD patients showed more variability in their get up times and lower interday stability, meaning higher variability of their rhythms. There was no correlation between amplitude assessed by actigraphy (relative amplitude or absolute amplitude) and the CTI L/V score. Correlation of sleep markers assessed by actigraphy and by the PSQI subcomponents Subjective and objective estimate measures of sleep duration correlated strongly ( ¼ 0.76) and this remained significant (p50.0001) after correcting for multiple testing. Other correlations between the objective actigraphic measures and the subjective markers measured using the PSQI were of moderate to strong magnitude (jj40.40), although they were not significant after correcting for multiple testing. These findings provide some degree of external validity for the PSQI total score and for the sleep quality, sleep latency and sleep disturbances component scores. Chronobiology International
!
Informa Healthcare USA, Inc.
Sleep disturbances (PSQI component)
Sleep duration (PSQI component) Habitual sleep efficiency (PSQI component)
Estimated duration from bedtime to sleep start Estimated sleep duration Ratio of time spent asleep (total sleep time) to the amount of time spent in bed Subjective sleep disturbances
Sleep latency (PSQI component)
Global indicator of sleep quality
Sleep markers PSQI total score
Subjective sleep quality assessed from good to bad
Indicator of rhythms’ stability
Stability markers CTI F/R score
Sleep quality (PSQI component)
Indicator of rhythms’ amplitude
Eveningness: preference for activity in the evening
Definition
Amplitude markers CTI L/V score
Phase markers CSM total score
Questionnaires
TABLE 1. Definitions of questionnaires and actigraphy markers.
Mean activity of sleep epochs Wake after sleep onset (WASO)
Sleep duration Sleep efficiency
Nighttime motor activity during sleep Wake after sleep onset duration
Duration from sleep start to sleep end Percentage of sleep between bedtime and get up time
Nighttime motor activity during sleep Duration from bedtime to sleep start
Mean activity of sleep active epochs Sleep latency
Wake after sleep onset (WASO)
Nighttime motor activity during sleep Percentage of sleep between bedtime and get up time Wake after sleep onset duration
Strength of coupling of the sleep–wake rhythm to the 24-h regularity in the environment Variability of the get up time between days
Quantifies the difference between daytime (M10) and nighttime (L5) activity levels Difference between the minimum and maximum of the motor activity
Beginning of the 10 most active hours of the 24-h cycle Beginning of the five less active hour of the 24-h cycle
Definition
Mean activity of sleep active epochs Sleep efficiency
standard deviation (SD) of get up time
Inter-day stability
Amplitude
Relative amplitude
L5 onset
M10 onset
Actigraphy markers
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder 701
702
C. Boudebesse et al.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
DISCUSSION We found varying levels of concordance between subjective and objective circadian and sleep assessment. Strong correlations were observed between actigraphic and questionnaires-based measures of circadian phase, sleep latency and sleep duration. Moderate correlations were also observed for some other parameters such as the stability of rhythms, sleep quality, sleep efficiency and sleep disturbances. While these findings support the use of such questionnaires in clinical practice for assessing some sleep/circadian markers in BD patients, further validation studies are clearly needed in larger and independent samples. The use of questionnaires to assess sleep or rhythmicity in clinical practice has long been questioned. In comparison to laboratory measures, such as PSG, they are thought to be too subjective despite their obvious practicality and cost-effectiveness. For example, it has been argued that compared to actigraphy questionnaires are likely to underestimate WASO or to overestimate sleep onset latency. Among in-patients with alcohol dependence, objective and subjective
TABLE 2. Patients’ medication. Treatment
Percentage of patients (%)
Lithium Anticonvulsivants Atypical antipsychotics Typical antipsychotics Benzodiazepines Antidepressants
46 58 27 15 15 19
measures of sleep efficiency correlate, but measures of sleep duration do not (Brooks et al., 2012). Similar misestimates have also been observed in PTSD (Kobayashi et al., 2012), insomnia (Gooneratne et al., 2011) and among women with fibromyalgia (Stuifbergen et al., 2010). However, such discrepancies remain controversial, for example in insomnia sleep-disordered patients (Kushida et al., 2001), leading to the recommendation that subjective data should be used as an adjunct to actigraphic data in estimating total sleep time and sleep efficiency. Little is known about such potential discrepancies in BD patients. Harvey et al. (2005) observed that BD patients, when assessed both objectively and subjectively, tend to overestimate sleep onset latency and underestimate sleep duration. Gonzalez et al. (2013) investigated the correlation between objective (7 d actigraphy) and subjective (estimated via sleep diaries) measures of total sleep time in a sample of 39 patients with bipolar type I disorder and found a significant correlation (r ¼ 0.51). In BD patients, actigraphy and PSG are highly correlated for sleep onset latency, wake after sleep onset, number of awakenings, total sleep time, and sleep efficiency (Kaplan et al., 2012). Thus it is, as yet, difficult to be sure about the external validity of sleep/circadian questionnaires and hence to definitively recommend their use in clinical practice. Nevertheless, our study provides evidence for substantial correlations for phase preference, sleep latency and sleep duration. Further investigations in larger samples are required for other variables such as the stability and amplitude of rhythms, sleep quality, sleep efficiency and sleep disturbances.
TABLE 3. Correlations between actigraphy and questionnaires for circadian and sleep markers in BD patients. Statistics Questionnaires Phase markers CSM total score Amplitude markers CTI L/V score Stability markers CTI F/R score Sleep markers PSQI total score PSQI components scores Sleep quality (PSQI) Sleep latency (PSQI) Sleep duration (PSQI) Habitual sleep efficiency (PSQI) Sleep disturbances (PSQI)
Actigraphy
p Value
M10 onset L5 onset
0.69 0.63
50.0001** 0.0005**
Relative amplitude Amplitude
0.09 0.15
0.64 0.46
Inter-day stability SD of get up time
0.42 0.47
0.03* 0.02*
Sleep efficiency Mean activity of sleep active epochs
0.46 0.38
0.02* 0.05*
Mean activity of sleep active epochs WASO Sleep latency Sleep duration Sleep efficiency WASO Mean activity of sleep epochs
0.46 0.23 0.50 0.76 0.38 0.35 0.48
0.02* 0.27 0.009* 50.0001** 0.05* 0.08 0.01*
**if p is significant after Bonferonni correction for multiple testing. *if p5 or ¼0.05 with jj40.35 but no longer significant after Bonferonni correction. Chronobiology International
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
Sleep and circadian markers in bipolar disorder Our methodological approach deserves some comments. Although the sample size is in the classical range for actigraphic studies, it may have reduced our ability to achieve significance for some of the parameters. Moreover, statistical considerations regarding the number of tests performed led us to restrict the range of variables examined, leaving some potentially important variables unexamined. The fact that our patients were normothymic may also have altered some results by reducing variance due to the absence of symptoms. However, this was done to avoid the risk of the discrepancies between subjective and objective assessments that have been suggested to be associated with depression (Gonzalez et al., 2013). We should also highlight the heterogeneity of our sample in terms of gender and BD subtypes distributions. Such heterogeneity may reduce the generalizability of our results: for example, sex difference can impact circadian rhythms and sleep parameters (Mong et al., 2011). Finally, some subjective measures do not have a ‘‘strict’’ actigraphic equivalent. For example, we specifically chose the absolute amplitude (based on minimum and maximum motor activity) and the relative amplitude (difference between daytime and night time activity) assessed by actigraphy as CTI L/V proxy measures. Since the languidity scale was designed to measure the amplitude of the underlying oscillator, other objective markers such as melatonin secretion may have been more appropriate. A more systematic assessment of sleep and circadian rhythms with valid and easy to use questionnaires in BD patients could enable us to better evaluate these parameters and follow these disturbances across time. For example, more severe sleep disturbances assessed by self-questionnaire during inter-critical periods were associated with a higher risk of depressive relapse (Franzen & Buysse, 2008). Sleep disturbances have also been associated with cardiovascular risk in BD patients (Soreca et al., 2009). Furthermore, biological rhythm dysfunction has been suggested as a potent predictor of functioning in interepisodic BD patients (Giglio et al., 2010). This non-exhaustive list indicates the crucial need to systematically assess sleep and circadian features in regular practice. This should enable the provision of personalized treatment care plans that integrate sleep/circadian perturbations and that may, in turn, help to improve other components of BD such as proneness to relapses, functioning and physical health. In this context, prospective studies using circadian questionnaires are required to explore the validity of such questionnaires as prognosis markers. BD patients display sleep and circadian disturbances even during remission periods that could impact their prognosis and their global functioning. This external validation study suggested that phase preference and sleep duration can be satisfactorily assessed using the CSM and the PSQI. The external validity of the CTI requires further investigation. Sleep and circadian !
Informa Healthcare USA, Inc.
703
research in BD could benefit from a more systematic use of the PSQI and the CSM in clinical assessments since they are more easily implementable in large-scale cohorts than actigraphy (Henry et al., 2011).
ACKNOWLEDGMENTS We thank the patients who agreed to participate to the study. We thank E. Abadie for her contribution.
DECLARATION OF INTEREST This work was supported by INSERM (Institut National de la Sante´ et de la Recherche Me´dicale), AP-HP (Assistance Publique des Hoˆpitaux de Paris) and the Fondamental foundation (RTRS Sante´ Mentale). The authors 1, 2, 3, 4, 6 and 7 are members of the FondaMental Foundation and as such this work was supported in part by the Investissements d’Avenir, program managed by the ANR under reference ANR-11-IDEX-0004-02. Laboratoires Servier provided funding to purchase actiwatches, but Servier had no role in the design of the protocol, or the acquisition, analysis and/or interpretation of data. The authors report no conflicts of interest related to the topic of this publication. The authors alone are responsible for the content and writing of the paper. The French version of the CTI is available on request: [email protected].
REFERENCES Blais FC, Gendron L, Mimeault V, Morin CM. (1997). Evaluation of insomnia: Validity of 3 questionnaires. Encephale. 23:447–53. Boudebesse C, Lajnef M, Geoffroy PA, et al. (2013a). Chronotypes of bipolar patients in remission: Validation of the French version of the circadian type inventory in the FACE-BD sample. Chronobiol Int. 30:1042–9. Boudebesse C, Leboyer M, Begley A, et al. (2013b). Comparison of five actigraphy scoring methods with bipolar disorder. Behav Sleep Med. 11:275–82. Brooks AT, Krumlauf MC, Whiting BP, et al. (2012). Are you sleeping? Pilot comparison of self-reported and objective measures of sleep quality and duration in an inpatient alcoholism treatment program. Subst Abuse. 6:135–9. Buysse DJ, Reynolds III CF, Monk TH, et al. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Res. 28:193–213. Caci H, Nadalet L, Staccini P, et al. (1999). Psychometric properties of the French version of the composite scale of morningness in adults. Eur Psychiatr. 14:284–90. Etain B, Milhiet V, Bellivier F, Leboyer M. (2011). Genetics of circadian rhythms and mood spectrum disorders. Eur Neuropsychopharmacol. 21:S676–82. Folkard S, Monk TH, Lobban M. (1979). Towards a predictive test of adjustment to shift work. Ergonomics. 22:79–91. Franzen PL, Buysse DJ. (2008). Sleep disturbances and depression: Risk relationships for subsequent depression and therapeutic implications. Dialogues Clin Neurosci. 10:473–81. Gershon A, Thompson WK, Eidelman P, et al. (2012). Restless pillow, ruffled mind: Sleep and affect coupling in interepisode bipolar disorder. J Abnorm Psychol. 121:863–73.
Chronobiol Int Downloaded from informahealthcare.com by Francis A Countway Library of Medicine on 03/13/15 For personal use only.
704
C. Boudebesse et al.
Giglio LM, Magalha˜es PVS, Kapczinski NS, et al. (2010). Functional impact of biological rhythm disturbance in bipolar disorder. J Psychiatr Res. 44:220–3. Gonzalez R, Tamminga C, Tohen M, Suppes T. (2013). Comparison of objective and subjective assessments of sleep time in subjects with bipolar disorder. J Affect Disord. 149: 363–6. Gooneratne NS, Bellamy SL, Pack F, et al. (2011). Case-control study of subjective and objective differences in sleep patterns in older adults with insomnia symptoms. J Sleep Res. 20: 434–44. Grutsch JF, Wood PA, Du-Quiton J, et al. (2011). Validation of actigraphy to assess circadian organization and sleep quality in patients with advanced lung cancer. J Circadian Rhythms. 9:4. Harvey AG, Schmid DA, Scarna` A, et al. (2005). Sleep-related functioning in euthymic patients with bipolar disorder, patients with insomnia, and subjects without sleep problems. Am J Psychiatr. 162:50–7. Henry C, Etain B, Mathieu F, et al. (2011). A French network of bipolar expert centres: A model to close the gap between evidence-based medicine and routine practice. J Affect Disord. 131:358–63. Jones SH, Hare DJ, Evershed K. (2005). Actigraphic assessment of circadian activity and sleep patterns in bipolar disorder. Bipolar Disord. 7:176–86. Kapczinski F, Frey BN, Vieta E. (2011). Sleep and circadian rhythm disturbances in bipolar disorder: An urgent need for objective assessment and systematic follow-up. J Clin Psychiatr. 72:724. Kaplan KA, Talbot LS, Gruber J, Harvey AG. (2012). Evaluating sleep in bipolar disorder: Comparison between actigraphy, polysomnography, and sleep diary. Bipolar Disord. 14:870–9. Kobayashi I, Huntley E, Lavela J, Mellman TA. (2012). Subjectively and objectively measured sleep with and without posttraumatic stress disorder and trauma exposure. Sleep. 35:957–65. Kushida CA, Chang A, Gadkary C, et al. (2001). Comparison of actigraphic, polysomnographic, and subjective assessment of sleep parameters in sleep-disordered patients. Sleep Med. 2: 389–96. Leboyer M, Kupfer DJ. (2010). Bipolar disorder: New perspectives in health care and prevention. J Clin Psychiatr. 71:1689–95. McKenna BS, Eyler LT. (2012). Overlapping prefrontal systems involved in cognitive and emotional processing in euthymic bipolar disorder and following sleep deprivation: A review of functional neuroimaging studies. Clin Psychol Rev. 32:650–63. Milhiet V, Etain B, Boudebesse C, Bellivier F. (2011). Circadian biomarkers, circadian genes and bipolar disorders. J Physiol Paris. 105:183–9.
Millar A, Espie CA, Scott J. (2004). The sleep of remitted bipolar outpatients: A controlled naturalistic study using actigraphy. J Affect Disord. 80:145–53. Mong JA, Baker FC, Mahoney MM, et al. (2011). Sleep, rhythms, and the endocrine brain: Influence of sex and gonadal hormones. J Neurosci. 31:16107–16. Monk R, Kupfer D, Buysse DJ, et al. (1994). The Pittsburgh sleep diary. J Sleep Res. 3:111–20. Montgomery SA, Asberg M. (1979). A new depression scale designed to be sensitive to change. Br J Psychiatr. 134:382–9. Morera-Fumero AL, Abreu-Gonza´lez P, Henry-Benı´tez M, et al. (2013). Chronotype as modulator of morning serum melatonin levels. Actas Esp Psiquiatr. 41:149–53. Morgenthaler T, Alessi C, Friedman L, et al. (2007). Practice parameters for the use of actigraphy in the assessment of sleep and sleep disorders: An update for 2007. Sleep. 30:519–29. Mullin BC, Harvey AG, Hinshaw SP. (2011). A preliminary study of sleep in adolescents with bipolar disorder, ADHD, and non-patient controls. Bipolar Disord. 13:425–32. Nurnberger Jr JI, Blehar MC, Kaufmann CA, et al. (1994). Diagnostic interview for genetic studies. Rationale, unique features, and training. NIMH Genetics Initiative. Arch Gen Psychiatr. 51:849–59; discussion 863–4. Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int. 27:1911–29. Ritter PS, Marx C, Lewtschenko N, et al. (2012). The characteristics of sleep in patients with manifest bipolar disorder, subjects at high risk of developing the disease and healthy controls. J Neural Transm. 119:1173–84. Rocha PMB, Neves FS, Correˆa H. (2013). Significant sleep disturbances in euthymic bipolar patients. Compr Psychiatr. 54: 1003–8. Salvatore P, Ghidini S, Zita G, et al. (2008). Circadian activity rhythm abnormalities in ill and recovered bipolar I disorder patients. Bipolar Disord. 10:256–65. Smith CS, Reilly C, Midkiff K. (1989). Evaluation of three circadian rhythm questionnaires with suggestions for an improved measure of morningness. J Appl Psychol. 74:728–38. Soreca I, Fagiolini A, Frank E, et al. (2009). Chronotype and body composition in bipolar disorder. Chronobiol Int. 26:780–8. St-Amand J, Provencher MD, Be´langer L, Morin CM. (2012). Sleep disturbances in bipolar disorder during remission. J Affect Disord. 146:112–19. Stuifbergen AK, Phillips L, Carter P, et al. (2010). Subjective and objective sleep difficulties in women with fibromyalgia syndrome. J Am Acad Nurse Pract. 22:548–56. Young RC, Biggs JT, Ziegler VE, Meyer DA. (1978). A rating scale for mania: Reliability, validity and sensitivity. Br J Psychiatr. 133:429–35.
Chronobiology International
