You are here:
Formaldehyde Exposure and Genotoxicity: Clearing the Fog using a Systematic Analysis (APHA)
Citation:
Glenn, B., J. Fritz, AND S. Vulimiri. Formaldehyde Exposure and Genotoxicity: Clearing the Fog using a Systematic Analysis (APHA). American Public Health Association Meeting, Denver, CO, October 29 - November 02, 2016.
Impact/Purpose:
Although findings reported in the literature appear heterogeneous, a consistent association of formaldehyde exposure with chromosomal and DNA damage emerges when accounting for exposure levels in exposed and comparison groups and methodological limitations. Our analysis illustrates how consideration of factors that may increase potential for bias and/or reduce study sensitivity can facilitate interpretation of seemingly disparate study results, thereby strengthening confidence in decisions pertinent to human health risk assessments.
Description:
Introduction: Formaldehyde has been linked to high exposures in mobile homes provided in response to Hurricane Katrina, imported engineered wood flooring, and hair straightening treatments. Measures of DNA and chromosomal damage in peripheral blood lymphocytes (PBLs) are established predictors of elevated cancer risk. Formaldehyde is associated with these measures in epidemiological studies, although some reviews concluded that the evidence was inconsistent. We conducted a systematic analysis of all epidemiological studies, considering methods and exposure levels to evaluate the consistency of the evidence. Methods: The literature search and screening were designed to answer the research question, is formaldehyde exposure to humans associated with elevated risk for genotoxicity? The literature search using defined terms identified 46 studies in diverse occupational and environmental settings that evaluated genotoxicity in nasal and buccal epithelial cells or PBLs. We evaluated the studies using predefined criteria related to study design, outcome and exposure measurements, sources of bias and reporting detail. Limitations in some studies (25%), including no blinding of sample analysis, potential confounding, inadequate analytic details, or very small sample size (N < 15) reduced confidence in their results. The evaluations informed interpretation of results, including potential over or underestimation of risks. Graphical displays of mean frequency ratios and confidence intervals by cell type showed DNA-protein crosslinks, micronuclei, chromosomal aberrations, and Comet assay results, stratified by possible explanatory variables for divergent results.