Intermittent Fasting During Pregnancy and Neonatal Birth Weight: A Systematic Review and Meta-Analysis

Abstract

Background/Objectives: Intermittent fasting (IF), such as Ramadan fasting, is common among pregnant women despite religious exemptions. The possible impact of fasting on pregnancy outcome and, in particular, on birthweight is uncertain and was documented with conflicting evidence. Methods: The aim of this meta-analysis and systematic review was to investigate the association between intermittent fasting during pregnancy and neonatal birthweight, along with low birthweight (LBW, <2500 g) risk. Studies evaluating fasting during pregnancy with reported neonatal outcomes were included. We searched (PubMed, Scopus, Web of Science) from 2004 through June 2025. All contributing studies were observational in design; no randomized trials were identified. The risk of bias was assessed using the Newcastle-Ottawa Scale. The pooled relative risks (RR) and mean differences (MD) were calculated according to random-effects models (DerSimonian-Laird method) and heterogeneity was quantified with the I² statistic. Results: Nineteen studies, all observational in design, were included for qualitative synthesis, and six studies yielded quantitative data to conduct meta-analyses. No randomized controlled trials were identified. Intermittent fasting during pregnancy, encompassing more than 1.3 million pregnancies, was associated with a small but statistically significant reduction in neonatal birth weight. The pooled mean difference was -94 g (95% CI: -176 to -12 g; p = 0.03; I² = 70%), indicating a minor but statistically significant effect that is unlikely to be clinically meaningful. The pooled RR for LBW was 0.96 (95% CI: 0.88-1.05; p = 0.38; I² < 10%), showing no association between fasting and low birthweight risk. Sensitivity analyses reduced heterogeneity (I² ≈ 55%) and confirmed the robustness of these findings. According to the GRADE approach, the certainty of evidence was moderate for birthweight and high for LBW. Conclusions: Intermittent fasting during pregnancy, including Ramadan fasting, was associated with a minor but statistically significant reduction in neonatal birthweight without increasing the risk of low birthweight. This difference was small and clinically negligible. Further prospective studies are needed to clarify trimester-specific effects and long-term developmental outcomes.

Keywords: Ramadan; birthweight; fetal growth; intermittent fasting; low birthweight; maternal metabolism; pregnancy.

Figure 1

PRISMA 2020 flow diagram illustrating the study selection process. The database search (PubMed, Scopus, Web of Science; 2004–June 2025) identified 49 records after deduplication. After title and abstract screening, 10 studies were excluded for irrelevance, leaving 39 full-text articles assessed for eligibility. Of these, 20 were excluded for specific reasons: non-human or animal studies (n = 6), narrative or umbrella reviews (n = 5), absence of relevant neonatal outcomes (n = 5), duplicate or overlapping datasets (n = 2), and incomplete or unavailable data (n = 2). Nineteen studies were included in the qualitative synthesis, and six provided quantitative data for meta-analysis [23].

Figure 2

Forest plot showing the pooled mean difference (MD) in neonatal birth weight between fasting and non-fasting pregnancies. Black circles represent the mean difference in birth weight for each study, with horizontal lines indicating 95% confidence intervals. The vertical dashed line corresponds to the line of no effect (MD = 0). The red square and red horizontal line represent the pooled random-effects estimate and its 95% confidence interval (−94 g; 95% CI −176 to −12 g). Heterogeneity was substantial (I² ≈ 70%), indicating variability mainly due to study design and population differences [10,13,14,20,24].

Figure 3

Forest plot showing the pooled relative risk for low birth weight (LBW < 2500 g) comparing fasting and non-fasting pregnancies. Blue squares represent the RR estimates for individual studies, with horizontal black lines indicating 95% confidence intervals. The vertical dashed line corresponds to the line of no effect (RR = 1). The red diamond represents the pooled random-effects estimate and its 95% confidence interval (RR = 0.96; 95% CI 0.88–1.05; p = 0.38). Between-study heterogeneity was low (I² < 10%), indicating consistent findings across studies [10,13,14,20,24].