Cai L, Zheng ZL, Zhang ZF. Risk factors for the gastric cardia cancer: A case-control study in Fujian Province. World J Gastroenterol 2003; 9(2): 214-218 [PMID: 12532434 DOI: 10.3748/wjg.v9.i2.214]
Corresponding Author of This Article
Lin Cai, Department of Epidemiology, Fujian Medical University, Fuzhou, 350004, Fujian Province, China. zjcailin@pub5.fz.fj.cn
Article-Type of This Article
Esophageal Cancer
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Lin Cai, Zong-Li Zheng, Department of Epidemiology, Fujian Medical University, Fuzhou, 350004, China
Zuo-Feng Zhang, Department of Epidemiology, UCLA School of Public Health, Los Angeles California, USA
ORCID number: $[AuthorORCIDs]
Author contributions: All authors contributed equally to the work.
Supported by the Natural Science Foundation of Fujian Province, China, No. C001009 and the Foundation for the Author of National Excellent Doctoral Dissertation of P. R. China, No. 200157
Correspondence to: Lin Cai, Department of Epidemiology, Fujian Medical University, Fuzhou, 350004, Fujian Province, China. zjcailin@pub5.fz.fj.cn
Telephone: +86-591-3569264
Received: June 14, 2002 Revised: July 4, 2002 Accepted: July 15, 2002 Published online: February 15, 2003
Abstract
AIM: The incidence of gastric cardia cancer has greatly increased in the past 2-3 decades, however, the risk factors for the disease are still not clearly understood. The investigations among Chinese population on the risk factors of gastric cardia cancer were also scarcely reported. We therefore conducted a case-control study in Fujian province, China, to investigate the potential risk and protective factors of this disease.
METHODS: 191 cardia and 190 non-cardia gastric cancer cases, and a total of 222 control cases were included in this study. Standard questionnaires were used in collecting epidemiological factors and the data were then analyzed by the unconditional logistic regression model.
RESULTS: As the factors such as age, gender, smoking, alcohol consumption, and family history of gastric cancer were controlled, a multivariable analysis was conducted, which revealed that there was a significant correlation between the dietary habits such as irregular meal, over and fast eating, and the gastric cardia cancer with the odds ratios (ORs) of 4.2 (95% confidence interval: 2.3-7.7), 4.7 (2.1-10.8), and 2.7 (1.3-5.3) respectively. Other correlations were also observed between the gastric cardia cancer and the consumption of salty fish or pickled vegetable, smoking, and the family cancer history with the ORs of 5.5 (1.4-19.5), 1.8 (1.0-3.0), 2.1 (1.3-3.5), and 3.8 (2.3-6.2) respectively. In contrast, the negative correlations were found existing between the intake of fresh vegetables and fruits, the use of refrigerator, and the gastric cardia cancer, with the ORs of 0.4 (0.2-0.9), 0.2 (0.1-0.5), and 0.2 (0.1-0.4), respectively. However, dietary habits were associated less with non-cardia gastric cancer compared with its cardia counterpart.
CONCLUSION: Dietary habits might be one of the risk factors for the cardia carcinogenesis among Chinese population.
Key Words: $[Keywords]
Citation: Cai L, Zheng ZL, Zhang ZF. Risk factors for the gastric cardia cancer: A case-control study in Fujian Province. World J Gastroenterol 2003; 9(2): 214-218
While the total incidence and mortality of gastric cancer manifested a declining trend world wide, the incidence of gastric cardia cancer has increased considerably in the past few decades[1-4]. The explanations for this phenomenon, however, have not been fully elucidated yet[5-6]. Although certain dietary factors have been suggested to be associated with the gastric carcinogenesis, there are few investigations performing the etiological analysis further on the anatomical sub-site of the cancer[7-11], whose divergent carcinogenesis patterns suggested that there might exist differences in etiological factors between the cardia and non-cardia gastric cancers. To verify this hypothesis, we have conducted a case-control study in Fujian Province, China.
MATERIALS AND METHODS
Subjects
191 cardia and 190 non-cardia primary gastric cancer cases, which were pathologically confirmed or diagnosed through surgical operation between April 2000 and June 2001 in six hospitals in Fuzhou, were employed as the observation group and the cancer controls respectively. The 222 cases of normal control, which came from the department of orthopedics in the same hospitals, were selected without cancers or gastric diseases and frequency-matched to the observation cases by age (± 3 year) and gender. All subjects in our study must have resided in Fujian Province for more than 20 years and can answer questions clearly.
Investigation
Face-to-face interviews were conducted in the hospitals by trained interviewers, in which a structured questionnaire containing socio-demographic characteristics, personal medical history, tobacco and alcohol consumption habits, family cancer history, dietary history and dietary habits was used. As for the dietary habits, the exposure time of 10 years before the interview was emphasized to ensure the assessment of potential risk factors with a reasonable latent period before the onset of cancer. Although the dietary habits could be changed presently by the course of the disease, they would never be influenced 10 years earlier. With the agreement to be interviewed, the observed case and his or her control counterparts were investigated by the same interviewer.
Statistical analysis
The data were statistically evaluated with Epi-info by the univariate and multivariate analysis. The relationship between the putative risk factors and the sub-sites of gastric cancer was assessed by odds ratios (ORs) and their 95% confidence intervals (CIs) derived from unconditional logistic regression model. Multivariate logistic regression analysis was used to evaluate simultaneously the effects of multiple factors and other potential confounding factors.
RESULTS
A total of 603 subjects from 30 to 79 years old were included in this study. In the observation group, there were 156 male and 35 female with the average age of 61.2 (± 9.7) and 57.3 (± 10.9) years old respectively. Table 1 shows the demographic characteristics of age and gender for the observed cases and controls. To minimize the possible confounding effects on the investigation results, the age and the gender were taken as the confounding factors to be controlled in all of our analyses.
Table 1 Age and sex distribution of the observed cases and the controls.
Age (yrs)
Controls
Cardia cancer
Gastric cancer
Total
P value
n
(%)
n
(%)
n
(%)
n
(%)
Female
30-
13
(22.8)
2
(5.7)
9
(15.8)
24
(16.1)
45-
15
(26.3)
15
(42.9)
13
(22.8)
43
(28.9)
55-
13
(22.8)
9
(25.7)
19
(33.3)
41
(27.5)
65-
16
(28.1)
9
(25.7)
16
(28.1)
41
(27.5)
0.209
Male
30-
29
(17.6)
10
(6.4)
17
(12.8)
56
(12.3)
45-
37
(22.4)
27
(17.3)
31
(23.3)
95
(20.9)
55-
44
(26.7)
49
(31.4)
42
(31.6)
135
(29.7)
65-
55
(33.3)
70
(44.9)
43
(32.3)
168
(37.0)
0.025
The relationship between the dietary habits and gastric cancers are shown in Table 2 with the factors such as age, gender, smoking, alcohol consumption and family cancer history being controlled. Irregular meal, over and fast eating could increase the risk of cardia carcinogenesis. Intake of salty fish, pickled vegetables, lard oil and drinking of the well water were also the risk factors for the cardia cancer. In contrast, the intake of fresh vegetables and fruits and the use of refrigerator were belong to the protective factors against the cancer. Besides, it was found that low income, poor education and bearing a family cancer history in the first-degree relatives were associated with the cardia cancer risk. However, the dietary habits were associated less with non-cardia cancer compared with its cardia counterpart. Alcohol consumption was not observed relating to the cardia cancer risk in the present study.
Table 2 Potential risk or protective factors for cardia and non- cardia gastric cancers, odds ratios & 95% confidence Interval (CI).
Variables
Controls
Cardia cancer
Non-cardia gastric cancer
n
n
OR*
95%CI
P
n
OR*
95%CI
P
Irregular meals
No
165
101
98
Sometimes
36
45
1.971
1.152-3.374
0.013
50
2.310
1.375-3.897
0.002
Frequently
21
45
4.184
2.260-7.747
0.000
42
3.662
1.997-6.717
0.000
P For trend
0.000
0.000
Over eating for each meal
No
105
103
156
Sometimes
55
58
1.864
1.139-3.050
0.013
63
1.806
1.122-1.908
0.015
Frequently
11
28
4.720
2.067-10.781
0.000
24
3.499
1.569-7.801
0.002
P For trend
0.034
0.000
Eating speed
Slow
33
18
20
Moderate
110
67
1.058
0.531-2.107
0.874
85
1.308
0.688-2.485
0.412
Fast
79
106
2.645
1.318-5.307
0.006
85
1.878
0.968-3.644
0.063
P For trend
0.000
0.043
Years of education
≤ 6
90
121
96
> 6
132
70
0.317
0.204-0.492
0.000
94
0.643
0.426-0.972
0.036
Monthly Food Expense (RMB)
< 80
62
79
65
80-
61
60
0.653
0.387-1.102
0.110
64
0.971
0.580-1.625
0.911
100-
59
34
0.397
0.223-0.710
0.002
46
0.697
0.405-1.199
0.192
200-
40
18
0.313
0.155-0.630
0.001
15
0.356
0.175-0.726
0.004
P For trend
0.000
0.044
Salty fish intake
< 1 times/M
170
139
134
< 3 times/W
49
41
t1.019
0.615-1.690
0.941
41
1.058
0.646-1.730
0.824
≥ 3 times/W
3
11
5.518
1.367-19.461
0.015
15
5.706
1.588-20.49
0.008
P For trend
0.111
0.020
Fresh vegetables intake
< 3 times/W
14
25
18
≥ 3 times/W
208
166
0.439
0.212-0.909
0.027
172
0.644
0.303-1.367
0.252
Fresh fruit intake
< 1 times/M
30
52
44
< 3 times/W
127
116
0.565
0.328-0.973
0.040
115
0.586
0.339-1.015
0.057
≥ 3 times/W
65
23
0.225
0.112-0.452
0.000
31
0.289
0.149-0.561
0.000
P For trend
0.000
0.000
Pickled vegetables intake
< 1 times/M
88
61
69
< 3 times/W
76
63
1.335
0.804-2.214
0.264
64
1.139
0.701-1.852
0.598
≥ 3 times/W
58
67
1.762
1.044-2.974
0.034
57
1.269
0.766-2.103
0.355
P For trend
0.038
0.367
Cooking oil
Other type
56
32
28
Peanut oil
149
103
0.533
0.347-0.820
0.004
115
0.738
0.485-1.124
0.157
Lard oil
93
117
2.327
1.521-3.562
0.000
107
1.853
1.230-2.794
0.003
Drinking water
Other type
27
30
32
Well water
60
98
2.822
1.826-4.360
0.000
76
1.804
1.176-2.768
0.007
Tap water
135
63
0.301
0.195-0.464
0.000
82
0.461
0.306-0.694
0.000
Use of refrigerator
No
83
108
86
< 10 yrs
45
46
0.764
0.449-1.302
0.323
53
1.155
0.689-1.935
0.585
≥ 10 yrs
94
37
0.242
0.143-0.409
0.000
51
0.426
0.263-0.693
0.001
P For trend
0.000
0.007
Family Cancer History
No
190
122
131
Yes
32
69
2.090
1.261-3.464
0.004
59
1.325
0.814-2.157
0.257
Table 3 shows the interrelation between cardia cancer and smoking. In smokers, the odds ratio for cardia cancer is 2.09 after adjusted by age, gender, alcohol drinking and family cancer history. For those who smoke more than 20 cigarettes a day, the risk for cardia cancer was 2-fold or more compared with non-smokers. We also noted a significant dose-risk relationship existed between the daily consumed cigarettes, smoking duration and the cardia cancer. However, it was not existed between smoking and the non-cardia gastric cancer.
Table 3 The interrelation between smoking and kinds of gastric cancers.
Variables
Controls
Cardia cancer
Non-cardia gastric cancer
n
n
OR*
95%CI
P
n
OR*
95%CI
P
Smoking
99
No
120
74
Yes
102
117
2.090
1.261-3.464
0.004
91
1.325
0.814-2.157
0.257
Daily consumed cigarettes
0
120
74
99
< 19
37
28
1.443
0.753-2.766
0.270
27
1.092
0.579-2.062
0.785
≥ 20
65
89
2.513
1.455-4.339
0.001
64
1.478
0.863-2.531
0.155
P For trend
0.000
0.472
Smoking duration (yrs)
0
120
74
99
< 30
53
37
1.408
0.755-2.625
0.282
36
1.073
0.596-1.933
0.814
≥ 30
49
80
2.717
1.548-4.768
0.000
55
1.600
0.906-2.825
0.105
P For trend
0.000
0.289
Unconditional logistic analyses were made by a backward elimination approach to select a possible best subset of risk factors for cardia cancer. The results showed that the best subset of risk factors included family cancer history, irregular meal, fast eating, well-water drinking, smoking, and poor-educated. Fresh fruit intake and the use of refrigerator have protective effect against cardia cancer. (Table 4)
Epidemiological studies have indicated that while the overall incidence of gastric cancer has been decreasing, the morbidity of cardia gastric cancer has constantly gone up, which might reflect a different etiology for the cardia gastric cancer[12-14]. However, little attention has been paid to the risk factors for this increasing trend of cardia gastric cancer. The correlation between the risk factors and the location of gastric cancer has also seldom been investigated[15-17].
In our study the dietary habits were identified as the risk factors for cardia cancer. Irregular meal, over-eating, and the fast eating may cause injures to the digestive tract mucosa and promote the carcinogenesis. The protective effects of taking in fresh vegetables and fruits against gastric cancer may be mediated by anti-oxidants such as ascorbic acid. Therefore, intake of more fresh fruits and vegetables may significantly lower the risk of gastric cancer, which is consistent with the former epidemiological studies[18-21]. The present study also showed that the more the pickled vegetables and salty fish were taken in, the more the risk for cardia cancer were increasing, which might be attributed to the potential carcinogens, such as nitroso compounds contained in these foods, and may also involved the increased cell replication and susceptibility to carcinogenesis from salt intake[22-24]. Lard oil intake was also associated with the increased risk of gastric ca cancer, probably because of excessive exposures of people to the saturated fatty acids. We found that in comparison of the cardia with non-cardia cancer, the carcinogenic risk with specific food or food groups were similar; however, the dietary habits associated more closely with the cardia cancer than with the other stomach carcinomas.
Long-term use of refrigerators could decrease the risk of both cardia and non-cardia gastric cancer with the adjusted OR of 0.242 (95% CI: 0.143-0.409) and 0.426 (95% CI: 0.263-0.693) respectively. Refrigerator may keep foods and vegetables fresh for longer period of time, reduce the possibility of generating nitroso compounds, and maintain vitamins and other antioxidants at a higher level, which in turn protect the individuals from exposure to nitroso compounds and other carcinogens. Effective food preservation, which slows down the conversion of nitrates ions to highly carcinogenic nitrites, has been suggested to be at least partially responsible for the overall decreasing trend of gastric cancer incidence.
Well-water, especially shallow well-water may be more easily polluted than tap water. In this study, the fact that well-water drinking was related to an increased risk of cardia cancer suggested that potentially detrimental materials in polluted drinking water might be involved in the cardia carcinogenesis.
After adjusted by age, gender, drinking water and family cancer history in the first-degree relatives, cigarette smoking manifested as a risk factor for the cardia cancer in this study. There was a longer smoking exposure history in the cardia cancer cases than in both cancer- free controls and non-cardia gastric cancer patients. Significant positive dose-effect relationship was found both in the daily cigarette consumption (P < 0.001) and the smoking duration (P < 0.001). These findings suggested that the smoking exposure was a specific risk factor for the cardia cancer, and the results were influenced less by the recall bias, as it is implausible that patients with cardia cancer would recall smoking exposure more accurately than the non-cardia gastric cancer cases. It is also impossible for the cardia cancer patients to think the more about smoking as the possible cause of their illness than the non-cardia gastric cancer patients, even though cancer cases were aware of their disease status. Therefore, taking multiple controls of different types can be valuable for exploring alternate hypotheses and for taking into account possible potential bias such as recall bias[25].
Low socio-economic status and poor education were also related to the gastric cancer risk[26]. The cancer-free controls tended to be better educated than the cancer cases in this study. Family aggregation of the cardia gastric cancer has been confirmed in this study with adjusted OR of 2.09. The correlation between the cardia gastric cancer and family cancer history in the first-degree relatives suggested that inherited genetic susceptibility and shared environmental risk factors might contribute to the cardia carcinogenesis[27,28].
In conclusion, our study indicated that dietary habits and family cancer history are the risk factors both for cardia and non-cardia gastric cancer. However, dietary habits and smoking were associated more with the cardia cancer than the non-cardia carcinomas. The different epidemiological features between the cardia and the non-cardia gastric cancer suggested that these tumors might have different etiologic profiles. Therefore, further studies on environmental and host determinants of gastric candia cancer are needed. Our study also suggested that interventions against cigarette smoking and bad dietary habits may be important for the prevention of gastric cancer. As part of the strategy for the cancer control, healthy lifestyle should be emphasized as every one's responsibility[29,30]. Daily intake of fresh vegetables and fruits should be encouraged[31-33].
Cai L, Yu SZ. A molecular epidemiologic study on gastric cancer in Changle, Fujian Province.Shijie Huaren Xiaohua Zazhi. 1999;7:652-655.
[PubMed] [DOI][Cited in This Article: ]
Cai L, Yu SZ, Ye WM, Yi YN. Fish sauce and gastric cancer: An ecological study in Fujian Province,China.World J Gastroenterol. 2000;6:671-675.
[PubMed] [DOI][Cited in This Article: ]
Cai L, Yu SZ, Zhang ZF. Helicobacter pylori infection and risk of gastric cancer in Changle County,Fujian Province,China.World J Gastroenterol. 2000;6:374-376.
[PubMed] [DOI][Cited in This Article: ]
Cai L, Yu SZ, Zhang ZF. Glutathione S-transferases M1, T1 genotypes and the risk of gastric cancer: A case-control study.World J Gastroenterol. 2001;7:506-509.
[PubMed] [DOI][Cited in This Article: ]
Cai L, Yu SZ, Zhan ZF. Cytochrome P450 2E1 genetic polymorphism and gastric cancer in Changle, Fujian Province.World J Gastroenterol. 2001;7:792-795.
[PubMed] [DOI][Cited in This Article: ]
Shen H, Xu Y, Zheng Y, Qian Y, Yu R, Qin Y, Wang X, Spitz MR, Wei Q. Polymorphisms of 5,10-methylenetetrahydrofolate reductase and risk of gastric cancer in a Chinese population: A case-control study.Int J Cancer. 2001;95:332-336.
[PubMed] [DOI][Cited in This Article: ][Cited by in F6Publishing: 4][Reference Citation Analysis (0)]
Liu XM, Wang QS, Ma J, Lin XP. A case-control study on the factors of stomach cancer in Tianjin city.Zhonghua Liuxing Xingxue Zazhi. 2001;22:362-363.
[PubMed] [DOI][Cited in This Article: ]
Li SP, Ding JH, Gao CM, Zhou JN, Wu JZ, Su P, Zang Y, Liu YT, Zhou XF, Ding BG. A case-control study of esophageal and stomach cancers in high incidence area of upper-digestive tract cancer.Zhongliu. 2001;21:277-279.
[PubMed] [DOI][Cited in This Article: ]
Deng DJ, E Z. Overview on recent studies of gastric carcinogenesis: human exposure of N-nitrosamides.Shijie Huaren Xiaohua Zazhi. 2000;8:250-252.
[PubMed] [DOI][Cited in This Article: ]
Ye WM, Yi YN, Luo RX, Zhou TS, Lin RT, Chen GD. Diet and gastric cancer: A casecontrol study in Fujian Province, China.World J Gastroenterol. 1998;4:516-518.
[PubMed] [DOI][Cited in This Article: ]
Gordis L. Case-control and cross-sectional studies. In: Epidemiology. Second edition, New York: W.B.Saunders Company. 2000;140-157.
[PubMed] [DOI][Cited in This Article: ]