Performance comparison of peripheral arterial tonometry–based testing and polysomnography to diagnose obstructive sleep apnea in military personnel

Study Objectives: Obstructive sleep apnea (OSA) is increasingly common among military personnel, but diagnostic capabilities are challenged by limited capability for polysomnography (PSG). We sought to evaluate the diagnostic properties of peripheral arterial tonometry (PAT)–based home sleep apnea testing (HSAT) to accurately identify and classify OSA among active-duty military personnel. Methods: This study was a retrospective review of all patients suspected of having OSA who completed an initial PAT-based HSAT followed by confirmatory PSG within 120 days. The diagnostic properties of a PAT-based, HSAT-derived apnea-hypopnea index (AHI) vs a PSG-derived AHI were assessed. Results: Two hundred eight matched pairs of asynchronous studies were analyzed. The prevalence of OSA was 63.5%. PAT-based HSAT overdiagnosed 27.4% of patients with OSA and underdiagnosed 46.6% of patients with OSA. The majority (n = 116, 55.8%) of patients changed OSA severity classification (absent, mild, moderate, severe) after PSG. OSA severity classification concordance between PAT-based HSAT and PSG was observed in 53.4%, 40.5%, 28.6%, and 40.0% of patients with absent, mild, moderate, and severe OSA, respectively. Receiver operating characteristic curve analysis showed an area under the curve of 0.715 and a proposed PAT-based, HSAT-derived AHI cutoff score for OSA diagnosis of 9.0 events/h. This PAT-based, HSAT-derived AHI provided a 52% sensitivity, 83% specificity, 84% positive predictive value, and 50% negative predictive value. Bland-Altman plots showed an unacceptable degree of agreement between PAT-based, HSAT-derived AHI and AHI. Conclusions: There is significant discordance between PAT-based HSAT and PSG among active-duty military personnel evaluated for OSA. PAT-based HSAT may have limited utility for diagnosing OSA and grading severity in this unique patient population.