Antibiotic exposure and the risk of colorectal adenoma and carcinoma: a systematic review and meta-analysis of observational studies

Aim The Incidence of colorectal cancer (CRC) is increasing, and evidence suggests that maladaptation of the bowel microbiome may be associated with colorectal carcinogenesis. Consumption of antibiotics may cause imbalance of the bowel microbiome but research assessing an association between antibiotic exposure and CRC is inconsistent. The aim of this systematic review and meta-analysis was to appraise and synthesize the available evidence. Method The MEDLINE, EMBASE and CINAHL databases were searched for published observational studies. We included eight studies of 3 408 312 patients. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) for the odds of CRC following antibiotic exposure were estimated. Sensitivity analyses were performed according to exposure definition, study design and risk of bias. Results A weak association between antibiotic exposure and CRC was demonstrated when exposure was assessed cumulatively by the number of prescriptions (OR 1.204, 95% CI 1.097-1.322, P < 0.001) or duration of antibiotic exposure (OR 1.168, 95% CI 1.087-1.256, P < 0.001). Antibiotic exposure assessed as a binary variable demonstrated no association with CRC. Conclusion The findings suggest a weak association between cumulative antibiotic consumption and risk of CRC but no causal conclusions can be made. Limitations include the heterogeneity and quality of the available research, particularly with regard to measurement of antibiotic exposure.