Abstract

Aim: Previous studies reported uncertain results on the relationship between non-steroidal anti-inflammatory drugs (NSAIDS) and colorectal adenomas. We performed a renewed meta-analysis investigating the efficacy and safety of NSAIDS in preventing the colorectal adenoma recurrence.

Methods: Data were retrieved from the Medicine, Cochrane library, and Emase database for randomized controlled trials (RCTs) up to August 2018. Data, bias, and quality were extracted and assessed by two independent reviewers. The random-effects model of Stata 12.0 was used for data analysis.

Results: This study included 13 RCTs. Aspirin and rofecoxib/celecoxib decreased the risk of recurrent adenomas with risk ratios (RR) of 0.83 (95% confidence interval (CI), 0.74-0.93, P=0.248, I2=26%, high-quality) and 0.63 (0.58-0.68, P=0.8, I2=0% high-quality), respectively. Subgroup analysis of 4- and 5-year follow-up for adenoma recurrence, respectively, 1.02 (0.84-1.24, P=0.127, I2=51.5% moderate-quality); and 1.15 (0.88-1.50, P=0.026, I2=72.6% very low-quality) and advanced adenomas, respectively, 0.88 (0.68-1.14, P=0.479, I2=0% high-quality); and 1.16 ( 0.82 -1.63, P=0.364, I2=1.0% high-quality).

Conclusion: NSAIDS decreased risks of recurrent adenomas, and results of 1- and 3-year follow-up were similar, but long-term 4- or 5-year followup showed that NSAIDS did not prevent the recurrence of adenomas. However, long-term rofecoxib and celecoxib were related to serious cardiovascular disorders and major bleeding; therefore, we suggest that the duration of rofecoxib and celecoxib prophylactic treatment for adenoma recurrence should not be more than 3 years.

Keywords: Non-steroidal anti-inflammatory drugs; Colorectal adenoma; Aspirin; Celecoxib; Rofecoxib; Meta-analysis

What Does This Paper Add to the Literature?

Our study first reports the pros and cons of long-term oral NSAIDs, and most benefits were achieved with short-term treatment of patients. Especially, we identified that the duration of administration should be limited to not more than 3 years, which paves the way for further research.