Using the No-Apnea score to screen for obstructive sleep apnea in adults referred to a sleep laboratory: comparative study of the performance of the instrument by gender

ABSTRACT

Objective:

To evaluate the performance of the No-Apnea score, a simplified screening instrument for obstructive sleep apnea (OSA), by gender.

Methods:

This was a cross-sectional study including adults undergoing full polysomnography. The No-Apnea model comprises two items (neck circumference and age) with a total score of 0 to 9. The severity of OSA was categorized, on the basis of the apnea-hypopnea index, as any (≥ 5 events/h), moderate-to-severe (≥ 15 events/h), or severe (≥ 30 events/h). The performance of the No-Apnea instrument was assessed by determining the area under the (ROC) curve (AUC) and by constructing contingency tables.

Results:

We evaluated a total of 6,606 adults (53.8% men). For categorizing the level of OSA severity, the No-Apnea score had a sensitivity of 83.9-93.0% and a specificity of 57.3-35.2%. At all OSA severity levels, the No-Apnea score exhibited higher sensitivity and lower specificity in men than in women. The No-Apnea score proved to be an appropriate screening model for patients in general or when separated by gender or severity of OSA (AUC > 0.7 for all). The discriminatory power of the No-Apnea score to predict any, moderate-to-severe, and severe OSA was similar between genders (p = 0.109, p = 0.698, and p = 0.094, respectively).

Conclusions:

In a sample of adults referred to the sleep laboratory, there was no significant difference between men and women in terms of the discriminatory power of the No-Apnea instrument in for screening for OSA severity.

Keywords:
Sleep apnea, obstructive/diagnosis; Polysomnography; Sex; Surveys and questionnaires