Table 4.

Main features of RCTs and observational studies assessing the relationship between CF introduction in infants and later onset of obesity

Author, Year	Study Design	Sample size	Results
Singhal, 2003 [64]	RCT	Measurement of fasting 32–33 split proinsulin concentration in adolescent participants born preterm and randomised to receive a nutrient-enriched or lower-nutrient diet (n = 216) vs. a control group born at term (n = 61)	Relative undernutrition early in life in premature infants may have beneficial effects on insulin resistance.
Sun, 2016 [41]	Observational study	Cross-sectional, population-based study on 3153 Australian infants	Introduction of CF at 5–6 months, compared with either early or delayed introduction, is associated with decreased odds of above normal BMI.
Baldassarre, 2020 [42]	Observational study	Prospective, population-based longitudinal study on 100 preterm infants	Half of preterm neonates experienced early adiposity rebound and featured significantly higher BMI at seven years compared to children with timely adiposity rebound (17.2 ± 2.7 vs. 15.6 ± 2.05, p = 0.021).
Gupta, 2017 [62]	RCT	RCT comparing CF starting at 4 vs. 6 months CA in 373 Indian preterm infants born < 34 weeks’ GA (n = 184 CF at 4 months CA vs. n = 189 CF at 6 months CA)	No difference was found in weight-for-age z score at 12 months CA between groups, but a higher hospitalization rate was documented in the 4 month CF group.
Morgan, 2004 [21]	Pooled RCTs results	Pooled results from 5 RCTs assessing early (< 12 weeks) vs. late (> 12 weeks) introduction of CF in 1694 term and preterm infants	Preterm infants weaned before 12 weeks featured slower increase in weight, length, and head circumference at 12 weeks − 18 months; by 18 months, there were no significant differences in size between the two groups.
Baldassarre, 2017 [43]	Observational study	Survey exploring the influence of neonatal features on the onset of non-communicable diseases: n = 6379 questionnaires were assessed	Preterm birth was not associated with the onset of asthma and allergy, celiac disease or diabetes, and acted as a protective factor in the development of obesity.
Kaul, 2019 [44]	Observational study	Results from n = 81,226 children with the aim to assess the association between gestational diabetes, being LGA at birth and breast feeding with overweight/obesity in early childhood	LGA is a strong risk factor for being overweight/obese in early childhood, especially in babies born to diabetic mothers. Breast feeding was a protective factor against overweight/obesity in childhood in the majority of children, except for LGA children of diabetic mothers.
Kapral, 2018 [45]	Observational study	Analysis of n = 10,186 term or preterm children in the Early Childhood Longitudinal Study-Kindergarten Cohort 2011 to assess relationships between BW and later obesity in childhood	High BW term and LGA preterm children had increased adjusted odds of obesity in childhood.
Brion, 2020 [47]	Observational study	Assessment of BMI, weight-for-length and head growth in 208 AGA infants born at 23–28 + 6 weeks’ GA	Infants started on ready-made CF ≤26 weeks CA had the highest BMI and weight-for-length at 12 months. Head growth from discharge to 12 months was the highest in infants either discharged on breastmilk or receiving home-made CF at ≤26 weeks' CA.
Fenton, 2021 [48]	Observational study	To examine the prevalence and risk factors for childhood overweight and obesity at 3-year CA in 911 preterm babies (BW < 1500 g or GA < 29 weeks)	Small size at birth or at 36 weeks’ GA in ELGAN is not associated with increased risk of early childhood overweight or obesity.

AGA appropriate for gestational age, BMI body mass index, BW birth weight, CA corrected age, CF complementary feeding, ELGAN extremely low gestational age neonates, GA gestational age, LGA large for gestational age, RCT randomized controlled trial