Interaction analysis of risk factors for long-term skeletal relapse following mandibular advancement with bilateral sagittal split osteotomy

Abstract The aim of this study was to identify interaction effects among risk factors for long-term skeletal relapse. The study sample consisted of 96 patients who underwent mandibular advancement with bilateral sagittal split osteotomy. Ten predictor variables were analyzed for an interaction effect: sex, age, preoperative temporomandibular joint symptoms, mandibular plane angle (MPA), single or double jaw surgery, clockwise or counterclockwise mandibular rotation, magnitude of mandibular advancement, concomitant genioplasty, type of fixation, and follow-up duration. Modeling interactions between pairs of covariates were applied to detect a significant interaction among these risk factors on horizontal and vertical long-term skeletal relapse, respectively. Stratification analyses and two-way full factorial interaction analyses were performed to demonstrate how the interaction influenced the associations between covariates and relapse. The interactions between sex and mandibular rotation (P = 0.006) and between MPA and mandibular rotation (P = 0.002) were statistically significant for horizontal long-term skeletal relapse. No significant interaction was identified for vertical relapse. This study showed that female patients and those with an MPA ≥30° undergoing counterclockwise mandibular rotation are predisposed to greater horizontal long-term skeletal relapse. Therefore, the judicious use of counterclockwise rotation is recommended in order to minimize the relapse, especially in female patients and those with a high MPA.