Development and validation of a nomogram predicting the risk of necrosis in fingertip replantation: a single-center retrospective study

Objectibve: Despite advanced microsurgical techniques, necrosis can still occur following fingertip replantation. This study aimed to systematically identify independent risk factors for tissue necrosis following fingertip replantation, specifically, the distal end of the finger including the distal interphalangeal joint, together with the development of a clinically applicable risk prediction model. Methods : The study retrospectively examined 1497 fingers from 1343 patients who underwent fingertip replantation procedures at Suzhou Ruihua Orthopaedic Hospital between August 2018 and August 2023. Patients were randomized into a training group and a validation group. In the training group, Log Lambda and Least Absolute Shrinkage and Selection Operator (LASSO) regression were employed to identify predictors associated with necrosis. A nomogram prediction model was then developed using multifactorial logistic regression. The performance characteristics of the resulting nomogram were evaluated, including discriminant analysis, calibration analysis, and clinical decision curve analysis (DCA). Models were internally validated in the validation group. Results: The study included 1,343 patients who underwent fingertip replantation, with 115 (13.9%) cases of necrosis in the training set and 48 (13.5%) in the validation set. The nomogram included 6 predictors from univariate and multivariate analyses, including platelet count,neutrophilic granulocyte percentage, D-dimer test, hypertension, vascular crises, and surgeon’s experience. The model had good discriminatory properties, with a well-calibrated consistency index of 0.9494 (95% CI: 0.9325-0.9662) for model construction (Hosmer-Lemeshow test, P=0.72). The nomogram of the validation set showed good discrimination with a concordance index of 0.9667 (95% CI: 0.9462-0.9872) and was well calibrated (Hosmer-Lemeshow test, P=0.5092). Clinical DCA demonstrated that the Nomogram model predicted the risk of necrosis in fingertip replantation with probability thresholds ranging from 1% to 91% and internal validation ranging from 1% to 99%, suggesting a net benefit in modeling and internal validation for predicting risk of replantation necrosis. Conclusion: Our Nomogram model can reliably identify patients who are at high risk of developing necrosis following fingertip replantation The model provides a practical early evaluation tool for healthcare professionals to assess the risk of necrosis.