Maternal Antioxidant Levels in Pregnancy and Risk of Preeclampsia and Small for Gestational Age Birth: A Systematic Review and Meta-Analysis

No authors available

PLoS ONE

Aug 2015

DOI: 10.1371/journal.pone.0135192

Citations:51

Influential Citations:2

Systematic Reviews / Meta-Analyses

Enhanced Details

Methods

Pregnant women from general populations and high-risk groups. Study designs included: prospective cohort, nested case-control, case-control, and cross-sectional; randomized trials included and analyzed as cohorts. Publications in English or French from 1970 to January 2013.

Results

Third-trimester levels of vitamins A, C, E and carotenoids tended to be lower in pregnancies with preeclampsia, but heterogeneity was high. Lipid-corrected vitamin E results were inconsistent. For SGA, evidence of lower carotenoids in the second trimester exists but data are limited. Observational data suggest lower antioxidant levels may reflect disease processes rather than causation; randomized trials of antioxidant supplementation largely show null effects on prevention, and publication bias and confounding limit conclusions. More robust observational studies with serial biomarker measurements before disease onset are needed; current evidence does not support routine antioxidant supplementation to prevent preeclampsia or SGA.

Limitations

Most studies were small and at high risk of bias; confounding not consistently addressed; substantial between-study heterogeneity; antioxidant measurements often late in pregnancy; variable assay methods and reporting; publication bias likely; the evidence base is observational and cannot establish causality.

Abstract

Background Oxidative stress in preeclampsia and small for gestational age (SGA) birth suggests antioxidant supplementation could prevent these conditions. However, it remains unclear whether maternal antioxidant levels are systematically lower in these pregnancies. Objective To conduct a systematic review of the association between maternal antioxidant levels during pregnancy and preeclampsia or SGA. Methods We searched PubMed, Embase, and several other databases from 1970–2013 for observational studies that measured maternal blood levels of non-enzymatic antioxidants (vitamins A, C, E, and carotenoids) during pregnancy or within 72 hours of delivery. The entire review process was done in duplicate. Study quality was assessed using the Newcastle-Ottawa Scale and additional questions. We pooled the standardized mean difference (SMD) across studies, stratified by outcome and pregnancy trimester, and investigated heterogeneity using meta-regression. Results We reviewed 1,882 unique citations and 64 studies were included. Most studies were small with important risk of bias. Among studies that addressed preeclampsia (n = 58) and SGA (n = 9), 16% and 66%, respectively, measured levels prior to diagnosis. The SMDs for vitamins A, C, and E were significantly negative for overall preeclampsia, but not for mild or severe preeclampsia subtypes. Significant heterogeneity was observed in all meta-analyses and most could not be explained. Evidence for lower carotenoid antioxidants in preeclampsia and SGA was limited and inconclusive. Publication bias appears likely. Conclusions Small, low-quality studies limit conclusions that can be drawn from the available literature. Observational studies inconsistently show that vitamins C and E or other antioxidants are lower in women who develop preeclampsia or SGA. Reverse causality remains a possible explanation for associations observed. New clinical trials are not warranted in light of this evidence; however, additional rigorous observational studies measuring antioxidant levels before clinical detection of preeclampsia and SGA may clarify whether levels are altered at a causally-relevant time of pregnancy. show less