Cost-effective analysis of primary prophylaxis of variceal bleeding with beta-blockers vs. endoscopic surveillance in cirrhotic patients with small esophageal varices.

INTRODUCTION/OBJECTIVES: Variceal hemorrage is an important clinical problem in patients with cirrhosis and large esophageal varices, but is rather unusual while varices remain small. Therefore it is reasonable to play every effort to avoid or delay the progression of varices and, consequently, the occurrence of bleeding. In cirrhotic with small varices most international guidelines recommend nonselective beta-blockers for the prevention of first variceal hemorrhage in patients with increased risk of hemorrage and suggest the use of them in patients without criteria for increased risk, but available evidence is rather limited, and no economical evaluation of this strategy is available. AIMS & METHODS: We performed a stochastic analysis based on decision trees to compare the cost-effectiveness of beta-blockers therapy to endoscopic screening and surveillance followed by beta-blockers when large varices develop, for preventing variceal growth and bleeding in cirrhotic with small esophageal varices. Event probabilities were derived from the a placebo-controlled trial of beta-blockers in this clinical setting [1] and the hospital administrative database of patients included in that trial. Where not available, data have been integrated by information from literature. RESULTS: The event rate (considering progression of varices, bleeding or death) in the beta-blockers group was 24.09% (95% C.I. 14.89–33.29%), significantly lower than in the observational group, [60.00% (95% C.I. 48.91–71.08%)]. The mean cost (considered until the first event) associated with the early beta-blockers treatment was 823£ (95% C.I. 106–2036£), which turned out to be not significantly different from the surveillance approach, being equal to 799£ (95% C.I. 0–3498£). The costeffectiveness ratio (CER) with respect to this endpoint was thus equal to 50.26£ (95% C.I. −504.37–604.89£) per event avoided over the four years of follow-up. When we also included bleeding episode and death occurring in subjects with varices growth, mean cost associated with treatment was 1028£ (95% C.I. 122–2581£), whereas in absence of treatment was higher, equal to 1699£ (95% C.I. 171–4674£). CONCLUSION: In conclusion, beta-blocker therapy, being more effective and less expensive, appears to be dominant over the endoscopic surveillance approach for primary prophylaxis in cirrhotic patients with small esophageal varices.