Datasets from 7 different TH-treated cohorts with 4724 Scr values in 1136 TH neonates were recently pooled, of whom 132 (11.6%) were classified as having AKI [5, 6]. To facilitate pooling, variables collected were restricted to birth weight, gestational age, neonatal survival, and the Scr values (no data on sex or body length). Postnatal day 1 was defined as the date of delivery [5]. In the original paper, we reported on centile (p10, p25, p50, p75, p90, and p95) Scr values from day 1 up to day 10.

These centile values were used to calculate the eGFR [eGFR (mL/min.1.73 m2) = 0.31 × height (cm)/Scr (mg/dL)], as reported by Smeets et al. [4], and recently applied by Muñoz et al. [2]. For ethics-related aspects, or additional methodological details or cohort characteristics, we refer to the original paper [5].

Centile eGFR values were calculated as suggested by Smeets et al. and as recently reported by Muñoz et al. [2, 4]. Since body length is only rarely collected and is notoriously known to show inter- and intra-observer variability in newborns, and since all cases were (near)term neonates, we arbitrarily used a fixed value (50 cm), extrapolated from the median birth weight (3350 g) of our dataset [5]. We hereby used the modified Schwartz formula, with the k value suggested by Smeets (k = 0.31) [4].

Finally, we are aware of a median mannitol clearance estimate in neonates undergoing TH and have recalculated this value (0.15 L/h, mean weight 3.25 kg [7]) to a measured GFR value (17 mL/min∙1.73 m2, on day 2) to enable comparison to the eGFR on day 2 in TH neonates [5, 7]. While this study reports on 77 mannitol measurements in 17 asphyxiated neonates, the study design (intermittent administration, at random sampling over the first 3 days of postnatal life), did not allow us to calculate mGFR values besides the reported median mannitol clearance value [7].

Postnatal day-specific eGFR values as reported for term neonates in the publications of Muñoz et al. (critical care), Wu et al. (mixed), and Smeets et al. (healthy) were extracted from the source documents [2,3,4]. For the Muñoz et al. and Smeets et al. papers, this was based on the values reported in the table(s) of their publications [2, 4]. For the Wu et al. paper, data were extracted from their Fig. 4 (gestational age 40.2 weeks, birth weight 3600 g), using extraction software (WebPlotDigitizer, version 5) [3]. For similarity, day 1 was defined as the day of delivery for all cohorts.

To estimate a postnatal eGFR equation in both TH cases and controls based on the 50th centile values, a regression-based approach was developed. Initially, we explored several models, including linear, logarithmic, exponential, and polynomial regression to determine the better model that fit the data. To evaluate the fit of each model, we used the Akaike information criterion (AIC) and the coefficient of determination (R2). These statistical metrics were used to assess the relative goodness of fit, with lower AIC values and higher R2 values indicating a better model.