Candidate for the Lee B. Lusted Student Prize Competition
Purpose: As prior cost-effectiveness analyses of primary PCI assumed locally available, existing cardiac catheterization laboratories (CCL's) in urban centers, the most cost-effective revascularization strategy for non-urban populations is not known. Moreover, the rate of rescue angioplasty for failed thrombolysis varies considerably in clinical practice, but it is unclear what impact this would have on the choice of revascularization strategy in non-urban communities.
Method: We compared 5 revascularization strategies for a hypothetical non-urban population (TNK thrombolysis, transfer to an existing CCL by either ambulance, helicopter, or fixed-wing air medical transport, and constructing a new CCL locally). Patient risk factor profiles and strategy-specific event rates were derived from the medical literature (all-cause mortality, ischemic stroke, hemorrhagic stroke, non-cerebral major bleeding, reinfarction, and ischemia-driven revascularization). The mortality benefit of PCI over thrombolysis was modeled as a decremental function of the PCI-related time delay. As it is not precisely established how the rate of rescue PCI correlates with outcomes, we conservatively limited the impact of varying the rate of rescue PCI to thrombolysis cost estimates only without any negative effect on efficacy.
Result: Under base case assumptions (rescue=26%), ambulance transport was the lowest cost strategy and TNK thrombolysis, as well as helicopter or fixed-wing transport, were dominated strategies. As would be expected, the probability of rescue PCI correlated positively with the estimated cost of the TNK strategy and this effect was more pronounced at greater inter-hospital distances. Under base case distance and capacity assumptions, however, ambulance transport was the least expensive strategy even when the probability of rescue PCI was 0% (i.e. no failure of TNK therapy). The TNK, helicopter and fixed-wing strategies were dominated irrespective of the rate of rescue (Figure). Moreover, after adjusting the utilisation rate of the new CCL down to 5% of capacity, TNK was still a dominated strategy and helicopter transport became the most cost-effective strategy at a willingness-to-pay (WTP) of 60,000 CAD.
Conclusion: Under base case geographical assumptions, the rate of rescue PCI after thrombolysis is unlikely to affect the choice of revascularization strategy with WTP's typical of developed countries. It is possible however that this parameter may be important in resource-poor settings, where WTP might be lower or certain transfer strategies are not available, or over greater transfer distances.
