Investigation of factors affecting the accuracy of portable monitors in Japanese professional drivers with suspected obstructive sleep apnea.

BACKGROUND

Portable monitors (PMs) are widely used for obstructive sleep apnea (OSA) screening in Japan. Appropriate interpretation of PM results and decisions regarding subsequent polysomnography (PSG) are essential to avoid underdiagnosis. However, validation studies of PMs in Japanese populations remain limited. We aimed to evaluate discordance between PM and PSG findings and identify factors contributing to this discrepancy.

METHODS

In this single-center retrospective study, 280 professional drivers were screened using type Ⅳ PMs. Participants with a respiratory event index ≥30 events/h or body mass index (BMI) ≥30 kg/m2 subsequently underwent PSG. Of these, 76.8% had a PSG apnea-hypopnea index (AHI) ≥20 events/h. We assessed the accuracy of the PM-derived 3% oxygen desaturation index (ODI) for predicting PSG AHI and assessed factors associated with PM-PSG discordance across BMI categories.

RESULTS

PM and PSG results showed a moderate correlation (r = 0.561, p < 0.001). Receiver operating characteristic curve analysis identified a PM 3% ODI cutoff of 19.8 events/h for predicting PSG AHI ≥20 events/h (area under the curve = 0.78), with 0.89 sensitivity and 0.60 specificity. A BMI ≥30 kg/m2 was independently associated with PSG AHI ≥20 events/h (odds ratio, 5.33; p = 0.001). Hypopnea-dominant patterns were also associated with PM-PSG discordance.

CONCLUSIONS

Although this study was limited by the high prevalence of OSA in the study population, BMI and hypopnea predominance were important factors for improving the interpretation of PM results. Consideration of these factors may enhance OSA phenotyping in PM-based screening.

TRIAL REGISTRATION: Not applicable.