Effects of thrombolysis during out-of-hospital cardiopulmonary resuscitation

Abstract

In this post hoc analysis, we assessed effects of thrombolysis during out-of-hospital cardiopulmonary resuscitation. The original study was designed as a double-blinded, prospective, multicenter, randomized, controlled clinical trial. In this report, 1,219 patients were randomized, but 33 patients were excluded due to missing study drug codes. Thus, 1,186 patients were analyzed based on receipt (n = 99) versus nonreceipt (n = 1,087) of thrombolysis; the primary end point was hospital admission, and the secondary end point was hospital discharge. Patients who received thrombolysis versus those who did not were significantly younger (mean +/- SD 62.7 +/- 13.3 vs 66.5 +/- 14.3 years of age, p = 0.01) and more likely to have had an acute myocardial infarction (75.3% vs 54.6%, p < 0.01) or pulmonary embolism (20.2% vs 12.0%, p = 0.03) as the suspected underlying cause for cardiac arrest. In patients who underwent thrombolysis versus those who did not, cardiac arrest was more often witnessed (86.9% vs 77.5%, p = 0.03), initial ventricular fibrillation was more likely (59.6% vs 38.0%, p < 0.01), and a short estimated interval (0 to 5 minutes) between collapse and initiation of basic life support was more likely (51.3% vs 29.2%, p < 0.01). In patients who received thrombolysis, sodium bicarbonate (45.5% vs 33.0%, p = 0.01), lidocaine (32.3% vs 18.1%, p < 0.01), and amiodarone (30.3% vs 12.2%, p < 0.01) were administered significantly more often. Hospital admission rates were significantly higher in patients who underwent thrombolysis than in patients who did not (45.5% vs 32.7%, p = 0.01), and there was a trend to higher hospital discharge rates (14.1% vs 9.5%, p = 0.14). In patients who had suspected myocardial infarction, hospital admission and discharge rates were significantly higher in patients who underwent thrombolysis than in patients who did not. In logistic regression models after adjusting for confounding variables (e.g., age, initial electrocardiographic rhythm, and initiation of basic life support), hospital admission and discharge rates did not differ significantly. In conclusion, even when being employed in patients with a potentially better chance to survive, thrombolysis in patients with cardiac arrest resulted in an increased hospital admission but not discharge rate in this post hoc analysis.