Thrombolysis during out-of-hospital cardiac arrest: a lesson in the law of diminishing returns

Article details 
Evidence-Based Medicine Journal Club 
 
Edited by Eric B Milbrandt. University of Pittsburgh Department of Critical Care Medicine

Micro circulatory thrombosis leading to a "no-refl ow" phenomenon after return of spontaneous circulation may contribute to poor neurological function after cardiac arrest [6,7]. A number of studies have evaluated the effi cacy of thrombolysis during out-of-hospital cardiopul mo nary resuscitation. A meta-analysis of these studies, including one prospective and seven retrospective studies, demonstrated an improvement in return of spontaneous circulation, survival to admission, 24-hour survival, hospital discharge, and neurological outcome [8]. Based on these results, the authors concluded that a large, randomized, multicenter study should be conducted to determine the effi cacy of thrombolysis during cardiac arrest.
Th e Th rombolysis in Cardiac Arrest (TROICA) trial investigators conducted a prospective double-blind, randomized, placebo-controlled trial in 66 European emergency medical-service systems (EMS) [1]. Adults with witnessed out-of-hospital cardiac arrest with an EMS response time of less than ten minutes were eligible for the study. Th e study protocol permitted open-label use of thrombolytics rather than randomization for cases in which pulmonary embolism was suspected as the cause of arrest. Patients with an initial rhythm of asystole or pulseless electrical activity were immediately randomized to weight-based tenecteplase or placebo, and patients with ventricular fi brillation or pulseless ventricular tachycardia were randomized after three failed attempts at defi brillation. Adjunctive antithrombotic and antiplatelet agents were not administered. Th e trial was suspended after futility analyses were performed on data from 653 patients. A total of 1050 patients were enrolled and no patient was lost to 30-day follow-up. Th e baseline characteristics of the two groups were well matched in terms of age, comorbidities, and long-term medications, including aspirin and warfarin. EMS response times were similar between groups and median time to study drug administration was 18 minutes. Th e circumstances of cardiac arrest were similar between groups, including the initial rhythm, cardiopulmonary resuscitation (CPR) by bystanders, and defi brillation administered by fi rst responder. Th ere was no diff erence between tenecteplase and placebo in the primary endpoint of 30-day survival or for any of the secondary endpoints, though there was a higher rate of intracranial hemorrhage in the tenecteplase group. Th e authors concluded that tenecteplase without an adjunctive antithrombotic during CPR does not improve outcome for out-of-hospital cardiac arrest.
Th e TROICA trial has several strengths, including the large sample size, multicenter design, evaluation of clinically important outcomes, and complete follow-up for the primary endpoint. Of particular note is the time to thrombolysis of 18 minutes from collapse, which represents a signifi cantly shorter time than the typical 30 minutes cited in previous studies. Despite these strengths, the study is subject to a few important limitations. Most detailled information regarding inhospital care was lacking, which may have aff ected the primary outcome of 30-day survival. In addition, survival data may be subject to selection bias as the authors allowed -for ethical reasons -the open-label use of thrombolytics for suspected pulmonary embolism, potentially excluding from randomization a subgroup of patients likely to benefi t from thrombolysis. Despite these limitations, the TROICA Trial convincingly demon strates no mortality benefi t from thrombolysis with tenecteplase and an increase risk of asymptomatic intracranial hemorrhage in patients with out-of-hospital cardiac arrest.
Th e search for new interventions to improve outcomes for out-of-hospital cardiac arrest remains elusive. Why did the current trial fail to show a benefi t for thrombolysis despite a strong biologic rationale and a suggestion of benefi t in prior, albeit smaller, studies? Decreased perfusion pressure may have prevented drug delivery and reduced the effi cacy of thrombolytics. Alternatively, the negative result seen in the TROICA trial could be ascribed to a lack of adjunctive antithrombotic or antiplatelet agents, given that all eight studies in the Li et al meta-analysis used heparin with or without aspirin [8]. Th e most likely explanation, however, may be the law of diminishing returns. Th e TROICA trial was conducted within a well-optimized EMS system, as evidenced by the rapid EMS response and time to thrombolysis. Furthermore, the authors selected a patient population with potential for a favorable outcome, as evidenced by the 30-day survival of 17% in the placebo group compared to 10% in most studies [9]. Th e corollary to this is that the incremental benefi t of pre-hospital advanced life support beyond early CPR and defi brillation tends to be minimal, a lesson learned from Th e Ontario Prehospital Advanced Life Support (OPALS) study [10].

Recommendation
Based on the results of the TROICA trial, there seems to be no benefi t from the use of tenecteplase without adjunctive antithrombotic therapy in out-of-hospital cardiac arrest. No such conclusion can be made regarding the subgroup of patients with suspected pulmonary embolism and the results should not be generalized to the inpatient setting.