Furosemide Response Predicts Acute Kidney Injury After Cardiac Surgery in Infants and Neonates.

OBJECTIVE: Cardiac surgery-induced acute kidney injury occurs frequently in neonates and infants and is associated with postoperative morbidity/mortality; early identification of cardiac surgery-induced acute kidney injury may be crucial to mitigate postoperative morbidity. We sought to determine if hourly or 6-hour cumulative urine output after furosemide in the first 24 hours after cardiopulmonary bypass could predict development of cardiac surgery-induced acute kidney injury and other deleterious outcomes. DESIGN: Retrospective chart review. SETTING: Pediatric cardiac ICU. PATIENTS: All infants younger than 90 days old admitted to the cardiac ICU from October 2012 to December 2015 who received at least one dose of furosemide in the first 24 hours after cardiopulmonary bypass surgery. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Ninety-nine patients met inclusion and exclusion criteria. In total, 45.5% developed cardiac surgery-induced acute kidney injury. Median time between cardiopulmonary bypass and furosemide was 7.7 hours (interquartile range, 4.4-9.5). Six-hour cumulative urine output was 33% lower (p = 0.031) in patients with cardiac surgery-induced acute kidney injury. Area under the curve for prediction of cardiac surgery-induced acute kidney injury was 0.69 (p = 0.002). Other models demonstrated urine output response to furosemide had significant area under the curves for prediction of peak fluid over load greater than 15% (0.68; p = 0.047), prolonged peritoneal dialysis (area under the curve, 0.79; p = 0.007), prolonged mechanical ventilation (area under the curve, 0.79; p < 0.001), prolonged hospitalization (area under the curve, 0.62; p = 0.069) and mortality (area under the curve, 0.72; p = 0.05). CONCLUSIONS: Urine output response to furosemide within 24 hours of cardiopulmonary bypass predicts cardiac surgery-induced acute kidney injury development and other important morbidity in children younger than 90 days old; prospective validation is warranted.

Decreasing IV Potassium in Pediatric Cardiac Intensive Care: Quality Improvement Project.

OBJECTIVES: IV potassium supplementation is commonly used in the pediatric cardiovascular ICU. However, concentrated IV potassium chloride doses can lead to life-threatening complications. We report results of a quality improvement project aimed at decreasing concentrated IV potassium chloride exposure. DESIGN: Retrospective evaluation of a quality improvement project aimed at reducing IV potassium chloride exposure. SETTING: Pediatric cardiac ICU. PATIENTS: All patients admitted to pediatric cardiac ICUs in April 2013 to September 2013 (preprotocol) and October 2013 to April 2014 (postprotocol). INTERVENTIONS: A quality improvement team developed a potassium protocol aimed at maintaining serum potassium levels 3.0-5.5 mEq/L, via algorithm focused on early enteral supplementation. All patients receiving IV diuretics who had a serum potassium level less than 4.5 mEq/L and urine output more than 0.5 mL/kg/hr had protocol initiated with potassium chloride-containing IV fluids or enteral potassium chloride. Concentrated IV potassium chloride infusions were limited to asymptomatic patients with serum potassium less than 2.0 mEq/L and high-risk patients at less than 3.0 mEq/L. Serum potassium levels were measured once daily, and protocolized adjustments were made based on this level and concurrent diuretic therapy. MEASUREMENTS AND MAIN RESULTS: Serum potassium, potassium chloride supplementation, patient cost, fluid administration, and arrhythmia incidence were compared pre and post protocol. Four hundred forty-three admissions were included (234 pre protocol and 209 post protocol). No significant differences were found in demographics. There was no difference in mean morning serum potassium after protocol implementation (3.85 [0.77] mEq/L before protocol and 3.89 [0.75] mEq/L after protocol; p = 0.90). Concentrated IV potassium chloride administration was decreased by 86% (331 vs 47 doses). With protocol, there was decreased incidence in days with one measured episode of hyperkalemia (11 vs 4/1,000 patient-days; p = 0.02) and a trend toward decreased hypokalemia (433 vs 400/1,000 patient-days; p = 0.05). Arrhythmia incidence was similar (p = 0.59). CONCLUSIONS: Protocolized potassium management in pediatric cardiac intensive care patients decreased concentrated IV potassium chloride exposure and incidence of hyperkalemia. Lower potassium treatment threshold for IV potassium chloride was not associated with increased arrhythmias.