Much has been said about the relevance of diet and socioeconomic factors in their association with peptic ulcer. As early as the seventh century AD, practitioners believed in the healing properties of milk when taken by patients whose digestive symptoms as then described suggested the likely possibility of peptic ulcer.(1) In 1959 Shull1 wrote that diets containing milk, cream and its products had become prominent in most dietary prescriptions at the time to reduce symptoms of peptic ulcer disease as it was widely believed that these substances perhaps inhibited gastric secretory function and/or neutralised acid gastric juice. Over time the role of diet in the pathogenesis of peptic ulcer has been disputed and various theories construed. However, epidemiologic studies(2,3,4,5,6) on the consumption of alcohol, coffee or tea, salt or salty foods, milk and fruit and their association with ulcer causation have produced inconsistent results.  Symptoms due to peptic ulcer may modulate the diet, therefore case-control or cross-sectional studies of this disease have implicit limitations. However, the literature evaluating the effect of diet on ulcer recurrence is inappreciable. Despite the fact that the main cause of duodenal ulcer incidence and recurrence is Helicobacter pylori, a majority of Helicobacter pylori-infected people do not develop peptic ulcers, which suggests that other factors may have an ancillary role.(7) In 1978, a clinical trial was conducted in India comparing a rice diet with an unrefined wheat diet among ulcer patients over a 5-year period.(8) Results showed that 81% of patients on the rice diet had recurrence of ulcer, whilst only 14% of patients on the unrefined wheat diet had recurrence of ulcer. A year later, Tovey’s paper was published which depicted higher incidence of peptic ulcer in rice eating areas and lower incidence in wheat eating areas in India. (9) The role of socioeconomic factors and the body mass index (BMI) index in peptic ulcer healing following eradication therapy for H.pylori has also not been accounted for. Chan et al(10) in 1997 undertook a study in 287 ulcer patients to investigate the role of smoking in ulcer recurrence after eradication of H. pylori without maintenance acid-suppression therapy. They concluded that cigarette smoking did not increase the recurrence of peptic ulcers after eradication therapy. A study by Watanabe et al(11) summarised an association for psychological stress with relapse, but none for sex, smoking habit, drinking habit, and past history of ulcer. In Thailand, Thong-Ngam et al(12) analysed the association of age, sex, eating habit, water drinking, number of children, and treatment regimens with ulcer recurrence and found none. Suzuki et al(13) conducted a meta-analysis of 22 published studies (5538 patients), including studies which provided information on eradication failure according to smoking status and demonstrated an increased treatment failure rate for H. pylori eradication in smokers. A study by Queiroz et al(14) evaluated factors associated with ulcer treatment failure in a developing country (Brazil), and found no association for smoking and alcohol consumption.  Baena et al(15) showed that daily alcohol consumption appeared to have an additive effect in a one-week eradication therapy consisting of omeprazole, clarithromycin and amoxicillin. What we have then is a potpourri of suggestions and opinions from diverse locations and no form of concord or unanimity. Indian studies evaluating the role of dietary factors in peptic ulcer disease were carried out  before the discovery of the role of H.pylori as a causative factor and their conclusions  require scrutiny in the context of H. pylori eradication therapy. There is a discernable absence of such studies from India and this study was undertaken to fill in this gap and understand better the dietary and socio-economic factors associated with duodenal ulcer healing and eradication of H.pylori.

PATIENTS AND METHODS
A cross-sectional study of socioeconomic factors, dietary habits, and anthropometric measurements of 67 consecutive duodenal ulcer patients attending the Ulcer Clinic at the Department of Gastroenterology and Human Nutrition, All India Institute of Medical Sciences, New Delhi, India, was carried out from January to July 2003. The cohort from which these consecutive patients were recruited consisted of duodenal ulcer patients with documented H.pylori infection who had received eradication therapy. The patients received standard triple eradication therapy consisting of lansoprazole 30 mg BD, amoxicillin 1 g BD and tinidazole 500 mg BD for 1 week. Follow-up endoscopy and testing for H.pylori were done 4 weeks following completion of the eradication therapy to assess ulcer healing and successful eradication. The patient was diagnosed with H.pylori infection if any two of the three diagnostic tests (rapid urease test, C14 urea breath test and haemotoxylin and eosin stain of antral biopsy) were positive. H.pylori was considered absent if all three tests were negative. Consecutive patients from this cohort were included and on enrollment the socioeconomic factors, personal habits, dietary habits and body mass index were recorded. Dietary habits were recorded using the food frequency questionnaire. The study was in accordance with the ethical principles laid down by the Declaration of Helisinki.

VARIABLES STUDIED
Socio-demographic factors: The factors considered under this category were age, occupation (according to the National Classification of Occupations); educational status; religion; type of family (nuclear/joint); total number of family members; total family income (monthly); per capita income (total family income/number of members) and residence (rural/urban).(16)

Personal habits: A detailed history of alcohol consumption, along with the age at which alcohol intake was started, and amount and frequency of alcohol consumption were enquired into.  A similar history was obtained for smoking habits.  Alcohol consumption was defined as consumption of more than 30 g (i.e. roughly equivalent to 230ml wine, 680ml beer and 86ml of 80-proof beverage) of ethanol in a week.

Dietary habits: Regular dietary intake was assessed using the food frequency questionnaire. Foods were grouped into seven sections (rice, wheat, pulses, vegetables, fruits, and non-vegetarian items). The frequency of intake of these food groups was recorded under seven categories: daily, alternate day, thrice a week, twice a week, once a week, once a fortnight or once a month. Food items consumed between thrice weekly to daily were included under one header. Type (saturated or unsaturated) and total amount of fat used for cooking was also noted. Information on amount and frequency of tea or coffee consumed was also obtained. All the information was collected by a trained dietician.

Anthropometry: Body mass index (BMI) calculated from height and weight [weight (kg)/height² (m2)] was computed as a measurement of the nutritional status.

Outcome variables were H.pylori eradication and duodenal ulcer healing at 4 weeks after completion of eradication therapy.

STATISTICAL ANALYSIS
Data were expressed as mean ± standard deviation (SD). Median (range) was used wherever variables did not have a normal distribution. Student’s t test was used for continuous variables. The difference in proportion was compared using the chi-square test. A p value less than 0.05 was taken as significant. SPSS version 10.0 statistical software (SPSS, Chicago, IL, USA) was used for analysis.

RESULTS
67 patients with mean age 39.9 +13.6 years (60 male, 7 female) were included in the study. Of these, 51 patients had healed duodenal ulcers and 16 patients had active duodenal ulcers despite therapy. In 31 patients H.pylori infection had been successfully eradicated and in 36 patients H.pylori infection was still present. The patients were divided into two groups based on ulcer healing and H.pylori eradication status.

H.pylori status post therapy
Table 1 shows the H.pylori status post therapy with reference to the main variables studied. Except for BMI, which was significantly higher in patients in whom H.pylori had been eradicated, all other variables studied were comparable in both groups.

There was no difference between the groups (Group A1: H.pylori eradicated, Group B1: H.pylori not eradicated) with regard to occupation, education, religion and place of residence i.e. urban or rural (not shown in the table). The frequency and number of cups of tea consumed did not affect H.pylori eradication. Only 7 patients, each, in both groups A1 and B1 were vegetarians. Duration of illness did not affect the eradication rate. 47.2% (17 of 36) of Group B1 consumed alcohol, as opposed to 29% (9 of 31) from Group A1, and the median duration of alcohol consumption was twice (10 vs. 5 years) in the former as compared with the latter, however, neither comparison was significant. The median per capita income was higher in Group A1 vs. B1, but not significantly so (p=0.101, Rs. 1000 vs. 645).

The dietary intake of major food groups was not significantly different in patients who had achieved H.pylori eradication as compared to patients with persistence of H.pylori (Table 2). A majority of patients (i.e. 58 of 67) used mustard oil as a cooking medium. Mean visible oil intake of the whole group was 24.5 + 14.1 g (5.6-100).

Duodenal ulcer status post therapy
Table 3 represents the duodenal ulcer status post therapy with reference to variables studied. Here too there was no difference between the two groups with regard to occupation, education, religion and place of residence (urban/rural). Only 18.7% and 23.4% patients were vegetarian in Groups A2 and B2 respectively. The frequency and number of cups of tea or coffee consumed did not affect duodenal ulcer healing.

The median per capita income was significantly higher in Group A2 (p=0.015, Rs.1000.00 vs. 612.50).

Intake of various food groups as assessed by the food frequency questionnaire had no influence on the healing status of duodenal ulcer. Major food group intake was similar in patients who had successful healing of duodenal ulcer as compared to patients with non healing ulcers (Table 4).

 

Table 1:  Patient characteristics: H.pylori status post therapy

 

DISCUSSION
Our study revealed that diet does not play a significant role in ulcer healing or H.pylori eradication; a number of studies do indicate however that the role of diet in the aetiology of ulcer formation is substantial. In a majority of studies that we will now discuss the central theme is dietary risk factors and possible association with ulcer occurrence; a few studies that pertain to ulcer eradication, recurrence and healing have already been mentioned. However, it may be presumed that to a large extent, factors which cause duodenal ulcer may also in part be responsible for its recurrence and resistance to treatment and we take up the discussion in this light.

Tovey(9) observed a higher incidence of peptic ulcer in rice eating areas and lower incidence in the wheat eating areas of India. Rydning et al(6) in 1982 proposed that the wide variation in the incidence of peptic ulcer in different native African populations may have been related to the apparent protective effect of high intake of fresh fruit and vegetables, while diets high in refined starch seemed detrimental. In China too Tovey(17) observed a reduced prevalence of duodenal ulcer in the wheat growing areas; however, there was not the marked north-south difference between the wheat and rice eating areas as was reported from India. The reason for this may have been that the white wheat flour used in making steamed bread is refined, whereas the wheat used in India to make ‘chappatis’ is unrefined. There is evidence from animal peptic ulcer models that wheat bran contains a potent protective factor against peptic ulceration(18,19,20).

Table 3: Patient characteristics : Duodenal ulcer status post therapy

 

 

In our study, wheat was consumed by all patients. Also in a number of cases both wheat and rice were consumed and hence the relation becomes further difficult to explain. Jalan(21) et al in Calcutta showed that acid output after a ‘chapatti’ meal (unrefined wheat) to that after a rice meal were similar, although the intra-gastric buffer content was higher after the former. One possible explanation was afforded by Malhotra(22,23,24,25,26) who suggested that the increased mastication required by the Punjabi ‘chapatti’ diet resulted in a greater volume of saliva with higher bicarbonate and mucus content than that resulting from the rice diets. This hypothesis also fits in with the low incidence of peptic ulcer reported in the Gond tribe where the diet of ‘rotis’ requires prolonged mastication.(9) The possible association of refined carbohydrate diets and sugar consumption with the incidence of peptic ulcer was postulated by Cleave and is supported by a considerable amount of geographical and historical data. Cleave elucidated the importance of the extra buffer content of unrefined foods, which has been confirmed experimentally(12).

In a study by Misciagna et al(27), research papers published in English from 1966 through October 1999 present in Medline, involving human subjects, and with duodenal ulcer as the outcome were reviewed. They concluded that soluble fibre from fruits and vegetables appeared to be protective against duodenal ulcer and refined sugar was a risk factor. The role of fibre and essential fatty acids in the treatment and prevention of recurrence of duodenal ulcers, they said, however, was uncertain. Aldoori et al(28) in Harvard documented 138 newly diagnosed cases of duodenal ulcer in a 6-year follow-up. They too noted that higher consumption of fruits and vegetables was associated with lower risk of duodenal ulcer following adjustment for age, body mass index, smoking, and use of aspirin or other nonsteroidal anti-inflammatory drugs. Total dietary fibre intake was inversely associated with the risk of duodenal ulcer. Likewise in Africa there are regions of high and low incidence, the incidence being low in millet and unrefined maize eating areas (29). Here the frequency of duodenal ulcer is rising in urban populations with their transition in diet and other aspects of lifestyle. The traditional diet of rural blacks was high in cereal, bean and spinach, with meat and dairy products seldom eaten. It was low in fat (supplying 10-15% energy) and sugar (5-8% energy) and high in fibre (30-40g or more daily).

Presently, in urban areas, fat intake has risen (supplying 10% or more of energy), whereas fibre intake has fallen to that of the white population, perhaps even lower, of 10g or less daily.(30)
Vitamin A was inversely related to the risk of duodenal ulcer, and this relation was statistically significant after adjustment for other risk factors.(31) None of the epidemiological studies on the relationship between diet and duodenal ulcer disease thus far have taken into account the effect of H.pylori eradication. In 2007, however, Salih et al(32) showed that in Turkish patients duodenal ulcer had significant association with a combination of factors, namely aspirin/smoking/NSAIDs (P=0.0259), aspirin/alcohol (P=0.0002) and aspirin/smoking (P=0.0233), also in the presence of H pylori. While DU in the absence of H pylori had significant association with smoking/alcohol/ NSAIDs (P=0.0013), aspirin/NSAIDs (P=0.0334), aspirin/alcohol (P=0.0360).

Hollander and Tarnawski(33) encouraged interest in the role of polyunsaturated fatty acids when they linked the decline in duodenal ulcer to the rise in dietary consumption of polyunsaturated fatty acids. They correlated the waning incidence of peptic ulcer disease with a 200% increase in the dietary availability of essential fatty acids during the same time span. We studied neither the role of fat as a broad group, nor the part played by saturated or unsaturated fats in ulcer healing. However as per the local custom, cooking mainly employed the use of mustard or vegetable oil or ghee and there was no difference in the amount of visible cooking oil between the groups.

Comments and studies on the influence of physical activity on the risk of duodenal ulcer are at variance and range from no effect to a beneficial effect. A case-controlled study by Katschinski et al(34) showed that the proportion of patients with duodenal ulcer who had highly active jobs was 38%. In a prospective study, Cheng et al(35) found that both moderately active and active men had a 45% to 60% reduced risk of duodenal ulcer. Cross-sectional data from the British Regional Heart Study indicated that body mass index and the proportion of obese people tends to fall with increasing exercise, so the effect of body mass index on the risk of duodenal ulcer would likely be similar to the effect of physical activity.(36) As with diet, these studies of the role of BMI and physical activity pertained to ulcer pathogenesis and not eradication, healing or relapse. We showed that a higher BMI was significantly associated with eradication of H.pylori.

Tovey(9) in 1979 hypothesised that duodenal ulcer was appearing in urbanised areas, perhaps due to their more westernised way of living and the fast pace of life. Thus reports from Delhi and Bombay showed a greater incidence of duodenal ulcer in the skilled and semi-skilled workers and those with white collar posts. Most reports from Madras, however, showed that the lower income groups were still principally affected, but the survey by Malhotra(24) of railway workers reported no difference between the social classes. In a study by Everhart et al in 1998, lower socio-economic status, as represented by low family income and lower educational attainment was strongly associated with incident ulcers. For example, persons who had not attended high school had 4.7 times the incidence of persons who had attended graduate school. Incidence by family income showed a clear separation at $20,000, where people with lower family income had about twice the incidence of ulcers as people with family income greater than $20,000.(37) Among the other factors studied we were unable to demonstrate any relationship between age, sex, alcohol consumption, smoking habit and eradication rate.  Everhart et al(37) demonstrated a strong relation for smoking and age-standardised prevalence of chronic active ulcers: 1.8% among non-smokers, 3% among smokers of less than a pack per day, 3.9% among smokers of a pack per day and 5.3% among smokers of more than a pack a day. Khuroo et al,(38) in a randomly selected population of 2763 persons perceived that peptic ulcer was not related to socio-economic status. Epidemiological data from India however suggest that peptic ulcer is more common in the poor(39,40). Raghavan39 found that the highest incidence (56.5%) of peptic ulcer was among the semi-skilled workers and the lowest (2.5%) in professional and managerial groups. Figures from the United Kingdom and the United States suggest that the duodenal ulcer tends to be more common in the poor than in the rich.37 Yodfat et al(41) in 1972 in a study from Israel demonstrated that ulcer prevalence rose with a rise in income and educational level. Per capita income may be one of the factors affecting the natural history of duodenal ulcer; even in our study the eradication of peptic ulcer in those with a median per capita income of Rs.1000 was significantly greater (p=0.015) than in those with a median per capita income of Rs.612.50.

Kato et al(42) (1992) found that the risk of both gastric and duodenal ulcers progressively increased with increasing pack-years of cigarette smoking. None of the variables of physical activity, body mass index, occupation, education and place of birth was associated with either type of ulcer. In a study by Jain et al(43) in 1999 from Delhi comprising 16 patients with duodenal ulcer, significantly greater number of smokers (80%) and alcoholics (58%) in the male population of the DU group was seen when compared to the controls (49% smokers and 15% alcoholics). The sample size in this case however was too small to make this data reliable. In a cohort study in Norway by Johnsen et al, to simultaneously analyse the risk factors associated with peptic ulcer in 328 patients, age, cigarette smoking, first-degree relatives with peptic ulcer, and low educational level were shared risk factors for peptic ulcer in both men and women. Dietary intake declines with smoking.3, 6, 7, 8 Smokers’ diets differ in other important respects, including high salt and low fibre intake19, so caution is required before implicating any particular dietary component. The increased risks associated with low educational background indicate that social strains, comprising lifestyle and diet habits, are part of the multifactorial aetiology of peptic ulcer.

The ulcer healing rate in our group of patients was disproportionately higher than the H.pylori eradication rate. This is not surprising as poor eradication rates of H.pylori in North India have been substantiated in previous studies conducted at our centre .(44,45) Moreover this study was aimed at evaluating the influence of dietary and sociodemographic factors on the healing of duodenal ulcers and eradication of H.pylori . This study was not constructed to estimate host and microbial factors associated with poor eradication rates of H.pylori in India.   

To conclude, we found that the body mass index and the per capita income were significantly associated with the eradication of H.pylori and duodenal ulcer healing, respectively, while dietary factors had no impact. Further epidemiological studies that control for Helicobacter pylori are imperative to establish predictive factors.

REFERENCES

1.     Harrison J Shull. Diet in the management of peptic ulcer. JAMA. 1959:170:1068–1071

2.     Friedman GD, Siegelaub AB, Seltzer CC. Cigarettes, alcohol, coffee and peptic ulcer. N Engl J Med. 1974;290:469–73.

3.     Paffenbarger RS Jr, Wing AL, Hyde RT. Chronic disease in former college students; 13. Early precursors of peptic ulcer. Am J Epidemiol. 1974;100:307–15.

4.     Kato I, Tominaga S, Ito Y, Kobayashi S, Yoshii Y, Matsuura A, et al. [Comparative case-control analysis of gastric and duodenal ulcers]. Nippon Koshu Eisei Zasshi. 1990;37:919–25.

5.     Chuong JJ, Fisher RL, Chuong RL, Spiro HM. Duodenal ulcer. Incidence, risk factors, and predictive value of plasma pepsinogen. Dig Dis Sci. 1986;31:1178–84.

6.     Rydning A, Berstad A, Aadland E, Odegaard B. Prophylactic effect of dietary fibre in duodenal ulcer disease. Lancet. 1982;2:736–9.

7.     Taylor DN, Blaser MJ. The epidemiology of Helicobacter pylori infection. Epidemiol Rev. 1991;13:42–59.

8.     Malhotra SL. A comparison of unrefined wheat and rice diets in the management of duodenal ulcer. Postgrad Med J. 1978;54:6–9

9.     Tovey F. Peptic ulcer in India and Bangladesh. Gut. 1979;20:329–47.

10.   Than FK, Sung JJ, Lee YT, Leung WK, Chan LY, Yung MY, et al. Does smoking predispose to peptic ulcer relapse after eradication of Helicobacter pylori? Am J Gastroenterol. 1997;92:442–5.

11.   Watanabe T, Higuchi K, Tominaga K, Fujiwara Y, Arakawa T. [Peptic ulcer recurrence after successful eradication of Helicobacter pylori—clinical characteristics and management] [Article in Japanese] Nippon Rinsho. 2004;62:495–8.

12.   Thong-Ngam D, Mahachai V, Kullavanijaya P. Incidence of Helicobacter pylori recurrent infection and associated factors in Thailand. J Med Assoc Thai. 2007;90:1406–10.

13.   Suzuki T, Matsuo K, Ito H, Sawaki A, Hirose K, Wakai K, et al. Smoking increases the treatment failure for Helicobacter pylori eradication. Am J Med. 2006;119:217–24.

14.   Queiroz DM, Dani R, Silva LD, Santos A, Moreira LS, Rocha GA, et al. Factors associated with treatment failure of Helicobacter pylori infection in a developing country. J Clin Gastroenterol. 2002;35:315–20.

15.   aena JM, López C, Hidalgo A, Rams F, Jiménez S, García M, et al. Relation between alcohol consumption and the success of Helicobacter pylori eradication therapy using omeprazole, clarithromycin and amoxicillin for 1 week. Eur J Gastroenterol Hepatol. 2002;14:291–6.

16.   Mishra D, Singh HP. Kuppuswamy’s socioeconomic scale – A Revision. Ind J Pediatrics 2003;273–4.

17.   Tovey FI. Duodenal ulcer in China. J Gastroenterol Hepatol. 1992;7:427–31.

18.   Jayaraj AP, Tovey FI, Clark CG. Possible dietary protective factors in relation to the distribution of duodenal ulcer in India and Bangladesh. Gut. 1980;21:1068–76.

19.   Tovey FI, Jayaraj AP. Peptic ulcer in India. Gut. 1990;31:123–4.

20.   Tovey FI, Jayaraj AP. Duodenal ulcer and carbohydrate. Gut. 1991;32:339.

21.   Jalan KN, Mahalanabis D, Maitra TK, Agarwal SK. Gastric acid secretion rate and buffer content of the stomach after a rice- and a wheat-based meal in normal subjects and patients with duodenal ulcer. Gut. 1979;20:389–93.

22.   Malhotra SL, Saigal ON, Mody GD.  Role of saliva in the aetiology of peptic ulcer. Br Med J. 1965;1:1220–2.

23.   Malhotra SL. A study of the effect of saliva on the concentration of mucin in gastric juice and its possible relationship to the aetiology of peptic ulcer. Gut. 1967;8:548–55.

24.   Malhotra SL. Peptic ulcer in India and its aetiology. Gut. 1964;5:412–6.

25.   Malhotra SL. Epidemiological study of peptic ulcer in the south of India. Observations from Madras on the changing incidence of peptic ulcer, with special reference to causation. Gut. 1967;8:180–8.

26.   Malhotra SL, Majumdar CT, Bardoloi PC. Peptic ulcer in Assam. Gut. 1964;5:355–8.

27.   Misciagna G, Cisternino AM, Freudenheim J. Diet and duodenal ulcer. Dig Liver Dis. 2000;32:468–72.

28.   Aldoori WH, Giovannucci EL, Stampfer MJ, Rimm EB, Wing AL, Willett WC. Prospective study of diet and the risk of duodenal ulcer in men. Am J Epidemiol. 1997;145:42–50.

29.   Tovey FI, Tunstall M. Duodenal ulcer in black populations in Africa south of the Sahara. Gut. 1975;16:564–76.

30.   Segal I, Walker AR, Vorster HH. Dietary factors associated with duodenal ulcer. Gut. 1991;32:456.

31.   Ryan-Harshman M, Aldoori W. How diet and lifestyle affect duodenal ulcers. Review of the evidence. Can Fam Physician. 2004;50:727–32.

32.   Salih BA, Abasiyanik MF, Bayyurt N, Sander E. H pylori infection and other risk factors associated with peptic ulcers in Turkish patients: a retrospective study. World J Gastroenterol. 2007;13:3245–8

33.   Hollander D, Tarnawski A. Dietary essential fatty acids and the decline in peptic ulcer disease—a hypothesis. Gut. 1986;27:239–42.

34.   Katschinski BD, Logan RF, Edmond M, Langman MJ. Duodenal ulcer and refined carbohydrate intake: a case-control study assessing dieary fibre and refined sugar intake. Gut. 1990;31:993–6.

35.   Cheng Y, Macera C, avis DR, Blair SN. Physical activity and peptic ulcers. West J Med. 2000;173:101–7

36.   Weatherall R, Shaper AG. Overweight and obesity in middle-aged British men. Eur J Clin Nutr. 1988;42:221–31

37.   Everhart JE, Byrd-Holt D, Sonnenberg A. Incidence and risk factors for self-reported peptic ulcer disease in the United States. Am J Epidemiol. 1998;147:529–36.

38.   Khuroo MS, Mahajan R, Zargar SA, Javid G, Munshi S. Prevalence of peptic ulcer in India: an endoscopic and epidemiological study in urban Kashmir. Gut. 1989;30:930–4.

39.   Raghavan P. Epidemiology and clinical behavior of peptic ulcer in Bombay, India. Gastroenterology. 1962;42:130–43.

40.   Konstam PG. Peptic ulcer in India. Indian J Med Sci. 1959;13:486–92.

41.   Yodfat Y.A population study of peptic ulcer. Its relation to various ethnic and socioeconomic factors. Isr J Med Sci. 1972;8:1680–4.

42.   Kato I, Nomura AM, Stemmermann GN, Chyou PH. A prospective study of gastric and duodenal ulcer and its relation to smoking, alcohol, and diet. Am J Epidemiol. 1992;135:521–30.

43.   Jain A, Buddhiraja S, Khurana B, Singhal R, Nair D, Arora P. Risk factors for duodenal ulcer in north India. Trop Gastroenterol. 1999;20:36–9.

44.   Bhatia V, Ahuja V, Das B, Bal C, Sharma MP. Use of imidazole-based eradication regimens for Helicobacter pylori should be abandoned in North India regardless of in vitro antibiotic sensitivity. J Gastroenterol Hepatol. 2004;19:619–25.

45.   Ahuja V, Sharma MP. High recurrence rate of Helicobacter pylori infection in developing countries. Gastroenterology. 2002 ;123:653–4.