Postoperative renal replacement therapy in patients with acute Stanford type A aortic dissection: risk factors and predictive modeling
10.3760/cma.j.cn112434-20190909-00309
- VernacularTitle:急性Stanford A型主动脉夹层患者手术后肾脏替代治疗：危险因素分析和预测模型建立
- Author:
Liang HONG
1
;
Xiao SHEN
;
Cui ZHANG
;
Fuhua HUANG
Author Information
1. 南京医科大学附属南京医院 南京市第一医院重症医学科 210006
- From:
Chinese Journal of Thoracic and Cardiovascular Surgery
2020;36(10):621-626
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To analyze the risk factors for postoperative renal replacement therapy(RRT) in patients with acute Stanford type A aortic dissection. Develop and validate a prediction model based on the risk factors with the purpose of early intervention.Methods:A retrospective analysis of 215 patients who underwent surgery for acute Stanford type A aortic dissection in our hospital from April 2016 to April 2019 were performed. Clinical variables including age, gender, basal blood pressure, preoperative serum creatinine, intraoperative blood pressure, operation time, cardiopulmonary bypass time, aortic occlusion time, intraoperative blood transfusion(including autologous blood), intraocular fluid infusion, colloidal fluid infusion, intraoperative urine volume, bleeding volume, total fluid balance, and postoperative blood lactate value were collected and their association with renal replacement therapy were analysed. Clinical variables were screened using lasso regression. Applying the post-filtering variables to construct a predictive model, calculating the area under the receiver operating characteristic curve( AUC) of the predictive model and the sensitivity and specificity under the optimal threshold for model evaluation. Results:In the 215 patients with acute Stanford type A aortic dissection, 38 patients required renal replacement therapy, accounting for 17.67%. Preoperative serum creatinine, operation time, cardiopulmonary bypass time, aortic occlusion time, intraoperative blood pressure less than 80mmHg time, intraoperative blood pressure less than 55% of basal blood pressure time, intraoperative blood transfusion, intraoperative crystal fluid dosage, intraoperative urine volume and lactate value after ICU admission were important risk factors for postoperative renal replacement therapy(RRT) in patients with acute Stanford type A aortic dissection. The AUC for the predictive model established using these variables was 0.955(95% CI: 0.897-1.000). The specificity under the optimal threshold was 96.1% and the sensitivity was 90.9%. Conclusion:Perioperative clinical variables can predict the possibility of RRT in patients with acute Stanford type A aortic dissection after surgery, which may provide the possibility for early intervention.
