Clinical Guidelines for the Use of Unattended Portable Monitors in the Diagnosis of Obstructive Sleep Apnea in Adult Patients

Abstract

Free AccessSurgeryClinical Guidelines for the Use of Unattended Portable Monitors in the Diagnosis of Obstructive Sleep Apnea in Adult Patients Portable Monitoring Task Force of the American Academy of Sleep Medicine Portable Monitoring Task Force of the American Academy of Sleep Medicine Search for more papers by this author Published Online:December 15, 2007https://doi.org/10.5664/jcsm.27032Cited by:852SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTBased on a review of literature and consensus, the Portable Monitoring Task Force of the American Academy of Sleep Medicine (AASM) makes the following recommendations: unattended portable monitoring (PM) for the diagnosis of obstructive sleep apnea (OSA) should be performed only in conjunction with a comprehensive sleep evaluation. Clinical sleep evaluations using PM must be supervised by a practitioner with board certification in sleep medicine or an individual who fulfills the eligibility criteria for the sleep medicine certification examination. PM may be used as an alternative to polysomnography (PSG) for the diagnosis of OSA in patients with a high pretest probability of moderate to severe OSA. PM is not appropriate for the diagnosis of OSA in patients with significant comorbid medical conditions that may degrade the accuracy of PM. PM is not appropriate for the diagnostic evaluation of patients suspected of having comorbid sleep disorders. PM is not appropriate for general screening of asymptomatic populations. PM may be indicated for the diagnosis of OSA in patients for whom in-laboratory PSG is not possible by virtue of immobility, safety, or critical illness. PM may also be indicated to monitor the response to non-CPAP treatments for sleep apnea.At a minimum, PM must record airflow, respiratory effort, and blood oxygenation. The airflow, effort, and oximetric biosensors conventionally used for in-laboratory PSG should be used in PM.The Task Force recommends that PM testing be performed under the auspices of an AASM-accredited comprehensive sleep medicine program with written policies and procedures. An experienced sleep technologist/technician must apply the sensors or directly educate patients in sensor application. The PM device must allow for display of raw data with the capability of manual scoring or editing of automated scoring by a qualified sleep technician/technologist. A board certified sleep specialist, or an individual who fulfills the eligibility criteria for the sleep medicine certification examination, must review the raw data from PM using scoring criteria consistent with current published AASM standards.Under the conditions specified above, PM may be used for unattended studies in the patient's home. A follow-up visit to review test results should be performed for all patients undergoing PM. Negative or technically inadequate PM tests in patients with a high pretest probability of moderate to severe OSA should prompt in-laboratory polysomnography.Citation:Collop NA; Anderson WM; Boehlecke B; Claman D; Goldberg R; Gottlieb DJ; Hudgel D; Sateia M; Schwab R. Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients. J Clin Sleep Med 2007;3(7):737–747.INTRODUCTIONThe current standard for clinical practice, established through evidence-based reviews by the American Academy of Sleep Medicine (AASM), is to confirm the diagnosis of obstructive sleep apnea (OSA) with in-laboratory polysomnography (PSG).1 This method has been proven to be accurate with a low failure rate because the study is attended by technical staff; PSG, however, is considered relatively expensive and technically complex. Portable monitoring (PM) has been utilized as an alternative diagnostic test for OSA based in part on the premise that it is less expensive and quicker to deploy compared to in-laboratory PSG. However, there is a paucity of evidence that shows PM is equivalent to PSG in regards to diagnosis, treatment, and outcomes. The available literature typically shows PM can be as accurate as PSG for diagnosis in selected populations; however, in practice it is often used without prior determination of whether the patient is an appropriate candidate for PM.The first practice parameter on PM was published in 1994.2 A subsequent paper on the indications for polysomnography was published in 1997.3 The Agency for Healthcare Research and Quality (AHRQ) reviewed articles and performed a meta-analysis of the literature on the diagnosis of OSA.4 An evidence review and practice parameter was written by a committee composed of representatives of the AASM, American College of Chest Physicians and the American Thoracic Society.1, 5 None of these documents supported broad use of PM due to lack of sufficient evidence. In 2004, the Centers for Medicare and Medicaid Services (CMS) reviewed its national coverage determination (NCD) 240.4 regarding the use of PM as a basis for prescribing CPAP therapy. Their final decision was released in April, 2005 stating that the evidence was not adequate to conclude that “the use of unattended portable multi-channel sleep testing with a minimum of 4 or 7 monitored channels was reasonable and necessary in the diagnosis of OSA; therefore these tests remain uncovered.”6 In 2006, the AASM released an interim position statement7 regarding the use of PM in the diagnosis of OSA in response to an Institute of Medicine report.8 In this statement, the AASM recommended that physicians who choose to use PMs should use them in combination with a clinical assessment and interpret them within the context of a comprehensive evaluation of the patient; that such devices should be used only by AASM-accredited sleep centers or laboratories or by board certified sleep specialists; and that decisions regarding therapy should be based on a comprehensive evaluation of the study results and the patient's symptoms.CMS NCD 240.4 states that inadequate evidence exists to support PM as a diagnostic tool for OSA, and it is not covered as a reasonable and necessary test.6 In 2007 CMS initiated a review of NCD 240.4 at the request of the American Academy of Otolaryngology – Head and Neck Surgery. In its testimony, the AASM presented evidence dismissing the assertion that patients experience unacceptable delays in accessing PSG, discussed the lack of available data on the efficacy of PM in the Medicare population and the lack of economic data in support of PM, and reiterated the AASM position that if PM is accepted as a diagnostic tool, it must be performed under the construct of AASM-accredited facilities or by specialists certified in sleep medicine. A decision from CMS is expected in March 2008.The AASM charged the Portable Monitoring Task Force with a reevaluation of the evidence on PM as an alternative to in-laboratory PSG. The Task Force performed a limited literature search to capture articles published since the last literature review5 and used evidence review and a consensus process to develop clinical guidelines for the use of PM in the diagnosis and management of OSA.METHODSThe Portable Monitoring Task Force was charged with answering the following questions:What are appropriate indications for PM?What types of PM should be used?How should PM data acquisition, analysis, and interpretation be performed?What is the proper application of PM results?The 1994 review2 divided PM into 4 types:Type 1: full attended polysomnography (≥ 7 channels) in a laboratory settingType 2: full unattended polysomnography (≥ 7 channels)Type 3: limited channel devices (usually using 4–7 channels)Type 4: 1 or 2 channels usually using oximetry as 1 of the parametersThat review included approximately 70 studies from 1960 to 1994. Inclusion criteria for studies were: 1) comparison of the PM device to in-laboratory PSG in adults age 18 and over; 2) publication in English; and 3) inclusion of at least 10 subjects.Since the charge of the Portable Monitoring Task Force was not limited to a review of the accuracy of PM compared to PSG, we sought to review a broader range of literature in which PMs were evaluated using patient outcomes, treatment variables, or other parameters. A MEDLINE search was conducted on articles published between 1997 and August 2006 using the following main terms in various combinations: “polysomnography,” “oximetry,” “physiologic monitoring,” and “sleep apnea.” These terms were then combined with “airway resistance,” “upper airway resistance syndrome,” “respiratory disturbance index,” “autoset,” “snoring,” and “respiratory event related arousal.” The combined search was then refined by combining with the following terms: “reproducibility of results,” “predictive value of tests,” “sensitivity,” and “specificity.” Additional searches were then conducted with “polysomnography” combined with the terms “home monitoring” and “home care services.”The search found 291 articles and the Task Force reviewed all abstracts to exclude studies that did not meet the following criteria: subjects ≥ 18 years of age; patient evaluated for OSA; patients had testing with a monitoring device that offered fewer channels (Type 3 devices) than polysomnography; and a minimum of 10 subjects. We restricted our review to Type 3 devices because these are used most frequently in the outpatient setting. The Task Force developed an extraction form to address the specific questions posed above. We also evaluated the monitors with respect to the physiologic signals monitored and devised a new technology classification system to assist us in assessing what leads are most valuable:Technology Used in Portable MonitoringOximetryRespiratory monitoring, including but not limited to:EffortAirflowSnoringEnd-tidal CO2Esophageal pressureCardiac monitoring, including but not limited to:Heart rate or heart rate variabilityArterial tonometryMeasures of sleep wake activity, including but not limited to:ElectroencephalographyActigraphyBody positionOtherReview of the 291 articles resulted in 36 meeting inclusion criteria9–43; abstraction data is summarized in the Evidence Table. A decision was made to review an additional paper published in 2007 as it had important outcome data.44 Therefore, the Evidence Table (available online at www.aasmnet.org/jcsm) shows data from 37 papers. The data were abstracted by an independent reviewer and reviewed for accuracy by a member of the Task Force. The Task Force included this data as well as data from all previous reviews in developing recommendations for the use of PMs.The Task Force held a face-to-face meeting to develop consensus-based guidelines. Each of the 4 questions posed were reviewed in detail by Task Force members and, using a modified nominal group technique, statements were developed and approved by the group. The AASM Board of Directors approved this guideline.RECOMMENDATIONS1. Indications for Portable Monitoring1.1. PM for the diagnosis of OSA should be performed only in conjunction with a comprehensive sleep evaluation. Clinical sleep evaluations using PM must be supervised by a practitioner with board certification in sleep medicine or an individual who fulfills the eligibility criteria for the sleep medicine certification examination. In the absence of a comprehensive sleep evaluation, there is no indication for the use of PM.The Task Force recommends that the comprehensive evaluation of patients follow the AASM Standards for Accreditation of Sleep Disorders Centers45 specifically with regard to the role of a sleep specialist board certified in sleep medicine, patient acceptance criteria, and quality assurance. This recommendation is consistent with the 2003 practice parameter defining indications for PSG.46 This recommendation emphasizes the role of a complete diagnostic evaluation to establish a differential diagnosis of sleep disorders. Any consideration of PM applicability must be taken in the context of this evaluation process. This recommendation is based on evidence as indicated in the Evidence Table. The majority of the study designs in the literature reviewed included a screening evaluation (stated in 32 articles). Seven studies specified physician involvement in pretest evaluation. Screening measures (stated in 34 studies) referenced snoring, suspected sleep related breathing disorders or sleepiness. Exclusion criteria were stated in 19 studies (4 “other sleep diagnoses,” 3 “logistics,” 6 “other cardiopulmonary diagnoses or supplemental oxygen,” and 6 “general medical or technical limitations”).It is the consensus of the Task Force that the clinical evaluation should be performed by a board certified sleep specialist or an individual who fulfills the eligibility criteria for the sleep medicine certification examination, and that interpretation of the PM study, supervision, and quality assurance be the responsibility of a sleep specialist board certified in sleep medicine as is required for sleep center accreditation.45 This is because the skill set required for board certification includes: an understanding of differential diagnosis of a broad array of symptoms; interpretation of PM results and sources of error; and the ability to use the PM results in the context of the individual patient's history and physical examination. There are currently 2 recognized sleep medicine certification pathways: the American Board of Sleep Medicine and the American Board of Medical Specialties. Either of these pathways meets the requirements for sleep medicine certification.The PM study should be 1 tool in the complete evaluation of the sleep disorders patient. This recommendation is an extension of the prior practice parameter for PM (section 16)1 emphasizing the specialty training of the interpreting physician. The consensus of the Task Force is that PM interpretation must be supervised by a trained sleep physician who must have access to the raw data. Although most of the studies reviewed were conducted at sleep centers, 26 of the 37 studies were performed outside the United States so that the effect of board certification and sleep center accreditation cannot be assessed.A study by Parthasarathy and colleagues47 reported that the absence of accreditation or provider certification was associated with higher rates of PAP discontinuation (odds ratio = 1.9). Improved patient education and treatment of nasal congestion by the certified physicians and accredited center personnel was associated with increased treatment utilization and patient satisfaction. It is critical to recognize that the utility of unattended portable monitoring in the diagnosis of OSA rests on more than the recording accuracy of a portable device. Historically, the field of sleep medicine has placed a high priority on the provision of comprehensive clinical care to patients with sleep disorders. Specifically, the standards for care emphasize that polysomnographic evaluation should only occur within the context of a full evaluation of the patient by a trained expert in sleep medicine. Comprehensive clinical assessment ensures several important facets of care:Appropriate health care utilization. By providing skilled assessment prior to study, sleep medicine clinicians ensure that use of portable monitoring is appropriate for a given patient, thereby avoiding overutilization or application of PM when attended study or alternate diagnostic assessments should be performed.Comprehensive diagnostic assessment. Patients undergoing PM for suspected OSA frequently present with comorbid medical and psychiatric conditions as well as other sleep disorders. Comprehensive evaluation is necessary to ensure that these comorbid conditions are reliably identified and addressed in a comprehensive therapeutic approach.Accurate data collection and scoring. Comprehensive sleep programs, particularly those accredited by the AASM, are expected to demonstrate adequate training of technologists, effective patient education regarding application and use of PM, and ongoing quality assessment programs that will maximize data quality, patient safety and satisfaction, and outcome. Accurate scoring is an additional consideration, which is addressed in a later section.Effective patient management. Positive outcomes for patients with OSA depend on adequate diagnosis as well as effective treatment planning and follow-up. Comprehensive sleep centers maintain the necessary organizational structure and the administrative, technical, and professional personnel to provide these services. Demonstration of effective therapy is often incorporated into patient management in sleep laboratories.1.2. Provided that the recommendations of 1.1 have been satisfied, PM may be used as an alternative to polysomnography (PSG) for the diagnosis of OSA in patients with a high pretest probability of moderate to severe OSA. PM should not be used in the patient groups described in 1.2.1, 1.2.2, and 1.2.3 (those with comorbidities, other sleep disorders, or for screening).The evidence to date shows that PM studies have been predominantly performed in high risk populations for moderate to severe OSA. The Task Force recommends that PM use should be limited to these groups. Clinical judgment remains the best method for determining OSA risk. The clinician must take into account the essential features of OSA: demographics; predisposing and precipitating factors; clinical features; and familial patterns. Treatment decisions must also rely on the judgment of an experienced clinician. PM results should be combined with clinical evaluation in determining whether additional testing is required or treatment should be initiated. The majority of studies reviewed included patients screened as “suspected OSA” as an entry criterion (see Evidence Table). The AASM practice parameter paper reviewed indications for polysomnography in the diagnosis of OSA.46 Risk factors included snoring, sleepiness, obesity, and witnessed apneas. All of these factors were strongly associated with OSA, and the severity of the risk factors often correlated with the severity of OSA. However, the predictive value of individual and combined risk factors is only moderate. The authors also reviewed clinical predictive models and concluded that none were sufficient to predict severity of OSA.No study has been specifically designed to distinguish mild from severe disease. The majority of studies evaluate patients with a high pretest probability for OSA, thereby eliminating those with mild disease. Therefore, no systematic research has been done to determine the discriminatory capacity of various PM devices to detect low levels of OSA vs. high levels of OSA.The taskforce recommendations specifically apply to adult populations. In addition, there are little data on the use of PM in the pediatric and older (> 65 years of age) populations. Most studies have been done in middle-aged adults; PM use in older patients who are more likely to have both comorbid conditions and comorbid sleep disorders should be approached cautiously. Clearly, more research is needed in these populations.1.2.1. PM is not appropriate for the diagnosis of OSA in patients with significant comorbid medical conditions that may degrade the accuracy of PM, including, but not limited to, moderate to severe pulmonary disease, neuromuscular disease, or congestive heart failure.Only 2 of the studies reviewed did not exclude patients with comorbid medical disorders.15, 43 The other 35 studies either excluded patients with comorbid medical disorders or did not state exclusion criteria. No new data has been published on this topic since the 2003 guidelines.1 Use of PM devices should be restricted to populations with data supporting its diagnostic accuracy, and therefore in-laboratory PSG remains the standard for patients with co-morbid medical disorders.1.2.2. PM is not appropriate for the diagnostic evaluation of OSA in patients suspected of having other sleep disorders, including central sleep apnea, periodic limb movement disorder (PLMD), insomnia, parasomnias, circadian rhythm disorders, or narcolepsy.This recommendation is consistent with the previous recommendations1; no new data are available to evaluate PM in patients with central sleep apnea or OSA with comorbid sleep disorders. In-laboratory PSG should be used in patients suspected of central sleep apnea or hypoventilation syndromes because there are no data evaluating the accuracy of PM devices for the detection of central apneas or hypoventilation. Furthermore, PM does not include data necessary to reach diagnostic criteria for PLMD, parasomnias, circadian rhythm disorders or narcolepsy.48 PM is not an appropriate methodology for the diagnosis of circadian rhythm disorders.1.2.3. PM is not appropriate for general screening of asymptomatic populations.It was the consensus of the Task Force that PM is not appropriate for general screening at this time. Even if screening may be appropriate for asymptomatic individuals in high risk populations (such as congestive heart failure,49,50 hypertensives,51,52 commercial truck drivers or patients undergoing bariatric surgery53) currently available PM devices are not acceptable tools. They have only been shown to have good specificity and sensitivity in populations evaluated by sleep specialists, considered to be at high risk for OSA based on clinical symptoms and without significant comorbid medical disorders or suspicion of comorbid sleep disorders. Although it was the consensus of the Task Force that there is not yet sufficient evidence to guide the use of PM in general screening even of high-risk populations, it is recommended that if such screening is performed, appropriate clinical assessment tools should be used to address potential false positives and false negatives.1.3. PM may be indicated for the diagnosis of OSA in patients for whom in-laboratory PSG is not possible by virtue of immobility, safety, or critical illness.This recommendation is a modification of the previous practice parameter.1 PM may be used when other forms of sleep evaluation are not possible, and, as stated in the previous paper, “clinical judgment made by the physician in light of individual circumstances has to be applied to individual patients.”1.4. PM may be indicated to monitor the response to non-CPAP treatments for obstructive sleep apnea, including oral appliances, upper airway surgery, and weight loss.This recommendation is based on Task Force consensus. PM may be used to monitor the efficacy of therapies other than CPAP when the diagnosis of OSA has already been made, either through PM or in-laboratory PSG.Summary of Indications for Portable Monitoring:Figure 1 summarizes a pathway for patients under consideration for PM. Patients appropriate for PM must be high risk for OSA and not have comorbid medical disorders or comorbid sleep disorders. Failure to meet these criteria in patients that are high risk for moderate to severe OSA should lead to an in-laboratory PSG. In laboratory PSG is also the standard in patients with other sleep disorders, such as narcolepsy and central sleep apnea.Figure 1 Flow chart depicting recommended pathway of patients considered for PM. Patients appropriate for PM should have moderate to high risk for OSA, have no comorbid medical conditions and no comorbid sleep disorders. Patients not considered appropriate for PM should have in-laboratory polysomnography. (BCSS = Board Certified Sleep Specialist or an individual who fulfills the eligibility criteria for the sleep medicine certification examination)Download Figure2. Technology for Portable Monitors2.1. At a minimum, the PMs must record airflow, respiratory effort, and blood oxygenation. The type of biosensors used to monitor these parameters for in-laboratory PSG are recommended for use in PMs.The previous reviews and the work of this Task Force uncovered little data on the validity and reliability of Type 2 PM devices. Therefore, our assessment focused on Type 3 devices. With the proliferation of these devices, the validity of a scheme based on the number of channels is no longer clear. The Task Force chose instead to focus on the types of signals used rather than their number. Although some monitors have not been adequately tested in the home environment, the Task Force chose to apply AASM recommendations for in-laboratory sensors to PM. The AASM Task Force on Respiratory Scoring recently completed a review of signals used in the detection of sleep related breathing disorders.54 The review was used to determine criteria for scoring respiratory events in the AASM Manual for the Scoring of Sleep and Associated Events.55 Technical considerations as well as the consensus of this Task Force support measurement of airflow, respiratory effort and blood oxygenation.The Task Force evaluated other sensor types in great detail. Four studies in the current review14,29,31,43 evaluated a PM that uses arterial tone, actigraphy, and oximetry. The evidence for their use has been rated level B or C in past reviews.57 Bar and colleagues14 reported good accuracy in 15 home unattended studies for respiratory disturbance indexes (RDIs) of 10 or 20 per hour with receiver operating characteristic areas under the curve of 0.82 and 0.87, respectively. In 30 patients studied both in-laboratory and at home31 and in a larger study of 98 individuals studied at home,43 similar degrees of accuracy were found in preselected populations of OSA and non-OSA subjects. However, Penzel and colleagues29 found a significant technical failure rate (4 of 21 or 19%). None of the articles noted any issues of safety, discomfort, or patient application. Unlike most PM devices, the arterial tone device uses a proprietary algorithm for scoring; although review of the raw data is possible, manual scoring is not.One article assessed the time of transmission of an arterial waveform from the ECG to oximeter-derived pulse sensation (pulse transit time) in 13 patients at home.30 Automated scoring could not distinguish normal from mild to moderate OSA, or mild to moderate OSA from severe OSA accurately. Ectopic beats were clearly felt to affect the results.End-tidal CO2, considered a standard polysomnographic measure in pediatric patients, was evaluated in 1 study.18 Stroke patients were screened for OSA in the ICU, where end-tidal CO2 monitoring is commonly available. The study was limited by the use of another PM as the gold standard rather than in-laboratory PSG, and inclusion of central with obstructive events in the AHI.Two investigations included the use of esophageal manometry13,28 as part of a PM array of bioparameters. In this context the technique had relatively poor sensitivity (64%) and specificity (78%) in the detection of OSA and is not recommended for routine use.Addition of biosensors that assist in determining sleep/wake state would clearly improve accuracy and provide a more accurate denominator for the apnea-hypopnea index. However a previous evidence review6 described only 3 studies, 2 in-laboratory and 1 at home, that could be used to provide support that this improves accuracy. Despite its overwhelming face validity, the subsequent practice parameter requested further research be done before Type 2 monitors could be recommended. Although numerous studies using unattended home polysomnography to assess sleep disordered breathing have been published since the last evidence review, there is a lack of new information comparing type 2 monitors to laboratory PSG. Therefore, the current committee only evaluated a PM if it did not include a measure of sleep stage. Our review also found that actigraphy28,29 was not a sufficiently accurate substitute measure of sleep time to recommend its routine use.2.2. The sensor to detect apnea is an oronasal thermal sensor and to detect hypopnea is a nasal pressure transducer. Ideally, PMs should use both sensor types.The recommendation for use of the thermal sensor is based on limited evidence and consensus of the Task Force on Respiratory Scoring.55 The use of a nasal pressure transducer is supported by consistent Level 1 to 5 evidence and consensus agreement of the Task Force on Respiratory Scoring. Both of these recommendations are based on in-laboratory studies.Previous literature review reveals that as a measure of airflow, nasal pressure is less accurate than pneumotachometer but more accurate than thermal sensors (thermocouples and thermistors).56 In spite of this, the most common signal used in portable monitors has been airflow measured by thermistor.5 These flow sensors have been shown to be nonlinearly related to actual airflow and may even overestimate ventilation. Although nasal pressure devices may be superior to thermistors for detection of flow, they are limited to only nasal flow assessment, leaving mouth flow undetected. The signals may be significantly dampened in mouth breathers. The current recommendation for use of both sensor types reflects these considerations. At the time of the earlier evidence review, nasal pressure monitors had not been tested in the unattended home setting,5 and the current literature review sheds no further light on the relative performance of PM devices using nasal pressure versus thermal sensors.2.3. Ideally the sensor for identification of respiratory effort is either calibrated or uncalibrated inductance plethysmography.This recommendation is based on consistent Level 1 to 5 evidence and consensus agreement of Task Force on Respiratory Scoring.55 Recommendations for techniques for in-laboratory PSG used to measure chest and abdominal effort were reviewed in 1999.56 These techniques are still current and include respiratory inductive plethysmography, piezo sensors, strain gauges, and thoracic impedance. None of these techniques have been thoroughly evaluated during PM. With the exce