Accuracy of a digital workflow for bimaxillary orthognathic surgery: comparison of planned and actual outcomes

Aim: The aim of the present prospective proof-of-concept study was to evaluate the accuracy of 3D orthognathic surgical planning and CAD/CAM splints by comparing planned with actual postoperative outcomes.
Materials and methods: Ten patients scheduled for bimaxillary orthognathic surgery to correct a skeletal Class III dentofacial deformity were recruited. All subjects had CBCT scans taken not more than 2 months preoperatively and within the 1-week postoperative period. The distance between six dental landmarks (midpoint of the maxillary and mandibular incisors, mesiobuccal cusps of the maxillary and mandibular first molars) and three intersecting symmetry planes (Frankfort horizontal plane [FHP], midsagittal plane [MSP], and coronal plane [CP]) were measured, and the differences between the virtually simulated and actual postoperative models were computed. The threshold for accuracy was set at 2 mm.
Results: Differences between the planned and actual outcomes were analyzed via chi-square tests and two-tailed paired student t tests. The overall mean linear difference for all six landmarks was 0.98 mm. The overall mean linear differences for both maxillary and mandibular landmarks relative to the FHP, MSP, and CP were 1.3, 0.7, and 0.9 mm, respectively. Four cases showed all linear differences of the six landmarks to be < 2.0 mm, while the other six cases had at least one linear difference of > 2.0 mm, the majority of which were in the superior-inferior direction. There were statistically significantly greater inaccuracies in the FHP compared with the MSP and CP (P < 0.05).
Conclusion: Most of the linear differences between the simulated and actual outcomes were clinically acceptable. However, greater linear differences were seen in the superior-inferior direction, indicating a greater surgical error in achieving the desired vertical position of the maxillomandibular complex. (Int J Comput Dent 2022;25(4):397–0; doi: 10.3290/j.ijcd.b2599749)