| 李志磊,乔存根,杨士雪,等.老年髋部骨折术后发生急性肾损伤的危险因素及列线图预测模型[J].安徽医药,2025,29(9):1791-1795. |
| 老年髋部骨折术后发生急性肾损伤的危险因素及列线图预测模型 |
| Risk factors for acute kidney injury after hip fracture surgery in the elderly and the establish. ment of a nomogram prediction model |
| |
| DOI：10.3969/j.issn.1009-6469.2025.09.019 |
| 中文关键词: 髋骨折 术后并发症 急性肾损伤 危险因素 列线图 95%，高风， |
| 英文关键词: Hip fractures Postoperative complications Acute kidney injury Risk factors Nomogram |
| 基金项目:邢台市重点研发计划自筹项目( 2021ZC069) |
|
| 摘要点击次数: 239 |
| 全文下载次数: 368 |
| 中文摘要: |
| 目的探讨老年髋部骨折( HF)病人术后发生;性肾损伤( AKI)的危险因素，并构建列线图预测模型。方法 2016年 10月至 2021年 10月华北医疗健康集团邢台总医院收治的 813例老年 HF病人按照 3∶7分成验证组( n=244)及建模组(n=569)。根据建模组病人术后 AKI的发生与否分为 AKI组( n=63)和无 AKI组( n=506)。收集 HF病人资料并采用多因素 logistic回归模型分析术后发生 AKI的影响因素； R4.3.1构建预测老年 HF病人术后 AKI发生风险的列线图模型。 ROC曲线及 H-L拟合度检验分别评估列线图模型预测老年 HF病人术后发生 AKI的区分度及一致性；临床决策曲线( DCA)评估该模型预测老年 HF病人术后 AKI发生风险的临床应用价值。结果验证组及建模组术前血肌酐( SCr)、手术时间、饮酒史、高血压、利尿剂使用等资料比较，均差异无统计学意义( P>0.05)。与无 AKI组比较， AKI组老年 HF病人年龄［(72.79±7.01)岁比( 68.65±5.95)岁］、术前 SCr水平［(84.31±10.80)μmol/L比( 77.04±9.32)μmol/L］、合并低血压( 22.22%比 8.50%)及糖尿病( 26.98%比 10.08%)的占比较高、手术时间较长［( 90.18±13.09)min比( 80.56±11.69)min］，术前肾小球滤过率( eGFR)［( 71.09±9.23)mL-1·min-1·(1.73 m2)-1比(77.92±9.74)mL-1·min-1·(1.73 m2)-1］显著下降( t/χ2=5.10、5.73、11.71、15.22、6.08、5.28，P<0.05)。多因素 logistic回归分析，年龄(OR=1.11)、术前 SCr(OR=1.08)、手术时间( OR=1.06)、低血压( OR=2.90)、糖尿病( OR=4.94)为老年 HF病人术后发生 AKI的独立危险因素( P<0.05)术前 eGFR(OR=0.94)为其保护因素( P<0.05)。建模组和验证组 ROC曲线下面积分别为 0.85，95%CI:(0.80，0.90)和 0.88， CI:(0.80，0.96)建模组和验证组 H-L检验结果分别为 χ2=5.64、P=0.687，χ2=7.64、P=0.469，校准曲线与理想曲线拟合度高。 DCA曲线所示，险阈值在 0.02~0.80时，临床应用价值高。结论基于年龄、术前 SCr、术前 eGFR、手术时间、低血压、糖尿病六个因素构建的预测老年 HF病人术后 AKI发生风险的列线图模型临床应用价值较高。 |
| 英文摘要: |
| Objective To explore the risk factors for postoperative acute kidney injury (AKI) in elderly patients with hip fracture(HF), and to construct a nomogram prediction model.Methods A total of 813 elderly HF patients admitted to Xingtai General Hospi-tal, North China Medical and Health Industry Group from October 2016 to October 2021 were assigned into a validation group (n=244) and a modeling group (n=569) at a ratio of 3∶7. According to the occurrence of postoperative AKI in the modeling group, patients in the modeling group were separated into AKI group (n=63) and non AKI group (n=506). HF patient data were collected and a multivari-ate logistic regression model was applied to analyze the influencing factors for postoperative AKI; R4.3.1 software was applied to con-struct a nomogram model for predicting the risk of postoperative AKI in elderly HF patients. ROC curve and H-L fit test were applied toevaluate the discrimination and consistency of the nomogram model in predicting postoperative AKI in elderly HF patients; DecisionCurve Analysis (DCA) was applied to collect and evaluate the clinical application value of this model in predicting the risk of postopera-tive AKI in elderly HF patients.Results There were no significant differences in preoperative serum creatinine (SCr), surgical time, al-cohol consumption history, hypertension, and diuretic use between the validation group and the modeling group (P>0.05). Compared with the non AKI group, the age [(72.79±7.01) years vs. (68.65±5.95) years], preoperative SCr level [(84.31±10.80) μmol/L vs. (77.04± 9.32) μmol/L], the proportions of concurrent hypotension (22.22% vs. 8.50%) and diabetes (26.98% vs. 10.08%) in the AKI group were higher, the operation time was longer [(90.18±13.09) min vs. (80.56±11.69) min], and the preoperative estimated glomerular filtration rate (eGFR) [(71.09±9.23) mL-1·min-1·(1.73 m2)-1 vs. (77.92±9.74) mL-1·min-1·(1.73 m2)-1] was significantly decreased (t/χ2=5.10, 5.73, 11.71, 15.22, 6.08, 5.28, respectively; P<0.05). Multivariate logistic regression analysis results showed that age (OR=1.11), preop-erative SCr (OR=1.08), operation time (OR=1.06), hypotension (OR=2.90), and diabetes (OR=4.94) were independent risk factors for postoperative AKI in elderly HF patients (P<0.05), and preoperative eGFR (OR=0.94) was the protective factor (P<0.05). The areas un-der the ROC curves of the modeling and validation groups were 0.85, 95%CI:(0.80,0.90) and 0.88, 95%CI:(0.80,0.96), respectively, and the H-L test results of the modeling and validation groups were χ2=5.64, P=0.687, χ2=7.64, and P=0.469, respectively, indicating ahigh fit of the calibration curve with the ideal curve. DCA curve showed that the clinical application value was high when the high-risk threshold was between 0.02 and 0.80.Conclusion The nomogram model based on age, preoperative SCr, preoperative eGFR, opera-tion time, hypotension, and diabetes to predict the risk of postoperative AKI in elderly HF patients is of high clinical value. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
