Diet and Risk of Gastric Cancer: An Umbrella Review

Abstract

Several dietary exposures have been associated with gastric cancer (GC), but the associations are often heterogenous and may be afflicted by inherent biases. In the context of an Umbrella Review (UR), we provide an overview and a critical evaluation of the strength and quality, and evidence classification of the associations of diet-related exposures in relation to the risk of GC. We searched PubMed and Scopus for eligible meta-analyses of observational studies published in English from inception to 12 December 2021, and for any identified association, we applied robust epidemiological validity evaluation criteria and individual study quality assessment using AMSTAR. We screened 3846 titles/abstracts and assessed 501 full articles for eligibility, of which 49 were included in the analysis, investigating 147 unique exposures in relation to GC, cardia (GCC) or non-cardia (GNCC) cancer. Supported by suggestive evidence, positive associations were found comparing the highest vs. lowest categories for: heavy (>42 g/day) alcohol consumption (Relative Risk (RR) = 1.42, 95% Confidence Interval (CI): 1.20−1.67), salted fish consumption (RR = 1.56, 95% CI:1.30−1.87) and waist circumference (RR = 1.48, 95% CI:1.24−1.78) and an inverse association for the healthy lifestyle index (RR = 0.60, 95% CI:0.48−0.74) in relation to GC. Additionally, a positive association was found comparing obese individuals (Body Mass Index (BMI) ≥ 30) to normal-weight individuals (BMI: 18.5−25) (RR = 1.82, 95% CI:1.32−2.49) in relation to GCC. Most of the meta-analyses were of medium-to-high quality (median items: 7.0, interquartile range: 6−9). Maintaining a normal body weight and adopting healthy dietary choices, in particular, limiting the consumption of salt-preserved foods and alcohol, can reduce the risk of gastric cancer.

Keywords: diet; gastric cancer; nutrition; risk factors; stomach cancer; umbrella review.

Figure 1

Flow chart of the study selection process (left) and overview of the dietary exposures that are evaluated in the present umbrella review, by category (right).

Figure 2

Summary of the significant associations (p-value < 0.05) between the dietary exposures and gastric and gastric non-cardia cancer, and the classification of the evidence using epidemiological validity criteria. The forest plot shows the random effect meta-analysis relative risks and the 95% confidence intervals. The heatmap shows the epidemiological validity criteria that are fulfilled per association (blue box indicates ‘yes’; red box indicates ‘no’). Bubble column to the right of the plot summarizes the study quality based on AMSTAR. Associations that were based on >2 studies are plotted.

Figure 3

Quality assessment of the included studies using AMSTAR. In the bubble plot (left), each bubble represents an association that is colored based on the classification output and the size is analogous to the number of citations per year that the study received. The bar plot (right) provides a summary of the assessment, per item, across the 48 studies.