Preoperative Evaluation Of Acute Kidney Injury After Resection Of Colorectal Cancers

Objective: Acute kidney injury(AKI)is one of the postoperative complications of colorectal cancer resection,which has a negative impact on the recovery and prognosis of patients.The incidence rate of colorectal cancer is increasing year by year in China,and there are a large number of patients who need surgery at present and in the future.This study investigated the risk factors of acute renal injury following resection of colorectal malignant tumors,and established a nomogram prediction model to provide evidence for the monitoring and prevention of AKI after such surgery.Methods: A total of 2339 patients who underwent resection of colorectal malignancies at our hospital from December 2012 to August 2021 were included in the study.According to the diagnostic criteria for AKI by the 2012 Kidney Disease: Improving Global Outcomes(KDIGO),the subjects were divided into two groups: non AKI group and AKI group.The clinical data of the two groups were compared,Single factor analysis was used to identify statistically significant indicators,and logistic regression was used to screen out independent risk factors for AKI after resection of colorectal malignant tumors.Nomogram prediction model and Poisson regression model were constructed.The accuracy of the evaluation model was evaluated using subject performance characteristic curve(ROC),and precision-recall curve(PRC)was drawn to evaluate the precision and recall of the model.Result: The incidence of AKI after resection of colorectal malignancies in 2339 patients was 4.28%(100/2339).Compared with the group without AKI,patients with AKI were older,more likely to have a history of coronary heart disease,lower baseline serum creatinine,and more likely to use aspirin during the perioperative period(all P<0.05);The proportion of preventive ileostomy was higher(P=0.01);The systolic blood pressure was higher,the level of venous blood chloride ion was higher,the proportion of ARB drugs,pond potassium diuretics and carbapenems used in perioperative period was higher,and the proportion of laparoscopic surgery was lower(all P<0.01).Multivariate logistic analysis showed that systolic blood pressure(P=0.02),serum creatinine(P<0.05),chloride ion(P<0.01),perioperative use of ARB drugs(P<0.01),perioperative use of carbapenems(P<0.01),and preventive ileostomy(P=0.03)were independent risk factors for AKI,while laparoscopic surgery(P<0.01)was a protective factor.Using the above indicators to construct a nomogram,the area under the ROC curve(AUC)of the training set is 0.695,and the AUC of the verification set is 0.794.Multivariate Poisson regression analysis showed that systolic blood pressure(P=0.03),chloride ion(P=0.01),perioperative use of ARB drugs(P<0.01),perioperative use of carbapenems(P<0.01),and preventive ileocolonostomy(P=0.04)were independent risk factors for AKI after this surgery,and laparoscopic surgery(P<0.01)was a protective factor.The training set AUC of Poisson regression model is 0.702,and the validation set AUC is 0.767.The precision recall curve shows that the average precision(AP)in the Nomogram model and Poisson model validation sets is 0.186 and 0.174,respectively.Both models have certain detection effects,which can provide support for the diagnosis of AKI after resection of colorectal malignant tumors and reduce the risk of missed diagnosis.Conclusion: This study shows that systolic blood pressure,serum creatinine,blood chlorine,perioperative use of ARB drugs,perioperative use of carbapenems,and preventive ileocolonostomy are independent risk factors for acute renal injury after resection of colorectal malignant tumors.The Nomogram prediction model and Poisson regression model have high accuracy in predicting postoperative AKI in patients undergoing colorectal cancer resection,and can provide reference and methods for preoperative evaluation,prevention,and monitoring of AKI secondary to such surgery.