Birth weight and hypertension: nature or nurture?

Abstract

For more than 30 years, observational studies have linked early life programming and low birth weight to blood pressure elevation and established hypertension. It all started in 1988 with studies from Sweden by Gensser [1] and the UK by Barker et al.[2], and later on, this relationship has been confirmed in the majority of studies, based on cohorts [3,4] or register linkages [5]. Low birth weight, prematurity and impaired foetal growth with small-for-gestational age babies, some of them experiencing rapid postnatal catch-up growth, are phenotypes all linked to blood pressure elevation, as well as a number of other cardiovascular risk factors and even increased risk for cardiovascular events [6–8]. However, also, U-shaped relationships have been documented for the association between both low and high birth weight with risk of type 2 diabetes [9]. A still unsettled questions is to what degree these associations reflect nature or nurture, that is the influence of genes or the environment. In favour of genetic explanations are reports that state that genes regulating maternal hypertension could also influence blood pressure elevation in the offspring when low birth weight is just a side phenomenon [10], and that genes associated with low birth weight also increase the risk of coronary artery disease and other disease conditions [11]. However, on the contrary, we know that environmental manipulations of maternal diet and living conditions during pregnancy in laboratory animals could influence offspring haemodynamics and blood pressure [12]. This corresponds to historical events with periods of war [13–15], famine [16] or stress caused by civil unrest [17] that could influence birth weight, as well as blood pressure elevation and cardiovascular risk in surviving children during long-term follow-up. This means that it is likely that environmental influences could interact with epigenetic mechanisms to programme organ development and function, including the cardiovascular system [18]. In this issue of the Journal, a group of authors report on findings from the STANISLAS cohort covering two generations of related individuals (n = 1028) in north-eastern France [19]. This study includes information on self-reported birth weight and screening in adult life for blood pressure, renal function, pulse wave velocity (PWV) and measures of target organ function (TOD) related to the heart and carotid artery. The finding in general was an inverse association between birth weight and blood pressure elevation, or hypertension in adult life, but less strong evidence for associations between birth weight and TOD, only found in some subgroups. No association was found with renal function using the CKD-EPI algorithm based on creatinine. The authors suggest that hypertension comes first, which in turn promotes vascular damage (increased atherosclerosis and arterial stiffness). Even if self-report of birth weight is highly correlated with register-based information on birth weight (r = 0.9) according to a systematic review and meta-analysis [20], the lack of information on pregnancy duration (i.e. gestational age) in the STANISLAS study is a shortcoming. In subgroups, self-report of prematurity and breastfeeding was available, but not for all, as is often the case in population-based studies when people are asked for such information based on recollection. The standard definition of low birth weight (LBW; <2500 g) and prematurity (<37 weeks of gestational age) in western populations is one way to use categories, standards in clinical medicine, but in epidemiology, one should more focus on continuous variables over a wider range. The problem is that some of these relationships are not linear, in itself a rare phenomenon in biology, but rather curvilinear or even nonlinear, for example the U-shaped association of birth weight with adult obesity and risk of type 2 diabetes [9]. No association was found between birth weight and PWV [19], a marker of arterial stiffness and vascular ageing [6]. In other studies, such an association was found, as in one study examining adolescents from Austria [21], but neither in 6-year-old children in Finland [22] nor in very premature children from the UK during follow-up [23]. This means that such associations with PWV are either lacking in most cases, or depending on age range and phenotype at birth. On the contrary, increased Augmentation Index (Aix) has repeatedly been shown to be inversely associated with birth weight and prematurity [24]. Aix is a far more complex variable than PWV, as it is influenced by a number of contributing factors, not only aortic stiffness, but also total peripheral resistance TPR, blood pressure, the aortic reflex wave and cardiac function [25]. In a study of very premature babies followed until age of 18 years, there was an association between all these three variables: prematurity, Aix and TPR [23]. In the STANISLAS study, no association was found between birth weight and renal function. However, other studies have found such an association if using other algorithms for renal function or different populations. For example, a study from Sweden used cystatin-C instead of creatinine for estimating renal function and could show an association [26]. It is also known that LBW is associated with a reduced number of nephrons, first proposed by the nephrologist Brenner [27]. If so, in early stages of renal dysfunction, there could well be compensatory hyperfiltration and thus a normal, or even supernormal estimated renal function, but later on replaced by impaired renal function. Another measure is albuminuria or the albumin/creatinine ratio, but no association was found in the French study. In summary, the STANISLAS study is of interest, as it could show an (expected) inverse association with blood pressure (even 24-h ABPM) and hypertension, but also some degree of heritability of birth weight (42–44%) across two generations. Associations with measures of TOD were either lacking (eGFR, PWV, diastolic dysfunction) or present only in subgroups, for example a higher left ventricular mass index (LVMI) in participants with a birth weight higher than 3 kg. The lack of information on Aix is regretful, as it could have shed more light on the association between birth weight and arterial function. We now have a wealth of data from observational, epidemiological studies, but more information is needed about mediating mechanisms between early life factors and adult outcomes (traits, events) within the so-called Developmental Origin of Health and Disease (DOHaD) concept [28]. Finally, also, intervention studies are needed to prove the importance of preventive maternal and child healthcare for these measures and outcomes. This was recently shown in a randomized study from India when such preventive efforts during preconception and pregnancy improved not only birth outcomes but also maternal health [29]. Thus, cardiovascular prevention should start early in life, even before birth. The health and lifestyle of pregnant women is of great importance, not only for the women themselves but also for their children, thereby linking reproductive health with family health and forming a basis for normal development of blood pressure and vascular function in offspring [30]. It is a truism that we all have been children and that we all undergo ageing, including vascular ageing, but this simple fact of life should inspire for more focus on early life programming of adult health, prevention and research [31], especially on mechanistic links – targets of interventions. ACKNOWLEDGEMENTS This editorial was supported by a grant from the Swedish Research Council (2013-2756). Conflicts of interest There are no conflicts of interest.