Normal saline versus Normosol™-R in sepsis resuscitation: A retrospective cohort study

Abstract

Objective To determine the effect of Normosol™-R as compared to normal saline on the outcomes of acute kidney injury and the need for renal replacement therapy in the resuscitation phase of sepsis. Design Our study is a retrospective before-and-after cohort study. Setting The study occurred at a 700-bed tertiary academic level 1-trauma center. Patients A total of 1218 patients were enrolled through emergency department admissions. The normal saline (before) cohort was defined as the dates between 1 March and 30 September 2014 and the Normosol™-R (after) cohort was assessed from 1 March to 30 September 2015. Interventions None. Measurements and main results Intravenous fluid volumes received during the first 24 h, 72 h, and total hospital stays were compared. Sodium, chloride, potassium, and bicarbonate levels at 72 h were also compared. The medical coded diagnosis of acute kidney failure, need for renal replacement therapy, hospital LOS, ICU admission, ICU LOS, in-hospital mortality, and need for mechanical ventilation were all compared. There was no significant difference in intravenous fluid volumes between groups. Regression modelling controlling for baseline characteristics and 24-h fluid intake volume found no differences between groups for the primary outcomes of acute kidney injury ( P = 0.99) and renal replacement therapy ( P = 0.88). Patients in the Normosol™-R cohort were found to have a lower rate of hyperchloremia at 72 h post-admission (28% vs. 13%, P < 0.0001). There was a trend toward a decrease in the hospital and ICU LOS in the Normosol™-R cohort; however, the data were not statistically significant. Conclusions This study was unable to detect any difference in outcomes between sepsis patients who received intravenous fluid resuscitation with either a balanced crystalloid (Normosol™-R) or normal saline, except for a decreased rate of hyperchloremia.