Hypertension

The greatest body of evidence to support an association between preterm birth and cardiometabolic outcomes is with regards to blood pressure (BP) in young adulthood. A large population-based study using antihypertensive prescriptions as a surrogate marker for hypertension found that in over 600 000 subjects for whom birth data were available, those born before 37 weeks’ gestation were more likely to be on medication and this increased with earlier gestational age.5 These results are supported by a recent large meta-analysis identifying higher BP in association with prematurity.6 A population-based study showing that the risk of elevated systolic BP increases with decreasing gestational age found that being small for gestational age (SGA) contributes to risk only in those subjects born after 33 weeks’ gestation,7 and several further cohort studies suggest that the risk of hypertension in individuals born preterm may indeed be independent of birth weight.8–10 Although this may be because moderately preterm individuals represent a large enough sample for the effect of growth restriction to become apparent in these studies, an alternative suggestion is that different mechanisms lead to an increase in the risk of hypertension as a consequence of intrauterine growth retardation and prematurity. Although the increase in BP in young adults born preterm may be small (in the region of 3–4 mm Hg),6 lowering diastolic BP by as little as 2 mm Hg can significantly reduce the risk of myocardial infarction and stroke,11 suggesting that there could be public health benefits of BP surveillance in the expanding ex-preterm adult population.

Vascular and cardiac health

Vascular endothelial dysfunction is associated with hypertension, hyperglycaemia and dyslipidaemia,12 and a number of studies have sought to determine whether there are long-term consequences of preterm birth on the vasculature (reviewed in ref. 13). Although some studies suggest that low birth weight associates with endothelial dysfunction, the evidence is not consistent.13–15 Increased intima-media thickness (IMT) is an early sign of atheroma formation, and cardiovascular risk factors in childhood including dyslipidaemia, hypertension and increased body mass index (BMI) are associated with increased IMT in adult life.16 Several studies show increased IMT in individuals born preterm,6 and in ex-preterm 19 year olds, increased carotid IMT is strongly associated with an unfavourable lipid profile and current body size.17 Impaired arterial elasticity is a further recognised risk factor for cardiovascular disease and some studies show a difference in arterial stiffness in association with prematurity, although in the immediate newborn period,18 where growth restriction may be the main contributory factor,19 while others report no increase in ex-preterm children and adolescents.20 Importantly, some of these effects may be dependent on the presence of fetal growth restriction, for example, in the Cardiovascular Risk in Young Finns Study, preterm infants had increased IMT and decreased flow-mediated dilatation as adults only if they had fetal growth restriction.21 Finally, data from MRI studies of preterm-born young adults suggest that prematurity is associated with altered cardiac structure and function, with increased left and right ventricular mass, a smaller right ventricle and impaired systolic and diastolic functional parameters.22 ,23

Insulin resistance/type 2 diabetes

The association between preterm birth and later insulin sensitivity has been the subject of a recent systematic review.4 In childhood, a number of studies suggest that preterm birth associates with reduced insulin sensitivity,4 ,24 ,25 with some studies suggesting that preterm-born infants may be up to 50% less insulin sensitive than term-born children.24 Although the extent to which insulin resistance relates to altered adiposity during childhood is unclear,26 ,27 a recent prospective study in 1358 children showing an inverse association between gestational age and elevated plasma insulin levels at birth and in early childhood suggests that insulin resistance may be present from early life.28 There is conflicting evidence for an association between prematurity and insulin resistance in adulthood, with some studies showing a 30%–40% reduction in insulin sensitivity in ex-preterm adults8 ,29 ,30 and two large retrospective cohort studies showing a weak association between prematurity and type 2 diabetes,31 ,32 while others report no persisting relationship.6 ,33 ,34 Any relationship between preterm birth and subsequent insulin resistance is likely to be multifactorial and dependent on risk factors such as excessive early weight gain, the presence of fetal growth restriction and altered adiposity. Nevertheless, since insulin resistance and type 2 diabetes risk increase with age, small differences in glucose homeostasis in young adulthood could translate into clinically significant outcomes in later life.

Altered adiposity

Increased total body and intra-abdominal fat are recognised important components of the metabolic syndrome in adults. A recent systematic review concluded that by term corrected age, preterm babies have a 3% greater percentage fat mass than those born at term, which is largely explained by a deficit in lean body mass.35 Furthermore, MRI studies have shown that preterm babies have more intra-abdominal36 and intrahepatocellular fat37 compared with term babies. Individuals born preterm have a lower BMI compared with those born at term throughout infancy, childhood and adolescence; however, the pattern of weight gain differs, with BMI z-scores increasing between childhood and adolescence in the extremely low birth weight group and remaining stable in term-born individuals.38 Whether these differences persist in adults born preterm is not clear. Altered adipose tissue partitioning, with a threefold increase in intrahepatocellular lipid content, as well as increased total body fat, has been reported in young adulthood;39 however, another recent systematic review concluded there were no significant differences in BMI between preterm and term-born adults.6 Although some studies suggest that BMI is strongly associated with measures of adiposity in children,40 it may not reliably distinguish between growth in lean and fat mass and beneficial changes in body composition, and vascular function can occur with exercise without any change in BMI41 so that other/additional measures of body composition may be more accurate. Preterm individuals who gain excessive weight up to 3 months of age have higher percentage body fat and waist circumference at age 21 years, with the fastest growing group also having higher cholesterol and low-density lipoprotein.42 Thus, the pattern of growth and fat acquisition in both the short and long term may differ between individuals born preterm and term which could contribute to the increased cardiometabolic disease risk.

Catch-up growth and nutrition

Catch-up growth, characterised by a height velocity above the limits of normal for age for at least 1 year after a transient period of growth inhibition, may lead to an increased risk of cardiometabolic disease.43 As in term infants, studies of the preterm population suggest that early growth patterns, in particular upwards centile crossing during childhood, are implicated in the development of higher systolic BP and insulin resistance.9 ,26 ,27 The timing of weight gain may be of particular importance; a study in adolescents born preterm and randomised to a nutrient-enriched or lower nutrient diet and a reference group born at term, early rapid weight gain—in the first weeks of life—was associated with an increased risk of insulin resistance.44 The nutritional management of preterm infants aims to ‘achieve growth similar to foetal growth coupled with satisfactory functional development’;45 however, the optimal targets for growth in preterm infants are unclear.46 Additionally, while there are recommendations for nutritional intake in infants up to 1800 g, a high degree of uncertainty remains45 and there are fewer data for optimal intake in moderately preterm infants who are also at risk of long-term complications.47 While some studies have suggested that feeding preterm infants with unsupplemented breast milk or standard formula may adversely affect growth and later cognitive function,48 ,49 the use of nutrient-enriched formulas has been associated with higher childhood BP50 and with insulin resistance in adolescence, which was greatest in those with rapid postnatal weight gain.44 Further, more recent studies in two independent cohorts of very preterm infants suggest that breast feeding at the time of discharge is associated with better neurodevelopment despite poorer initial weight gain.51 Clearly, this is an important area for ongoing research.

Altered hypothalamic-pituitary-adrenal axis activity

Preterm infants have altered plasma cortisol levels in the first two years52 and there is a strong positive correlation with salivary cortisol response to pain at 4 months of age, suggesting an effect on hypothalamic-pituitary-adrenal (HPA) axis functioning.53 This is important since altered HPA axis activity in adulthood is associated with cardiovascular risk factors including central adiposity, glucose intolerance, hypertension and dyslipidaemia (reviewed in ref. 54), so that programmed alterations in HPA axis activity may be one mechanism linking preterm birth with later cardiometabolic disease risk.

Extensive data from animal studies show that fetal overexposure to excess glucocorticoids has long-term effects on the HPA axis and cardiometabolic sequelae (reviewed in ref. 55) and preterm babies are at increased risk of perinatal exposure to excess glucocorticoids for a number of reasons. Preterm babies are potentially exposed to a highly stressful environment both in utero and ex utero, with concomitant increased endogenous glucocorticoid release.53 While the administration of antenatal glucocorticoids to women with threatened preterm labour has clear benefits in the neonatal period, there are concerns over potential longer-term adverse effects, particularly following multiple doses. There is some evidence for long-term adverse effects of early glucocorticoid overexposure in humans: studies in preterm infants exposed to exogenous glucocorticoids in utero have reported evidence of reduced birth weight (with repeated dosing),56 increased behavioural disorders,57 higher systolic (mean difference 4.1 mm Hg) and diastolic BP (mean difference 2.8 mm Hg) in adolescence58 and increased insulin resistance in early adulthood.59 However, other studies report no adverse effects of antenatal steroid use on growth and neurodevelopment in young children.60 Since many children exposed to antenatal glucocorticoids are born preterm, it has been difficult to differentiate the effects of prematurity from those of glucocorticoid overexposure per se. While a large systematic review found no persisting effects of antenatal glucocorticoid administration on HPA axis activity at least until infancy,61 a further study did find long-term effects in infants who received glucocorticoids for fetal lung maturation but who were born at term.62

Potential mechanisms

Although epidemiological studies inform us of associations between the early life environment and later disease risk, the mechanisms remain unclear. Since disease risk varies between individuals, understanding of the mechanisms would potentially facilitate the development of predictive markers of later disease risk in this group of vulnerable infants.

Telomeres are regions of repetitive nucleotides at the ends of chromatids that function to protect genomic stability by preventing deterioration or fusion of chromosomes. Telomere length varies among individuals at birth and shortening occurs with each mitotic cycle and secondary to environmental factors throughout life such as oxidative stress.63 Importantly, telomere shortening is associated with cell senescence, tissue atrophy and an increased risk of age-associated disease.63 Programmed effects on telomere attrition have been proposed as one mechanism linking early life with later disease risk.64 Telomere shortening occurs rapidly in young children,65 and in preterm babies this occurs more rapidly than unborn fetuses of comparable age;66 however, it is not known whether this results in long-term differences in telomere length in babies born preterm.

There has been much recent interest in the role of changes in the epigenome in mediating the association between early life factors and later disease risk.67 Altered DNA methylation has been in adulthood in term-born individuals in association with exposure to an adverse prenatal nutritional environment,67 ,68 and changes in DNA methylation have been reported in ex-preterm individuals in adulthood.69 Whether these differences are identifiable at birth is unclear; a recent epigenome-wide study using DNA extracted from stored neonatal blood spots and blood from the same individuals at age 18 years identified differences in DNA methylation profiles between preterm and term individuals at birth but found that these largely resolved by adulthood, presumably reflecting immaturity in the preterm group.70 Nevertheless, some loci were identified which showed differential methylation at birth and at age 18, suggesting that such changes might potentially be useful as biomarkers of disease risk.70

Implications for public health

Although there is growing evidence that preterm birth is associated with an increased risk of non-communicable diseases (NCDs), the degree of risk is difficult to quantify. NCDs represent a considerable economic burden, and the predicted increase in the prevalence of NCDs during the next two decades will particularly affect countries in the developing world, as their economies and populations grow (United Nations General Assembly, 2011). Although the major focuses for the prevention of NCDs are adult lifestyle risk factors, those working in the DOHaD field continue to advocate for increased recognition of the role of the early life environment in mediating the risk of NCDs.71 In 2010, 92% of the 14.9 million births before 37 weeks of gestation occurred in low-income and middle-income countries and the rates of preterm birth continue to increase in most countries with reliable data.72 The rapid development of perinatal care in many middle-income countries is likely to result in a substantial increase in the numbers of preterm babies surviving through to adulthood.72 Thus, the predicted increase in survival and the associated increased risk of NCDs in those born preterm, coupled with the increasing rates of preterm birth, could place significant additional burdens on healthcare costs, particularly in low-to-middle-income countries. Indeed, given that many of the cardiovascular risk factors emerge with ageing and the dramatic increase in survival of preterm infants has occurred mainly within the last quarter of the 20th century, the magnitude of any effect may only become apparent in future years.