既往疾病史与胰腺癌关系的流行病学研究

Medical history and the risk of pancreatic cancer:a case-control study in Shanghai,China

目的：探讨既往疾病史与胰腺癌的关系。方法：采用全人群病例一对照研究，研究对象为1990年10月～1993年6月期间确诊的、年龄在35～74岁的上海市区451例胰腺癌新发病例以及按性别、年龄（5岁1组）频数配对的1552例人群对照。采用非条件logistic回归模型分析既往疾病史与胰腺癌的关系。结果：男性中，糖尿病患者（OR=4．49，95％CI：2．21～9．12）、慢性胰腺炎患者（OR=7．05，95％CI：1．93～25．75）、胆囊切除者（OR=4．02。95％CI：1．75～9．26）以及肝硬化者（OR=4．22，95％CI：1．68～10．59）发生胰腺癌的危险性显著增加；女性糖尿病患者（OR=3．58，95％CI：1．89～6．76）、慢性胰腺炎患者（OR=6．15，95％CI：1．88～20．06）、胆囊切除者（OR=2．91，95％CI：1．54～5．48）以及肝硬化者（OR=6．92，95％CI：1．08～44．49）。去除胰腺癌诊断前或调查前5年内出现上述疾病史的研究对象后，男性糖尿病患者胰腺癌危险性仍显著增加（OR=3．47，95％CI：1．09～11．06）。其余结果虽多数危险性也升高，但均未达显著水平。结论：糖尿病、慢性胰腺炎、胆囊切除以及肝硬化病史可能会增加胰腺癌的发病风险，上述疾病史可能在胰腺癌病因学中起一定的作用。

Objective： To explore the relationship between medical history and the risk of pancreatic cancer. Methods： A population-based case-control study was conducted in urban Shanghai from October 1990 to June 1993 involving interviews with 451 incident cases of pancreatic cancer aged from 35 to 74 years and a population control （including 1 552 normals） frequencymatched to cases by gender and age in five year groups. An unconditional logistic regression was performed to calculate the adjusted odds ratios （ORs） and 95 % confidence intervals （CIs）. Results：Risks increased significantly among the male subjects with diabetes mellitus（OR=4.49,95%CI：2.21-9. 12）, chronic pancreatitis（OR= 7.05,95 %CI ： 1.93-25.75 ）, cholecystectomy （OR= 4.02,95%CI：1.75-9.26） and liver cirrhosis （OR=4.22,95%CI：1.68-10.59）. The corresponding ORs for women were 3.58 （95%CI：1. 89-6. 76） ,6.15（95%CI：1.88-20.06）,2.91（95%CI：1.54-5.48）, and 6.92（95%CI：1.08-44.49） with diabetes mellitus,chronic pancreatitis,cholecystectomy, and liver cirrhosis, respectively. Similar results were observed after further analysis with exclusion of those with the above medical conditions within 5 years prior to cancer diagnosis or interview. Risk associated with diabetes in male subjects wag still significantly increased （OR= 3.47,95%CI： 1.09-11.06）. The risk with other diseases also increased but the difference was not significant. Conclusion：Diabetes, chronic pancreatitis, cholecystectomy, and liver cirrhosis appear to be associated with an increased risk of pancreatic cancer and such medical conditions seem to play a role in the etiology of pancreatic cancer.