Free AccessCircadian RhythmsClinical Practice Guideline for the Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders: Advanced Sleep-Wake Phase Disorder (ASWPD), Delayed Sleep-Wake Phase Disorder (DSWPD), Non-24-Hour Sleep-Wake Rhythm Disorder (N24SWD), and Irregular Sleep-Wake Rhythm Disorder (ISWRD). An Update for 2015An American Academy of Sleep Medicine Clinical Practice Guideline R. Robert Auger, MD, Helen J. Burgess, PhD, Jonathan S. Emens, MD, Ludmila V. Deriy, PhD, Sherene M. Thomas, PhD, Katherine M. Sharkey, MD, PhD R. Robert Auger, MD Address correspondence to: R. Robert Auger, MD, Mayo Center for Sleep Medicine, Rochester, MN E-mail Address: [email protected] Mayo Center for Sleep Medicine, Rochester, MN Search for more papers by this author , Helen J. Burgess, PhD Rush University Medical Center, Chicago, IL Search for more papers by this author , Jonathan S. Emens, MD Portland VA Medical Center, Portland, OR Search for more papers by this author , Ludmila V. Deriy, PhD American Academy of Sleep Medicine, Darien, IL Search for more papers by this author , Sherene M. Thomas, PhD American Academy of Sleep Medicine, Darien, IL Search for more papers by this author , Katherine M. Sharkey, MD, PhD Brown University, Providence, RI Search for more papers by this author Published Online:October 15, 2015https://doi.org/10.5664/jcsm.5100Cited by:243SectionsAbstractPDFSupplemental Material ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTA systematic literature review and meta-analyses (where appropriate) were performed and the GRADE approach was used to update the previous American Academy of Sleep Medicine Practice Parameters on the treatment of intrinsic circadian rhythm sleep-wake disorders. Available data allowed for positive endorsement (at a second-tier degree of confidence) of strategically timed melatonin (for the treatment of DSWPD, blind adults with N24SWD, and children/ adolescents with ISWRD and comorbid neurological disorders), and light therapy with or without accompanying behavioral interventions (adults with ASWPD, children/adolescents with DSWPD, and elderly with dementia). Recommendations against the use of melatonin and discrete sleep-promoting medications are provided for demented elderly patients, at a second- and first-tier degree of confidence, respectively. No recommendations were provided for remaining treatments/ populations, due to either insufficient or absent data. Areas where further research is needed are discussed.Citation:Auger RR, Burgess HJ, Emens JS, Deriy LV, Thomas SM, Sharkey KM. Clinical practice guideline for the treatment of intrinsic circadian rhythm sleep-wake disorders: advanced sleep-wake phase disorder (ASWPD), delayed sleep-wake phase disorder (DSWPD), non-24-hour sleep-wake rhythm disorder (N24SWD), and irregular sleep-wake rhythm disorder (ISWRD). An update for 2015. J Clin Sleep Med 2015;11(10):1199–1236.SUMMARYPurposeThe present document replaces/updates the previous American Academy of Sleep Medicine (AASM) Practice Parameters pertaining to the intrinsic CRSWDs (i.e., ASWPD, DSWPD, N24SWD, and ISWRD). The treatment of remaining CRSWDs is not addressed.MethodologyThe AASM commissioned a Task Force (TF) of 4 members with expertise in the field of CRSWDs, appointed a Board of Directors (BOD) liaison, and assigned a Science and Research Department staff member to manage the project. PICO (Patient, Population or Problem, Intervention, Comparison, and Outcomes) questions were developed by the TF and approved by the BOD. Extensive literature searches were performed to identify articles of interest, and relevant data were extracted by the TF. The TF developed consensus-based relevant outcomes, rated their relative importance, and determined clinical significance thresholds. Extracted data were pooled across studies for each outcome measure in accordance with PICO questions, and based upon CRSWD diagnosis, study design, patient population, outcome of interest, and method of derivation. Statistical analyses were performed using dedicated software, and meta-analyses were completed when applicable. The GRADE (Grading of Recommendations Assessment, Development, and Evaluation) approach was used to develop recommendation statements and to determine the direction and strengths of these recommendations based upon a composite assessment of evidence quality, benefits versus harms analyses, and patient values and preferences.FindingsAvailable data allowed for positive endorsement (at a second-tier degree of confidence) of strategically timed melatonin (for the treatment of DSWPD, blind adults with N24SWD, and children/adolescents with ISWRD and comorbid neurological disorders), and light therapy with or without accompanying behavioral interventions (adults with ASWPD, children/ adolescents with DSWPD, and elderly with dementia and ISWRD). Recommendations against the use of melatonin and discrete sleep-promoting medications are provided for demented elderly patients, at a second- and first-tier degree of confidence, respectively. No recommendations were provided for remaining treatments/populations, due to either insufficient or absent data.Recommendations are as FollowsASWPD5.1.4a The TF suggests that clinicians treat adult ASWPD patients with evening light therapy (versus no treatment). [WEAK FOR]DSWPD5.2.6.1a The TF suggests that clinicians treat DSWPD in adults with and without depression with strategically timed melatonin (versus no treatment). [WEAK FOR]5.2.6.2.1a The TF suggests that clinicians treat children and adolescents with DSWPD (and no comorbidities) with strategically timed melatonin (versus no treatment). [WEAK FOR]5.2.6.2.2a The TF suggests that clinicians treat children and adolescents with DSWPD comorbid with psychiatric conditions with strategically timed melatonin (versus no treatment). [WEAK FOR]5.2.9.2a The TF suggests that clinicians treat children and adolescents with DSWPD with post-awakening light therapy in conjunction with behavioral treatments (versus no treatment). [WEAK FOR]N24SWD5.3.6.1a The TF suggests that clinicians use strategically timed melatonin for the treatment of N24SWD in blind adults (versus no treatment). [WEAK FOR]ISWRD5.4.4a The TF suggests that clinicians treat ISWRD in elderly patients with dementia with light therapy (versus no treatment). [WEAK FOR]5.4.5a The TF recommends that clinicians avoid the use of sleep-promoting medications to treat demented elderly patients with ISWRD (versus no treatment). [STRONG AGAINST]5.4.6.1a The TF suggests that clinicians avoid the use of melatonin as a treatment for ISWRD in older people with dementia (versus no treatment). [WEAK AGAINST]5.4.6.2a The TF suggests that clinicians use strategically timed melatonin as a treatment for ISWRD in children/ adolescents with neurologic disorders (versus no treatment). [WEAK FOR]5.4.9.1a The TF suggests that clinicians avoid the use of combined treatments consisting of light therapy in combination with melatonin in demented, elderly patients with ISWRD (versus no treatment). [WEAK AGAINST]ConclusionUse of the GRADE system for this updated Clinical Practice Guideline represents a major change. This update should provide clinicians with heightened confidence with respect to prescribing select treatments and, equally importantly, should serve as a roadmap for future studies that will propel higher quality, more sophisticated therapies for the intrinsic CRSWDs.1.0 INTRODUCTIONThe American Academy of Sleep Medicine (AASM) produced the first Practice Parameters (and associated reviews) for the evaluation and treatment of circadian rhythm sleep-wake disorders (CRSWDs) in 2007.1–3 The purpose of the present publication is to provide an evidence-based update of existing recommendations for the treatment of the intrinsic CRSWDs: advanced sleep-wake phase disorder (ASWPD), delayed sleep-wake phase disorder (DSWPD), non-24-hour sleep-wake rhythm disorder (N24SWD), and irregular sleep-wake rhythm disorder (ISWRD). The extrinsic or predominantly environmentally influenced CRSWDs, namely shift work and jet lag disorder, are not addressed in this paper.2.0 BACKGROUNDReviewed studies that included patients with an explicitly stated CRSWD predominantly utilized the International Classification of Sleep Disorders, Second Edition (ICSD-2)4 diagnostic criteria, despite the fact that International Classification of Sleep Disorders, Third Edition (ICSD-3)5 nomenclature is referenced throughout the manuscript. Important modifications to the International Classification of Sleep Disorders include incorporation of the word “wake” (the ICSD-2 referred solely to circadian rhythm sleep disorders), which highlights the significant impairments these conditions exert on daytime functioning. Caregiver input is also emphasized in the ICSD-3, particularly with respect to diagnostic assessments among cognitively impaired and pediatric patients. Other major changes include the recommendation that CRSWD diagnoses are ascertained via actigraphy derived data when possible (with inclusion of both work/school and free days), to provide objective longitudinal documentation of sleep-wake patterns. Consistent with this emphasis on objective measures, circadian phase assessments (e.g., dim light melatonin onset, or DLMO) are also recommended, if feasible. Other changes include a de-emphasis on “conventional” and “socially acceptable” clock times (recognizing the relative nature of these terms, and instead highlighting patients' subjective concerns), extensive additions to the “Pathology and Pathophysiology” and “Polysomnographic and Other Objective Findings” sections, and new descriptions of “Developmental Issues” and “Clinical and Pathophysiologic Subtypes.”5In many instances, this review incorporated trials with participants who were not recruited in strict accordance with International Classification of Sleep Disorders criteria, but who nonetheless described symptoms consistent with a CRSWD (based upon Task Force consensus). Examples include pediatric/adolescent patients with “idiopathic sleep-onset insomnia,” whose symptoms were consistent with DSWPD, as well as select populations of institutionalized elderly patients, among whom varied descriptions of insomnia, nighttime wakefulness, and daytime napping selectively appeared to be representative of ISWRD, despite the fact that this condition was not named explicitly. A similar approach was taken for this latter group of patients during literature review and development of the previous Practice Parameters.1,3A brief scientific background is required. The two-process model for sleep regulation delineates two principle mechanisms for the governance of sleep and wakefulness: “Process S” and “Process C.”6 The homeostatic drive to sleep (Process S) is proportional to the duration of wakefulness. In contrast, Process C creates a drive for wakefulness that variably opposes Process S and is dependent upon circadian (“approximately daily”) rhythms intrinsic to the individual. Master coordination of this sleep/wake rhythm is provided by the neurons of the suprachiasmatic nuclei located within the hypothalamus.7–10 As this intrinsic period is typically slightly longer than 24 hours in humans11,12 synchronization to the 24-hour day (entrainment) is accomplished by various environmental inputs, the most important of which is light and dark exposure.13 Failure to synchronize can alter the phase relationships between internal rhythms and the light/dark cycle, which may manifest in the form of circadian rhythm sleep-wake disorders (CRSWDs). The intrinsic CRSWDs refer to those conditions that are thought to exist predominantly due to innate phenomena, although exogenous components contribute to varying degrees in all of these disorders.The intrinsic CRSWDs are briefly characterized as follows. DSWPD manifests as a delay of the major sleep episode with respect to the patient's desired timing or the timing required to attend to social, educational, and/or occupational demands. Patients report extreme difficulty both with falling asleep at bedtimes considered typical among their peers, and with waking at the required or desired times, but sleep quality is typically reported as normal when the individual sleeps at the delayed times. In contrast, an advance of the major sleep episode with respect to the patient's desired or required sleep-wake times characterizes ASWPD. ASWPD patients report extreme difficulty staying awake during evening hours and frequently note falling asleep before completion of pertinent work, social, or family obligations. In addition, wake time is undesirably early, and considered atypical in comparison to peers. For both conditions, symptoms must be present for at least 3 months and schedules need to be documented with sleep diaries and/or wrist actigraphy for a period of at least 7 days.N24SWD is diagnosed when patients fail to entrain to the 24-hour light-dark cycle and clock times. Thus, patients exhibit sleep-wake patterns that show a progressive delay (usually) or advance, depending upon the period length (tau) of their endogenous circadian rhythms. During a symptomatic period, the time of high sleep propensity gradually shifts, such that patients experience daytime hypersomnolence and nighttime insomnia. Most patients with N24SWD are totally blind, but this disorder also occurs among sighted individuals. In contrast to the other CRSWDs, a N24SWD diagnosis requires at least 14 days of documentation of progressively shifting sleep-wake times with sleep diaries and/or actigraphy.Patients with ISWRD lack a clear circadian pattern of sleep-wake behavior. Thus, afflicted individuals experience prolonged periods of wakefulness during the typical nocturnal sleep episode in addition to excessive sleepiness and prolonged sleep bouts during daytime hours. Sleep is fragmented and frequently insufficient. ISWRD is observed more commonly among patients with neurodevelopmental or neurodegenerative disorders, and can pose particular challenges for caregivers. Documentation (sleep diaries and/or actigraphy) of multiple non-circadian sleep-wake bouts for a period of at least 7 days is required for diagnosis.Interventions for CRSWDs can be broadly categorized as follows: (1) prescribed timing of sleep-wake and/or physical activity/ exercise, (2) strategic receipt and/or avoidance of light, (3) use of medications and/or supplements to phase shift and/or to promote sleep or wakefulness, and (4) alternate interventions that exert effects by altering bodily functions to impact sleep/wake behaviors (i.e., somatic interventions).Light is strategically timed according to phase response curves (PRCs).2 In brief, light can suppress melatonin secretion14 and phase shift circadian timing in humans,15 leading to the use of timed light exposure as a treatment for CRSWDs. Light timed in the evening and before the core body temperature minimum (CBTmin) leads to phase delays, and light timed after the CBT-min in the morning leads to phase advances.15 Larger effects are observed with greater intensities of light and longer durations of light, but the increases are nonlinear.16,17 Additionally, the response to light is modified by prior exposure to light or “light history,”18,19 such that a history of less light exposure leads to a greater response to light. Just as light exposure can shift circa-dian timing, so too can the strategic avoidance or reduction of light.20,21 Finally, the human circadian system is most sensitive to short wavelength blue light (∼480 nm),22,23 although at bright intensities phase shifts to white broad spectrum light and blue enriched light are similar, presumably due to a saturation of photoreceptors.24,25Less is known about the variables contributing to melatonin response.2 The melatonin PRC is approximately 180 degrees out of phase with the light PRC, such that dosing in the afternoon/ evening shifts rhythms earlier and dosing in the morning shifts rhythms later. As the CBTmin serves as the “inflection point” between delaying and advancing effects for light, the DLMO serves as the approximate inflection point for advancing and delaying effects of melatonin. Optimal dosing of melatonin for circadian effects remains unclear, and studies suggest that timing is more important than dose (PRCs for doses above 5 mg have not been published). In addition to phase shifting effects, melatonin may also have direct soporific effects, particularly at higher doses.3.0 METHODS3.1 Expert Task ForceIn order to develop these Clinical Practice Guidelines, the American Academy of Sleep Medicine (AASM) commissioned a Task Force (TF) of four members with expertise in the field of CRSWDs, appointed an AASM Board of Directors (BOD) liaison, and assigned an AASM Science and Research Department staff member to manage the project. Prior to appointment, the content experts were required to disclose all potential conflicts of interest according to AASM policy. None were declared. The TF performed an extensive review of the scientific literature and assessed the available evidence employing the methodology of evidence-based medicine (specifically, meta-analysis and the Grading of Recommendations Assessment, Development and Evaluation system, or GRADE) to draft recommendations. The present paper was approved by the AASM BOD and replaces the previous Practice Parameters.1 The AASM expects these guidelines to have a positive impact on clinical decision-making and patient outcomes. These recommendations reflect the state of knowledge at the time of publication and will be revised when the availability of new information necessitates.3.2 PICO QuestionsEight PICO (Patient, Population or Problem, Intervention, Comparison, and Outcomes) questions were developed, based on both the inquiries raised in the previous AASM publications1,3 and an investigation of systematic reviews, meta-analyses, and guidelines published subsequently (Table 1). The AASM BOD ultimately approved these questions. In addition, combination treatments were also reviewed for the four intrinsic CRSWDs included in this guideline.Table 1 PICO question parameters.Table 1 PICO question parameters.3.3 Literature SearchesLiterature search #1 was performed in PubMed using broad terms (see Appendix), in order to identify systematic reviews, meta-analyses or relevant practice guidelines published subsequent to availability of the previous AASM Practice Parameters. Examination of discovered papers (n = 93) enabled elucidation of Practice Parameter recommendations requiring revisions, and also assisted with further refinement of the PICO questions. The next literature search (#2) targeted treatment trials involving intrinsic CRSWDs that addressed at least one PICO question. This search utilized PubMed, Embase, and PsycInfo databases.At least two TF members carefully assessed the abstract of each retrieved article (n = 2,063), to determine whether the publication should be included for further consideration. The following list of general exclusion criteria was used: Diagnosis or not treatmentNot CRSWDNot intrinsic CRSWD (shift work or jet-lag disorder)Wrong publication type (review, editorial, etc.)Not human subjectsMechanistic or methodological studyStudy was published before October 2006When there were questions or disagreements, the full text of the article was reviewed in detail until consensus was reached. The same search terms, databases and inclusion/exclusion criteria were used for literature search #3, although new date limitations were applied (June, 2012–March, 2014), with the intention to capture articles published after completion of search #2. Four hundred fifty-three additional publications were retrieved, and TF assessments occurred in the same manner described above. Finally, TF members selected several literature reviews (by consensus), and screened reference lists to identify other articles of potential interest. This “pearling” process served as a “spot control” for the previous searches, and ensured that important articles were not missed. All duplicate references were eliminated.Since new inclusion/exclusion criteria were used in this project, investigations cited in the previous Practice Parameters1 were not necessarily incorporated into the current analysis. Studies that did not meet inclusion criteria were selectively used for discussion purposes, but were neither included in the GRADE reports nor used as a basis for recommendations. The TF made a particular effort to discuss those studies (containing either patients or healthy subjects) that might spur and/or improve future clinical research for the reviewed CRSWDs.A final PubMed search was conducted to identify harms or adverse effects attributed to the relevant interventions: light therapy (PICO 4), hypnotics (PICO 5), and melatonin (PICO 6) (see Appendix). Limitations were imposed to select for English-language “meta-analyses” and “systematic reviews” pertaining to human subjects. The titles and abstracts of articles produced by these searches were reviewed for relevance, and pertinent publications were examined. Other cited articles from the “Harms and Adverse Effects” section were culled from prior searches (but deemed ineligible for quantitative analysis) or were provided via TF members' preemptive awareness and consensus regarding relevancy. Adverse effects of combined treatments were addressed based on the singular components of combinations.3.4 Treatment Efficacy OutcomesDuring the process of data extraction, the TF developed a list of patient-oriented clinically relevant outcomes and rated their relative importance. Physiologic circadian phase markers, total sleep time (TST), initial sleep latency (ISL), sleep onset time (SOT), and sleep offset time (SOffT) were deemed CRITICAL for making recommendations, and a significance threshold was defined for each outcome based upon consensus (Table 2), as no published standards exist. Sleep parameters were alternately evaluated with polysomnography (PSG), wrist actigraphy, or sleep diaries. Although both wakefulness after sleep onset and sleep efficiency were also commonly reported, the two variables were rated as IMPORTANT (but not CRITICAL) by the TF. As such, related data did not factor into clinical recommendations. A unique scenario arose for N24SWD, for which entrainment status (i.e., whether the biological clock is synchronized to the 24-hour day) was solely utilized as a CRITICAL outcome measure, as it defines this CRSWD physiologically (see section 5.3 in this paper).Table 2 CRITICAL outcomes and their TF-defined clinical significance thresholds.Table 2 CRITICAL outcomes and their TF-defined clinical significance thresholds.3.5 Extraction of EvidenceQuantitative data pertaining to the outcomes of interest as well as information necessary for systematic evaluation and grading of the evidence were extracted from accepted articles using a dedicated spreadsheet. Articles determined to lack quantitative data or with data presented in a format incompatible with desired statistical analyses were rejected at this stage, but used selectively for further discussion. In instances where desired data were available but not presented in the desired format, the authors were contacted, and raw data were acquired if possible. Data were pooled across the studies for each outcome measure in accordance with PICO questions and based on diagnosis, study design, patient population, clinical outcome of interest, and method of derivation (e.g., PSG derived data were analyzed separately from data derived from actigraphy or sleep diaries).3.6 Statistical AnalysesMeta-analyses were completed (in the few instances possible) using the random effects model. All computations were performed using the Review Manager software,26 and included calculations of the mean difference (MD) ± standard deviation (SD) for CRITICAL outcomes. Values analyzed in this manner are displayed to the hundredths place. Age demographics and information regarding melatonin doses are presented in the format provided by the associated study (mean ± SD if available) but, in an effort to maintain consistency, are displayed only to the tenths place in instances where the authors provided values with numerical place values of lower hierarchy. The results of meta-analyses are depicted in figures within the text, in association with a “forest plot.” Summary of findings tables for all investigations are presented in the Appendix.When studies contained placebo/control groups, the evaluation of the effect of treatment was performed by comparison of averaged posttreatment and averaged post-placebo/control group values, regardless of the authors' approaches. In studies with crossover or “before-after” designs where there was no placebo/control group, posttreatment values were compared to baseline values. Our use of this methodology occasionally produced results that differed from those reported in the original publications (e.g.27–29).3.7 Interpretation of Clinical Significance of ResultsInterpretation of clinical significance was ascertained via comparisons with predefined thresholds (see Table 2 and Figure 1). For meta-analyses, the pooled MD (black diamond) on “forest plots” depicts the average response or magnitude of effect across all studies, the width of the diamond represents the associated 95% confidence interval (CI), and the “no effect” line represents nil benefit from the intervention. The dotted lines on the left and right sides (equidistant from the “no effect” line) represent clinical decision thresholds defined by the TF (Figures 1A, 1B, and 1C). The right side represents an increase in the outcome measure, while the left represents a decrease. An increase in some outcome measures, such as TST, represents improvement. If the black diamond of TST data lies beyond the clinical significance threshold on the right side, the result of a treatment is interpreted as a clinically significant improvement (Figure 1A). If, however, the diamond lies to the left of the “clinical significance” line, the decrease is regarded as a clinically significant undesired outcome, and the treatment is deemed contraindicated. When improvement is signified by a decrease in the outcome measure (e.g., ISL), the interpretation is reversed.Figure 1: Guide for interpretation of clinical significance of the results.Examples of (A) clinically significant improvement; (B) “serious” imprecision, grade one level down; (C) “very serious” imprecision, grade two levels down.Download FigureWhen the confidence interval crosses the clinical significance threshold on one side, the evidence is graded one level down (Figure 1B) for “serious imprecision.” When the confidence interval crosses the clinical significance threshold on both sides of the no effect line, the evidence is graded two levels down for “very serious imprecision” (Figure 1C). Since the Review Manager software does not operate with clinical significance thresholds, these dotted lines are not depicted in the figures associated with the actual data. The interpretation of clinical significance from results of individual studies was accomplished in the same manner, but forest plots were not created.3.8 Quality of EvidenceThe GRADE approach was used for the assessment of quality of evidence30–37 (also see: http://www.gradeworkinggroup.org/publications/JCE_series.htm). In contrast to other methods, an estimate of effect is generated for critical outcomes across studies, as opposed to an evaluation of individual studies. Multiple aspects of quality of evidence are assessed, with downgrading of evidence as applicable (see Table 3).Table 3 Summary of GRADE approach to rating quality of evidence (from Guyatt et al.30).Table 3 Summary of GRADE approach to rating quality of evidence (from Guyatt et al.30).GRADE assigns high quality to evidence from randomized controlled trials while evidence from observational studies is considered low quality. However, high quality evidence can be graded down, and low quality evidence can be graded up, based upon the factors described below (see Table 3). The analysis of risk of bias includes review of the presence/absence of blinding, allocation concealment, loss to follow up, or selective outcome reporting. Indirectness occurs when the question being addressed is different than the available evidence in terms of population, intervention, comparator, or outcome. There is inconsistency when there is unexplained heterogeneity of the results. Imprecision is described in section 3.7 in this paper.In GRADE, there are 4 specific categories for assessing the quality of a body of evidence.High: corresponds to a high level of certainty that the true effect lies close to that of the estimate of the effect.Moderate: corresponds to a moderate level of certainty in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.Low: corresponds to a low level of certainty in the effect estimate; the true effect may be substantially different from the estimate of the effect.Very low: corresponds to very little certainty in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.The body of evidence for each outcome was assessed and graded, taking into account quality considerations based on the quantitative analysis and other major factors described above. CRITICAL outcome results are presented as summary of findings tables organized by PICO question and patient population (see Appendix, Tables S1–S12).A cumulative quality grade for a particular PICO question and patient population is predicated upon the lowest level of evidence assigned to one of the C
