Comparison of Nasal Pressure Transducer and Thermistor for Detection of Respiratory Events During Polysomnography in Children

Abstract

The results of small studies have suggested that a nasal-cannula pressure transducer has a higher sensitivity than a thermistor in detecting hypopneas and diagnosing sleep-disordered breathing in both adults and children. We compared a thermistor alone, and in conjunction with a pressure transducer, for detection of sleep-disordered breathing in children during in-home polysomnography. Retrospective analysis of a subsample of a prospective cohort study. Students attending elementary school in the Tucson Unified School District. A subsample of the Tucson Children's Assessment of Sleep Apnea study population. Polysomnographic recordings of 40 children (24 girls and 16 boys, mean age 9.2 +/- 1.7 years; range 6-11 years) were analyzed to compare the detection of sleep-disordered breathing events by 2 different methods of measuring airflow: thermistor alone and thermistor with nasal-cannula pressure transducer (transducer) used simultaneously. The transducer detected all the respiratory events detected by the thermistor, but the thermistor detected only 84% of the transducer-defined events. Consequently, the transducer-derived mean respiratory disturbance index was higher than that detected by the thermistor (7.0 +/- 3.8 vs 5.9 +/- 3.4, P < .001). The bias error between transducer respiratory disturbance index and thermistor respiratory disturbance index on a Bland-Altman plot was 1.08 (95% confidence interval, 0.8 - 1.4). There was good agreement between the thermistor and the transducer for making the diagnosis of sleep apnea using a cutoff of a respiratory disturbance index greater than 5 (kappa = 0.69). The quality of the tracings with the transducer was comparable to that of the thermistor, but the transducer dislodged more frequently. The use of a nasal transducer in conjunction with a thermistor was more sensitive than the thermistor alone in detecting sleep-disordered breathing in children during unattended polysomnography.