Different protein intake in the first year and its effects on adiposity rebound and obesity throughout childhood: 11 years follow-up of a randomized controlled trial

BACKGROUND AND OBJECTIVES

Infant feeding affects child growth and later obesity risk. We examined whether protein supply in infancy affects the adiposity rebound, body mass index (BMI) and overweight and obesity up to 11 years of age.

METHODS

We enrolled healthy term infants from five European countries in a double blind randomized trial, with anticipated 16 examinations within 11 years follow-up. Formula-fed infants (n = 1090) were randomized to isoenergetic formula with higher or lower protein content within the range stipulated by EU legislation in 2001. A breastfed reference group (n = 588) was included. Adiposity rebound and BMI trajectories were estimated by generalized additive mixed models in 917 children, with 712 participating in the 11 year follow-up.

RESULTS

BMI trajectories were elevated in the higher compared to the lower protein group, with significantly different BMI at adiposity rebound (0.24 kg/m², 0.01-0.47, p = 0.040), and an increased risk for overweight at 11 years (adjusted Odds Ratio 1.70; 1.06-2.73; p = 0.027) but no significant difference for obesity (adjusted Odds Ratio 1.47; 0.66-3.27). The two formula groups did not differ in the timing of adiposity rebound, but all children with obesity at 11 years had an early adiposity rebound before four years.

CONCLUSIONS

Compared to conventional high protein formula, feeding lower protein formula in infancy lowers BMI trajectories up to 11 years and achieves similar BMI values at adiposity rebound as observed in breastfed infants.

Usefulness of the waist-to-height ratio for predicting cardiometabolic risk in children and its suggested boundary values

BACKGROUND & AIMS

Only limited information is available on the usefulness of the waist-to-height ratio (WHtR) as an abdominal obesity marker in children. Our aim was to compare the ability of a WHtR >90th percentile, a WHtR ≥0.50, a WHtR ≥0.55 and a BMI z-score ≥2 SD to predict cardiometabolic risk in children followed-up at different ages.

METHODS

We evaluated data from 660 children at 5, 8 and 11 years of age who participated in the Childhood Obesity Project trial in 5 European countries. We classified children with or without cardiometabolic (CMet) risk (yes vs. no) according to the presence of ≥2 parameters (blood pressure, HOMA-IR, triglyceride levels and high-density lipoprotein (HDL) cholesterol levels) ≥90th percentile.

RESULTS

The odds ratio for CMet risk in children at all followed-up ages was statistically significant for all measures. The OR for the WHtR≥0.55 cut-off was 29.1 (5.6, 151.7) at 5 years of age, 11.8 (4.1, 33.8) at 8 year of age and 3.6 (1.7, 7.7) at 11 years of age, compared to the WHtR<0.55 cut-off. The WHtR≥0.55 cut-off showed a higher OR at younger ages than the BMI z-score ≥2SD, WHtR ≥90th percentile and WHtR≥0.50 cut-offs and a higher positive predictive value (82% at 5 years of age compared to 55%, 36% and 41%, respectively).

CONCLUSION

A WHtR≥0.55 is a suitable cut-off for screening children at high cardiometabolic risk in the general young European population.

Influence of total sugar intake on metabolic blood markers at 8 years of age in the Childhood Obesity Project

Purpose

We aimed to characterize the association of dietary sugar intake with blood lipids and glucose-related markers in childhood.

Methods

Data from the multicentric European Childhood Obesity Project Trial were used. Three-day weighed dietary records were obtained at 8 years of age along with serum concentrations of triglycerides, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol (HDL-C), glucose, and insulin. Total sugar intake comprised all mono- and disaccharides; different sugar sources were defined. Linear regression models were applied to investigate the cross-sectional association of total sugar intake with blood lipids and glucose-related markers with adjustment for total energy intake using the residual method.

Results

Data were available for 325 children. Children consumed on average 332 kcal (SD 110) and 21% (SD 6) of energy from total sugar. In an energy-adjusted model, an increase of 100 kcal from total sugar per day was significantly associated with a z score HDL-C decrease (− 0.14; 95% CI − 0.01, − 0.27; p value = 0.031). Concerning different food groups of total sugar intake, 100 kcal total sugar from sweetened beverages was negatively associated with z score HDL-C (− 1.67; 95% CI − 0.42, − 2.91; p value = 0.009), while total sugar from milk products was positively related to z score HDL-C (1.38, 95% CI 0.03, 2.72; p value = 0.045). None of the other blood lipids or glucose-related markers showed a significant relationship with total sugar intake.

Conclusion

Increasing dietary total sugar intake in children, especially from sweetened beverages, was associated with unfavorable effects on HDL-C, which might increase the long-term risk for dyslipidemia and cardiovascular disease.

Clinical trial registry

ClinicalTrials.gov Identifier: NCT00338689; Registered: June 19, 2006. URL: https://clinicaltrials.gov/ct2/show/NCT00338689?term=NCT00338689&rank=1.

Electronic supplementary material

The online version of this article (10.1007/s00394-020-02229-w) contains supplementary material, which is available to authorized users.

Effects of screen time and playing outside on anthropometric measures in preschool aged children

Objective

In view of the current obesity epidemic, studies focusing on the interplay of playing outside (PO), screen time (ST) and anthropometric measures in preschool age are necessary to guide evidence-based public health planning. We therefore investigated the relationship between average time spent PO and ST from the ages 3 to 6 years and anthropometric measures at 6 years of age.

Methods

PO and ST of 526 children of the European Childhood Obesity Project (CHOP) were annually assessed by questionnaire from 3 until 6 years of age. Body weight, waist circumference and height were measured at 3 and 6 years of age to calculate Body-Mass-Index z-Scores (zBMI) and waist-to-height ratio (WTH). Linear, logistic and quantile regressions were used to test whether average time spent PO and ST in the 4 year period had an effect on anthropometric measures at age 6 years.

Results

Longer daily ST was associated with a higher zBMI (P = 0.002) and WTH (P = 0.001) at 6 years of age. No significant associations were found for time spent PO. Each additional hour of average ST during the 4 year period resulted in a 66% higher risk of having a zBMI score over 1 (P < 0.001) and almost twice the risk (94% higher risk) of having an zBMI score over 2 (P < 0.001) at 6 years.

Conclusions

Excessive ST during preschool age is a risk factor for increased zBMI at 6 years, regardless of time spent PO. Reducing high levels of ST during preschool age, for e.g. at least 1h per week, could help preventing childhood obesity.

Effect of Lower Versus Higher Protein Content in Infant Formula Through the First Year on Body Composition from 1 to 6 Years: Follow-Up of a Randomized Clinical Trial

OBJECTIVE

The objective of this study was to investigate the effect of lower protein (LP) versus higher protein (HP) content in infant formula on body composition from 3 months to 6 years.

METHODS

In a multicenter, double-blind European trial, healthy infants (N = 1,090) were randomly assigned to different protein content formulas (upper [HP] and lower [LP] limits of the European Union regulations in 2001) during the first year; breastfed infants (N = 588) were recruited for reference values. Weight, height, and triceps and subscapular skinfold (SF) thickness were measured repeatedly (N = 650 at 6 years), and body composition was estimated (Slaughter). The 99th percentile of fat mass index reference data were used to assess excess body fat at 6 years.

RESULTS

At 2 and 6 years, the study observed greater sum of SFs (Δ 2 years: 0.5 mm, P = 0.026, Δ 6 years: 0.6 mm, P = 0.045), fat mass index (Δ 2 years: 0.12 kg/m², P = 0.008, Δ 6 years: 0.15 kg/m², P = 0.011), and fat-free mass index (Δ 2 years: 0.17 kg/m², P = 0.003, Δ 6 years: 0.18 kg/m², P = 0.010) in the HP group compared with the LP group. At 6 years, the HP group had a twofold higher risk than the LP group for excess body fat (adjusted odds ratio: 2.13, P = 0.019).

CONCLUSIONS

Infant formula with HP levels induced greater fat mass in children from 2 to 6 years. Lowering the protein content of infant formula may result in a healthier body composition in early childhood.

Unhealthy Dietary Patterns Established in Infancy Track to Mid-Childhood: The EU Childhood Obesity Project

Background

Dietary habits established in infancy may persist into adulthood and determine long-term health.

Objectives

The aims of this work were to describe dietary patterns, predictors of adherence to them, and their tracking from ages 1 to 8 y in European children.

Methods

Three-day food diaries were prospectively collected at ages 1, 2, 3, 4, 5, 6 and 8 y. Foods were allocated to 1 of 29 food groups, which were included in exploratory factor analyses at each children's age. The tracking of patterns through childhood was assessed by an estimated general equation model.

Results

At age 1 y (n = 633), 2 patterns were identified. One was labeled "core foods" (CORE), since it was positively loaded for vegetables, fish, olive oil, and white and red meat, and negatively loaded for ready-to-eat infant products, sugar, and confectioneries. The other was positively loaded for saturated spreads, sugar, fruit juices, and confectioneries, and negatively loaded for olive oil, fish, and cow milk; this was labeled as the "poor-quality fats and added sugars" (F&S) pattern. From ages 2 to 8 y, 3 patterns were repeatedly identified: CORE, F&S, and a "high protein sources" (PROT) pattern that was positively loaded for milk, flavored milks, fish, eggs, white and processed meat, chips, and olive oil, and negatively loaded for fresh fruits at almost all time points. Of those children in the highest quartiles of the CORE, F&S, and PROT patterns at 2 y, 45%, 72%, and 36%, respectively, remained in the highest quartile at 8 y [OR = 2.01 (1.08, 3.8), OR = 3.6 (1.5, 8.4) and OR = 0.80 (0.4,1.6), respectively; P = 0.510].

Conclusions

Dietary patterns are established between 1 and 2 y of age and track into mid-childhood. A dietary pattern characterized by added sugars, unhealthy fats, and poor consumption of fish and olive oil was the most stable throughout childhood. Further analyses will reveal whether those dietary patterns are associated with metabolic disease risk.