实用器官移植电子杂志 ›› 2022, Vol. 10 ›› Issue (2): 122-128.DOI: 10.3969/j.issn.2095-5332.2022.02.006

• 论著 • 上一篇    下一篇

肾功能正常肝移植受者术后急性肾损伤风险预测模型的构建

翟慧敏 1 ,姜英俊 2 ,孔心涓 1 ,蔡金贞 3 ,饶伟   

  1. 1. 青岛大学附属医院消化内科,山东 青岛 266000 ;

    2. 青岛大学附属医院普外科,山东 青岛 266000 ;

    3. 青岛大学附属医院器官移植中心肝脏病中心,青岛大学附属医院器官捐献与移植研究院,山东 青岛 266000

  • 出版日期:2022-03-20 发布日期:2022-05-16

The establishment of predictive model for postoperative acute kidney injury in liver transplant recipients with normal renal function 

Zhai Huimin1,Jiang Yingjun2,Kong Xinjuan1,Cai Jinzhen3,Rao Wei3.    

  1. 1.Department of Gastroenterology,the Affiliated Hospital of Qingdao University,Qingdao 266000,Shandong,China ;

    2. Department of General Surgery,the Affiliated Hospital of Qingdao University,Qingdao 266000,Shandong,China ;

    3.Center for Organ Transplantation,the Affiliated Hospital of Qingdao University,Institute of Transplantation Science,Qingdao University,Qingdao 266000,Shandong,China.

  • Online:2022-03-20 Published:2022-05-16

摘要:

目的 肝移植(liver transplantation,LT)术后早期急性肾损伤(acute kidney injury,AKI)的发生是影响患者长期预后的常见问题之一,本研究试图创建一个列线图以精确预测术前肾功能正常的 LT 受者术后早期 AKI 的发生。方法 回顾性分析本院 2017 年 7 月 1 日至 2020 年 12 月 31 日期间行 LT 术后患者 369 例的临床病历资料,最终纳入 349 例患者,根据是否发生 AKI 分为 AKI 组和非 AKI 组,通过多元Logistics 回归分析确定 AKI 的独立危险因素,并将其用于创建列线图风险预测模型。通过使用内部验证一致性指数(concordance index,C-index)来评估列线图的效能。结果 AKI LT 术后 7 d 内发生率为 53% 185/349)。多元 Logistics 回归分析显示年龄(OR 2.049 ;95% CI 1.252 ~ 3.352)、BMI(OR 2.041 ;95% CI 1.251 ~ 3.329)、 术 前 γ- 酶(γ-glutamyl transpeptidase,GGT)(OR 2.261 ;95% CI 1.288 ~ 3.970)、术后乳酸峰值(OR 2.917 ;95% CI 1.798 ~ 4.733)、腹水(OR 1.874 ;95% CI 1.104 ~ 3.180)、术前白蛋白(OR 0.475 ;95% CI 0.271 ~ 0.832)是 LT 术后 AKI 的独立危险因素,在预测 LT 术后早期 AKI 的列线图中,其内部验证 C-index 0.755,并且校准曲线斜率接近于1表明校准良好。结论 本研究所得列线图在预测肝移植术后早期肾功能不全方面具有较好的临床应用价值。

关键词:

急性肾损伤 , 肝移植 , 列线图 , 风险预测模型

Abstract:

Objective The occurrence of early acute kidney injury(AKI)after liver transplantation(LT is one of the common problems affecting the longterm prognosis of patients. This study tried to create a nomogram to accurately predict the occurrence of AKI after LT. Methods We retrospectively analyzed 369 patients who were treated in our hospital between July 1,2017 and December 31,2020,and 349 patients were included in this study. We used univariate and binary logistic regression analysis to identify the significantly related predictors for occurrence,and used these predictors to create a nomogram risk prediction model and evaluated the effectiveness of the nomogramby using the internally verified consistency index(C-index). Results The occurrence rate of AKI after LT was 53%(185/349)within 7 days after LT. Multiple logistic regression analysis found age(OR = 2.049 ;95% CI =1.252 ~ 3.352),BMI(OR = 2.041 ;95%CI = 1.251 ~ 3.329),preoperative γ-glutamyl transpeptidase(GGT)(OR = 2.261;95%CI = 1.288 ~ 3.970),postoperative lactic acid peak(OR = 2.917;95%CI = 1.798 ~ 4.733),ascites(OR = 1.874;95%CI = 1.104 ~ 3.180)and preoperative albumin(OR = 0.475;95% CI = 0.271 ~ 0.832)were independent risk factors for AKI after LT. The nomogram for predicting early AKI after LT verified C-index was 0.755,the slope of the calibration curve was close to 1 which was well calibrated. Conclusion The nomogram obtained in this study has good clinical application value in predicting early AKI after LT. 

Key words:

Acute kidney injury, Liver transplantation, Nomogram, Risk prediction model