Neonatal hypoxic-ischemic encephalopathy (HIE) affects 1–8 neonates per 1000 live births in both high- and low-income countries and is a leading cause of death and long-term neurological disabilities in neonates, worldwide [1-4]. Neonatal HIE results from cerebral dysfunction and injury due to hypoxia and hypoperfusion caused by perinatal asphyxia. Treatment options for this condition are limited and are essentially based on therapeutic hypothermia (TH) which has been shown to modestly reduced long-term neurological disabilities [5,6].

In addition to impairing the central nervous system, perinatal asphyxia often leads to multiorgan dysfunction, with the kidneys being among the most vulnerable organs affected. Acute kidney injury (AKI) develops in 30%–40% of neonates with moderate or severe HIE and associates with mortality and neurological disorders [7-9]. Its effects on the severity of brain injury are debated [10,11]. Neonatal AKI is mostly defined using the Kidney Disease: Improving Global Outcome (KDIGO) criteria, which include an increase in serum creatinine (SCr) levels (SCr ≥0.3 mg/dL within 48 hours) and low urinary output (oliguria ≤1 mL/kg/day) [12]. However, the AKI-KDIGO definition is not sufficiently sensitive to detect alterations in renal function in the days following birth due to the physiological decrease in SCr levels and the difficulties in rigorously assessing urine output which is often low in the postnatal period. An alternative definition based on the rate of SCr decline during the 7 days after birth has been proposed to identify neonates with impaired renal function (IRF) [13-15]. IRF may reflect haemodynamic disorders and hypovolemia, resulting in renal hypoperfusion, drug-induced nephrotoxicity, and kidney injury. Some studies have demonstrated that the IRF and AKI-KDIGO definitions are equally associated with adverse neonatal outcomes [14,15]. However, the prognostic value of IRF based on the rate of SCr level decline in neonatal morbidity and mortality, especially in neonates with HIE, requires further investigations.

This study aimed to investigate the association between IRF and brain injury severity in neonates treated with TH for HIE.

During the study period, 161 neonates were treated with TH for moderate or severe HIE. Fourteen neonates were excluded because of missing SCr level data (n=6), death before the third day of life (n=5), or not interpretable cerebral MRI (n=3) (Fig. 1). Among the 147 included neonates, 4 (2.7%) died early without cerebral MRI assessment (ED) and 28 (19%) had severe brain injury (SevereBI).

During the study period, 67 neonates (46%) had IRF and 22 of them (32.8%) had ED/SevereBI. In the subset of neonates meeting the KDIGO-AKI criteria (n=15), the composite outcome of ED/SevereBI was observed in 8 neonates (53.3%). But all of them were also classified as having IRF.

Compared to neonates without severe brain injury, those with ED/SevereBI were more likely to have lower Apgar scores (P<0.01), higher blood lactate levels at birth (P=0.02) and within the first hour after birth (+34%, P<0.001), and to require delivery room intubation (50% vs. 75%, P=0.01) (shown in Table 1). GA, sex, birth weight z score, mode of delivery, and perinatal sentinel events did not differ between the 2 groups of neonates. Nevertheless, the rate of IRF was higher in neonates with ED/SevereBI (39% vs. 69%, P<0.01). Fig. 2 illustrates the evolution of SCr in the 2 groups during the week following birth.

Neonates with IRF had more frequent severe brain injury on MRI (28.6% vs. 12.5%, P=0.01) compared to neonates without IRF. The mortality rate was also higher (25.4% vs. 10%, P=0.01). Characteristics of neonates with IRF are shown in Supplementary Table 1 and Supplementary Fig. 1.

The multivariable logistic regression analysis showed that neonates with IRF had 2- to 3-fold higher odds of ED/SevereBI than their counterparts (adjusted odds ratio [aOR] (95% confidence interval [CI]): 2.66 [1.09-6.84]; P=0.03).

In this study of neonates treated with TH for HIE, those who had IRF defined by a low rate of SCr level decline during the first week of life were at a high risk of ED/SevereBI.

This definition has been proposed as a surrogate marker of IRF in neonates [13,14]. The rate of SCr level decline considers the maturational process of the glomerular filtration rate (GFR) and the high susceptibility of the immature kidney to an adverse environment [20,21]. Following birth, there is a rapid increase in GFR that results in an exponential decrease in SCr levels, by more than 50% during the first week of life [13]. However, GFR is low, making newborns highly susceptible to various factors, including hemodynamic and oxygenation changes, nephrotoxic drug exposure, and fluid depletion. The neonatal AKI-KDIGO definition which is based on the rise in the SCr level (≥0.3 mg/dL within any 48-hour period) and/or low urine output (<1 mL/kg/hr), does not integrate this evolution and can underestimate newborns with mild kidney injury [14]. In addition, urine output is usually low in the first few days after birth, and oliguria is not always observed in infants with AKI. In a population of 106 neonates treated with TH for HIE, Gupta et al. [13] showed that neonates diagnosed with IRF based on SCr level decline or with the neonatal AKI-KDIGO definition have similar adverse outcomes, including high use of vasopressors, prolonged mechanical ventilation, fluid overload, longer hospital stay, high blood gentamycin levels, and high urinary markers of kidney injury compared to control newborns. In summary, during the postnatal period, the rate of SCr level decline may be considered a reliable marker of renal injury, with greater sensitivity than that of the neonatal AKI-KDIGO criteria.

The prognostic significance of IRF, as indicated by a low rate of SCr level decline during the first week after birth, in mortality and brain injury severity in newborns with HIE, has not yet been thoroughly investigated. In this study, we found an association between IRF and the composite outcome of ED/SevereBI (aOR [95% CI]: 2.66 [1.09–6.84]; P=0.03). Similar findings have been observed in a recent retrospective study of neonates treated with TH for HIE in a neonatal intensive care unit in the USA. The authors found that neonates with IRF based on SCr level decline had increased odds of the same composite outcome compared to the control group (aOR [95% CI]: 2.72 [1.18–6.26]; P=0.02) [15]. This study also showed that this definition captured more neonates with early-stage AKI than the KDIGO criteria: 51% of neonates had IFR (SCr level decline definition) versus 28.4% using the neonatal AKI-KDIGO criteria. The mechanism by which IRF associates with severe brain injury is unknown and complex. IRF can be considered a marker of the severity of perinatal asphyxia with renal injury due to a profound decrease in fetal systemic blood pressure and renal blood flow [22,23]. IRF may also act as a contributing factor for brain injury through diverse pathways. IRF induces systemic inflammation by stimulating proinflammatory cytokines release and reducing their clearance [24,25]. Metabolic changes associated with renal injury including electrolyte disturbances (hyponatremia), the accumulation of inorganic molecules and reduced drug clearance, in conjunction with a proinflammatory state, can affect the integrity of the blood brain barrier. These changes may perpetuate brain injury through enhancing cerebral edema, oxidative stress, microglial activation and accumulation of inflammatory cells, excitatory metabolism and endothelial dysfunction [26,27]. These findings, including those of the present study, indicate that IRF, based on the rate of SCr decline during the first week of life, could be considered a reliable marker of adverse neonatal outcomes in neonates with HIE.

Our study had some limitations. This was a retrospective, single-center study, and potential confounders were not considered. GFR measurements using creatinine clearance, urine output, fluid intake and electrolyte balance were lacking. Electrolyte imbalance, particularly hyponatremia, should also be considered; given its potential link to IRF. Long-term neurological and kidney outcomes were not collected; also, we could not evaluate the long-term consequences of IRF on neurodevelopment and kidney function [9,28,29].

Although the period of inclusion extended from 2015 to 2023, no substantial changes in healthcare management occurred during this period that might have significantly affected the incidence of IRF. In addition, the contemporary characteristics of the studied population managed with current recommendations strengthen the relevance of our findings. We believe that the results of this study may help clinicians identify newborns at high risk of neonatal morbidity and rapidly adapt clinical practices, including fluid therapy, electrolyte balance, and therapeutic options, all of which can improve neonatal healthcare.

In conclusion, in this study of neonates treated with TH for moderate or severe HIE, IRF based on SCr level decline was significantly associated with adverse neonatal outcomes. These findings underscore the clinical relevance of IRF during the days following birth. Further studies are required to investigate the long-term neurological and renal consequences of IRF.