Correlation between Nasal Anatomy and Objective Obstructive Sleep Apnea Severity

Abstract

Objectives

To determine if a correlation exists between nasal anatomical obstruction and obstructive sleep apnea severity as measured by overnight polysomnogram (PSG).

Study Design

Cross-sectional study.

Setting

Tertiary medical center.

Subjects and Methods

Subjects were recruited immediately prior to an overnight, in-lab PSG. All subjects who agreed to participate underwent a standardized nasal examination performed by the senior author and then completed the Nasal Obstruction Symptom Evaluation (NOSE) questionnaire, the Snore Outcomes Survey (SOS), and the Epworth Sleepiness Scale (ESS) prior to their sleep study. In addition, tonsil size, Mallampati score, Friedman tongue position, neck circumference, uvula length, and occlusion were assessed and documented. Nasal anatomy assessments were then compared with PSG, NOSE, SOS, and ESS results. Bonferroni correction was used to account for multiple comparisons.

Results

One hundred subjects were included in the study. Fifty-nine subjects (59%) were found to have obstructive sleep apnea syndrome (OSAS) (mean apnea-hypopnea index, 13.1; range, 0-64). Severity of OSAS was associated with age (Spearman’s ρ = 0.386, P = .0001). No single nasal anatomy measurement or combined nasal anatomy index was found to correlate with objective sleep-disordered breathing severity measured by PSG. The sample size should have provided 90% power to detect a significant correlation if one existed. After accounting for multiple comparisons, turbinate hypertrophy was found to correlate with the NOSE score (0.3577, P = .0305 corrected), and external and internal nasal valve collapse correlated with each other (0.4986, P < .0001 corrected).

Conclusions

Objectively assessed abnormal nasal anatomy was not found to be significantly correlated with PSG-measured OSAS severity. Specific objective measurements of obstructive nasal anatomy were correlated to subjective measures of nasal obstruction.