Surveillance colonoscopy or chemoprevention with COX-2 inhibitors in average-risk post-polypectomy patients: a decision analysis

Objectives: Clinical trials are currently underway evaluating the efficacy of COX-2 inhibitors in decreasing the incidence of adenomas and colorectal carcinoma in 'average' risk individuals. Aim: To use decision analysis to compare the cost-effectiveness of celecoxib to surveillance colonoscopy in 'average' risk patients who had undergone prior adenoma resection. Methods: A model of the natural history of adenomas after endoscopic polypectomy was constructed using probabilities from the literature. Cost estimates were obtained from available Medicare reimbursement rates and supplemented by the literature. Three strategies were evaluated: (i) no surveillance; (ii) colonoscopic surveillance; and (iii) celecoxib chemoprevention. We compared total costs and performed cost-effectiveness analysis between these strategies. The outcome measures were years of life saved and 'high-grade' adenoma prevented. Sensitivity analyses were performed on selected variables. Results: Our base-case analysis assumed a 50% risk reduction in the incidence of adenomas among patients using cerecoxib. No surveillance was associated with a cost of $1014 per patient, and colonoscopic surveillance with a cost of $1572 per patient, whereas celecoxib use was associated with a total cost of $11 503. Ten years after the index colonoscopy, 15% of patients in the no surveillance strategy developed 'high-grade' lesions compared to 13% of patients in the colonoscopic surveillance group and 6% in the celecoxib group. There was a small gain in years of life saved (0.006) favouring celecoxib over colonoscopic surveillance. The incremental cost-effectiveness ratio of celecoxib vs. colonoscopy was $141 871 per 'high-grade' adenoma prevented and $1 715 199 per year of life saved. The most important variables in determining the cost-effectiveness of celecoxib were its cost and its efficacy. Conclusion: Chemoprevention with COX-2 inhibitors in 'average-risk' postpolypectomy patients is a more expensive strategy compared to colonoscopic surveillance.