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.

Peritoneal Dialysis International 2022 Vol 42(5) 460-469. doi: 10.1177/08968608221091023.

Sidharth Kumar Sethi, et al…Rupesh Raina.

Reviewed by: Sangeeta Hingorani


Neonatal AKI occurs commonly in the NICU and is associated with an increased morbidity and mortality. Management of AKI in the NICU is often supportive and dialysis is often challenging in newborns especially those born prematurely or with other comorbidities


The goal of this study was to prospectively follow babies in 11 centers in India who were admitted to Level 2-3 NICUs and determine risk factors for AKI and need for PD and to describe outcomes in these babies.

Methods/Study design:

Multicenter, national prospective cohort study (TINKER) of 11 centers with level 2-3 NICU across India. Neonates were admitted from 2018-2021.

Inclusion criteria– neonates <28 days who were admitted to the NICU and received IV fluids and/or nutrition for a minimum of 48 hours.

Exclusion criteria – death within 48 hours of admission, presence of a lethal chromosomal anomaly, cardiac surgery within the first week of life, and neonates receiving normal newborn care.

A total of 2839 neonates were screened and 1600 were enrolled into the study. 1109 neonates did not have AKI and 491 neonates did have AKI. Of those with AKI, 44 (9%) required PD.

PD was performed using a stiff/straight catheter and solutions were 1.5-1.7% dextrose with an average dwell time of 30 minutes, fill of 10, and drain of 20 minutes.


Indications for PD in the neonates with AKI were fluid overload, oliguria, acidosis, and/or hyperkalemia. PD was done for a median time of 58 hours (Interquartile range of 52-65 hours).  Complications included a leak, and impaired drainage due to a clot or catheter position. They did not have any episodes of peritonitis. They conclude that PD can be done successfully in neonates with AKI. Need for PD was higher in neonates with cardiac disease and higher creatinine in the first 12 hours after birth, whereas higher Apgar score at 5 minutes and higher hemoglobin were associated with a lower odds of needing PD. Mortality was higher in neonates who needed PD in the NICU – 70.5% vs 7.6%.


This is a large prospective, multicenter cohort study evaluating the use of PD for management of AKI in India. The findings of neonates who are sicker from the beginning are more likely to develop AKI and to need dialysis and that those neonates who need dialysis have a much higher mortality rate are not new. However, the use of PD successfully in neonates in this study provides more data to support its use in the US or other countries where hemodialysis is often the modality of choice.

It is interesting that the investigators included creatinine in the first 12 hours of life in the model because conventional thinking is that this number usually reflects maternal creatinine and not the neonate.  When defining AKI, many studies ignore the creatinine in the first 2-3 days of life. They offer no explanation as to why this might be an important variable. The values for creatinine between the groups were 1.8 mg/dL for those who needed PD compared to 1.2 mg/dL for those with AKI who did not need PD. It is unclear if this difference is clinically relevant (though they report statistical significance) and how to interpret it in this context. The analysis plan used a backward stepwise regression which may explain why some of the variables were chosen and included in the multivariable model, such as creatinine in the first 12 hours. None of the variables were chosen a priori. Of note, the decision when and on whom to start PD was left to the discretion of the center. Duration of PD but not on what hospital day PD started was reported. Volume overload in the first 12 hours of life was predictive of needing PD suggesting that aggressive hydration in these neonates soon after birth may be detrimental.