Objective Based on the nutrition-related indexes of patients with hip fracture,a risk model of poor prognosis was established and its predictive efficiency was analyzed.Methods From February 2020 to February 2022,a total of 172 patients with hip fracture who were treated in the Department of Orthopaedics,Wenzhou TCM Hospital of Zhejiang Chinese Medical University were selected as the research objects.According to the follow-up results,they were divided into survival group and death group.The nutritional related indexes of the two groups were compared.Logistic regression analysis was used to build a risk model for predicting poor prognosis of patients with hip fracture,and ROC curve analysis was conducted to evaluate the predictive efficiency of the model.Results Up to the last follow-up time,3 patients were lost,51 patients died and were included in death group,118 patients were included in the survival group.Multivariate logistic regression showed that increasing age and reducing hemoglobin,albumin,prealbumin levels and decreasing serum creatinine were independent risk factors for poor prognosis in patients with hip fracture(P＜0.05).The ROC curve dem-onstrated that the joint prediction had higher predictive efficiency(AUC=0.989,sensitivity was 94.1％,specificity was 96.6％,accuracy was 95.86％)compared to single index.Conclusion The risk model based on the four nutritional in-dicators has good diagnostic efficiency in predicting the mortality of patients with hip fracture.
Hip fracturePrediction modelAlbuminPrealbuminHemoglobin
维普
万方数据
