Objective Obesity and excess adiposity are leading causes of metabolic and cardiovascular morbidity and mortality. Early identification of individuals at risk is key for preventive strategies. We examined the relationship between infant body composition (0–2 years of age) and later (>2 years) health outcomes using a systematic review.
Design We preregistered the study on PROSPERO (ID 288013) and searched Embase, PubMed and Cochrane databases for English language publications using the Medical Subject Headings (MeSH) terms ‘infant’ and ‘body composition’ and ‘risk’ between January 1946 and February 2022. We included studies which assessed infant body composition using predetermined in vivo methods other than body mass index (BMI).
Results We identified 6015 articles. After abstract screening to assess eligibility, we reviewed 130 full text publications. 30 were included in the final assessment and narrative synthesis. Meta-analysis was not possible due to heterogeneity of results. All 30 studies were of high quality and reported associations between infant body composition and 19 different health outcomes after 2 years of age. Outcome measurements ranged from 2 years to 16 years. The strongest associations were found between infant fat mass and later fat mass (7 studies), and later BMI (5 studies). For 11 of the outcomes assessed, there was no relationship to infant adiposity detected.
Conclusions Current evidence, from a small number of studies, suggests a positive association between infant adiposity and future adiposity or BMI, but the validity of infant body composition as a biomarker of future health remains inconclusive. Carefully designed, standardised studies are required to identify the value of infant body composition for predicting later health.
Trial registration PROSPERO: 288013
- child health
Data availability statement
No data are available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
WHAT IS ALREADY KNOWN ON THIS TOPIC
The association between infant body composition and later health outcomes is unclear.
Early identification of individuals at risk of later disease is key for preventive strategies to reduce mortality and morbidity.
WHAT THIS STUDY ADDS
The strongest associations were found between infant fat mass and later fat mass (seven studies), and later body mass index (five studies).
There are limited studies which assess other health outcomes.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Carefully designed, standardised studies are required to identify the value of infant body composition for predicting later health.
The importance of a child’s first 1000 days, from conception to their second birthday, to long-term health outcomes such as non-communicable diseases (NCDs) has been widely discussed.1 2 The Global Disease Burden report highlights that mortality and morbidity are highly attributable to NCDs, with both undernutrition and overnutrition playing fundamental roles.3 The top four NCDs of global public health interest are cardiovascular disease, cancer, chronic respiratory disease and metabolic disease (type 2 diabetes and obesity). These account for about 74% of all NCD mortality and are associated with suboptimal early life conditions.3 4
It is of relevance to global public health to be able to identify groups or individuals at risk of developing NCDs, to allow for early detection and appropriate interventions to reduce risk of future disease, before ill-health is established. One of the measures used as a physiological marker of interest for adverse prenatal and early life exposures is infant adiposity, measured as the balance between accurate measures of fat-free mass (FFM) and fat mass (FM) (table 1). After infancy, FM is positively associated with risk of NCD and negatively associated with cognitive function in children.5 In contrast, low FM, such as in growth-restricted infants is associated with poorer short-term outcomes, which especially, if followed by a period of compensatory rapid weight gain in infancy, can predispose to excessive visceral adiposity and associated adverse health outcomes.6 7 Fetal (eg, gestational age, birth weight for gestational age, intrauterine environment), maternal (eg, diet, gestational diabetes, smoking status, placental function) and infant feeding (eg, type of milk feeding, complimentary food quality) factors can affect infant body composition.8–11 As reviewed elsewhere, these factors are also associated with increased risk of cardiometabolic disease in adulthood.12 13
Adult body composition is strongly associated with many risk factors,14 however, the association between infant body composition and later health outcomes is unknown. A systematic review by Bander et al15 assessed the association between body composition in the first 5 years of life and later cardiometabolic diseases.15 They found that low body mass index (BMI) in infancy and higher BMI in childhood seems to be associated with non-communicable disease in adulthood, however few studies measured body composition using validated methods, and instead used BMI.16
We hypothesise that actual infant body composition before a child’s second birthday, measured by validated methods of body composition, is associated with later health outcomes. Adiposity in early life and its impact on later health outcomes are not straightforward to assess. FM accretion is sensitive to in utero and early life exposures, and changes with age and sex.17 Human babies are born with much more adipose tissue compared with other mammals and infant body composition is complex. Previous work has reviewed the challenges in measuring infant body composition, with considerable variability between groups according to gestational birth age, length, weight and mode of feeding.18
The aim of this review was to evaluate the evidence of associations between infant body composition and later health outcomes, and identify gaps in the literature to help direct future research.
This systematic review included studies that reported the association between infant body composition and subsequent health outcomes. This study was registered with PROSPERO (ID 288013).
Eligible studies were human studies published in English from January 1946 to February 2022, focused on measuring infant body composition using a predetermined in vivo method (table 1) and reported at least one health outcome after 2 years of age. Retrospective and prospective observational studies were included. Studies that focused on BMI as a measure of body composition in infancy, conference abstracts and other review studies were excluded. BMI was included as an outcome.
A comprehensive search of published articles in the Embase, PubMed and Cochrane (via Imperial OVID Medline) databases was performed. This study was conducted aligning with the PRISMA guidelines. The search used the following MEDLINE Medical Subject Headings terms: (((Infant) OR (Neonat*)) AND ((Body Composition) OR (fat mass)) AND ((Risk) OR (Predict*))). Search filters used were humans and abstracts available. We manually searched the bibliographies for pertinent studies.
Data extraction and quality assessment
Three independent investigators (LM, EP and FA) evaluated the studies for eligibility. Publications were double screened based on title and abstract screening then based on assessment of the full text using systematic review software Covidence under Imperial College London license.
Study data were extracted including general information, surname of the first author, the year of publication, study population, type of outcome and definition, study design timing of outcome evaluation, statistical method, and adjusted ORs or mean differences between the study groups by FA and LM. When adjusted ORs or risk ratios were not reported by the studies included, odds, risk or other measure of outcome were extracted for analysis.
Two researchers independently reviewed each article for methodological quality using the validated Joanna Briggs Institute (JBI) quality checklist for observational studies.19 This tool was used to assess four domains with seven questions. Each study was rated as low (0–2), moderate3–5 or high quality (>5). Inconsistencies in study inclusion, data extraction or quality assessment were resolved by discussion and consensus among the three researchers.
Data synthesis and analysis
A narrative synthesis of the findings from the included studies alongside summaries of exposure effects for each study with risk (for dichotomous outcomes) or standardised mean differences (for continuous outcomes) is presented. The impact of infant adiposity on reported outcomes is described for each paper and a table summarising the effects of infant body composition on each outcome created for an overview of the evidence.
Figure 1 presents a flow chart of the search. Initially, the search strategy resulted in 6014 articles. A further study was added through manual searching. Forty-three studies were excluded as they measured body composition using BMI, or they did not measure it in infancy. Thirty-four were removed due to incorrect study design (eg, narrative reviews, cross-sectional analyses and protocols). After abstract screening, 130 full-text publications were reviewed for eligibility and 30 were selected for inclusion.
A meta-analysis was not possible due to the heterogeneity of reporting. The studies often included multiple measures of exposures and multiple outcomes, making the clear grouping of change in OR based on infant adiposity difficult. Infant body composition, although measured in all the studies, was not always clearly linked to outcome in the analyses.
Since running the search, a study has been published on the topic which does report infant body composition and outcomes in a standardised way.20 We have therefore taken this study into consideration in the discussion.
All studies were reported as being of high quality in the quality assessment (>5 score out of 7), making them low risk of bias. The number of included infants per study ranged from n=34 to n=9876, with a total n=20 975 infants included in the review. Most of the studies (24 out of the 30) measured infant body composition within the first 3 months of age. The age of outcome measurement spanned from 2 years to 16 years of age. Most of the studies were prospective cohort designs. There was heterogeneity in the methods used to measure infant body composition, and in the reporting of outcomes; for example, for body composition some studies reported FM at a specific timepoint, others reported FM accretion. Many of the studies investigated the association of several parameters with several outcomes. For example, Catalano et al reported on over 15 different outcomes.21 Studies separately assessed male and female infants and mostly recruited healthy infants. The complete data extraction table can be found as web only data (see online supplemental table).
Despite the heterogeneity and complexity of the studies, some themes emerged (table 2). There was a relationship between increased infant adiposity and future anthropometry. From the nine studies that assessed the relationship between FM in infancy and later in life, six found a solely significant positive relationship and one found no significant relationship. Duncan et al found a mixed result; infant FM correlated with FM at 3 years, but not at 2 years.22 Wibaek et al found no link with FM at birth but did find a link with FM accretion from 0 months to 3 months and later FM.23 Santos et al also found that the association was dependent on age, with FM at 24 months but not 1.5 months associated with FM at 6 years.24 There was a relatively consistent association between infant FM and later BMI, with higher FM in infancy associated with higher BMI in later life in five studies.
FM in infancy had some association with blood pressure; of the six studies that assessed blood pressure as an outcome, none showed a clear association other than Krishnaveni et al25 Pfister et al’s26 findings supported these but only when body composition was measured before 4 months.27 Most of the other reported outcomes had no association with FM in infancy (table 2), although the number of studies assessing each outcome was even more limited.
Higher FM in infancy was associated with higher FM and BMI at later ages in seven studies, but was not predictive of hypertension, hypertriglyceridaemia, cognitive function, or other cardiometabolic outcomes between ages 2 years and 16 years. No studies followed up cohorts longer than this. The interpretation of the results was limited by the heterogeneity of results, limited number of studies for each outcome, and lack of consistent reporting for infant FM and the health outcomes, which precluded a meta-analysis. Carefully designed, standardised studies are required to confirm or refute the predictive value of infant adiposity for defined health measures, particularly into adulthood.
Bander et al’s systematic review on the association between early childhood body composition and long-term health outcomes also concluded similar findings, despite differing from ours in that it assessed body composition in the first 5 years of life.15 They found a lack of studies with standardised measures of childhood body composition and adult health outcomes. They found that low BMI at birth and higher BMI in later childhood could be associated with poorer health outcomes in adulthood, aligning with the principles of the capacity-load model, which suggests that accelerated fat accretion is a factor to be further investigated.28
Infant FM and later body composition
Our findings suggest that infants with a higher FM at birth could be at higher risk of later high FM and obesity. Studies which measured FM at several time points from birth to second birthday show that body composition after 4 months may be a good indicator of later FM. Most studies show raised BMI is associated with increased FM in infancy. In future studies, it will be useful to differentiate visceral from subcutaneous FM as a marker of later health since the former is known to be associated with more unfavourable health.29 In addition, the speed at which FM is gained seems to play an important role in later FM, so measuring FM accretion in the first months of life could be a more useful measure than a single cross-sectional measure.
The low number of included studies means there is insufficient evidence to draw conclusions about other cardiometabolic outcomes. Future studies that assess FM distribution and speed of FM accretion may look to address whether there is a difference in cardiometabolic outcomes for children who rapidly accumulate excess visceral fat from birth to second birthday, as a group who may be more at risk when considering the Developmental Origins of Health and Disease principles.30–32
Some strength and limitations are to be addressed. We conducted a comprehensive search strategy with 30 included studies and over 20 000 children with a diverse study population and high overall study quality. The studies reported on a variety of cardiometabolic health outcomes of interest for later health.
Several factors precluded a meta-analysis. The types of outcomes and the way the outcomes were measured differed between studies. We had 21 outcomes and of the similar reported outcomes, such as BMI, different papers used different measures including correlation, BMI z-score and BMI. Most of the studies did not include infant body composition as their primary exposure. This limits publication bias but made interpreting the association complicated. All the studies followed up cohorts to 16 years or younger.
Knowledge gaps and future work
Studies designed to assess the relationship between infant body composition and later health are needed, with a standardised approach in terms of exposures, outcomes and timeframes and consideration of potential interactions with sex.
There is a need to define ideal time points to measure infant adiposity as infant feeding, complementary feeding and catch-up growth results in body composition variations.33 34 A baseline between birth and 4 months of age would coincide with immunisation visits when weight is already commonly measured, and would mean that environmental factors have not compounded the association. A second measurement at around 1 year of age would allow assessment of fat accretion. Once standardised body composition measures during infancy are in place, additional efforts to assess FM and FFM up to 2 years of age would be of value to assess their predictive value for future health outcomes. We suggest using skinfold thickness until other methods are more accessible, as this is currently the most accessible measurement method which can be more easily repeated to track changes over time.
Currently, no studies have followed infants across the life course. Therefore, it is unknown whether there is any association sustained into adulthood. Similarly, the association between infant body composition and outcomes that typically present in adulthood (eg, type 2 diabetes) have not been assessed. Modern technologies that accurately measure body composition in infants are only recently widely available. As other technological advances in assessing metabolic outcomes develop, our ability to address these knowledge gaps will increase.
Our review found higher FM in infancy was frequently associated with higher FM and BMI between ages 2 years and 16 years, but no evidence of other associations within this age cohort. We identify several gaps in the current literature, including the limited evidence on FM in infancy and later health outcomes, FM accretion in infancy and health outcomes, and certain body composition measurements. These findings highlight the need to standardise body composition measurements in infancy, track body composition over regular time periods and follow-up health outcomes into adulthood. Further understanding of the interplay between infant body composition and later health outcomes is needed to identify at-risk individuals, implement interventions and optimise health outcomes.
Data availability statement
No data are available.
Patient consent for publication
The authors thank Matthieu Flourakis for help with the initial idea for the project and organisation of meetings for the working group to get the project started, and Rebecca Jones for help with the initial search strategy. The authors also thank Celine Tabche and Noor Al-Rubaye for their contributions to the screening process.
Twitter @dremilyprior, @NeenaModi1
Contributors FA and LM contributed to the final manuscript equally; EC-G drafted the first proposal with input from the taskforce members; NM was the lead supervising author; FA led the registration and conduction of the research; FA, LM and EP conducted paper screening and quality assessment; FA wrote the initial manuscript and led further revisions; LM created the tables, and led the data presentation and manuscript revisions post review; FA is the guarantor; All authors had contributed to the refining of research questions, and read and approved the final manuscript.
Funding This work was supported by ILSI Europe, coordinated by the Early Nutrition and Long-Term Health Task Force.
Competing interests FA was paid an honoraria to complete the work. C-YC is a full-time employee of ILSI Europe.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.