(Screen 1 - 6:00 PM Friday) Early postoperative weight-based fluid overload is associated with worse outcomes after neonatal cardiac surgery: a report from the multicenter NEPHRON collaborative

Find board locations here.

Abstract:

Introduction: Excess fluid balance (FB) is associated with morbidity and mortality after neonatal cardiac surgery in single center studies. FB is frequently measured as intake minus output (FB-IO). Yet a recent multicenter analysis of 2223 postoperative neonates showed neither early nor peak cumulative FB-IO was associated with worse outcomes. Alternatively, FB can be measured as change from preoperative weight (FB-W), which may avoid inaccuracies specific to FB-IO. Single-center studies have associated FB-W with worse outcomes, but multicenter data is lacking. This secondary NEPHRON analysis aimed to describe the multicenter epidemiology of early fluid overload, as defined by FB-W.

Methods: This observational study included a subset of 2235 neonates (≤30 days) undergoing cardiac surgery with or without cardiopulmonary bypass (CPB) from 22 hospitals in the NEonatal and Pediatric Heart and Renal Outcomes Network (NEPHRON) data registry; only those with postoperative day (POD) 2 weight recorded were included. POD2 FB-W was calculated as: 100 x (POD2 weight – preoperative weight)/preoperative weight). Patients were stratified by >10% and ≤10% POD2 FB-W and characteristics were compared. A generalized linear model assessed for independent association of POD2 FB-W ≥10% with mortality and duration of postoperative mechanical ventilation, respiratory support, inotropes and hospital length of stay (LOS).

Results: 998 neonates (CPB n=658, non-CPB n=340) were weighed on POD2, with 13% (n=129) premature, 23% (n=225) single ventricles, 18% (n=177) chromosomal anomalies, and 65% (n=649) STAT 4-5. Median POD FB-W was 8.1% (IQR 0.5-17.9); 45% (n=444) had FB-W >10%. Odds of POD2 FB-W >10% were higher for single ventricles (OR 1.55, 95% CI 1.023-2.35), STAT 4-5 (1.88, 95% 1.29-2.74), CPB (2.13, 95% 1.39-3.26), delayed sternal closure (1.83, 95% 1.16-2.87), and POD1 mechanical ventilation (2.97, 95% 2.01-4.40).
Hospital mortality was 2.8% (n=28). POD2 FB-W >10% was not associated with mortality (OR 1.04, 95% CI 0.29-3.68) but independently associated with longer durations of mechanical ventilation by 19% (multiplicative rate 1.19, 95% CI 1.04-1.36), respiratory support by 28% (1.28, 95% CI 1.07-1.54), inotropic support by 38% (1.38, 95% CI 1.10-1.73), and postoperative hospital LOS by 15% (1.15, 95% CI 1.03-1.27). POD2 FB-W >5% and 20% were also associated with all utilization outcomes while POD2 FB-IO and severe postoperative AKI were not.

Conclusions: This is the first multicenter study to evaluate the impact of weight-based fluid balance on outcomes after neonatal cardiac surgery. POD2 FB-W >10% is associated with longer cardiorespiratory support and postoperative hospital length of stay, but not with mortality. Even low thresholds of fluid accumulation (FB-W >5% and >10%) were associated with worse outcomes. Concurrent intake-output based fluid balance in the same population has no association with outcomes. Initiatives aimed at mitigating weight change based postoperative fluid accumulation are warranted and have potential to improve outcomes. To perform these initiatives and extend our POD2 FB-W findings to earlier time periods and the entire neonatal cohort, increasing the proportion of neonates weighed in the early postoperative period is necessary.