Risk of pancreatic cancer in workers exposed to chlorinated hydrocarbon solvents and related compounds: A meta-analysis

This is a meta-analysis of occupational exposures to chlorinated hydrocarbo n (CHC) solvents and pancreatic cancer, based primarily on studies that add ressed exposure directly (agent studies) and secondarily on studies that re ported data without verification of individual CHC exposures (job title stu dies), all of which were listed in databases for the period January 1969 to May 1998. Standardized extraction of data and double-checking of consisten cy of data extraction by five extractors were done. Simple random models es timated meta-relative risks. Suggestive weak excesses were found for trichl oroethylene (meta-relative risk (MRR) = 1.24, 95% confidence interval (CI): 0.79, 1.97), polychlorinated biphenyls (MRR = 1.37, 95% CI: 0.56, 3.31), m ethylene chloride (MRR = 1.42, 95% CI: 0.80, 2.53), and vinyl chloride (MRR = 1.17, 95% CI: 0.71, 1.91) but not for carbon tetrachloride. One study ad dressed tetrachloroethylene (MRR = 3.08, 95% CI: 0.63, 8.99); another inves tigated chlorohydrin production:(MRR = 4.92, 95% CI: 1.58, 11.4). Exposure- response meta-analyses for trichloroethylene and methylene chloride failed to reveal trends. Job title studies on metal degreasing and dry cleaning re vealed significant MRRs (2.0 and 1.4, respectively). Publication bias was u nlikely. Confounding may have remained insufficiently controlled. Unless th e results are seriously biased by exposure or endpoint misclassification or by confounding, strong causal associations between CHC compounds and pancr eatic cancer can be judged unlikely. Interactions between environmental and occupational agents, lifestyle factors, and genetic susceptibility remain a possibility, but the data for this meta-analysis did not address interact ions.