A Comparative Case-Control Analysis of Stomach Cancer and Atrophie Gastritis1

We conducted a comparative case-control analysis of stomach cancer and atrophie gastritis involving 427 cases with stomach cancer, 1414 cases with atrophie gastritis, and 3014 control subjects based on a questionnaire survey conducted for the subjects who received gastroscopie examination at Aichi Cancer Center Hospital from April 1985 to March 1989. The risk of atrophie gastritis in both males and females was not associated with any environmental factors. The risk of stomach cancer compared with the control subjects was positively associated with an intake of salted fish guts or cod roe (relative risk (RR) = 1.52, 95% confidence interval (CI) = 1.08-2.15) and smoking (RR for 20 or more cigarettes per day = 2.84; 95% Cl = 1.79-4.51) and inversely associated with Western-style breakfast (RR = 0.68; 95% CI = 0.48-0.96) in males. Additionally, the risk of stomach cancer was inversely associated with a daily intake of raw vegetables (RR = 0.56; 95% CI = 0.34-0.94) in males when compared with the patients with atrophie gastritis as controls. Several environmental factors, such as intake of green-yellow vegetables, fruit, and meat, and a family history of stomach cancer, were only associated with intestinal types of cancer in females, whereas a clear difference between diffuse and intestinal types was not observed in males. The results of the present study suggest that risk factors for stomach cancer may be different from those for premalignant lesions. INTRODUCTION Japan has been a high-risk area for stomach cancer. The mortality rate from this cancer during 1978-1979 was highest among 39 countries (1). The previous pathological and epidemiological evidence has suggested that the intestinal type of stomach cancer is predominant in high-risk areas and closely associated with atrophie gastritis and intestinal metaplasia (24). Correa et al. (5, 6) proposed a hypothesis linking stomach cancer with atrophie gastritis and intestinal metaplasia and with the conversion of nitrate to mutagenic 7V-nitroso com pounds. In their hypothesis, the development of atrophie gas tritis is the first important stage caused by environmental agents that injure the gastric mucosa. According to their hypothesis, it is considered that identification of risk factors for atrophie gastritis may lead to a more effective primary prevention for stomach cancer in high-risk populations. Furthermore, identi fication of specific and common risk factors for both atrophie gastritis and stomach cancer may provide useful information to interrupt each part of the sequence to stomach cancer. There fore, we conducted a comparative case-control analysis of atrophie gastritis and stomach cancer based on a cohort that received gastroscopie examination. MATERIALS AND METHODS We conducted a questionnaire survey for the patients who received gastroscopie examination at Aichi Cancer Center Hospital from April Received 12/7/89; accepted 7/13/90. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' Presented at the Sixth Symposium on Epidemiology and Cancer Registries in the Pacific Basin, Kauai, Hawaii. November 13-17, 1989. This study was supported by Grant-in-Aid for a Comprehensive 10-year Strategy for Cancer Control. Japan, from the Ministry of Health and Welfare. 1To whom requests for reprints should be addressed. 6559 1985 to March 1989. A total of 7019 questionnaires was distributed and 6226 were collected (88.7%). After excluding the patients with cancers other than stomach cancer and the patients with resected stomach, a total of 5859 subjects comprised the original cohort. Pa tients with resected stomach were not included in the study because diagnosis for the presence and degree of atrophie gastritis was difficult in those patients. The results of endoscopie examination were judged by the 6 physi cians specialized for gastroenterology in the Aichi Cancer Center Hos pital. The diagnosis of stomach cancer was based on histological examination and that of atrophie gastritis was based on endoscopie findings, because histological examination, which was done for about 20% of the subjects without stomach cancer, was performed primarily to negate cancer, but not to determine the degree of atrophie gastritis. This was supported by the previous report about the consistency be tween endoscopie and histological findings on atrophie gastritis (7). Atrophie gastritis was classified into 3 groups, mild, moderate, and severe, according to the size of transparent blood vessels and discolor ation in the gastric mucosa. Mild atrophie gastritis was defined as gastric mucosa with transparent fine blood vessels and yellowish dis coloration. Moderate atrophie gastritis was defined as gastric mucosa with clearly transparent blood vessels and yellow-grayish discoloration. Severe atrophie gastritis was defined as gastric mucosa with transparent large blood vessels and gray-greenish discoloration. Because of difficul ties of discrimination between normal mucosa and mild atrophie gas tritis and between moderate and severe atrophie gastritis, the patients with moderate or severe atrophie gastritis without other gastric lesions were defined as atrophie gastritis group and the patients with normal gastric mucosa and the patients with mild atrophie gastritis only were defined as the control group in this case-control analysis. The analysis was conducted for 3 groups: (a) 427 patients with stomach cancer; (b) 1414 patients with atrophie gastritis; and (c) 3014 patients in the control group. The rest of the 1004 patients with other gastroduodenal diseases, such as peptic ulcer, polyp, and other types of gastritis (super ficial or acute), were excluded from the present analysis. Age distribu tions of the study subjects are shown by sex in Table 1. Compared with the controls, the proportion of the subjects aged 55 and over was higher in the patients with atrophie gastritis and much higher in the patients with stomach cancer. Females were predominant in the controls, while males were predominant in the patients with stomach cancer. The proportion of the subjects who lived outside of Aichi prefecture was higher in the patients with stomach cancer. The questionnaire was self-recorded and included items on symp toms, medical history, family history, dietary habits, smoking and drinking habits, and psychological factors. Frequencies of intakes of Table 1 Age distribution of study subjects Controls (%)Age group-2930-3435-3940-4445-4950-5455-5960-6465-6970-Male8.05.911.414.514.513.211.29.16.45.8Female5.44.912.916.614.715.612.89.34.53.3Atrophie gastritis (%)Male1.42.25.113.615.716.213.114.88.59.5Female0.91.79.011.112.216.519.314.58.06.8Stomach cancer (%)Mal 0. 2. 3. 5.9 .710.714.219 412.521.5Female0.7 .66. . 8.715.213.012.314.517.4 Total(no.) 1247 1767 766 648 289 138 on July 21, 2017. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from STOMACH CANCER AND ATROPHIC GASTRITIS Table 2 Relative risk estimates for atrophie gastritis and stomach cancer in males Atrophie gastritis (n = 766) compared with control (n = 1247)Stomach cancer (n = 289) compared withControl («= 1247)FactorRiceSoy bean paste soupPickled vegetablesPickled Japanese apricotsSalted or dried fish Salted fish gut. cod roe Foods boiled down in soy Green-yellow veg tabl sRaw vegetablesFruitsMeatBreakfast Supper timeHot green teaDrinki g habitsSmoking habitsPast history Family history OccupationDefinition3 cups/day £4 cups/day2-3 times/wk Daily2-3 times/wk Daily2-3 times/wk DailyS2-3 times/wk £2-3 times/wk S2-3 times/wk Daily2-3 times/wk Daily2-3 times/wk Daily2-3 times/wk DailyWestern style After 8 p.m.1-4 cups/day 35 cups/dayOccasional DailyExsmoker 1-19 cigarettes/day S20 cigarettes/dayGastric ulcer Stomach cancer Professional or administrativeRR"1.151.201.01 0.961.11 0.990.78 0.900.96 0.96 0.91 0.960.87 0.981.11 1.131.22 1.260.92 1.101.04 1.001.01 1.040.96 0.80 0.891.09 1.12 0.8095% CI0.90-1.47 0.95-1.520.67-1.52 0.65-1.420.85-1.45 0.75-1.300.64-0.96 0.69-1.180.78-1.17 0.77-1.20 0.75-1.10 0.79-1.160.64-1.19 0.70-1.360.82-1.52 0.82-1.550.80-1.85 0.81-1.960.75-1.120.88-1.380.83-1.30 0.78-1.290.79-1.29 0.82-1.310.75-1.23 0.56-1.13 0.70-1.140.84-1.42 0.89-1.41 0.64-0.99RR0.90 1.160.91 1.04.54.37.04 .35.24 .50 0.87 0.810.77 0.591.02 0.831.67 1.260.71 1.461.14 1.010.77 0.991.811.93 2.812.38 0.92 0.7095% CI0.63-1.29 0.82-1.630.50-1.65 0.60-1.811.00-2.39 0.88-2.130.76-1.42 0.93-1.970.91-1.68 1.09-2.07 0.65-1.16 0.61-1.090.51-1.150.37-0.930.64-1.62 0.51-1.330.90-3.09 0.65-2.450.52-0.97 1.01-2.090.82-1.60 0.70-1.470.53-1.110.71-1.371.17-2.79 1.13-3.30 1.83-4.291.70-3.34 0.65-1.30 0.50-0.98Atrophie gastritis («= 766)RR0.80