Tiny Kidneys, Big Impact

Since the seminal work by Brenner et al.1 in the 1990's, nephrologists have recognized the association between reduced nephron number because of prematurity or low birth weight and future risk of CKD and hypertension. As more infants are surviving at earlier and earlier gestational ages with advanced neonatal and nephrology care, this is going to be a growing issue for pediatric and adult nephrologists alike. Factors that may affect the risk of these infants developing future CKD or hypertension are poorly defined. One leading hypothesis is that postnatal steroid exposure may affect ongoing nephrogenesis and is supported by some evidence in animal models. Premature infants are commonly exposed to steroids in the antenatal and postnatal stages of development. Antenatal steroids (ANS) are indicated for all pregnant women between 22 and 33 weeks gestation who are at risk of premature delivery to accelerate lung development and reduce the risk of respiratory distress syndrome.2–4 Indications for postnatal steroids are more varied, but, similar to ANS, are associated with improved lung function and are commonly given in premature neonates to prevent or treat bronchopulmonary dysplasia.5,6 The effect of postnatal corticosteroids on development of other organ systems, such as the kidneys, is largely unknown. Schuh et al.7 sought to further analyze the effects of steroids on neonates born at <28 weeks gestational age in patients identified through a secondary analysis of patients enrolled in the Preterm Erythropoietin Neuroprotection Trial.8 This prospective multicenter study enrolled infants born at a gestational age of 24–27 weeks and collected data on their neonatal intensive care unit course and exposures and followed outcomes through 2 years of age including eGFR as determined by Chronic Kidney Disease in Children under 25 creatinine and cystatin C equation. A total of 831 infants survived to discharge and were included in this analysis with a high burden of exposure to steroids identified; 90% of the cohort was exposed to ANS and 47% were exposed to steroids postnatally. Dexamethasone was the most commonly used (n=238), followed by hydrocortisone (n=232). Fewer than half of the patients completed their planned follow-up at 22–26 months. Of them, 51.1% were exposed to postnatal steroids and 17.5% demonstrated reduced eGFR. Overall, there was no significant association between any steroid exposure and reduced eGFR. Not all steroids are created equal, however. Of the infants seen in follow-up, a majority were exposed to dexamethasone (33%) or hydrocortisone (31.3%). There was no association to reduced eGFR with either dexamethasone exposure or prednisone/solone; however, a significant two-fold increased odds of reduced eGFR was observed in hydrocortisone exposure (adjusted odds ratio, 1.97; 95% confidence interval CI, 1.01 to 3.87). Neonates exposed to hydrocortisone for 1–7 days had increased odds of reduced eGFR (adjusted odds ratio, 2.84; 95% CI, 1.06 to 7.62) compared with those never exposed to hydrocortisone. This association was interestingly NS in neonates exposed to hydrocortisone for 8–14 days. Timing of postnatal steroid exposure after birth may also affect ongoing nephrogenesis. When assessing the risk of reduced eGFR at follow-up to postmenstrual age (PMA) at the time of steroid initiation, there was no difference in risk between infants exposed at ≤25 weeks PMA, 26–28 weeks PMA, and ≥29 weeks using crude data. However, when adjusted for sex, 5-minute APGAR+size for gestational age, there was a significant association in neonates exposed to initial steroids at ≤25 weeks PMA to dexamethasone with a nine-fold increased odds (adjusted odds ratio, 3.4; 95% CI, 1.6 to 9.92), indicating that exposure to steroids in the first week of life had the highest association with reduced GFR. A strength of this study is its multicenter nature and rigorous prospective data collection. Some weaknesses were addressed by the authors themselves. Although type, timing, and duration of steroids were evaluated—the overall lifetime dosing of steroids was not addressed and indications for steroids were not well delineated. Indications for use of steroids could introduce additional unknown significant confounders—certain indications such as adrenal insufficiency or sepsis may indicate poor renal perfusion and may contribute to kidney injury and reduction of eGFR. There were a relatively small number of patients with 2-year follow-up and even fewer infants with blood and urine metrics at follow-up which may have biased results. This analysis of the Preterm Erythropoietin Neuroprotection Trial study by Schuh et al. highlights the potential impact of postnatal corticosteroid exposure on development of CKD in neonates, particularly those exposed at ≤25 PMA—a ripe timeline for nephrogenesis. Limiting duration of steroid exposure and timing of dosing has the potential to improve kidney outcomes in this high-risk population but needs to be balanced by potential impact to respiratory outcomes. We are not ready to change our standard of care on the basis of the results of this analysis. Large, prospective trials are needed to illuminate the effects of antenatal and postnatal corticosteroids on nephrogenesis and development of CKD in premature neonates balanced with the risk of chronic lung disease. With higher survival rates of premature infants, continued collaboration with neonatologists and pediatric nephrologists is necessary. This will allow for more insight in nephron endowment and modifiable risk factors such as steroid exposure on postnatal kidney development. The work of the Neonatal Kidney Collaborative has helped to move this field forward via development of a multidisciplinary team with a strong infrastructure for research, educational efforts for the next generation of pediatric nephrologists and neonatologists, and advocacy efforts.9 Understanding at the molecular and cellular level how common neonatal drug exposures, such as steroids, affect nephrogenesis may allow for individualized neonatal care and better long-term outcomes for this high-risk population.