Risk factors and outcomes of neonates with acute kidney injury needing peritoneal dialysis: Results from the prospective TINKER (The Indian PCRRT-ICONIC Neonatal Kidney Educational Registry) study

Background: Acute kidney injury (AKI) is common in neonates admitted to neonatal intensive care units (NICUs). There is a need to have prospective data on the risk factors and outcomes of acute peritoneal dialysis (PD) in neonates. The use of kidney replacement therapy in this population compared to older populations has been associated with worse outcomes (mortality rates 17-24%) along with a longer stay in the NICU and/or hospital.

Methods: The following multicentre, prospective study was derived from the TINKER (The Indian PCRRT-ICONIC Neonatal Kidney Educational Registry) database, assessing all admitted neonates ≤28 days who received intravenous fluids for at least 48 h. The following neonates were excluded: death within 48 h, presence of any lethal chromosomal anomaly, requirement of congenital heart surgery within the first 7 days of life and those receiving only routine care in nursery. Demographic data (maternal and neonatal) and daily clinical and laboratory parameters were recorded. AKI was defined according to the Neonatal Kidney Disease: Improving Global Outcomes criteria.

Results: Of the included 1600 neonates, a total of 491 (30.7%) had AKI. Of these 491 neonates with AKI, 44 (9%) required PD. Among neonates with AKI, the odds of needing PD was significantly higher among those with significant cardiac disease (odds ratio (95% confidence interval): 4.95 (2.39-10.27); p < 0.001), inotropes usage (4.77 (1.98-11.51); p < 0.001), severe peripartum event (4.37 (1.31-14.57); p = 0.02), requirement of respiratory support in NICU (4.17 (1.00-17.59); p = 0.04), necrotising enterocolitis (3.96 (1.21-13.02); p = 0.03), any grade of intraventricular haemorrhage (3.71 (1.63-8.45); p = 0.001), evidence of fluid overload during the first 12 h in NICU (3.69 (1.27-10.70); p = 0.02) and requirement of resuscitation in the delivery room (2.72 (1.45-5.12); p = 0.001). AKI neonates with PD as compared to those without PD had a significantly lower median (interquartile range) duration of stay in NICU (7 (4-14) vs. 11 (6-21) days; p = 0.004), but significantly higher mortality (31 (70.5%) vs. 50 (3.2%); p < 0.001). This discrepancy is likely attributable to the critical state of the neonates with AKI.

Conclusions: This is the largest prospective, multicentre study specifically looking at neonatal AKI and need for dialysis in neonates. AKI was seen in 30.7% of neonates (with the need for acute PD in 9% of the AKI group). The odds of needing acute PD were significantly higher among those with significant cardiac disease, inotropes usage, severe peripartum event, requirement of respiratory support in NICU, necrotising enterocolitis, any grade of intraventricular haemorrhage, evidence of fluid overload more than 10% during the first 12 h in NICU and requirement of resuscitation in the delivery room. AKI neonates with PD as compared to AKI neonates without PD had a significantly higher mortality. There is a need to keep a vigilant watch in neonates with risk factors for the development of AKI and need for PD.