Multi setting evaluation of Watch-PAT in obstructive sleep apnea diagnosis in patients with atrial fibrillation.

Abstract

Study objectives: Patients with atrial fibrillation (AF) have elevated risk of obstructive sleep apnea (OSA). A peripheral arterial tonometry (PAT)-based device, the Watch-PAT (WP) has been increasingly used as a home OSA test. We conducted a multi-setting evaluation of WP in OSA diagnosis by comparing results from in-lab PSG, in-lab WP, and home WP in AF patients.

Methods: AF patients underwent an in-lab PSG with concurrent in-lab WP and two consecutive nights of home WP. OSA was categorized into no OSA, mild, moderate, and severe OSA by apnea hypopnea index (AHI; <5, ≥5-<15, ≥15-<29, ≥30). The primary outcome was OSA severity agreement between in-lab PSG vs. home WP. Secondary outcomes included the agreement between in-lab PSG vs. in-lab WP, as well as in-lab WP vs. home WP. Night-to-night variability of consecutive nights of home WP was also evaluated. Results were expressed by Cohen's Kappa (κ) for AHI scoring rules 1a and 1b.

Results: 27 patients were included for analysis. There was slight and fair agreement between in-lab PSG and home WP (κ=0.15, 0.40 for score rule 1a and 1b, respectively), substantial agreement between in-lab PSG vs. in-lab WP (κ=0.63, 0.71), moderate agreement between in-lab WP and home WP (κ=0.59, 0.59), and almost perfect agreement between first night home WP and second night home WP (κ=0.87, 0.89).

Conclusions: We demonstrate overall slight to fair agreement between home WP vs. in-lab PSG, suggesting inadequate performance against the standard reference test. This should be considered when interpreting the results of home WP. Other evaluations demonstrated moderate agreement for in-lab WP and home WP in the classification of OSA severity, substantial agreement between in-lab PSG and in-lab WP, and almost perfect agreement in the comparison both home WP.