ABSTRACT
Objective To investigate the risk factors of gastric cancer in Hanzhong area and establish a gastric cancer risk prediction model based on gastric function indexes. Methods A questionnaire survey was conducted on 227 patients with gastric and non-gastric cancer admitted to the Department of Gastroenterology of Hanzhong Central Hospital from January 2018 to January 2019. The diet, lifestyle habits, family history, and past history of the subjects were collected. Serological tests (including serum PGⅠ,Ⅱ, and G-17), gastroscopy and pathological examination of the samples were carried out. Multi-factor Logistic regression analysis was performed to analyze the high risk factors. The efficacy of serum gastric function for gastric cancer screening was assessed through ROC curve. Results There were no significant differences in the distribution of age (χ2=0.054, P=0.816), gender (χ2=3.823, P=0.051) and BMI (χ2=5.569, P=0.135) between the gastric cancer and non-gastric cancer groups, suggesting that the clinical data between the two groups was comparable. Multivariate logistic regression analysis showed that smoked food intake (P=0.031, 95% CI: 1.150 ~ 17.545) and hyperlipidemia (P=0.039, 95% CI:0.216~0.960) were high risk factors for gastric precancerous lesions, other factors did not show significant high risk (P>0.05). The results of ROC curve showed that the diagnostic efficacy of combined PG I, PG II, and G-17 was significantly higher than that of PG I, PGⅡ, or G-17 alone (P<0.05). Conclusion The risk factors of gastric cancer in Hanzhong area mainly included the intake of smoked food and hyperlipidemia. Patients can be advised to pay attention to diet during clinical treatment. Based on gastric function indexes PG, PGⅡ and G-17, the diagnosis of gastric precancerous lesions had a good prediction. The combined diagnosis is more effective, which provided a theoretical basis for the early diagnosis of clinical gastric cancer.