Chloride-liberal vs. chloride-restrictive intravenous fluid administration and acute kidney injury: an extended analysis

Purpose: In a previous study, restricting intravenous chloride administration in ICU patients decreased the incidence of acute kidney injury (AKI). To test the robustness of this finding, we extended our observation period to 12 months. Methods: The study extension included a 1-year control period (18 August 2007 to 17 August 2008) and a 1-year intervention period (18 February 2009 to 17 February 2010). During the extended control period, patients received standard intravenous fluids. During the extended intervention period, we continued to restrict all chloride-rich fluids. We used the Kidney Disease: Improving Global Outcomes (KDIGO) staging to define AKI. Results: We studied 1,476 control and 1,518 intervention patients. Stages 2 and 3 of KDIGO defined AKI decreased from 302 (20.5 %; 95 % CI, 18.5-22.6 %) to 238 (15.7 %; 95 % CI, 13.9-17.6 %) (P < 0.001) and the use of RRT from 144 (9.8 %; 95 % CI, 8.3-11.4 %) to 103 (6.8 %; 95 % CI, 5.6-8.2 %) (P = 0.003). After adjustment for relevant covariates, liberal chloride therapy remained associated with a greater risk of KDIGO stages 2 and 3 [hazard ratio 1.32 (95 % CI 1.11-1.58); P = 0.002] and use of RRT [hazard ratio 1.44 (95 % CI 1.10-1.88); P = 0.006]. However, on sensitivity assessment of each 6-month period, KDIGO stages 2 and 3 increased in the new extended intervention period compared with the original intervention period. Conclusions: On extended assessment, the overall impact of restricting chloride-rich fluids on AKI remained. However, sensitivity analysis suggested that other unidentified confounders may have also contributed to fluctuations in the incidence of AKI.