The Effect of Preterm Birth on Renal Development and Renal Health Outcome

1. Clinicians should be aware that preterm birth is associated with a reduced functional nephron mass, predisposing an individual to chronic kidney disease and quicker progression to end-stage chronic kidney disease.

2. Patients and health-care professionals must take an active role in protecting the kidneys of preterm survivors from further injury throughout life.

After completing this article, readers should be able to:

1. Improve awareness of NICU professionals on how to protect kidneys of preterm infants from further injury, particularly during the period of ex utero nephrogenesis.

2. Encourage surveillance of preterm survivors in childhood and adulthood to reduce obesity and observe for evidence of hypertension or insulin resistance to minimize further “hits” to an already reduced nephron mass.

3. Stimulate research in areas where there are gaps in the evidence, in particular long-term follow-up of cohorts of preterm survivors into late adulthood.

Fifteen million births occur before 37 weeks’ gestational age (GA) worldwide each year. (1) Over the past decade, preterm birth rates in the United States have varied between 9.8% and 11.4% and currently are at 9.9%. (2)(3) Around 2.8% of births are early preterm births occurring before 34 weeks’ gestation. (2) Although this group experiences a greater proportion of long-term morbidity, all preterm infants have higher rates of morbidity than their term-born contemporaries. (3) The preterm population is diverse in both GA and the underlying cause for prematurity. Conditions associated with an adverse intrauterine environment include preeclampsia, multiple births, and chorioamnionitis, and account for 38% of preterm deliveries. (3)

The renal consequences of preterm birth are becoming increasingly recognized and include a higher risk of chronic kidney disease (CKD), (4) a quicker progression of renal pathology, (5) and a predisposition toward hypertension. (6) CKD is associated with significant morbidity and mortality and is estimated to affect 14% of the US population. (7) The global burden of disease study highlighted that CKD is a growing public health problem that is increasingly contributing to global mortality. (8) As the preterm population surviving into childhood and adulthood grows, there is an increasing need to better understand the renal long-term effects of preterm birth and to identify ways to reduce morbidity associated with preterm survival.

Renal development begins with the differentiation of the pronephros in the third week of gestation followed by the mesonephros at 4 weeks of gestation. (9) The pronephros is nonfunctional in humans and involutes whereas the mesonephros transiently functions as an excretory structure until the definitive kidney develops from the metanephros. (10) The metanephros gives rise to the ureteric bud which undergoes branching; through reciprocal signaling, the branching of the ureteric bud progresses, and the mesenchyme differentiates to form the renal vesicles. The renal vesicles ultimately form the glomeruli, tubules, and loop of Henle of the mature nephron, with the first glomeruli present from 9 to 10 weeks’ gestation and fetal urine production beginning at 10 weeks. (9) The fetus becomes the main producer of amniotic fluid from around 16 weeks’ gestation. (9) Maximal human kidney growth has been shown to occur toward the end of the second and beginning of the third trimester between 26 and 34 weeks’ GA, (11) with nephrogenesis complete by term GA. (12)

Perinatal autopsy studies have demonstrated that there is individual variation in the GA at which nephrogenesis is complete; 1 group examined normally grown fetuses, demonstrating ongoing nephrogenesis at 31 weeks’ GA and complete nephrogenesis in all infants by 38 weeks’ GA, with some variability in those between 35 and 37 weeks. (12) Another group demonstrated that nephrogenesis was complete by 32 weeks’ GA in some fetuses. (13) Term infants are born with the total number of nephrons that they will have in their lifetime; optimizing nephrogenesis is advantageous for the long-term renal health of an individual.

Many preterm infants are born at a time when they are still undergoing nephrogenesis, with some born before reaching their maximal period of nephron development. Nephron development continues after very preterm birth with active glomerulogenesis, characterized by the presence of basophilic S-shaped bodies, and continuing until 40 days’ postnatal age. (14) Extremely preterm infants more than 40 days old who experienced acute renal failure (defined as a sustained elevation in creatinine >2.0 mg/dL) have fewer nephrons than preterm infants without a history of renal failure. During normal kidney development, nephrons grow from the corticomedullary junction toward a nephrogenic zone situated below the renal capsule, forming nephron “generations.” (13) The newest nephrons are therefore situated in the outer renal cortex. An autopsy study has recently shown that many preterm infants have abnormal glomeruli in their outer renal cortex, implying that nephrons that develop ex utero may not develop normally. (13) Autopsy studies have also shown that preterm birth is associated with accelerated renal maturation and the presence of a larger proportion of morphologically abnormal glomeruli compared with gestation-matched stillborn controls. (13) It is clear that the number of nephrons that form in the first 4 to 6 weeks after delivery is important in establishing the lifelong nephron mass of a preterm infant and that this should be supported as much as possible by avoiding postnatal kidney injury.

Preterm birth and reports of its effect on the glomerular filtration rate (GFR) have been contradictory. Preterm infants born before 32 weeks’ GA were found to have a reduced estimated GFR (eGFR), measured using cystatin C (CysC) at term postmenstrual age. (15) In contrast, an extremely preterm cohort of less than 28 weeks’ gestation at birth was shown to have an eGFR comparable to that of term infants using urinary CysC despite having smaller kidneys, leading the authors to conclude that this may be evidence of glomerular hyperfiltration and compensation for reduced nephron mass. (15)

Nephrons are the functional units of the kidney; the nephron mass of an individual refers to the total number of functioning nephrons a person has at any given time. (16) Data regarding normal human nephron number and density have predominantly been determined histologically from autopsy studies or from renal transplant donors. There is large variation in nephron number and density between individuals with 4- to 8-fold variations in nephron number observed at autopsy. (10) The reason for this degree of variability between individuals is not fully understood but is likely related to an interplay among genetics, the environment of an individual during nephrogenesis, and renal insults that occur over the course of a lifetime. (12)(17) Nephron mass naturally declines with age; a study examining the kidneys of healthy renal transplant donors demonstrated a 48% difference in the nonsclerotic glomerular number between donors aged 19 to 28 years and those aged 70 to 75 years. (18) Older donors have more sclerotic glomeruli. (18)

A reduced nephron mass has been associated with CKD (19)(20) and hypertension (21) in adulthood. Because it is not possible to directly measure nephron mass in living subjects, kidney size, and in particular, kidney length and volume, are often used as surrogate markers of nephron mass. (22) Unfortunately, ultrasonography has been shown to underestimate true renal volume by a median of 24% (range 5%–48%) in an animal model. (23) It has been proposed that measuring the renal parenchyma using ultrasonography may be a more reproducible and potentially more accurate tool for estimating normal and abnormal renal development. (24) Normal kidney size and volume do not differentiate between normal growth and compensatory hypertrophy of an oligonephronic kidney. (25) Renal parenchymal thickness has been shown to be closely associated with renal volume, (26) and renal cortical thickness has been shown to be correlated with renal function in adults. (27) Currently, renal volume as seen on ultrasonography cannot be considered a surrogate for nephron number.

Examining the effects of nephron mass on renal function is challenging; noninvasive estimates of GFR vary in accuracy. The gold standard plasma and urinary clearance studies are accurate but invasive, whereas other noninvasive estimates of GFR have been shown to be accurate at predicting a population mean but not individual GFR, (28) limiting their clinical usefulness at the bedside. Noninvasive calculated eGFR values are often used in clinical studies, because they are more acceptable to study participants. A number of alternative biomarkers to creatinine have been used to measure GFR, in particular CysC (29); however, again, the accuracy of this biomarker to estimate GFR is variable. The development of new real-time measured GFR using transdermal sensors that detect fluorescent filtration markers may change the way in which both clinicians and researchers are able to measure and study the effect of nephron mass on renal function, as well as the impact on both acute kidney disease and CKD. (30)

David Barker first observed that in utero events are associated with adult disease and described the “Barker hypothesis,” which underpins current theories around the developmental origins of health and disease. (31) Brenner et al hypothesized that this principle could be applied to kidney development, postulating that reduced nephron numbers could predispose a person to hypertension. (32) Glomerular hypertrophy is thought to occur in response to reduced nephron mass and glomerular hypertension (maintaining glomerular surface area), leading to hyperfiltration, salt sensitivity, sodium and water retention, hypertension, and glomerulosclerosis.

It has been demonstrated that glomerular surface area is similar between individuals with different kidney sizes, supporting the theory that glomerular hypertrophy occurs in oligonephronic kidneys. (33) Glomerular hypertrophy has been shown to be more prevalent in populations with a higher risk of CKD, including Australian aboriginal and black populations. (5) Glomerular hypertrophy has also been associated with poor post-transplantation outcomes. (5) Autopsy studies have shown a large variability in nephrogenesis in fetal life, thus supporting the hypothesis that the intrauterine environment influences kidney development. (12)

Birthweight has been demonstrated to be an important predictor of nephron and glomerular mass. (12)(21)(34)(35) Low birthweight (LBW <2.5 kg) has been associated with the development of end-stage renal failure, (36)(37) CKD, (19)(38)(39) and a more rapid deterioration in renal function in patients with underlying kidney disease. (5) Some infants are LBW as a result of prematurity, others are term but have experienced fetal growth restriction (FGR), and some are both growth restricted and preterm. Preterm birth is estimated to account for up to 80% of the LBW population. (40) FGR has various causes, including placental insufficiency and maternal smoking, which are often associated with an adverse intrauterine environment for a developing fetus. (41) The hypotheses underlying why LBW is associated with CKD vary depending on the etiology. In normally grown preterm infants, it is likely a result of an interruption in normal organ and vascular growth followed by ex utero nephrogenesis altered by the postnatal environment. In term infants born after a pregnancy complicated by placental insufficiency, the fetal kidneys may not have received adequate nutrition or oxygenation for nephron development. (42)(43) Many preterm infants are subjected to both the effects of being born during nephrogenesis and by an adverse intrauterine environment for organ development.

Maternal diet and nutrition during pregnancy have been shown to influence fetal nephrogenesis. (44) Micronutrient deficiencies, including vitamin A, iron, and folate, may affect human renal growth while macronutrient deficiency, in particular protein deficiency, has been shown to affect fetal renal growth in animal studies. (44)

Smoking is a risk factor for preterm birth. Cigarette exposure in pregnancy has a demonstrated dose effect, with the more cigarettes smoked per day increasing the risk of preterm birth. (45) Nicotine causes vasoconstriction and reduces placental blood flow. Maternal smoking during pregnancy has been associated with a smaller kidney volume and lower eGFR in school-age children (46) and a smaller kidney size in both fetuses and newborns. (47) A large retrospective cohort study in Japan found that maternal smoking was an independent risk factor for proteinuria at age 3 years in offspring. (48) In animal studies, ethanol use has been shown to reduce nephron number in offspring. (49) An Australian cohort study has demonstrated an association between maternal alcohol use in pregnancy and mild CKD in offspring during their 30s. (50)

Many preterm infants are exposed to in utero medications aimed at improving their survival and outcomes, a number of which may have an impact on nephrogenesis. Atosiban, which is an oxytocin receptor antagonist sometimes used as a tocolytic to delay preterm labor, may reduce renal cell growth, renal vasodilation, and the carbonic anhydrase activity. (51) Corticosteroids have been shown to reduce nephrogenesis in animal models and upregulate the expression of angiotensin II and its receptors. (51) Antibiotics (in particular aminoglycosides) and nonsteroidal anti-inflammatory drugs also cross the placenta and may affect fetal renal function. (52) Indomethacin is known to reduce amniotic fluid volume as a result of fetal renal dysfunction. Indomethacin in a neonatal rat model has been shown to cause glomerular injury and reduce glomerular number in adulthood, suggesting that indomethacin should be avoided if other less nephrotoxic medications are available. (53)

Significant physiologic changes occur after preterm birth, which lead to structural and functional changes in the developing kidney. The renal perfusion pressure of a fetal kidney is estimated to be around 3% of the cardiac output, which increases to 15% after delivery. (54) A sudden change in glomerular vascular resistance has been proposed as a possible mechanism for glomerular injury after preterm birth. (55)

Nephrogenesis and ureteric branching ordinarily occur in a hypoxic environment, with animal studies showing that mice undergoing nephrogenesis in a physiologically hypoxic environment had more ureteric branches and larger kidneys than those in a physiologically hyperoxic environment. (43) Conflicting results demonstrating worsened ureteric branching and smaller kidneys in a hypoxic environment have been seen in a similar mouse model, raising questions about whether kidneys respond to hyperoxia differently during different stages of nephrogenesis. (43) Another mouse model examining the effects of hyperoxia after preterm birth demonstrated a reduced glomerular count in pups with retinopathy of prematurity exposed to a hyperoxic environment, (56) again suggesting that hyperoxia impairs nephrogenesis. The arterial oxygen tension increases suddenly after preterm birth, altering the environment in which organogenesis occurs. Hyperoxia and hypoxia have both been shown to be deleterious to the outcomes of preterm infants and both should be avoided in the NICU whenever possible. (57)

Nephrotoxic medications are widely used in NICUs worldwide, (58)(59)(60) with the smallest and least mature infants often experiencing the largest exposures. In a retrospective review of infants born weighing less than 1,500 g, it was demonstrated that the majority of nephrotoxin exposure occurred in the first 40 days of age, which coincides with the period of ex utero nephrogenesis. (60) It has been shown that up to 87% of very-low-birthweight infants in NICUs are exposed to more than 1 nephrotoxic medication, most commonly gentamicin, indomethacin, or vancomycin, up to a total of 2 weeks in their hospital stay. (59)

Aminoglycoside antibiotics are widely used in the NICU, often as empiric therapy for suspected early- or late-onset infection. Aminoglycosides can accumulate in the kidney, leading to high concentrations in the renal cortex and ultimately leading to tubular injury. (61) Aminoglycoside-associated acute kidney injury (AKI) is generally felt to be less prevalent in newborns than in older children or adults; however, it can still lead to significant tubular damage and injury. (62) Nonsteroidal anti-inflammatory medications are often used in preterm infants to treat a patent ductus arteriosus, and indomethacin may be used as prophylaxis for intraventricular hemorrhage. Both act as prostaglandin inhibitors, which lead to vasoconstriction and reduced blood flow in renal and mesenteric vessels in addition to the intended vasoconstriction of the patent ductus arteriosus. (63) Animal studies have demonstrated that both ibuprofen and indomethacin reduce cyclooxygenase 2 and vasodilator prostanoids, with indomethacin having a more profound effect than ibuprofen. (64) This reduces renal blood flow and often leads to oliguria. Ibuprofen use in preterm infants has been shown to reduce GFR in the first month of age, (65) whereas indomethacin use has been shown to increase urinary podocyte and albumin concentration in the urine of preterm infants of less than 33 weeks’ GA compared with preterm and term controls, suggesting that it causes glomerular injury. (66) Although ibuprofen was found to cause glomerular injury in a neonatal rats, it did not result in reduced glomerular number in adulthood. (67)

A recent retrospective review of newborns weighing less than 1,500 g and of less than 30 weeks’ gestation found that the prevalence of AKI increased with decreasing GA, decreasing birthweight, and increasing nephrotoxic medication exposure. (60) Given that nephrotoxin administration is a potentially controllable risk factor for AKI, this has become the focus of a clinical practice improvement project: the Nephrotoxin Injury Negated by Just-in-time Action (NINJA). Implementing an initiative that triggers daily creatinine measures in noncritically ill children who are either receiving 3 simultaneous nephrotoxins or have received an aminoglycoside for more than 3 days has been shown to reduce AKI by 68%. (68) Trials of this approach need to be conducted in neonatal populations.

Preterm infants have a high incidence of AKI; extremely preterm infants of less than 29 weeks’ GA have been shown to have an incidence of 47.9% using the modified Kidney Disease: Improving Global Outcomes (KDIGO) criteria. (69) The preterm population of 29 to 36 weeks’ GA, in whom adverse long-term outcomes have been traditionally less concerning, has an incidence of AKI of 18.3%. (69) Although AKI has been independently associated with mortality in a preterm population (<1,200 g or <31 weeks’ gestation), (70) as previously mentioned, it is also associated with reduced nephron mass. (14)

Although some epidemiologic evidence exists to support an association between AKI and CKD in pediatric populations, (71)(72), more work needs to be done to show a causal link between pediatric or neonatal AKI and adult CKD. The IRENEO prospective cohort study has shown a reduced kidney volume but no difference in eGFR at a median age of 6.6 years in former preterm newborns (<33 weeks’ GA) who survived neonatal AKI compared with control subjects. (73) Although the investigators demonstrated no difference in eGFR between groups, the overall population had a high incidence of albuminuria and diminished eGFR. Similarly, the Follow-up of AKI in Neonates during Childhood Years (FANCY) study showed that extremely low-birthweight newborns who experienced AKI in the NICU had a more than 4-fold increased risk of renal dysfunction at a median age of 5 years compared with control subjects (74) and again demonstrated a high incidence of low eGFR of less than 90 mL/min per 1.73 m² (26%) in their entire cohort. When possible, efforts should be made to prevent AKI in preterm populations.

The evidence to guide clinicians about what constitutes optimal postnatal nutrition for ex utero nephrogenesis is limited. Animal models have shown that increasing growth through overfeeding postnatally may lead to an increase in nephron mass without any improvement in the risk for hypertension or renal dysfunction. (75) A rat model demonstrated that early postnatal overnutrition led to an increase in weight gain and 20% increase in nephron number in overfed offspring compared with normally fed offspring. (75) Despite their increased nephron number, overfed male rats were more likely to have hypertension, proteinuria, and glomerulosclerosis relative to controls. Neonatal high-protein diets have been shown to be associated with glomerular hypertrophy and glomerulosclerosis in an FGR rat model, (76) but to a lesser extent in an FGR piglet model. (77)

A single-center prospective cohort study demonstrated that postnatal growth failure is associated with a reduced GFR relative to controls in a human preterm population. (78) Ensuring optimal nutrition in this population is challenging because human evidence supports the idea that catch-up growth leading to obesity in preterm infants may worsen the progression of proteinuric kidney disease and that obesity and prematurity are additive factors. (79) A cohort study examining 153 former preterm infants at a median age of 11.5 years demonstrated that rapid weight gain in early childhood (>1 year of age) was associated with a higher body fat percentage and worsened metabolic markers of insulin resistance and hypertension relative to controls whereas rapid weight gain in infancy (<1 year of age) was not. (80)

Growing evidence demonstrates that prematurity has an impact on long-term renal health (Table). As large cohorts of preterm survivors continue to reach adulthood, more information is being gained about their long-term medical outcomes. Although a number of preterm cohorts have been followed up into early adulthood, we are yet to fully understand the effects of preterm birth on morbidity in mid-to-late adulthood.

Preterm infants are at an increased risk for calcium deposition in the kidney, known as “nephrocalcinosis.” This has a variable incidence in the literature and is thought to be associated with decreasing birthweight (81)(82); increased calcium, phosphorus, and ascorbic acid intake; and increased exposure to furosemide, dexamethasone, thiazides, and theophylline. (81) A matched cohort study in Greece examined 105 infants born before 36 weeks’ GA and found evidence of mild tubular dysfunction and a reduced kidney length in those with nephrocalcinosis relative to controls at 1 year of age. (83) Kist-van Holthe et al also demonstrated that former preterm children born at less than 32 weeks’ GA with nephrocalcinosis were more likely to have mild renal insufficiency and tubular dysfunction compared with control subjects at a mean age of 7.5 years. (84) These findings suggest that preterm infants with nephrocalcinosis may be at an increased risk of developing long-term renal dysfunction into adulthood; however, this has not yet been demonstrated in the literature.

As previously discussed, kidney size and volume are used as noninvasive markers of nephron mass. Preterm infants have smaller kidneys at term postmenstrual age than their term-born contemporaries; however, preterm infants have been shown to have larger kidney-to-bodyweight ratios, (13)(14)(15) potentially because of glomerular hypertrophy. A population of preterm infants has recently been shown to exhibit catch-up renal growth over the first 6 months of age. (85) Preterm infants born at 30 to 32 weeks’ GA were shown to have smaller kidneys at term postmenstrual age than term-born controls; however, they subsequently exhibited an increased renal growth rate, leading to similar kidney length at 6 months’ corrected age. (85) Their renal cortical regions grew more during this period of catch-up growth while their medullary regions remained smaller. The cause for this catch-up growth is uncertain; however, this suggests that while catch-up growth may occur, this is not necessarily associated with normal renal development.

A number of groups have examined former preterm infants’ kidney sizes in adulthood. The renal volumes of 20-year-old former premature female survivors (small for GA and appropriate for GA) have been shown to be smaller (both length and volume) compared with term controls. This association was seen irrespective of whether the study participants were small or appropriate for GA at birth. (86) More recently, the Health of Adults Born Preterm (HAPI) cohort study performed in Canada showed that adults born at less than 29 weeks’ GA had smaller kidneys as young adults (average age 23 years) compared with matched term-born controls. (6) Their renal volume corrected to body surface area was 10% lower than controls. There was no difference in eGFR between groups; however, young adults born preterm had higher albumin-to-creatinine ratios (albeit still within the normal range) relative to term-born controls, suggesting that they may have reduced glomerular endothelial integrity. (6)

Prematurity has been demonstrated to have an impact on kidney function. At 11 years of age, former preterm infants have been shown to have a reduced eGFR measured by urinary CysC and symmetric dimethylarginine compared with term controls. (87) A Polish cohort study has shown that in children born extremely preterm, eGFR measured using CysC was significantly higher in former preterm infants at both 7 and 11 years of age compared with matched term controls. Their eGFR worsened slightly over time, implying that long-term follow-up would be prudent. (88) Not all longitudinal studies have shown a reduction in eGFR during mid-childhood. Vieux et al’s cohort in France (27–31 weeks’ gestation evaluated at 5 years of age) all had normal eGFRs in follow-up. (89) As many former preterm cohorts continue to age, it is important to monitor their renal function longitudinally to gain a better understanding of how these early changes in kidney function and structure manifest themselves in mid-to-late adulthood.

A recent large Swedish cohort study demonstrated that extremely preterm infants (<28 weeks’ GA) have a 3-fold risk of developing CKD (odds ratio 3.01 [1.67–5.45]) and that late preterm infants have a risk that is almost twice that of term infants (odds ratio 1.84 [1.62–2.08]). (4) This association was strongest in childhood and persisted until midadulthood in all groups. There was a strong inverse association between GA and CKD, which included children born at what is typically considered full term at 37 and 38 weeks’ GA. This study relied on medical coding data to make a diagnosis of CKD, which means that the investigators could have under- or overestimated the true incidence of CKD in this population. Importantly, 24% of those who experienced acute renal failure in the neonatal period (27% of whom were preterm) were subsequently diagnosed with CKD.

Preterm birth is associated with an increased risk of hypertension. The underlying etiology of this association is not fully understood and is likely to be multifactorial. Prematurity and exposure to antenatal steroids have been associated with alterations in the function of the renin-angiotensin system. (90) A recent cohort study showed that angiotensin I levels are elevated in preterm infants relative to term infants and that changes in blood pressure in preterm infants occurred in association with changes in renin and angiotensin peptides, (6) which were not seen in term controls. Alamandine, a vasodilatory counterregulatory peptide moderated by the renin-angiotensin system, (91) has also been shown to be elevated in hypertensive preterm survivors but not in hypertensive term-born participants. It has been theorized that this represents a counterregulatory cardiovascular and renoprotective activation in preterm survivors.

A meta-analysis of observational studies found that preterm birth is associated with a 2.2 mm Hg increase in systolic blood pressure (SBP) compared with term controls. (92) These studies were from different countries and reported outcomes at varied ages until early adulthood. The incidence of hypertension was also increased in those born preterm. The reason for this association is not fully understood. Studies published since this systematic review have continued to demonstrate increases in both SBP and diastolic blood pressure (DBP) in those born preterm. Edstedt Bonamy et al showed that both SBP and DBP were higher (although not abnormal) in children born preterm compared with control subjects at 6 years of age. (93) An individual patient meta-analysis including 9 international cohort studies recently demonstrated a 3.4 mm Hg increase in SBP and a 2.1 mm Hg increase in DBP in adulthood in those born with very low birthweight compared with control subjects, (94) with a larger hypertensive effect noted in former preterm women than men and in those infants whose mothers had preeclampsia. A cross-sectional study of 5,232 young adult women in Sweden (mean age, 18 years) showed that those born preterm had an elevated SBP and mean blood pressure relative to those born at term. (95) A meta-analysis performed by Parkinson et al also showed an association between increased SBP and DBP affecting women more than men. (96) Some observational evidence shows that some preterm infants may exhibit salt sensitivity (ie, changes in blood pressure with a higher salt diet). (97)

Hypertension causes glomerulosclerosis and is a risk factor for CKD and progression to end-stage CKD. (98) In a population that begins life with fewer nephrons, ensuring that hypertension is both detected and treated appropriately is important. Evidence shows that preterm infants, particularly those born small for GA, are at an increased risk for insulin resistance and metabolic syndrome. (99) Diabetes is also strongly associated with end-stage CKD, (98) making it important to identify those with insulin resistance early to avoid nephron loss.

Interventions aimed at improving the long-term renal health of preterm infants need to be multipronged (Figure). The starting point of any preterm infant’s increased risk of mortality and morbidity is the point at which they are born early; any intervention aimed at reducing the risk of preterm birth is therefore of huge potential benefit. A large proportion of preterm birth remains unexplained; however, there are interventions that have reduced the incidence of preterm birth in specific high-risk groups, the details of which are beyond the scope of this review. (100) Given that many preterm infants experience the “double hit” of both preterm birth and growth restriction, interventions that reduce risk factors for FGR are also of benefit; this includes smoking cessation, advice during (and ideally before) pregnancy with regard to smoking, drug use, and good maternal nutrition both around the time of conception and throughout pregnancy. The first 1,000 days is an international public health initiative aimed at optimizing the nutrition of infants from conception to their second birthday to reduce their risk of programmed noncommunicable diseases. (101) Other obstetric interventions such as screening for the risk of preterm preeclampsia and providing aspirin to those who qualify could potentially reduce iatrogenic preterm birth. (102)

Once preterm birth has occurred, attention should be given to reducing renal injury during ex utero nephrogenesis. This includes reducing the risk of sepsis in the NICU, which is associated with AKI, avoiding hyperoxia, and reducing exposure to nephrotoxins. The evidence around what constitutes optimal nutrition for nephrogenesis is limited; however, human milk feeding (either expressed maternal milk or banked breast milk) has been associated with a reduction in hypertension in adolescents born preterm in nonrandomized studies. (103)

Ideas for novel drug and stem cell therapies aimed at prolonging nephrogenesis in preterm infants are being discussed and explored in the literature (104)(105)(106); however, at this stage (to our knowledge) these have not progressed to preclinical trials.

Given the increased risk that preterm survivors have for hypertension and decreased insulin sensitivity, this population should be followed throughout childhood and adulthood to ensure that they receive appropriate diagnosis and treatment. Health-care professionals and people born preterm should receive education and advice regarding their increased risk for CKD and the importance of lifestyle interventions.

Many gaps remain in our understanding of the effects of preterm birth on long-term renal health. As preterm cohorts are followed up into mid-to-late adulthood, we will gain further knowledge of their risk profiles and renal complications, which will continue to help guide future research, intervention, and therapies. As new technology such as real-time measured transcutaneous GFR becomes available, we may be able to better identify and target our highest risk populations and ensure that they receive appropriate follow-up.

Preterm birth leads to a reduced functional nephron mass and to maladaptation of the kidney. This in turn predisposes preterm survivors to reaching the threshold of glomerulosclerosis at which renal function declines earlier in life. There is still a lot to learn about the effects of prematurity on the kidney and how to best support the kidneys of preterm survivors, which should remain a focus for ongoing research. Currently many opportunities exist for clinicians to avoid further contributing to nephron loss in preterm infants. These include avoiding neonatal AKI and nephrotoxin exposure during nephrogenesis; educating preterm survivors and their families about lifestyle risks that may contribute to nephron loss; and educating primary health care professionals about the importance of reducing obesity, screening for hypertension and insulin resistance, and avoiding nephrotoxins in former preterm children and adults.

AKI

acute kidney injury

CKD

chronic kidney disease

CysC

cystatin C

DBP

diastolic blood pressure

eGFR

estimated glomerular filtration rate

FGR

fetal growth restriction

GA

gestational age

GFR

glomerular filtration rate

LBW

low birthweight

SBP

systolic blood pressure