The sex and age distribution
During the period there were 8 559 males and 5 648 females with EC. The incidence rate among males was 142.80/100 000, while females was 95.18/100 000. The sex ratio was 1.50:1. The sex ratios (males to females) in 1970s, 1980s and 1990s respectively was 1.57:1, 1.56:1 and 1.35:1.
   The minimum age group of incidence was 1-year group over 23 years. The incidence rate of EC increased with age after 30 years old. It reached the highest at 80 years old group. In 1970s, the minimum age group was 25 years old group. The incidence rate had significantly increased after 35 years of age, and reached the highest level at 70 years old, then declined in 75-year old group. In 1980s, the minimum incidence group was one-year group. The incidence rate increased with age after that 25-year old group. The male incidence rate reached the highest at 80-year-old group. In females at the 60-year-old group it declined a little and reached the highest at 80-year-old group. In 1990s the minimum incidence group was 10-year-old group. The incidence rate increased with age after 25-year-old group and reached the highest at 80-year-old group. The male incidence rate reached the highest at 80 years old group, while the female declined little at 60-year-old group and reached the highest at 80-year-old group.

Geographic Distribution
In the mountainous area there were 2 106 EC cases from 1974 to 1996. The annual average incidence rate was 112.14/100 000. The incidence rate of EC in 1974 was 213.60/100 000, which declined to 82.55/100 000 in 1996. Decreased by 131.05/100 000, and the decline rate was 61.35 percent. From Table 2 and Figure 3 we could find that the incidence rate of EC in mountainous area had significantly decreased. The result of trend test was x²=149.93, P<0.001. The incidence rates of EC in 1970s, 1980s, and 1990s were 150.44/100 000, 109.85/100 000 and 75.55/100 000, respectively. The U test result between 1970s and 1990s, between 1980s and 1990s was P<0.01. Both of them had significant difference.
   In the hilly area there were 4 633 EC cases. The annual average rate was 117.16/100 000. From 1974 to 1996, the incidence rate changed from 165.71/100 000 to 96.33/100 000. Decreasing in number was 69.38/100 000. And declining rate was 41.86 percent. The result of trend test was x²=42.70, P<0.001. From Table 2 and Figure 3, we could see that from 1970s to 1980s the incidence rate of EC declined significantly, from 141.62/100 000 to 106.30/100 000 (U test: P<0.01). While comparing the incidence rate of EC of 1980s with that of the 1990s' it increased from 106.30/100 000 to 111.24/100 000 (U test: P<0.01).
   On level land there were 7 416 EC cases over 23 years. The annual average incidence rate was 122.30/100 000. According to Table 2 and Figure 3 we can see that during the past 23 years incidence rate of EC increased steadily. The result of trend test was x²=22.39, P<0.001. Comparing 1970s?incidence rate with 1990s? it increased from 116.14/100 000 to 130.09/100 000. The increasing in number was 13.95/100 000 and the increasing rate was 12.01 percent (U test: P<0.01). Comparing the incidence rate of 1980s with the 1990s? which increased from 120.50/100 000 to 130.09/100 000. The increasing in number was 9.59/100 000 and the increasing rate was 7.95 percent (U test: P<0.01). We could see that the incidence rate of EC in level land was increasing steatidly.
   There are 35 townships in Cixian County. In the 1970s, Linfeng township had the highest incidence rate of EC, which was 211.90/100 000. While Dudang township had the lowest, which was 66.57/100 000. The highest male incidence rate existed in Linfeng and the mumber was 298.02/100 000. While the highest female incidence rate existed in Guanglu township and the number was 164.49/100 000. The lowest incidence rate of EC of both male and female existed in Dudang, which was 79.71/100 000 and 51.51/100 000, respectively. The highest rate was 3.18 times as much as the lowest. Cixian County had two high risk areas, one was the mountainous area which was centered around Baitu district (132.23/100 000), the other was hilly area which was centered around Lintan district (128.64/100 000). Level land had the lowest incidence rate (116.01/100 000). After 1980s, the highest incidence rate existed in Guanglu township (232.04/100 000) and the lowest existed in Dudang (55.08/100 000). Linfeng had the highest male incidence rate of EC (285.37/100 000). While Cizhou town had the lowest (64.77/100 000). Guanglu had the highest female incidence rate of EC (214.28/100 000), while Dudang had the lowest (38.65/100 000). The highest EC incidence rate was 4.21 times as much as the lowest. There were 20 towmships whose incidence rate were higher than the average of county wide, forming a high-risk hilly area, which was centered around Lintan district. After 1990s, Ducun township had the highest EC incidence rate of 275.80/100 000. While Dudang township had the lowest whose number was 45.92/100 000. The highest male incidence rate existed in Ducun township which was 341.08/100 000. The highest female incidence rate existed in Guanglu which was 213.26/100 000. While the lowest male incidence rate lay in Wuhe township which number was 36.20/100 000. The lowest female incidence rate lay in Huangsha township which was 26.98/100 000. The highest EC incidence rate was 6.01 times as much as the lowest. There were 19 townships whose incidence rate were higher than the average of county wide, forming a high-risk level land area which was centered around Ducun township.
   In the 1970s, the high incidence area existed in mountainious areas which are centered around Baitu district and in hilly area which was centered around Lintan district. After 1990s, the incidence rate of EC in Cixian declined significantly. But the hilly area still had high incidence rate, decilining slowly. While the level land area had increasing trend, forming a high risk area which was centered around Ducun township.


Table 2 The geographic distribution of esophageal cancer in Cixian from 1974 to 1996 (1/100 000)

Figure 1 (PDF) Cixian EC Incidence Rate State from 1974 to 1996.
Figure 2 (PDF) Cixian EC Incidence Rate of Groups from 1983 to 1997.
Figure 3 (PDF) The Geography Distribution of Cixian EC Incidence from 1974 to 1996.

DISCUSSION
Cixian county is one of the highest EC incidence rate areas in China, as well as in the world. Since the early 1970s, cancer registry system has been established where began to collected the cancer incidence in Cixian county^[10-21]. Until now Cixian Cancer Registry have piled up the incidence data of more than twenty years. In this study we found the trend of the incidence rate of EC from 1974 to 1996 had declined after two decades. As for the possible causes of EC^[22-27], we advocate the inhabitant to take follow prevention: (1) To improve the quality of the drinking water condition; (2) To administer the farm products storage, not to eat the food with mold; (3) Eating more vegetables and fruits, changing the bad life style; (4) Conducting screening survey to find the carcinoma in situ or intramucosal carcinoma especially esophageal epithelium dysplasia (EED). EED is a precancerous lesion which can either develop further into a more severe stage or cancer, stay unchanged, or reverse back to normal again for a period of several years or even a decade^[28-34]. It is therefore very promising to detect patients with EED and treat the precancerous lesions before they transform into the irreversible malignant stage. There are several techniques and chemicals or nutrients that have been reported to be effective in blocking precancerous lesions from transforming into cancer^[35-38].
   The incidence rate among males declined significantly however, females remained steady. As to the geographic distribution, the incidence rate in mountainous area and hilly area showed a declining trend. The incidence rate of EC in level land area had increased. The reasons why these happened needs to be further studied and analysed.
   In conclusion, the trend of the EC incidence rate in Cixian county had declined after two decades, especially in mountainous areas. But compared to the other regions in the world Cixian county still had a high incidence rate of EC. By using a register the information can provide the scientific data for cancer prevention and control.

REFERENCES
1    Puttawibul P, Chanvitan A, Pornpatanarak C, Sangthong B. Esophageal carcinoma in Southern Thailand. J Med Assoc 
      Thai 2001; 84: 1-5
2    Adanja B, Gledovic Z, Pekmezovic T, Vlajinac H, Jarebinski M, Zivaljevic V, Pavlovic M. Mortality trends of malignant tumours 
      of digestive organs in Belgrade, Yugoslavia, 1975-1997. Dig Liver Dis 2000; 32: 386-391
3    Vega KJ, Jamal MM. Changing pattern of esophageal cancer incidence in New Mexico. Am J 
      Gastroenterol 2000; 95: 2352-2356
4    Martin IG. Gastro-oesophageal malignancy in New Zealand: 1995-97. N Z Med J 2002; 115: 64-67
5    Corley DA, Buffler PA. Oesophageal and gastric cardia adenocarcinomas: analysis of regional variation using the 
      Cancer Incidence in Five Continents database. Int J Epidemiol 2001; 30: 1415-1425
6    Aksel' EM, Davydov MI, Ushakova TI. Statistics of lung, stomach and esophageal cancer: status of oncological care, 
      morbidity and mortality. Vestn Ross Akad Med Nauk 2001; 9: 61-65
7    Kocher HM, Linklater K, Patel S, Ellul JP. Epidemiological study of oesophageal and gastric cancer in south-east England. 
      Br J Surg 2001; 88: 1249-1257
8    Hou J, Lin PZ, Chen ZF, Ding ZW, Li SS, Men FS, Guo LP, He YT, Qiao CY, Guo CL, Duan JP, Wen DG. Field Population-
      based blocking treatment of esophageal epithelia dysplasia. World J Gastroenterol 2002; 8: 418-422
9    World health organization. International classification of diseases of the international statistical classification of 
      diseases, injuries, and causes of death. Volume 1. Geneva 1977: 115-163
10  Zhang WH, Bailey-Wilson JE, Joan EB, Li WD, Wang XQ, Zhang CL, Mao XZ, Liu ZH, Zhou CN, Wu M. Segregation analysis 
      of esophageal cancer in a moderately high-incidence area of northern China. Am J Hum Genet 2000; 67: 110-119
11  Wang G, Hao C, Lai S. Endoscopic study on cancr of gastric cardia in the high incidence areas of China. Zhonghua 
      Zhongliu Zazhi 2002; 24: 381-383
12  Wu MY, Chen MH, Liang YR, Meng GZ, Yang HX, Zhuang CX. Experimental and clinicopathologic study on the 
      relationship between transcription factor Egr-1 and esophageal carcinoma. World J Gastroenterol 2001; 7: 490-495
13  Tan LJ, Jiang W, Zhang N, Zhang XR, Qiu DH. Fas/FasL expression of esophageal squamous cell carcinoma, dysplasia 
      tissues and normal mucosa. Shijie Huaren Xiaohua Zazhi 2001; 9: 15-19
14  Wu QM, Li SB, Wang Q, Wang DH, Li XB, Liu CZ. The expression of COX-2 in esophageal carcinoma and its relation 
      to clinicopathologic characteristic. Shijie Huaren Xiaohua Zazhi 2001; 9: 11-14
15  Liu HF, Liu WW, Fang DC. Study of the relationship between apoptosis and proliferation in gastric carcinoma and 
      its precancerous lesion. Shijie Huaren Xiaohua Zazhi 1999; 7: 649-651
16  Liu J, Su Q, Zhang W. Relationship between HPV-E6 P53 protein and esophageal squamous cell carcinoma. Shijie 
      Huaren Xiaohua Zazhi 2000; 8: 494-496
17  Lin J, Deng CS, Sun J, Zhou Y, Xiong P, Wang YP. Study on the genetic susceptibility of HLA-DQB1 alleles in esophageal 
      cancer of Hubei Chinese Hans. Shijie Huaren Xiaohua Zazhi 2000; 8: 965-968
18  Ma QF, Jiang H, Feng YQ, Wang XP, Zhou YA, Liu K, Jia ZL. Detection of human papillomavirus DNA in squamous cell 
      carcinoma of the esophagus. Shijie Huaren Xiaohua Zazhi 2000; 8: 1218-1224
19  Dong Z, Tang P, Li L, Wang G. The strategy for esophageal cancer control in high-risk areas of China. Jpn J Clin 
      Oncol 2002; 32 (Suppl): S10-12
20  Lu Z, Chen K, Guo M. Detection of HPV in human esophageal cancer in high-incidence area and its correlation with 
      p53 expression. Zhonghua Zhongliu Zazhi 2001; 23: 220-223
21  Chen KN, Xu GW. Diagnosis and treatment of esophageal cancer. Shijie Huaren Xiaohua Zazhi 2000; 8: 196-202
22  Wang AH, Sun CS, Li LS, Huang JY, Chen QS. Relationship of tobacco smoking CYP1A1 GSTM1 gene polymorphism 
      and esophageal cancer in Xi'an. World J Gastroenterol 2002; 8: 49-53
23  Su M, Lu SM, Tian DP, Zhao H, Li XY, Li DR, Zheng ZC. Relationship between ABO blood groups and carcinoma of 
      esophagus and cardia in Chaoshan inhabitants of China. World J Gastroenterol 2001; 7: 657-661
24  Yu GQ, Zhou Q, Ivan D, Gao SS, Zheng ZY, Zou JX, Li YX, Wang LD. Changes of p53 protein blood level in esophageal 
      cancer patients and normal subjects from a high incidence area in Henan, China. World J Gastroenterol 1998; 4: 365-366
25  Liu DW, Wanf BY, Zhou Y, Cui WF, Liu B, Zhou Q, Ying JY, Zheng S, Gao SS, Jin XX, Min FZ, Min NH, Hao ZZ, Yang CS, Min 
      BY, Jun QY. Endoscopic screening and determination of p53 and proliferating cell nuclear antigen in esophageal 
      multistage carcinogenesis: a comparative study between high- and low-risk populations in Henan, northern China. 
      Dis Esophagus 2002; 15: 80-84
26  Bachmann MO, Alderson D, Edwards D, Wotton S, Bedford C, Peters TJ, Harvey IM. Cohort study in South and West 
      England of the influence of specialization on the management and outcome of patients with oesophageal and gastric 
      cancers. Br J Surg 2002; 89: 914-922
27  Li T, Lu ZM, Chen KN, Guo M, Xing HP, Mei Q, Yang HH, Lechner JF, Ke Y. Human papillomavirus type 16 is an 
      important infectious factor in the high incidence of esophageal cancer in Anyang area of China. 
      Carcinogenesis 2001; 22: 929-934
28  Zuo L, Lin P, Qi F, Zhang L, Guo J, Liu J. Quantitive detection of multi-gene expressions and DNA content in the 
      precancerous cells of esophageal carcinoma Zhonghua Zhongliu Zazhi 2002; 24: 30-33
29  Griffin SM, Shaw IH, Dresner SM. Early complications after Ivor Lewis subtotal esophagectomy with two-
      field lymphadenectomy: risk factors and management. J Am Coll Surg 2002; 194: 285-297
30  Zhang J, Yan XJ, Yan QJ, Duan J, Hou Y, Su CZ. Cloning and expression of HPV16 L2 DNA from esophageal carcinoma in E. 
      coli. Shijie Huaren Xiaohua Zazhi 2001; 9: 273-278
31  Wei ZB, Wang LB, Tian BS, Wang JL, Sun XF, Wei JP, Liu N, Wang JH. Lugol's staining with p53 oncoproteins in detecting 
      early esophageal cancer. Shijie Huaren Xiaohua Zazhi 2001; 9: 495-498
32  Gu HP, Shang PZ, Su H, Li ZG. Association of CD15 antigen expression with cathepsin D in esophageal carcinoma tissues. 
      Shijie Huaren Xiaohua Zazhi 2000; 8: 259-261
33  Huang ZZ, Wu XY, Liang YR, Li QS, Shen J. Immunohistochemical study of esophageal basaloid squamous cell 
      carcinoma. Shijie Huaren Xiaohua Zazhi 2000; 8: 1097-1100
34  Liu J, Chen SL, Zhang W, Su Q. P21WAF1 gene expression with P53 mutation in esophageal carcinoma. Shijie Huaren 
      Xiaohua Zazhi 2000; 8: 1350-1353
35  Corley DA, Levin TR, Habel LA, Weiss NS, Buffler PA. Surveillance and survival in Barrett's adenocarcinomas: 
      a population-based study. Gastroenterology 2002; 122: 633-640
36  Anderson MR, Jankowski JA. The treatment, management and prevention of oesophageal cancer. Expert Opin Biol 
      Ther 2001; 1: 1017-1028
37  Wilson KS, Wilson AG, Dewar GJ. Curative treatment for esophageal cancer: Vancouver Island Cancer Centre experience 
      from 1993 to 1998. Can J Gastroenterol 2002; 16: 361-368
38  Zhang LJ, Chen KN, Xu GW, Xing HP, Shi XT. Congenital expression of mdr-1 gene in tissues of carcinoma and its relation 
      with pathomorphology and prognosis. World J Gastroenterol 1999;5: 53-56

Edited by Xia HHX