Cost-Effectiveness of TIPS vs Large-Volume Paracentesis in Refractory Ascites: Results of a Markov Model Incorporating Individual Patient-Level Meta-Analysis and Nationally Representative Cost Data
Refractory ascites is a common complication of decompensated cirrhosis which causes severe limitations to patient quality of life as well as extensive health care costs. The main therapeutic options are serial large volume paracentesis (LVP) and transjugular intrahepatic portosystemic shunt (TIPS) creation, each of which have their own set of risks and benefits. In this study, a decisional Markov model was developed to estimate payer cost and quality-adjusted life-years (QALY) with both of these treatment options using survival and symptomatology estimates from multiple prospective randomized clinical trials. Procedural and hospital cost values were derived from a national claims database and pharmaceutical costs from the 2015 Medicare Part D Prescriber Public Use Files. These costs were then inflated to 2017 US dollars with annual inflation rates for medical expenditures. Outcomes were measured in quality adjusted life years gained and an incremental cost-effectiveness ratio of each therapeutic strategy. LVP resulted in 1.72 QALYs gained at $41,391 and TIPS resulted in 2.76 QALYs gained at $100,538. The incremental cost-effectiveness ratio of TIPS vs LVP was $57,003/QALY. The cost effectiveness acceptability curve showed a 62% probability of TIPS being acceptable relative to LVP at a willingness to pay (WTP) ratio of $100,000/QALY. A threshold of 2-3 times the per capita annual income has been argued by the World Health Organization. This would imply a WTP ratio of $110,000-$160,000/QALY for the US. Using these parameters, TIPS would be considered a more cost-effective therapy for refractory ascites in decompensated cirrhosis for the United States.
Figure 1: Markov model for different treatment strategies and associated outcomes. Various health utility values were given depending on the strength of an individual’s preferences for specific health outcomes
Figure 2: Probability sensitivity analysis of cost effectiveness acceptability curve for willingness to pay ranging $0-$100,000. At $50,000-$55,000 the probability of cost effectiveness begins to trend in favor of TIPS. A 62% probability of TIPS being more cost effective is demonstrated at $100,000.
The global prevalence of cirrhosis has been estimated to effect up to 9.5% of the general population. Cirrhosis causes a significant mortality and morbidity which reduce patient’s quality of life and cost the United States approximately $1.4 billion annually (excluding hepatitis C virus treatment). With repeated large volume paracentesis and TIPS creations as the mainstays for therapeutic options, it is an important and relevant question as to which treatment is more appropriate. TIPS has been shown to achieve higher rates of ascites control but a greater incidence of encephalopathy than LVP. However, the initial creation is high risk and costly upfront. Whereas serial LVP costs and risks are low initially, they will continue to accrue as well as potentially have a less positive impact on quality of life. The study suggests TIPS should be considered more cost effective when high monetary value is placed on health improvements, while LVP may be more effective in countries with lower levels of health care resources.
Click here for abstract
Kwan, S. W., Allison, S. K., Gold, L. S., & Shin, D. S. (2018). Cost-Effectiveness of Transjugular Intrahepatic Portosystemic Shunt versus Large-Volume Paracentesis in Refractory Ascites: Results of a Markov Model Incorporating Individual Patient-Level Meta-Analysis and Nationally Representative Cost Data. Journal of Vascular and Interventional Radiology, 29(12), 1705-1712.
Bradley Unruh, MD PGY-3
Department of Radiology
Wake Forest Baptist Medical Center