Background Despite several case reports describing liver toxicity of nitrosamines and the fact that some N-nitroso compounds are used to induce cirrhosis of the liver in animal models, this association has not been investigated in epidemiological studies. Methods A cohort of 2,875 female rubber workers who were active on January 1, 1976, or hired thereafter, and who had been employed for at least 1 year in one of five plants producing tires or technical rubber goods was followed for mortality from January, 1976, through December, 1991. Work histories were reconstructed using routinely documented cost center codes and classified into six work areas. Age and calendar year standardized mortality ratios (SMR) and 95% confidence intervals (95% CI) were calculated and stratified by plant, work area, year of hire, and years of employment in the respective work area. Results The excess mortality from cirrhosis of the liver was most pronounced for nonalcohol-related cirrhosis of the liver (ICD-9 571.4-571.9: 10 deaths, SMR 202; 95% CI 97-372). Mortality from alcohol-related cirrhosis of the liver (ICD-9 571.0-571.3: 3 deaths, SMR 153; 95% CI 31-446) and from other alcohol-related diseases (organic psychoses, injury, and poisoning) was not statistically significantly elevated. All 10 cases of nonalcohol-related cirrhosis had worked in production of technical rubber goods (SMR 279; 95% CI 134-514) and risks increased with earlier years of hire and with longer duration of employment in this work area. Discussion Although our results must be interpreted with caution, they suggest that the observed excess deaths from cirrhosis of the liver are associated with occupational risk factors. In light of additional evidence from case reports and animal data, exposure to nitrosamines may be a plausible risk factor for the observed excess mortality.

Support: Paper