Determination of prognostic factors for orthognathic surgery in children with cleft lip and/or palate

Objective To determine the prognostic factors for orthognathic surgery (OGS) in children with cleft lip and/or palate (CL/P) using artificial intelligence (AI) systems. Design Retrospective cohort study. Setting An orthodontic department at a university dental hospital. Participants This study included 126 patients with bilateral and unilateral CL/P for whom lateral cephalograms were obtained at three time points: 7 (T1), 10 (T2) and 15 (T3) years of age. Main outcome measures Cleft type, severity of lip separation at birth, number of missing teeth, sex, palatal repair methods and surgeons, cephalometric variables at T1 and T2, and the total duration of orthodontic treatment were examined as predictors. The need for OGS and skeletal and dental discrepancies at T3 was examined as outcomes. Results A total of six models were developed, with a mean area under the receiver operating characteristic curve of 0.93. Multiple prognostic factors for OGS were identified. In particular, the number of clefts in the lip and alveolus showed relatively high odds ratios, as did anterior crossbite at T3. Achieving palatal closure with the push-back method, rather than Furlow's method, was also found to be a predictive factor for anterior crossbite at T3, with high odds ratios. Conclusions The prognostic factors for OGS determined by the AI systems were the number of clefts in the lip and alveolus, the palatal repair method, male sex, several cephalometric variables for the sagittal and vertical dimensions, growth patterns and the number of missing teeth.