| Sumario: | Background Previous studies have shown that Percutaneous coronary intervention (PCI) is a better treatment than thrombolysis for patients with acute myocardial infarction admitted to an invasive centre. However, the economic consequences of PCI as the primary strategy has not yet been assessed. One should also consider different strategies when there is a long distance from local hospital to the centre offering PCI. The aim of the study was to explore the long-term cost-effectiveness of PCI. Methods Based on a systematic review of clinical trials, we developed a state-transition model that follows patients from when they develop an acute ST-elevation infarction until they die. The model encompassed events (re-infarction, stroke, medical interventions, etc.) and health states (sequalae stroke, heart failure, angina, well, and death). Transition probabilities and utilisation of health care were taken from the systematic review, population statistics, and expert judgement. Unit costs were based on market prices, DRG charges, and reimbursement schedules. One-way sensitivity analyses and Monte-Carlo-simulations were undertaken to explore uncertainty. Results For a 65-year old man living close to a PCI unit, life expectancy was 8.3 years with PCI and 7.6 with thrombolysis, while lifetime costs were 162 000 NOK and 234 000 NOK., respectively. For patients who would need helicopter ambulance to get to a PCI unit, the expected costs were NOK 192 000 for PCI (all costs undiscounted). In sensitivity analyses PCI entailed lower costs and greater health benefits for any realistic change in the parameters. Conclusion PCI seems to entail greater health benefits at lower costs than thrombolysis whether the patients live close to PCI hospital or not. The conclusion is unchanged after sensitivity analysis. However, a change to primary PCI as a strategy will have consequences for personnel and organisation of the health care system. These secondary consequences have not been considered in these analyses. In clinical practice combined strategies are actual, that is thrombolysis in ambulance or in local hospital before sending the patient to an invasive centre for PCI. For the present, the clinical data supporting an analysis of combined strategies are still too weak to be analysed by our model. When more robust data are available, it is possible to modify the model to include combinations.
|
|---|
