Objective: To investigate the influence of fluid balance and model of renal replacement therapy (RRT) on renal function and prognosis of patients suffering from septic acute kidney injury (AKI).
Methods: A retrospective cohort analysis of 117 septic AKI patients who had undergone RRT between January 2009 and December 2014 was performed in the Second Affiliated Hospital of Nanjing Medical University. The patients were divided into positive fluid balance group (n = 52) and negative fluid balance group (n = 65) according to the total amount of fluid calculated from the difference between fluid administered and fluid lost during the first 1 week of RRT. The incidence of renal recovery and death of the patients by 60 days as the endpoint events were taken to judge the prognosis of two groups. RRT strategies included continuous renal replacement therapy (CRRT) and intermittent renal replacement therapy (IRRT). Multiple factors including estimated glomerular filtration rate (eGFR), sequential organ failure assessment (SOFA) score, RRT model, the accumulation of fluid before initiation of RRT, and negative fluid balance during RRT were analyzed for outcome predictors by Cox proportional hazards model.
Results: There were no differences between two groups regarding clinical characteristics. The percentage of receiving CRRT in the negative fluid balance group was slightly higher than that of the positive fluid balance group (52.31% vs. 36.54%, χ² = 2.899, P = 0.089). With Kaplan-Meier survival curves, it was shown that the patients of negative fluid balance group had a higher rate of recovery of renal function (χ² = 4.803, P = 0.028) and significantly lower mortality rate (χ² = 9.505, P = 0.002). The rate of recovery of renal function by 60 days was higher in the negative fluid balance group than that in the positive fluid balance group (47.69% vs. 28.85%, χ² = 3.991, P = 0.046), while the mortality rate was significantly lowered in the negative fluid balance group compared with that of the positive fluid balance group (40.00% vs. 67.31%, χ² = 4.378, P = 0.036). Cox multivariate regression was used for excluding confounding factors. After adjusting for the clinically relevant variables, RRT negative fluid balance was significantly associated with recovery of renal function [ hazard ratios (HR) = 2.440, 95% confidence intervals (95%CI) = 1.089-5.464, P = 0.030] and mortality (HR = 0.443, 95%CI = 0.238-0.822, P = 0.010]. Higher eGFR before RRT and CRRT were independent favorable factors for recovery of renal function (HR = 1.014, 95%CI = 1.003-1.026, P = 0.012; HR = 3.138, 95%CI = 1.765-7.461, P = 0.002), and higher SOFA score was associated with a significantly higher risk of death (HR = 1.115, 95%CI = 1.057-1.177, P < 0.001).
Conclusions: Once the patients with septic AKI showed the signs of fluid overload, timely RRT and effective removal of excessive liquid may reverse the adverse prognosis. RRT with negative fluid balance is beneficial for the recovery of renal function, and reduce the mortality in patients with septic AKI, and CRRT model is a good choice.