Practical and rapidly-implemented parameters for assessing APAP titration failure

This study aimed to show the predictive value of simple polysomnographic parameters including latency of deep sleep (nREM3), latency of rapid eye movement sleep (REM), and minimum oxygen saturation (SpO₂) for predicting failure of autoadjusting positive airway pressure (APAP) titration.

Out of 1470 patients with moderate to severe obstructive sleep apnea syndrome (OSAS) who underwent APAP titration between July 1, 2016, and December 31, 2017, 22 patients with titration failure were enrolled in the study. The demographic and polysomnographic characteristics of this group were compared with 44 patients with an adequate APAP titration who were matched with the titration failure group by age, sex, and OSAS severity. The periods between the start of sleep and the start of REM and nREM3 stages were noted as REM latency and nREM3 latency, respectively.

The between group differences in the parameters including nREM3 latency, REM latency, and minimum SpO₂ during the titration test were statistically significant (p = 0.004, p = 0.008, p <0.001 respectively). Possible threshold values to predict failure of APAP titration were found as 40 min and 135 min for nREM3 and REM latencies, respectively. The best threshold for minimum SpO₂ was 86% with an 86.4% of sensitivity. Despite the high negative predictive values (81% for nREM3 latency, 75% for REM latency), the sensitivities (63.6% for nREM3 latency, 54.5% for REM latency) of the prior parameters in determining titration failure were relatively low.

The minimum SpO₂ < 86% can be used as a readily available indicator of APAP titration failure. nREM3 and REM latencies may be regarded as supplementary indicators in detecting the patients who may need an advanced PAP device.