Objective: To analyze multiple factors that may affect renal function in septic shock patients with acute kidney injury (AKI) in the intensive care unit (ICU), in order to find factors of predictive value for renal function change in those patients.
Methods: Septic patients with AKI admitted to department of critical care medicine of Wuhan University Zhongnan Hospital from January 2017 to June 2019 were enrolled, and the patients were divided into renal function improvement group and renal function non-improvement group according to their renal function change. Baseline, laboratory and clinical indicators of them were collected to conduct retrospective analysis. Comparing the difference of each index between the two groups, the statistically significant indexes in the univariate analysis were selected to perform ridge regression analysis. The receiver operating characteristic (ROC) curve and its 95% confidence interval (95%CI) were used to analyze the predictive value of each influencing factor on the recovery of renal function in patients.
Results: A total of 323 patients met the inclusion criteria, and 195 of them were divided into renal function improvement group while the other 128 of them into the renal function non-improvement group. Univariate analysis showed that, there was significantly difference in acute physiology and chronic health evaluation II (APACHE II), sequential organ failure assessment (SOFA), Glasgow coma score (GCS), heart rate (HR), serum creatinine (SCr), blood urea nitrogen (BUN), potassium (K+), white blood cell count (WBC), maximum central venous pressure (CVPmax), maximum-minimum central venous pressure distance (ΔCVP), fluid balance, maximum lactic acid (LACmax), and maximum norepinephrine infusion speed (NEmax) between the renal function improvement group and the renal function non-improvement group. Ridge regression analysis of those indexes found that APACHE II, SOFA, SCr, BUN, HR, WBC, fluid balance, and NEmax were influential factors of non-improvement renal function (t values were 5.507, 3.690, 2.026, 4.815, 2.512, 2.114, 3.532, 3.735, all P < 0.05). ROC analysis found the predictive value combining the APACHE II, SOFA, BUN, NEmax was the highest [the area under ROC curve (AUC) and 95%CI: 0.863 (0.821-0.899)], which had a higher AUC than any of APACHE II, SOFA, BUN, SCr and NEmax [AUC and 95%CI: 0.863 (0.821-0.899) vs. 0.755 (0.705-0.801), 0.722 (0.670-0.770), 0.738 (0.686-0.785), 0.743 (0.692-0.790), 0.748 (0.697-0.794), all P < 0.01], and so did it when compared to APACHE II, SOFA, SCr and NEmax combination [AUC and 95%CI: 0.863 (0.821-0.899) vs. 0.825 (0.799-0.865), P < 0.01].
Conclusions: APACHE II, SOFA, SCr, BUN, HR, WBC, fluid balance, and NEmax are the positive influencing factors for patients without renal function improvement. The combination of APACHE II, SOFA, BUN, and NEmax had a relatively high predictive value for renal function recovery.