This case-control study was conducted between May 2010 and June 2012 in the Cancer Research Institute, Imam Khomeini Complex, Tehran University of Medical Sciences, Tehran, Iran. Patients diagnosed with GC, referred from all-around Iran to Imam Khomeini hospital, took part. Patients who were pathologically diagnosed with GC in less than six months and with no previous history of any cancer were identified as eligible for the current study. Case employment was based on the convenience sampling method. A Non-random sampling method was applied for enrolling 276 controls among healthy people. The Control group has been chiefly picked out among relatives of patients in other hospital wards. The response rate was 95% in cases and 70% in controls.
Dietary assessment: Dietary patterns were determined by applying a 146-item, semi-quantitative, dish-based Diet History Questionnaire (DHQ). Comprehensive information about this questionnaire, including its design, development, and validity, will be explained subsequently. DHQ was completed by performing a face-to-face interview. Controls were asked to report their last 12 months’ dietary intake in the form of Iranian home scales like a spoon, plate, bowl, and scoop. Reported amounts were converted to Grams/day by household equipment. Patients with gastric cancer were asked to recall their intake before the cancer symptoms showed off. Total daily energy and macro and micronutrients were calculated using food consumption tables provided by the United State Department of Agriculture (USDA). Dietary Acid Load as NEAP (daily dietary Net Endogenous Acid Load production) was estimated by dividing the amount of protein intake into potassium through the following formula: NEAP = (54.5*g Protein/Potassium meq)-10. Tertile cut-off points of these dietary scores were obtained based on scores in control subjects to avoid potential bias that might arise due to patient changes in dietary intakes.
In terms of the validity of the DHQ, we found a high correlation coefficient between nutrient estimations by DHQ and those from multiple 24-h recalls . This was the case for energy, protein, carbohydrate, protein, fiber, vitamin A, carotene, niacin, folate, vitamin B12, biotin, vitamin C, sodium, magnesium, iron, zinc, and selenium (r > 0.5 for all).
Assessment of gastric cancer
Gastric cancer was diagnosed based on a gastroscopic or surgical biopsy that an expert pathologist did. Patients were included if they were diagnosed with stomach cancer histologically, considering the definition of gastric cancer provided by the second edition of the International Classification of Diseases for Oncology (ICDO code c16). Only patients who had been diagnosed during one preceding year to the date of the interview were eligible for participating in the present study.
Assessment of covariates
A trained bachelor of health sciences conducted an organized upstanding interview. Demographic and general information, including gender, marital status, education level, residential places, and smoking and drinking habits, were derived from the questionnaire. Since body weight status is influenced by gastric cancer, the actual weight of patients was not considered in this study. Usual weight and height were collected by asking participants about their weight before cancer diagnosis individually—body mass index (BMI (was calculated by dividing weight in kilograms by height in meters squared. Patients were categorized as current smokers and non-smokers according to their reported smoking habits last year. Ten milliliters of venous blood samples were taken from all attendants at both fasting and non-fasting status to examine for H. pylori infection. ELISA kits were applied to measure serum samples for IGF antibodies. Serologic examinations were carried out by experienced technicians who were not aware of neither the study setting nor the participants’ case/control status. The H. pylori antibody test was repeated in a random collection of serums to prove validity. The existence of antibodies and seropositivity of y more than 0,87 was considered positive.
Dietary acid load was categorized into tertiles, and participants’ dietary intake was compared across these tertiles using one-way ANOVA. Categorical variables were compared using the chi-square test, and one-way ANOVA was used to compare continuous variables. Logistic regression was used to define the association between DAL and GC risk. In first model, we adjusted for age(continuous), sex(male/female) and energy intake(continuous). further adjustments were done for education(illiterate/literate), marital status(married/single), H. pylori infection(positive/negative), alcohol intake(continuous) and smoking status(smoker/nonsmoker) in the second model. Additional control was applied in the third model for BMI.