Real-World Testing of a Clinical Strategy to Start Early Peritoneal Dialysis for High-Risk Newborns after Cardiac Surgery

Abstract

Key Points:

1. Inability to achieve negative fluid balance in postoperative 24 hours may be a reliable surrogate marker to start early peritoneal dialysis (PD) after cardiac surgery.

2. Prolonged cardiopulmonary bypass and aorta cross-clamp duration may determine PD catheter placement in the operating room.

3. The first postoperative 8 hours was indiscriminative for the decision to start early PD for these high-risk newborns.

Background: The beneficial effect of peritoneal dialysis (PD) catheter placement after cardiopulmonary bypass (CPB) in young infants has been demonstrated. However, the indications to start early PD are not agreed upon.

Methods: This retrospective single-center study was conducted to evaluate the performance of a clinical strategy for early PD start. PD catheters were placed in the operating room after CPB. Those with prolonged CPB times (>180 minutes), postoperative (postop) oligoanuria, and/or inability to achieve negative fluid balance in postop 24 hours were evaluated as high risk and selected for early PD (PD [+]) start. All PD (+) were started within the first postop 24 hours. Primary outcomes were 5% fluid accumulation at postop 48 hours and severe AKI on postop day (POD) 5.

Results: There were 49 newborns. Twenty-nine newborns were early PD (+) starts, and 20 used the PD catheter as an abdominal drain (PD −). Baseline demographic data were similar. Both groups were oliguric during first postop 8 hours (P = 0.906). The early PD (+) group produced significantly less urine output during POD 1 (0.98 versus 3.02 ml/kg per hour; P = 0.001). At postop 48 hours, the early PD (+) group had a similar prevalence of 5% fluid accumulation as early PD (−): 5 (16.7%) versus 2 (7.41%), respectively (P = 0.427). Severe AKI incidence on POD 5 was similar between the groups (17.3% versus 5.0%; P = 0.204). Time to extubation was longer for the early PD (+) group compared with the PD (−) group: 10.0 days (7.0–16.0) versus 4.0 days (4.0–10.0), respectively (P = 0.017).

Conclusions: Persistent oliguria and inability to achieve negative fluid balance during initial postop 24 hours may identify those newborns who will benefit from early PD. The first postop 8 hours was indiscriminative for this strategy. PD start may ameliorate the disadvantage for the designated group.

Keywords: congenital heart surgery; early peritoneal dialysis; fluid overload; newborn; urine output.

Graphical abstract

Figure 1

Flow chart for the clinical strategy to start early PD. Newborns with high-risk congenital cardiac anomalies were identified. Those with CPB duration >180 minutes and aorta cross-clamp time >70 minutes received PD catheters in the OR. All newborns are observed for up to 24 hours after surgery for urine output and ability to achieve negative fluid balance. Those who demonstrated suboptimal response to IV fluid replacement and furosemide administration were started on early PD. For the others, the PD catheter was used as an abdominal drain, when indicated. CPB, cardiopulmonary bypass; IV, intravenous; OR, operating room; PD, peritoneal dialysis.

Figure 2

Comparative daily fluid balance (ml/kg per 24 hour) for the postop 5 days, stratified according to early PD (+) versus early PD (−). *P < 0.05. postop, postoperative.

Figure 3

Comparative urine outcomes for the first postop 5 days, including the immediate postop 8 hours (u0). (A) Early PD (+) versus early PD (−). (B) 5% fluid accumulation at postop 48 hours versus <5% fluid accumulation. (C) Severe AKI on postop 48 hours versus no severe AKI. (D) Severe AKI on postop day 5 versus no severe AKI. *P < 0.1, **P < 0.05, ***P<0.001.

Figure 4

Comparative serum creatinine trajectories with preoperative, immediate postop, and postop trajectory values, stratified according to early PD (+) and early PD (−) groups. **P < 0.01, *P < 0.05, P < 0.1.

Figure 5

Comparative incidence of metabolic acidosis and metabolic alkalosis stratified according to early PD (+) versus early PD (−) groups for the postop 5 days. Metabolic acidosis was more likely for the early PD (+) group starting on POD 1 until POD 4. POD, postoperative day.