Low Birthweight, Body Composition and Cardiometabolic Risk: Nature Versus Nurture?

It has long been recognised that being born at low birthweight carries a risk of metabolic syndrome and later coronary artery disease, stroke and type 2 diabetes [1]. Obesity, in particular abdominal obesity, is strongly related to insulin resistance and metabolic syndrome [1]. The associations between low birthweight and metabolic syndrome are best established for individuals who were born full-term. However, in the absence of congenital malformations or chromosomal anomalies, low birthweight is the result of early birth (preterm) or fetal growth restriction (or both). There is mounting evidence that the metabolic risk factors differ between individuals born preterm and those born full-term and low birthweight [2]. In this issue of Paediatric and Perinatal Epidemiology, Sandboge and colleagues [3] investigated a cohort of 77 adults born at very low birthweight (VLBW, < 1500 g) and 70 term-born sibling controls, sex-matched and within 10 years of the VLBW participant. The VLBW cohort was recruited from three sources: the Helsinki Study of Very Low Birth Weight Adults, the ESTER Preterm Birth Study and the Finnish Medical Birth Registry. Birth years ranged from 1985 until 1989, which reflected an era of neonatal intensive care that predated significant advancements in perinatal care, including surfactants to treat respiratory distress syndrome. This is particularly relevant as VLBW cohorts born in the 1980s included more mature but growth-restricted individuals because the survival of more immature babies was much lower than in the ‘post surfactant’ era [4]. The participants in Sandboge et al.'s study were assessed at a mean age of 29–30 years and underwent adiposity measurement using dual-energy X-ray absorptiometry (DEXA). There were group differences in lean body mass, limb/trunk fat mass ratio, appendicular skeletal mass and its ratio to body mass index (BMI): measures that, respectively, reflected shorter stature, more centralised fat distribution and lower relative appendicular muscle mass in the VLBW group compared with controls. In a further exploratory analysis, the differences were restricted to the VLBW and small for gestational age subgroup, suggesting that fetal growth restriction rather than gestational age may have been driving most of the group differences. The authors highlighted that by using matched sibling controls, their study design accounted for familial shared genetic or environmental influences. How do these findings compare with other studies? A systematic review and meta-analysis published in 2019 reported higher fat mass, blood pressure and insulin resistance, but similar BMI, waist circumference or waist–hip circumference in adults born preterm compared with those born at term [2]. Since then, an Australian geographical cohort of young adults born extremely preterm or extremely LBW and term controls reported lower total body lean mass but higher visceral fat volume measured using DEXA in the preterm group compared with controls at age 25 years [5]. Flahault et al. [6] found no group differences between Canadian adults aged 18–29 years born preterm (≤ 29 weeks' gestation) and healthy full-term controls. In that study, rates of adiposity were 31% in both groups as measured using BMI, waist circumference or waist–hip ratio based on World Health Organization expert consultation cut-offs [6]. One of the challenges of comparing studies is using different methods to ascertain adiposity and the comparability of these measures. Body composition assessment is an essential component of measuring obesity and nutritional status [7]. Body composition is usually calculated using a two-compartment model that separates the body into fat and fat-free components. The commonly used methods in clinical and research settings include DEXA, bioelectrical impedance analysis (BIA), measurement of skinfolds and sophisticated imaging such as computed tomography or magnetic resonance imaging. However, indirect measures of body composition are also widely reported. These include measures of anthropometry such as BMI, waist and hip circumferences. However, these proxies for body composition have important limitations; BMI does not account for fat and fat-free components, nor does it reflect the pattern of fat distribution. As such, BMI may lead to misclassification of overweight/obesity [7]. Meanwhile, waist and hip circumferences, including waist–hip ratio and waist–height ratio, may better reflect excess adiposity and regional distribution of adipose tissue. So, how does Sandboge et al.'s study [3] add to the current evidence? Differences in body composition were consistent with previous studies that did not use matched sibling controls, indicating that VLBW status alone is likely an important risk factor for suboptimal body composition in young adulthood. As the differences were restricted to the VLBW and small for gestational age subgroup (rather than appropriate for gestational age subgroup), this exploratory finding further supports that preterm birth may have distinct associations with body composition and later cardiometabolic risk. Interestingly, a previous study by the same group, using the same cohort, showed no group differences in the distribution of abdominal adipose tissue, albeit measured with a different method, for example, magnetic resonance imaging [8]. That finding was in contrast with previous reports by others on smaller cohorts. The authors speculated that a possible explanation for their different findings could have been due to familial confounding. It would have benefited the authors to integrate findings from both studies to provide a more comprehensive picture of adiposity. Many questions remain unaddressed, however. With the increasing survival of infants born very or extremely preterm, it is critical to determine the risk of preterm birth with the components of metabolic syndrome and later cardiovascular health. Existing evidence suggests that preterm birth is likely to have different risk relationships and potential pathways to metabolic syndrome and cardiovascular risk than low birthweight. Study designs that recruit inception cohorts based on gestational age rather than birthweight alone would be ideal, as this approach circumvents the inevitable inclusion of a higher proportion of individuals with intrauterine growth restriction in cohorts defined by birthweight status. We also need data from more contemporary cohorts. Much of the existing data, by necessity, draws from older cohorts born before major advances in neonatal intensive care, including exogenous surfactants to treat respiratory distress syndrome, antenatal corticosteroids for pregnant women at risk of preterm birth and improved nutritional practices for preterm newborns. These older cohorts may not reflect the risk of more contemporary cohorts that have benefited from perinatal and neonatal care advances. Further, data on the trajectories of adiposity with increasing age are scarce. BMI trajectories continue to increase with age until the mid-30s, with studies showing a disparity in change over time between those born extremely preterm and term [9, 10]. One study has shown that this disparity is due to a more rapid increase in z-scores for weight and BMI but not in height z-scores [9]. Reporting group differences at one-time point may give an erroneous reflection of how the preterm or low birthweight differs from the term group. More studies are needed to understand the dynamic nature of BMI and body composition trajectories during early adulthood and where interventions may be most effective. Finally, does altered body composition indeed result in poorer cardiometabolic health? This remains to be seen and points to the importance of longer-term follow-up to determine if the increased risk of metabolic syndrome with low birthweight does result in later disease. Supporting the follow-up of such cohorts is paramount. The author takes full responsibility for this article. The author declares no conflicts of interest. The author has nothing to report.