A preliminary study on application of portable monitoring for diagnosis of obstructive sleep apnea.

Portable monitoring (PM) for diagnosis of obstructive sleep apnea has developed recently. Many studies were performed in the laboratory setting, with simultaneous polysomnographic recordings or required technical assistance in the home setting. And the data were automatically analyzed. In this study, we attempted to evaluate whether PM could be useful in fully unattended home setting, and whether the auto analysis of the data is reliable. Home setting examination by Stardust II, a novel PM device, was performed unattendedly on consecutive 62 patients who complained of snoring. The questionnaire survey on the difficulty of fitting and operation, and the discomfort was conducted by visual analog scale. Automatically and manually analyzed results were compared. The examination was successfully performed by all subjects. The difficulty of fitting and operation, and the discomfort were 2.9+/-1.9, 1.8+/-1.2 and 3.6+/-2.1, respectively. Auto analysis differed significantly from manual analysis not only in apnea/hypopnea index (AHI), but also in the construction of sleep disordered respiratory events. Although AHI in automatic and manual analysis had a good correlation (r=0.949; P<0.001), their agreement was poor, especially in mild and moderate cases. However, setting AHI=50 as a cut-off point in auto analysis, sensitivity and specificity could reach 100% and 92.5%, respectively. Accordingly, PM is useful to identify obstructive sleep apnea in an unattended home setting condition. Considering the significant difference between automatic and manual analysis, we suggest that the data analysis should be performed manually.