Comparison of pharyngeal airway changes on plain radiography and cone-beam computed tomography after orthognathic surgery.

[PURPOSE] The purpose of the present prospective study was to develop a 3-dimensional analysis of the airway using cone-beam computed tomography (CBCT) and to determine whether changes in the airway before and after orthognathic surgery correlate on 2-dimensional lateral cephalogram and 3-dimensional CBCT images.

[MATERIALS AND METHODS] Patients requiring orthognathic surgery during 2004 to 2005 were recruited for the present study. Lateral cephalograms and CBCT scans were obtained at 3 points: preoperatively, within 1 month postoperatively, and after 6 months postoperatively. The nasopharynx, oropharynx, and hypopharynx were segmented on both the radiograph and the CBCT scan for each patient in a repeatable manner at each point. For the lateral cephalogram, linear measurements in the middle of each of the 3 segments were obtained. For the CBCT, volumetric measurements of each of the 3 segments were obtained. The intrarater variability was assessed, and Pearson's correlation was used to compare the 2 imaging modalities.

[RESULTS] A total of 20 patients scheduled for orthognathic surgery were recruited for the present study. Of the 20 patients, 13 were female and 7 were male. The mean age at surgery was 23.85 years (range 14 to 43). Of the 20 patients, 6 underwent maxillary advancement only, 8 underwent mandibular advancement with or without genioplasty, and 6 underwent 2-jaw surgery or mandibular setback. We examined the entire cohort without separation into procedure or examination point and found a weak, but statistically significant, correlation between the linear and volume measurements in the nasopharyngeal and oropharyngeal regions but not in the hypopharyngeal region (r = 0.43, P < .002; r = 0.49, P < .0002; r = 0.16, P = .26, respectively). The maxillary advancement group (n = 6) demonstrated a correlation between the linear and volume measurements in the nasopharyngeal region (r = 0.53, P = .03). The mandibular advancement with or without genioplasty group (n = 8) showed a correlation in the nasopharyngeal and oropharyngeal regions (r = 0.55, P < .02, and r = 0.46, P = .05, respectively). For the combination/setback procedures (n = 6), a correlation was found in the oropharyngeal region (r = 0.64, P < .01). All other comparisons between the linear and volume measurements did not correlate. Additionally, no correlations were found between the linear and volumetric change in airway size between 6 months postoperatively and preoperatively, except for the oropharyngeal region (r = 0.67, P < .01).

[CONCLUSION] We present a method of measuring the airway that could be used for both 2-dimensional and 3-dimensional images. It includes segmentation of the pharyngeal airway into its nasopharyngeal, oropharyngeal, and hypopharyngeal components. Correlations were found between the linear and volumetric measurements of the segmented airway in patients who had undergone orthognathic surgery; however, the correlations were generally weak.