Article Text
Abstract
Purpose We investigated the role of occupational exposure to endotoxin in lung cancer risk in a French population based case-control study.
Methods Lifelong work histories of 2926 cases and 3555 controls were collected using standardised questionnaires. Estimation of endotoxin exposure was based on a publication issued by the Institut National de Recherche et de Sécurité (INRS), which reported measurements in biological aerosols workstations. Several exposure-related variables were created: intensity (<1000UE/m³; >1000UE/m³), total duration (years), life time cumulative exposure (UE/m³*years) and time since cessation (years). Smoking history was combined into a comprehensive smoking index (CSI) that included mean number of cigarettes/day, duration and time since cessation. Odds ratios (OR) and 95 % confidence intervals (CI) were estimated using unconditional logistic regression models and controlled for main confounding factors including CSI.
Results We found an inverse significant association between high endotoxin exposure and lung cancer risk (151 cases/336 controls; OR = 0.64, 95% Cl = 0.5–0.82). The main activity sectors entailing high exposure included dairy, cattle, poultry and pig farms. In contrast, a statistically significant positive association was found for low exposure (OR = 1.33, 95% Cl = 1.08–1.64) (eg. sawmills, wool industry, meat processing). Using restricted splines cubic models, we observed that the association between cumulative endotoxin exposures and lung cancer risk departed significantly from linearity. In addition, the risk decreased with duration and increased with time since cessation, irrespective of the exposure intensity. Analysis by subgroups showed an interaction with tobacco consumption, and light smokers appeared to be more sensitive to endotoxin protective effect.
Conclusions Our findings suggest that high exposure to endotoxin confers protection against lung cancer.