Validations of a portable home sleep study with twelve-lead polysomnography: comparisons and insights into a variable gold standard

Abstract

Objectives: Accurate and timely diagnosis for patients with obstructive sleep apnea (OSA) is imperative. Unfortunately, growing interest in this diagnosis has resulted in increased requests and waiting times for polysomnography (PSG), as well as a potential delay in diagnosis and treatment. This study evaluated the accuracy and viability of utilizing SNAP (SNAP Laboratories, LLC, Wheeling, Illinois), a portable home sleep test, as an alternative to traditional PSG in diagnosing OSA.

Methods: This prospective clinical trial included 59 patients evaluated at our institution's sleep laboratory. Concurrent PSG and SNAP testing was performed for 1 night on each patient. Independent, blinded readers at our institution and at an outside-accredited institution read the PSG data, and 2 independent, blinded readers interpreted the SNAP data at SNAP laboratories. The apnea-hypopnea index (AHI) was used to compare the 2 testing modalities. The correlation coefficient, receiver operating characteristic curve analysis, and the Bland-Altman curves, as well as sensitivity, specificity, inter-reader variability, positive predictive value, and negative predictive value, were used to compare SNAP and PSG.

Results: There is a definitive, statistically sound correlation between the AHIs determined from both PSG and SNAP. This relationship holds true for all measures of comparison, while displaying a concerning, weaker correlation between the different PSG interpretations.

Conclusions: There is a convincing correlation between the study-determined AHIs of both PSG and SNAP. This finding supports SNAP as a suitable alternative to PSG in identifying OSA, while accentuating the inherent variation present in a PSG-derived AHI. This test expands the diagnostic and therapeutic prowess of the practicing otolaryngologist by offering an alternative OSA testing modality that is associated with not only less expense, decreased waiting time, and increased convenience, but also statistically proven accuracy.