ABSTRACT
Obstructive sleep apnea (OSA) is highly associated with cardiovascular diseases, but most patients remain undiagnosed. Cyclic variation of heart rate (CVHR) occurs during the night, and R-R interval (RRI) analysis using a Holter electrocardiogram has been reported to be useful in screening for OSA. We investigated the usefulness of RRI analysis to identify OSA using the wearable heart rate sensor WHS-1 and newly developed algorithm. WHS-1 and polysomnography simultaneously applied to 30 cases of OSA. By using the RRI averages calculated for each time series, tachycardia with CVHR was identified. The ratio of integrated RRIs determined by integrated RRIs during CVHR and over all sleep time were calculated by our newly developed method. The patient was diagnosed as OSA according to the predetermined criteria. It correlated with the apnea hypopnea index and 3% oxygen desaturation index. In the multivariate analysis, it was extracted as a factor defining the apnea hypopnea index (r = 0.663, p = 0.003) and 3% oxygen saturation index (r = 0.637, p = 0.008). Twenty-five patients could be identified as OSA. We developed the RRI analysis using the wearable heart rate sensor WHS-1 and a new algorithm, which may become an expeditious and cost-effective screening tool for identifying OSA.
ABSTRACT
Differences in successive R-R intervals (RRIs) were normalized by RRIs before and after the indexing beats (normalized DRs) in individuals with normal sinus rhythm (NSR) and 98.89% of normalized DRs were found to distribute within mean±0.100 (âmean±3SD), whereas 73.47% were out of this range in atrial fibrillation (AF). When 7 out 20 normalized DRs fell outside of 0.000±0.100, NSR (n=129) and AF (n=108) could be discriminated with high sensitivity, specificity, and predictive values (>99.0% for all). This method will be used in detecting AF candidates from a small number of heart beats or arterial pulses.
