Variability in intraoperative and postoperative fluid removal strategies in neonates undergoing cardiac surgery and associations with clinical outcomes

Background: Various fluid removal strategies are utilized to mitigate the deleterious effects of fluid accumulation in neonates following cardiac surgery. We aim to describe practice patterns and identify associations of fluid management strategies with outcomes in neonates undergoing cardiac surgery using the Neonatal and Pediatric Heart and Renal Outcomes Network dataset. Methods: Multicentre retrospective study including neonates who underwent cardiac surgery between 2015 and 2018. Primary outcomes were duration of mechanical ventilation, delay in first negative daily fluid balance, peak cumulative fluid balance, and severe persistent acute kidney injury. Multivariable logistic and ordinal regression models assessed associations between fluid removal strategies and clinical outcomes. Results: Of 2,240 neonates, 25% ( n = 560) received furosemide in the operating room. Peritoneal catheters were placed in 24% ( n = 527), and 8.6% ( n = 193) underwent prophylactic peritoneal dialysis. Diuretic use on postoperative day 0 ranged 6–95% across institutions. Diuretics on postoperative day 0 were associated with shorter time to negative daily fluid balance (odds ratio 0.55; 95% confidence interval 0.30–0.995) and shorter duration of mechanical ventilation (odds ratio 0.58; 95% confidence interval 0.39–0.86). Intraoperative diuretics were also associated with shorter duration of mechanical ventilation (odds ratio 0.60; 95% confidence interval 0.42–0.86). Receipt of continuous diuretic infusion on postoperative day 0 or 1 was associated with longer duration of mechanical ventilation (odds ratio 3.02; 95% confidence interval 1.98–4.58). No association of diuretic use with severe persistent acute kidney injury was found. Conclusions: Early use of diuretics in neonatal cardiac surgery is highly variable and has a differential association with outcomes. Future studies that standardize the dose, timing, and mode of administration (intermittent vs continuous) of diuretics are warranted to identify potential associations with outcomes.