A comparison of a computer-based orthognathic surgery prediction system to postsurgical results.

This retrospective study analyzes the accuracy of a prediction algorithm from Quick Ceph Image. Pre- and postsurgical lateral cephalograms of 35 adult patients were computer imaged and specific landmarks digitized. Digitization error was assessed from duplicate digitizations. Sixteen measurements of the predicted and actual postsurgical hard and soft tissue landmarks were compared using Student's t test, one- and two-way analysis of variance, and a modified Bradley-Blackwood test. Results showed a good correlation between repeated digitization of all variables except soft tissue point B and E-plane. Comparison of the predicted and actual changes found that all differences were less than 1.8 mm or 3.1 degrees. Student's t test showed 10 of the 16 measurements did not differ significantly, and interclass correlation demonstrated moderate to good correlations for 13 of 16 variables. Overall the results of this study suggest that the magnitude of the differences were within clinically acceptable limits.