Ritz Beate, Zhao Yingxu, Krishnadasan Anusha, Kennedy Nola, Morgenstern Hal
Department of Epidemiology, School of Public Health, University of California-Los Angeles, 90095-1772, USA.
Epidemiology. 2006 Mar;17(2):154-61. doi: 10.1097/01.ede.0000199323.55534.fb.
Animal studies suggest that hydrazine is a lung carcinogen, but human studies have been rare, rather small, and limited to cancer mortality.
We examined cancer mortality and incidence in a cohort of aerospace workers with varying exposure to hydrazine contained in rocket fuels-extending previous mortality follow-up from 1994 to 2001 and investigating cancer incidence for the period 1988-2000 using population-registry data. We newly estimated hydrazine effects adjusting for occupational exposures to other carcinogens assessed through a job-exposure matrix. Rate-ratio estimates were derived from Cox proportional hazards and random-effects models using time-dependent exposure measures for hydrazine adjusting for trichloroethylene, polycyclic aromatic hydrocarbons, benzene, and mineral oil exposures.
Exposure to hydrazine was positively associated with lung cancer incidence (estimated rate ratio for high vs low exposure with 20-year lag = 2.5; 95% confidence interval = 1.3-4.9) and with colorectal cancer incidence (2.2; 1.0-4.6). Dose-response associations were observed for both outcomes; similar associations were found for lung cancer mortality but not for colorectal cancer mortality. Effect estimates for cancers of the pancreas, blood and lymph system, and kidneys were based on small numbers rendering our analyses uninformative, and patterns considering exposure levels and lags were inconsistent. Use of random-effect models did not change our results.
The findings reported here are consistent with our previous results for lung cancer mortality; our new results suggest that exposure to hydrazine increases the risk of incident lung cancers. We also found, for the first time, an increased risk of colon cancers. Results for other cancer sites are inconclusive.
动物研究表明肼是一种肺部致癌物，但人体研究较少，规模较小，且仅限于癌症死亡率。
我们调查了一组接触火箭燃料中所含肼程度不同的航空航天工人的癌症死亡率和发病率——将先前的死亡率随访从1994年延长至2001年，并利用人口登记数据调查1988 - 2000年期间的癌症发病率。我们新估计了肼的影响，并针对通过工作暴露矩阵评估的其他致癌物的职业暴露进行了调整。率比估计值来自Cox比例风险模型和随机效应模型，使用肼的时间依赖性暴露测量值，并对三氯乙烯、多环芳烃、苯和矿物油暴露进行了调整。
接触肼与肺癌发病率呈正相关（高暴露与低暴露且有20年滞后的估计率比 = 2.5；95%置信区间 = 1.3 - 4.9），与结直肠癌发病率也呈正相关（2.2；1.0 - 4.6）。两种结果均观察到剂量反应关系；肺癌死亡率也发现了类似关系，但结直肠癌死亡率未发现。胰腺、血液和淋巴系统以及肾脏癌症的效应估计值基于少量病例，使我们的分析无信息量，且考虑暴露水平和滞后的模式不一致。使用随机效应模型并未改变我们的结果。
此处报告的结果与我们先前关于肺癌死亡率的结果一致；我们的新结果表明，接触肼会增加患肺癌的风险。我们还首次发现患结肠癌的风险增加。其他癌症部位的结果尚无定论。
