Noninvasive detection of reperfusion after thrombolysis based on serum creatine kinase MB changes and clinical variables
Coronary artery patency after thrombolytic therapy has important prognostic implications for survival after acute myocardial infarction. The ability to noninvasively identify patients early after thrombolysis may therefore allow other strategies, such as adjunctive therapy or rescue angioplasty, to be used to restore patency of the infarct-related artery. This study examined the use of a rapid creatine kinase (CK)-MB assay in conjunction with selected clinical variables for noninvasive detection of reperfusion after thrombolysis. Patients were enrolled in a study evaluating accelerated plasminogen activator dose regimens with patency assessments by first angiographic injection during acute angiography at a median and interquartile range (25th and 75th percentiles) 142 (96, 195) minutes after starting thrombolytic therapy. Serum CK-MB samples measured by a rapid dual monoclonal antibody assay were obtained in 207 patients before (baseline) and 30 minutes, 90 minutes, and 3 hours after starting thrombolytic therapy. In 109 patients a CK-MB sample was obtained within 10 minutes of acute angiography (angio sample). At acute angiography the infarct-related artery was patent (Thrombolysis In Myocardial Infarction trial grade 2 to 3 flow) in 71%. Baseline CK-MB values were similar in patients with and without later reperfusion at acute angiography: 3 (0,8) ng/ml and 0 (0,4) ng/ml, respectively. At acute angiography, patients with successful reperfusion had higher CK-MB values [46 (20,138) ng/ml] compared with patients with persistent occlusion of the infarct-related artery [8 (3,63) ng/ml; (p = 0.002). The changes in CK-MB values in association with reperfusion status were best defined by the slope of CK-MB (angio minus baseline value divided by time between samples) or the ratio of CK-MB (angio divided by baseline value); χ2 12.9, p = 0.0003, for both slope and ratio. Selected clinical markers were evaluated to construct a combined CK-MB and clinical model for noninvasive detection of reperfusion. By logistic regression a combined model was generated that yielded a highly significant association (χ2 26.3, concordance probability index 0.85, p < 0.00001) with patency of the infarct-related artery. In this combined model CK-MB slope contributed the most information (χ2 15.7, p < 0.0001), followed by chest pain before acute angiography (χ2 6.5, p = 0.01), and time from onset of symptoms to thrombolytic therapy (χ2 5.6, p = 0.02). In conclusion, rapid noninvasive detection of reperfusion is possible by combining a fast CK-MB analysis with clinical variables obtained at the bedside. By using a logistic regression model for noninvasive detection of reperfusion, it is possible to ascribe probabilities of persistent occluded infarct-related artery. This approach may be helpful in decision-making for acute angiography after thrombolysis. © 1993.
Ohman, EM; Christenson, RH; Califf, RM; George, BS; Samaha, JK; Kereiakes, DJ; Worley, SJ; Wall, TC; Berrios, E; Sigmon, KN; Lee, K; Topol, EJ
Volume / Issue
Start / End Page
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)