ABSTRACT
Importance Childhood obesity increases cardiometabolic risk during childhood among individuals aged 2-18 years. These cardiometabolic outcomes include glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and type 2 diabetes (T2D). In the current state of research, a comprehensive review identifying all early-life exposures and risk factors that could predict cardiometabolic risk in children is lacking.
Objective To identify and evaluate the predictive early-life risk factors during the first 1,000 days of life, including preconception, pregnancy and birth, and early infancy periods for cardiometabolic risk outcomes in childhood.
Evidence review The present systematic review of existing literature was conducted to revise and search selected electronic databases (Medline, EMBASE, WEB OF SCIENCE, SCOPUS, and Cochrane CENTRAL) for longitudinal studies published between the database’s inception and August 17, 2022. This systematic review protocol was registered to PROSPERO, CRD42022355152, and following the PRISMA guidelines. We selected articles that studied the risk factors in mothers, fathers and infants, from preconception to infancy, for childhood cardiometabolic outcomes between 2 and 18 years.
Findings In 68 studies, we identified 229 associations between exposures and childhood cardiometabolic outcomes namely glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and T2D. The majority of associations (n=162) were positively associated with cardiometabolic risk factors. Pregnancy and birth risk factors were the categories with the most reported associations (86%). Among them, the most frequently assessed characteristics were birth anthropometrics (n=75), sociodemographics data (n=47), and pregnancy complications factors (n=34). However, we only identified few risk factors during preconception. In infancy period, breastfeeding (n=16) and infant anthropometrics (n=15) were consistently associated with cardiometabolic outcomes. In all periods, the most studied associations were identified for hypertension and metabolic syndrome.
Conclusions and relevance Birth anthropometrics, sociodemographics, and pregnancy complication factors were the most frequently reported predictive factors associated with a higher risk for cardiometabolic outcomes in children, particularly hypertension and metabolic syndrome. These results of this study are useful for predicting the risk for childhood cardiometabolic outcomes and for the modifiable factors. They also may facilitate the design of approaches aimed at the alteration of several behaviours from birth to infancy, encompassing both the maternal and paternal influences, as well as the preconception to infancy transition period. Future studies evaluating early-life risk factors with scarce scientific evidence, such as paternal and preconception factors, are urgently needed.
Question What is the existing evidence of early-life risk factors during the first 1000 days of life that are associated with cardiometabolic dysfunction between 2 and 18 years?
Findings Birth anthropometrics, pregnancy complications, sociodemographics and lifestyle factors are the most frequently reported associated exposures with hypertension and metabolic syndrome in children.
Meaning Identifying early-life risk factors and changing behavior patterns throughout preconception and infancy can contribute to prevent metabolic diseases later in childhood.
INTRODUCTION
Obesity increases the risk and severity of several cardiometabolic conditions, such as glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and type 2 diabetes (T2D). More than 50% of children and adolescents with obesity are diagnosed with one cardiometabolic risk condition, and 25% are diagnosed with two and more cardiometabolic risk conditions 1,2. The global prevalence of T2D by 2030 is estimated to increase to 10.2% from 9.3% in 2019 3. Furthermore, estimations from 2021 indicate that T2D or its complications is main cause of death for nearly 7 million people worldwide, leading to an economic cost of approximately US$ 966 billion in 2021, which could reach US$ 1.03 trillion by 2030 and US$ 1.05 trillion by 2045 4. Early-life exposures or risk factors during the preconception period, pregnancy and the early years of life could lead to the development of cardiometabolic features during childhood 5. Children rarely suffer adverse cardiovascular events as those experienced by adults, but cardiovascular and metabolic disease risk factors may already be present and related to future adult cardiovascular diseases (CVD) development 6,7, which are responsible for about 31% of deaths worldwide 8. Several studies emphasize intrauterine conditions are an important factors of cardiometabolic risk. In fact, birth weight is a sentinel marker of fetal health, reflecting both intrauterine growth and length of gestation. Interestingly, both low birth weight (LBW) (2500 g) and high birth weight (HBW) (>4000 g) are linked to cardiometabolic risk factors later in childhood 7. Extrauterine growth retardation seems to be also associated with a greater risk of cardiometabolic risk later in life 9,10. Several studies have shown that maternal factors might influence cardiometabolic health in children11. Thus, maternal body mass index (BMI) before pregnancy, maternal weight gain during pregnancy, gestational diabetes, maternal malnutrition, and maternal smoking during pregnancy, are independently associated with neonatal, infant, and child adiposity, and might impact the development of cardiometabolic risk factors. In addition, pregnancy complications such as preeclampsia and gestational diabetes also predict cardiovascular morbidity and mortality, and the cardiometabolic risk is often transmitted to the next generation12. Children who were exposed to gestational diabetes in utero may also be more susceptible to developing metabolic syndrome, since they have higher systolic blood pressure, BMI, and blood glucose than those who were not exposed to gestational diabetes 13. Although there is less information about how paternal factors might imprint cardiometabolic conditions during childhood, evidence indicates a relevant role for males in affecting cardiometabolic outcomes in their offspring 14. It has been proven that paternal BMI during preconception impairs the cardiometabolic profile of children 15–17.
To halt the global rise in cardiometabolic disease, prevention of risk factor development in early life is of key importance. A suboptimal fetal and postnatal development needs to be addressed as early as possible according to evidence-based recommendations to prevent the perpetuation of adiposity, which can lead to obesity in adolescence, adulthood, and subsequent generations 18,19. Moreover, a better understanding of early life exposures and risk factors that are associated with an increased cardiometabolic disease risk is crucial.
Hence, we conducted this systematic review to determine whether/which early-life exposures result in increased rates of cardiometabolic comorbidities in children between 2 and 18 years. By analyzing the first 1,000 days of life, we aimed to assess the quality of potential crucial early-life risk factors during preconception, pregnancy and birth, and early infancy periods for the prediction of children at risk for cardiometabolic outcomes such as glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and T2D.
METHODS
Systematic review protocol development
This study was embedded in a large collaborative project to conduct a systematic review on risk factors in the first 1,000 days of life for the development of cardiometabolic disorders in childhood. The purpose of our systematic review protocol was to include and evaluate all individual research studies on risk factors and noninvasive biomarkers in children and adolescents for cardiometabolic disorders during preconception, pregnancy and infancy. To investigate early childhood risk factors for cardiometabolic outcomes between 2 and 18 years, we aimed to identify case‒control, prospective observational, retrospective observational, and randomized control trials (RCTs) focusing on those risk factors and specific outcomes. Studies that included diseased populations only, such as mothers with type 1 diabetes, were excluded.
Early exposure variables related to sociodemographic factors, lifestyle factors, physical factors, environmental factors, pregnancy-related factors, and noninvasive biomarkers are referred to as risk factors. Here, risk factors correspond to exposure variables of interest that are associated with cardiometabolic outcomes such as glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and T2D. We assessed risk factors in mothers, fathers, and offspring during preconception, pregnancy, and infancy up to 2 years of age, together covering the first 1000 days of life. The outcomes of interest in this current review were self-reported, physician- or researcher-diagnosed; glucose intolerance; dyslipidemia; hypertension; metabolic syndrome; and T2D.
Data sources, search strategy, and screening criteria
This systematic review protocol was registered to PROSPERO, CRD42022355152. Our search terms for Medline, EMBASE, WEB OF SCIENCE, SCOPUS, and Cochrane CENTRAL (Text S1) relate to eligible citations published in English before August 17, 2022. We included prospective and retrospective longitudinal observational studies which identified risk factors associated with offspring outcomes of our interest. A prospective comparison of treatment allocation effects on outcome was also considered. The exclusion criteria were: studies with outcomes before 2 years, studies reporting only intermediate phenotypes or continuous traits, or studies assessing only endpoints other than the cardiometabolic outcomes of interest. Cross-sectional studies and studies of diseased populations were excluded, except for genetics studies. Two independent reviewers screened at title and abstract level. For accepted citations, two independent reviewers examined the entire manuscript. A third reviewer resolved conflicts at all screening levels. The consensus of the full group was followed on conflicts not resolved by the third reviewer. The covidence online systematic review tracking platform was used for all screenings.
Data extraction and synthesis of results
For eligible manuscripts, we developed and piloted a data extraction template. In addition to manuscript information, the data-set included details about the study design, demographics of the population, exposure and outcome assessment, and diagnostic criteria. A total of 8 broad categories of exposures were recorded: (i) parental lifestyle factors, (ii) parental physical factors, (iii) environmental factors, (iv) pregnancy-related factors, (v) birth anthropometrics, (vi) feeding patterns, (vii) infant anthropometrics, and (viii) cord blood biomarkers.
Quality assessment (risk of bias) and synthesis
For cohort studies, case‒control studies, and randomized clinical trials (RCTs), the Joanna Briggs Institute (JBI) critical appraisal tool was used to assess the quality of the included studies 20. For case‒control studies, 9 items were used to assess quality, including group’s comparison, assessment of the exposure and outcome, confounding, exposure period of interest, and statistical methodology. For cohort studies, 11 items were used to assess quality, including recruitment, assessment of exposure and outcome, confounding, statistical methodology, and follow-up. In RCTs, 10 items were evaluated according to the JBI criteria, including selection and allocation, intervention, administration, outcome assessment, follow-up, and statistical analysis. Using the guidelines, each JBI item was categorized as ‘yes’, ‘no’, ‘unclear’ or ‘not applicable’. Any uncertainties in the assessment were further discussed by the entire research team. Studies with overall scores of 50% or less on questions answered with ‘Yes’, 50% until 70% questions answered with ‘Yes’, and 70% or more questions answered with ‘Yes’, were considered low, moderate and high quality, respectively.
Quality assessment of risk factors for prediction and prevention
For quality assessment, we selected those risk factors we considered consistently associated with childhood obesity. Generally, we defined consistently associated risk factors as those for which more than 50% of studies report an association in the same direction, in at least two studies of moderate or high quality. Based on the International Life Sciences Institute (ILSI) Europe Marker Validation Initiative, we developed and piloted a criteria template for quality assessment of risk factors 21 (Table S1). Based on methodological aspects, the study objective, modifiability, predictiveness, and early-life risk factors were evaluated. Two independent reviewers scored the risk factors at four different levels: proven, strong, moderate, and low. The whole group discussed conflicts until consensus was reached.
RESULTS
Characteristics of participants and the screening process
Figure S1 shows the flow chart for screening all studies (n=35584), of which about one-half were excluded as duplicates (n=17974). A total of 17610 full-text or abstract articles were screened by two independent reviewers, and 16059 were scored irrelevant. Full-text screening was conducted on 1551 studies; 302 were extracted, and 112 were analyzed for associations. Ultimately, 68 articles were meeting the inclusion criteria and were included for the current systematic review.
Positive, negative, and null associations between the main categories of risk factors preconception, pregnancy and birth, and early infancy and cardiometabolic outcomes: glucose intolerance (A), dyslipidemia (B), hypertension (C), metabolic syndrome (D), and type 2 diabetes (E).
The studies were published between 1991 and 2022 in 31 different countries, with the majority of the data coming from America, Asia, and Europe. Among the 68 articles included, 59 were observational studies, of which 38 were prospective and 21 were retrospective studies. Seven were case-control studies, and only 2 were RCT interventional studies. The sample sizes of the different studies ranged from 50 to 155,411 participants. Table 1 shows the characteristics, the main results and the cardiometabolic outcome definitions for each study.
Associations of exposures and risk factors for cardiometabolic outcomes in childhood
We reported 229 associations between the main categories of risk factors in the first 1,000 days of life for cardiometabolic outcomes in childhood, including glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and T2D (Table S2). These associations were classified as positive, negative, or none. The majority of associations were positive (n=162) for all cardiometabolic risk factors, with pregnancy and birth categories exhibiting the greatest associations (Figure 1).
We found that risk factors including birth anthropometrics, pregnancy complications, cord blood biomarkers, lifestyle, physical pregnancy, infant anthropometrics, and feeding patterns categories were associated with glucose intolerance later in childhood. Of 16 evaluated studies, 2 were case-control studies22,23, 11 were prospective studies24–34, 1 was a retrospective study35, and 2 were RCTs36,37. These associations were mainly studied in pregnancy and birth period, and to a lesser extent in the early infancy period; however, investigation of the association of exposure and risk factors during preconception and later glucose intolerance was not reported in any of the reviewed studies. There were 23 associations (82% were positive associations between risk factors and glucose intolerance in children). Among them, 18 were identified positively associated pregnancy and at birth period category22,23,25–28,32–34,37,38 (Figure 1A).
11 unique studies reported data on dyslipidemia in children, 2 case-control studies22,39, 4 prospective studies24,28,32,40, 4 retrospective studies35,41–43, and 1 RCT37. Within these studies, 22 different risk factors included in birth anthropometrics, pregnancy complications, lifestyle, sociodemographics, physical pre- and pregnancy, infant anthropometrics, and feeding patterns categories, were investigated. Pregnancy and birth factors were the most reported period for associations with dyslipidemia. Indeed, 14 risk factors were significantly associated with an increased risk for dyslipidemia in childhood 22,28,32,37,39 (Figure 1B).
We identified 37 studies evaluating risk factors for hypertension in children. The majority were prospective studies (n=21) 24,28,29,32,40,44–59, followed by retrospective studies (n=12) 35,41,43,60–68, case-control studies (n=2) 22,69 and RCTs (n=2) 36,37. We identified 127 associations between early life factors and childhood hypertension, of which 66% significantly associated with increased risk for hypertension in childhood22,28–32,37,43,45–50,53,54,56,57,61–68 (Figure 1C).
We identified 22 studies related to childhood metabolic syndrome, including 4 case-control studies22,70–72, 12 prospective studies31–33,58,73–80, 5 retrospective studies66,81–84, and 1 RCT37. Figure 1D shows how associations were distributed during preconception, pregnancy, and infancy. There were 64 different risk factors (including physical pre- and pregnancy, birth anthropometrics, sociodemographics, pregnancy complications, cord blood biomarkers, lifestyle, infant anthropometrics, and feeding patterns categories) that associated with metabolic syndrome as outcome, from which the majority investigated risk factors during pregnancy and birth. A total of 77% of the associations were positive (49 of 64 total associations), observed in the pregnancy and birth period (40 positive associations)22,31–33,37,58,66,70–72,74,76,77,81–83, as well as the preconception (3 positive associations) 74,75 and early infancy period (6 positive associations) 33,70,73,78,79 were identified.
There were 9 studies that reported associations between risk factors (including physical prepregnancy, birth anthropometrics, sociodemographics, pregnancy complications, cord blood biomarkers, lifestyle, environmental, and feeding patterns categories) for developing T2D in children. Of these, 1 was a case-control study85, 4 were prospective studies27,30,38,86, and 4 were retrospective studies42,87–89. For this outcome, we did not find RCTs. Thirty-one associations were evaluated. Among them, 23 risk factors were positively associated with T2D27,30,38,85–89, 7 were not associated42,85,88,89, and only 1 reported a decreased risk for this outcome 87. Most of the associations analyzed were risk factors during pregnancy and birth periods27,30,42,85,86,88,89, followed by early infancy86–88 and the preconception period87 (Figure 1E).
Preconception risk factors associated with childhood cardiometabolic outcomes
During the preconception period, 7 studies identified exposures and risk factors to childhood cardiometabolic outcomes, 6 of which were observational studies, 5 of which were prospective studies45,46,74,75,80, 1 of which was a retrospective study87, and 1 of which was a case-control study39. No RCTs were reported. We identified 4 risk factors measured during preconception, 3 of which were associated with a higher cardiometabolic risk in children. The prepregnancy maternal BMI and the prepregnancy paternal BMI as exposures were linked positively to cardiometabolic risk factors39,45,46,74,75. Moreover, pregestational diabetes and diabetes were also positively associated with cardiometabolic risk87. Finally, 1 study reported that higher concentrations of plasma folate, serum vitamin B12 and plasma homocysteine were associated with metabolic syndrome80 (Table S2). Our current review of the available evidence shows that physical prepregnancy risk factors during the preconception period, which includes both paternal and maternal exposures, were mostly associated with metabolic syndrome and hypertension in childhood45,46,74,75 (Figure 2A).
Positive, negative, and null associations between risk factors physical prepregnancy (A), birth anthropometrics (B), pregnancy complications (C), sociodemographic (D), cord blood biomarkers (E), lifestyle (F), physical pregnancy (G), environmental (H), infant anthropometrics (I), feeding patterns (J) and cardiometabolic outcomes: glucose intolerance, dyslipidemia, hypertension, metabolic syndrome and type 2 diabetes.
Pregnancy and birth risk factors associated with childhood cardiometabolic outcomes
The vast majority of articles included in this study (58 of 68) reported on associations between pregnancy and birth exposures and cardiometabolic risk factors in childhood. Of those, 48 were observational studies (31 prospective studies25–34,38,44–58,74,76,77,80,86 and 17 retrospective studies41,42,60–68,81–83,87–89), 8 were case-control studies22,23,39,69–72,85, and 2 were RCTs36,37. Risk factors identified during pregnancy and birth period were the most frequently studied, with 197 (86%) associations for all cardiometabolic outcomes. 7 categories of risk factors were evaluated: birth anthropometrics, sociodemographics, pregnancy complications, lifestyle, physical, environmental, and cord blood biomarkers (Figure S2B). The most frequently assessed characteristics were birth anthropometrics (n=75), sociodemographic data (n=47), and pregnancy complications factors (n=34). Concerning the birth anthropometric category, we found that the majority of the associations were positively associated with cardiometabolic risk, highlighting hypertension, and metabolic syndrome as the most frequently associated22,28,34,39,45,46,54,56–58,61,63–66,68,71,82,83,86–89. However, we reported a high number of associations that were null33,41,45,48,65,66,68,69,71,72,85, most of which were of risk factors with childhood hypertension, and only 5 were negative associations44,52,63,64. Alike the preconception period, hypertension 22,28,34,45–48,54,56,57,61,63–66,68 and metabolic syndrome 22,28,34,45–48,54,56–58,61,63–66,68,71,72,82,83 were the two outcomes most frequently associated with birth anthropometrics (Figure 2B). We identified 7 birth anthropometric risk factors correlated with cardiometabolic outcomes. Among them, the birth weight of the child had the most frequently associated with all of outcomes (n=25) 28,33,39,44,45,48,52,54,58,61,63–66,68,69,71,72,82,83,87, 13 of which were linked to an increased risk for childhood cardiometabolic issues28,39,45,48,54,58,61,63,82,83,87, while 9 had no relationship33,45,65,66,68,69,71,72, and 3 had inverse associations 44,52,64 (Figure S3A). The most frequent cardiometabolic outcome associated with birth weight was hypertension. Furthermore, low birth weight or small size for gestational age (LBW/SGA) was identified as a risk factor for glucose intolerance, hypertension, metabolic syndrome and T2D: 10 positive associations15,34,54,58,61,64,65,68,86,88,89 and 6 no associations were observed45,48,66,72,85 (Figure S3B). High birth weight (HBW) or large for gestational age (LGA), was evaluated by several studies as a potential risk factor for cardiometabolic outcomes. In fact, we identified 15 positive associations22,34,58,63,66,72,87,89 for HBW/LGA and cardiometabolic disorder risk and 4 no associations48,68,69,85 (Figure S3C). As part of the birth anthropometrics group, other exposures were identified, such as gestational age39,45,56,58,63,71,82,83, preterm birth41,45,47,56,57,71,87, birth length44,61,63,82,83, and mode of delivery (cesarean delivery)46,58,87. Despite the small amount of data, most associations were positive, highlighting both hypertension and metabolic syndrome as the most adverse cardiometabolic outcomes.
Regarding the sociodemographics category, most associations reported a risk of developing all cardiometabolic outcomes28,39,45–48,51,53,54,57,58,61,64,77,81,87–89. Figure 2C shows that sociodemographic factors increased the risk of hypertension28,45–48,54,57,61,64 and metabolic syndrome35,58,77 during childhood. The strongest associations were found between ethnicity28,45,47,48,51,54,57,58, maternal education39,45–47,54,58,61,64,77, and income45,61,64,77,87,88. Ethnicity was mostly associated with hypertension28,45,57,58 (Figure S3D). Similarly, low maternal education was positively associated with the risk to develop hypertension, and metabolic syndrome45–47,54,61,64 (Figure S3E). We also reported that low parental income is mostly associated with higher risk to develop hypertension, metabolic syndrome and T2D45,61,64 (Figure S3F). Cardiometabolic outcomes were also influenced by other parental factors such as employment61,64,77, socioeconomic status47,77,89, and marital status47. Pregnancy complications were associated with all cardiometabolic outcomes22,23,25,27,29,30,32,33,39,42,46,47,49,57,62- 64,70,81,85-87, highlighting hypertension22,29,32,46,47,49,57,62–64 and glucose intolerance22,23,25,27,29,30,32,33 as the most reported associations (Figure 2D). Among them, we identified gestational diabetes, pregnancy hypertension, and preeclampsia as the main exposures associated with cardiometabolic outcomes in children. Almost all associations between these risk factors were positive. All 27 studies on gestational diabetes had a positive association with all outcomes22,23,25,27,29,30,32,33,39,47,49,63,70,81,85–87 (Figure S3G). Regarding pregnancy-related hypertension, we found associations for dyslipidemia39 (1 positive association), hypertension46,63,64 (3 positive associations), and metabolic syndrome81 (1 no association) (Figure S3H). We evaluated the associations of preeclampsia with dyslipidemia42 (1 for no association), hypertension42,46,47,57,62,81,87 (4 for positive associations and 1 for no association), metabolic syndrome81 (1 for no association), and T2D42,87 (1 for positive association and 1 for no association) (Figure S3I).
In addition to the above categories, we also evaluated cord blood biomarkers, lifestyle, physical pregnancy, and environmental exposures (Figure 2). For the cord blood biomarkers, we found 1 study reporting that telomere length was positively associated with hypertension48; plasma folate, serum vitamin B12, and plasma homocysteine were negatively associated with metabolic syndrome80; and umbilical insulin levels were positively associated with glucose intolerance and T2D28. No evidence was found between plasma folate concentrations and hypertension51 (Figure 2E). The lifestyle category (including exposures such as smoking, alcohol, drugs, and diet) presented 24 associations with cardiometabolic outcomes36,37,39,47,51,53,54,58,64,67,77. The most frequent associations were related to hypertension during childhood36,37,47,51,53,54,64,67 (Figure 2F). Concerning physical pregnancy (such as parental anthropometrics, parental BMI, gestational weight gain), we identified 15 associations22,39,45,47,53,58,63,64,74,76,81, 14 of which were related to an increased risk of cardiometabolic outcomes, especially hypertension22,45,47,63,64 and metabolic syndrome22,58,74,76,81 (Figure 2G). The results in the environmental category contrasted, being mostly associated with hypertension. In this category, there are 7 associations, 3 of which were related to an increased risk of hypertension 61,64,67, while the other 2 were not associated with T2D69,88. Finally, two negative associations were also found with hypertension55,64 (Figure 2H).
Early infancy-related risk factors associated with childhood cardiometabolic outcomes
We identified 19 studies reporting associations between risk factors in early infancy and adverse cardiometabolic outcomes in childhood. 2 were case-control studies58,70, 7 were retrospective studies35,43,61,64,84,87,88, and 10 were prospective studies24,40,44,48,53,59,73,78,79,86. We found 24 associations between risk factors and cardiometabolic outcomes in children, of which 15 positive associations were reported33,40,43,48,53,61,68,70,73,78,79,86–88, 4 negative associations were observed44,64,70,87, and 5 no associations were found24,35,51,59,84 (Figure S2C). During the period between birth and 2 years of age, feeding patterns and infant anthropometrics were the main risk factors evaluated. Anthropometrics in infancy were associated with dyslipidemia40,43 in 2 out of 3 studies, with hypertension40,43,48,53,61 in 5 out of 8 studies and with metabolic syndrome73,78,79 in 3 out of 3 studies. Only one study investigated anthropometrics and glucose intolerance, and did not report a significant association24, for T2D outcomes, no studies were identified. We did not find any study reporting infant anthropometrics or cardiometabolic outcomes (Figure 2I). Feeding patterns, which means breastfeeding, were the only category correlated mostly with a decreased risk for cardiometabolic outcomes. Indeed, 16 associations were reported, 8 of which were associated with a lower risk for glucose intolerance33,68, hypertension68, metabolic syndrome33,70,78 and T2D86–88. Only one study investigated breastfeeding and dyslipidemia, but did not report any significant associations 35, while we found 3 associations reporting a lower risk for hypertension64, metabolic syndrome70 and T2D87 (Figure 2J).
Risk of bias assessment
Case-control studies, observational studies (prospective and retrospective studies), and RCTs were evaluated for risk of bias. Of the total studies analyzed (n=68), only 15% had low risk of bias. Regarding case-control studies, we found that 38% of studies showed low bias risk (n=8). For observational studies, the rate was 10% of low risk of bias (n=58). The percentage of these were distributed between prospective studies, which showed 16% of low bias risk (n=38). However, for retrospective studies, all exhibited at least one item unclear and/or high risk of bias (n=20). For the randomized control trials, 50% showed low risk of bias (n=2) (Figure S3-S13). In case-control studies, we identified that unreliable exposure measurements, and inadequate confounding strategies were the main conditions associated with high risk of bias (Figure S14A). In observational studies, we identified several conditions associated with high risk of bias, such as loss to follow-up, unreliable exposure measurements, and inadequate confounding strategies (Figure S14B). In the evaluated RCTs, we did not find high risk of bias for any of the included studies.
Assessment of risk factors for prevention and prediction
We evaluated the quality of the assessment of risk factors included in the present systematic review (Figure 3). Risk factors with high scores on reproducibility, accuracy, standardization, stability, and technical variation were for the preconception period: maternal prepregnancy weight, in pregnancy and birth: maternal age, maternal and paternal education, income, parity, anthropometrics, paternal BMI, and gestational weight gain continuous; and for the early infancy period: breastfeeding, and infant BMI change. Risk factors quality scores were generally moderate to strong, except for cord blood biomarkers, which were low. A change in the marker that is linked with a change in the endpoint in one or more target populations was proven for maternal prepregnancy weight, birth weight, LBW/SGA, HBW/LGA, and gestational diabetes. The strongest theoretically modifiable risk factors on an individual level were diet, and the quality of the majority of risk factors were scored low based on evidence from the available intervention studies. Finally, for prediction, the proven risk factors were maternal prepregnancy weight, gestational diabetes, pregnancy hypertension, maternal education, and parental BMI. Concentrations of plasma homocysteine, parity, and marital status had the lowest scores (Figure 3).
Quality assessment of identified risk factors based on ‘Guideline on quality assessment of risk factor’ from Calder et al. 2017. Four different levels has been represented: proven (hard green), strong (light green), moderate (yellow), and low (red).
DISCUSSION
In the present systematic review, we examined the associations of early-life exposures during preconception, pregnancy and birth, and the early infancy period, with adverse cardiometabolic outcomes during childhood such as glucose intolerance, dyslipidemia, hypertension, metabolic syndrome, and T2D. We identified 59 observational studies (of which 38 prospectives, and 21 retrospectives), 7 case-control and 2 intervention studies, and we found 228 reported associations between risk factors and all cardiometabolic outcomes, of which 162 were positive. Exposures during pregnancy and at birth were the most frequently studied in relation to later life cardiometabolic risk factors. More specifically, high or low birth weight, sociodemographics factors including education, income and ethnicity, and pregnancy complications (including gestational diabetes and preeclampsia) were the most frequently reported risk factors associated with cardiometabolic risk. Additionally, we identified metabolic syndrome and hypertension as the most frequently and positive associated cardiometabolic outcomes reported.
This systematic review offers relevant information on the association between risk factors in the first 1,000 days and childhood cardiometabolic health. It encompasses 3 different timings of exposure: preconception, pregnancy and birth, and the early life period, as well risk factors such as parental and offspring conditions that might affect the cardiometabolic profile for up to 18 years. In preconception period, both paternal and maternal BMI were associated with adverse cardiometabolic health in childhood, including dyslipidemia, hypertension, metabolic syndrome and T2D. Maternal obesity before pregnancy was associated with impaired cardiometabolic health. Furthermore, paternal prepregnancy BMI was associated with adverse cardiometabolic profiles in offspring such as a higher childhood BMI, systolic blood pressure, insulin levels, and lower high-density lipoprotein cholesterol75, although strongest/most profound associations were observed for maternal prepregnancy BMI. In addition, during the preconception period, we found that pregestational diabetes was positively associated with T2D in children in one study. Thus, either pregestational or gestational diabetes in the mother has been associated with childhood-onset T2D73,87.
Pregnancy and birth were the most frequently studied period to investigate associations of early life exposures and later in life cardiometabolic outcomes. Our results add support to the developmental origins of health and disease (DOHaD) hypothesis, which states that adverse intrauterine exposure causes several metabolic changes during fetal growth, increasing the risk of cardiometabolic disease development 90. Birth anthropometrics, particularly having a high or low birth weight, showed a consistent positive association with all cardiometabolic outcomes, being hypertension the most frequently reported. A cross-sectional study suggests that obesity and hypertension in children and adolescents may be modifiable by birth weight91. Using Mendelian randomization, a study examined whether birth weight could be a risk factor independent of the intrauterine environment for cardiometabolic outcomes92. In the first 1,000 days of life, children born outside the normal birth weight range tend to regain their genetically determined growth trajectory to compensate for both restricted and excessive intrauterine growth93. In this way, the rapid weight gain in SGA infants increases the risk of obesity and metabolic diseases such as insulin resistance, impaired glucose tolerance, T2D, hypertension, and ultimately cardiovascular diseases94,95. It has also been demonstrated that birth weight has a direct impact on plasma triacylglycerols and insulin resistance, whereas obesity has a direct impact on hypertension, impaired glucose metabolism, and hypertriglyceridemia. Considering that birth weight and obesity cannot be separated from metabolic syndrome and its components, large prospective studies involving normal-weight individuals are needed. A recent meta-analysis showed that LBW might increase metabolic syndrome risk in childhood and adulthood96. To assess the potential interaction effects of LBW on metabolic syndrome, more high-quality studies are needed.
At least 30% of women exhibit pregnancy complications, which are among the leading causes of mortality among pregnant women97. These complications can be caused by the natural changes in the mother’s body as a response to her pregnancy, including increased cardiac output, an increased inflammatory response, and metabolic abnormalities related to insulin resistance98. As a consequence, pregnant women are more prone to develop metabolic disorders such as hypertension and gestational diabetes mellitus 97,99. Worldwide, pregnancy-related hypertension is the third leading cause of maternal mortality, affecting between 3% and 10% of pregnancies. A higher incidence of stillbirths and large for gestational-age babies is associated with gestational diabetes, which affects approximately 7–10% of pregnancies worldwide100,101. Both pregnancy complications are strongly associated with an increased risk of T2D and hypertension. In this systematic review, we found that gestational diabetes, which is one of the most common current diseases during pregnancy102, is associated with all cardiometabolic outcomes, with glucose intolerance having the greatest association. Hypertension during pregnancy is associated with dyslipidemia, hypertension and metabolic syndrome in children. We found associations between preeclampsia and dyslipidemia, hypertension, metabolic syndrome and T2D. Preeclampsia, like pregnancy hypertension, is associated with increased risks for hypertension, vascular disease, stroke, and venous thromboembolism later in life103. Therefore, gestational diabetes, pregnancy hypertension, and preeclampsia are risk factors for metabolic diseases, and the importance of pregnancy history to future health should be discussed with all care providers and women.
In the current research, we observed that several parental sociodemographic factors have been linked to adverse cardiometabolic outcomes. Among these factors, several risk factors, such as ethnicity, maternal education, and income, were associated with hypertension and other cardiometabolic outcomes. We found that ethnicity is mostly linked to hypertension, metabolic syndrome, dyslipidemia, and glucose intolerance. It is well known that ethnicity differs in cardiovascular disease prevalence104, with minorities having a more adverse cardiovascular profile than some European countries. Parental prepregnancy factors, pregnancy factors, and childhood BMI all contribute to these differences. In terms of income, we found a positive association with hypertension in children. The effect of household income has been analyzed for Black adolescents. This showed that households with incomes greater than $75,000 were more likely to have hypertensive-range blood pressure due to their black race compared to households with incomes less than $75,000 in United States of America, however, further research is needed 105. Both maternal and paternal education were associated with cardiometabolic outcomes, especially hypertension. In particular, evidence suggests that educational status might affect the development of adverse cardiometabolic health106,107. The majority of the studies show that a low level of maternal and paternal education is associated with a greater risk of hypertension during childhood108,109. Taken together, these data indicate that parental education level, rather than the individual’s education level, is independently better, showing that infants with low parental education levels are more likely to develop hypertension110.
Regarding the early infancy period, we found that breastfeeding is associated with a lower risk of cardiometabolic outcomes during childhood. Exclusive breastfeeding is highly recommended for infants for at least the first 6 months of life111, and breastfeeding has been shown to reduce the risk of obesity and chronic noncommunicable diseases, such as T2D112. The protein content of infant formulas is traditionally higher than that of human milk, and all essential amino acids are present in sufficient amounts. While these factors promote weight gain and growth, they also increase the risk of obesity and metabolic diseases113. For a formulation to be effective, its protein content needs to be as close as possible to that of human milk. Concerning the early infancy period, we also found that infant anthropometrics, especially weight gain and changes in BMI, are associated with hypertension and metabolic syndrome. These findings are relevant since previous studies have reported that rapid weight gain during infancy is associated with high blood pressure at 17 years43. Furthermore, rapid weight gain during infancy until 6 months could predict metabolic risk at 17 years73; therefore, maintaining or controlling early infancy weight gain might help to prevent future adult cardiovascular disease risks.
The main strengths of this systematic review has been the extensive search of the literature until 2022 (17610 articles in the first screening), and the implementation of at least two independent reviewers across the phases of screening and extraction, along with a third independent reviewer to resolve conflicts. The quality of the manuscript has been performed by evaluating the risk of bias assessment based on JBI guidelines. Using a standardized procedure, at least two independent reviewers conducted the review. All reviewers discussed to resolve the conflicts. As part of our attempt to minimize bias in the selection of studies, we adhered to predefined inclusion and exclusion criteria. By including a large body of evidence, we were able to provide a comprehensive and robust overview of the available evidence. We only identified 2 RCTs and all the associations have been based on observational research, either prospective or retrospective studies. Hence, causality cannot be established between the identified risk factors and cardiometabolic outcomes. However, there is a huge evidence about the potential risk factor to develop hypertension and metabolic syndrome later in childhood.
Conclusions
We found that parental prepregnancy weight, birth weight, SGA, LGA, pregnancy complications such as hypertension and gestational diabetes, infant weight gain, and sociodemographics factors such as income, maternal education level and ethnicity, and not breastfeeding, were associated with cardiometabolic outcomes, highlighting hypertension and metabolic syndrome in children. These findings provide relevant information early-life risk prediction and modifying several behavioral patterns from conception to infancy, not only from mother to father but also from preconception to infancy. Finally, this systematic review provide useful information that could be applied in clinical practice to prevent or predict the metabolic complications later in childhood. Future studies focused on preconception factors, paternal factors, and sociodemographics are urgently needed to prevent metabolic disease.
Article information
Author contributions
MBH: provided extraction of the data and provided critical intellectual feedback and completed quality assessment of papers, and drafted the final version of manuscript. FJRO: Drafted the initial version of the manuscript, completed the review, extraction, and quality assessment of papers, and drafted the final version of manuscript. SMB: Drafted the initial version of the manuscript, completed the review, extraction, and quality assessment of papers. ASJK: Drafted the initial version of the manuscript, completed the review, extraction, and quality assessment of papers. EFV: provided extraction of the data and provided critical intellectual feedback. MAB: provided extraction of the data and provided critical intellectual feedback. MCC: provided extraction of the data and provided critical intellectual feedback. JvD: Set up the initial aim of the study and study protocol, provided extraction of the data and provided critical intellectual feedback. PI: provided extraction of the data and provided critical intellectual feedback. KK: provided extraction of the data and provided critical intellectual feedback. CAvLB: provided extraction of the data and provided critical intellectual feedback. RG: Set up the initial aim of the study and study protocol, completed the review, extraction, and quality assessment of papers. AG: Set up the initial aim of the study and study protocol, completed the review, extraction, and quality assessment of papers, and revised the final version of the manuscript. All authors approved the final version for publication.
Conflict of interest disclosures
None reported.
Data Availability
All data produced in the present work are contained in the manuscript
Acknowledgements
RG received funding from the Netherlands Organization for Health Research and Development (ZonMW, grant number 543003109). RG has received funding from the European Union Horizon 2020 research and innovation program under the ERA-NET Cofund action (No. 727565), European Joint Programming Initiative A Healthy Diet for a Healthy Life (JPI HDHL), EndObesity, ZonMW Netherlands (No. 529051026). SMB was supported by the KNAW Ter Meulen Grant/KNAW Medical Science Fund, Royal Netherlands Academy of Arts & Sciences, and by ILSI Europe. ASJK was supported by ISLI Europe. MCC acknowledges the support by the Spanish Ministry of Science and Innovation (MCIN) research grant (ref. PID2022-139475OB-I00) and the award from MCIN/AEI to the Institute of Agrochemistry and Food Technology (IATA-CSIC) as Centre of Excellence Severo Ochoa (CEX2021-001189-S MCIN/AEI /10.13039/ 501100011033). EFV was supported by a Predoctoral grant awarded by the Spanish Ministry of Science and Innovation and the European Social Fund Plus (ESF+) for the training of doctors within the framework of the State Plan for Scientific, Technical and Innovation Research 2021 - 2023. (ref. CEX2021-001189-S-20-1). This paper will be part of Marco Brandimonte-Hernández’s doctorate, which is being completed as part of the “Nutrition and Food Sciences Program” at the University of Granada, Spain. FJRO was supported by the Spanish Government (Juan de la Cierva program, Ministry of Science and Innovation). This Project was executed in collaboration with the Early Nutrition and Long-term Health Task Force of the European Branch of the International Life Sciences Institute (ILSI Europe). Industry members of this task force are listed on the website www.ilsi.eu. For more information, please contact ILSI Europe by email info{at}ilsieurope.be or call +32 2 771.00.14. The opinions expressed herein, and the conclusion of the project are those of the authors and do not necessarily represent the view of ILSI Europe nor those of its member companies.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵