Establishment and validation of a nomogram model for surgical site infections after posterior lumbar interbody fusion: a retrospective observational study.

Abstract

Surgical site infection is a serious complication of posterior lumbar interbody fusion surgery and is influenced by various factors. To construct a predictive nomogram of the risk of surgical site infection among patients after posterior luminal interbody fusion surgery. A total of 496 patients who underwent posterior lumbar interbody fusion surgery between January 2019 and December 2023 were included, and randomly assigned to a training or a validation queue following a 7:3 ratio. A nomogram prediction model was established based on the training queue, and evaluation of its accuracy and discriminative ability was done using calibration curves and receiver operating characteristic analysis. Decision curve analysis was used to estimate the clinical value of the nomograms. Seventeen cases (3.43%) of SSI were observed. The predictive factors included preoperative hypoalbuminemia (P = 0.048), drainage tube retention time (P = 0.002), number of fusion segments(P < 0.001), and postoperative white blood cell count (P = 0.003). The receiver operating characteristic analysis indicated that the model had good predictive performance (training cohort: 0.95; validation cohort: 0.903). The calibration curves showed good consistency between the predicted and actual values, and the decision curve indicated good clinical benefits. Preoperative hypoalbuminemia, drainage tube retention time, number of fusion stages, and postoperative white blood cell count were independent risk factors of surgical site infection in patients undergoing posterior lumbar interbody fusion surgery. The nomogram model had a good predictive performance and can provide an effective evaluation method to improve prediction accuracy.