Correlation between cardiovascular health and vascular phenotype

Javascript is currently disabled in your browser. When javascript is disabled, some functions of this website will not work.
Register your specific details and specific drugs of interest, and we will match the information you provide with articles in our extensive database and send you a PDF copy via email in a timely manner.
The relationship between ideal cardiovascular health and vascular phenotype of obese mothers and their 6-year-old children
Authors: Litwin L, Sundholm JKM, Meinilä J, Kulmala J, Tammelin TH, Rönö K, Koivusalo SB, Eriksson JG, Sarkola T
Linda Litwin,1,2 Johnny KM Sundholm,1,3 Jelena Meinilä,4 Janne Kulmala,5 Tuija H Tammelin,5 Kristiina Rönö,6 Saila B Koivusalo,6 Johan G Eriksson,7–10 Taisto Sarkola1,31Children’s Hospital, University of University of Helsinki and Helsinki University Hospitals, Helsinki, Finland; 2 Department of Congenital Heart Defects and Pediatric Cardiology, Silesian Medical University, Katowice, Poland, Zabrze FMS; 3 Minerva Foundation Medical Research Institute, Helsinki, Finland; 4 Department of Food and Nutrition, University of Helsinki, Helsinki, Finland; 5LIKES Sports Activity and Health Research Center, Jyvaskyla, Finland; 6 University of Helsinki Women’s Hospital and Helsinki University Hospital in Helsinki, Finland; 7 Folkhälsan Research Center, Helsinki, Finland; 8 University of Helsinki and Helsinki Department of General Practice and Primary Health Care, University Hospital, Helsinki, Finland; 9 Human Potential Transformation Research Program and Department of Obstetrics and Gynecology, Yang Luling School of Medicine, National University of Singapore, Singapore; 10 Singapore Institute of Clinical Sciences (SICS), Science, Technology and Research Bureau (A*STAR), Singapore Communications: Linda Litwin Department of Congenital Heart Defects and Pediatric Cardiology, Zabrze FMS, Silesian Medical University, M.Sklodowskiej-Curie 9, Zabrze, 41-800, Poland Tel +48 322713401 Fax +48 322713401 Email [email protected] Background: Genetics and family-shared lifestyles can cause cardiovascular risks, but the extent to which they affect the structure and function of arteries in early childhood is unclear. We aimed to assess the association between ideal cardiovascular health in children and mothers, maternal subclinical atherosclerosis, and arterial phenotypes in children. Methods: From the Finnish Gestational Diabetes Prevention Study (RADIEL) longitudinal cohort, a cross-sectional analysis of 201 mother-child children at 6.1 ± 0.5 years of age assessed ideal cardiovascular health (BMI, blood pressure, fasting blood glucose, total cholesterol, diet quality, Physical activity, smoking), body composition, carotid ultra-high frequency ultrasound (25 and 35 MHz) and pulse wave velocity. Results: We found that there was no correlation between the ideal cardiovascular health of the child and the mother, but reported evidence of the correlation of specific indicators: total cholesterol (r=0.24, P=0.003), BMI (r=0.17, P=0.02), Diastolic blood pressure (r=0.15, P=0.03) and diet quality (r=0.22, P=0.002). The pediatric arterial phenotype has nothing to do with the ideal cardiovascular health of the child or mother. In a multivariate regression interpretation model adjusted for children’s gender, age, systolic blood pressure, lean body mass, and body fat percentage, the thickness of the carotid artery intima-media in children was only independently correlated with the thickness of the maternal carotid artery intima-media (an increase of 0.1 mm [95 %] CI 0.05, 0.21, P=0.001] The thickness of the maternal carotid artery intima-media increased by 1 mm). Children of mothers with subclinical atherosclerosis decreased carotid artery dilatation (1.1 ± 0.2 vs 1.2 ± 0.2%/10 mmHg, P=0.01) and increased carotid artery intima-media thickness (0.37 ± 0.04 vs 0.35) ± 0.04 mm, P=006) Conclusion: Ideal cardiovascular health indicators are heterogeneously related to mother-child pairs in early childhood. We found no evidence of the effect of children’s or mother’s ideal cardiovascular health on children’s arterial phenotypes. Maternal carotid artery intima-media thickness can predict the carotid artery intima-media thickness in children, but its underlying mechanism is still unclear. Maternal subclinical atherosclerosis is related to local carotid artery stiffness in early childhood. Keywords: cardiovascular disease, atherosclerosis, carotid artery intima-media thickness, risk factors, children
Traditional cardiovascular risk factors contribute to the occurrence and development of atherosclerosis. 1,2 Risk factors tend to cluster together, and their combination seems to be more predictive of individual cardiovascular risk. 3
The American Heart Association defines ideal cardiovascular health (ICVH) as a set of seven health indicators (body mass index (BMI), blood pressure (BP), fasting blood glucose, total cholesterol, diet quality, physical activity, smoking) to promote primitive prevention Cardiovascular disease in children and adults. 4 ICVH is negatively correlated with subclinical atherosclerosis in adulthood. 5 ICVH and adverse vascular phenotypes are reliable predictors of cardiovascular events and mortality in adults. 6-8
The cardiovascular disease of the parents increases the risk of cardiovascular events in the offspring. 9 Environmental factors related to genetics and common lifestyles are both considered as potential mechanisms, but their contribution has not yet been determined. 10,11
The correlation between parent and child ICVH is already evident in children 11-12 years old. At this stage, children’s ICVH is related to carotid artery elasticity and negatively related to cervical femoral pulse wave velocity (PWV), but it is not reflected in the carotid artery intima-media thickness (IMT). 12 However, cardiovascular risk between 12-18 years of age is associated with an increase in carotid IMT in middle-aged life, and has nothing to do with risk factors during the same period. 13 Evidence regarding the strength of these associations in early childhood is missing.
In our previous work, we did not find the effects of gestational diabetes or maternal lifestyle interventions on early childhood anthropometry, body composition or arterial size and function. 14 The focus of this analysis is the cross-generational trend of cardiovascular risk aggregation. Class and its effect on the arterial phenotype of children. We hypothesize that maternal ICVH and vascular substitutes for cardiovascular disease will be reflected in childhood ICVH and arterial phenotypes in early childhood.
The cross-sectional data are from a six-year follow-up of the Finnish Gestational Diabetes Prevention Study (RADIEL). The initial research design has been proposed elsewhere. 15 In short, women who plan to become pregnant or are in the first half of pregnancy and have an increased risk of gestational diabetes (obesity and/or history of gestational diabetes) were recruited (N=728). The 6-year cardiovascular follow-up was designed as an observational study of mother-infant pairs, with an equal number of mothers with and without gestational diabetes, with a pre-specified cohort size (~200). From June 2015 to May 2017, continuous invitations were sent to participants until the limit was reached, and 201 pairs of two-tuples were recruited. The follow-up is designed for children aged 5-6 years to ensure cooperation without sedation, including maternal-infant binary group assessment of body size and composition, blood pressure, fasting blood glucose and blood lipids, physical activity using accelerometers, diet quality and smoking questionnaires (mothers), blood vessels Ultrasound and intraocular pressure measurement and echocardiography in children. The availability of data is listed in Supplementary Table S1. The Ethics Committee of Obstetrics and Gynecology, Pediatrics and Psychiatry of Helsinki University Hospital approved the research protocol (20/13/03/03/2015) for a six-year follow-up evaluation. All mothers’ informed written consent was obtained at the time of registration. The study was conducted in accordance with the Declaration of Helsinki.
A skilled researcher (TS) uses 25 MHz and 35 MHz transducers with Vevo 770 system, and uses UHF22, UHF48 (similar center frequency) and Vevo MD system (VisualSonics, Toronto, Canada) as the final 52 pairs of mother and child. The common carotid artery was imaged 1 cm proximal to the bilateral carotid bulbs, and the resting position was in the supine position. Use the highest frequency that can visualize the far wall to obtain high-quality film images covering 3-4 cardiac cycles. Use Vevo 3.0.0 (Vevo 770) with manual electronic calipers and VevoLab (Vevo MD) software to analyze the images offline. 16 Lumen diameter and IMT were measured by an experienced observer (JKMS) at the end of diastole using cutting-edge techniques), unaware of subject characteristics (Supplementary Figure S1). We have previously reported that the intra-observer coefficient of variation measured by ultra-high-resolution ultrasound in children and adults is 1.2-3.7% in the lumen diameter, IMT is 6.9-9.8%, and the inter-observer coefficient of variation is 1.5-4.6% in the lumen diameter. , 6.0-10.4% of IMT. The carotid IMT Z score adjusted for age and gender was calculated using the reference of healthy white non-obese children. 17
Carotid artery lumen diameter was measured at peak systole and end-diastole to evaluate carotid artery β stiffness index and carotid artery expansion coefficient. Using an appropriately sized cuff, the oscillometric method (Dinamap ProCare 200, GE) was used to record the systolic and diastolic blood pressure for elastic performance calculations during ultrasound imaging in the supine position of the right arm. The carotid artery expansion coefficient and the carotid artery β-stiffness index are calculated from the carotid artery using the following formula:
Among them, CCALAS and CCALAD are the common carotid artery lumen area during systole and diastole respectively; CCALDS and CCALDD are the common carotid artery lumen diameter during systole and diastole respectively; SBP and DBP are systolic and diastolic blood pressure. 18 The coefficient of variation of the carotid artery expansion coefficient in the observer is 5.4%, the coefficient of variation of the carotid artery β stiffness index is 5.9%, and the inter-observer coefficient of variation of the carotid artery expansion is 11.9% coefficient and 12.8% of the carotid artery β stiffness index .
The traditional high-resolution ultrasound Vivid 7 (GE) equipped with a 12 MHz linear transducer was used to further screen the maternal carotid artery for plaque. Starting from the common carotid artery near the bulb, the carotid artery is screened bilaterally through the bifurcation and the proximal part of the internal and external carotid arteries. According to the Mannheim consensus, plaque is defined as 1. Local thickening of the vessel wall by 0.5 mm or 50% of the surrounding IMT or 2. The total arterial wall thickness exceeds 1.5 mm. 19 The presence of plaque was assessed by a dichotomy. The primary observer (JKMS) independently performs repeated measurements on a subset of images (N = 40) to evaluate intra-observer variability, and the second observer (TS) evaluates inter-observer variability. The Cohen κ of intra-observer variability and inter-observer variability were 0.89 and 0.83, respectively.
PWV was measured by a trained research nurse to assess regional arterial stiffness using a mechanical sensor (Complior Analyse, Alam Medical, Saint-Quentin-Fallavier, France) while resting in the supine position. 20 Sensors are placed on the right carotid artery, right radial artery, and right femoral artery to evaluate the central (right carotid artery-femoral artery) and peripheral (right carotid artery-radial artery) transit time. Use a tape measure to measure the direct distance between the recording points to the nearest 0.1 cm. The right carotid femoral artery distance is multiplied by 0.8 and then used in the center PWV calculation. Repeat the recording in the supine position. Two records were obtained when the third record was performed in a setting where the difference between the measurements was greater than 0.5 m/s (10%). In the setting of more than two measurements, the result with the lowest tolerance value is used for analysis. Tolerance is a quality parameter that quantifies the variability of the pulse wave during recording. Use the average of at least two measurements in the final analysis. The PWV of 168 children can be measured. The coefficient of variation of repeated measurements was 3.5% for the carotid-femoral artery PWV and 4.8% for the carotid-radial artery PWV (N=55).
A set of three binary indicators is used to reflect the mother’s subclinical atherosclerosis: the presence of carotid artery plaque, carotid artery IMT adjusted age and exceed the 90th percentile in our sample, and more than 90 percent The PWV of the neck and femur is matched with age and optimal blood pressure. twenty one
ICVH is a set of 7 binary indicators with a cumulative range from 0 to 7 (the higher the score, the more in line with the guidelines). 4 The ICVH indicators used in this study are consistent with the original definition (three modifications have been made)-Supplementary Table S2) and include:
The quality of the diet is assessed by the child’s Finnish Child Healthy Eating Index (range 1-42) and the mother’s healthy food intake index (range 0-17). Both indexes cover 4 of the 5 categories included in the original diet indicator (except sodium intake). 23,24 The critical value of ideal and non-ideal diet quality is defined as 60% or more to reflect the quality of the original diet. Indicator definition (it is ideal if more than 3 of the 5 criteria are met). With reference to the recent healthy Finnish pediatric child population (87.7% for girls, 78.2% for boys), if the gender-specific threshold for overweight children is exceeded, the child’s BMI is defined as non-ideal, which is slightly different from 85% of the Finnish population. 22 Due to a large number of school dropouts and a very low discriminant value (Supplementary Table S1, 96% of mothers meet the ICVH criteria), physical activity of pregnant and lying-in women was excluded. ICVH is subjectively divided into the following categories: low (children 0-3, mothers 0-2), medium (children 4, mothers 3-4) and high (children and mothers 5-6), providing an opportunity to compare different categories.
Use electronic equipment (Seca GmbH & Co. KG, Germany) to measure height and weight to the nearest 0.1 cm and 0.1 kg. Children’s BMI Z scores are generated with reference to the most recent Finnish population data set. 22 Body composition passed bioelectrical impedance assessment (InBody 720, InBody Bldg, South Korea).
The resting blood pressure was measured by the oscillometric method from the right arm in a sitting position (Omron M6W, Omron Healthcare Europe BV, The Netherlands) with a sufficient cuff. The average systolic and diastolic blood pressure are calculated from the two lowest measurements (a minimum of three measurements). Children’s blood pressure Z value is calculated according to the guidelines. 25
Blood samples of plasma glucose and lipids were collected under fasting conditions. Results from 3 children with uncertain fasting compliance (excessive high triglycerides, fasting blood glucose, and glycosylated hemoglobin A1c (HbA1c)) were excluded from the analysis. Total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol and triglycerides are determined by enzymatic method, plasma glucose and enzymatic hexokinase determination, and HbA1c and immunoturbidimetric analyzer (Roche Diagnostics , Basel, Switzerland) for evaluation.
The mother’s dietary intake was assessed by the food frequency questionnaire and further assessed by the healthy food intake index. The Healthy Food Intake Index has previously been validated as a useful tool to reflect compliance with Nordic Nutrition Recommendation 26 in the original RADIEL cohort. 24 In short, it contains 11 ingredients, covering the consumption of vegetables, fruits and berries, high-fiber cereals, fish, milk, cheese, cooking oil, fatty sauces, snacks, sugary drinks and fast food. The higher the score reflects the higher the degree of compliance with the recommendations. The quality of the children’s diet was assessed through 3-day food records and further assessed by the Finnish Children’s Healthy Eating Index. The Finnish Children’s Healthy Eating Index has previously been validated in the Finnish pediatric population. 23 It includes five types of food: vegetables, fruits and berries; oil and margarine; foods high in sugar; fish and fish and vegetables; and skimmed milk. The food consumption is scored so that the higher the consumption, the higher the score. Except for foods that contain a lot of sugar, the score is reversed. Before scoring, adjust the energy intake by dividing the intake (grams) by the energy intake (kcal). The higher the score, the better the quality of the children’s diet.
Moderate to vigorous physical activity (MVPA) was measured using a child hip accelerometer (ActiGraph GT3X, ActiGraph, Pensacola, USA) and a mother’s armband (SenseWear ArmBand Pro 3). Instructed to wear the monitor during awake and sleep time, but sleep time was excluded from the analysis. The child monitor collects data at a sampling rate of 30 Hz. The data is usually filtered, converted to a 10-second epoch count, and analyzed using the Evenson (2008) cut point (≥2296 cpm). 27 The mother monitor collects MET values ​​in the 60-second epoch. MVPA is calculated as the MET value exceeds 3. Effective measurement is defined as at least 2 working days and 1 weekend (recording at least 480 minutes per day) and 3 working days and 1 weekend (recording at least 720 minutes per day) for the mother. MVPA time is calculated as a weighted average [(average MVPA minutes/day on weekdays × 5 + average MVPA minutes/day on weekends × 2)/7], in addition, as a percentage of total wearing time. The most recent physical activity data of the Finnish population was used as a reference. 28
The questionnaire was used to obtain information about the mother’s smoking, chronic diseases, medications, and education.
Data are expressed as mean ± SD, median (interquartile range) or counts (percentage). Evaluate the normal distribution of all continuous variables based on the histogram and normal QQ plot.
Independent sample t test, Mann-Whitney U test, one-way analysis of variance, Kruskal-Wallis, and chi-square test were used as appropriate for comparison groups (mother and child, boy and girl, or low and medium and high ICVH).
The Pearson or Spearman rank correlation coefficient was used to explore the univariate association between the characteristics of the child and the mother.
The multivariate linear regression model was used to establish an explanatory model for children’s HDL cholesterol and carotid IMT. Variable selection is based on correlation and expert clinical judgment, avoids significant multicollinearity in the model, and includes potential confounding factors. Multicollinearity is evaluated using the variance inflation factor, with a maximum value of 1.9. Multivariate linear regression was used to analyze the interaction.
Two-tailed P ≤ 0.05 was set to be significant, except in the correlation analysis of determinants of carotid artery IMT in children with P ≤ 0.01.
Participant characteristics are shown in Table 1 and Supplementary Table S3. Compared with the reference population, children’s BMI Z score and BP Z score increased. Our previous work reported detailed data on arterial morphology in children. 14 Only 15 (12%) children and 5 (2.7%) mothers met all ICVH criteria (Supplementary Figures 2 and 3, Supplementary Tables S4-S6).
The maternal and infant cumulative ICVH score is only related to boys (boys: rs=0.32, P=0.01; girls: rs=-0.18, P=0.2). When analyzed as a continuous variable, the maternal-infant univariate correlation analysis has significant significance in the measurement of blood lipids, HbA1C, obesity, diastolic blood pressure, and diet quality (Supplementary Figures S4-S10).
Children’s and mother’s LDL, HDL, and total cholesterol are correlated (r=0.23, P=0.003; r=0.35, P<0.0001; r=0.24, P=0.003, Figure 1). When stratified by child’s gender, the correlation between child’s and mother’s LDL and total cholesterol remained significant only in boys (Supplementary Table S7). Triglycerides and HDL cholesterol are correlated with girls’ body fat percentage (rs=0.34, P=0.004; r=-0.37, P=0.002, respectively, Figure 1, Supplementary Table S8).
Figure 1 The relationship between child and mother’s blood lipids. Scatter plot with linear regression line (95% confidence interval); (AC) maternal and infant blood lipid levels; (D) girl’s body fat percentage and high-density lipoprotein cholesterol. Significant results are shown in bold (P ≤ 0.05).
Abbreviations: LDL, low-density lipoprotein; HDL, high-density lipoprotein; r, Pearson correlation coefficient.
We found that there was a significant correlation between the HbA1C of the child and the mother (r=0.27, P=0.004), but it was not related to fasting blood glucose (P=0.4). Children’s BMI Z score, but not body fat percentage, is weakly correlated with mother’s BMI and waist-to-hip ratio (r=0.17, P=0.02; r=0.18, P=0.02, respectively). The Z value of children’s diastolic blood pressure is weakly correlated with the mother’s diastolic blood pressure (r=0.15, P=0.03). The Finnish children’s healthy diet index is correlated with the mother’s healthy food intake index (r=0.22, P 0.002). This relationship was only observed in boys (r=0.31, P=0.001).
After excluding mothers who were treated for hypertension, hypercholesterolemia, or hyperglycemia, the results were consistent.
The detailed arterial phenotype is shown in Supplementary Table S9. The vascular structure of children is independent of the characteristics of children (Supplementary Table S10). We did not observe any association between childhood ICVH and vascular structure or function. In the analysis of children stratified by ICVH scores, we observed that the carotid IMT Z scores of children with only moderate scores increased compared with children with low scores (mean ± SD; moderate score 0.41 ± 0.63 vs low score- 0.07 ± 0.71, P = 0.03, Supplementary Table S11).
Maternal ICVH is not associated with the vascular phenotype of children (Supplementary Tables S10 and S12). Children and maternal carotid artery IMT are correlated (Figure 2), but the maternal-child correlation between different vascular stiffness parameters is not statistically significant (Supplementary Table 9, Supplementary Figure S11). In a multivariate regression interpretation model adjusted for children’s gender, age, systolic blood pressure, lean body mass, and body fat percentage, maternal carotid IMT is the only independent predictor of children’s carotid IMT (adjusted R2 = 0.08). For every 1 mm increase in maternal carotid IMT, childhood carotid IMT increased by 0.1 mm (95% CI 0.05, 0.21, P = 0.001) (Supplementary Table S13). The gender of the child did not mitigate this effect.
Figure 2 Correlation between carotid artery intima-media thickness in children and mothers. Scatter plot with linear regression line (95% confidence interval); (A) maternal and child carotid IMT, (B) maternal carotid IMT percentile and child carotid IMT z-score. Significant results are shown in bold (P ≤ 0.05).
The maternal blood vessel score is correlated with the carotid artery expansion coefficient and β stiffness index in children (rs=-0.21, P=0.007, rs=0.16, P=0.04, Supplementary Table S10, respectively). Children born to mothers with a vascular score of 1-3 have a lower coefficient of carotid artery expansion than those born to mothers with a score of 0 (mean ± standard deviation, 1.1 ± 0.2 vs 1.2 ± 0.2%/10 mmHg, P=0.01) and there is a tendency to increase carotid artery β stiffness index (median (IQR), 3.0 (0.7) and 2.8 (0.7), P=0.052) and carotid artery IMT (mean ± SD, 0.37 ± 0.04 and 0.35 ± 0.04 mm, P=0.06) (Figure 3), Supplementary Table S14).
Figure 3 Child vascular phenotype stratified by maternal vascular score. Data are expressed as mean + SD, P with independent sample t test (A and C) and Mann-Whitney U test (B). Significant results are shown in bold (P ≤ 0.05). Maternal blood vessel score: range 0-3, a set of three binary indicators: the presence of carotid plaque, the thickness of the carotid artery intima-media adjusted by age and exceeded 90% in our sample, and the cervical-femoral pulse wave velocity exceeded 90% are age-matched and optimal blood pressure. twenty one
The maternal score (ICVH, vascular score) and the combination of child and maternal scores are not related to the arterial phenotype of children (Supplementary Table S10).
In this cross-sectional analysis of mothers and their 6-year-old children, we investigated the association between childhood ICVH, maternal ICVH, and maternal subclinical atherosclerosis with the structure and function of children’s arteries. The main finding is that only the mother’s subclinical atherosclerosis, while the children’s and mother’s conventional cardiovascular risk factors are not related to the adverse changes in early childhood vascular phenotypes. This new insight into early childhood vascular development increases our understanding of the intergenerational impact of subclinical atherosclerosis.
We report evidence of decreased carotid artery dilatation and trends in carotid artery beta stiffness and carotid artery IMT in children of mothers with cardiovascular disease vascular substitutes. However, there is no direct correlation between maternal and infant vascular function indicators. We hypothesize that including maternal plaque into the vascular score significantly increases its predictive value.
We have observed a positive correlation between the carotid artery IMT in children and mothers; however, the mechanism is still unclear because the carotid artery IMT in children is independent of the characteristics of the child and the mother. The association between the children’s ICVH score and carotid IMT showed inconsistency, because we did not observe any difference between low ICVH and high ICVH.
We know that other factors may play a role, including children’s head circumference, which may be an important predictor of carotid artery size in the early stages of growth. In addition, our results may be attributable to unmeasured factors that affect fetal vascular development. However, we have previously reported that pre-pregnancy overweight/obesity and gestational diabetes have no effect on early childhood carotid IMT. 14 Further research is necessary to explore the influence of arterial structure and function on children’s growth and genetic background.
The reported associations are consistent with previous studies conducted in adolescents, which provided evidence of associations between parent-child vascular phenotypes, including carotid IMT, although body size was not adjusted in the analysis. 29 The considerable heredity of carotid IMT further confirms this and adult arterial stiffness. 30,31
The observed association between maternal subclinical atherosclerosis and childhood vascular phenotype was not extended by maternal ICVH. This is consistent with previous studies in which a large part of the variation in children’s vascular phenotype is explained by genetic factors independent of the conventional cardiovascular risk factors of parents and children. 29
In addition, the observed vascular changes have nothing to do with childhood ICVH, indicating the main influence of early childhood genetic background. The contribution of environmental factors seems to change with the age of children, as a previous large cross-sectional cohort study of children aged 11-12 reported a significant association between children’s vascular function and ICVH. 12