Several studies have assessed the risk of cervical abnormalities including dysplasia and cancer, and more recently urinary cancer in inflammatory bowel disease (IBD) patients. However the results of these studies were controversial. The aim of our study was to perform a meta-analysis of existing clinical trials to compare the risk of developing genitourinary cancers, or abnormalities in patients with IBD compared to normal population.Methods:
A systematic search of PubMed, CINAHL, Scopus, Cochrane database of systematic reviews, and clinical trials.gov registry was performed in September/2016. Two independent reviewers systematically identified trials that compared the effect of IBD versus normal population in developing cervical abnormalities including dysplasia and cancer, beside risk of developing bladder cancer. A meta-analysis was performed using a fixed effects model to assess the primary outcome (development of cervical abnormalities), and secondary outcomes (development of cervical dysplasia, cervical cancer, and urinary cancer). Subgroup analysis was performed to assess the risk of cervical abnormalities in Crohn's disease and Ulcerative Colitis separately. Review Manager 5.3 software program was utilized for statistical analysis.Results:
seven studies met the inclusion criteria. Six studies looked at developing cervical abnormalities, and 2 studies evaluated the risk of bladder cancer. A total of 2,023,946 patients were included in the meta-analysis assessing the cervical abnormalities. Our meta-analysis demonstrated a significant increased risk of cervical abnormalities in general in IBD patients (OR: 2.409; 95% CI, 1.455–3.988, P-value: 0.001), and cervical dysplasia in IBD patients (OR: 2.607, 95% CI, 1.247–5.452, P = 0.011). However, There was no statistically significant difference in the incidence of cervical cancer (OR 1.690; 95% CI, 0.852–3.353, P = 0.133), or bladder cancer (OR: 2.282, 95% CI, 0.519–10.033, P = 0.275). Subgroup analysis showed a significantly increased risk of cervical abnormalities in CD (OR: 2.319, 95% CI, 1.105–4.869, P = 0.026), and in UC patients (OR: 1.798, 95% CI, 1.202–2.690, P = 0.004). A significant heterogeneity was observed, most likely to the significant difference in the follow up time between studies, and random effects model used in analysis to decrease the heterogeneity.Conclusions:
Our results indicated an increased risk of cervical abnormalities in IBD, CD, and UC patients. It also showed an increased risk of cervical dysplasia in IBD patients, but no increased risk of cervical cancers, or bladder cancer in IBD patients compared to normal population. A limitation to our study is the significant heterogeneity that we tried to overcome by using a random effects model.