Early life factors and their relevance to intima-media thickness of the common carotid artery in early adulthood.

BACKGROUND: Early life factors may predispose an offspring to cardiovascular disease in later life; relevance of these associations may extend to ‟healthy" people in Western populations. We examined the prospective associations between early life factors and adult carotid intima-media thickness (IMT), a surrogate marker of atherosclerosis, in a healthy German population.
METHODS: We studied term participants (n = 265) of the DONALD Study, with bilateral sonographic measurements of IMT (4-8 measurements on both left and right carotid artery) at age 18-40 years and prospectively collected data on early life factors (maternal and paternal age at child birth, birth weight, gestational weight gain and full breastfeeding (>17weeks). Mean IMT values were averaged from mean values of both sides. Associations between early life factors and adult IMT were analyzed using multivariable linear regression models with adjustment for potential confounders.
RESULTS: Adult mean IMT was 0.56mm, SD 0.03, (range: 0.41 mm-0.78 mm). Maternal age at child birth was of relevance for adult IMT, which was sex specific: Advanced maternal age at child birth was associated with an increased adult IMT among female offspring only (β 0.03, SE 0.009 mm/decade, P = 0.003), this was not affected by adult waist circumference, BMI or blood pressure. Other early life factors were not relevant for IMT levels in males and females.
CONCLUSION: This study suggests that advanced maternal age at child birth is of prospective relevance for adult IMT levels in a healthy German population and this association may be of adverse relevance for females only.

1. Introduction

Interest in research on early life exposures as possible determinants of disease in later life greatly increased since the discovery of the Barker hypothesis that cardiovascular disease (CVD) has its origins in early life [1]. The atherosclerotic process begins early in life, i.e. many years before cardiovascular complications develop later in life [2]. Fatty streaks which are initial lesions of atherosclerosis have been observed in arteries of children less than a year old which increases with age [3]. Also increases in carotid intima media thickness (IMT) and endothelial dysfunction have been suggested as preliminary indications of atherosclerotic plaque development [4, 5] Similarly, alterations in the IMT have been indicated as a marker of subclinical atherosclerosis [6-8] with a high IMT shown to correlate with CV risk factors [6, 9, 10], and to predict CVD [10, 11].

CVD has been associated with low birthweight mostly in populations exposed to maternal undernutrition during gestation [12, 13]. Though the actual pathophysiological mechanisms underlying the association is not fully clarified, it has been associated with the theory of developmental plasticity, i.e. that the developing offspring’s system is plastic and sensitive to the nutritional, hormonal and the metabolic milieu in utero, resulting in various physiological or morphological states due to different conditions during development [14, 15]. A decrease in cell numbers in organs due to changes in the intrauterine environment could also account for CVD in adulthood [15, 16].

It is hence plausible that a range of exposures in early life may be associated with later CVD and thus also with adult IMT. The relevance of these associations may also extend to ‟healthy” people in Western populations due to the presence of such exposures in this population. Specifically, four groups of early life factors remain to be addressed in these populations: (i) Birthweight, which reflects fetal in utero environment, a potential predictor for which evidence is now also emerging in relation to IMT [17-19]. (ii) gestational weight gain may give rise to an obesogenic intrauterine environment specifically among overweight or obese mothers [20], however, there is no data on its relevance for later IMT; (iii) maternal age, since women are now giving birth at a more diverse age range, with a tendency towards older age in Western countries; (iv) early postnatal nutrition, specifically breastfeeding, which has been related to IMT, however producing controversial results [21-23].

To address these four sets of hypotheses we examined the prospective relevance of (i) indicators of intrauterine growth such as birth weight, birth weight-for-gestational-age (i.e. adequate, small and large for gestational age), (ii) pregnancy duration and gestational weight gain, (iii) parental age and (iv) full breastfeeding for adult IMT as a surrogate of CVD among healthy term-born German participants.

2. Methods

2.1. Study population

The Dortmund Nutritional and Anthropometric Longitudinally Designed (DONALD Study), is a continual, open cohort study undertaken in Dortmund, Germany. Since its commencement in 1985, elaborate records on diet, growth, development, and metabolism has been gathered from over 1,700 children between infancy and adulthood. About 35–40 infants are newly enrolled each year while initial examination commences at the age of 3–6 months, afterwards each child returns for 2–3 more visits in the first year, 2 in the second year and then once yearly until adulthood.

The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki. The study was approved by the Ethics Committee of the University of Bonn. The data protection officer is Dr. Jörg Hartmann and requests for data access may be sent to the local data protection coordinator Heinz Rinke, email: rinke@uni-bonn.de at the University of Bonn. All examinations are performed with written consent of parent and adult participant [24-28].

The children who were initially recruited for the DONALD Study differed considerably in age and prospectively collected data on breastfeeding was not always available. Follow-up into adulthood was not planned at the inception of the study. Since 2004 participants are invited to return for further visits at ages 18, 21, 25, 30, 35 etc. However, not all participants followed this invitation. In addition, due to the open cohort design, many DONALD participants had not yet reached young adulthood by the time of this analysis. IMT measurements are offered to adolescents and adult participants since 2008. In this analysis only IMT measurements in adulthood (≥18 years of age) are used. Mean follow-up until IMT measurement is equivalent to the mean age at IMT-measurement. 607 IMT measurements were available with two persons excluded due to the presence of plaques and stenosis in their measurement. 58 others did not have a minimum of four measurements each on the right and left common carotid artery to be included in the analysis whilst 178 persons were not considered because the images did not fulfill the quality control criteria (see below). Among the remaining 369 persons with acceptable IMT measurements, data from 349 persons were considered who were born term (37–42 weeks of gestation) singletons with a birthweight ˃ = 2500g. A further 84 persons were excluded because they did not fulfill the following minimum requirements: Parents had to have provided information on maternal age at birth, only available for 262, paternal age at birth, only available for 256, birth year, birth weight, gestational weight gain only available for 258 and gestational duration. Hence the sample considered for this analysis includes 265 participants with information on IMT collected between 2009 and 2014. See Fig 1, for the sample size of early life factors and for relevant covariates see Table 1.

Fig 1

Participant flowchart diagram for IMT and early life factors.

Table 1

Characteristics of participants in early life and young adulthood.

Variables		Males		Females
	N		N
Measurements in young adulthood
Average IMT (mm)1	120	0.57 (0.06)	145	0.55 (0.05)
Age at IMT measurement (yrs)	120	23.3 (5.7)	145	23.9 (5.1)
Waist circumference (cm)	120	83.6 (8.3)	145	75.3 (8.3)
BMI at IMT measurement (kg/m2)	104	24.2 (3.3)	142	23 (4.2)
Systolic blood pressure (mm Hg)	103	121.3 (10.7)	140	110.5 (9.8)
Diastolic blood pressure (mm Hg)	103	75.9 (9.4)	140	72.8 (8.1)
Participation in sport (yes/no) 2	91	88 (96.7%)	123	118 (96.0%)
Energy expenditure (kcal) 3	55	641 (263)	68	363 (260)
Early life factors
Maternal age at child birth (yrs)	120	30.6 (4.2)	142	30.3 (4.2)
Paternal age at child birth (yrs)	119	33.3 (5.1)	137	33.2 (5.0)
Pregnancy duration (wks)	120	40 (40, 41)	145	40 (39, 41)
Gestational weight gain (kg)	117	12.9 (4.0)	141	12.9 (3.6)
High gestational weight gain n (%)4	117	20 (17.1%)	141	20 (14.2%)
Birthweight (g)	120	3583 (443)	145	3411 (437)
Birth weight < 3000 (g)		5 (4.2%)		28 (19.3%)
Birth weight ≥ 3000 to ≤ 4000 (g)		93 (77.5%)		104 (71.7%)
Birth weight > 4000 (g)		22 (18.3%)		13 (9%)
Birth weight by gestation age5	120		145
SGA		12 (10%)		16 (11.0%)
AGA		92 (76.7%)		111 (76.6%)
LGA		16 (13.3%)		18 (12.4%)
Full breastfeeding	104		130
Never (0–2 weeks)		29 (27.8%)		37 (28.4%)
Short duration (3–17 weeks)		40 (38.5%)		49 (37.7%)
Long duration (˃17 weeks)		35 (33.7%)		44 (33.9%)
Additional potential confounders
Birth year	120	1989 (1985, 1992)	145	1988 (1985, 1991)
Firstborn status (yes/no)	104	58 (55.8%)	129	74 (57.4%)
Maternal overweight (yes/no)6	116	38 (32.8%)	141	41 (29.1%)
High paternal educational status (yes/no)7	119	70 (58.8%)	136	73 (53.7%)
Smokers in the household (yes/no)	106	37 (34.9%)	134	55 (41.0%)

Values are presented as means (SD) medians (IQR) or frequencies (percentage).

AGA: appropriate for gestational age, LGA: large for gestational age, SGA: small for gestational age.

AGA, LGA and SGA defined according to German sex-specific birth weight and length-for-gestational-age curves.

Average IMT: mean of intima media thickness (IMT).

Participation in organized or unorganized sport: (yes/no).

Estimated energy expenditure during participation in organized or unorganized sport.

High gestational weight gain: yes (>16kg), no (≤16kg).

Birth weight by gestation age

Maternal overweight: yes (≥ 25kg), no (<25kg).

High educational status: yes (≥12yrs of school attendance), no (<12yrs of school attendance).

2.2. Early life exposures

Child birth and maternal characteristics were extracted from the “Mutterpass,” a standard document given to all pregnant women in Germany. Gestational duration is calculated according to the mother’s last menstrual period.

Maternal weight at first visit at the gynecologist during pregnancy and at the end of pregnancy weight were abstracted from the “Mutterpass,” and from these the gestational weight gain was computed.

Birth weight and birth length were recorded at birth. Birth weight-for-gestational-age is defined according to the German sex-specific birth weight and length-for-gestational-age curves [29]. Small-for-gestational-age (SGA) is defined as birth weight and length <10th percentile, and large-for-gestational-age (LGA) is defined as birth weight and length ˃90th percentile. All other infants were classified as appropriate-for gestational age (AGA).

Maternal and paternal age at the time of child birth is assessed at first visit.

Breastfeeding data was assessed upon the child’s admission to the study. During first visit either at 3 or 6 months the study pediatrician and/or dietitian enquired from the mothers the duration (in weeks) the infant had been fully breastfed (not given solid foods and no liquids daily except breast milk, tea, or water). If the child is still being fully breastfed, the length of breastfeeding is assessed at successive visits at ages 6, 9, 12 and 18 months until commencement of complementary feeding. The duration of feeding formula or solid foods is also assessed during the visits. A coherent check is conducted on all breast feeding information collected such as the recording of breast milk in 3-day dietary records and information acquired by the dietitians before analysis to minimize errors. From this information the duration (in weeks) of full breastfeeding is calculated [30].

2.3. Early adulthood outcome variable: Intima media thickness

Vascular conditions of the left and right common carotid artery (CA) were studied ultrasonographically using high-resolution technology. The Mindray DP3300, tragbares portable digital system was used for this study. The participants were measured in a supine position, head slightly to the right or left after having rested for 10 min. The start point of the measurement was at the beginning of the bifurcation at the left edge of the image with a horizontal vessel course. IMT was measured at 4 points 1 cm before the carotid bifurcation. Images were always taken in the systole. Two images were first taken each on the right and left CA on the participants and the images were frozen. Subsequently, measurements were taken at four measurement points on each image.

Quality control was carried out on all images and only images that met the criteria were used for analysis. The criteria for IMT measurement quality control are based on 1) clear representation of the Intima-Media-Complex of the “far wall “shown as echo-rich/echo-poor uninterrupted line, 2) echo-free imaging of the vessel lumen and clear separation of the intima from the lumen, 3) localization of the image section at the beginning of the bifurcation and 4) horizontal course of the vessel within the image. Individual measurement points were discarded, if they were set incorrectly (i.e. the point for measurement was set below or above the visible IMT lines). Mean IMT values were firstly averaged for the right and left sides (i.e. 4–8 measurements), and then an overall mean was calculated from the two averages.

All measurements were performed by the study physicians. Each year, a quality control of IMT measurement by the physicians is carried out. Coefficient of variation (CV) which considers the precision of the measurements within and between the physicians is computed and from 2009 to 2015, the values are CVintra = 6.95 and CVinter = 3.70. An acceptable precision is given at a value less than 10.

2.4. Potential covariates

Anthropometry of study participants were taken at each visit using standard protocol by trained nurses. The participants are dressed in only underwear and are barefooted. Recumbent length of children until 2 years of age is measured to the nearest 0.1 cm using a Harpenden (UK) stadiometer, whilst standing height is measured in children aged older than 2 years to the nearest 0.1 cm with a digital stadiometer (Harpenden Ltd., Crymych, UK). Body weight is measured to the nearest 100 g using an electronic scale (Seca 753E; Seca Weighing and Measuring Systems, Hamburg, Germany). Waist circumference is measured at the midpoint between the lower rib and iliac crest to the nearest 0.1 cm. The trained nurses who perform the measurements undergo quality control, conducted with healthy young adult volunteers [28]. This same measurement procedure is used to measure anthropometry of parents at regular intervals.

The number of smokers in the household was enquired and from this smoking exposure was assessed. The years of schooling was also enquired and from this a proxy of parental socioeconomic status was created. A high educational status is defined as (≥ 12 years of schooling).

2.5. Statistical analysis

All statistical analysis was conducted using SAS 9.4. Prospective association between early life factors and IMT during young adulthood were analyzed using multivariable linear regression models. IMT was adjusted for age and sex using the residual method.

To evaluate whether sex modifies the association between early life factors and IMT, an interaction analysis was carried out and if a significant sex difference existed, analysis was carried out separately for men and women. Interaction analysis indicated sex interactions for maternal age at child birth and breastfeeding (Pinteraction = 0.03 to 0.09).

Initial regression models (A) included IMT as the dependent continuous variable and individual inclusion of an early life predictor as the independent variable, adjusted for age at IMT measurement, sex and the physician measuring IMT.

Next, multivariable adjusted models (B) were constructed considering covariates individually for potential confounding in the models in a hierarchical manner [31]. Covariates which substantially modified the predictor–outcome associations by (≥10%) or significantly predicted the outcome were included in the final multivariable adjusted models.

These early life factors were considered as mutual potential covariates in this model (1) early life factors: birth weight (g) considered as both a continuous and categorical variable (i.e., <3000g, ≥3000g to ≤4000g and ˃4000g), birth weight-for-gestational-age as a three level categorical variables (AGA, SGA, LGA), gestational age (weeks), maternal and paternal age at child birth (years) were considered as continuous variables, gestational weight gain (kg) as a continuous and categorical variable (i.e.≤16kg and >16kg), breastfeeding for >2 weeks (Yes/No and >16 weeks (Yes/No), first born status (Yes/No) and birth year regressed on age at IMT measurement as a continuous variable. (2) Socioeconomic factors: paternal school education ≥12 years (Yes/No), presence of an overweight parent BMI≥25 kg/m2, (Yes/No), smokers in the household (Yes/No). Sensitivity analyses were conducted in subsamples who had provided either information on participation in sport (Yes/No) (n = 123) or on estimated energy expenditure during participation in sports (n = 68) in early adulthood, so as to account for potential confounding arising from adult physical activity levels.

Finally, four sets of conditional models were constructed adding adult waist circumference, adult BMI or adult systolic or diastolic blood pressure to the models, so as to investigate whether observed associations were partly attributable to these variables in adulthood.

Results from regression analysis are presented as adjusted least-square means (95% confidence interval (CI) by tertiles of the respective predictor while P-value is obtained from models using the predictors as continuous and categorical variables. Significance was determined at a p-value of 0.05.

3. Results

Early life and young adulthood characteristics of the participants in this analysis are presented in Table 1 according to sex. The minimum and maximum average IMT ranged from 0.41mm to 0.78mm. The mean age at IMT measurement was 23 years in males and 24 years in females. Participants lost to follow up differed slightly from those included in this analysis: mothers of males and females were younger i.e. 29.4 and 29.8 years, children were born earlier i.e. in 1987 and fewer offspring were fully breastfed for a long duration, i.e. 31% and 27% among male and female offspring (for details see S1 Table).

Paternal age at birth was not related to adult IMT (S2 Table). In multivariable analysis, increased maternal age at child birth was associated with an increased IMT among female offspring during young adulthood (P = 0.003, Fig 2), but not in males (P = 0.2, Fig 3) (S2 Table). These associations were not affected by adjusting for paternal age at birth. In addition, inclusion of adult waist circumference, BMI, systolic or diastolic blood pressure in separate conditional models did not affect the relationships (S3 Table).

Fig 2

Association between maternal age at child birth and intima media thickness.

Maternal age at child birth by tertiles of average IMT adjusted for age by the residual method in young adulthood among female participants. Data are means and 95% CI adjusted for adult age at IMT measurement, the physician taking the IMT measurement and birth year (residuals of birth year were calculated on age at IMT measurement), (n = 142).

Fig 3

Association between maternal age at child birth and intima media thickness.

Maternal age at child birth by tertiles of average IMT adjusted for age by the residual method in young adulthood among male participants. Data are means and 95% CI adjusted for adult age at IMT measurement, the physician taking the IMT measurement and birth year (residuals of birth year were calculated on age at IMT measurement), (n = 120).

Sensitivity analyses in the subsample of females, for whom data on participation in sport (n = 123) or data on estimated energy expenditure during participation in sport (n = 68) was available, additional consideration of these variables did not affect the association of maternal age at birth with IMT (S4 Table).

In multivariable analysis on the relevance of full breastfeeding for adult IMT, there was no association in females (P = 0.1, model A and B, Table 2). In males, there was a trend for an association between full breastfeeding and IMT in young adulthood (P = 0.09, model B), which remained when considering adult waist circumference, BMI, systolic or diastolic blood pressure in separate conditional models (data not shown).

Table 2

Association of full breastfeeding categories and IMT in young adulthood among females and males.

		Average IMT (mm)
	N	Full breastfeeding categories		P trend
		0–17 wks	>17 wks
Females	130	86 (76.7)1	44 (23.3)1
Model A		0.55 (0.54, 0.56)2	0.56 (0.55, 0.58)2	0.1
Model B3		0.55 (0.54, 0.56)	0.56 (0.55, 0.58)	0.1
Males	104	69 (61.1)	35 (38.9)
Model A		0.57 (0.56, 0.58)	0.56 (0.54, 0.58)	0.6
Model B4		0.57 (0.56, 0.59)	0.55 (0.54, 0.57)	0.0917

Average IMT: average of means of right and left side intima media thickness (IMT).

Linear trends (P trend) were obtained in linear regression models with IMT as a continuous variable.

Values are frequencies (percentages) of breastfeeding durations.

Values are adjusted least squares means (95% Confidence Interval (CIs)) of IMT.

Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement.

Model B among females additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement) and maternal age at child birth.

Model B among males additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

When analyses were repeated in subsamples of males with data on participation in sport (n = 91) or estimated energy expenditure during participation in sport (n = 55) in adulthood, full breastfeeding was no longer associated with adult IMT levels, irrespective of considering adult physical activity (data not shown).

There was no association of IMT with other early life factors i.e. pregnancy duration, gestational weight gain, birthweight, birthweight according to gestational age (S5 Table and S6 Table).

4. Discussion

This study indicates that older maternal age at child birth is associated with an increased IMT of the offspring in young adulthood. Whereas increased maternal age at child birth was associated with an increased IMT in female offspring’s, this was not evident in males.

Other early life factors were not associated with IMT in this study. Birthweight and birthweight according to gestational age were not associated with IMT, probably because term infants with mostly adequate birthweight were included in our study.

It is likely advanced maternal age at child birth may programme an offspring for the onset of cardiovascular disease later in life: maternal age at birth was found to be associated with higher infant systolic blood pressure [32] and impaired adult glucose metabolism [33]. In our study, IMT was not associated with paternal educational status (data not shown) and the relevance of maternal age at birth for adult IMT was not influenced by paternal age at birth, hence it is likely the association with maternal age is due to the intra uterine environment rather than socioeconomic factors.

In line with our study, a study among Chinese participants reported that older maternal age at delivery was adversely associated with IMT among females only [34], with an effect size comparable to ours (ß: of 0.02mm per decade compared to ß: 0.03mm per decade) in our study. Mothers in that cohort were approximately 4 years younger than mothers in our cohort, which may explain the slightly more pronounced effect size in our cohort.

The overall mechanisms between maternal age at child birth and IMT of offspring are not known, yet state of health in mothers advanced in age and placental nutrition could affect fetal development due to variations in uterine artery flow or hormone synthesis in comparison to younger mothers. Likewise, mothers advanced in age are prone to higher levels of CVD risk factors such as higher blood pressure, dyslipidemia or increased oxidative stress [34, 35]. A study using a rat model of advanced maternal age, reported that pregnancy complications were partly due to development of maternal hypertension and altered vascular function in the aged female rats [36]. Hence, both the study in humans and animals suggest that advanced maternal age at child birth may programme an offspring for cardiovascular disease later in life.

In terms of the sex-specific nature of our results it should firstly be noted that men and women have similar CVD risk factors, however, there are considerable variations in the first manifestation and in clinical signs [37]. There is evidence suggesting that IMT among females are more vulnerable to metabolic disorder: Insulin resistance is related to IMT and atherosclerosis solely among females [34], and blood glucose and triglycerides levels associate strongly with IMT among females only [38]. Such sex differences could be due to hormonal differences as well as genetics [39]. Specifically, sex chromosomes may be involved in sex variations in disease development [40]. The X chromosomes in females carries clues related to inflammation [41] and the unstable nature of the X chromosome in the outer blood cells (due to loss of the second X chromosome) is associated with autoimmune diseases as well as cardiovascular diseases in females [42, 43].

Secondly, mechanisms linking early life factors to adult IMT levels in a sex-specific way are not clear, yet female placentas have been found to be more responsive to maternal emotional distress: in female fetuses, high emotional distress of the mothers was associated with lower mRNA levels of fetal genes which prevent glucocorticoid transfer to the fetuses as well as higher mRNA levels of placental glucocorticoid receptors. In turn, in male placentas, high emotional distress was associated with high mRNA levels of genes which raise glucocorticoid inactivation providing a protective effect [44]. Animal studies have shown that high glucocorticoid exposure can result in structural alterations that can hamper heart function [45-47]. Additionally, increased maternal emotional distress was associated with high insulin like growth factor IGF2 and IGF2R mRNA levels in female but not in male placenta [44].

The strength of this study is its long follow up of participants and the fact that exposure variables were collected prospectively; careful consideration was given to the prospective assessment of potential confounders.

The study is limited by its observational nature, thus any conclusion drawn should be done with circumspection and the relatively small sample size. Attrition bias is a possibility, yet those lost to follow-up differed only slightly in their early life characteristics from those included in the analysis. Participants in the DONALD study are from a relatively high educational and socioeconomic status, so the results cannot be uncritically extrapolated to other socioeconomic (sub) populations. Data on physical activity in young adulthood were only available for subsamples, yet their consideration did not change our main findings. Nonetheless, the non-availability of behavioral variables for all participants as well as variables in other moments throughout childhood and adolescence is a limitation of our study.

Conclusion

In conclusion, our study suggests a sex specific association between older maternal age at child birth and increased IMT in early adulthood among females.

Characteristics of participant’s loss to follow up in adulthood.

Values are presented as means (SD), medians (IQR) or frequencies (percentage). Full breastfeeding defined as breast milk including water given to the child.

(DOCX)

Click here for additional data file.

Association of maternal or paternal age at child birth and IMT in young adulthood.

Average IMT: average of means of right and left side intima media thickness (IMT). T: tertile, n: sample size in tertile. Linear trends (P trend) were obtained in linear regression models with IMT as a continuous variable. Values are medians (25th, 75th percentiles) of early life factors. Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement. Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

(DOCX)

Click here for additional data file.

Conditional models for the association of maternal age at child birth with adult IMT in females.

Average IMT: average of means of right and left side intima media thickness (IMT). T: tertile, n: sample size in tertile. Linear trends (P trend) were obtained in linear regression models with IMT as a continuous variable. Values are medians (25th, 75th percentiles) of maternal age at child birth. Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement. Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

(DOCX)

Click here for additional data file.

Association of maternal age at child birth and IMT in young adulthood among females with data on physical activity.

Average IMT: average of means of right and left side intima media thickness (IMT). T: tertile, n: sample size in tertile. Linear trends (P trend) were obtained in linear regression models with IMT as a continuous variable. Values are medians (25th, 75th percentiles) of maternal age at child birth. Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement. Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

(DOCX)

Click here for additional data file.

Association of pregnancy duration or gestational weight gain with IMT in young adulthood.

Average IMT: average of means of right and left side intima media thickness (IMT). T: tertile, n: sample size in tertile. Linear trends (P trend) were obtained in linear regression models with IMT as a continuous variable. 1p values less than 0.025 are considered significant according to Bonferroni adjustment. Values are medians (25th, 75th percentiles) of early life factors. Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement. Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

(DOCX)

Click here for additional data file.

Association of birthweight for gestational age or birthweight and adult IMT.

AGA: appropriate for gestational age, LGA: large for gestational age, SGA: small for gestational age. AGA, LGA and SGA defined according to German sex-specific birth weight and length-for-gestational-age curves. Linear trends (P difference) were obtained in linear regression models with IMT as a continuous variable and birthweight according to gestational age as a 3 level categorical variable (0 = SGA; 1 = AGA; 2 = LGA with AGA set as the reference category in the models). 1p difference and trend less than 0.025 are considered significant according to Bonferroni adjustment. Values are frequencies (percentages) of birthweight according to gestational age. Values are adjusted least squares means (95% CIs) of IMT. Values are medians (25th, 75th percentiles) of birthweight. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement. Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

(DOCX)

Click here for additional data file.

20 Jan 2020

PONE-D-19-30815

Early life factors and their relevance to intima-media thickness of the common carotid artery in early adulthood

PLOS ONE

Dear Dr. Buyken,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As you will recognize from the comments of the reviewers major points of critique were raised especially regarding study design and methodology as well as presentation of results.

We would appreciate receiving your revised manuscript within 2 months. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Rudolf Kirchmair

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Acknowledgments Section of your manuscript:

The DONALD Study is supported by the Ministry of Innovation, Science,

Research and Technology of the State of North Rhine-Westphalia, Germany. Juliana Nyasordzi is

a PhD candidate co-sponsored by the government of Ghana (Ministry of Education) and the

German government (Deutscher Akademischer Austauschdienst German Academic Exchange

Service (DAAD).

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

The author(s) received no specific funding for this work.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

3. We noticed you have some minor occurrence(s) of overlapping text with the following previous publication(s), which needs to be addressed:

https://doi.org/10.3390/nu10040488

https://doi.org/10.1038/oby.2007.57

https://doi.org/10.1016/j.atherosclerosis.2014.09.027

https://doi.org/10.1007/s00394-016-1286-x

https://doi.org/10.1093/ajcn/84.6.1449

https://doi.org/10.3945/ajcn.2009.28259

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the Methods section. Further consideration is dependent on these concerns being addressed.

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This was a long follow-up study on early life factors and their relevance to early adulthood IMT German participants. None of the early life factors was associated with IMT except the maternal age only among females. The manuscript gives insight on a major public health problem; however, some concerns exist when interpreting the results.

1- It is not clear what the aim of study was. Whether constructing a prediction model of IMT based on early life factors or investigating the association of a specific factor with IMT. These two approaches are totally different in the statistical analysis. The manuscript should be unified based on its objective.

2- Too many factors were investigated with IMT. Multiple testing (multiplicity) is a common phenomenon and would lead to probability of false positive results. How the authors considered this event?

3- Strong evidence has been shown that one of the important factors determining the IMT is physical activity. The current research lacks measuring physical activity. Those with higher IMT may have lower physical activity. And we cannot relate the higher IMT to only maternal age.

4- Nothing has been mentioned on the characteristics of loss to follow-up participants. The high rate of loss to follow-up may confound the results.

5- The mean/median follow-up time was not stated. As authors declared that inclusion of participants in the cohort and IMT measurement were conducted both in different ages of participants, it more logical to use the survival analysis and include censoring time.

6- Birth year really does not have any added value. The range of the inclusion of participants was less than a decade. Unless investigators believe that something had happened during the recruitment years in the situation.

7- Blood pressure and body mass index have main effect on atherosclerosis in particular carotid IMT. Were these factors considered during analysis or not?

8- Does testing the interaction of sex and early life factors on early adulthood IMT sound biologically? The discussion actually did not explain this main finding from the early life point of view. The sex difference finding of studies in adult age was mentioned. What are the probable mechanisms for the sex difference effect of early life factors such as maternal age?

9- It is not clear why adulthood waist circumference was adjusted. There are many other important factors in adulthood, which affect adulthood IMT.

10- For non-significant findings, it is not necessary to discuss the effect of early life factors on IMT. More importantly, the explanation of non-significant results is required.

11- To test the trend of baseline characteristics, analysis of variance is not appropriate. It is mainly shows the difference and is used for comparison.

Reviewer #2: Dear Editor and Authors,

thank you for the opportunity to review this nice manuscript. I have some comments in order to improve the manuscript, but in general perspective, the manuscript brings relevant data about the issue.

Introduction: In general, the section is well organized and provides all the relevant details that the readers need to understand the background leading the authors to perform this research. My only comment is the absence of any data about other relevant behaviors / factors during adolescence (also a sensitive period to development of cardiovascular diseases). I will bring this aspect into the light later in the discussion section as well.

Methods

The section is well described. I do not have many comments. I only comment is to insert a flowchart describing all variables considered, as well as the the time point where those variables were collected.

Results

The section is particularly confusing. I suggest a different approach to present the results.

Table 1: it would be nice to compare all variables according to sex, mainly because the findings seems sex-dependent.

The comparison according to IMT tertile seems a weak stratification (there is no data about the number of participants with altered values for IMT). Even a few number, split out the sample according altered and normal IMT value seems more appropriate.

Is there any relationship between maternal age and the other early life variables considered (eg. low birthweight)? Due to the relevance of maternal age in your findings, a table describing that would welcome.

Discussion: The section need to the reduced in size. There was only one significant association, but there are four pages of discussion. Please, focus only on significant findings. Moreover, the ways the authors built the sections leads the reader to believe that behavioral variables in other moments of life are not relevant into this phenomenon. For instance, IMT in adulthood is affected by sports participation during adolescence (Werneck et al. 2018). Please, I would suggest to highlight the relevance of these data and also its absence in the models created.

References

Werneck AO, Lima MCS, Agostinete RR, Silva DR, Turi-Lynch BC, Codogno JS, Fernandes RA. Association between Sports Participation in Early Life and Arterial Intima-Media Thickness among Adults. Medicina (Kaunas). 2018;54(5). pii: E85. doi: 10.3390/medicina54050085.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Mar 2020

POINT-BY-POINT RESPONSE TO EDITORIAL BOARD MEMBER’S AND REVIEWERS COMMENTS

Manuscript: [PONE-D-19-30815] - [EMID:15f54e8fc6e19a2a]

Title of paper: Early life factors and their relevance to intima-media thickness of the common carotid artery in early adulthood.

Authors: Juliana Nyasordzi, Katharina Penczynski, Thomas Remer, Anette E.

Buyken.

General Comments

We cherish the suggestions made by the editor and reviewers all in an attempt to improve the quality of the manuscript. We have carefully considered the concerns raised by the editor and reviewers and amended the manuscript appropriately.

Editor’s Query 1

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response to EQ1

We have ensured that the manuscript meets PLOS ONE’S style requirements.

Editor’s Query 2

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

The author(s) received no specific funding for this work.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Response to EQ2

The funding information has been removed from the manuscript and is reported in the funding statement section of the online submission form as follows: The DONALD Study is supported by the Ministry of Innovation, Science, Research and Technology of the State of North Rhine-Westphalia, Germany. Juliana Nyasordzi is a PhD candidate co-sponsored by the government of Ghana (Ministry of Education) and the German government (Deutscher Akademischer Austausch-dienst German Academic Exchange Service (DAAD).

The authors declare no conflict of interest. The funders have no role in the design of the study, collection, analyses and interpretation of the data, the writing of the manuscript, and in the decision to submit the findings for publication.

Editor’s Query 3

We noticed you have some minor occurrence(s) of overlapping text with the following previous publication(s), which needs to be addressed:

https://doi.org/10.3390/nu10040488

https://doi.org/10.1038/oby.2007.57

https://doi.org/10.1016/j.atherosclerosis.2014.09.027

https://doi.org/10.1007/s00394-016-1286-x

https://doi.org/10.1093/ajcn/84.6.1449

https://doi.org/10.3945/ajcn.2009.28259

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the Methods section. Further consideration is dependent on these concerns being addressed.

Response to EQ3

The overlapping texts have been reworded accordingly and duly cited.

This has resulted in a complete rewording of paragraphs � see lines 30-31, 39-40, 49-56, 57-64, 85-87, 92-95, 98-105, 112-114, 116-117, 119, 122, 134-143, 169-173 in the manuscript.

In addition, multiple minor changes of the text were performed throughout the manuscript.

All changes are highlighted in the marked up copy in red.

Editor’s Query 4

We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Response to EQ4

In response to the General data protection regulation issued by the European Union, the official data policy has been adapted by the DONALD study group including the decision that datasets can generally only be obtained upon request. The main reason for this being that the study is ongoing (i.e. data cannot be anonymized) and due to its comparatively small sample size there is a higher risk for identification of persons even in pseudonomized data sets. This regulation has been developed as specified by the data protection officer of the University of Bonn Dr. Jörg Hartmann.

Our text dealing with this now reads:

Data from this study are available on request. The DONALD Study is still ongoing and the comparably small sample requires specific precautions to avoid potentially identifying participant information. This has been imposed by the data protection officer of the University of Bonn as an official ethical restriction. Requests for data access may be sent to the local data protection coordinator Heinz Rinke, email: rinke@uni-bonn.de at the DONALD Study Dortmund of the University of Bonn.

Reviewers’ comments to the Author:

Reviewer #1 (Remarks to the Author):

Query 1

It is not clear what the aim of study was. Whether constructing a prediction model of IMT based on early life factors or investigating the association of a specific factor with IMT. These two approaches are totally different in the statistical analysis. The manuscript should be unified based on its objective.

Response 1

We apologize for not having been clear with respect to our aims. We were not interested in constructing a prediction model of IMT. Instead, our aim was to assess the relevance of specific early life factors for adult intima media thickness (IMT), a surrogate of cardiovascular disease (CVD). Our hypothesis was that it is plausible that a range of early life factors could be associated with CVD. Specifically, we had four domains of hypotheses which we tested for,

1. Early life factors related to indicators of intrauterine growth (i.e. birth weight, birth weight-for-gestational-age (SGA, AGA and LGA) and their association with IMT.

2. Early life factors related to pregnancy (i.e. pregnancy duration and gestational weight gain) and their association with IMT.

3. Early life factors related to parental age at child birth (i.e. maternal and paternal age) and their association with IMT.

4. Postnatal nutrition (i.e. breastfeeding) and its association with IMT.

The aim of the study in the manuscript has been reworded to reflect these four domains of hypotheses tested. Please see lines 69-79 in the manuscript.

Query 2

Too many factors were investigated with IMT. Multiple testing (multiplicity) is a common phenomenon and would lead to probability of false positive results. How the authors considered this event?

Response 2

We agree that multiple testing is a concern, enhancing the probability of false positive results.

As outlined above, we did however address four different sets of hypotheses, i.e. for each of these different mechanisms have been proposed.

For the first (intrauterine growth) and the second (pregnancy) hypotheses we do however, concur that their relevance was assessed by two variables each, i.e. the intrauterine growth hypothesis is addressed by birthweight and birthweight according to gestational age (please, note that SGA, AGA, LGA represent one 3 level categorical variable)) and the pregnancy hypothesis was addressed by both pregnancy duration and gestational weight gain. We therefore now apply Bonferroni correction for these two hypotheses in that we include a footnote explaining that the respective results are considered significant if p<0.025.

In our view, maternal age at birth and paternal age at birth address two separate potential mechanisms, hence no Bonferroni adjustment was applied to the variables reflecting this hypothesis.

Please see line 542 Table S5 and line 553 Table S6 in the supplementary data section.

Query 3

Strong evidence has been shown that one of the important factors determining the IMT is physical activity. The current research lacks measuring physical activity. Those with higher IMT may have lower physical activity. And we cannot relate the higher IMT to only maternal age.

Response 3

We reckon that other factors such as physical activity can influence the IMT. Unfortunately, assessment of physical activity was incorporated into the DONALD study protocol in 2004 for participation in sport and 2012 for estimated energy expenditure during participation in sport. Hence information on participation in sport in young adulthood is only available in a subsample of n=214 (n=123 female and n=91 male) and information on estimated energy expenditure during sport in young adulthood only in a subsample of n=123 (n=68 female and n=55 males).

Note that participation in sport refers to both organized and unorganized sports, similarly, estimated energy expenditure is based on information acquired for both organized and unorganized sport.

Please, find below the results from the two sensitivity analyses in the respective samples, which reflect that our main finding (association of maternal age at birth with IMT) was retained in the smaller samples and not affected by additional consideration of participation in sport or estimated energy expenditure during participation in sport. Please, note that the change in the ß for maternal age at birth was < 1% upon additional consideration of participation in sport (comparing model B without vs model B with additional consideration of sport) and ˂ 4% upon additional consideration of estimated energy expenditure during participation in sport (comparing model B without vs model B with additional consideration of energy expenditure during sport).

S4 Table. Association of maternal age at child birth and IMT in young adulthood among females with data on physical activity

Average IMT (mm)

N P trend

Females 123 T1 (n=32) T2 (n=46) T3 (n=45)

Maternal age at child birth (yrs)1 27 (25, 28) 30 (29, 31) 34 (33, 36)

Model A2 0.54 (0.53, 0.56) 0.56 (0.54, 0.57) 0.57 (0.56, 0.58) 0.0017

Model B3 0.55 (0.53, 0.56) 0.56 (0.54, 0.57) 0.57 (0.56, 0.58) 0.0016

Model B including sport 0.54 (0.53, 0.56) 0.56 (0.54, 0.57) 0.57 (0.56, 0.58) 0.0014

Females 68 T1 (n=26) T2 (n=18) T3 (n=24)

Maternal age at child birth (yrs)1 28 (27, 29) 31 (30, 32) 36 (34, 37)

Model A2 0.55 (0.53, 0.57) 0.56 (0.54, 0.58) 0.58 (0.56, 0.60) 0.0144

Model B3 0.55 (0.53, 0.57) 0.56 (0.54, 0.58) 0.58 (0.56, 0.60) 0.0092

Model B including energy expenditure during sport 0.55 (0.53, 0.57) 0.56 (0.54, 0.58) 0.58 (0.56, 0.60) 0.0095

T: tertile, n: sample size in tertile.

1Values are medians (25th, 75th percentiles) of maternal age at child birth.

2Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement.

3Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

The results of this sensitivity analysis have been included as additional supplementary material in the manuscript (Table S4 line 532).

Additional sensitivity analyses were performed in subsamples of males, adjusting for participation in sport (n=91) and estimated energy expenditure during participation in sport (n=55), so as to investigate whether the trend association of breastfeeding with IMT was retained. However, no trends were observed in these smaller subsamples irrespective of additional consideration of sport or energy expenditure.

In view of the fact that this change in association is clearly attributable to the reduction in sample size (and not to the consideration of physical activity variables in adulthood), we refrain from including the results of this sensitivity analyses as additional supplementary material.

The results of the sensitivity analyses are however reported in the paper. Please see lines 269-272.

Query 4

Nothing has been mentioned on the characteristics of loss to follow-up participants. The high rate of loss to follow-up may confound the results.

Response 4

We agree that loss to follow-up may result in attrition bias. Please, note some specifics arising for the DONALD Study:

• The children who were initially recruited for the DONALD Study differed con-siderably in age and prospectively collected data on breastfeeding was not always available.

• Due to the open cohort design, many DONALD participants had not yet reached young adulthood by the time of this analysis.

• Follow-up into adulthood was not planned at the inception of the study. Partici-pants are invited since 2004 onwards, to return for further visits at ages 18, 21, 25, 30, 35 etc. However, not all participants followed this invitation.

To address your query, we assessed the characteristics (see table S1) provided by the par-ticipants at inclusion in the study among those who

- had provided data on birth year, birth weight, gestational duration, and maternal age at birth

- had reached the age of 18 years by 2014

- yet had not returned for an IMT measurement in young adulthood.

This comparison reveals that there are minor differences between the sample included in our analysis and those lost to follow up (see Table 1 in the main paper for comparison): mothers were younger at birth i.e. 29.4 and 29.8 years for male and female offspring ver-sus 30.6 and 30.3 years, children were born earlier, i.e. 1987 in both sexes versus 1989 and 1988, respectively and fewer offspring were breastfed for a long duration, i.e. 31% and 27% among male and female offspring in the lost-to follow-up sample versus 34% among both sexes in the analyzed sample.

This information has been included in the manuscript, please see lines 224-227.

In addition, the specifics of the DONALD study and the selection of the sample as out-lined above are now reported together in one paragraph, so as to make this information more easily understandable for the reader � see lines 98-103.

Taken together, in spite of the minor differences attrition bias remains a possibility and this is now acknowledged in the limitations � see lines 342-344.

Table S1. Early life characteristics of participants lost to follow up

Variables Males Females

Early life factors N N

Maternal age at child birth (yrs) 157 29.4 (4.2) 177 29.8 (4.0)

Paternal age at child birth (yrs) 145 32.7 (5.1) 159 33.0 (6.2)

Pregnancy duration (wks) 157 40 (40, 41) 177 40 (39, 41)

Gestational weight gain (kg) 150 13.2 (4.4) 171 12.9 (4.2)

Birthweight (g) 157 3561 (476) 177 3445 (417)

Full breastfeeding 135 144

Never (0-2 weeks) 53 (39.3%) 45 (31.3%)

Short duration (3-17 weeks) 40 (29.6%) 60 (41.7%)

Long duration (>17 weeks) 42 (31.1%) 39 (27.0%)

Birth year 157 1987 (1982, 1990) 177 1987 (1983, 1991)

Full breastfeeding defined as breast milk including water given to the child.

Query 5

The mean/median follow-up time was not stated. As authors declared that inclusion of participants in the cohort and IMT measurement were conducted both in different ages of participants, it more logical to use the survival analysis and include censoring time.

Response 5

We apologize for the confusion, which may have arisen from the fact that the open-cohort design of our study is very uncommon. The study is ongoing and participants are still being recruited onto the study each year. However, those participants that were born until 1996 could have been included in this analysis, since they reached the age of 18 by 2014. Hence, for this analysis the mean follow-up is equivalent with the mean age at IMT measurement, which was 23 years in males and 24 years in females.

This has been stated in the methods of the manuscript � see lines 104-105.

Please, note that our outcome was a continuous variable (IMT), representing a surrogate marker of CVD risk, i.e. we were not following them up for a disease outcome per se in which case survival analysis and censoring time would have been appropriate. In addition, as explained above age at IMT measurement is largely attributable to the changes in the study design (invitation for adult visits since 2004, invitation to IMT measurements since 2008), i.e. censoring time is not related to the outcome. Hence use of survival analysis appears inappropriate, which is why we performed a multiple linear regression analysis. Please, note that we adjusted for age at IMT measurement.

Query 6

Birth year really does not have any added value. The range of the inclusion of participants was less than a decade. Unless investigators believe that something had happened during the recruitment years in the situation.

Response 6

We politely disagree. The data presented for birth year in table 1 are interquartile ranges (as indicated in the footnote). As explained above due to the open-cohort design of the DONALD study, birth year ranged from 1985-1996, i.e. slightly more than a decade.

Selection of birth year as a potential confounder emerged from our strict strategy of confounder selection outlined in the method section. To our surprise, birth year was identified as a relevant confounder hence its inclusion in the model. Removal of this variable results in stronger associations between the early life factors and IMT, i.e. we believe that it would not be correct to report the stronger associations, knowing that they may be confounded by birth year.

Query 7

Blood pressure and body mass index have main effect on atherosclerosis in particular carotid IMT. Were these factors considered during analysis or not?

Response 7

We agree that – similar to waist circumference – BMI or blood pressure may act as a factor explaining the link between early life factors and adult IMT. We hence carried out additional sensitivity analyses adding either BMI or systolic or diastolic blood pressure (in a slightly smaller sample) in adulthood in the respective conditional models for early life factors that were found to be significantly associated with adult IMT. As illustrated below, the association between maternal age at birth and IMT in young adulthood was retained in all three additional conditional models. Please, note that the change in ß for maternal age at birth was <3% in the conditional models (when compared to model B among n=142) with systolic or diastolic blood pressure and <1% in conditional models with BMI (when compared to model B among n=140).

S3 Table. Conditional models for the association of maternal age at child birth with adult IMT in females

Average IMT (mm)

N P trend

Females 142 T1 (n=44) T2 (n=50) T3 (n=48)

Maternal age at child birth (yrs) 1 27 (25, 28) 30 (29, 31) 34 (33, 36)

Model A2 0.54 (0.53, 0.56) 0.55 (0.54, 0.57) 0.57 (0.56, 0.58) 0.0010

Model B3 0.54 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.55, 0.58) 0.0025

Conditional model including waist circumference 0.54 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.56, 0.58) 0.0025

Conditional model including BMI 0.54 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.56, 0.58) 0.0025

140 T1 (n=44) T2 (n=50) T3 (n=46)

Maternal age at child birth (yrs) 1 27 (25, 28) 30 (29, 31) 34 (33, 36)

Model A2 0.54 (0.53, 0.56) 0.55 (0.54, 0.57) 0.57 (0.55, 0.58) 0.0012

Model B3 0.54 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.55, 0.58) 0.0029

Conditional model including systolic blood pressure 0.55 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.55, 0.58) 0.0036

Conditional model including diastolic blood pressure 0.54 (0.53, 0.56) 0.55 (0.54, 0.56) 0.57 (0.55, 0.58) 0.0024

T: tertile, n: sample size in tertile.

1Values are medians (25th, 75th percentiles) of maternal age at child birth.

2Values are adjusted least squares means (95% CIs) of IMT. Model A adjusted for adult age at IMT measurement and the physician taking the IMT measurement.

3Model B additionally adjusted for birth year (residuals of birth year were calculated on age at IMT measurement).

Since none of the conditional models affected the main result, the conditional model for waist circumference was removed from tables 2 and S2, S5-S6 and all conditional models are reported in Appendix table S3 lines 522 and the text only � lines 244-246 and lines 266-268.

Query 8

Does testing the interaction of sex and early life factors on early adulthood IMT sound biologically? The discussion actually did not explain this main finding from the early life point of view. The sex difference finding of studies in adult age was mentioned. What are the probable mechanisms for the sex difference effect of early life factors such as maternal age?

Response 8

We apologize for not having separated observations in adulthood more clearly from potential mechanisms acting in early life. We now separately appraise that

1.) There is a considerable body of evidence that biological factors are responsible for sex dif-ferences in chronic diseases (Kautzky-Willer, 2016) (Kuznetsova, 2018) (Den Ruijter et al., 2015) and potential genetic differences are discussed in this context � lines 317-327.

2.) Epigenetic mechanisms and early nutritional factors and psychosocial stress acting in early life may contribute to sex-differences in adult IMT. Specifically, we discuss

• maternal emotional distress during pregnancy, which can result in adverse changes in placental gene expression, in female fetuses only.

• susceptibility in female animal fetuses to higher transfer of bioactive glucocorticoids that can impair cardiac structure and function

� lines 328-337.

Query 9

It is not clear why adulthood waist circumference was adjusted. There are many other important factors in adulthood, which affect adulthood IMT.

Response 9

Waist circumference is a measure of central adiposity and is a useful predictor of atherosclerosis (Knowles et al., 2010). It is hence plausible, that waist circumference may act as a link between early life factors and adult IMT – similar to BMI and blood pressure, please see query 7. This is why we had included waist circumference in a conditional model.

As outlined above, neither inclusion of adult waist circumference, nor BMI, nor systolic or diastolic blood pressure affected the significant association between maternal age at birth and adult IMT. To avoid presenting a large number of non-informative tables, we removed all conditional models from the tables. Instead, we now include a statement in the results, that additional consideration of these variables did not explain our finding. All conditional models are presented in table S3 in the appendix of the manuscript.

Query 10

For non-significant findings, it is not necessary to discuss the effect of early life factors on IMT. More importantly, the explanation of non-significant results is required.

Response 10

Agreed. We substantially reduced our discussion on the non-significant findings, moving some of the content to the introduction to explain our lines of hypotheses.

Query 11

To test the trend of baseline characteristics, analysis of variance is not appropriate. It is mainly shows the difference and is used for comparison.

Response 11

Agreed. We removed all tests in the baseline characteristics.

Reviewer #2 (Remarks to the Author):

Reviewers’ comments to the Author:

Query 12

My only comment is the absence of any data about other relevant behaviors / factors during adolescence (also a sensitive period to development of cardiovascular diseases). I will bring this aspect into the light later in the discussion section as well.

Response 12

We acknowledge that physical activity can influence the IMT. However, assessment of physical activity was included in the DONALD study protocol in 2004 for participation in sport and 2012 for estimated energy expenditure during participation in sport, thus information on participation in sport is only available for a subsample of n=123 female and n=91 male and information on estimated energy expenditure during sport in young adulthood only for n=68 female and n=55 males.

We performed a sensitivity analysis in the subsamples with information on participation in sport or estimated energy expenditure during participation in sport. The results indicate the association of maternal age at birth with IMT was retained and not affected by additional consideration of physical activity. Please see response to query 3.

Query 13

Insert a flowchart in the methods describing all variables considered, as well as the time point where those variables were collected

Response 13

A flowchart of IMT measurement has been inserted in the methods section� lines 117-119.

In terms of the time points, we apologize for not having been clear.

• Early life factors referring to gestation and birth were inquired at the first visit to the study (usually at ages 0.25 or 0.5 years)

• Data on full breastfeeding (not given solid foods and no liquids daily except breast milk, tea, or water) were inquired at 3 or 6 months and prospectively at subsequent visits until weaning.

• Data on IMT were measured during young adulthood at ages 18, 21, 25, 30, 35 and 40 years

This information has been added to the manuscript � lines 99-100.

Query 14

I suggest a different approach to present the results.

Table 1: compare all variables according to sex, mainly because the findings seems sex-dependent.

Response 14

Thanks for your suggestion, Table 1 has been revised and the results are presented according to sex.

Query 15

Is there any relationship between maternal age and the other early life variables considered (eg. low birthweight)? Due to the relevance of maternal age in your findings, a table describing that would welcome.

Response 15

Maternal age at birth was related to paternal age at birth (p˂0.0001) and pregnancy duration (p=0.03) but not to birth weight according to gestational age (i.e. SGA, AGA, LGA), gestational weight gain, birthweight or breastfeeding (p>0.05).

Following our strict procedure for confounder selection, paternal age at birth or pregnancy duration did not emerge as confounders for models examining the relevance of maternal age at birth for adult IMT. Please, note that despite its high correlation with maternal age at birth, there was no collinearity with maternal age at birth (i.e. tolerances of the model including both variables were acceptable).

Since the fact that paternal age at birth did not affect the association may have implication for the interpretation of the results, this was stressed in the results � see lines 241 and 244 and referred to in the discussion � see lines 298-300.

We would like to abstain from presenting a table on the relation between maternal age at birth with other early life factors for the following reasons:

• Maternal age at birth emerged as relevant for IMT during analyses, i.e. our aim was not to describe the association between maternal age at birth and other early life factors.

• The other early life factors were not of relevance for the main association that emerged during analyses, i.e. the relation of maternal age with adult IMT.

Should the Editor feel that the inclusion of such a table in the Appendix would be of interest, we are however prepared to do so.

Query 16

The discussion needs to be reduced and focused only on significant findings. Moreover, the ways the authors built the sections leads the reader to believe that behavioral variables in other moments of life are not relevant into this phenomenon. For instance, IMT in adulthood is affected by sports participation during adolescence (Werneck et al. 2018). Please, I would suggest to highlight the relevance of these data and also its absence in the models created.

Response 16

Agreed. We substantially reduced our discussion on the non-significant findings, moving some of the content to the introduction to explain our lines of hypotheses. �see response to queries 1 & 10, reviewer 1.

The reviewer correctly points out that other behavioral variables such as physical activity are relevant for adult IMT. We expanded our analysis carrying out a sensitivity analysis including participation in physical activity as an additional covariate but this did not change the result that maternal age at child birth is relevant for IMT in early adulthood in females but not in males. � see response to query 3, reviewer 1.

We have added the non-availability of behavioral variables in other moments of life that are relevant to IMT as a limitation of our study in the discussion. � see lines 347-349.

References

Den Ruijter HM, Haitjema S, Asselbergs FW, Pasterkamp G. Sex matters to the heart: A special issue dedicated to the impact of sex related differences of cardiovascular diseases. Atherosclerosis. 2015;241:205–7. doi:10.1016/j.atherosclerosis.2015.05.003.

Kautzky-Willer A, Harreiter J, Pacini G. Sex and Gender Differences in Risk, Pathophysiology and Complications of Type 2 Diabetes Mellitus. Endocr Rev. 2016; 37:278–316. doi:10.1210/er.2015-1137.

Knowles KM, Paiva LL, Sanchez SE, Revilla L, Lopez T, Yasuda MB, et al. Waist Circumference, Body Mass Index, and Other Measures of Adiposity in Predicting Cardiovascular Disease Risk Factors among Peruvian Adults. Int J Hypertens. 2011; 2011:931402. doi:10.4061/2011/931402.

Kuznetsova T. Sex Differences in Epidemiology of Cardiac and Vascular Disease. Adv Exp Med Biol. 2018;1065:61–70. doi:10.1007/978-3-319-77932-4_4.

Submitted filename: RESPONSE to Editor and Reviewers.docx

Click here for additional data file.

1 May 2020

Early life factors and their relevance to intima-media thickness of the common carotid artery in early adulthood

PONE-D-19-30815R1

Dear Dr. Buyken,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Rudolf Kirchmair

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: All the comments have been fully addressed and responded. The important restrictions have been added to limitation part.

Reviewer #2: Comparing with the previous version, all my comments were properly addressed by the author. Therefore, there are no further comments.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

5 May 2020

PONE-D-19-30815R1

Early life factors and their relevance to intima-media thickness of the common carotid artery in early adulthood

Dear Dr. Buyken:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof Rudolf Kirchmair

Academic Editor

PLOS ONE