Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Screening tests are not as accurate as overnight polysomography for the diagnosis of adult sleep apnea


ACP J Club. 2000 Mar-Apr;132:69. doi:10.7326/ACPJC-2000-132-2-069

Source Citation

Ross SD, Allen IE, Harrison KJ, et al. Systematic review of the literature regarding the diagnosis of sleep apnea. Evidence Report/Technology Assessment No. 1. (Prepared by MetaWorks Inc. under contract no. 290-97-0016.) AHCPR publication no. 99-E002. Rockville, MD: Agency for Health Care Policy and Research; February 1999.



How do the accuracies of diagnostic and screening tests (radiologic imaging, laboratory assays, partial-night polysomnography [PSG], daytime PSG, reduced-channel PSG, and clinical signs and symptoms) compare with overnight sleep laboratory PSG for identifying sleep apnea in adults?

Data sources

Studies were identified by searching MEDLINE (1980 to November 1997) and Current Contents (1997) using the terms sleep apnea syndrome; monitoring, physiologic; airway resistance; and index. Bibliographies were scanned.

Study selection

English-, French-, German-, Spanish-, and Italian-language studies were selected if a diagnostic test or intervention was used to support a diagnosis of sleep apnea and ≥ 10 adults with any form of sleep apnea were studied. Studies of patients with other potentially confounding diseases were excluded.

Data extraction

Data were extracted on study quality and description, patient and test characteristics, and results.

Main results

71 studies (7572 participants) compared overnight PSG (diagnostic standard) with portable monitoring devices (25 studies), reduced-channel PSGs (3 studies), daytime PSGs (3 studies) partial-night PSGs (4 studies), oximetry (12 studies), radiologic tests (5 studies), clinical measures (17 studies), chemical assay (1 study), questionnaires (3 studies), and multivariate models (8 studies). Sensitivity and specificity for the tests vary (Table).


Screening tests are not as accurate as over-night polysomnography for diagnosing sleep apnea in adults.

Source of funding: Agency for Health Care Policy and Research.

For correspondence: Dr. S.D. Ross, MetaWorks, Inc., 470 Atlantic Avenue, Boston, MA 02210, USA. FAX 617-368-3590.

Table. Diagnostic and screening tests for sleep apnea in adults using overnight laboratory polysomnography (PSG) as the diagnostic standard*

Test Weighted sensitivity Weighted specificity +LR -LR
Oximetry 87% 65% 2.5 0.2
Partial-night PSG 42% to 93%† 70% to 100%†
Daytime PSG 66% to 100%† 50% to 100%†
Reduced-channel PSG 82% to 94%† 82% to 100%†
FEF50/FIF50 20% 79% 1.0 1.0
Sawtooth sign 62% 63% 1.7 0.6
FEF50/FIF50 and sawtooth sign 39% 61% 1.0 1.0
Global impression 59% 66% 1.7 0.6
Prediction equations 67% 89% 6.1 0.4

*FEF50/FIF50 = measure of extrathoracic airway obstruction. LRs defined in Glossary and calculated from data in article.
†Unweighted data.


The review by Ross and colleagues attempts to determine whether sleep studies without electrophysiologic monitoring of sleep are clinically useful. It includes a serious methodologic flaw: Polysomnography is used as the diagnostic standard, but no evidence has shown that it is such. Polysomnography became accepted as the usual standard for diagnosing sleep apnea in North America, but this conclusion was not evidence based. Studies that have attempted to determine whether electrophysiologic monitoring adds to the accuracy of diagnosing sleep apnea have found that it does not (1). Further, electrophysiologic variables are not the best predictors of improvement with therapy (2).

Despite this flaw, the review is useful because it shows the lack of agreement between diagnostic techniques and highlights the need for more carefully done studies. If the major clinical conclusion of the review is reduced to the statement that no severely symptomatic patient should have the diagnosis excluded on the basis of a negative limited sleep study, then I believe this is a reasonable, although not evidence-based, statement. The finding of a pristine sleep and breathing pattern in all postures in a patient who is known to be asleep remains the best excluder of sleep apnea. This belief has been reinforced by evidence from randomized controlled trials of continuous positive airway pressure (CPAP) therapy showing that large placebo responses can be obtained in patients with sleep apnea (3, 4). Thus, a response to a trial of CPAP cannot be used as evidence of disease, despite recent trends in this direction.

Neil Douglas, MD
Royal Infirmary
Edinburgh, Scotland, UK


1. Douglas NJ, Thomas S, Jan MA. Clinical value of polysomnography. Lancet. 1992;339:347-50. [PubMed ID: 1346422]

2. Bennett LS, Langford BA, Stradling JR, Davies RJ. Sleep fragmentation indices as predictors of daytime sleepiness and nCPAP response in obstructive sleep apnea. Am J Respir Crit Care Med. 1998;158:778-86. [PubMed ID: 9731004]

3. Engleman HM, Kingshott RN, Wraith PK, et al. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep Apnea/Hypopnea syndrome. Am J Respir Crit Care Med. 1999;159:461-7. [PubMed ID: 9927358]

4. Jenkinson C, Davies RJ, Mullins R, Stradling JR. Comparison of therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised prospective parallel trial. Lancet. 1999;353:2100-5. [PubMed ID: 10382693]