Performance of peripheral arterial tonometry–based testing for the diagnosis of obstructive sleep apnea in a large sleep clinic cohort

Abstract

Study Objectives: Peripheral arterial tonometry (PAT)-based technology represents a validated portable monitoring modality for the diagnosis of OSA. We assessed the diagnostic accuracy of PAT-based technology in a large point-of-care cohort of patients studied with concurrent polysomnography (PSG). Methods: During study enrollment, all participants suspected to have OSA and tested by in-laboratory PSG underwent concurrent PAT device recordings. Results: Five hundred concomitant PSG and WatchPat tests were analyzed. Median (interquartile range) PSG AHI was 18 (8-37) events/h and PAT AHI 3% was 25 (12-46) events/h. Average bias was + 4 events/h. Diagnostic concordance was found in 42%, 41%, and 83% of mild, moderate, and severe OSA, respectively (accuracy = 53%). Among patients with PAT diagnoses of moderate or severe OSA, 5% did not have OSA and 19% had mild OSA; in those with mild OSA, PSG showed moderate or severe disease in 20% and no OSA in 30% of patients (accuracy = 69%). On average, using a 3% desaturation threshold, WatchPat overestimated disease prevalence and severity (mean + 4 events/h) and the 4% threshold underestimated disease prevalence and severity by −6 events/h. Conclusions: Although there was an overall tendency to overestimate the severity of OSA, a significant percentage of patients had clinically relevant misclassifications. As such, we recommend that patients without OSA or with mild disease assessed by PAT undergo repeat in-laboratory PSG. Optimized clinical pathways are urgently needed to minimize therapeutic decisions instituted in the presence of diagnostic uncertainty.

Pulse Arterial Tonometry Evaluation of Reliability study

“…Using a 3% desaturation threshold for both peripheral arterial tone AHI and AHI, we found that 20% of the PAT studies overestimated by > 15 events/h and 9% underestimated by > 15 events/h, leading to clinically relevant misclassifications in both directions. 2 Dr. Cook states that the accuracy of 53% in our study represents "a finding that differs significantly from all other published PAT AHI validation studies that reflect correlation above 85%." 1,[3][4][5][6] The reader must be cautioned about this statement, because the existing literature may show a much more nuanced view.…”

“…The argument that one's PSG pulse oximeter "lacks accuracy" does not hold water even without any further analyses, because this type of unidirectional, systematic error would lead to either consistent overestimation or predominant underestimation of the results by PAT vs gold-standard, supervised, and curve-verified PSG oximetry. However, this situation is not what was found in the PATER study, 2 where there were significant percentages of both false alarm and false reassurance. If one compares the mean/median/interquartile range/90th percentile of the hypoxic burden (percentage of total sleep time with SpO 2 < 90%), the results show 11/2/0-10/40% vs 4/0.2/0-2/10% using PSG vs PAT, respectively, or a significantly higher value using PSG (Wilcoxon signed rank test, P < .0001).…”

“…On behalf of the PATER study investigators, we thank Dr. Jennifer Cook, Global Medical Director at Itamar Medicine, for her letter. 1 In the PATER study, 2 we investigated the performance of the WatchPAT200 (Itamar Medical, Ltd., Caesarea, Israel) at the point of care in 500 simultaneous polysomnographic (PSG) in-laboratory recordings. The WatchPAT200 is an approved device based on the pulse arterial tonometry (PAT) signal and is widely used for the diagnosis of obstructive sleep apnea (OSA) by home sleep apnea testing.…”

“…The indices of central tendency for the residuals (PAT-based apnea-hypopnea index 3% -apnea-hypopnea index [AHI] 3% ) were found to be acceptable, showing a median (mean) of bias of +4 events/h (Figure 3 in the study). 2 However, their dispersion, which really tells another story about performance, was far from desirable: The differences were anywhere between -61 and +62, with a standard deviation of the residuals of 15 events/h. Using a 3% desaturation threshold for both peripheral arterial tone AHI and AHI, we found that 20% of the PAT studies overestimated by > 15 events/h and 9% underestimated by > 15 events/h, leading to clinically relevant misclassifications in both directions.…”

Autopilot and algorithms: accidents, errors, and the current need for human oversight

“…4 In this issue of the Journal of Clinical Sleep Medicine, Ioachimescu and colleagues performed concurrent laboratory-based polysomnography and WatchPAT-200 studies comprising 500 veterans, demonstrating that differences in severity levels between the two studies occurred in a sizeable number of patients. 5 The authors noted that 5% of patients with a WatchPAT-based diagnosis of moderate or severe OSA who did not have OSA on laboratory-based polysomnography and the 20% of patients with WatchPAT-diagnosed mild OSA who had moderate or severe OSA on laboratory-based polysomnography. It is of concern that these erroneous diagnoses can lead to poor patient outcomes, either related to potential mistreatment or nontreatment.…”