A Matched Comparison of theCombination of Prehospital Thrombolysis

and Standby Rescue Angioplasty WithPrimary Angioplasty

Jean-Michel Juliard, MD, Dominique Himbert, MD, Pascal Cristofini, MD,Jean-Charles Desportes, MD, Monique Magne, MD, Jean-Louis Golmard, MD, PhD,

Pierre Aubry, MD, Hakim Benamer, MD, Albert Boccara, MD, Gaetan J. Karrillon, MD,and P. Gabriel Steg, MD

This study sought to assess the rate of acute Thrombol-ysis In Myocardial Infarction (TIMI) trial grade 3 patencythat can be achieved with the combination of prehospitalthrombolysis and standby rescue angioplasty in acutemyocardial infarction. No large angiographic study hasbeen performed after prehospital thrombolysis to deter-mine the 90-minute TIMI 3 patency rate in the infarct-related artery. Hospital outcome and artery patencywere compared to 170 matched patients treated withprimary angioplasty. Prehospital thrombolysis was ap-plied 151 6 61 minutes after the onset of pain in 170patients (56 6 12 years, 86% men), using recombinanttissue-type plasminogen activator, streptokinase, oreminase. Emergency 90-minute angiography was per-formed in every case. All patients in whom thrombolysisfailed underwent rescue angioplasty. After thrombolysisalone, TIMI grade 3 flow in the infarct-related artery wasobserved in 108 patients (64%), TIMI grade 2 in 12 (7%),and TIMI grade 0 or 1 in 50 (29%). Rescue angioplastywas successful in 47 of 50 attempts. Overall, TIMI 3

patency was achieved in 91%, and additionally TIMI 2flow in 7% of patients, an average of 113 6 39 minutesafter thrombolysis and 55 6 19 minutes after admis-sion. Therefore, <2 hours after thrombolysis, only 2% ofpatients had persistent occlusion (TIMI 0 or 1) of theinfarct-related artery. In-hospital mortality was 4% over-all (7 of 170), and 3% in the 155 patients in whom TIMI3 was obtained during the acute phase. Severe hemor-rhagic complications occurred in 14 patients (8%) with 2fatal cerebral hemorrhages (7% of patients requiredtransfusions). The matched comparison with primaryPTCA showed no significant difference in hospital out-come. Combined prehospital thrombolysis, 90-minuteangiography, and rescue angioplasty yield a high rateof acute TIMI 3 patency rate early after thrombolysis andhospital admission. A randomized, prospective compar-ison between these 2 reperfusion strategies may be nowwarranted. Q1999 by Excerpta Medica, Inc.

(Am J Cardiol 1999;83:305–310)

Thrombolytic therapy is the standard treatment forongoing acute myocardial infarction. The greatest

benefit, in terms of survival and limitation of infarctsize, is obtained when treatment is started early afterthe onset of symptoms.1 In addition, there are data tosuggest that streptokinase is more effective at achiev-ing patency when administered early in the course ofinfarction.2 Prehospital thrombolysis gains an esti-mated 30 to 120 minutes, and yields a slight butsignificant reduction in mortality compared with in-hospital thrombolysis.3–13 After thrombolysis, early(90-minute) and complete Thrombolysis In Myocar-dial Infarction (TIMI) trial flow grade 3 patency of theinfarct-related artery is a strong predictor of survival,but is obtained in only about half the patients; forexample, in the Global Utilization of Streptokinaseand t-PA for Occluded Coronary Arteries (GUSTO)

angiographic substudy, accelerated tissue-type plas-minogen activator (t-PA) achieved only a 54% TIMI 3patency rate on 90-minute angiography.14 Moreover,successful emergency rescue percutaneous translumi-nal coronary angioplasty (PTCA) may improve theprognosis of patients with failed thrombolysis.15,16

This study explored whether it was possible and safeto maximize the acute TIMI 3 patency rate by com-bining prehospital thrombolysis, routine 90-minuteangiography, and standby rescue PTCA. In-hospitalresults of this strategy were compared with those of170 primary PTCA-treated patients, matched accord-ing to age, sex, and infarct location.

METHODSStudy patients and enrollment criteria: Prehospital

thrombolysis was applied to 170 patients with acutemyocardial infarction of,6-hour duration. Therapywith aspirin ($250 mg) and thrombolysis were begunin the mobile intensive care unit. Patients were imme-diately transferred to the coronary care unit and un-derwent 90-minute coronary angiography on admis-sion; rescue PTCA was performed if there was TIMIgrade 0 or 1 flow in the infarct-related artery. Patientswith TIMI grade 3 or 2 were treated medically. In

From the Cardiology Department, Hopital Bichat, Paris; Service AideMedicale Urgente, Hopital Necker, Paris; Brigade de Sapeurs Pompi-ers, Paris; Service Medical Urgence et Reanimation, Hopital Beaujon,Clichy; and INSERM U436, Medical Biostatistics, Hopital Pitie-Sal-petriere, Paris, France. Manuscript received May 26, 1998; revisedmanuscript received August 20, 1998, and accepted August 21.

Address for reprints: Jean-Michel Juliard, MD, Service de Cardi-ologie A, Hopital Bichat, 46 rue Henri-Huchard, 75018 Paris,France.

305©1999 by Excerpta Medica, Inc. 0002-9149/99/$–see front matterAll rights reserved. PII S0002-9149(98)00858-3

Paris, prehospital triage of patients with suspectedmyocardial infarction is generally done by physiciansin mobile emergency units. If indicated, thombolysisis often initiated at home or in the ambulance andpatients are generally transferred directly to the coro-nary care unit, bypassing the emergency room. In ourcardiology department, the catheterization laboratoryis located in the coronary care unit and operatesround-the-clock with trained physicians on call. Inclu-sion in this study was based on conventional criteria,i.e., chest pain lasting.30 minutes and resistant tonitrates, with typical electrocardiographic changes($1-mV ST-segment elevation in$2 limb leads or$2-mV ST-segment elevation in$2 continguous pre-cordial leads on a 12-lead electrocardiogram) andeligibility for thrombolysis. The diagnosis was con-firmed by creatine kinase elevation (more than twicenormal levels) in every case. Exclusion criteria wererelated to the risk of bleeding, i.e., prolonged cardio-pulmonary resuscitation (.30 minutes), systolicblood pressure.200 mm Hg, oral anticoagulant ther-apy, history of stroke or transient ischemic attack,known bleeding disorder, inability to communicate,recent intramuscular or intraarterial puncture, and gas-trointestinal bleeding, surgery, major trauma, urologicbleeding, or hemoptysis within the previous 3 months.

Coronary angiography and rescue PTCA: Coronaryangiography was performed using 6Fr sheaths andcatheters via the femoral artery 90 minutes after ini-tiation of thrombolytic therapy; 5,000 IU heparin wasgiven immediately after sheath insertion. Angio-graphic patency of the infarct-related artery was im-mediately graded using the TIMI score.17 Multivesseldisease was defined as the presence of$50% luminaldiameter stenoses in at least 2 major epicardial arter-ies. All patients undergoing emergency rescue PTCAfor failed thrombolysis (TIMI 0 to 1) received anadditional bolus of at least 5,000 IU heparin at thebeginning of the procedure. Patients with TIMI 3 or 2flow at 90 minutes were treated medically.

In-hospital course: All patients received aspirin (atleast 250 mg/day) and intravenous heparin until pre-discharge angiography (started at 1,000 IU/hour), ad-justed to an activated partial thromboplastin time of 2to 3 times control levels. All patients receivedb-ad-renergic blocking agents, first intravenously thenorally unless contraindicated. Before hospital dis-charge, patients underwent repeat coronary angiogra-phy, and left ventricular ejection fraction was mea-sured on radionuclide ventriculography. Elective cor-onary artery bypass grafting was performed whensevere 3-vessel disease or left main coronary arterydisease was present.

Matched comparison: To compare in-hospital out-come of patients treated with prehospital thrombolysisas opposed to patients eligible for thrombolysis buttreated with primary PTCA, the 170 patients in thisstudy were compared with a group of 170 treated withprimary PTCA (without contraindication to thrombol-ysis) from the same cohort, matched for age, sex, andinfarct location.

Statistical analysis: All continuous variables are ex-pressed as mean6 SD. Groups were compared bymeans of chi-square analysis for categorical variablesand Student’st test for numerical studies. Fisher’sexact test was used if there was an expected cell value,5. Statistical analysis was performed on SAS soft-ware (version 6.04, SAS Institute, Cary, North Caro-lina). A p value,0.05 was considered significant.

RESULTSOne hundred seventy patients were enrolled in the

study and received prehospital thrombolysis. The di-agnosis of myocardial infarction was confirmed inevery case. Mean age was 566 12 years (range 27 to83), most patients were men (86%), and the infarctlocation was anterior in 49%. Risk factors for coro-nary artery disease are given in Table I. The averagetime from onset of chest pain to thrombolysis was1516 61 minutes, and the average time from onset ofchest pain to hospital admission was 2096 92 min-utes (Figure 1). Figure 2 shows the distribution of theinterval from pain onset to initiation of thrombolytictherapy. Standard regimens of thrombolytic therapywere used, according to the preference of the physi-cian in the mobile care unit, as follows: rt-PA (n5110; 100 mg over 90 minutes in 46 patients, acceler-ated rt-PA [15 mg bolus followed by an infusion of0.75 mg/kg over 30 minutes and then an infusion of0.5 mg/kg over 60 minutes] in 61 patients; and doublebolus rt-PA [double bolus of 50 mg given 30 minutesapart] in 3 patients included in the Continuous Infu-sion versus Double-Bolus Administration of Alteplase[COBALT] study); streptokinase (n5 45; 1.5 millionIU/60 minutes); or eminase (n5 15; 30 IU bolus). At90-minute coronary angiography, infarct-related ar-tery patency was TIMI grade 3 in 108 patients (64%),TIMI grade 2 in 12 (7%), and TIMI grade 0 or 1 in 50(29%) (Figure 3). Patients with TIMI 2 or 3 flow(71%) were treated medically. All 50 patients inwhom thrombolysis failed underwent immediate res-cue PTCA, with 47 successes (TIMI 3 flow and re-

TABLE I Baseline Patient Characteristics

PrehospitalThrombolysis(n 5 170)

PrimaryAngioplasty(n 5 170)

Age (mean yr 6 SD) 56 6 12 57 6 13Male sex 147 (86) 147 (86)Anterior MI location 84 (49) 84 (49)Previous MI 14 (8) 26 (15)Shock upon admission 4 (2) 1 (0.5)Risk factors

History of smoking 130 (77) 113 (67)Dyslipidemia 76 (45) 60 (36)Family history of CAD 59 (35) 56 (33)Hypertension 56 (33) 60 (35)Diabetes mellitus 24 (14) 22 (13)

Time from pain to admission (min) 209 6 92 181 6 90*

* p ,0.03Values are expressed as number (%) unless otherwise noted.CAD 5 coronary artery disease; MI 5 myocardial infarction.

306 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 83 FEBRUARY 1, 1999

sidual stenosis,50%) and 3 failures. Four patientsrequired stent implantation because of a large dissec-tion. Overall, acute TIMI 3 patency was achieved in91% of the 170 patients, acute TIMI 2 in 7%, on anaverage 113 minutes after the start of thrombolysisand 55 minutes after hospital admission. Therefore,only 2% of the patients had a persistent occlusion(TIMI 0 or 1) of the infarct-related artery,1 hourafter arrival to the hospital. The average interval be-tween the onset of chest pain and angiographic evi-dence of TIMI 3 flow was 2646 78 minutes (Figure1). Multivessel disease was identified in 22 patients(13%). The infarct-related vessel was the left anteriordescending in 49%, the right in 40%, and the leftcircumflex coronary artery in 11%.

Cardiac events: Ventricular fibrillation occurred in3 patients before hospital admission and thrombolytictherapy (they received the thrombolytic drug afterat-home resuscitation), and in another 2 patients dur-ing coronary angiography. All patients underwent suc-cessful defibrillation by electrical countershocks.Three other cases of ventricular fibrillation occurredduring hospital stay. Twelve patients with symptom-

atic recurrent ischemia required emer-gency catheterization. Angiographyidentified 5 reocclusions and 7 patent(TIMI grade 3) infarct-related arteries,but with a tight stenosis in the infarct-related artery. All these patients un-derwent revascularization: 10 weresuccessfully treated with PTCA and 2underwent emergency coronary arterybypass grafting because of severe3-vessel disease. Cardiac surgery wasrequired in 9 cases during hospitalstay, once for a ventricular septal de-fect, twice for recurrent ischemia (aspreviously mentioned), and 6 times asan elective procedure.

Hemorrhagic complications: Severehemorrhagic complications occurredin 14 patients (8%): 2 fatal cerebralhemorrhages occurred despite the ab-

sence of risk factor for bleeding, 2 patients had gas-trointestinal bleeding, and 2 patients had a retroperi-toneal hematoma. The 4 latter patients required bloodtransfusions and had no further complication. Accesssite hematomas required transfusion in 8 patients(5%), but did not lead to surgery and were never fatal.Overall, 7% of patients (n5 12) received transfu-sions.

Mortality: The in-hospital mortality rate was 4% (7deaths among 170 patients). No death occurred beforecoronary care unit admission. Among the 7 patientswho died, 2 women aged 69 and 66 years died ofhemorrhagic stroke. Three patients with a large ante-rior infarct died of heart failure, 2 of whom werealready in cardiogenic shock on admission. One pa-tient died of free wall rupture a few hours after re-ceiving thrombolytic therapy, and 1 died of a ventric-ular septal defect,24 hours after successful rescuePTCA of the left anterior descending coronary arterydespite emergency cardiac surgery and cardiopulmo-nary support. The mortality rate was 3% (5 of 155)among patients in whom acute TIMI 3 flow in theinfarct-related artery was demonstrated angiographi-

FIGURE 1. Distribution oftime from symptom onsetto TIMI 3 flow.

FIGURE 2. Distribution of time from symptom onset to thrombolysis.

CORONARY ARTERY DISEASE/PREHOSPITAL THROMBOLYSIS AND RESCUE ANGIOPLASTY 307

cally compared with 8% in patients with TIMI 2 flow(1 of 12) and 33% in patients with persistent occlusion(TIMI grade 0 or 1) (1 of 3) (p5 0.054).

Reocclusion: Among 163 survivors, 145 patientswho did not undergo emergency in-hospital repeatangiography or coronary artery bypass surgery werescheduled for predischarge angiography, which wasperformed in 140 cases (5 patients declined). Silentreocclusion was confirmed angiographically in 11 pa-tients (8%). The infarct-related artery remained patent(TIMI grade 3) in 125 (89%) and TIMI 2 flow wasobserved in 4. The combined silent (n5 11) andsymptomatic in-hospital reocclusion (n5 5) rate was10% (16 of the 155 patients with acute TIMI 3 flow).Calculated left ventricular ejection fraction was 51614% (n5 117 patients).

Comparison with primary PTCA: Baseline character-istics were similar among the 2 groups in terms ofhistory of previous myocardial infarction and riskfactors for coronary artery disease (Table I). Timefrom pain to admission was longer in thrombolysis

group, leading to earlier reperfusion in the group ofpatients treated with primary PTCA. However, thefinal acute TIMI 3 patency rate was the same in the 2groups (91%). There was no significant difference inhospital outcome between the 2 groups of patients(Table II), except for a trend toward a lower rate oftransfusion (p5 0.06) and angiographic reocclusion(p ,0.05) in patients with primary PTCA.

DISCUSSIONThe main result of this study is that the combina-

tion of prehospital thrombolysis with routine 90-minute angiography and rescue PTCA if thrombolysisfails yielded a high early patency rate (91% TIMI 3,7% TIMI 2) of the infarct-related artery. In takingadvantage of prehospital thrombolysis, this wasachieved, on average, 1136 39 minutes after the startof thrombolysis and only 556 19 minutes after hos-pital admission. To the best of our knowledge, this isthe first large angiographic patency study after pre-hospital thrombolysis. The few existing angiographicstudies after prehospital thrombolysis involved selec-tive and delayed rather than routine immediate an-giography, and did not discriminate patients withTIMI 2 from patients with TIMI 3 flow.5,7

Prehospital thrombolysis: There were no misdiag-noses in this study, and the net time gain comparedwith in-hospital thrombolysis could be estimated to 58minutes (not taking into account delays inside thehospital). Because this was not a randomized compar-ison of prehospital versus hospital thrombolysis, theprobability that prehospital thrombolysis actually de-layed transportation to the hospital cannot be excludedand therefore the rapidity gain of 58 minutes may beoverestimated.

Coronary angiography after thrombolysis: Ninety-minute TIMI 3 patency of the infarct-related arteryafter thrombolysis is a major predictor of survival andpreserved left ventricular function.14 We have previ-ously reported the results of a patency-oriented strat-egy in the management of acute myocardial infarctionusing, among other approaches, routine 90-minute an-giography after thrombolysis for early detection offailed thrombolysis; there were no major complica-tions related to vascular access, and the success rate ofrescue PTCA was high.18 Early real-time noninvasivediagnosis of reperfusion within the first 3 hours afterthrombolysis (75% of patients in this cohort) is diffi-cult: routine clinical and electrocardiographic markersare unreliable,19 and biochemical markers are morereliable in patients receiving thrombolysis later than 3hours after symptom onset.20 Moreover, coronary an-giography early after thrombolysis has substantial ad-vantages,21 including identification of high-risk pa-tients (3-vessel disease or significant narrowing of theleft main coronary artery) for triage to early coronaryartery bypass surgery; and identification of patientswith no significant coronary artery lesions, who can bedischarged rapidly. In addition, angiography itself hasa small but tangible positive impact on the patency ofthe infarct-related artery.22 Finally, the Thrombolysisand Angioplasty in Myocardial Infarction trial showed

FIGURE 3. Acute angiographic evaluation of the 170 patients.

308 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 83 FEBRUARY 1, 1999

that angiographic detection of failed thrombolysis fol-lowed by rescue PTCA was associated with betterregional wall motion relative to conservative manage-ment.20 Bleeding around the vascular access site wasinfrequent and never fatal.

Rescue PTCA: The mortality rate among patientswith failed thrombolysis is nearly twice as high as thatin patients successfully treated with thrombolysisalone.23 The encouraging results of rescue PTCA inseveral cohort studies15,24were confirmed in the Ran-domized Evaluation of Salvage angioplasty withCombined Utilization of Endpoints (RESCUE) trial.15

In contrast, the TIMI 4 study, despite a higher inci-dence of TIMI grade 3 flow after successful rescuePTCA (performed at 120 minutes) compared withsuccessful thrombolysis (87% vs 65%, p5 0.002)failed to show any improvement in clinical outcomesin a nonrandomized comparison with patients treatedwithout rescue PTCA (mortality 12% vs 11%, respec-tively), although this may be related to relatively late(5 hours) intervention after chest pain onset.25 Thepresent results are consistent with our previous expe-rience of rescue PTCA,15 demonstrating a low overallmortality in patients in whom rescue PTCA is at-tempted (4%). These results, combined with the re-sults of the randomized but relatively small RESCUEtrial, provide impetus for continued evaluation of res-cue PTCA. In the present study, all 50 patients inwhom thrombolysis failed underwent rescue PTCA,which achieved acute TIMI 3 patency in 47 (94%); thein-hospital mortality rate was only 4% among these 50patients.

Mortality: The in-hospital mortality observed hereappears low, but comparison with other trials shouldbe cautious given possible baseline differences and thesmall size of the present study. However, it is lowerthan the death rate reported in other nonrandomizedstudies of rescue PTCA after failed thrombolysis.25

Comparison with primary PTCA: Because it was nota prospective randomized trial, this matched compar-ison suggests that combination of rescue PTCA andprehospital thrombolysis led to in-hospital results sim-ilar to primary PTCA in thrombolysis-eligible pa-tients. Acute TIMI 3 patency rate of the infarct-relatedartery was high in the 2 groups (91%), but with asmaller reocclusion rate in patients with primary

PTCA. Actually, times from pain toreperfusion in both groups of pa-tients were probably very close if weconsider that initiation of thrombol-ysis was performed on an average 30minutes before admission, and reper-fusion probably occured earlier, be-fore 90-minute angiography inthrombolyzed patients. Moreover,patients treated with primary PTCAgot to the hospital earlier. The trendfor a lower reocclusion rate (al-though p,0.05 may not be mean-ingfully significant in a nonrandom-ized comparison) observed in pa-tients with primary PTCA may be

due to a higher rate of stent implantation (36 of 170)than that in patients treated with rescue PTCA (4 of50).

Study limitations: This was a single-center, uncon-trolled study, performed in a tertiary care centertrained in round-the-clock use of interventional cardi-ology for acute myocardial infarction. The variety ofthrombolytic regimens is also a major limitation butreflects a pragmatic approach of each physician in themobile care unit. TIMI flow was not graded in acentral core laboratory, but in real time by an experi-enced physician, since therapeutic decisions werebased on immediate patency. This setting correspondsto real-life application of a patency-oriented strategy.In addition, whereas TIMI 3 patency grading is oftenpoorly reproducible, previous experience from theTIMI 4 trial has suggested that there is little overlapbetween TIMI 0 to 1 and TIMI 2 to 3 patency (K 50.846 0.05).26 In this study, TIMI 2 to 3 was obtainedin 98% of patients. Whereas prehospital thrombolysiscan be implemented in a wide variety of clinicalsettings.3, 9–11widespread use of this aggression reper-fusion strategy after prehospital thrombolysis may belimited by the need for immediate and permanentavailability of catheterization laboratory facilities andtrained physicians. Advances in the early noninvasivediagnosis of reperfusion are therefore needed to avoidunnecessary coronary angiograms in patients in whomthrombolysis is successful. Electrocardiographic orvectorcardiographic monitoring (especially beforehospital admission), or direct imaging of myocardialreperfusion using contrast echocardiography,27,28 or99mTc-hexakis-2-methoxy-2-methylpropyle-isonitrilsingle-photon emssion computed tomography29 mayimprove early detection of failed thrombolysis.

Acknowledgment: We thank the cardiac catheter-ization laboratory team for their participation in theround-the-clock effort, and Laurence Toussaint forpreparing the manuscript.

1. Simoons ML, Serruys PW, van den Brand M, Res J, Verheugt FWA, KrausXH, Remme WJ, Ba¨r F, de Zwaan C, van der Laarse, Vermeer F, Lubsen J. Earlythrombolysis in acute myocardial infarction: limitation of infarct size and im-proved survival.J Am Coll Cardiol1986;7:717–728.2. Steg PG, Laperche T, Golmard JL, Juliard JM, Benamer H, Himbert D, Aubry

TABLE II In-Hospital Outcome

PrehospitalThrombolysis

PrimaryAngioplasty

Angiographically-proven TIMI 3 patency 91% 91%Time from pain to reperfusion (min) 264 6 78 232 6 94*Death 7 (4.1%) 8 (4.7)Recurrent ischemia 12 (7%) 11 (6.4)Coronary bypass 8 9Reocclusion (angiographic) 16 (10%) 7% (4.5)†Transfusion .1U without coronary bypass 12 (7%) 2% (1)‡VF 6 5Ejection fraction 51 6 14% (n 5 117) 51 6 13% (n 5 142)

*p ,0.02; †p ,0.05; ‡p 5 0.06.VF 5 ventricular fibrillation.

CORONARY ARTERY DISEASE/PREHOSPITAL THROMBOLYSIS AND RESCUE ANGIOPLASTY 309

P, for the PERM Study Group. Efficacy of streptokinase, but not tissue-typeplasminogen activator, in achieving 90-minute patency after thrombolysis foracute myocardial infarction decreases with time to treatment.J Am Coll Cardiol1998;31:776–779.3. Koren G, Weis AT, Hasin Y, Appelbaum D, Welber S, Rozenman Y, Lotan C,Mosseri M, Sapoznikov D, Luria MH, Gotsman MS. Prevention of myocardialdamage in acute myocardial ischemia by early treatment with intravenous strep-tokinase.N Engl J Med1985;313:1384–1389.4. Sauval P, Artigou JY, Cristofini P, Jannie`re D, Barrier G. Fibrinolyse pre´hos-pitaliere par rt-PA dans l’infarctus aigu du myocarde.Arch Mal Coeur1989;82;1957–1961.5. Castaigne AD, Herve´ C, Duval-Moulin AM, Gaillard M, Dubois-Rande JL,Boesch C, Wolf M, Lellouche D, Jan F, Vernant P, Huguenard P. Prehospital useof APSAC: results of a placebo-controlled study.Am J Cardiol1989;64:30A–33A.6. Schofer J, Bu¨ttner J, Geng G, Gutschmidt K, Herden HN, Mathey DG, MoeckeHP, Polster P, Raftopoulo A, Sheehan FH, Voelz P. Prehospital thrombolysis inacute myocardial infarction.Am J Cardiol1989;66:1429–1433.7. Roth A, Barbash GI, Hod H, Miller HI, Rath S, Modan M, Har-Zahav Y, KerenG, Bassan S, Kaplinsky E, Laniado S. Should thrombolytic therapy be adminis-tered in the mobile intensive care unit in patients with evolving myocardialinfarction? A pilot study.J Am Coll Cardiol1990;15:932–936.8. BEPS Collaborative Group. Prehospital thrombolysis in acute myocardialinfarction: the Belgian Eminase Prehospital Study (BEPS).Eur Heart J 1991;12:965–967.9. GREAT Group. Feasibility, safety and efficacy of domiciliary thrombolysis bygeneral practitioners: Grampian Region Early Anistreplase Trial.Br Med J1992;305:548–553.10. The European Myocardial Infarction Project Group. Prehospital thrombolytictherapy in patients with suspected acute myocardial infarction.N Engl J Med1993;329:383–389.11. Weaver WD, Cerqueira M, Hallstrom AP, Litwin PE, Martin JS, KudenchukPJ, Eisenberg M, for the Myocardial Infarction Triage and Intervention ProjectGroup. Prehospital-initiated vs hospital-initiated thrombolytic therapy. The Myo-cardial Infarction Triage and Intervention Trial.JAMA 1993;270:1211–1216.12. Gruseels EWM, Bouten MJM, Lenderink T, Deckers JW, Hoes AW, HartmanJAM, van der Does E, Simoons ML. Pre-hospital thrombolytic therapy witheither alteplase or streptokinase. Practical applications, complications and long-term results in 529 patients.Eur Heart J1995;16:1833–1838.13. Linderer T, Schro¨der R, Arntz R, Heineking ML, Wunderlich W, Kohl K,Forycki F, Henzgen R, Wagner J. Prehospital thrombolysis: beneficial effects ofvery early treatment on infarct size and left ventricular function.J Am CollCardiol 1993;22:1304–1310.14. The GUSTO Angiographic Investigators. The effects of tissue plasminogenactivator, streptokinase, or both on coronary artery patency, ventricular functionand survival after acute myocardial infarction.N Engl J Med1993;329:1615–1622.15. The CORAMI Study Group. Outcome of attempted rescue coronary angio-plasty after failed thrombolysis for acute myocardial infarction.Am J Cardiol1994;74:492–494.16. Ellis SG, da Silva ER, Heyndrickx G, Talley JD, Steg G, Spaulding C,Nobuyoshi M, Erbel R, Vassanelli C, Topol EJ, for the RESCUE Investigators.Randomized comparison of rescue angioplasty with conservative management ofpatients with early failure of thrombolysis for acute anterior myocardial infarc-tion. Circulation 1994;90:2280–2284.17. Chesebro JM, Knatterud G, Roberts R, Borer J, Cohen LS, Dalen J, Dodge

MT, Francis CK, Lubrook P. Thrombolysis myocardial infarction (TIMI) trial,phase I: a comparison between intravenous tissue plasminogen activator andintravenous streptokinase. Clinical findings through hospital discharge.Circula-tion 1987;76:142–154.18. Juliard JM, Himbert D, Golmard JL, Aubry P, Karrillon GJ, Boccara A,Benamer H, Steg PG. Can we provide reperfusion therapy to all unselectedpatients admitted with acute myocardial infarction?J Am Coll Cardiol1997;30:157–164.19. Califf RM, O’Neill W, Stack RS, Aronson L, Mark DB, Mantell S, GeorgesBS, Candela RJ, Kereiakes DJ, Abbotsmith C, Topol EJ. Failure of simpleclinical measurements to predict reperfusion status after intravenous thromboly-sis.Ann Intern Med1988;108:658–662.20. Laperche T, Steg PG, Dehoux M, Benessiano J, Grollier G, Aliot E, MossardJM, Aubry P, Coisne D, Hanssen M, Iliou MC, for the PERM Study Group. Astudy of biochemical markers of reperfusion early after thrombolysis for acutemyocardial infarction. Circulation 1995;92:2079–2086.21. Califf RM, Topol EJ, Stack RS, Ellis SG, Kereiakes DJ, Samaha JK, WorsleySJ, Anderson JL, Harrelson-Woodlief LH, Wall TC, Phillips HR, AbbottsmithCW, Candela RJ, Flanagan WH, Mantel SJ, Lee KL, for the TAMI Study Group.Evaluation of combination thrombolytic therapy and timing of cardiac catheter-ization in acute myocardial infarction. Results of Thrombolysis and Angioplastyin Myocardial Infarction-Phase 5 Randomized Trial.Circulation 1991;83:1543–1556.22. Topol EJ, Holmes DR, Rogers WJ. Coronary angiography after thrombolytictherapy for acute myocardial infarction.Ann Intern Med1991;114:877–885.23. Califf RM, Topol EJ, George BS, Boswick JM, Lee KL, Stump D, Dillon J,Abbottsmith C, Candela RJ, Kereiakes DJ, O’Neill WW, Stack RS, and theTAMI Study Group. Characteristics and outcome of patients in whom reperfusionwith intravenous tissue-type plasminogen activator fails: results of the Throm-bolysis and Angioplasty in Myocardial Infarction (TAMI) I Trial.Circulation1988;77:1090–1099.24. Ross AM, Lundergan CF, Rohrbeck SC, Deneane HB, van den Brand M,Buller CH, Holmes DR, Reiner JS, for the GUSTO-1 Angiographic Investigators.Rescue angioplasty after failed thrombolysis: technical and clinical outcomes ina large thrombolysis trial.J Am Coll Cardiol1998;31:1511–1517.25. Gibson CM, Cannon CP, Greene RM, Sequeira RF, Margorien RD, Leya F,Diver DJ, Baim DS, Braunwald E, for the TIMI 4 Study Group. Rescue angio-plasty in the Thrombolysis in Myocardial Infarction (TIMI) 4 Trial.Am J Cardiol1997;80:21–26.26. Gibson CM, Cannon CP, Daley WL, Dodge JT, Alexander B, Marble SJ,McCabe CH, Raymond L, Fortin T, Poole WK, Braunwald E, for the TIMI 4Study Group. TIMI frame count. A quantitative method of assessing coronaryartery flow.Circulation 1996;93:879–888.27. Porter TR, D’Sa A, Turner C, Jones LA, Minisi AJ, Mohanty PK, VetrovecGW, Nixon JV. Myocardial contrast echocardiography for the assessment ofcoronary blood flow reserve: validation in humans.J Am Coll Cardiol1993;21:349–355.28. Himbert D, Seknadji P, Karila-Cohen D, Juliard JM, Steg PG. Myocardialcontrast echocardiography to assess spontaneous reperfusion during myocardialinfarction. Lancet1997;349:617–619.29. Gibbons RJ, Verani MS, Berhenbeck T, Pellikka PA, O’Connor PK, Mah-marian JJ, Chesebro JH, Wackers FT. Feasibility of tomographic 99mTc-hexakis-2-methoxy-2-methylpropyle-isonitrile imaging for the assessment of myocardialarea at risk and the effect of treatment in acute myocardial infarction.Circulation1989;80:1277–1286.

310 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 83 FEBRUARY 1, 1999