Abstract
Background: Optimal management of asymptomatic pancreatic cystic neoplasm is not known.
Objective: In a decision analysis, the cost-effectiveness of different strategies for managing solitary, asymptomatic pancreatic cystic neoplasm were compared.
Intervention: Three strategies were examined in a Markov model with a third-party-payer perspective. In strategy I, the natural history of the lesion was followed without any specific intervention. In strategy II, an aggressive surgical approach was considered in that all patients were considered for resection. In strategy III, an initial EUS-guided FNA with cyst fluid analysis was performed for risk stratification, and patients with mucinous cysts were considered for resection. Transitional probabilities, discontinued cost, and utility values to estimate quality-adjusted life years were obtained from published information. An operability risk score based on patients age, comorbidity, and size and location of the cyst was developed to estimate the probability of surgical resection.
Results: In the baseline analysis, strategy III yielded the highest quality-adjusted life years with an acceptable incremental cost-effectiveness ratio. In a Monte Carlo analysis, the relative risk of patients developing unresectable pancreatic cancer was decreased in strategy III compared to the other strategies. Although threshold analyses identified few important parameters influencing the conclusion of the analysis, operability risk score was the critical determinant of the optimal management strategy.
Limitations: Indirect costs were not considered in this analysis.
Conclusions: For asymptomatic patients with incidental solitary pancreatic cystic neoplasm, a blanket policy of surgical resection for all patients cannot be justified. A strategy based on risk stratification of malignant potential by EUS-guided FNA and cyst fluid analysis is the most-effective strategy. (Gastrointest Endosc 2009;70-690-9.)