Furosemide response predicts acute kidney injury in children after cardiac surgery.

OBJECTIVE

A standardized assessment of response to furosemide is predictive of acute kidney injury progression in adults, but a paucity of data exists in pediatric patients. We evaluate furosemide responsiveness in a multicenter cohort of pediatric patients after cardiac surgery.

METHODS

Children who underwent cardiac surgery with a Society of Thoracic Surgeons-European Association for Cardiothoracic Surgery score of 3 or greater were retrospectively identified. The first dose of furosemide after surgery was recorded, and hourly urine output for 6 hours was recorded after the index dose. Urine flow rate calculated as urine output per hour was used to predict development of acute kidney injury.

RESULTS

A total of 166 patients from 4 institutions (median age, 6.3 months; interquartile range, 0.4-27.7) were included. Acute kidney injury occurred in 54 patients (33%). Compared with those without acute kidney injury, the 2- and 6-hour urine flow rates were significantly lower in patients in whom acute kidney injury developed: 2.9 (0.9-6.5) versus 5.0 (2.5-9.0) mL/kg/h for 2-hour urine flow rate, P = .004, and 2.4 (1.2-4.0) versus 4.0 (2.3-5.9) mL/kg/h for 6-hour flow rate, P = .001. In multivariable regression analysis, 2-hour (odds ratio, 1.2, P = .002) and 6-hour (odds ratio, 1.40, P < .001) urine flow rates were independently associated with acute kidney injury development. Lower urine flow rate at both 2 and 6 hours was also independently associated with longer hospital length of stay.

CONCLUSIONS

Lower urine flow rate after furosemide administration, when evaluated in a heterogeneous cohort of children from multiple institutions after pediatric cardiac surgery, was independently associated with subsequent acute kidney injury and longer length of stay. Future prospective studies are needed to validate furosemide responsiveness as a predictor of acute kidney injury.