|
Does this patient have obstructive sleep apnea? The Rational Clinical Examination systematic review |
Myers KA, Mrkobrada M, Simel DL |
|
|
CRD summary This review concluded that the clinical examination of patients with suspected obstructive sleep apnoea was useful for selecting patients for more definitive testing – nocturnal gasping or choking was the most reliable indicator. The authors' discussion of the evidence was appropriately cautious, but this was not reflected in their overall conclusions, which were overoptimistic. Authors' objectives To review the accuracy of clinical examination in diagnosing obstructive sleep apnoea. Searching MEDLINE was searched to June 2013; search terms were reported. Text books, review articles, and reference lists were searched. Published papers, in English, were considered for inclusion. Study selection Studies that compared clinical examination with full, attended, nocturnal polysomnography, as the reference standard, for diagnosing obstructive sleep apnoea, were included. Studies had to report sufficient data to calculate the likelihood ratios for at least one symptom or sign associated with sleep apnoea. Studies that included patients who were referred for a range of other suspected disorders were excluded, as were studies in which the diagnostic threshold was an apnoea-hypopnoea index of more than 15 events per hour. About three quarters of the included studies focused on referred or selected participants; the remaining quarter were based on a community population. The main outcome was the accuracy of symptoms and signs, or composite tools, for the diagnosis of obstructive sleep apnoea. The symptoms evaluated included: nocturnal choking or gasping; morning headache; reported apnoea; excessive daytime sleepiness; and snoring. Signs assessed included: Mallampati classification; pharyngeal narrowing; and overall clinical impression. Composite clinical measurement tools included: the STOPBang questionnaire; the Snoring Severity Scale; the Berlin Questionnaire; and the Neck, Airway, Comorbidities, Epworth, and Snoring (NAMES)-2 assessment. Most studies were conducted in Europe or North America, and three quarters of participants were men. Where reported, the mean participant age was generally between 40 and 55 years. One author reviewed titles and abstracts for inclusion. Assessment of study quality A published quality grading system, developed for the Rational Clinical Examination Series, was used. Studies were assigned a level from 1 to 5, based on the number of participants, prospective or retrospective design, consecutive recruitment, use of a gold standard test, and diagnostic status of patients. It seems that two reviewers assessed study quality. Data extraction Study characteristics including the apnoea-hypopnoea index threshold, age, body mass index, and signs or symptoms with definitions, where reported, were extracted for each paper. It was unclear how many reviewers extracted the data. Methods of synthesis For the clinical symptoms and signs, the authors calculated summary positive and negative likelihood ratios, sensitivity, specificity, and positive and negative predictive values. Studies were grouped according to apnoea-hypopnea index threshold. Where only one study was available, the point estimate and confidence interval was reported. The findings from two studies were summarised as a range. The findings from three studies were summarised using a univariate random-effects model. The findings evaluated in four or more studies were summarised using a bivariate random-effects model; univariate measures were used when the model did not converge on a solution. Heterogeneity was explored using Ι² for different apnoea-hypopnoea index thresholds. As confidence intervals for the likelihood ratios were considered to be narrow, single summary likelihood ratios were presented for studies with apnoea-hypopnoea index thresholds of 10 events per hour or higher, and those with 15 events per hour or higher. Results of the review In total, 42 studies were included and 10 of them were of community populations. Sample sizes ranged from 42 to 2,677 people. Four community studies of sleep physicians’ ability to diagnose obstructive sleep apnoea, based on history and physical examination, were judged to be level 1 quality (high). Six studies at level 1 reanalysed data from these four studies, and 32 studies at level 3 were of referred patients. The prevalence of sleep apnoea in community-screened patients ranged from 2% to 14% (sample sizes 360 to 1,741), and in referred patients, it ranged from 21% to 90% (sample sizes 42 to 2,677). Individual symptoms: At thresholds of 10 or 15 events per hour, nocturnal choking or gasping was the most useful diagnostic symptom (LR+ 3.3; 95% CI 2.1 to 4.6; five studies). Other individual symptoms had likelihood ratios of less than 2.0, at thresholds of 10 or 15 events per hour or higher, and were judged not to be useful predictors of obstructive sleep apnoea. These included snoring (six studies), excessive daytime sleepiness (10 studies), morning headache (four studies), reported apnoea (nine studies), and oropharyngeal examination (five studies). Clinical impressions: Four studies assessed sleep medicine specialists’ ability to diagnose apnoea, based on the patient's history and physical examination. The pooled positive likelihood ratio was not more than two (LR+ 1.7, 95% CI 1.5 to 2.0; Ι²=0; LR- 0.67, 95% CI 0.60 to 0.74; Ι²=10%; sensitivity 58%; specificity 67%). Composite tools based on symptoms and signs: Three studies evaluated the STOP-Bang questionnaire, one evaluated the Snoring Severity Scale plus body mass index, four evaluated the Berlin Questionnaire, and one evaluated the NAMES and NAMES-2 tools. None of the combinations of findings or measurement scales, at either of the selected diagnostic thresholds, was judged to make the diagnosis of sleep apnoea more likely, with summary likelihood ratios of less than two, at index thresholds of five and 15 events per hour or higher. Authors' conclusions Nocturnal gasping or choking was the most reliable indicator of obstructive sleep apnoea. Clinical examination of patients with suspected obstructive sleep apnoea was useful for selecting patients for more definitive testing. CRD commentary The review addressed a clear question with reasonable inclusion criteria. The literature searches were limited to one database and excluded unpublished papers and those not in English. It was unclear if two reviewers were independently involved in each review stage, making it difficult to rule out reviewer error and bias. Studies were assessed for quality, but few results were presented, making it impossible to evaluate the bias in the studies. The included studies were suitably grouped by clinical variation and diagnostic criteria, but significant heterogeneity remained for some analyses and the positive likelihood ratios were generally low. The authors discussion of the evidence was appropriately cautious, but this was not reflected in their overall conclusions, which were overoptimistic. Implications of the review for practice and research Practice: The authors stated that recently published multi-item questionnaires could be useful in ruling out obstructive sleep apnoea, but they could not be considered helpful in positively identifying affected patients. Research: The authors stated that the Sleep Apnoea Clinical Score (SACS) was a promising tool for the identification of patients likely to have obstructive sleep apnoea, and this required validation by primary care clinicians, in the general population, using the usual diagnostic thresholds. Bibliographic details Myers KA, Mrkobrada M, Simel DL. Does this patient have obstructive sleep apnea? The Rational Clinical Examination systematic review. JAMA 2013; 310(7): 731-741 Indexing Status Subject indexing assigned by NLM MeSH Airway Obstruction /etiology; Body Mass Index; Humans; Medical History Taking; Physical Examination; Polysomnography; Prevalence; Referral and Consultation; Sleep Apnea, Obstructive /complications /diagnosis /epidemiology; Snoring /etiology AccessionNumber 12013049250 Date bibliographic record published 22/08/2013 Date abstract record published 09/09/2013 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|