Thrombolytic Therapy with Streptokinase and Tissue Plasminogen Activator in a Patient with Suspected Acute Myocardial Infarction: A Decision Analysis

Two commonly used thrombolytic agents are streptokinase (SK) and tissue plasminogen activator (t-PA), which have different impacts on the incidence of mortality and thrombolysis-related acute intracranial hemorrhage. A decision-analytic model was developed to compare the use of SK and t-PA in the treatment of a patient with suspected acute myocardial infarction (AMI). The outcome was health-related quality of life as quantified in a measure of utility from the patient's point of view. The model included three outcome states: death, nonfatal yet disabling stroke, and survival with no disabling stroke. The utility for disabling stroke was determined relative to the reference states of no disabling stroke (1.00) and death (0.00) by means of the time trade-off estimation technique. Probabilities were derived from the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Artery trial, which revealed that although administering t-PA results in a lower percentage of deaths compared to SK, it may lead to a higher percentage of strokes. A decision tree was constructed to model the options and outcomes. The tree was analyzed by standard decision analytic techniques using SMLTREE software, and the stability of the results was examined as values of parameters were varied systematically in a sensitivity analysis. In the baseline analysis, SK yielded 0.9235, whereas t-PA yielded 0.9329. The sensitivity analysis revealed that if the probability of a disabling stroke attributable to t-PA were greater than 2.08%, SK would yield the higher expected utility. This threshold value, however, was much greater than the probability established in major trials. The administration of t-PA leads to a slightly better outcome than does the administration of SK in a patient with suspected AMI.