Systematic review and meta-analysis based on the composition of risk factors of chronic atrophic gastritis under gastroscopy detection.

Background: This meta-analysis explored the diagnostic value of gastroscopy in the diagnosis of patients with chronic atrophic gastritis (CAG).

Methods: The Chinese and English medical databases were searched using a combination of the following search terms: "gastroscope", "chronic atrophic gastritis", "endoscope", and "diagnose". The Review Manager 5.3 software was used for meta-analysis.

Results: A total of 10 articles were included. The funnel charts showed that the circles of some articles were basically symmetrical with the midline, suggesting that there was no publication bias and the conclusions are credible. There were 6 randomized controlled trials (RCTs) analyzing clinical symptoms [Chi2 =5.33; degree of freedom (df) =5; I2=6%; P=0.38] and tested the overall heterogeneity [Z=7.94; odds ratio (OR) =9.42; 95% confidence interval (CI): 5.42 to 16.39; P<0.01]. A total of 10 articles analyzed the efficacy of gastroscopy (Tau2 =2.96; Chi2 =70.00; df =9; I2=87%; P<0.01) and tested the overall heterogeneity (Z=4.38; OR =13.44; 95% CI: 4.20 to 42.98; P<0.01). A total of 8 articles analyzed the pathological efficacy (Chi2 =11.93; df =7; I2=41%; P=0.10) and tested the overall heterogeneity (Z=10.59; OR =13.36; 95% CI: 8.27 to 21.59; P<0.01). The combined sensitivity and the specificity were 69.34% (95% CI: 0.526 to 0.784) and 96.45% (95% CI: 0.785 to 0.952), respectively. The area under the receiver operating characteristic (ROC) curve (AUC) of the combined sensitivity and combined specificity was 0.759 and 0.846, respectively.

Discussion: The results of the GRADE quality assessment revealed that although the sample size of this meta-analysis was small, the research was of sufficient medium and high quality, and the included experiments were of relatively good quality.