The Navigation Guide - evidence-based medicine meets environmental health: systematic review of human evidence for PFOA effects on fetal growth

Abstract

Background: The Navigation Guide methodology was developed to meet the need for a robust method of systematic and transparent research synthesis in environmental health science. We conducted a case study systematic review to support proof of concept of the method.

Objective: We applied the Navigation Guide systematic review methodology to determine whether developmental exposure to perfluorooctanoic acid (PFOA) affects fetal growth in humans.

Methods: We applied the first 3 steps of the Navigation Guide methodology to human epidemiological data: 1) specify the study question, 2) select the evidence, and 3) rate the quality and strength of the evidence. We developed a protocol, conducted a comprehensive search of the literature, and identified relevant studies using prespecified criteria. We evaluated each study for risk of bias and conducted meta-analyses on a subset of studies. We rated quality and strength of the entire body of human evidence.

Results: We identified 18 human studies that met our inclusion criteria, and 9 of these were combined through meta-analysis. Through meta-analysis, we estimated that a 1-ng/mL increase in serum or plasma PFOA was associated with a -18.9 g (95% CI: -29.8, -7.9) difference in birth weight. We concluded that the risk of bias across studies was low, and we assigned a "moderate" quality rating to the overall body of human evidence.

Conclusion: On the basis of this first application of the Navigation Guide systematic review methodology, we concluded that there is "sufficient" human evidence that developmental exposure to PFOA reduces fetal growth.

Figure 1

Overview of Navigation Guide systematic review methodology used for rating the quality and strength of the human evidence.

Figure 2

Flowchart showing the literature search and screening process. The primary goal of our search was to obtain comprehensive results; therefore, our search was not limited by language or publication date. The search terms for each database are provided in Supplemental Material, Table S1.

Figure 3

Summary of the risk of bias judgments (low, probably low, probably high, and high risk) for each included human study (A) and (B) given as percentages across all included human studies. Risk of bias designations for individual studies are assigned according to criteria provided in Supplemental Material, “Instructions for Making Risk of Bias Determinations.”

Figure 4

Summary of data extracted from all studies of PFOA exposure that included continuous outcome of birth weight. The PFOA increase is the exposure contrast being compared in each study. Squares represent data for which there was an exposure gradient that can be evaluated in considering dose response in upgrading the quality of the evidence. Error bars indicate 95% CIs. Savitz et al. (2012b) presented additional alternative estimates based on different modeling assumptions that are not included here due to space limitations. Covariate abbreviations: bmi, body mass index; bsp, blood sampling period; cot, serum cotinine; delmode, delivery mode; dia, diabetes; edu, maternal education level; exposyr, year of exposure estimate; ga, gestational age; gabd, gestational age at blood draw; geomean(stdev), geometric mean (geometric SD); grav, gravidity; ma, maternal age; ht, maternal height; hyp, hypertension; mwt, maternal prepregnancy weight; NA, not applicable: ND, not detected; par, parity; PFOS, serum perfluoro­octane sulfonic acid; SES, socioeconomic status; sex, infant sex; smk, smoking status; state, state of residence; wtg, maternal weight gain during pregnancy. This figure was created using Meta Data Viewer (http://ntp.niehs.nih.gov/help/browsers/metadata/index.html) (Boyles et al. 2011).

Figure 5

Results of meta-analysis for birth weight (n = 9 studies, 4,149 births) shown as effect estimates [change in birth weight in grams per nanogram of PFOA per milliliter of serum or plasma (95% CIs)]. The percentages are weightings of the individual studies in the meta-analysis according to the inverse of the variance, and the sizes of the boxes are scaled accordingly. The dashed line indicates the overall effect estimate derived from the DerSimonian-Laird random effects meta-analysis, and the diamond indicates the 95% CI of the overall effect estimate. Heterogeneity statistics: Cochran’s Q = 12.92; p = 0.12; I² = 38%. Estimates were adjusted as follows: Apelberg et al. (2007): maternal age and gestational age; Fei et al. (2007): maternal age, gestational age, quadratic gestational age, infant sex, socio-occupational status, parity, smoking, pre­pregnancy body mass index, and gestational week at blood draw; Hamm et al. (2010): maternal age, gestational age, race, gravidity, maternal pre­pregnancy weight, maternal height, smoking status, and infant sex; Washino et al. (2009): maternal age and gestational age; Fromme et al. (2010): unadjusted; Kim S et al. (2011): maternal age, gestational age, and parity; Whitworth et al. (2012): maternal age, gestational age, prepregnancy body mass index, and parity; Maisonet et al. (2012): smoking, prepregnancy body mass index, previous live birth, and gestational age; Chen et al. (2012): maternal age and gestational age.