Fibular free flap necrosis after mandibular reconstruction surgery with osteoradionecrosis: Establishment and verification of an early warning model.

ABSTRACT

OBJECTIVE: Fibular free flap necrosis (FFFN) is the most common complication in patients with osteoradionecrosis (ORN) after mandibular reconstruction surgery. However, there are no effective forecasting tools at present. This research is aimed to establish and verify a nomogram model to predict the risk of FFFN after mandibular reconstruction surgery in ORN patients. METHODS: A total of 193 ORN patients with mandibular reconstruction using fibular free flap (150 cases in the model group and 43 cases in the validation group) were enrolled in this study. In the model group, the variables were optimized by lasso regression. Then the prediction model was established by binary logistic regression analysis, and the nomogram was drawn. The bootstrap self-sampling method was used for internal verification. Moreover, 43 cases in the validation group were used for external validation. RESULTS: The results of lasso regression and binary logistic regression analysis showed that the radiotherapy interval (≤2 years), trismus, diabetes, without deep venous anastomoses, and American society of anesthesiologists (ASA) III were the independent risk factors for FFFN after mandibular reconstruction surgery in ORNJ patients (P<0.05). Based on the above-mentioned risk factors, the nomogram model was established. The AUC values of the model group and the validation group were 0.936 and 0.964, respectively. The curve analysis showed that when the probability thresholds of the model group and the validation group were 5.699%∼98.229% and 0.413%∼99.721%, respectively. So the patient's clinical net profit rate was the highest. CONCLUSION: A nomogram combining the factors of radiotherapy interval (≤2 years), trismus, diabetes, without deep venous anastomoses, and ASA III provided a comparatively effective way to predict the risk of FFFN after mandibular reconstruction surgery in ORN patients, which has distinct applied clinical value.