Dose Prescription and Delivery in Neonates With Congenital Heart Diseases Treated With Continuous Veno-Venous Hemofiltration

OBJECTIVES: Renal replacement therapy may be required for acute kidney injury treatment in neonates with complex cardiac conditions. Continuous veno-venous hemofiltration is applied safely in this population but no published recommendations for dose prescription in neonates currently exist. The aim of our study was to evaluate the effects of a relatively small dialysis dose on critically ill neonates. DESIGN: Retrospective analysis of clinical charts. SETTING: Pediatric Cardiac ICU. PATIENTS: Ten critically ill neonates with severe acute kidney injury were analyzed. The primary indication for continuous veno-venous hemofiltration initiation was severe fluid overload with oligoanuria. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The median (range) age and weight were 3 days (1-12 d) and 2.6 kg (2.1-4.2 kg), respectively, whereas the median continuous veno-venous hemofiltration duration was 17 days (3-63 d). Median prescribed blood flow rate, replacement fluid rate, and net ultrafiltration rate were 12 mL/min (9-50 mL/min), 100 mL/hr (40-200 mL/hr), and 20 mL/hr (5-45 mL/hr), respectively. The median effluent-based continuous veno-venous hemofiltration dose was 35 mL/kg/hr (11-66 mL/kg/hr), whereas the median delivered daily Kt/V per session (24 hr) was 0.5 (0.01-1.8). However, for treatment sessions lasting less than or equal to 12 versus greater than or equal to 12 hours per session, the median prescribed effluent dose was 41 (11-66) and 32 (17-60) mL/kg/hr, respectively (p = 0.06), whereas the delivered creatinine daily Kt/V values were 0.3 (0.01-0.9) and 0.9 (0.5-1.8), respectively (p < 0.0001). An inverse correlation was found between delivered daily Kt/V and the blood concentration differences of both creatinine (r = -0.3; p = 0.0093) and urea (r = -0.3; p = 0.0028) measured at the end and the beginning of a 24-hour treatment. The decrease of creatinine concentration was significantly greater during 24-hour treatment sessions with a delivered daily Kt/V greater than 0.9 than during those with daily Kt/V less than 0.9. CONCLUSIONS: Based on these findings, we propose on a provisional basis the use of daily Kt/V as a measure of continuous renal replacement therapy adequacy for critically ill neonates.