Acoustic pharyngometry no additional benefit in OSA diagnosis

Assessment of upper airway cross-sectional area using acoustic pharyngometry is no better than the use of clinical variables to diagnose obstructive sleep apnea (OSA), according to a study in the Annals of the American Thoracic Society.

Tetyana Kendzerska, PhD, of the Institute for Clinical Evaluative Sciences, Toronto, and her colleagues found that the median upper airway cross-sectional area at functional residual capacity when sitting was significantly reduced in individuals with OSA, compared with those without the condition (3.3 cm² vs. 3.7 cm²).

^{©designer491/Thinkstock}

The researchers found that, at a cut-off value of 3.75 cm², which struck the best balance of sensitivity and specificity, upper airway cross-sectional area had a sensitivity of 73% and specificity of 46%. Varying the apnea-hypopnea index to define OSA or varying the analysis of upper airway cross-sectional area did not improve its predictive or discriminative ability, nor was there any benefit to measuring upper airway cross-sectional area when an individual was supine, compared with sitting.

Dr. Kendzerska and her colleagues had hypothesized that acoustic pharyngometry could play a role in screening for OSA, based on previous studies suggesting significant differences in upper airway cross-sectional area measures in individuals with and without the condition. Their cross-sectional study included 576 subjects with suspected OSA who underwent acoustic pharyngometry within 35 days of standard diagnostic polysomnography (Ann Am Thorac Soc. 2016 Aug 16. doi: 10.1513/AnnalsATS.201601-056OC).

“Although the mean [upper airway cross-sectional area] at [functional residual capacity] when sitting was a significant predictor of OSA controlling for important confounders, it had only fair discriminant validity for identifying those with OSA in a clinic population and had no significantly greater discriminant value than the use of clinical variables,” the researchers reported. “Therefore, it is probably of no clinical utility in this setting.”

The investigators said that they had no conflicts of interest.