The CDC/ACSM (1995) recommendations for moderate physical activity (PA) (3–5 METs, ≥30 min, ≥5 d·wk−1) in healthy adults are currently advocated for pregnant women by the American College of Obstetricians and Gynecologists (ACOG, 2002). However, further research on birth outcome may be needed before ACOG endorses theCDC/ACSM recommendations for vigorous PA (≥ 6 METs, ≥20 min, ≥3 d·wk−1) during pregnancy. PURPOSE Using meta-analysis, we examined whether the relationship between maternal PA and birth weight was moderated by PA intensity. Additionally we examined the impact of maternal weight gain on birth weight. METHODS The literature search was performed using NLM Medline, Embase, and manual searching of bibliographies. Studies were screened to ensure they included a measure of physical activity during pregnancy (with a sedentary control group) and reported birth weight as a continuous variable within the normal birth weight range. A random effects model was used to calculate the pooled effect size (ES) of birth weight by maternal PA. Pre-specified subgroups (e.g., maternal age, maternal weight gain, study design) were used to examine potential heterogeneity. Meta-regression was utilized to estimate the amount of variance in birth weight between physically active and sedentary pregnant women that can be accounted for by differences in maternal weight gain. RESULTS Of 2418 potentially relevant studies located, 23 satisfied inclusion criteria after full-text review. Study participants were mostly Caucasian, nonsmokers, and nulliparous. Significant heterogeneity was found between PA and birth weight across all studies (P<0.01). In subgroup analyses, moderate PA was not significantly related to birth weight (ES=0.05; P=0.49). There was evidence of a negative relationship between vigorous PA and birth weight (ES=0.27; P<0.01) (∼150 g difference); however, significant heterogeneity precludes further interpretation, i.e., the studies included in this subgroup were too disparate to compare. Differences in maternal weight gain significantly and positively accounted for ∼17% of the variance in birth weight between moderate and vigorous exercisers (P<0.01). CONCLUSIONS Moderate physical activity performed throughout pregnancy did not affect birth weight. The impact of vigorous PA on birth weight remains unclear due to significant heterogeneity of study results.
Read full abstract