|| Checking for direct PDF access through Ovid
The authors investigated the accuracy of virtual surgical planning in predicting airway volume changes after mandibular distraction in patients with Pierre Robin sequence and associated tongue-based airway obstruction.The authors completed a single-institution retrospective review of patients for whom virtual surgical planning was used during mandibular distraction osteogenesis for treatment of tongue-based airway obstruction. Preoperative airway volume, virtual surgical planning–predicted airway volume, and postoperative airway volume were calculated from three-dimensional computed tomographic scans using industry software. A blinded institutional radiologist also calculated pre- and post-operative airway volumes. Pre- and post-operative polysomnography was used to titrate the endpoint of mandibular lengthening.Eleven patients were included in the study. Mean apnea-hypopnea index (5.42 ± 4.53 versus 44.96 ± 20.57; p < 0.001) and mean nadir oxygen saturation (70.3 ± 9.72 percent versus 82.9 ± 9.62 percent; p = 0.003) improved with mandibular distraction. There was moderate correlation between predicted and actual mandibular distraction lengths (R2 = 0.65; p = 0.003). There was a strong correlation between predicted and industry-calculated actual post–distraction airway volume (R2 = 0.99; p < 0.001). There was no significant correlation between actual mandibular distraction length and industry-calculated actual post–distraction airway volume for the entire cohort (R2 = 0.05; p = 0.49), but correlation approached significance by institutional calculations. No significant correlation existed between industry and institutional-calculated percentage change in post–distraction airway volume (R2 = 0.06; p = 0.57).Predictive airway volume calculation may be an effective adjunct to determine anatomic endpoint of mandibular distraction but small sample size, operator and software variability, and patient airway morphology may confound firm conclusions. Further studies are warranted.