Autonomic Functioning in Young Adults Born at Extremely Low Birth Weight

Abstract

Autonomic functioning is altered in infants born at extremely low birth weight (ELBW; <1000 g), but we know little about how such alterations manifest in adulthood. We examined associations between birth weight and resting heart rate (R-R interval), high-frequency (HF) and low-frequency (LF) heart rate variability, and systolic (SBP) and diastolic (DBP) blood pressure, in 60 participants (ages 22-26) born at ELBW and free of major neurosensory impairment, and 79 controls born at normal birth weight (NBW; >2500 g). HF in the smallest-born ELBW participants was significantly lower than in NBW controls. In both groups, greater birth weight was associated with higher HF. Among ELBW survivors, lower birth weight predicted faster heart rate and higher DBP, but neither heart rate nor DBP appeared to be well-coordinated with baroreflex activity (LF), the principal mechanism for short-term blood pressure regulation. Adult autonomic regulation may be significantly altered in those born extremely preterm.

Keywords

extremely low birth weight R-R interval heart rate variability baroreflex adult blood pressure SBP DBP

Introduction

Infants born at extremely low birth weight (ELBW) are exposed to multiple forms of prenatal and early postnatal adversity, including the stresses that contribute to premature delivery, and exposure to postnatal medical procedures administered to improve their odds of survival. Because heart rate variability,¹ baroreflex sensitivity,² and myelination of the vagus nerve³ undergo substantial development during the third trimester of pregnancy, the consequences of extremely preterm birth may include alterations in autonomic nervous system (ANS) functioning. Autonomic indices such as heart period (R-R interval)⁴ and high-frequency (HF) and low-frequency (LF) heart rate variability^2,5,6 are reduced in infants born very preterm (28-32 weeks’ gestational age) or small for gestational age (SGA; birth weight <10th percentile for gestational age⁷), increasing their risk of poor systemic perfusion and adverse neurological events in early infancy. Moreover, autonomic development does not appear to “catch up” between birth and term-equivalent age.^8,9 Accordingly, autonomic responsiveness has been shown to be reduced in children born preterm at age 9 or 10 years.^10,11 Researchers have also reported higher blood pressure (BP) in young adults,¹² adolescents,¹³ and children¹⁴ born very preterm. In children born extremely preterm,¹⁵ increases in BP may be evident as early as 2.5 years of age.

Relatively little is known about the longer term cardiovascular effects of being born extremely preterm. We do know that cardiac arrhythmias¹⁶ and subclinical atherosclerosis¹⁷ are more prevalent in adults born at low birth weight. Individuals born extremely small may also be prone to increased arterial stiffness^18,19 and elevated BP^20,21 in adolescence and young adulthood, especially males,²⁰ and those who showed rapid weight gain in early childhood.^21-23 Sequelae of extremely preterm birth appear to include reduced physical fitness in young adulthood.²⁴

In clinical settings, birth weight serves as an easily measured predictor of neonatal mortality and morbidity.²⁵ Birth weight reflects gestational age, genetic inheritance, and various intrauterine environmental exposures, including poor nutrition, restricted placental blood flow, and maternal factors such as gestational diabetes.²⁶ A first step to ascertaining whether ELBW may influence autonomic functioning in adult life is to explore associations between birth weight and adult indices of heart rate, autonomic regulatory capacity, and BP in participants born at ELBW and their counterparts born full term and at normal birth weight (NBW).

Most studies of associations between birth weight and autonomic outcomes have defined birth weight as a categorical rather than a continuous variable. Because assessing groups as categories may fail to capture differences in biological or functional relations among variables, it may be more useful to assess individual differences along a continuum. Using continuous variables preserves more information in the data, thereby providing greater statistical power for finding effects.²⁷ Therefore, we used birth weight as both a categorical variable and as a continuous variable, adjusting group means and correlations for growth restriction because of its high prevalence in preterm infants (eg, 24% of ELBW infants are born SGA)²⁸ and adverse consequences. Separating the relative contributions of being born preterm from those of growth restriction is important because preterm birth and growth restriction represent separate etiological pathways.²⁰

In healthy young adults, resting heart rate exhibits substantial variability. Higher levels of resting heart rate variability (HRV) have been linked to dynamic cardiovascular adaptation to environmental demands²⁹ and a lower risk for cardiovascular disease.³⁰ Resting heart rate is predominantly under the control of the parasympathetic nervous system,³¹ which is responsible for effecting rapid, beat-to-beat changes in heart rate. The cardiac signal may be decomposed to ascertain the power density of its major frequencies using spectral frequency analysis. HF HRV (above 0.12 Hz) in the cardiac signal is centered on the respiratory frequency and primarily reflects parasympathetic modulation of heart rate. LF HRV (about 0.10 Hz) may reflect activity of the arterial baroreflexes, which adjust heart rate and peripheral vascular resistance to offset sudden variations in BP whenever they occur in response to environmental or internal demands.^32,33 Individual differences in resting heart rate and HRV are thought to originate in the prenatal period,³⁴ persist in infancy,³⁵ and remain stable in adulthood.³⁶

Both HF and LF are reported to increase substantially between 24 weeks’ gestation and term.^1,2,37 In preterm neonates, higher resting HF signifies greater maturation of parasympathetic control.² In adults, resting HF is positively associated with cardiovascular health,³⁸ and competent emotion regulation,³⁹ executive functioning,^40,41 and attentional control.⁴² In neonates born preterm, baroreflex activity appears to be underdeveloped and matures more slowly than it does in full-term infants.^2,37, signifying increased vulnerability to potentially dangeroushypotensive events.⁶ In adults, lower resting LF is associated with various forms of autonomic failure (eg, congestive heart failure³³). Like HF, baroreflex sensitivity normally shows progressive increases between 28 weeks’ gestation and term.² However, preterm birth may delay the normal maturational increase in baroreceptor sensitivity, resulting in substantially reduced baroreflex control at 5 or 6 months corrected age.⁶ In adults, baroreflex sensitivity is considered a major prognostic indicator of cardiovascular health, that is reduced in hypertensive conditions. Importantly, HF and LF reflect 2 aspects of autonomic regulatory capacity, whereas heart rate (R-R) and BP (systolic blood pressure [SBP] and diastolic blood pressure [DBP]) are net outcomes of autonomic regulatory processes.

Although the extant literature suggests that cardiovascular regulation may be altered in infants and children born at ELBW, relatively little is known about whether these autonomic differences persist in adulthood. We tested whether birth weight predicted resting measures of autonomic regulation (R-R interval, HF, LF, SBP, and DBP) in young adults born at ELBW and their peers born at NBW, as a first step in examining the effects of extremely preterm birth on cardiovascular functioning. If being born at ELBW confers increased long-term cardiovascular risk, we would expect to see group differences in autonomic parameters, and/or differential associations between birth weight and autonomic parameters in adult ELBW survivors versus NBW controls. Specifically, we hypothesized that higher birth weights within the ELBW group would be associated with greater capacity for autonomic regulation (greater R-R interval, HF, and LF, and lower SBP and DBP) in adulthood, whereas the restricted range represented by NBW control participants would result in smaller correlations in this group. A third prediction was that associations among the autonomic parameters would differ between groups, suggesting altered functional relationships among these variables.

Method

Cohort and Participants

The cohort included ELBW survivors recruited at birth, and a group of NBW controls recruited when ELBW participants and controls were 8 years old. Originally, the ELBW group included 397 mainly Caucasian infants born between 1977 and 1982 within a geographically defined area in south western Ontario, Canada Weighing between 500 and 1000 grams at birth, these infants have been assessed longitudinally at ages 3, 5, 8, 14, 22 to 26, and 29 to 36 years. Gestational ages for the ELBW participants ranged from 23 to 34 weeks, with 68% (n = 121) being born <28 weeks’ gestational age. Of the original cohort, 179 (45%) survived to hospital discharge, and 13 children died subsequently. Of the 166 survivors, 142 were contacted and invited to participate in the young adult assessment at age 22 to 26⁴³ at which time a subset of 71/142 right-handed participants free of major neurosensory impairments (eg, blindness, deafness, and cerebral palsy) was tested in the Child Emotion Laboratory at McMaster University.44 Usable electrocardiogram (ECG) data were obtained from 67/71. Resting HF (2 participants) or socioeconomic status (SES) information (5 participants) was unobtainable from some participants, leaving data from 60/71 (85%) of the ELBW survivors available for analysis. Blood pressure measures were missing for 1 ELBW participant and 1 NBW participant.

The age-, sex-, and SES-matched control group consisting of 145 children born at NBW has been assessed in tandem with the ELBW cohort at ages 8, 14, 22 to 26, and 29 to 36. Of these, 133 were assessed at age 22 to 26. From this group, a subset of 83/133 right-handed control participants was tested in our laboratory, yielding usable ECG data from 80/83. Resting HF was unobtainable from 1 participant, leaving data from 79/83 (95%) of the NBW controls available for analysis.

Procedure

Participants were introduced to the laboratory and briefed about the study procedures. Resting ECG and BP measures were collected after obtaining written consent. Resting R-R interval, HF, and LF were quantified noninvasively by spectral frequency analysis of the ECG. All laboratory procedures were approved by the participating university and hospital research ethics boards.

ECG Collection and Measures

ECG Recording

ECG was continuously recorded during 2 minutes of rest via 2 disposable ECG electrodes placed on the medial forearms while the participant sat quietly in a comfortable chair. Recordings made during the resting state, before any task has started, provide measures that are characteristic of the individual, as this state is assumed to be relatively free of external demands. A minimum recording length of 2 minutes is required for accurate assessment of LF.45 This recording length also meets the minimum required for calculating HF.⁴⁵ Because spontaneous respiration rates during seated rest are generally slow enough to avoid any undue influence of respiration,^32,46 respiration rate was not monitored in this study. ECG signals were amplified by an individual SA Instrumentation Bioamplifier, filtered between 0.1 Hz (high pass) and 1000 Hz (low pass), and digitized at a sampling rate of 512 Hz. The acquisition software was Snapshot-Snapstream (HEM Data Corp, Southfield, MI). Measures of maximum (SBP) and minimum (DBP) arterial pressure were collected by an ambulatory BP monitor while each participant was comfortably seated.

ECG Reduction and Quantification

Mean levels of R-R interval, HF, and LF were derived offline from each resting ECG recording in its entirety. Indices of R-R interval are preferred over heart rate in autonomic research, as heart rate is a nonlinear transform of the R-R interval.⁴⁷ Cardiac R-waves were detected offline with a 4-pass algorithm that produced a visual display of the cardiac signal with the R-waves marked (IBI Analysis, James Long Company, Caroga Lake, NY). Missing or spurious R-waves were edited manually according to Berntson and Stowell.⁴⁸ The edited file of R-wave onset times was converted to interbeat intervals and prorated into equal time intervals of 125 milliseconds using a moving polynomial filter. This series was then detrended using a high-pass filter with a period of 25 seconds (frequency cut-off, 0.04 Hz), to allow quantification of both HF variability (ie, 0.12 to 0.40 Hz) and LF variability (0.04 to 0.12 Hz) in the signal. A discrete Fourier transform analysis, with a 32-second Hanning window and 50% consecutive overlap, was applied to quantify HF variability in the signal, summed across the range of respiratory frequencies, and also LF variability, summed across the LF range. In keeping with common practice in current psychophysiological research,^{49, 50} HF and LF values were then submitted to natural log transformations to normalize their distributions (ln ms²) before analysis.

Statistical Analysis

The variables of interest were R-R interval, HF, LF, SBP, and DBP. Group differences in mean levels were assessed using independent, 2-sided t tests. Separate regression analyses were performed to ascertain the variance explained by birth weight for each autonomic variable and group (Model 1, unadjusted). To assess the unique contribution of birth weight to predictions of autonomic activity, we ran a second, hierarchical, regression analysis for each autonomic variable with age, sex, SGA status, familial SES, and body mass index (BMI), entered as covariates on the first step, followed by birth weight on the second step (Model 2, adjusted). The unique variance attributable to each independent predictor was calculated and reported as semipartial correlations (sr²). Partial correlations between birth weight and each of the autonomic parameters, and among the autonomic parameters themselves, were then computed for each group, controlling statistically for intrauterine growth restriction (ie, SGA status⁷). Fisher’s r-to-z transformations were used to test for significant between-group differences in the partial correlations (2-tailed). SPSS version 22.0 (IBM SPSS Inc, Chicago, IL) was used for all analyses.

Covariates

Birth weight, rather than gestational age, served to differentiate between groups born at term and those born extremely prematurely because accurate information on gestational age was not available for the control participants, who had entered the study at age 8. In the ELBW group, however, birth weight and gestational age were significantly correlated, r(60) = .42, P < .01. Given that the risks conferred by growth restriction may be independent of those associated with being born at ELBW,⁵¹ all analyses were adjusted for SGA status. Age on the date of assessment was included as a covariate in each regression analysis, as HF is sensitive to differences in age within even a narrow range.⁵² Sex was included because HF⁵³ and LF⁵⁴ have been reported to vary by sex in some studies. Because low SES has previously been associated with lower levels of HF⁵⁵ and higher BP,⁵⁶ familial SES at age 8 was included as a covariate to account for the home rearing environment during childhood development. The 2-factor Hollingshead scale,⁵⁷ which incorporates parental education level and employment status, was used to assess familial SES. In this 5-point scale, lower SES was indicated by higher values. Finally, BMI at the time of assessment was included as an index of adult body size, because greater body mass has been associated with lower HF in some studies,⁵⁸ and overweight is a risk factor for cardiovascular disease. Incorporating both mass (kg) and height (m²), BMI captured variance associated with adult weight and genetic differences in growth potential (height).

Results

Demographic and cardiovascular measures are reported in Table 1. The ELBW participants, included (n = 60) and excluded (n = 11), were similar with respect to demographic characteristics (birth weight, age, sex, SGA status, adult height, weight, and BMI, and familial SES, all Ps > .16). The NBW participants, included (n = 79) and excluded (n = 4), were also similar with respect to these characteristics (all Ps > .45).

Table 1.

Sample Characteristics.

Variable	ELBW (n = 60), Mean (SD)	NBW (n = 79), Mean (SD)	Mean Difference^a (95% CI)
Mean gestational age***	27.4 (2.1) weeks	Term
SGA/AGA***	19/41	2/77
Sex (male/female)	25/35	34/45
Birth weight (g)***	872 (114)	3403 (465)	−2530.7 (−2638.7 to −2422.6)
Age in years*	23.1 (1.2)	23.6 (1.1)	−0.49 (−0.88 to −0.10)
Adult height (cm)**	164.6 (10.3)	169.2 (10.2)	−4.65 (−8.14 to −1.17)
Adult weight (kg)	65.5 (16.3)	70.4 (16.6)	−4.91 (−10.48 to 0.67)
Body mass index (kg/m²)	24.1 (5.7)	24.5 (5.1)	−0.35 (−2.16 to 1.46)
Familial SES^b (age 8)	3.35 (0.8)	3.05 (1.0)	0.30 (−0.01 to 0.61)
Resting R-R interval (ms)	841 (142)	852 (132)	−10.86 (−57.13 to 35.41)
Resting HF (ln ms²)	6.65 (0.9)	6.93 (1.1)	−0.29 (−0.63 to 0.05)
Resting LF (ln ms²)	6.31 (1.0)	6.49 (0.9)	−0.18 (−0.50 to 0.15)
Resting SBP^c (mm Hg)	118.7 (14.6)	119.2 (14.0)	2.47 (−5.34 to 4.41)
Resting DBP^c (mm Hg)	69.9 (10.1)	68.9 (9.1)	1.65 (−2.26 to 4.25)

Abbreviations: ELBW, extremely low birth weight; NBW, normal birth weight; SD, standard deviation; CI, confidence interval; SGA, small for gestational age; AGA, appropriate for gestational age; SES, socioeconomic status; R-R interval, heart period; HF, high-frequency heart rate variability; LF, low-frequency heart rate variability; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Group differences were assessed by t tests or χ² tests.

Familial SES was assessed by the 2-factor scale of Hollingshead.⁵⁵ Higher values = lower SES, ie, 3 = middle SES level, 4 = second-lowest SES level.

BP data were unavailable for 1 ELBW participant and 1 NBW participant.

Birth weight was normally distributed in the NBW group (Shapiro–Wilk, P > .70), but right-skewed in the ELBW group because comparatively fewer ELBW survivors were born at the very lowest birth weights (Shapiro–Wilk, P < .001). Each of the cardiovascular and BP variables was normally distributed in the ELBW survivors (Shapiro–Wilk, Ps > .08) and their NBW (Ps > .12) peers. Although no group differences in the mean resting autonomic parameters reached significance (Shapiro–Wilk, Ps > .09), R-R, HF, and LF were numerically lower in ELBW survivors. In addition, a comparison of NBW controls with the smallest-born ELBW survivors revealed that HF in the controls was significantly higher than in the smallest ELBW survivors (divided by a median split at 920 g; NBW: mean = 6.93 ln ms², SD = 1.05; ELBW: mean = 6.44 ln ms², SD = 0.84), although no other subgroup differences reached significance (Ps > .09).

Regression Models of Adult Autonomic Measures on Birth Weight in ELBW Survivors

Results for the ELBW group are presented in Table 2 and Figure 1. In both regression models (unadjusted, adjusted), higher birth weight was associated with longer resting R-R intervals (slower heart rates), accounting for 13% of the variance in the R-R interval. Although the association of birth weight with R-R interval was significant for the ELBW group as a whole, this association depended mainly on the smallest survivors (≤920 g; r[30] = .42, P < .03). The correlation did not reach significance for ELBW participants with relatively higher birth weights (920-1000 g; r[30] = .01, P > .95; Fisher r-to-z, P < .09). ELBW birth weight also explained 7% to 8% of the total variance in resting HF, and 4% to 7% of the variance in SBP and DBP. In contrast, ELBW birth weight accounted for little variance (<1%) in resting LF in either model. We note also that associations between birth weight and autonomic measures did not differ from those in the original analyses when adult weight was substituted for adult BMI.

Table 2.

Results of Regression Models of Resting Autonomic Variables on Birth Weight for the ELBW Group^a.

	Predictors	R ²	FΔ	df	B	SE	sr ²
R-R interval
Model 1	Birth weight	.13	8.55**	1, 58	0.45	0.15	.13
Model 2	Step 1	.05	0.52	5, 54
	Sex				−30.69	38.48	.01
	Age				16.72	15.61	.02
	SGA status				−19.50	41.50	<.01
	BMI				−2.20	3.68	.01
	SES				11.99	25.05	<.01
	Step 2
	Birth weight	.13	8.47**	1, 53	0.47	0.16	.13
HF
Model 1	Birth weight	.08	5.10*	1, 58	.002	.001	.08
Model 2	Step 1	.10	1.25	5, 54
	Sex				−0.07	0.24	<.01
	Age				0.06	0.10	.01
	SGA status				−0.44	0.26	.05
	BMI				−0.03	0.02	.04
	SES				0.10	0.16	.01
	Step 2
	Birth weight	.07	4.60*	1, 53	0.002	0.001	.07
LF
Model 1	Birth weight	<.01	0.02	1, 58	.000	.001	<.01
Model 2	Step 1	.26	3.74**	5, 54
	Sex				−0.83**	0.24	.16
	Age				0.15	0.10	.03
	SGA status				−0.36	0.26	.03
	BMI				−0.04	0.02	.04
	SES				−0.03	0.16	<.01
	Step 2
	Birth weight	<.01	0.05	1, 53	0.001	0.001	<.01
SBP
Model 1	Birth weight	.06	3.72^†	1, 57	−0.03	.02	.06
Model 2	Step 1	.26	3.77**	5, 53
	Sex				−8.50*	3.54	.08
	Age				−3.76*	1.42	.10
	SGA status				−0.49	3.79	<.01
	BMI				0.77*	0.33	.07
	SES				0.58	2.34	<.01
	Step 2
	Birth weight	.04	3.10^‡	1, 52	−0.03	0.02	.04
DBP
Model 1	Birth weight	.07	3.94^†	1, 57	−.02	.01	.06
Model 2	Step 1	.19	2.40*	5, 53
	Sex				−4.03	2.57	.04
	Age				−1.39	1.03	.03
	SGA status				−4.86^‡	2.75	.05
	BMI				0.42	0.24	.05
	SES				0.90	1.69	<.01
	Step 2
	Birth weight	.07	4.77*	1, 52	−0.02	0.01	.07

Abbreviations: ELBW, extremely low birth weight; df, degrees of freedom; R-R interval, heart period; SGA, small for gestational age; BMI, body mass index; SES, socioeconomic status; HF, high-frequency heart rate variability; LF, low-frequency heart rate variability; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Birth weight predictions did not change when regressions were performed on untransformed HF and LF data. HF, Model 1: B = 2.92 (SE = 1.19), P < .02. Model 2, Step 2: B = 2.84 (SE = 1.23), P < .03. LF, Model 1: B = 0.93 (SE = 1.12), P > .40. Model 2, Step 2: B = 0.93 (SE = 1.09), P > .35.

P < .01. *P < .05. ^†P < .06. ^‡P < .09.

Figure 1.

Pearson (r) correlations between birth weight and resting R-R, HF, LF, SBP, and DBP, by group.

Regression Models of Adult Autonomic Measures on Birth Weight in NBW Controls

Regression results for the NBW control group are presented in Table 3 and Figure 1. Birth weight in this group explained very little variance (2%) in the R-R interval, resting SBP, or DBP (<1% to 1%), but accounted for 4% to 6% of the variance in resting HF, and 10% to 13% of the variance in resting LF. Associations between birth weight and autonomic measures were retained when adult weight was substituted for adult BMI.

Table 3.

Results of Regression Models of Resting Autonomic Variables on Birth Weight for the NBW Group^a.

	Predictors	R ²	FΔ	df	B	SE	sr ²
R-R interval
Model 1	Birth weight	.02	1.48	1, 77	0.04	0.03	.02
Model 2	Step 1	.09	1.37	5, 73
	Sex				11.91	31.83	<.01
	Age				−0.42	15. 01	<.01
	SGA status				−130.00	97.01	.02
	BMI				−6.64*	2.93	.06
	SES				−9.99	16.74	<.01
	Step 2
	Birth weight	.02	1.36	1, 72	0.04	0.03	.02
HF
Model 1	Birth weight	.06	4.58*	1, 77	0.001	.000	.06
Model 2	Step 1	.18	3.15**	5, 73
	Sex				0.28	0.24	.02
	Age				−0.04	0.11	<.01
	SGA status				−1.70	0.73	.06
	BMI				−0.06*	0.02	.07
	SES				−0.26*	0.13	.05
	Step 2
	Birth weight	.04	3.48^‡	1, 72	0.000	0.000	.03
LF
Model 1	Birth weight	.13	11.24**	1, 77	0.001	.000	.13
Model 2	Step 1	.18	3.12*	5, 73
	Sex				−0.15	0.21	.01
	Age				0.05	0.10	<.01
	SGA status				−0.98	0.64	.03
	BMI				−0.04*	0.02	.05
	SES				−0.29*	0.11	.08
	Step 2
	Birth weight	.10	10.15**	1, 72	0.001	0.000	.10
SBP
Model 1	Birth weight	<.01	0.02	1, 76	0.00	.003	<.01
Model 2	Step 1	.35	7.64***	5, 72
	Sex				−13.74***	2.87	.21
	Age				1.05	1.36	.01
	SGA status				17.80*	8.71	.04
	BMI				0.76**	0.27	.07
	SES				−1.32	1.53	.01
	Step 2
	Birth weight	<.01	0.02	1, 71	0.000	0.003	<.01
DBP
Model 1	Birth weight	.01	0.71	1, 76	−0.002	0.002	.01
Model 2	Step 1	.16	2.69*	5, 72
	Sex				−2.86	2.12	.02
	Age				0.25	1.00	<.01
	SGA status				17.38**	6.41	.09
	BMI				0.46*	0.20	.06
	SES				1.43	1.12	.02
	Step 2
	Birth weight	<.01	0.13	1, 71	−0.001	0.002	<.01

Abbreviations: NBW, normal birth weight; df, degrees of freedom; R-R interval, heart period; SGA, small for gestational age; BMI, body mass index; SES, socioeconomic status; HF, high-frequency heart rate variability; LF, low-frequency heart rate variability; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Birth weight predictions did not change when regressions were performed on untransformed HF and LF data, except Model 1 birth weight prediction of HF now reached significance. HF, Model 1: B = 1.08 (SE = 0.41), P < .02. Model 2, Step 2: B = 0.99 (SE = 0.43), P < .03. LF, Model 1: B = 0.68 (SE = 0.20), P < .01. Model 2, Step 2: B = 0.64 (SE = 0.21), P < .01.

***

P < .001. **P < .01. *P < .05. ^‡P < .09.

Group Differences in Association Strengths

Group differences in the partial correlations between birth weight and the autonomic parameters were tested using Fisher’s r-to-z transformation. Results for each group are presented in Table 4. Correlation strengths did not differ between groups, with one important exception: the association between birth weight and LF was significantly greater in NBW controls than ELBW survivors (z = 2.58, P < .01; see Figure 1). As well, NBW controls showed stronger partial correlations between LF and the R-R interval (z = 2.73, P < .01) and between LF and DBP (z = 2.07, P < .04; see Figure 2) than did ELBW survivors, who showed no such associations (Ps > .45).

Table 4.

Partial Correlations (pr) Among Resting Autonomic Measures and Blood Pressure Measures by Group, Controlled for Growth Restriction (SGA Status)^a.

	Birth Weight	RR	HF	LF	SBP	DBP
ELBW (n = 59)
Birth weight	—
RR	.34**	—
HF	.23	.32*	—
LF	−.09	−.01	.44**	—
SBP	−.25^†	−.22	−.12	.06	—
DBP	−.30*	−.45***	−.14	.10	.69***	—
NBW (n = 78)
Birth weight	—
RR	.11	—
HF	.20	.56***	—
LF	.34 **	.44***	.68***	—
SBP	.05	−.31**	−.18	−.09	—
DBP	−.03	−.45***	−.39**	−.26 *	.54***	—

Abbreviations: SGA, small for gestational age; ELBW, extremely low birth weight; NBW, normal birth weight; R-R interval, heart period; HF, high-frequency heart rate variability; LF, low-frequency heart rate variability; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Bold values identify significantly stronger correlations in NBW controls than ELBW survivors, by Fisher’s r-to-z transformation (2-tailed): LF-Birth Weight, z = 2.54, P < .02. LF-RR: z = 2.73, P < .01. LF-DBP: z = 2.07, P < .04.

***

P < .001. **P < .01. *P < .05. ^†P < .06.

Figure 2.

Pearson (r) correlations between LF heart rate variability and R-R interval, DBP, by group, without controlling for SGA status.

In sum, for young adults who were born weighing less than 1000 g, relatively lower birth weight was associated with faster resting heart rate and greater DBP, and lower HF activity (parasympathetic regulatory capacity). In young adults born weighing >2500 g, birth weight was positively associated with HF and LF regulatory activity. Importantly, among NBW controls, LF regulatory activity was inversely correlated with resting heart rate and DBP, links that were absent in the ELBW survivor group.

Discussion

The goal of this study was to examine whether being born at ELBW conferred increased autonomic risks in adulthood. As a first step to answering this question, we tested for associations among birth weight and resting autonomic measures in young adults born at ELBW and their NBW peers. Although ELBW participants in this sample were slightly younger, shorter, and socioeconomically more disadvantaged than NBW participants, mean resting autonomic measures did not differ significantly between groups (in contrast to other studies^5,20). However, HF heart rate variability was significantly higher in NBW controls than in the smallest-born ELBW young adults, indicating that adult parasympathetic regulatory capacity may indeed be lower following extremely preterm birth. In addition, the groups showed distinctive associations between birth weight and autonomic regulation that were broadly consistent with other cardiovascular conditions reported in adults born extremely preterm (eg, cardiac arrhythmias¹⁶ and hypertension²²). In adult ELBW survivors, smaller physical size at birth predicted faster resting heart rate, higher DBP, and lower capacity for parasympathetic regulation (HF), after adjusting for potential confounds. In control adults born at full term and at NBW, relatively lower birth weights were associated with lower levels of parasympathetic control (HF), and lower resting baroreflex activity (LF), but were unrelated to heart rate or BP measures.

Resting Heart Rate (R-R Interval)

As birth weight accounted for significant variance in resting heart rate in ELBW survivors after controlling for relevant covariates, the association between birth weight and heart rate must be attributable to some factor other than age, sex, SGA status, familial SES, or adult body size.^*,† One possibility is that ELBW survivors may be prone to alterations in cardiac morphology and function that promote faster resting heart rates.

The birth event induces profound changes in infants’ circulatory functioning.^59,60 In humans, early birth may result in increased ventricle mass and other changes in gross cardiac morphology (eg, shorter ventricles with smaller internal volumes^61,62), changes that are likely to be irreversible.²⁶ Moreover, the severity of adverse cardiac remodeling increases with shorter gestational periods.⁶¹,⁶² Evidence from nonhuman animal models suggests that a premature increase in the hemodynamic load instigated by early birth may lead to increased collagen deposition (fibrosis or stiffening) in myocytes of the heart’s left ventricle⁵⁹ that may affect its structure, and potentially, its function. Like the heart, the vasculature may be subject to remodeling in individuals born very small. Deficient elastin synthesis in the large arteries and aorta has been proposed as a causal mechanism for vascular changes that predispose individuals to increased risk of hypertension and cardiovascular disease.^63,64,65 Structural alterations that decrease flexibility in the heart or vasculature may necessitate increased cardiac activity, including a faster heart rate, in order to maintain adequate stroke volume and BP in ELBW populations.

Although the association of birth weight with R-R interval was significant for the ELBW group as a whole, scatterplots suggested that the association depended mainly on the smallest-born survivors (≤920 g). No association was apparent for those ELBW participants with relatively higher birth weights (≥920 g), nor for control participants, whose intrauterine cardiac development had proceeded without being interrupted by early birth. These differential associations suggest a possible threshold effect of gestational age, beyond which cardiac remodeling is unlikely to occur and birth weight no longer predicts resting heart rate in preterm survivors.

Resting Parasympathetic Regulation (HF)

Correlations between birth weight and resting parasympathetic activity were of similar magnitude in adults born at ELBW and at NBW. These parallel findings suggest that the development of parasympathetic regulation of heart rate may begin comparatively early in gestation.³¹ Fetal heart rate is known to rise to 175 bpm by 9 weeks’ gestational age, declining thereafter to about 150 bpm by 16 weeks⁶⁶ and 140 bpm by term.¹ Diminishing fetal heart rate during the second and third pregnancy trimester reflects ongoing development of the parasympathetic nervous system, which gradually overtakes the accelerating effects of developing sympathetic control.¹ If parasympathetic control of heart rate increases from 9 weeks’ gestational age, then vagal development is likely to be well underway at the time of birth even in individuals born before the third trimester.^1,67. The absence of group differences in resting HF across the whole sample suggests that vagal function may “catch up” by young adulthood in many ELBW survivors

Resting Baroreflex Regulation (LF)

The main mechanism for modulating heart rate and vascular resistance in response to sudden variations in BP is the baroreflex. Like parasympathetic regulation (HF), baroreflex sensitivity shows a linear increase with gestational age in nonhuman animal models,^68,69 and in human infants.^2,37 Baroreflex activity tends to be underdeveloped in infants born preterm, and its postnatal maturation may remain delayed, even by theoretical term.³⁷ Here, a robust, positive association between birth weight and baroreflex activity (LF) was evident among NBW controls (P < .01), even after familial SES and adult body size were accounted for, while no association was observed in ELBW survivors (P > .85). These differential associations suggest the dependence of LF development on reaching an advanced gestational age. In addition, consistent with reports of reduced HRV in groups with lower SES,⁵⁵ LF was lower among controls who experienced lower SES during development. Among ELBW survivors, female sex was associated with lower LF, in line with studies showing that baroreflex activity is sexually dimorphic in humans⁵⁴ and nonhuman animals,⁷⁰ and lower in females.

Blood Pressure (SBP and DBP)

Much evidence links relatively lower birth weights to elevated BP levels in individuals born very preterm^{13,15,20, 61, 71} or even near term.^14,23 Stroke volume, systolic and diastolic functioning, and right ventricular ejection fraction all appear to be adversely affected in adults born preterm.^61,62 Studies of children and adolescents indicate that these associations appear early.^{14, 72} They may also amplify with age, given that BP increases occurring between childhood and young adulthood are larger in individuals born at low birth weights.⁷³ Like cardiac remodeling, the degree of BP deviation in young adulthood is graded with the degree of prematurity.^{13,20, 61, 62}

While some studies have found stronger links between very low birth weight and SBP than DBP,^13,20 others report only altered DBP in adults born very preterm.⁷¹ Here, we found that with SGA status controlled, lower birth weights in the ELBW group predicted higher minimum arterial pressures during the relaxation segment of the cardiac cycle (DBP, Table 4). This association suggests that diastolic relaxation in ELBW survivors may be sluggish or incomplete, with possible negative consequences for cardiovascular functioning⁶¹. Indeed, evidence suggests that diastolic functioning in young adults who were born very preterm is comparable to that of control adults who are 10 years older.⁶¹

Associations Among Autonomic Variables

Although the ELBW and NBW groups did not differ with respect to mean levels of resting heart rate, HRV, or blood pressure, they exhibited important differences with respect to associations involving resting baroreflex (LF) activity. Whereas NBW controls exhibited strong positive correlations between LF and R-R interval, and significant negative correlations between LF and DBP, both heart rate and diastolic relaxation in the ELBW group appeared to be independent of baroreflex activity. These differential findings suggested that in adult ELBW survivors, resting heart rate and DBP may not be well-controlled by the baroreflexes. Baroreflex insufficiency may have a role in explaining the development of higher BP,²⁰ and possibly, lower fitness²⁴ in adults born at ELBW, although these hypotheses remain to be explored in future studies.

Limitations

Some study limitations should be borne in mind when interpreting the findings. First, the exclusion of ELBW survivors with significant neurosensory impairments limits the generalizability of our results to adults born extremely preterm with relatively intact levels of functioning. However, given that our general goal was to examine autonomic effects of being born extremely small, it was deemed prudent not to introduce the confound of serious neurosensory impairment (e.g., cerebral palsy).

Second, although the Task Force of the European Society of Cardiology guidelines⁴⁵ state that 2-minute recordings meet the minimum requirements for assessing HF and LF, we acknowledge that the recommended length for short ECG recordings is 5 minutes, and that conforming to this standard allows easier comparisons across studies. Measures of LF are adversely affected if too few cycles are recorded or if the detrending period is too short (eg, 10 seconds, frequency cutoff, 0.10 Hz). Adherence to the minimum recording length will have reduced this possibility, and we took care to alter the period of the high-pass filter from the usual 10 seconds to 25 seconds when detrending the cardiac signal, to allow accurate calculation of LF.

Third, we acknowledge that it is preferable to record respiration when assessing HF, in order to demonstrate that any reported group differences or condition effects involving HF were not simply attributable to changes in respiratory rate or volume. Fortunately, the effects of spontaneous breathing on resting HF are likely to be small.^32,46,50 As well, because there were no obvious group differences in resting HF, we inferred that both resting parasympathetic activity and respiration rates were similar across groups.

Fourth, our indices of cardiovascular functioning were limited to resting heart rate, spectral measures of HRV (HF and LF), and BP. Examining whether birth weight predicts autonomic reactivity—ie, autonomic changes between rest and physical or psychological stress tests—would allow ecologically valid observations of the autonomic system “in action.” Finally, although our sample size was reasonably large for studies of extremely preterm birth, the present findings require replication with adult ELBW cohorts from other locations, representing other demographic categories. Nonetheless, after controlling for intrauterine growth restriction and other key variables, the data from this sample were sufficient to identify significant group differences in the associations between birth weight and autonomic regulatory parameters, more than 20 years after birth.

Conclusions

The strengths of this study include the availability of a geographically determined cohort (reducing the impact of referral biases) and reasonably high follow-up rates in adulthood. The major findings were that adult parasympathetic regulatory capacity was significantly reduced in the very smallest-born ELBW survivors, and that associations between birth weight and resting autonomic measures differed between ELBW survivors and their NBW counterparts, suggesting increased cardiovascular risk in individuals born extremely preterm. Within the group of ELBW survivors, those born at relatively lower birth weights had faster resting heart rates, lower resting parasympathetic activity (HF), and higher resting DBP. Within the NBW control group, relatively higher birth weights were associated with greater parasympathetic (HF) and baroreflex (LF) activity, and both resting heart rate and DBP were closely aligned with baroreflex control (LF). Notably, and reported for the first time in young adults born at ELBW, neither resting heart rate nor BP appeared to be well-coordinated with baroreflex activity (LF), the principal mechanism for short-term, reflexive control of BP. We conclude that autonomic functioning may be significantly altered in young adulthood for those born extremely preterm. Given these circumstances, early monitoring for symptoms of hypertension and cardiac disease may be warranted in this special population as its members grow older.

Footnotes

Acknowledgements

We wish to thank the many participants and their families for continued participation over the years.

Author Contributions

KJM contributed to conception and design, acquisition, analysis, and interpretation; drafted manuscript; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy. RJVL contributed to analysis and interpretation; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy. SS contributed to interpretation; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy. KMM contributed to interpretation; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy. MHB contributed to analysis and interpretation; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy. LAS contributed to conception and design; contributed to analysis and interpretation; critically revised manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a Canadian Institutes of Health Research (CIHR) Team Grant (2009H00529, awarded to LAS) and a National Institute of Child Health and Human Development (NICHD) Grant (R01HD40219, awarded to SS).

*

A covariate representing “age at hospital discharge” was added to the regression analyses of R-R interval, HF, LF, SBP, and DBP for the ELBW group, to account for possible differences in lung development and perinatal comorbidities. Age at hospital discharge was not directly associated with any of the autonomic parameters (all Ps > .55), and including it in the regression analyses did not change the significance of associations between birth weight and the autonomic variables in ELBW survivors.

†

Self-reported smoking status, alcohol use, and sleep quality were collected from both groups at the young adult assessment. The groups did not differ on any of these adult lifestyle variables (Ps > .19), and none of them was associated directly with the autonomic indices in either group (all Ps > .08), with 2 exceptions: in the NBW group, heavy alcohol use was associated with shorter resting R-R interval, accounting for 6% of the variance, B = −113.8 (49.7), P < .03, and sleep restlessness explained 3% of the variance in SBP, B = 3.05 (1.6), P < .06. Adding the lifestyle variables to the covariates did not change associations between birth weight and any of the autonomic parameters in either group.

References

Van Leeuwen

Lange

Betterman

Grönemeyer

Hatzman

. Fetal heart rate variability and complexity in the course of pregnancy. Early Hum Dev. 1999;54:259-269.

Andriessen

Oetomo

Peters

Vermeulen

Wijn

PFF

Blanco

. Baroreceptor reflex sensitivity in human neonates: the effect of post-menstrual age. J Physiol. 2005;568:333-341.

Sachis

Armstrong

Becker

Bryan

. Myelination of the human vagus nerve from 24 weeks postconceptional age to adolescence. J Neuropathol Exp Neurol. 1982;41:466-472.

Spassov

Curzi-Dascalova

Clairambault

. Heart rate and heart rate variability during sleep in small-for-gestational-age newborns. Pediatr Res. 1994;35:500-505.

Patural

Barthelemy

Pichot

. Birth prematurity determines prolonged autonomic nervous system immaturity. Clin Auton Res. 2004;14:391-395.

Witcombe

Yiallourou

Sands

Walker

Horne

RSC

. Preterm birth alters the maturation of baroreflex sensitivity in sleeping infants. Pediatrics. 2012;129:e89-e96.

Kramer

Platt

Wen

. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108:e35.

Patural

Pichot

Jaziri

. Autonomic cardiac control of very preterm newborns: a prolonged dysfunction. Early Hum Dev. 2008;84:681-687.

Yiallourou

Witcombe

Sands

Walker

Horne

RSC

. The development of autonomic cardiovascular control is altered by preterm birth. Early Hum Dev. 2013;89:145-152.

10.

Rakow

Katz-Salamon

Ericson

Edner

Vanpeé

. Decreased heart rate variability in children born with low birth weight. Pediatr Res. 2013;74:339-343.

11.

De Rogalski

Roche

Pichot

. Autonomic nervous system functioning in premature and full-term infants from theoretical term to 7 years. Auton Neurosci. 2007;136:105-109.

12.

Hovi

Andersson

Räikkönen

. Ambulatory blood pressure in young adults with very low birth weight. J Pediatr. 2010;156:54-59.

13.

Doyle

Faber

Callanan

Morley

. Blood pressure in late adolescence and very low birth weight. Pediatrics. 2003;111:252-257.

14.

Martinez-Aguayo

Aglony

Bancalari

. Birth weight is inversely associated with blood pressure and serum aldosterone and cortisol levels in children. Clin Endocrinol. 2012;76:713-718.

15.

Bonamy

AKE

Källén

Norman

. High blood pressure in 2.5-year-old children born extremely preterm. Pediatrics. 2005;129:1199-1204.

16.

Bassareo

Fanos

Puddu

Cadeddu

Balzarini

Mercuro

. Significant QT interval prolongation and long QT in young adult ex-preterm newborns with extremely low birth weight. J Matern Fetal Neonatal Med. 2011;24:1115-1118.

17.

Skilton

Viikari

JSA

Juonala

. Fetal growth and preterm birth influence cardiovascular risk factors and arterial health in young adults: the Cardiovascular Risk in Young Finns study. Arterioscler Thromb Vasc Biol. 2011;31:2975-2981.

18.

Mzayak

Sherwin

Hughes

. The association of birth weight with arterial stiffness at mid-adulthood: the Bogalusa Heart Study. J Epidemiol Community Health. 2009;63:729-733.

19.

Rossi

Laurent Tauzin

Marchand

Boussuges

Gaudart

Frances

. Respective roles of preterm birth and fetal growth restriction in blood pressure and arterial stiffness in adolescence. J Adolesc Health. 2011;48:520-522.

20.

Johansson

Norman

Legnevall

Dalmaz

Lagercrantz

Vanpée

. Increased catecholamines and heart rate in children with low birth weight: perinatal contributions to sympathoadrenal overactivity. J Intern Med. 2007;261:480-487.

21.

Law

Schiell

Newsome

. Fetal, infant, and childhood growth and adult blood pressure: a longitudinal study from birth to 22 years of age. Circulation. 2002;105:1088-1092.

22.

Eriksson

Forsén

Tuomilehto

Osmond

Barker

. Fetal and childhood growth and hypertension in adult life. Hypertension. 2000;36:790-794.

23.

Dalziel

Parag

Rodgers

Harding

. Cardiovascular risk factors at age 30 following pre-term birth. Int J Epidemiol. 2007;36:907-915.

24.

Svedenkrans

Henckel

Kowalski

Norman

Bohlin

. Long-term impact of preterm birth on exercise capacity in healthy young men: a national population-based cohort study. PLoS One. 2013;8:e80869.

25.

Duryea

Hawkins

McIntire

Casey

Leveno

. A revised birth weight reference for the United States. Obstet Gynecol. 2013;124:16-22.

26.

Kajantie

Osmond

Barker

DJP

Forsén

Phillips

DIW

Eriksson

. Size at birth as a predictor of mortality in adulthood: a follow-up of 350,000 person-years. Int J Epidemiol. 2005;34:655-663.

27.

Altman

Royston

. The cost of dichotomising continuous variables. BMJ. 2006;332:1080.

28.

Saigal

Stoskopf

Boyle

. Comparison of current health, functional limitations, and health care use of young adults who were born with extremely low birth weight and normal birth weight. Pediatrics. 2007;119:e562-e573.

29.

Thayer

Lane

. The importance of inhibition in dynamical systems models of emotion and neurobiology. Behav Brain Sci. 2005;28:218-219.

30.

Dekker

Crow

Folsom

. Low heart rate variability in a 2-minute rhythm strip predicts risk of coronary heart disease and mortality from several causes. Circulation. 2000;102:1239-1244.

31.

Levy

. Neural control of cardiac function. Baillieres Clin Neurol. 1997;6:227-244.

32.

Berntson

Bigger

Jr Eckberg

. Heart rate variability: origins, methods and interpretive caveats. Psychophysiology. 1997;34:623-648.

33.

Goldstein

Bentho

Park

Sharabi

. Low-frequency power of heart rate variability is not a measure of cardiac sympathetic tone but may be a measure of modulation of cardiac autonomic outflows by baroreflexes. Exp Physiol. 2011;96:1255-1261.

34.

DiPietro

Costigan

Pressman

Doussard-Roosevelt

. Antenatal origins of individual differences in heart rate. Dev Psychobiol. 2000;37:221-228.

35.

Propper

Moore

Mills-Koonce

. Gene-environment contributions to the development of infant vagal reactivity: the interaction of dopamine and maternal sensitivity. Child Dev. 2008;79:1377-1394.

36.

Pitzalis

Mastropasqua

Massari

Forleo

Di Maggio

Passantino

. Short- and long-term reproducibility of time and frequency domain heart rate variability measurements in normal subjects. Cardiovasc Res. 1996;32:226-233.

37.

Gournay

Drouin

Rozé

. Development of baroreflex control in preterm and full-term infants. Arch Dis Child. 2002;86:F151-F154.

38.

Thayer

Lane

. The role of vagal function in the risk for cardiovascular disease and mortality. Biol Psychol. 2007;74:224-242.

39.

Friedman

. An autonomic flexibility-neurovisceral integration model of anxiety and cardiac vagal tone. Biol Psychol. 2007;74:185-199.

40.

Gianaros

Van der Veen

Jennings

. Regional cerebral blood flow correlates with heart period and high-frequency heart period variability during working memory tasks: implications for cortical and subcortical regulation of cardiac autonomic activity. Psychophysiology. 2004;41:521-530.

41.

Thayer

Hansen

Saus-Rose

Johnsen

. Heart rate variability prefrontal neural function and cognitive performance: the neurovisceral integration perspective on self-regulation adaptation and health. Ann Behav Med. 2009;37:141-153.

42.

Porges

. Autonomic regulation and attention. In: Campbell

Hayne

Richardson

, eds. Attention and Information Processing in Infants and Adults: Perspectives From Human and Animal Research. Hillsdale, NJ: Lawrence Erlbaum; 1992:201-223.

43.

Saigal

Stoskopf

Boyle

Streiner

Pinelli

Paneth

. Transition of extremely low birth weight infants from adolecence to adulthood. JAMA. 2006;2925:667-675.

44.

Schmidt

Miskovic

Boyle

Saigal

. Shyness and timidity in adults who were born at extremely low birth weight. Pediatrics 2008;122: 181-187.

45.

Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement physiological interpretation and clinical use. Eur Heart J. 1996;17:354-381.

46.

Denver

Reed

Porges

. Methodological issues in the quantification of respiratory sinus arrhythmia. Biol Psychol. 2007;74:286-294.

47.

Quigley

Berntson

. Autonomic interactions and chronotropic control of the heart: heart period versus heart rate. Psychophysiology. 1996;33:605-611.

48.

Berntson

Stowell

. ECG artifacts and heart period variability: don’t miss a beat! Psychophysiology. 1998;35:127-132.

49.

Chambers

Allen

JJB

. Vagal tone as an indicator of treatment response in major depression. Psychophysiology. 2002;39:861-864.

50.

Lewis

Furman

McCool

Porges

. Statistical strategies to quantify respiratory sinus arrhythmia: are commonly used metrics equivalent? Biol Psychol. 2012;89:349-364.

51.

Regev

Lusky

Dolfin

Litmanovitz

Arnon

Reichman

. Excess mortality and morbidity among small-for-gestational-age premature infants: a population-based study. J Pediatr. 2003;143:186-191.

52.

Lenard

Studinger

Mersich

Kocsis

Kollai

. Maturation of cardiovagal autonomic function from childhood to young adult age. Circulation. 2004;110:2307-2312.

53.

Umetani

Singer

McCraty

Atkinson

. Twenty-four hour time domain heart rate variability and heart rate: relations to age and gender over nine decades. J Am Coll Cardiol. 1998;31:593-601.

54.

Dart

Kingwell

. Gender, sex hormones and autonomic nervous control of the cardiovascular system. Cardiovasc Res. 2002;53:678-687.

55.

Lampert

Ickovics

Horwitz

Lee

. Depressed autonomic nervous system function in African Americans and individuals of lower social class: a potential mechanism of race- and class-related disparities in health outcomes. Am Heart J. 2005;150:153-160.

56.

Paulson

Andersen

Munck

. Socio-economic status influences blood pressure control despite equal access to care. Fam Pract. 2012;29:503-510.

57.

Hollingshead

. Two-Factor Index of Social Position. New Haven, CT: Yale University Press; 1969.

58.

Molfino

Fiorentini

Tubani

Martuscelli

Fanelli

Laviano

. Body mass index is related to autonomic nervous system activity as measured by heart rate variability. Eur J Clin Nutr. 2009;63:1263-1265.

59.

Bensley

Stacy

De Matteo

Harding

Black

. Cardiac remodelling as a result of preterm birth: implications for future cardiovascular disease. Eur Heart J. 2010;31:2058-2066.

60.

Rudolph

. Myocardial growth before and after birth: clinical implications. Acta Paediatr. 2000;89:129-133.

61.

Lewandowski

Augustine

Lamata

. Preterm heart in adult life: cardiovascular magnetic resonance reveals distinct differences in left ventricular mass geometry and function. Circulation. 2013;127:197-206.

62.

Lewandowski

Bradlow

Augustine

. Right systolic ventricular dysfunction in young adults born preterm. Circulation. 2013;128:713-720.

63.

Tauzin

. Alterations in viscoelastic properties following premature birth may lead to hypertension and cardiac disease development later in life. Acta Pediatr. 2015;104:19-26.

64.

Martyn

Greenwald

. Impaired synthesis of elastin in walls of aorta and major conduit arteries during early development as an initiating event in pathogenesis of hypertension. Lancet. 1997;350:953-955.

65.

Martyn

Greenwald

. A hypothesis about a mechanism for the programming of blood pressure and vascular disease in early life. Clin Exp Pharmacol Physiol. 2001;28:948-951.

66.

Van Lith

Visser

GHA

Mantingh

Beekhuis

. Fetal heart rate in early pregnancy and chromosomal disorders. Br J Obstet Gynecol. 1992;99:741-744.

67.

Hoyer

Nowack

Bauer

. Fetal development of complex autonomic control evaluated from multiscale heart rate patterns. Am J Physiol Regul Integr Comp Physiol. 2013;304:R383-R392.

68.

Shinebourne

Vapaavuori

Williams

Heymann

Rudolph

. Development of baroreflex activity in unanesthetized fetal and neonatal lambs. Circ Res. 1972;31:210-218.

69.

Frasch

Müller

Wicher

. Fetal body weight and the development of the control of the cardiovascular system in fetal sheep. J Physiol. 2007;579:893-907.

70.

Berry

Jaquiery

Oliver

Harding

Bloomfield

. Preterm birth has sex-specific effects on autonomic modulation of heart rate variability in adult sheep. PLoS One. 2013;8:e85468.

71.

Pyhälä

Räikkonen

Feldt

. Blood pressure responses to psychosocial stress in young adults with very low birth weight: Helsinki study of very low birth weight adults. Pediatrics. 2009;123:731-734.

72.

Bonamy

AKE

Endito

Martin

Andolf

Sedin

Norman

. Preterm birth contributes to increased vascular resistance and higher blood pressure in adolescent girls. Pediatr Res. 2005;58:845-849.

73.

Moore

Cockington

Ryan

Robinson

. The relationship between birth weight and blood pressure amplifies from childhood to adulthood. J Hypertens. 1999;17:883-888.