BACKGROUND: Urine alkalinisation with sodium bicarbonate decreases renal oxidative stress and might attenuate sepsisassociated acute kidney injury (s-AKI). The safety and feasibility of urine alkalinisation in patients at risk of s-AKI has never been tested.
METHODS: We randomly assigned patients at risk of s-AKI (those with systemic inflammatory response syndrome [SIRS], oliguria and elevated [≥150 µg/L] serum neutrophil gelatinase-associated lipocalin [sNGAL] concentration) to receive sodium bicarbonate (treatment group) or sodium chloride (placebo group) in a 0.5 mmol/kg bolus followed by an infusion of 0.2 mmol/kg/hour.
RESULTS: Among 50 patients with SIRS and oliguria, 25 (50%) had an elevated sNGAL concentration. Of these, 13 were randomised to receive sodium bicarbonate and 12 to receive sodium chloride infusion. Study drugs were infused for a mean period of 25.9 hours (SD, 10 hours). Severe electrolyte abnormalities occurred in seven patients (28%) (four [30.8%] in the treatment group and three [25%] in the placebo group). These abnormalities resulted in early protocol cessation in six patients (24%) and study drug suspension in one patient (4%). This adverse event rate was judged to be unacceptable and the study was terminated early. There was no difference between the two groups in sNGAL or urinary NGAL concentrations over time, occurrence of acute kidney injury, requirement for renal replacement therapy, hospital length-of-stay or mortality.
CONCLUSION: Administration of sodium bicarbonate and sodium chloride solutions to patients at risk of s-AKI was associated with frequent major electrolyte abnormalities and early protocol cessation. The tested protocol does not appear safe or feasible.