Validation of ApneaLink Plus for the diagnosis of sleep apnea

Abstract

PURPOSE: The aim of this study is to evaluate the validation of ApneaLink Plus (ALP) based on a large number of subjects in a prospective design. METHOD: We recommended enrolling of 200 consecutive patients who had been referred because of habitual snoring or witnessed apnea during sleep. If consent was obtained, patients underwent standard polysomnography (PSG) and ALP evaluation simultaneously at the hospital (ALPlab), and repeated ALP at home (ALPhome) within 1 month. The parameters of ALP were scored based on three different systems, Manual, Auto AASM, and Auto scoring systems. RESULT: Among the 200 patients who were recommended for enrollment, 149 completed the study. Sensitivity was good for all three scoring systems, while specificity was poor for the Auto scoring system. The area under curve was highest for the manual scoring system and lowest for the auto scoring system, and increased as the apnea-hypopnea index (AHI) cutoff value increased from 5 to 30. Regarding agreement with PSG, the manual scoring system was most superior, followed by the Auto AASM, and Auto scoring systems. The degree of agreement between PSG and ALP was affected by sleep efficiency and the arousal index. Moderate agreement between PSG and ALP based on central apnea index was observed. CONCLUSION: ALP is an accurate and reliable device for the diagnosis of OSA. Manual scoring is recommended: however, Auto AASM is also acceptable. ALP detects central sleep apnea to a certain degree. Both sleep efficiency and the arousal index affect the results of ALP.