Development and validation of a clinical nomogram for predicting postoperative hypoalbuminemia in anemic hip fracture patients

Background Major orthopedic interventions induce significant blood loss and systemic inflammation, frequently precipitating acute postoperative hypoalbuminemia. However, individualized risk assessment tools for predicting this complication remain limited. This study aimed to develop and internally validate a clinical nomogram using machine learning to predict postoperative hypoalbuminemia risk in anemic hip fracture patients. Methods A retrospective analysis of 359 anemic patients undergoing hip fracture surgery was conducted. The Boruta algorithm was utilized for feature selection, followed by multivariable logistic regression to pinpoint independent predictors. A predictive nomogram was constructed and its performance evaluated via the area under the receiver operating characteristic curve (AUC), calibration plots with 1,000 bootstrap resamples, and decision curve analysis (DCA). Results Postoperative hypoalbuminemia occurred in 200 patients (55.7%). Five independent predictors were identified: advancing age (odds ratio [OR] = 1.192 per 5 years, p = 0.010), elevated intraoperative blood loss (OR = 1.313 per 100 mL, p = 0.001), decreased preoperative hemoglobin (Hb) (OR = 0.741 per 10 g/L, p = 0.001), increased preoperative monocyte count (OR = 1.200 per 0.1 × 10 9 /L, p = 0.001), and lower preoperative serum albumin (OR = 0.509 per 5 g/L, p < 0.001). The nomogram demonstrated moderate discrimination with an initial AUC of 0.764. After internal validation using 1,000 bootstrap resamples, the optimism-corrected AUC was 0.751 (95% CI: 0.702–0.801), with a mean optimism of 0.013. Calibration showed good agreement (mean absolute error = 0.012). DCA indicated continuous positive net clinical benefits across threshold probabilities of 0.10 to 0.86. Conclusion This internally validated nomogram predicts postoperative hypoalbuminemia utilizing five independent factors. This tool may assist in early risk stratification and rational allocation of human serum albumin, supporting risk stratification, albumin resource planning and efficient medical resource allocation.