Development and Validation of Nomogram to Predict Frailty for Older Patients Undergoing Abdominal Surgery

开发和验证用于预测老年腹部手术患者虚弱程度的列线图

阅读:1

Abstract

BACKGROUND: Frailty is a critical geriatric syndrome associated with adverse surgical outcomes, yet preoperative risk prediction models for older adults undergoing abdominal surgery remain underdeveloped. This study aimed to identify frailty risk factors and establish a predictive nomogram in this population. METHODS: We enrolled 790 older patients undergoing abdominal surgery at Hunan Provincial People's Hospital from February 2022 to September 2022. Frailty was assessed using the Tilburg Frailty Index. Univariate analysis, LASSO regression and multivariate analysis were used in turn to identify independent risk factors for frailty. The nomogram was developed based on the independent risk factors. The sample was randomly divided into a test group (75%) and a validation group (25%). The area under the curve (AUC) of the receiver operating characteristic (ROC) was calculated to assess the predictive performance of the nomogram. RESULTS: The prevalence of frailty among older patients undergoing abdominal surgery was 74.18%. Eight independent risk factors were identified: advanced age (OR=1.32), lower BMI (OR=1.28), limited education (OR=1.45), laparoscopy (OR=1.67), tumor comorbidity (OR=2.01), diabetes (OR=1.89), antihyperlipidemic drug use (OR=1.53), and elevated interleukin-6 (OR=1.76). The nomogram demonstrated acceptable discrimination, with AUCs of 0.748 (the test group) and 0.707 (the validation group). CONCLUSION: Our findings demonstrate a nomogram to predict the probability of frailty for older patients undergoing abdominal with acceptable predictive performance. The nomogram is helpful in guiding further targeted and effective intervention and prevention efforts to decrease frailty and improve health outcomes.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。