Accuracy of cephalometric pharyngeal analysis for diagnosis of obstructive sleep apnea syndrome (OSAS)

Abstract

The objective of this study was to verify the accuracy of lateral radiographic cephalometry for the diagnosis of obstructive sleep apnea syndrome (OSAS). The sample was composed of 60 patients, all with clinical suspicion of OSAS. A positive polysomnographic diagnosis requires an apnea-hypopnea index [(A+H)I] per hour of sleep equal to or above five events. A positive cephalometric diagnosis requires superior posterior airway space (PASs) < 26 mm and/or middle posterior airway space (PASm) < 9 mm and/or inferior posterior airway space (PASi) < 11 mm. The statistical study verified the accuracy of the cephalometric analysis in relation to the polysomnographic examination (gold standard). The comparison was established with the McNemar and Cohen’s Kappa Tests. The sensitivity of the cephalometric analysis was 91.5%. The specificity was 38.5%. The positive predictive value was 84.3%. The negative predictive value observed was 55.6%. False negative results occurred in four situations and false positive in eight situations. The McNemar test showed no statistically significant difference between the results obtained by cephalometry and by polysomnography (p=0.388); and the measure of agreement calculated by Cohen’s kappa coefficient was 33.7% (p=0.007), significant for values of p < 0.05. The high sensitivity and the elevated positive predictive value of radiographic cephalometry make it a trustworthy tool, capable of diagnosing OSAS patient with a high degree of certainty, and indicating the obstructive site related to the manifestation of the condition. However, the low specificity of radiographic cephalometry and its negative predictive value do not allow its use as an exclusive examination.