OBJECTIVE: To assess the economic implications of implantation with a cardiac resynchronization device (CRT) versus optimum pharmacologic treatment (OPT) for NYHA class III or IV heart failure. METHODS: A discrete event simulation of the course of heart failure was used to compare 1000 pairs of identical patients — one receiving CRT, the other OPT — for two years, in terms of hospitalizations for heart failure, device-related complications, NYHA class, and QOL (mortality assumed equal). All input rates were derived by specific analyses of the data obtained in the Multicenter InSync Randomized Clinical Evaluation (MIRACLE). Changes in NYHA class were translated to QOL via the corresponding distributions of Minnesota Living with Heart Failure scores and the related utilities. Direct medical costs to the NHS, reported in 2002 British Pounds Sterling (£), discounted at 6%/yr, were obtained from HRG codes E18-19. Benefits were discounted at 1.5%. Extensive multivariate sensitivity analyses were done. RESULTS: In these patients with mean age of 64 yrs, 91% in class III, 23% dead at two yrs, CRT reduced hospitalization for heart failure by 42%, leading to total costs of £3500 per patient vs. £3000 with OPT. Based on 100 replications, mean improvement of 0.16 QALY (SD 0.009) is achieved with CRT at mean net cost of £526 (SD £167) per patient, a mean cost-effectiveness ratio of £3379 /QALY. Extensive sensitivity analyses revealed the greatest cost/QALY variability when the length of stay for heart failure was varied ±25% (£562-£6354). CONCLUSION: Despite the cost of implantation, cardiac resynchronization therapy decreases hospitalizations and increases QOL sufficiently to be cost-effective in treating advanced heart failure.