The extent of myocardial damage occurring during acute myocardial infarction is time dependent, and there is abundant evidence from most clinical trials that mortality reduction is greatest in patients treated early with thrombolytic agents, although beneficial effects have been shown with treatment initiated up to 12 hours after onset of symptoms. This temporal dependence of benefit was most clearly seen with the 47% mortality reduction obtained with streptokinase given within the first hour in the GSSI-1 trial (Table 1). The process of infarction may be completely aborted if reperfusion is initiated within 30 minutes after symptom onset. Current approaches designed to reduce the time delay between onset of symptoms and the inhibition of thrombolytic treatment include increasing public awareness of the need to seek prompt medical attention in the event of chest pain, reducing in-hospital delays in initiation of thrombolytic therapy, and administration of thrombolytic agents in the prehospital setting. Since the first paper on prehospital thrombolysis by Koren et al. in 1985 there have been numerous studies published, including several randomized trials comparing prehospital and in-hospital initiation of thrombolytic therapy (Table 2). In randomized studies patients received in the prehospital phase either a bolus or infusion with the thrombolytic agent followed by placebo after hospital admission, or placebo when first seen outside the hospital, followed by the thrombolytic agent after arrival at the hospital. This was performed in a double-blind fashion in some of the trials. Some of the feasibility and safety, others on time again, differences in ejection fraction and mortality benefit. Almost all have conclusively shown the practicability and safety of patient selection and administration of the thrombolytic agent. This positive result was observed in different organizational types of the emergency system: In the Myocardial Infarction Triage and Intervention trial (MITI) patients with chest pain were screened by paramedics using a checklist of clinical inclusion and exclusion criteria. The paramedics then obtained a computer-interpreted ECG which was transmitted to the emergency department in the hospital where a physician made the decision on the form of treatment. The thrombolytic agent was then administered by the paramedic. In the European Myocardial Infarction Project (EMIP) an emergency physician was personally present and responsible in the prehospital setting, whereas in the Grampian Region Early Anistreplase Trial (GREAT) general practitioners made the decision for enrolling the patient. The accuracy of diagnosis in the prehospital setting was comparable to trials of in-hospital thrombolysis. E.g. in MITI 98% of the patients enrolled had subsequent evidence of acute myocardial infarction. Table 3 shows the final diagnosis in EMIP, the largest trial of prehospital thrombolysis up to date, with about 90% correct diagnoses. In the swedish Thrombolysis Early in Acute Heart Attack Trial (TEAHAT) no ECG criteria were used and only 42% of prehospitally treated patients later developed definite signs of acute myocardial infarction. Therefore the use of ECG criteria for correct identification of patients suitable for prehospital thrombolysis seems to be warranted. Another question of trials of prehospital thrombolysis was whether the rate of complications would be comparable to in-hospital initiation of treatment. In the EMIP-study there was a significant excess of ventricular fibrillation in the prehospital group as compared with the hospital group during the period before admission which was offset by a higher incidence of this complication in the hospital group during the hospital period. The incidence of cardiac arrest and shock showed a similar trend. The incidence of bleeding complications and stroke were not statistically different between the two groups. With comparable results of other trials the prehospital initiation of thrombolytic therapy appears to be as safe as in-hospital treatment. With regard to time savings all randomized studies showed positive results (Table 4). The smallest time gain was observed in the MITI-trial: prehospitally treated patients received thrombolytic therapy a median of 33 minutes earlier than those treated in-hospital. In EMIP the difference in time between prehospital and hospital treatment was a median of 55 minutes. However, none of these trials were able to show a significant short-term mortality difference between the two groups (Table 4). The ability e.g. of the MITI-study to detect a significant difference in mortality was considerably reduced by the astonishing fact that the time gain of prehospital thrombolysis was with 33 minutes less than to be expected in real life. This was in part due to the fact that hospital treatment was accelerated by ECG transmission to the receiving hospital by the paramedics. A meta-analysis of five randomized studies with a combined median time gain of about 60 minutes showed a significant 17% reduction in short-term mortality for patients who received thrombolytic therapy in the prehospital. phase (Table 4). In GREAT, a study performed in a more rural area than other studies, the time gain by prehospital initiation of thrombolysis was a median of 130 minutes. GREAT was the only study up to date publishing long-term mortality results: There was a significant mortality benefit for prehospitally treated patients after 3 months and 1 year (Table 5). In conclusion, prehospital thrombolysis is feasible and safe. Patients with acute myocardial infarction can be correctly identified and treated with thrombolytic agents in the prehospital setting with the same rate of complications as expected for in-hospital thrombolysis provided basic resuscitation equipment including a defibrillator is available. The results of randomized studies comparing the results of prehospital and in-hospital thrombolysis seem to justify the prehospital institution of thrombolytic therapy especially in rural areas where transport times to the hospital are long and the expected time gain is largest.
