Objective: To assess the clinical utility of the frequencies of transient increases of pulse rate, non-invasively measured with a pulseoximeter, as an indirect indication of the degree of cortical arousal, measured conventionally on an electroencephalogram (EEG), in obstructive sleep apnea-hypopnea syndrome (OSAHS) patients.
Patients and methods: Thirty-three consecutive patients referred with suspected OSAHS were studied. Polysomnography (PSG) with determination of esophageal pressure (Pes) and pulseoximetry was monitored to identify breathing-related EEG arousal (B-Ar) associated with apnea, hypopnea or respiratory effort and the frequencies of pulse rate increases. We also assessed the association of B-ArI (defined as the number of B-Ar per hour) with the pulse rate rise index (PRRI)-X(X=4-10) (defined as the number of pulse rate increases per hour). In addition, the sensitivity and specificity of PRRI for the assessment of a B-ArI cutoff point of 30 were calculated.
Results: The sensitivity and specificity of pulseoximetry for different thresholds of PRRI-X(X=4-10) demonstrated that the greatest diagnostic accuracy for detecting frequent arousal (B-ArI > or =30) occurs at a cutoff point of 40 PRRI-6 with a sensitivity of 0.88 and specificity of 0.86. This point shows a significant area under the curve of 0.84. In addition, a statistically significant correlation between PRRI-6 and B-ArI (r=0.68, P<0.0001) was observed.
Conclusions: The transient increases in pulse rate measured by pulseoximetry during sleep may be a useful clinical marker for predicting the degree of arousal in OSAHS patients, and may, in addition, prevent cases with frequent respiratory effort related arousals from being overlooked. However, further studies are required to improve the confidence level of the PRRI and to investigate the causes of overestimation of EEG arousals.