Dr. Sehrish Ali (@sehrish_alii) is an Assistant Professor at Baylor College of Medicine in Houston, TX, where she completed her fellowship under the Transition of Care clinical pathway. At BCM, she serves as the Associate Program Director of Nephrology Fellowship. Her clinical interests include women’s health and kidney disease, gender disparities and kidney disease, education, and wellness.

Dr. Sarwar Khan is a new Assistant Professor at Baylor College of Medicine in Houston, Texas. He completed his medical school and residency at McGovern Medical School, also in Houston. He split his fellowship years between UTSW in Dallas and Baylor College of Medicine. He is passionate about teaching trainees at all levels.

Maternal health has long-lasting implications on fetal, pediatric, and adult outcomes. It is evident that what happens during gestation can affect long-term kidney and overall health, but the factors that influence these outcomes are understudied and lead to a lack of evidence-based care in this population.

The kidneys start development at three weeks of gestation and nephron development is not complete until 36 weeks of gestation. During this development period, there is a risk of kidney maldevelopment, which is more formally known as congenital anomalies of the kidney and urinary tract (CAKUT). These anomalies include kidney hypoplasia, renal agenesis, and ureteral abnormalities. Known maternal risks for CAKUT are cocaine, ethanol, diabetes, obesity, vitamin A deficiency, folic acid deficiency, and renin-angiotensin-aldosterone system inhibitors. Other than maternal-environmental factors, known genetic mutations cause approximately 20% cases of CAKUT.

The same factors that can cause CAKUT may result in reduced nephron number without overt anatomical changes in the kidney. We see patients progress in their kidney disease despite being on cutting edge treatments. This could be in part because we have already missed the initial insult, occurring in utero, that may make a person more prone to CKD in the long run. This concept of congenitally reduced nephron mass/number gives a simple yet powerful explanation of why certain people develop chronic kidney disease (CKD) while others don’t, even with the same underlying medical conditions. This is highlighted in diabetics of equal severities where completely different levels of kidney disease are expressed. Imagine two patients with uncontrolled diabetes and retinopathy where one has nephrotic range proteinuria and declining estimated glomerular filtration rate (eGFR), while the other has preserved eGFR and minimal proteinuria. One potential explanation proposed for this phenomenon is congenitally reduced nephron number. Fewer nephrons allow for less adaptability to kidney injury, less filtration and ability to secrete sodium, therefore accelerating hypertension and causing glomerular hyperfiltration. The most important predictor of this is birth weight. Therefore, it is important to maximize the number of nephrons to guard against CKD.

The study by Tain et al provides a potential new cause of aberrant fetal kidney development that may influence the course of CKD development and progression. Their study explored how gestational exposure to maternal systemic corticosteroids could be associated with pediatric kidney outcomes.

Taking a step back to put this study into context, there are established uses of corticosteroids in pregnancy, the most common being inducing fetal lung maturity in pre-term deliveries. The scope of this intervention is relatively large with approximately 10% of children being born pre-term and the question of steroid administration being asked in each of these births. The American College of Obstetricians and Gynecologist (ACOG) recommends a single course of steroids for pregnant patients between 24 and 34 weeks gestation that have an impending risk of pre-term delivery within one week. There is a large amount of evidence for this intervention at the level of a Cochrane review, indicating reductions in neonatal mortality, perinatal death, respiratory distress, and other major outcomes.  The long-term effects of steroids are less well understood. Figure 1 outlines some of the potential side effects of gestational corticosteroid use.

Figure 1. Potential long-term effects of gestational corticosteroid use. © Ali and Khan

In the study by Tain et al, the investigators retrospectively designed cohorts from a Taiwanese database comparing exposure and non-exposure to corticosteroids in pregnant women and their eventual child. The primary outcome of interest was rates of CKD development in the children grouped into CAKUT-related and other causes of CKD. The follow-up period for these outcomes was ten years. Further secondary analysis was done subdividing the cohorts into dichotomous variables of sex, gestational weight, and pre-term (before 37 weeks) vs term birth (after 37 weeks). This helps understand which specific fetal characteristics when exposed to steroids may predispose to the development of CKD. The total dose, duration, and timing of corticosteroids exposure was also assessed.

The cohorts were relatively large with 966 in the steroid-exposed group and 22,397 in the steroid-free group. With a study like this, confounding by indication can play a significant role as the reason patients get the intervention may, in fact, be the cause of the outcomes, i.e. mothers/fetus pairs who end up needing steroids already have comorbidities that can cause CKD such as gestational diabetes, hypertension, and pre-eclampsia. To control for these confounding effects, inverse probability of treatment-weighting (IPTW) was done.

The primary outcome showed gestational steroid exposure was significantly associated with higher rates of childhood CKD. Higher event rates of CAKUT and non-CAKUT CKD were seen in the exposure group vs the non-exposure group, although when divided into the two specific outcomes, statistical significance was not achieved. In the secondary heterogeneity analysis, it was noted that birth before the 37^th week of gestation significantly increased the risk of CKD development. Additionally, second trimester exposure was also associated with higher rates of CKD. Total dose of steroids also was associated with higher CKD development.

This study is an excellent step towards understanding the factors that may influence fetal kidney development. However, given the clear benefits of steroids in fetuses 24-34 weeks regarding fetal lung maturity and other resultant outcomes, there would have to be convincing data to make changes to current clinical practice. Less routine indications or multiple courses of steroids may be seen as more problematic and called into question given this signal to harm regarding CKD development. Past 34 weeks gestation, there is much less consensus to steroid benefit in terms of mortality and pulmonary development. With the results of this present study and the other signal towards neurodevelopmental harm, the scales may tip further into cautioning against the routine use of steroids past 34 weeks gestation and offer further exploration into fetal determinants of kidney health. More research is needed on gestational factors that can affect fetal kidney outcomes for clinicians to better help patients. Addressing issues as early as fetal development would give a stronger foundation for long-term kidney health. A task which, though challenging, has great potential.