Diagnosis of glossopharyngeal obstruction using nasopharyngeal tube versus CT scan in obstructive sleep apnea-hypopnea syndrome.

The polysomnography after nasopharyngeal tube insertion (NPT-PSG) was used to assess glossopharyngeal airway obstruction in patients with moderate-to-severe obstructive sleep apnea-hypopnea syndrome (OSAHS), and was compared with that obtained using spiral computed tomography (CT). A total of 125 patients diagnosed with moderate-to-severe OSAHS using PSG were subjected to PSG after NPT insertion, and spiral CT was used for continuous scan of the upper respiratory tract. The NPT-PSG and CT scan results were subjected to correlation analysis. In addition, the two sets of data were used separately to determine whether there was glossopharyngeal airway stenosis or obstruction, and the results were compared. Neither apnea hypopnea index (AHI) nor lowest oxygen saturation (LSaO2) obtained from the first PSG was significantly correlated with the cross-sectional area or the inner diameter of the glossopharyngeal airway. NPT-PSG AHI was significantly correlated with the cross-sectional area and anteroposterior diameter of the glossopharyngeal airway, with correlation coefficients of 0.264 and 0.185, and P values of 0.003 and 0.039, respectively. NPT-PSG AHI was not significantly correlated with the left-right diameter of the airway, and NPT-PSG LSaO2 was not significantly correlated with the cross-sectional area or the inner diameter of the glossopharyngeal airway. With NPT-PSG, 52 patients showed obvious glossopharyngeal airway stenosis while 73 patients did not, and with CT scan 41 patients showed obvious glossopharyngeal airway stenosis while 84 patients did not. The two methods reached the same diagnosis in 86 cases, with a consistency rate of 68.8 %. Spiral CT and NPT-PSG show certain degrees of consistency by assessing the presence of glossopharyngeal airway stenosis or obstruction, and there are also notable differences. Clinical assessment on glossopharyngeal airway obstruction should be based on a combination of multiple methods.