317 Disease group differences in benzene exposure in a pooled analysis of petroleum worker case-control studies
本研究汇总英国、加拿大和澳大利亚石油行业队列，比较不同淋巴造血系统疾病亚型（如急性髓系白血病、骨髓增生异常综合征等）的苯暴露水平，发现总体暴露较低，但骨髓增生异常综合征病例的累积暴露高于对照组和急性髓系白血病病例。
<h3>Objectives</h3> To compare exposure to benzene across different lymphohaematopoietic subtypes in a pooled nested cased control study. <h3>Methods</h3> Cases of acute myeloid leukaemia (AML), chronic myeloid leukaemia (CML), chronic lymphocytic leukaemia (CLL) myelodysplastic syndrome (MDS) and myeloproliferative disease (MPD) were identified in petroleum industry cohorts from UK, Canada and Australia and pooled with matched controls into a single study. For each job held by participants, the intensity of full shift exposure to benzene (ppm), the certainty of the estimate, job duration, probability of dermal and peak exposure were estimated using study defined metrics. Cumulative benzene exposure was calculated for each participant. <h3>Results</h3> Benzene exposure was low, 90% of participants accumulated < 20 ppm-years. Mean cumulative exposure was 5.15 ppm-years, mean duration was 22 years and mean exposure intensity was 0.2 ppm. 46% of participants were allocated a peak exposure and 40% had a high probability of dermal exposure at least weekly. For AML, mean duration was slightly longer for cases but maximum exposure intensity was slightly higher for controls. The mean cumulative and maximum exposure estimates were higher for MDS cases than for controls or for AML cases. CLL and CML cases had slightly higher mean estimates of duration and cumulative benzene exposure than their controls. MPD cases had lower exposure estimates than controls. AML cases and controls had longer employment duration than other disease groups. Higher exposure certainty scores confirmed the MDS case/control and MDS/AML differences. <h3>Conclusions</h3> Benzene exposure estimates were lower overall than those in previous studies for all the disease groups. There was little difference in exposure between AML cases and controls and in general the exposure was less than for MDS cases. Certainty scores for exposure assessments can be used in sensitivity analyses to strengthen interpretations of the observed associations between exposure and risk.