The risk of thrombolysis in association with cardiopulmonary resuscitation: no reason to withhold this causal and effective therapy

Thrombolysis is a causal and effective therapy in patients suffering from massive pulmonary embolism or acute myocardial infarction [1]. In 50% to more than 70% of patients suffering from cardiac arrest, one of these two diseases is the underlying cause of deterioration, and about 10–20% of all patients suffering from acute myocardial infarction require cardiopulmonary resuscitation (CPR) [2-5]. Nevertheless, in the past and because of the fear of severe bleeding complications, thrombolytic therapy during and in the early phase after CPR has been contraindicated. However, this contraindication, which has repeatedly been underlined by various authors and medical societies [6-8], was never based on sufficient scientific evidence [9, 10]. The recommendations of the American College of Cardiology and the American Heart Association [7] refer to the Fibrinolytic Therapy Trialists' study [1] as the basis for this contraindication. However, this meta-analysis did not identify a subgroup of patients receiving CPR and lists the overall incidence of bleeding only [1]. Therefore, the bleeding risk of thrombolysis during and/or after CPR needs to be defined more precisely and discussed in the light of the potential benefits. Increasing clinical experience and data from recent clinical studies suggest that thrombolysis during ongoing CPR can contribute to haemodynamic stabilization and survival in patients with massive pulmonary embolism and acute myocardial infarction [9-14]. Following the administration of thrombolytic agents during CPR, some patients have been stabilized even after more than 90 min of CPR [9, 10]. Besides the specific causal action of thrombolytic agents at the site of pulmonary emboli and coronary thrombosis, experimental data indicate that thrombolysis during and after CPR can improve microcirculatory reperfusion, which may be most important in the brain [9, 15]. In accordance with these data, marked activation of blood coagulation without adequate activation of endogenous fibrinolysis has been demonstrated during reperfusion after cardiac arrest [9, 16]. Therefore, massive coagulation activation with subsequent fibrin formation is responsible for microcirculatory reperfusion disorders, and thrombolytic therapy may be indicated. In addition, early clinical experience with thrombolysis after CPR in nearly 200 patients suffering from acute myocardial infarction revealed no markedly increased risk of severe bleeding complications [17-20]. Preliminary data from two open studies, however, demonstrate that the survival rate in patients with acute myocardial infarction and cardiac arrest who received thrombolytic therapy in the early phase after CPR was significantly higher than in those patients who did not receive reperfusion therapy [21, 22]. Of 14 980 patients with acute myocardial infarction who were enrolled in ‘The 60-Minute Myocardial Infarction Project’, a subgroup of 681 patients received CPR. In-hospital mortality was significantly lower in patients who were treated with thrombolysis after CPR (n = 308) than in those in whom thrombolytic therapy was withheld (n = 373; 48% vs. 62%, respectively) [21]. Comparable data from the MITI Registry reveal that, of 12 984 patients suffering from an acute myocardial infarction, 599 were resuscitated [22]. Thrombolysis or percutaneous transluminal coronary angioplasty (PTCA) after CPR was performed in 53 and 59 of these patients, respectively, but coronary revascularization was withheld in 487 patients. The in-hospital mortality in these subgroups was 13.5% and 24.1%, but 58.2% in the group without thrombolysis or PTCA, respectively [22]. In a recently published subgroup analysis of the Spanish multicentre registry (ARIAM; n = 13 704 patients), two haemorrhagic strokes and two haematomas that required transfusion occurred in the subgroup of 303 patients with acute myocardial infarction and CPR before hospital admission, 67 of whom received thrombolysis after CPR and hospital admission [23]. None of these bleeding events, however, was CPR-related or fatal [23]. Most interestingly, the overall outcome in this survey was significantly improved with thrombolytic intervention after cardiac arrest: mortality in patients receiving thrombolysis after CPR was 18% (vs. 46% in patients not receiving thrombolysis), the incidence of cardiogenic shock was 14% (vs. 39% in patients without thrombolysis) and 9% of treated patients developed hypoxic brain damage (vs. 40% of patients not treated with thrombolysis) [23]. In a retrospective study of 68 patients resuscitated after cardiac arrest caused by acute myocardial infarction, Voipio et al. performed thrombolytic therapy after return of spontaneous circulation [24]. Of these, 36 patients were discharged from the hospital, 21 without any severe neurological disability. In one case, fatal intracranial haemorrhage occurred and there was one patient with generalized bleeding. In addition, two gastrointestinal bleedings and one case of epistaxis were described. However, none of these bleeding complications was clearly attributed to CPR and no patient received a blood transfusion [24]. Most recently, Schreiber et al. reported on 157 patients with witnessed out-of-hospital cardiac arrest following acute myocardial infarction [25]. Thrombolytic therapy was applied in 42 patients (27%). With thrombolytic therapy, good functional neurological recovery was achieved significantly more frequently (50%; vs. 69% in patients not treated with thrombolysis). After controlling for age, prehospital dosage of adrenaline and the duration of cardiac arrest, these authors found a nonsignificant trend towards good neurological recovery when thrombolytic therapy was given. Bleeding complications were seen in five patients but, again, none was clearly related to cardiocompressions [25]. Therefore, and although thrombolysis per se can cause severe and potentially fatal haemorrhages, there is no evidence that severe bleeding events occur far more often when thrombolysis is combined with cardiocompressions, and most of the bleeding complications can be treated effectively. Moreover, CPR-associated bleeding complications do not appear to outweigh the potential benefits that are associated with thrombolysis, particularly in this specific setting [9, 10]. However, considering that not enough controlled clinical trials have been conducted yet, further data and data from large randomized multicentre studies are needed to definitely answer the question of the relationship between safety and efficacy of this promising treatment option during and after CPR [26]. In the present issue of the Journal of Internal Medicine, Kurkciyan and coworkers report on the data from a retrospective cohort study in 265 cardiac arrest patients [27]. They have carefully evaluated the complication rate of systemic thrombolysis using recombinant tissue-type plasminogen activator (rt-PA) after out-of-hospital cardiac arrest in patients with acute myocardial infarction. Major haemorrhage was observed in 13 of 132 (10%) patients who received thrombolysis and in seven of 133 (5%) patients without thrombolytic intervention. Therefore, the risk of bleeding was increased if thrombolytics were used, but this increase was not significant. Moreover, there was no clear association between the duration of CPR [thrombolysis group, 12 (4–24) min; group without thrombolysis, 17 (10–31) min] and bleeding complications. Therefore, the authors conclude that in the light of possible benefits on outcome, thrombolytic treatment should not be withheld in carefully selected patients after cardiac arrest. This further supports the view that the risk of thrombolysis in association with CPR is no reason to withhold this causal and effective therapy in most of the patients. Received 20 November 2002; accepted 26 November 2002. Bernd W. Böttiger MD DEAA, Associate Professor of Anaesthesiology, Co-Chair of the Department of Anaesthesiology, University of Heidelberg, Im Neuenheimer Feld 110, D-69120 Heidelberg, Germany (fax: +49-6221-565345; e-mail: bernd_boettiger@med.uni-heidelberg.de)