Optimal Management of Pancreatic Cysts - A Cost-Effectiveness Analysis

Abstract

The management of asymptomatic pancreatic cysts, increasingly identified as incidental findings on abdominal imaging tests is a dilemma. A small proportion of mucinous cysts may progress to invasive cancer and an aggressive strategy of surgical resection of all cysts is often pursued with the attendant risk of surgery. Objective: We performed a decision analysis comparing the cost-effectiveness of different strategies for managing solitary, asymptomatic pancreatic cysts. Methods: Incremental cost-effectiveness ratio (ICER) of three primary strategies were examined in a Markov model with a societal perspective. Strategy I was a wait and watch approach with follow-up MRI every 3 years. Development of specific alarm symptoms or imaging features suggesting malignant transformation during surveillance would trigger further evaluation with EUS-FNA and appropriate management, including resection. In Strategy II, an initial EUS-FNA was performed and patients with increased cyst fluid CEA or positive cytology underwent resection. A. Strategy III was surgical resection of all cysts without prior EUS evaluation. Results: In a 60 year old person Strategy I yielded highest quality adjusted life years (QALY) with an acceptable ICER (Table) in both the baseline and a second order Monte Carlo analysis. Strategy II dominated strategy III, and was both cheaper and yielded more QALYs. Test characteristics of EUS-FNA, surgical risk score and the annual mortality rate of malignant mucinous cystic neoplasm were the three important threshold variables that influenced the optimal strategy. Conclusion: A conservative strategy is generally more cost-effective than surgical resection in managing incidental asymptomatic pancreatic cysts. An initial EUS-FNA based strategy is favored in selecting patients who may benefit from resection.