Observational studies have reported inconsistent results on the association between fruit and vegetable intake and the risk of pancreatic cancer. We carried out a meta-analysis of epidemiological studies to summarize available evidence. We searched PubMed, Scopus, and ISI Web of Science databases for relevant studies published until the end of January 2015. Fixed-effects and random-effects models were used to estimate the summary relative risks (RRs) and 95% confidence intervals (CIs) for the associations between fruit and vegetable intake and the risk of pancreatic cancer. A total of 15 case–control studies, eight prospective studies, and one pooled analysis fulfilled the inclusion criteria. The summary RR for the highest versus the lowest intake was 0.73 (95% CI=0.53–1.00) for fruit and vegetables, 0.73 (95% CI=0.63–0.84) for fruit, and 0.76 (95% CI=0.69–0.83) for vegetables, with significant heterogeneities (I2=70.5, 55.7, and 43.0%, respectively). Inverse associations were observed in the stratified analysis by study design, although the results of prospective studies showed borderline significance, with corresponding RR=0.90 (95% CI=0.77–1.05) for fruit and vegetable intake, 0.93 (95% CI=0.83–1.03) for fruit intake, and 0.89 (95% CI=0.80–1.00) for vegetable intake. Besides, significant inverse associations were observed in the majority of other subgroup analyses by study quality, geographic location, exposure assessment method, and adjustment for potential confounders. Findings from the present meta-analysis support that fruit and vegetable intake is associated inversely with the risk of pancreatic cancer. However, study design may play a key role in the observed magnitude of the aforementioned association. Future well-designed prospective studies are warranted to confirm these findings.
aDepartment of Clinical Epidemiology
bDepartment of Obstetrics and Gynecology, Shengjing Hospital of China Medical University, Shenyang, China
cCenter for Clinical and Translational Science, Mayo Clinic, Rochester, Minnesota, USA
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (www.eurjcancerprev.com).
Correspondence to Qi-Jun Wu, MD, PhD, Department of Clinical Epidemiology, Shengjing Hospital of China Medical University, No. 36, San Hao Street, Shenyang, Liaoning 110004, China Tel: +86 24 96615 13648; fax: +86 24 96615 13648; e-mail: email@example.com
Received December 11, 2014
Accepted May 14, 2015