Obesity is Associated with Higher Blood Pressure and Higher Levels of Angiotensin II but Lower Angiotensin-(1-7) in Adolescents Born Preterm

Abstract

Objectives

To evaluate if obesity is associated with increased angiotensin II (Ang II) and decreased angiotensin-(1–7) [Ang-(1–7)] in the circulation and urine among adolescents born prematurely.

Study design

In a cross-sectional analysis of 175 14-year-olds born preterm with very low birth weight, we quantified plasma and urinary Ang II and Ang-(1–7) and compared their levels between subjects with overweight/obesity (body mass index ≥85^th percentile, n = 61) and those with body mass index <85^th percentile (n = 114) using generalized linear models, adjusted for race and antenatal corticosteroid exposure.

Results

Overweight/obesity was associated with higher systolic blood pressure and a greater proportion with high blood pressure. After adjustment for confounders, overweight/obesity was associated with an elevated ratio of plasma Ang II to Ang-(1–7) (β: 0.57, 95% CI 0.23 to 0.91) and higher Ang II (β: 0.21 pmol/l, 95% CI 0.03 to 0.39) but lower Ang-(1–7) (β: −0.37 pmol/l, 95% CI −0.7 to −0.04). Overweight/obesity was associated with a higher ratio of urinary Ang II to Ang-(1–7) (β: 0.21, 95% CI −0.02 to 0.44), an effect that approached statistical significance.

Conclusions

Among preterm-born adolescents, overweight/obesity was associated with increased Ang II but reduced Ang-(1–7) in the circulation and the kidney as well as higher blood pressure. Obesity may compound the increased risk of hypertension and cardiovascular disease in individuals born prematurely by further augmenting the prematurity-associated imbalance in the renin-angiotensin system.

The survival of infants born preterm has increased due to advances in prenatal and neonatal care, but premature birth and other early life events increase the lifetime risk for chronic health conditions such as hypertension and cardiovascular disease.^1–3 The mechanisms behind this elevated risk are described incompletely but may include programmed alterations in various hormonal pathways. Perinatal events have been shown to induce changes in the renin-angiotensin system (RAS), a critical regulator of blood pressure and overall cardiovascular function⁴. In addition to the traditional angiotensin-converting enzyme (ACE) / angiotensin II (Ang II) / Ang II type 1 receptor pathway, the RAS also consists of the regulatory ACE2 / Ang-(1–7) / Mas receptor pathway which acts in part to counteract Ang II’s actions by promoting vasodilation and sodium excretion as well as inhibiting inflammation and fibrosis^{5, 6}. Alterations in the RAS that promote Ang II and suppress Ang-(1–7) are now recognized to be linked to hypertension^{7, 8}.

Obesity is also associated with increased Ang II expression which may lead to higher blood pressure but its effect in individuals born prematurely is unknown⁹. Race and antenatal corticosteroid exposure are associated with alterations in the RAS and may affect obesity’s influence^{10, 11}. We recently demonstrated that adolescents born preterm with very low birth weight have higher blood pressure and a higher ratio of plasma Ang II relative to reduced Ang-(1–7) as compared with term-born peers, an association that is stronger in subjects with overweight/obesity as compared with subjects without overweight/obesity¹². Therefore, we further evaluated the effect of obesity on the RAS among adolescents born preterm. We hypothesized that obesity is associated with higher Ang II and lower Ang-(1–7) in the circulation and kidney of adolescents born preterm compared with those without obesity.

Methods

Subjects were recruited from a prospective birth cohort of 193 patients born prematurely with very low birth weight (birth weight <1500 grams) between January 1, 1992 and June 30, 1996 at a regional perinatal center (Forsyth Medical Center in Winston Salem, NC). As part of a broader study of cardiovascular outcomes in adolescents born prematurely, we evaluated the cohort at 14 years of age (Figure 1; available at www.jpeds.com). Inclusion criteria were singleton birth, birth weight < 1500 grams, follow-up clinical information through at least 1 year of age corrected for gestation, and successful contact at age 14 years. Exclusion criteria were major congenital anomalies or genetic syndromes and being a ward of the state. Subjects were assessed at three study visits; we report measurements conducted at the third study visit. The Institutional Review Boards of Wake Forest School of Medicine and Forsyth Medical Center approved the study. Written parental or legal guardian informed consent and subject assent were obtained. Subjects were compensated upon completion of all study visits, and parents/guardians received compensation for travel expenses.

Figure 1;

online only Study cohort

A research nurse reviewed medical records and research databases to obtain maternal and subjects’ birth characteristics, including antenatal corticosteroid exposure, maternal hypertensive pregnancy, maternal smoking during pregnancy, cesarean delivery, sex, gestational age, and birth weight. Subjects were categorized as small-for-gestational age if their birth weight was less than the 10^th percentile for gestational age.¹³ At the 14-year visit, demographic data were recorded including parental-reported race (black or non-black), current Medicaid use, and patient smoking status. Measurements of the subjects’ height, weight, and waist circumference were obtained, and the waist-to-height ratio and the body mass index (BMI) were calculated. Subjects were categorized as having overweight/obesity if their BMI was ≥85^th percentile for age and sex according to established pediatric recommendations¹⁴. Subjects rated their sexual maturity in private using a self-reported questionnaire on a scale of 1–5, and we reported the percentage of subjects with a score of 5 in either of the two determinants of secondary sexual characteristics (pubic hair in all subjects plus breast development in females and external genitalia development in males)¹⁵.

Blood pressure was measured according to established pediatric guidelines by trained staff using a mercury manometer and an appropriately sized blood pressure cuff¹⁶. Each subject was seated quietly for a minimum of five minutes with their arm fully supported. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were recorded as the average of three measurements (taken 1 minute apart), and blood pressure z-scores were calculated according to age, sex, and height-specific values¹⁷. According to pediatric recommendations, we defined high blood pressure as SBP or DBP ≥ 120/80 mmHg and further classified the stages of hypertension as i) elevated blood pressure if SBP was 120 to 129 mmHg but DBP was < 80 mmHg; ii) stage 1 hypertension if SBP was 130 to 139 mmHg or DBP was 80 to 89 mmHg; and iii) stage 2 hypertension if SBP was ≥ 140 mmHg or DBP was ≥ 90 mmHg.

Detailed descriptions of RAS measurements have been extensively described in previous publications¹². Briefly, blood samples were collected with subjects seated for measurement of plasma renin activity, aldosterone, Ang II, and Ang-(1–7). Peptide samples were immediately collected in a tube containing a cocktail of inhibitors. Spot urine samples were immediately collected in a tube with HCl to prevent peptide degradation. Ang II, Ang-(1–7), plasma renin activity, and aldosterone were analyzed using radioimmunoassays. ACE and ACE2 were assayed in non-acidified urine samples, and ACE and ACE2 protein content were based on human standards (R&D Systems, Minneapolis, Minnesota, U.S.A.). Urine creatinine levels were determined by a modified Jaffe assay traceable to isotope dilution mass spectrometry. We calculated the ratio of Ang II to Ang-(1–7) in the plasma and urine, calculated the ratio of ACE to ACE2 in the urine, and corrected the urinary concentrations of Ang II, Ang-(1–7), ACE, and ACE2 for urine creatinine. For sample results that were below the laboratory’s lower limit of detection, the sample’s measurement was assigned a value calculated as the lower limit of detection divided by the square root of two¹⁸.

Statistical Analyses

We used descriptive statistics including measures of central tendency and dispersion and frequencies. We reported continuous variable distributions as mean with standard deviation or median with interquartile range. We used Chi-square test, Fisher exact test, t-test, and Wilcoxon rank-sum test for between-group comparisons and Pearson or Spearman correlation coefficients to test for correlations between continuous variables. A 2-sided alpha level less than 0.05 was considered statistically significant.

We utilized generalized linear models to estimate the relationship between overweight/obesity and the RAS outcomes. Distributional characteristics of the outcomes were improved with natural logarithmic transformation in the models. We employed directed acyclic graphs to determine potentially confounding factors based on the literature a priori and identified race and antenatal corticosteroid exposure in our minimally sufficient set of confounders for inclusion in the models^{10, 11, 19}. We used Enterprise Guide software, Version 7.11 of the SAS System for Windows (SAS Institute Inc, Cary, NC) for all analyses.

Results

Within the cohort of 193 enrolled subjects, 5 were excluded (3 had significant medical problems and 2 were twins). One hundred and seventy-seven subjects participated in the third study visit; 175 subjects provided urine samples and 121 subjects provided blood specimens. Subjects’ perinatal and adolescent characteristics are noted in Table I. Thirty-five percent of subjects had overweight/obesity at age 14 years. Among the perinatal characteristics, subjects with overweight/obesity had higher rates of maternal hypertensive pregnancy and maternal smoking during pregnancy compared with those with BMI <85^th percentile. At age 14 years, subjects with overweight/obesity had greater median values for weight, BMI, and waist-to-height-ratio, and were more likely to have a sexual maturity rating of 5. The overweight/obesity group had higher mean systolic blood pressure (109.9 ± 9.1 vs 104.4 ± 9.7 mmHg) and mean systolic blood pressure z-score (0.04 ± 0.8 vs. −0.43 ± 0.9) at age 14 years. Subjects with overweight/obesity also had a greater rate of high blood pressure (20% vs. 8%).

TABLE 1.

Clinical characteristics by body mass index status

	Overweight/Obesity, n = 61	BMI <85th %ile, n = 114
Perinatal
Female	34 (56%)	63 (55%)
Black	29 (48%)	45 (39%)
Antenatal corticosteroid exposure	28 (46%)	64 (56%)
Maternal hypertensive pregnancy	34 (56%)*	41 (36%)
Maternal smoking	16 (26%)*	15 (13%)
Cesarean section	35 (57%)	57 (50%)
Gestational age (weeks)	28.2 (2.6)	27.6 (2.6)
Birth weight (g)	1066 (285)	1051 (256)
Small for gestational age	10 (16%)	9 (8%)
Adolescent
Height (cm)	162.7 (8.7)	161.0 (9.5)
Weight (kg)	74.5 [68.5, 83.5]*	50.4 [44.4, 56.0]
BMI (kg/m2)	28.7 [26.2, 30.8]*	19.5 [17.7, 21.1]
Waist-to-height ratio	0.59 [0.53, 0.63]*	0.43 [0.41, 0.46]
Sexual maturity rating of 5	46 (77%)*	59 (52%)
Medicaid	26 (45%)	43 (39%)
Current smoker	2 (3%)	3 (3%)
SBP (mmHg)	109.9 (9.1)*	104.4 (9.7)
DBP (mmHg)	60.3 (9.5)	61.9 (8.6)
SBP z-score	0.04 (0.8)*	−0.43 (0.9)
DBP z-score	−0.35 (0.85)	−0.18 (0.79)
High blood pressure	12 (20%)*	9 (8%)
Elevated blood pressure	9 (15%)*	6 (5%)
Stage 1 hypertension	3 (5%)	2 (2%)
Stage 2 hypertension	0 (0%)	1 (1%)
Plasma RAS (N = 121)	n = 42	n = 79
Ang II / Ang-(1–7)	6.09 [2.96, 12.22]*	3.46 [1.59, 6.16]
Ang II (pmol/l)	26.4 [18.9, 33.9]*	20.15 [16.2, 30.5]
Ang-(1–7) (pmol/l)	3.25 [1.98, 7.9]*	7.9 [2.6, 14.6]
PRA (pmol Ang I/l/h)	2.3 [1.1, 3.02]	2.31 [1.2, 3.3]
Aldosterone (pmol/l)	11.15 [6.16, 18.19]	8.77 [4.85, 13.29]
Urinary RAS
Ang II / Ang-(1–7)	0.16 [0.1, 0.31]	0.15 [0.1, 0.2]
Ang II / Cr (pmol/g)	0.06 [0.04, 0.1]	0.05 [0.04, 0.1]
Ang-(1–7) / Cr (pmol/g)	0.41 [0.31, 0.56]	0.42 [0.3, 0.6]
ACE / Cr (ng/mg)a	3.73 [2.07, 7.86]	3.43 [1.04, 7.3]
ACE2 / Cr (ng/mg)a	0.53 [0.33, 1.46]	0.55 [0.18, 1.23]
ACE / ACE2	6.68 [3.65, 14.82]	6.34 [3.01, 15.73]

N (%), mean (SD), or median [IQR]. Cr, creatinine.

^aN = 123 (overweight/obesity 44, BMI <85^th percentile 79).

^*P < 0.05 for comparison by BMI status via chi-square test, Fisher’s exact test, t-test, or Wilcoxon rank-sum test.

BMI correlated positively with plasma Ang II but inversely with Ang-(1–7) (Figure 2). Subjects with overweight/obesity had a higher median ratio of plasma Ang II to Ang-(1–7), higher median Ang II, and lower median Ang-(1–7) compared with subjects with BMI <85^th percentile (Table 1). There were no differences in median plasma renin activity or aldosterone. The differences in the Ang II to Ang-(1–7) ratio, Ang II, and Ang-(1–7) in the plasma persisted after adjustment for the potentially confounding factors race and antenatal corticosteroid exposure (Table 2). Compared with those with a BMI <85^th percentile, overweight/obesity was associated with a higher Ang II to Ang-(1–7) ratio (adjusted β: 0.57, 95% CI 0.23 to 0.91) as well as higher Ang II (adjusted β: 0.21 pmol/l, 95% CI 0.03 to 0.39) and lower Ang-(1–7) (adjusted β: −0.37 pmol/l, 95% CI −0.7 to −0.04).

Figure 2.

Body mass index is associated with higher plasma angiotensin II and lower plasma angiotensin-(1–7).

Plasma angiotensin-(1–7) on the left Y-axis and in black with circles; plasma angiotensin II on the right Y-axis and in gray with x’s; BMI on the X-axis. Loess fitted curves with cubic interpolation. Angiotensin-(1–7) – BMI Spearman correlation coefficient r = −0.29, P = .002. Angiotensin II – BMI Spearman correlation coefficient r = 0.29, p = 0.001.

TABLE 2.

Differences in the plasma and urinary renin-angiotensin system profiles according to body mass index status

	Crude	Adjusteda
	β (95% CI)	β (95% CI)
Plasma Ang II / (1–7)	0.63 (0.29, 0.97)	0.57 (0.23, 0.91)
Plasma Ang II (pmol/l)	0.17 (−0.01, 0.35)	0.21 (0.03, 0.39)
Plasma Ang-(1–7) (pmol/l)	−0.46 (−0.8, −0.13)	−0.37 (−0.7, −0.04)
Urinary Ang II / (1–7)	0.18 (−0.04, 0.41)	0.21 (−0.02, 0.44)

Generalized linear models with natural log-transformed outcomes.

^aAdjusted for race and antenatal corticosteroid exposure.

On initial bivariate analyses there were no between-group differences in the median values of the components of the RAS in the urine, including the ratio of Ang II to Ang-(1–7), Ang II/creatinine, Ang-(1–7)/creatinine, ACE/creatinine, ACE2/creatinine, and the ratio of ACE to ACE2 (Table 1). However, after controlling for race and antenatal corticosteroid exposure, overweight/obesity was associated with a higher ratio of urinary Ang II to Ang-(1–7) compared with those with a BMI <85^th percentile (β: 0.21, 95% CI −0.02 to 0.44), a result that approached statistical significance (Table 2).

Discussion

Among adolescents born preterm with very low birth weight, overweight/obesity was associated with increased blood pressure, a greater proportion of high blood pressure, and alterations in the circulatory and renal RAS compared with subjects with BMI <85^th percentile. Subjects with overweight/obesity had higher Ang II and lower Ang-(1–7) as well as a higher ratio of Ang II relative to Ang-(1–7) in the plasma. Compared with term-born peers, adolescents born preterm have a higher ratio of Ang II relative to Ang-(1–7) in the plasma, and obesity magnifies this effect wherein preterm-born adolescents with overweight/obesity have a higher ratio of plasma Ang II relative to Ang-(1–7) compared with term-born adolescents with overweight/obesity¹². In the current study we confirmed that within subjects born preterm, obesity has additional detrimental effects on the circulatory RAS that may compound those attributed to perinatal programming events.

A shift in the circulatory RAS towards higher Ang II expression, or tone, and lower Ang-(1–7) tone may partially explain the higher blood pressure in adolescent subjects with overweight/obesity and could serve as a mechanism for further increasing the risk of hypertension and cardiovascular disease during adulthood that is known to be associated with prematurity. Upregulation of Ang II at the expense of Ang-(1–7) is associated with persistently elevated blood pressure as well as inflammation and fibrosis in various organ systems such as the vasculature, kidneys, and heart that occur in part via vasoconstriction and reduced nitric oxide production leading to increased oxidative stress^{6, 20, 21}. Obesity is associated with dysregulation of the RAS which may contribute to the increased risk of hypertension and cardiovascular disease seen in patients with obesity^{22, 23}. Our finding that overweight/obesity is associated with higher blood pressure in adolescents born preterm supports similar findings reported by Rotteveel et al, although this study did not assess the Ang II or Ang-(1–7) pathways in the plasma and urine²⁴. Thus, obesity may provide an additive effect on the development of hypertension and cardiovascular disease in individuals born preterm that augments prematurity-induced perinatal programming of the RAS.

Subjects with overweight/obesity also had a higher ratio of urinary Ang II to Ang-(1–7) that approached statistical significance, indicating a possible shift in the renal RAS that increases the deleterious Ang II pathway but reduces the beneficial Ang-(1–7) pathway. Evidence from experimental models in rodents indicates that urinary angiotensin peptide levels are independent of glomerular filtration of peptides within the circulation and thus indicate generation within the tubules²⁵. The exact mechanisms are unclear but could be due to increased tubular formation or reduced metabolism of Ang II or decreased formation or increased metabolism of Ang-(1–7). We found no differences in urinary ACE or ACE2 in the current study, but other peptidases such as neprilysin, thimet oligopeptidase, or dipeptidyl peptidase 3 may contribute to the tubular processing of angiotensins²⁶. Additional study is warranted to better define this shift in the renal RAS by expanding our analysis of these peptidases in the urine.

The shift in the renal RAS that we describe could further contribute to the development of hypertension and cardiovascular disease by stimulating sodium retention, raising blood pressure, and ultimately promoting renal inflammation and fibrosis in part through lower nitric oxide production and increased oxidative stress^{27, 28}. Experimental models and human studies of the fetal-programming effect of antenatal corticosteroid exposure on the renal RAS also demonstrate this shift in the kidney towards Ang II, leading to elevated blood pressure and, in sheep, decreased renal function^{7, 11}. Our findings suggest that among adolescents born prematurely, a group already at increased risk for disease²⁹, having overweight/obesity may augment this risk in part through the shift in the renal RAS towards Ang II and away from Ang-(1–7).

A limitation of our study is that although plasma renin activity and aldosterone were measured, neither levels of the precursor protein angiotensinogen nor the peptidases neprilysin or, in the circulation, ACE and ACE2 that can generate and metabolize Ang II and Ang-(1–7) were assessed³⁰. Further study is warranted to establish whether alterations in the synthesizing components of the RAS are associated with higher Ang II or lower Ang-(1–7) in the overweight/obesity group. We also did not fully examine additional factors that may influence the RAS, including dietary salt or potassium intake or genetic profiles. Furthermore, the sex hormones including estrogen and testosterone are known to influence the RAS, and we did not measure their levels.

Potential strategies to prevent or treat disease in individuals born preterm could target the RAS to inhibit Ang II and augment expression of Ang-(1–7).

Figure 3.

Plasma ratio of angiotensin II to angiotensin-(1–7) according to body mass index status. Overweight/obesity on left, BMI <85th percentile on right. Bar denotes median, diamond denotes mean, box indicates IQR, and whiskers include ≤1.5x IQR. P < 0.05 via Wilcoxon rank-sum test. IQR, interquartile range.

Abbreviations

ACE

angiotensin-converting enzyme

ACE2

angiotensin-converting enzyme 2

Ang II

angiotensin II

Ang-(1–7)

angiotensin-(1–7)

BMI

body mass index

DBP

diastolic blood pressure

RAS

renin-angiotensin system

SBP

systolic blood pressure