Abstract
Objectives:
Rapid weight gain in infancy is associated with an increased risk of later adiposity. Very rarely, however, exclusively breastfed infants experience excessive weight gain (EWG) during the period of exclusive breastfeeding (EBF) when breast milk is the only source of nutrition. We investigated growth and body composition at 36 months in children experiencing EWG during EBF.
Methods:
Ten infants with EWG during the first 6 months of EBF were followed up at 36 months. The infants had been followed from age 5 months. Examinations included anthropometry, body composition by bioimpedance, and blood samples. Body composition and plasma leptin concentrations were available for eight and five children, respectively.
Results:
From 5 to 36 months, body mass index-for-age z-scores (BAZ) decreased from (mean ± standard deviation) 2.33 ± 0.94 to 1.48 ± 0.57, and only one child still being overweight (BAZ >2). Fat mass and fat mass index (FMI) decreased from 18 to 36 months (4.71 ± 0.56 to 3.57 ± 0.67 kg and 6.50 ± 0.34 to 3.66 ± 0.72 kg/m2, respectively) with a concurrent 45% decrease in leptin. The increase in lean mass was higher than the increase in weight (5.27 versus 3.65 kg, respectively) due to fat loss. There was substantial conformity within the sample in the patterns of body composition change.
Conclusion:
This unusual group of children continued to decrease in BAZ and FMI with a concomitant increase in fat-free mass, indicating an ongoing normalization of body weight and composition.
Introduction
It is important to understand the mechanisms involved in the development of obesity in order to develop effective prevention strategies hereof. Many studies have reported a consistent positive association between rapid weight gain during infancy and increased risk of adiposity later in life.1–3 Postnatal nutrition may be a modulating factor.
Breastfeeding is considered the optimal nutrition for infant growth and development 4 and has been associated with a reduced risk of later obesity.5–7 Overall, breastfed infants show a slower growth velocity after the first 2–3 months compared with formula-fed infants,8,9 and thus breastfeeding has been recommended for obesity prevention. 10 At the same time, a special group of relatively few infants who are exclusively breastfed experience excessive weight gain (EWG) within the period of exclusive breastfeeding (EBF). This means that the EWG occurs when breast milk is the only source of nutrition.
The consequences of this EWG in these EBF infants are poorly understood and not examined extensively.11–13 Furthermore, this group of children is very difficult to recruit and study due to their low prevalence and inability to identify them before the EWG has occurred. It has been assumed that being overweight that has developed in EBF infants is not a risk factor for overweight later in life. 14 However, some recent studies have indicated that EBF infants with high weight at the age of 5–6 months maintained overweight into early childhood.14,15 In contrast, other studies have reported catch-down growth in heavy EBF infants after the introduction of complementary foods, indicating a normalization of body weight in late infancy or early childhood.11,12,16
More information is needed for clarifying the risk of later obesity for infants experiencing EWG during EBF. We established a small exploratory cohort of 13 infants with EWG during EBF to examine factors associated with EWG during EBF and later growth (the SKOT-III cohort). 13 At the last examination at 18 months, the 12 children showed a continued decrease in weight-for-age z-score (WAZ) and body mass index (BMI)-for-age z-score (BAZ) from 9 months but it is not known if and how this decrease will continue. We therefore conducted a follow-up examination at 36 months to investigate growth and body composition within this rare group of children.
Subjects and Methods
Subjects
The study presents data from the 36-month follow-up of the SKOT-III cohort previously described in detail. 13 Briefly, inclusion criteria were healthy infants EBF for at least 4 months with a WAZ above +2 standard deviations score (SDS) at the age of 4–6 months and an increase in WAZ of at least +1 SDS within the first 5 months postpartum. A reference group of EBF infants with normal weight gain was also established but not followed up at 36 months.
The SKOT-III study complied with the guidelines for human studies and the Declaration of Helsinki. It was approved by the Committees on Biomedical Research Ethics for Capital Region of Denmark (H-15008948). Written informed consent was obtained from all parents or legal guardians of the children included in this study.
Anthropometry
The anthropometric measurements were performed at the Department of Nutrition, Exercise and Sports, University of Copenhagen, Denmark. Weight was measured to the nearest 0.1 kg using a Tanita WB-100MA digital weight (Tanita Corporation, Tokyo, Japan). Height was measured to the nearest 0.1 cm using a transportable stadiometer (Tanita HR001, Tanita Corporation, Tokyo, Japan). Height was calculated as the mean of three measurements, and BMI as weight/height. 2 z-Scores for weight, height, and BMI were calculated using the World Health Organization (WHO) growth standards as a reference and the WHO Anthro software. 17
A BODYSTAT®500 device (Bodystat Ltd., Douglas, Isle of Man, British Isles) was used to measure whole-body bioelectrical impedance as described previously. 13 The impedance value was used to estimate fat-free mass (FFM) via a predictive equation developed among 99 healthy 3-year-old Danish children. 18 Fat mass index (FMI) and FFM index (FFMI) were calculated by dividing fat mass (FM) and FFM with the squared height. z-Scores for FM, FFM, FMI, and FFMI were calculated using body composition reference data for UK infants and children. 19
Blood samples
Venous blood samples of 10 mL were taken after approximately 2 hours of fasting at the 36 months visit. Leptin and insulin were analyzed as described previously. 13
Statistics
Characteristics are given as means and standard deviations (SD) or medians and interquartile ranges (IQR) for continuous variables and as n (%) for categorical variables. Changes in anthropometry and body composition during the follow-up period were determined by paired t-test or Wilcoxon sign rank test and effect sizes by Cohen’s d. Significance was defined as p-values <0.05. Data analyses were performed using IBM SPSS statistics (version 28.0; IBM, New York, NY, USA).
Results
Of the 13 infants included in the EWG group in the SKOT-III cohort, 10 children participated in the 36-month examination, and data from these children were used in the present article (Table 1). Of these, body composition measurements were available for eight children, hereof five boys. At 5, 9, and 18 months, body composition data were available for eight, nine, and five children, respectively. Blood analyses were only available for five to six infants depending on the analyses.
Characteristic for Children Participating at the 36-Month Follow-Up Visit a
Data are shown as mean ± SD or n (%) as appropriate and for blood analysis median [interquartile range] (n).
z-Scores for anthropometry refers to WHO standards. 17
z-Scores for body composition refers to Wells et al. 19
BAZ, BMI-for age z-score; FFM, fat-free mass; FFMI, fat-free mass index; FM, fat mass; FMI, fat mass index; LAZ/HAZ, length/height-for-age z-score; WAZ, weight-for-age z-score.
WAZ decreased by 44% from 5 to 36 months (p < 0.001), while mean BAZ decreased by 36% (p = 0.024) (Table 1). However, both were still above +1, and one child was still overweight (BAZ = 2.582), as seen in Figure 1 showing BAZ and WAZ curves over time for each child.

Plot showing the change in
Mean FM decreased both in absolute value and percentage of body weight from 18 to 36 months (Table 1 and Fig. 2). Indexing for height did not change the FM z-score and FFM z-score curves so the FMI z-score and FFMI z-score curves are not shown. No children showed an increase in FM or FMI (Fig. 2). From 18 to 36 months, the FM and FMI z-scores declined, whereas the z-scores for FFM and FFMI increased. This is consistent with the overall fall in BAZ in this period.

Plots of
For the five children (two boys) with body composition measurements at both 18 and 36 months, FM decreased by 1.1 kg from 4.71 ± 0.56 to 3.59 ± 0.52 kg (p = 0.037), and FM% decreased by 15% (36.0 [1.52]% to 21.0 [3.4]%, p = 0.020). FFM increased by 5.08 kg from 8.36 ± 0.52 to 13.44 ± 1.03 kg (p ≤ 0.001). When looking at height-indexed body composition, the changes for FMI (6.50 ± 0.44 to 3.72 ± 0.65 kg/m2) and FFMI (11.6 ± 0.3 to 13.9 ± 0.6 kg/m2) remained significant with large differences for both (effect size [95% CI] FMI:−3.28 [−5.08; −0.84]; FFMI: 3.31 [0.94; 5.66]; both p = 0.02). For these five children, the mean increase in weight was 3.91 kg, suggesting that the weight gained within the 18 months period was lean mass.
Leptin and insulin continued to decrease, showing a mean decline of 45% and 20%, respectively, from 9 to 36 months (Table 1).
Discussion
In this follow-up study of children with EWG during the first 5 months of EBF, we found that the catch-down which began after the start of complementary feeding continued through to 36 months. Both WAZ and BAZ decreased further but remained above +1 SDS. Of the 10 children attending the 36-month follow-up examination, only one child was overweight. This indicates an ongoing normalization of body weight.
Body composition also seemed to continue to normalize with increasing FFM and decreasing FM from 18 to 36 months. Thus, the weight gain in this period was only lean mass. This was also reflected in the z-scores, as FFM z-score increased to around +1, whereas FM z-score decreased to around 0. We used reference data for UK children in order to calculate the z-scores for body composition and refer to a comparable population. 19 This reference is contemporary and based on British children of white European ethnicity. We found the same pattern irrespective of controlling for height. Overall, the sample demonstrated remarkable conformity regarding the changes in both adiposity and relative FFM.
Being a composite measure including FFM and FM, BMI is an inadequate index of body composition, especially during periods of growth. 20 Before the age of 6 years, the correlation between FMI and BMI is very weak but after 6 years it becomes stronger and positive. According to Wright et al., children below 6 years with raised BMI have relatively low FMI but seem to have more lean mass, and most do not have excess adiposity. However, the children in our cohort initially showed a high FM with a high FMI and an FMI z-score above 1.7 for the first 18 months. Though the FFM z-score was very high at 5 months (+2.1) it was mostly due to their greater length as the FFMI z-score was around +1 and declined to −1 at 18 months, that is, for these children the high BAZ seemed to reflect a high FM up to 18 months. From 18 months to 36 months, the pattern reversed so the growth reflected loss of FM z-score and gain of FFM z-score. This was accompanied by a fall in BAZ.
The fall in FM is in accordance with the decrease observed in leptin. This hormone is mainly produced by adipocytes, suppressing appetite and stimulating energy expenditure. It is known to correlate strongly with body weight and FM both in adults and children.21–25 Leptin concentrations at 36 months were in the same range as reference values reported for normal weight 3.0–3.9 years old European children (median [IQR] girls: 2.0 [1.5, 3.8] ng/mL; boys: 1.5 [1.1, 2.5] ng/mL). 26 Previously, these children with EWG showed considerably higher leptin values than the group of children with normal weight gain during EBF. 13 However, leptin values appear to have normalized toward values comparable to those of normal weight children of the same age.
To our knowledge, very few studies have assessed the growth of children with EWG during EBF beyond the first year of life. We have previously reported a case followed to the age of 42 months, and in line with the present results, BAZ declined from >+3 SDS at 6 months to <+2 SDS at 42 months also showing a normal body weight at 36 months. 12 Other studies on cases or small cohorts of children with EWG do not report data beyond the age of 18 months so it is not possible to compare our findings with these studies.11,16,27,28 Thus, our data provide unique information on the longer-term effects of EWG during EBF.
Another approach for investigating the effect of EWG during EBF on later growth is analyzing data from prospective birth cohorts. A Dutch study compared children according to being EBF for ≥3 months, formula-fed from the age of 2 weeks, or mixed-fed, and if they were overweight or normal weight at 6 months in relation to their weight status at 5–6 years. 14 They reported that being overweight in infancy was associated with an increased risk for childhood overweight irrespective of infant feeding mode. Similarly, Morgen et al. examined the risk of being overweight in 7- and 11-year-old Danish children with a high WAZ at 5 months according to the duration of EBF. 15 They found that being overweight in infancy was associated with a higher risk of being overweight in childhood and with no influence on the duration of EBF. This is contrary to our own results, that is, that infants with EWG during EBF normalize their weight in childhood.
However, the mechanism behind EWG in EBF infants might be different from what is seen in larger cohorts of infants. Furthermore, the studies are difficult to compare to our findings due to differences in methodology, sample size, and follow-up age. In addition, body composition data were not available in these birth cohort studies so it was not possible to ascertain if the higher BMI was primarily due to high FFMI rather than fatness. Also, the WAZ for the infants at 5–6 months differed. The average WAZ was around +3 SDS in our study which was considerably higher than the ≥+1 SDS and ≥+2.5 SDS WAZ thresholds used to categorize overweight by van der Willik et al. and Morgen et al., respectively.
Further research is needed to clarify why some infants attain excessive weight during EBF and studies are sparse. EWG is likely a multifactorial process, and different mechanisms have been suggested. In this cohort, the leptin concentrations in breast milk were low which indicates that appetite regulation may be involved. At the first examination at 5 months, 24-hour milk intake was over 15% higher in the infant with EWG compared with the reference group with normal weight gain, but the difference was not significant (p = 0.19) which may be due to the low sample size and the high variability in the milk intake. 13 Another study found lower fat content in breast milk of mothers to infants with EWG during EBF compared with the general population. The milk volume consumed may therefore be higher for these children according to the hypothesis that infants adjust milk intake in proportion to the energy content in the breast milk, 27 and the total intake of protein would be higher. This was found in another case report where normal macronutrient and energy concentrations were reported but a higher milk intake resulted in a higher protein dose. 16 According to the “early protein hypothesis” high amounts of protein in early life would stimulate growth, especially fat tissue, which could lead to later increased risk of obesity. 29 No difference in macronutrients or energy content was found in breast milk between mothers in this cohort and the reference group at 5 and 9 months. 13 A high milk protein content was described in a case report of an infant with EWG during EBF but milk volume was not reported. 11 In another small cohort study of infants with EWG during EBF no alteration in macronutrients in human milk was found but gestational weight gain, age of the mother, and co-sleeping of the mother and the child were associated with EWG during EBF. 28 Furthermore, a high milk intake has been associated with fat accretion. 30 Thus, the contributing factors are not clear or definitive, and the phenomenon is challenging to study as it concerns only a few infants. The possibility that early fetal growth faltering might contribute to a drive for rapid postnatal growth also merits attention.
The primary limitation of the study is the sample size. This was already a limitation when the cohort was established and reduced the statistical power of analyses and the generalizability of the findings. The number of children was further reduced in the follow-up study with 10 out of 13 children participating and body composition available for only eight children. Blood samples were not available for all children, making it difficult to compare longitudinally and reducing the power further. Furthermore, the reference group of children with normal weight gain during EBF was not followed up so it was not possible to compare the growth with this group as previously. Aside from these limitations, we succeeded in following up with the majority of this special and rare cohort, and the patterns of tissue accretion were very similar across the sample.
Conclusions
This study is as far as we know the only one that has examined the consequences of EWG in infancy during EBF by reporting follow-up data on growth and body composition. The results support a continued normalization of body weight and composition in childhood, but further follow-up is needed to assess the risk of later overweight and inform relevant health policy in cases where normal weight is not reached.
Footnotes
Acknowledgments
The authors thank the participating children and families of the SKOT-III cohort and Birgitte Hermansen, Vivian Anker, and Inge Rasmussen for technical assistance in the data collection.
Authors’ Contributions
Conceptualization and methodology: K.F.M., C.M., M.W.L., and A.L. Formal analysis: A.L. Investigation and project administration: A.L. and M.W.L. Writing—original draft preparation: A.L. Writing—review and editing: K.F.M., C.M., J.W., J.I.L., and S.C.H. Funding acquisition: K.F.M., C.M., M.W.L., and A.L. All authors have read and agreed to the published version of the article.
Data Availability Statement
The Danish Act on Data Protection does not allow for personal data to be made available to other researchers without prior individual approval from the Danish Data Protection Agency.
Disclosure Statement
The authors declare no conflicts of interest.
Funding Information
The study was supported by grants from the University College Copenhagen, the Family Larsson-Rosenquist Foundation, and the Governing Obesity program funded by the University of Copenhagen.
