Abstract
Aims:
Epidemiologic studies have examined the possible associations between diabetes mellitus
(DM) and gastric cancer (GC), but the results are inconclusive. We conducted a meta-analysis
to assess the evidence regarding the associations between diabetes and incidences
of, or mortality from, gastric cancer.
Methods:
PubMed, Embase and Web of Science were searched up to Oct 20, 2011. We identified
studies that included effects estimates with 95% confidence intervals (CIs) of the
associations between GC and diabetes. Summary RRs for the GC incidence and mortality
were calculated using random-effects model; subgroup analyses were also performed.
Heterogeneity among studies was examined using Q and I2 statistics.
Results:
A total of 7 case-control and 18 cohort studies met the inclusion criteria. The summary
RR showed a slightly statistical link between history of DM and GC incidence (RR=1.11,
95% CI: 1.00–1.24, p=0.045, I2=79.5%). In the subgroup analyses, a positive association was noted among the studies
conducted in Asia (summary RR=1.19, 95% CI: 1.07–1.32, I2=29.8%). Additionally, slight associations between DM and GC were observed by pooling
the data of type 2 DM, cohort studies and the studies controlling more confounders,
respectively. Furthermore, mortality from GC with diabetes was increased compared
with individuals without diabetes (summary RR=1.29, 95% CI: 1.04–1.59). No publication
bias was found.
Conclusion:
Individuals with diabetes have an increased risk of developing gastric cancer, and
are positively associated with gastric cancer mortality. Large better-designed cohort
studies are needed to verify this conclusion.
Key words
diabetes mellitus - gastric neoplasm - epidemiology - meta-analysis