Birth weight and long-term overweight risk: systematic review and a meta-analysis including 643,902 persons from 66 studies and 26 countries globally

Abstract

Background: Overweight is among the major challenging health risk factors. It has been claimed that birth weight, being a critical indicator of prenatal developmental conditions, is related to long-term overweight risk. In order to check this important assumption of developmental and preventive medicine, we performed a systematic review and comprehensive meta-analysis.

Methods and findings: Relevant studies published up to January 2011 that investigated the relation between birth weight and later risk of overweight were identified through literature searches using MEDLINE and EMBASE. For meta-analysis, 66 studies from 26 countries and five continents were identified to be eligible, including 643,902 persons aged 1 to 75 years. We constructed random-effects and fixed-effects models, performed subgroup-analyses, influence-analyses, assessed heterogeneity and publication bias, performed meta-regression analysis as well as analysis of confounder adjusted data. Meta-regression revealed a linear positive relationship between birth weight and later overweight risk (p<0.001). Low birth weight (<2,500 g) was found to be followed by a decreased risk of overweight (odds ratio (OR) =0.67; 95% confidence interval (CI) 0.59-0.76). High birth weight (>4,000 g) was associated with increased risk of overweight (OR=1.66; 95% CI 1.55-1.77). Results did not change significantly by using normal birth weight (2,500-4,000 g) as reference category (OR=0.73, 95% CI 0.63-0.84, and OR=1.60, 95% CI 1.45-1.77, respectively). Subgroup- and influence-analyses revealed no indication for bias/confounding. Adjusted estimates indicate a doubling of long-term overweight risk in high as compared to normal birth weight subjects (OR=1.96, 95% CI 1.43-2.67).

Conclusions: Findings demonstrate that low birth weight is followed by a decreased long-term risk of overweight, while high birth weight predisposes for later overweight. Preventing in-utero overnutrition, e.g., by avoiding maternal overnutrition, overweight and/or diabetes during pregnancy, might therefore be a promising strategy of genuine overweight prevention, globally.

Figure 1. Flow diagram of selection process.

Course of systematic literature review on birth weight and risk of overweight later in life, 1966–January 2011.

Figure 2. Low birth weight (<2,500 g) and subsequent risk of overweight.

ORs for overweight in subjects with birth weights <2,500 g as compared with subjects with birth weights ≥2,500 g. Studies are ordered alphabetically by first author. The point estimate (center of each black square) and the statistical size (proportional area of square) are represented. Horizontal lines indicate 95% confidence intervals. The pooled odds ratio (diamond) was calculated by means of a random effects model. OR, odds ratio; CI, confidence interval.

Figure 3. High birth weight (>4,000 g) and subsequent risk of overweight.

ORs for overweight in subjects with birth weights >4,000 g as compared with subjects with birth weights ≤4,000 g. Studies are ordered alphabetically by first author. The point estimate (center of each black square) and the statistical size (proportional area of square) are represented. Horizontal lines indicate 95% confidence intervals. The pooled odds ratio (diamond) was calculated by means of a random effects model. OR, odds ratio; CI, confidence interval.

Figure 4. Relationship between birth weight and risk of overweight.

Continuous relation between birth weight and later risk of overweight, calculated by fractional polynomial regression. Studies are represented by black dots. Grey shading indicates the 95% confidence interval around the fitted line. The model was estimated from a robust regression model based on second-order fractional polynomial (−1, −0.5) functions weighted by/variance.