Articles

Primary versus tenecteplase-facilitated percutaneous coronary intervention in patients with ST-segment elevation acute myocardial infarction (ASSENT-4 PCI): randomised trial Assessment of the Safety and Efficacy of a New Treatment Strategy with Percutaneous Coronary Intervention (ASSENT-4 PCI) investigators*

Summary Background Primary percutaneous coronary intervention (PCI) is more effective than fibrinolytic therapy for ST-segment elevation acute myocardial infarction (STEMI), but time to intervention can be considerable. Our aim was to investigate whether the administration of full-dose tenecteplase before a delayed PCI could mitigate the negative effect of this delay. Methods We did a randomised study in which we assigned patients with STEMI of less than 6 h duration (scheduled to undergo primary PCI with an anticipated delay of 1–3 h) to standard PCI (n=838) or PCI preceded by administration of full-dose tenecteplase (n=829). All patients received aspirin and a bolus, without an infusion, of unfractionated heparin. Our primary endpoint was death or congestive heart failure or shock within 90 days. Analyses were by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00168792. Findings We planned to enrol 4000 patients, but early cessation of enrolment was recommended by the data and safety monitoring board because of a higher in-hospital mortality in the facilitated than in the standard PCI group (6% [43 of 664] vs 3% [22 of 656], p=0·0105). Of those enrolled, six were lost to follow-up in the facilitated PCI group and seven in the other group. Median time from randomisation to first balloon inflation was similar in both groups. The median time from bolus tenecteplase to first balloon inflation was 104 min. We noted the primary endpoint in 19% (151 of 810) of patients assigned facilitated PCI versus 13% (110 of 819) of those randomised to primary PCI (relative risk 1·39, 95% CI 1·11–1·74; p=0·0045). During hospital stay, significantly more strokes (1·8% [15 of 829] vs 0, p0·0001), but not major non-cerebral bleeding complications (6% [46 of 829] vs 4% [37 of 838], p=0·3118), were reported in patients assigned facilitated rather than standard PCI. We also noted more ischaemic cardiac complications, such as reinfarction (6% [49 of 805] vs 4% [30 of 820], p=0·0279) or repeat target vessel revascularisation (7% [53 of 805] vs 3% [28 of 818], p=0·0041) within 90 days in this study group.

Lancet 2006; 367: 569–78 Published Online February 14, 2006 DOI:10.1016/S0140-6736(06) 68147-6 See Comment page 543 *Members are listed at end of paper Correspondence to: Professor Frans Van de Werf, Department of Cardiology, Gasthuisberg University Hospital, Herestraat 49, B-3000 Leuven, Belgium [email protected]. ac.be

Interpretation A strategy of full-dose tenecteplase with antithrombotic co-therapy, as used in this study and preceding PCI by 1–3 h, was associated with more major adverse events than PCI alone in STEMI and cannot be recommended.

Introduction Primary percutaneous coronary intervention (PCI) is better than fibrinolytic therapy when delivered soon after onset of symptoms by an experienced team.1 As such, US and European guidelines2,3 recommend that primary PCI is done within 90 min of presentation or within 60 min of the time to fibrinolysis. Although such a timeframe is generally achievable in randomised studies, in general practice it is often not met. The logistics of transporting patients between hospitals greatly increase time to treatment. In the latest report of the National Registry of Myocardial Infarction,4 for example, the median delay between arrival in a community hospital and first balloon inflation after transfer to a tertiary care hospital was more than 3 h. Fibrinolytic therapy has a similar effect on survival as primary PCI when administered to patients within 2–3 h after onset of their symptoms.5–7 The greatest benefits of reperfusion are noted in this early time window. These considerations, and the fact that www.thelancet.com Vol 367 February 18, 2006

outcomes after primary PCI are much improved when the procedure is done on an open vessel,8 provide the rationale for the notion of facilitated PCI. The use in facilitated PCI of fibrinolytic agents,9–14 glycoprotein IIb/IIIa inhibitors,15–22 or a combination of the two23–25 has been studied previously, but only in small numbers of patients; no clear clinical benefit has been noted. Our aim was to ascertain whether facilitated PCI is more effective than delayed standard primary PCI in patients who have large myocardial infarctions.

Methods Patients Between Nov 10, 2003, and April 22, 2005, we did a randomised trial to which we enrolled patients from 24 countries at three locations: tertiary care hospitals, community hospitals without facilities for PCI, and ambulances in which prehospital fibrinolysis was available. Our inclusion criteria were age 18 years or 569

Articles

Procedures 1667 enrolled and randomised

829 assigned facilitated PCI 6 lost to follow-up 823 completed 90-day follow-up 13 vital status only 810 composite endpoint

838 assigned standard PCI 7 lost to follow-up 831 completed 90-day follow-up 12 vital status only 819 composite endpoint

Figure 1: Trial profile

older; symptom onset 6 h or less before randomisation; intention to undertake primary PCI; total ST-segment elevations of 0·6 mV or more across multiple leads or, for inferior infarction, ST-segment deviations of 0·6 mV or more provided ST-segment elevations of 0·4 mV or more were present in leads II, III, and aVF; or new left bundle-branch block with concordant STsegment elevation of 0·1 mV or more. Our exclusion criteria were expected arrival at the catheterisation laboratory less than 1 h or more than 3 h after randomisation; anticipated problems with vascular access; previous enrolment to the study; and the usual contraindications to fibrinolytic therapy, as described in ASSENT-3.26 The institutional review board of every hospital involved approved the protocol, and every patient had to provide written informed consent.

Age (years) Age 75 years Women Weight (kg) Height (cm) Killip class I II/III IV Congestive heart failure at randomisation Heart rate (bpm) Systolic blood pressure (mm Hg) Infarct location Anterior Inferior Other Previous infarction Previous congestive heart failure Previous PCI Previous coronary artery bypass graft Hypertension Diabetes

Tenecteplase and PCI (n=829)

PCI alone (n=838)

61 (12·1) 98/829 (12%) 193/829 (23%) 77·9 (14·7) 170·3 (8·2)

60 (12·0) 106/837 (13%) 189/838 (23%) 77·7 (14·9) 169·8 (8·4)

753/829 (91%) 66/829 (8%) 10/829 (1%) 42/806 (5%) 74·3 (16·9) 133·7 (24·2)

773/834 (93%) 53/834 (6%) 8/834 (1%) 41/814 (5%) 76·1 (17·1) 133·7 (22·2)

403/828 (49%) 418/828 (50%) 7/828 (1%) 108/822 (13%) 7/824 (1%) 70/819 (9%) 18/825 (2%) 391/829 (47%) 144/828 (17%)

389/837 (46%) 429/837 (51%) 19/837 (2%) 90/834 (11%) 13/834 (2%) 68/829 (8%) 16/833 (2%) 391/837 (47%) 131/836 (16%)

Data are mean (SD) or number (%). Calculations based on available data.

Table 1: Baseline characteristics

570

We randomly assigned patients via a central computerised telephone system to routine primary PCI or to PCI preceded by the administration of a standard, bodyweight-adjusted bolus of tenecteplase as used in ASSENT-227 and ASSENT-3.26 In addition to study medication, all patients received 150–325 mg aspirin and a single intravenous bolus of unfractionated heparin (70 U/kg without a maximum dose in the primary PCI group and 60 U/kg with a maximum of 4000 U in the facilitated PCI group). Although an infusion of heparin was not given, additional intravenous bolus could be administered to obtain an activated clotting time of 350–400 s in the primary PCI group (or 250–300 s if a glycoprotein IIb/IIIa inhibitor was used) or of 300–350 s in the facilitated PCI group during the intervention. We left use of glycoprotein IIb/IIIa antagonists to the discretion of the investigator in primary PCI patients, but did not allow their use in the facilitated PCI group, except in bailout situations (large residual clot with no, or suboptimum, recanalisation of the infarct artery). Coronary angiography was done as soon as possible after randomisation. Although we encouraged coronary intervention on the culprit lesion, we left the decision as to whether to do the procedure to the investigator. We asked that TIMI (thrombolysis in myocardial infarction) flow grades before and after the intervention be recorded. If a stent was deployed, patients were given a loading dose of 300 mg clopidogrel as soon as possible followed by a maintenance dose of 75 mg, irrespective of group. We left all other treatment decisions to the discretion of the responsible doctor. Our primary endpoint was the combination of death, congestive heart failure, or shock within 90 days of randomisation. Congestive heart failure and shock were centrally adjudicated by a clinical event committee unaware of the allocated treatment. We defined cardiogenic shock as: 1) systolic blood pressure less than 90 mm Hg for at least 30 min (or the need for supportive measures to maintain a systolic blood pressure of greater than 90 mm Hg) in the presence of a heart rate of more than 60 bpm and in association with signs of hypoperfusion (cool extremities, or urinary output of less than 30 mL/h or mental confusion, or both); or 2) a cardiac index of less than 2·2 L/min/m2 in the presence of a pulmonary capillary wedge pressure of more than 15 mm Hg. A diagnosis of congestive heart failure consisted of at least one of the following conditions, requiring treatment with diuretics: 1) pulmonary oedema or congestion on chest radiograph without suspicion of a non-cardiac cause; 2) rales of more than 1/3 up from the lung base (Killip class 2 or higher); 3) pulmonary capillary wedge pressure of more than 25 mm Hg; or 4) dyspnoea with an oxygen pressure of less than 80 mm Hg or oxygen saturation of less than 90% without supplemental www.thelancet.com Vol 367 February 18, 2006

Articles

Symptom onset to randomisation (min) Symptom onset to unfractionated heparin bolus (min) Symptom onset to tenecteplase (min) Symptom onset to first balloon (min) Randomisation to first balloon (min) Tenecteplase to first balloon (min)

Tenecteplase and PCI (n=829)

PCI alone (n=838)

p

140 (90, 210)

135 (91, 210)

0·5545

150 (101, 220)

145 (100, 215)

0·9391

153 (105, 225)

··

··

263 (213, 339)

255 (200, 335)

0·7178

115 (94, 150)

107 (85, 140)

0·7042

104 (82, 135)

··

··

Data are median (Q1, Q3).

Table 2: Time intervals between onset of symptoms, randomisation, and the catheterisation procedure

oxygen and in the absence of known lung disease. The stroke committee of previous ASSENT trials, blinded to treatment assignment, reviewed all of the stroke cases reported up to day 90 in this study. Major clinical endpoints, such as reinfarction, refractory ischaemia, and non-cerebral bleeding complications were not adjudicated, but definitions were provided that were identical to those used in previous ASSENT trials.26,27 We defined reinfarction in the first 18 h as recurrent symptoms of ischaemia at rest accompanied by new or recurrent ST-segment elevations of 0·1 mV or more in at least two contiguous leads, lasting for at least 30 min. After 18 h, the definition was new Q waves in two or more leads or further increases in concentrations of creatine kinase MB, troponins, or total creatine kinase above the upper limit of normal and higher than the previous value. We defined non-cerebral bleeding complications as major (requiring transfusion or intervention, or both, because of haemodynamic compromise) or minor.

Statistical analysis Our primary aim was to show a significant reduction in the primary endpoint in patients randomised to facilitated PCI as compared with those assigned primary PCI. Assuming an incidence of the primary endpoint of 16% in the primary PCI group, we needed a sample size of 2000 patients per group to show (by means of a 2 test) a 22% relative or a 3·5% absolute reduction with facilitated PCI with 85% power at a two-sided  level of 5%. Baseline characteristics are reported as mean (SD) or number (%) as appropriate. Time differences are reported as medians (lower [Q1] and upper [Q3] quartiles). We calculated relative risks with two-sided asymptotic 95% CIs and p values of a Fisher’s Exact test on proportions for all single and combined endpoints. For the primary endpoint we also undertook prespecified subgroup analyses—age, sex, Killip class, time delays, infarct location, location of first medical contact, history www.thelancet.com Vol 367 February 18, 2006

Before catheterisation Tenecteplase Bolus heparin Glycoprotein IIb/IIIa inhibitor During catheterisation Glycoprotein IIb/IIIa inhibitor Abciximab Eptifibatide Tirofiban Clopidogrel or ticlopidine Additional heparin After catheterisation Clopidogrel or ticlopidine Oral anticoagulants Low molecular weight heparins Fibrinolytics other than study drug Heparin Glycoprotein IIb/IIIa inhibitor Diuretics (oral) Diuretics (intravenous)  blockers Angiotensin-converting enzyme inhibitors Angiotensin receptor blockers Statins Positive inotropic agents

Tenecteplase and PCI (n=829)

PCI alone (n=838)

p

818/829 (99%) 767/821 (93%) 2/825 (1%)

·· 784/829 (95%) 25/834 (3%)

·· 0·3516 0·0001

78/819 (10%) 49/78 (63%) 10/78 (13%) 19/78 (24%) 517/818 (63%) 551/817 (67%)

417/827 (50%) 272/417 (65%) 32/417 (8%) 113/417 (27%) 526/828 (64%) 579/825 (70%)

0·0001

716/819 (87%) 61/818 (7%) 392/818 (48%) 7/819 (1%) 409/819 (50%) 108/819 (13%) 154/797 (19%) 160/792 (20%) 676/804 (84%) 602/801 (75%) 41/795 (5%) 696/801 (87%) 92/793 (12%)

754/832 (91%) 63/831 (8%) 398/831 (48%) 15/831 (2%) 381/831 (46%) 414/831 (50%) 112/814 (14%) 139/815 (17%) 701/820 (85%) 628/816 (77%) 30/811 (4%) 741/821 (90%) 87/816 (11%)

0·0405 0·9260 0·9999 0·1314 0·1039 0·0001 0·0031 0·1093 0·4476 0·4146 0·1817 0·0350 0·5792

0·9185 0·2413

Data are number (%). Calculations based on available data.

Table 3: Medications given before, during, and after catheterisation

of diabetes, infarction, PCI or coronary bypass surgery, stent replacement, and country (European vs nonEuropean). We checked the interactions between treatment and subgroups. For the primary endpoint and mortality, we present Kaplan-Meier curves and results of log-rank tests. All analyses were by intention to treat and done with SAS (version 8.02). We did not foresee any imputation of missing data or interim analyses. However, we planned to enrol 4000 patients, but early cessation of enrolment was recommended by the data and safety monitoring board because of a higher inhospital mortality in the facilitated than in the standard PCI group (6% [43 of 664] vs 3% [22 of 656] ; p=0·0105). This study is registered with ClinicalTrials.gov, number NCT00168792.

Before PCI TIMI 0 TIMI 1 TIMI 2 TIMI 3 Not assessable After PCI TIMI 0 TIMI 1 TIMI 2 TIMI 3 Not assessable

Tenecteplase and PCI (n=812 before and n=719 after PCI)

PCI alone (n=821 before and n=763 after PCI)

196/812 (24%) 83/812 (10%) 172/812 (21%) 353/812 (43%) 8/812 (1%)

512/821 (62%) 70/821 (9%) 107/821 (13%) 124/821 (15%) 8/821 (1%)

15/719 (2%) 15/719 (2%) 55/719 (8%) 631/719 (88%) 3/719 (1%)

13/763 (2%) 4/763 (1%) 68/763 (9%) 677/763 (89%) 1/763 (1%)

p 0·0001

0·0471

Data are number (%). Calculations based on available data.

Table 4: TIMI flow grades of participants

571

Articles

Proportion with primary endpoint (%)

20

Facilitated PCI Primary PCI Death Congestive heart failure Shock Reinfarction Repeat target vessel revascularisation Rehospitalisation for congestive heart failure Rehospitalisation for shock Rehospitalisation for other cardiac reason

15

10 Logrank test: p=0·0042

5

0 0

10

20

30

40

50

60

70

80

90

Tenecteplase and PCI (n=829)

PCI alone (n=838)

55/823 (7%) 97/807 (12%) 51/807 (6%) 49/805 (6%)

41/831 (5%) 75/818 (9%) 39/817 (5%) 30/820 (4%)

0·1412 0·0640 0·1933 0·0279

53/805 (7%)

28/818 (3%)

0·0041

15/807 (2%)

11/818 (1%)

0·4356

0/807 83/806 (10%)

1/817 (1%) 90/819 (11%)

p

0·9999 0·6878

Data are number (%). Calculations based on available data.

Table 5: Single clinical endpoints within 90 days of randomisation

Time since randomisation (days) Number at risk Facilitated PCI Primary PCI

829 838

703 747

696 741

691 736

685 730

678 726

675 725

673 724

673 724

672 722

Figure 2: Kaplan-Meier curves for primary endpoint

Role of the funding source The sponsors participated in the discussions on study design and study logistics. They had no role in data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results 1667 patients were enroled, of whom 334 (20%) were first treated in ambulances, 578 (35%) in community hospital without catheterisation facilities, and 749 (45%) in hospitals with catheterisation facilities. Figure 1 shows the trial profile and table 1 the baseline characteristics of the two groups of participants. Table 2 shows the 20

Proportion death (%)

Facilitated PCI Primary PCI 15

Logrank test: p=0·1187

10

5

0 0

10

20

30

40

50

60

70

80

90

Time since randomisation (days) Number at risk Facilitated PCI

829

785

780

778

772

770

768

768

768

768

Primary PCI

838

815

810

805

799

796

794

793

792

790

Figure 3: Kaplan-Meier curves for mortality

572

different time intervals between onset of symptoms, randomisation, and the catheterisation procedure. There were no great differences in baseline characteristics between the two groups. The median time delay between randomisation and the first balloon inflation was 115 min in the facilitated PCI group and 107 min in the primary PCI group. The median time interval between symptom onset and bolus tenecteplase was 153 min and an additional 104 minutes expired before first balloon inflation in the facilitated PCI group. The median time delay between symptom onset and PCI was more than 4 h in both groups. Table 3 shows the medications given before, during, and after the catheterisation procedure. A similar proportion of patients received additional heparin bolus and clopidogrel in both groups. The use of glycoprotein IIb/IIIa inhibitors was significantly higher, as expected, in the primary PCI group. Table 4 shows TIMI flow grades. Significantly more patients in the facilitated PCI group than in the standard PCI group had an open infarct artery on arrival in the catheterisation laboratory. A PCI of the culprit lesion was done in 762 patients randomised to primary PCI, of whom 714 (94%) received a stent. 719 patients assigned facilitated PCI underwent a PCI with stent placement in 667 (93%). The proportions of drug eluting stents used were similar: 22% (149 of 667) in the facilitated and 23% (162 of 714) in the primary PCI group. After PCI, patency rates were slightly higher in the primary PCI group, but TIMI flow grade 3 rates were very similar (table 4). Figure 2 and figure 3 show the Kaplan-Meier curves for the primary endpoint and for mortality. The individual components of this primary endpoint at 90 days together with other major clinical endpoints within 90 days are listed in table 5. Although we did not note a significant difference for any of the individual components, the combined endpoint of death, congestive heart failure, or shock at 90 days was significantly higher in the facilitated PCI group (p=0·0045). Figure 4 and figure 5 show results of prespecified subgroup analyses of the primary endpoint and 90-day mortality. There was consistency of www.thelancet.com Vol 367 February 18, 2006

Articles

Mortality, shock, or congestive heart failure within 90 days

Facilitated PCI n/N (%)

Primary PCI n/N (%)

Relative risk (95% CI)

Overall event rate Age (years) 75 75 Sex Male Female Killip class I II-III IV Time from symptoms onset to randomisation (h) 0–2 2–4 4 Time from symptoms onset to 1st balloon inflation (h) 4 4 Time from randomisation to 1st balloon inflation (h) 2 2 Infarct location Anterior Other First medical contact PCI-hospital Non-PCI hospital/ambulance Ambulance Diabetes Yes No Previous infarction Yes No Previous PCI Yes No Previous coronary artery bypass graft Yes No Stent placement Yes No Country European Non-European

151/810 (18·6%)

110/819 (13·4%)

1·39 (1·11–1·74)

113/714 (15·8%) 38/96 (39·6%)

76/716 (10·6%) 34/103 (33·0%)

1·49 (1·14–1·96) 1·20 (0·83–1·74)

93/620 (15·0%) 58/190 (30·5%)

81/637 (12·7%) 29/182 (15·9%)

1·18 (0·89–1·56) 1·92 (1·29–2·85)

125/736 (17·0%) 21/64 (32·8%) 5/10 (50·0%)

91/759 (12·0%) 15/51 (29·4%) 3/8 (37·5%)

1·42 (1·10–1·82) 1·12 (0·64–1·94) 1·33 (0·45–3·96)

56/338 (16·6%) 47/317 (14·8%) 48/155 (31·0%)

39/344 (11·3%) 49/323 (15·2%) 22/151 (14·6%)

1·46 (1·00–2·14) 0·98 (0·68–1·41) 2·13 (1·35–3·34)

32/258 (12·4%) 92/395 (23·3%)

35/312 (11·2%) 51/397 (12·8%)

1·11 (0·70–1·73) 1·81 (1·33–2·48)

55/364 (15·1%) 69/289 (23·9%)

52/450 (11·6%) 34/260 (13·1%)

1·31 (0·92–1·86) 1·83 (1·25–2·66)

98/393 (24·9%) 53/416 (12·7%)

70/382 (18·3%) 40/437 (9·2%)

1·36 (1·04–1·79) 1·39 (0·94–2·05)

67/373 (18·0%) 84/435 (19·3%) 31/161 (19·3%)

40/362 (11·0%) 70/454 (15·4%) 23/165 (13·9%)

1·63 (1·13–2·34) 1·25 (0·94–1·67) 1·38 (0·84–2·26)

37/140 (26·4%) 114/669 (17·0%)

27/126 (21·4%) 82/692 (11·8%)

1·23 (0·80–1·90) 1·44 (1·11–1·87)

23/105 (21·9%) 126/698 (18·1%)

20/86 (23·3%) 89/730 (12·2%)

0·94 (0·56–1·60) 1·48 (1·15–1·90)

16/67 (23·9%) 135/733 (18·4%)

10/65 (15·4%) 99/747 (13·3%)

1·55 (0·76–3·17) 1·39 (1·09–1·76)

4/18 (22·2%) 147/788 (18·7%)

6/15 (40·0%) 103/800 (12·9%)

0·56 (0·19–1·61) 1·45 (1·15–1·83)

113/655 (17·3%) 36/150 (24·0%)

78/700 (11·1%) 32/118 (27·1%)

1·55 (1·18–2·02) 0·89 (0·59–1·33)

92/527 (17·5%) 59/283 (20·8%)

75/527 (14·2%) 35/292 (12·0%)

1·23 (0·93–1·62) 1·74 (1·18–2·56)

Facilitated PCI better

Facilitated PCI worse

0·5

1·5

1·0

2·0

2·5

Figure 4: Relative risks (95% CI) for primary endpoint in total population and in subgroups

the primary endpoint result across most subgroups. However, we did note a significant treatment interaction for stent placement (p=0·0376). We noted no significant treatment interactions for 90-day mortality. Mortality rates in the facilitated PCI group were lowest in patients first treated in ambulances and highest in those first treated in PCI centres. We noted more ischaemic cardiac complications, such as reinfarction and repeat target vessel revascularisations, in the facilitated PCI group than in the primary PCI group and most arose during the first day. Table 6 shows the number of strokes and major bleeding complications in the two study groups up to www.thelancet.com Vol 367 February 18, 2006

day 90. We noted a significant difference in total and haemorrhagic stroke between the two groups. No ischaemic strokes but one haemorrhagic stroke (after discharge) were reported in patients randomised to primary PCI. There was a slight excess of major noncerebral bleeding complications and blood transfusions in patients assigned to facilitated PCI, but the differences were not significant. Table 7 lists the causes of death. None of the 55 deaths that occurred in the facilitated PCI group was due to non-cerebral bleeding complications, but eight were due to stroke (four intracranial haemorrhages). 573

Articles

Mortality within 90 days

Facilitated PCI n/N (%)

Primary PCI n/N (%)

Relative risk (95% CI)

Overall event rate Age (years) 75 75

55/823 (6·7%)

41/831 (4·9%)

1·35 (0·91–2·01)

37/726 (5·1%) 18/97 (18·6%)

23/725 (3·2%) 18/106 (17·0%)

1·61 (0·96–2·68) 1·09 (0·60–1·98)

31/631 (4·9%) 24/192 (12·5%)

31/642 (4·8%) 10/189 (5·3%)

1·02 (0·63–1·65) 2·36 (1·16–4·80)

37/748 (4·9%) 14/65 (21·5%) 4/10 (40·0%)

26/768 (3·4%) 11/52 (21·2%) 3/8 (37·5%)

1·46 (0·89–2·39) 1·02 (0·51–2·05) 1·07 (0·33–3·45)

20/344 (5·8%) 15/321 (4·7%) 20/158 (12·7%)

14/347 (4·0%) 15/331 (4·5%) 12/152 (7·9%)

1·44 (0·74–2·81) 1·03 (0·51–2·07) 1·60 (0·81–3·17)

14/260 (5·4%) 28/402 (7·0%)

13/313 (4·2%) 17/403 (4·2%)

1·30 (0·62–2·71) 1·65 (0·92–2·97)

22/369 (6·0%) 20/293 (6·8%)

21/454 (4·6%) 9/263 (3·4%)

1·29 (0·72–2·31) 1·99 (0·92–4·30)

38/399 (9·5%) 17/423 (4·0%)

27/386 (7·0%) 14/445 (3·1%)

1·36 (0·85–2·19) 1·28 (0·64–2·56)

32/379 (8·4%) 23/442 (5·2%) 5/163 (3·1%)

19/365 (5·2%) 22/463 (4·8%) 7/170 (4·1%)

1·62 (0·94–2·81) 1·10 (0·62–1·94) 0·74 (0·24–2·30)

17/143 (11·9%) 38/679 (5·6%)

11/131 (8·4%) 29/699 (4·1%)

1·42 (0·69–2·91) 1·35 (0·84–2·16)

9/107 (8·4%) 45/709 (6·3%)

9/88 (10·2%) 31/740 (4·2%)

0·82 (0·34–1·98) 1·52 (0·97–2·37)

6/69 (8·7%) 49/744 (6·6%)

6/67 (9·0%) 34/757 (4·5%)

0·97 (0·33–2·86) 1·47 (0·96–2·24)

1/18 (5·6%) 54/801 (6·7%)

3/15 (20·0%) 37/812 (4·6%)

0·28 (0·03–2·40) 1·48 (0·99–2·22)

37/663 (5·6%) 16/153 (10·5%)

27/708 (3·8%) 14/121 (11·6%)

1·46 (0·90–2·38) 0·90 (0·46–1·78)

30/534 (5·6%) 25/289 (8·7%)

27/534 (5·1%) 14/297 (4·7%)

1·11 (0·67–1·84) 1·84 (0·97–3·46)

Facilitated PCI better

Facilitated PCI worse

Sex Male Female Killip class I II-III IV Time from symptoms onset to randomisation (h) 0–2 2–4 4 Time from symptoms onset to 1st balloon inflation (h) 4 4 Time from randomisation to 1st balloon inflation (h) 2 2 Infarct location Anterior Other First medical contact PCI-hospital Non-PCI hospital/ambulance Ambulance Diabetes Yes No Previous infarction Yes No Previous PCI Yes No Previous coronary artery bypass graft Yes No Stent placement Yes No Country European Non-European

0·5

1·0

1·5

2·0

2·5

Figure 5: Relative risks (95% CI) for mortality alone in the total population and in subgroups

Discussion Our findings indicate that full-dose tenecteplase given 1–3 h before planned PCI results in a worse outcome than PCI alone. This finding was evident even though participants in the facilitated PCI group had a higher infarct artery patency rate than those in the standard PCI group. Results of previous studies of routine early PCI after fibrinolytic therapy also indicate no convincing clinical benefit of facilitated PCI over standard PCI, but most were done before the introduction of stents, low-profile 574

guiding catheters, steerable wires, and smaller-sized sheaths, and before glycoprotein IIb/IIIa inhibitors and thienopyridines became available.9–12 Suboptimum antithrombotic co-therapy in the facilitated PCI arm could in part explain the worse clinical outcome noted in these patients. The absence of an infusion of heparin after bolus administration and of an up-front loading dose of clopidogrel coupled with prohibition to use glycoprotein IIb/IIIa inhibitors except in bailout situations, might have contributed to a lower than expected infarct artery patency rate on the www.thelancet.com Vol 367 February 18, 2006

Articles

one hand and an excess of early thrombotic complications, mitigating the benefit of earlier reperfusion by tenecteplase, on the other. Although caution is required in view of the multiple subgroups tested, the significant interaction between treatment and stent placement raises the possibility that stent placement in a prothrombotic, post-fibrinolytic environment,28 especially in the absence of optimum antithrombotic co-therapy, caused early thrombotic complications. When designing the trial, we were concerned about the intracranial and other bleeding risks, and we kept the pre-PCI pharmacological treatment as simple as possible to facilitate and accelerate transport to the catheterisation laboratory. No data on the efficacy and safety of a loading dose of clopidogrel given together with fibrinolytic agents was available at that time and the combination of a full dose of tenecteplase with a glycoprotein IIb/IIIa inhibitor was (and still is) considered to carry an unacceptable risk of bleeding complications.29,30 We asked that no infusion of heparin be given after the bolus administration, since we assumed a single bolus of 60–70 U/kg would provide sufficient anticoagulation for 1–3 h and since an additional bolus of heparin could be given according to the activated clotting time measured on arrival at the catheterisation laboratory. Treatment delays might also provide an insight into our unexpected results. First, the median time between bolus tenecteplase and PCI was only 104 min. Assuming a delay of about 30–45 min between injection and recanalisation of the infarct vessel, the time gain for reperfusion with this treatment was probably short in many patients. Second, the bolus tenecteplase was given a median of 2·6 h after the onset of symptoms; hence, most patients received the drug after 2 h. In these patients, the benefit of an earlier recanalisation of the infarct vessel by tenecteplase is probably attenuated, even in the absence of a thrombotic or ischaemic complication.31,32 The reduced mortality with facilitated PCI in ambulance patients and the much higher mortality with this treatment in PCI centres can be partly explained by the treatment delays discussed above. The expected and higher stroke rates in the facilitated PCI group (although very similar to those in previous ASSENT trials) also contributed to the negative outcome. Stroke was the primary cause of death in eight of 55 patients who died before day 90 and were assigned facilitated PCI. The absence of a stroke in the primary PCI group during admission further enhanced the contrast between the two study groups. Non-cerebral bleeding complications did not contribute to the negative outcome of the study. As in two more recent studies with full-dose tenecteplase13,14 followed by PCI, a significantly higher rate of major bleeding complications or need for blood transfusion was not noted, possibly because of the high degree of fibrin specificity of tenecteplase. Finally, particularly related to www.thelancet.com Vol 367 February 18, 2006

Tenecteplase and PCI (n=829) In-hospital stroke Intracranial haemorrhage Primary ischaemic stroke Unclassified stroke Total Stroke after discharge (up to day 90) Intracranial haemorrhage Primary ischaemic stroke Unclassified stroke In-hospital bleeding complications Major Minor Blood transfusions

PCI alone (n=838)

8 (1·0%) 5 (0·6%) 2 (0·2%) 15 (1·8%)

p

0 0 0 0

0·0037 0·0302 0·2472 0·0001

1 (0·1%) 4 (0·5%) 2 (0·2%)

1 (0·1%) 0 0

0·9999 0·0602 0·2466

46 (5·6%) 210 (25·3%) 48 (6·2%)

37 (4·4%) 159 (19·0%) 33 (4·2%)

0·3118 0·0021 0·0873

Data are number (%).

Table 6: Strokes and non-cerebral bleeding complications within 90 days

early termination, our results could have been in part due to the play of chance. Our study has several limitations. The most important one is its open-label design. However, major clinical endpoints besides death were either centrally adjudicated by an event committee blinded to the allocated treatment, or stringent definitions for reporting were provided. The reporting of TIMI flow rates by investigators is another limitation.We intended to do a central blinded analysis of angiograms from 600 patients in a substudy. Because of the premature termination of the main study, and its unexpected results, a core laboratory will now analyse all available angiograms. In conclusion, we cannot recommend a strategy of routine PCI preceded by full-dose tenecteplase with antithrombotic co-therapy as used in this study. Indeed, this combined therapy could be harmful. Whether such a strategy could be beneficial with an up-front loading dose of a thienopyridine, with better antithrombin cotherapy, in patients treated earlier (eg, in ambulances within 2 h of onset of symptoms), with longer delay to PCI, or with a more selective use of PCI (eg, only after no effect with fibrinolysis) is unknown and should be studied. Finally, full-dose fibrinolytic therapy is only

Reinfarction Cardiogenic shock Arrhythmia or sudden death Asystole or cardiac arrest Cardiac rupture or electromechanical dissociation Stroke or intracranial haemorrhage Other cardiac event Other non-cardiac event

Tenecteplase and PCI (n=55)

PCI alone (n=40)*

4 22 1 6 8

4 17 3 5 5

8

0

1 5

3 3

*Cause of death missing for one patient assigned primary PCI.

Table 7: Causes of death

575

Articles

one of the possible pre-PCI pharmacological therapies under study. As such, we eagerly await the results of the ongoing Facilitated Intervention with Enhanced Reperfusion to Stop Events (FINESSE) trial,33 in which up-front abciximab alone or in combination with halfdose reteplase are compared with standard PCI. ASSENT-4 PCI investigators Executive committee—F Van de Werf (chairman), Leuven, Belgium; A Ross (co-chairman), Washington, DC, USA; P Armstrong, Edmonton, Canada; C Granger, Durham, USA. Steering committee—F Van de Werf (chair), Belgium; A Ross (co-chair), USA; R Diaz, Argentina; P Aylward, Australia; K Huber, Austria; L Soares Piegas, Brazil; P Armstrong, Canada; RL Gao, China; P Widimsky, Czech Republic; A Vahanian, France; P Goldstein, France; U Zeymer, Germany; C Stefanadis, Greece; A Rónaszéki, Hungary; D Sugrue, Iceland; S Iliceto, Italy; D Antoniucci, Italy; YS Jang, South Korea; LA Elizondo Sifuentes, Mexico; F Verheugt, Netherlands; H White, New Zealand; R Fanebust, Norway; W Ruzyllo, Poland; R Seabra-Gomes, Portugal; A Dalby, South Africa; C Macaya Miguel, Spain; O Hess, Switzerland; FT Chiang, Taiwan; G Veerakul, Thailand; C Erol, Turkey; AS Binbrek, United Arab Emirates; J Adgey, UK; MA DeBelder, UK; C Granger, USA; P Berger, USA; M Gibson, USA; C Grines, USA. Data safety and monitoring board—K Fox (chairman), Edinburgh, UK; R Brower, Rotterdam, Netherlands; AP Maggioni, Milano, Italy; J Tijssen, Amsterdam, Netherlands; P de Feijter, Rotterdam, Netherlands; D Weaver, Detroit, USA. Cardiac event committee—P Sinnaeve (chair), C Dubois, W Droogne, J Van Cleemput, Leuven, Belgium. Core laboratories—ECG: Paul Armstrong, Edmonton, Canada; angiography: Walter Desmet, Leuven, Belgium. Stroke review panel—W Hacke, Heidelberg, Germany; R von Kummer, Dresden, Germany. Statistical analysis committee—Biostatistical Centre, Leuven, Belgium: E Lesaffre, K Bogaerts, A Belmans, S Cecere, D Rizopoulos, R Tsonaka, and E Bluhmki (Boehringer Ingelheim). Global study coordination—Leuven Coordinating Centre, Leuven, Belgium: M Moreira (project manager), R Brower (data safety and monitoring board liaison), C Luys, L Goffin, K Broos, H De Tollenaere, P Stevens. Operations committee—Boehringer Ingelheim, Ingelheim, Germany (H Sarelin, T Danays, E Bluhmki, J Kaye, G Goetz), and Leuven Coordinating Centre, Leuven, Belgium (M Moreira). Countries and investigators Australia (65 patients)—P Tideman, C Aroney, G Cope, Y Nagree, A Goudie, J French, N Jepson, J Lefkovits, G New, D Ashby, R Waltham. Austria (31 patients)—K Huber, J Slany, W Benzer, T Brunner, H Mayr, D Glogar, G Winder, M J Scheier, D Striberski, H Schmied, G Röggla, H Frank, W Schreiber. Belgium (144 patients)—W Desmet, P Sinnaeve, C Dubois, H Celen, U Van Walleghem, L De Wolf, D Temmerman, K Dujardin, R Haspeslagh, M Vrolix, G Jeurissen, A Van Dorpe, J Tack, C Vrints, I Bekaert, Y Taeymans, P Gheeraert, J Geraedts, S De Maeseneire, M Rosseel, J Van Besien, F Charlier, J Beys, M Herssens, C Van Haesendonck, L Janssens, L Missault, P Willems, P Van Iseghem, L Muyldermans, M Vincent, K Erard, O De Coster, D Djian, T Eeman, G Mazairac, and the staff of the SMUR CHR Namur, C Laruelle, J Lalmand, F De Man, P Vandergoten, P Merveille, I Liebens, V Legrand, C Berthe, D El Allaf, P Lefebvre, M Carlier, C Bastin, P Dermine, D Dourte, S Dietrich, E El Khouny. Brazil (208 patients)—AC Palandri Chagas, AC Carvalho, A de Pádua Mansur, E Ribeiro da Silva, R Bicudo Ramos, J Fabri Jr, J Takada, E Silva dos Santos, LA Piva Mattos, EA Pelloso, ER Fernandes Manenti, O Pereira Dutra, PE Leães, N Toazza Duda, R Tadeu Tumelero, AP Morales Kormann, R Maestri, PR Ferreira Rossi, ÊE Guérios, CR Frack Costantini, AJ Muniz, RL Marino, MA Marino, P Roussef, MA Neves Rati, RA Bento Ribeiro, PA Marra da Motta, JA Jatene, J Klauber Roger Carneiro Marglecia, M Vasconcelos Coutinho,

576

M Sanali Moura de Oliveira Paiva, A Martins Pacheco Filho, R D´Aurea Mora Jr. Canada (57 patients)—P Costi, P Klinke, C Wells, N Joanis, R Zimmermann, B Wong, E Schampaert, H Comtois, R Larochelliere, M DeGrace. Czech Republic (24 patients)—J Dvorak, P Formanek, M Zelizko, M Aschermann, A Herman, J Vojacek, Z Felix, D Vencour, J Povolny, K Sochor, A Mandovec, E Mandakova, N Turkova, M Sofr, J Kotous, L Kvasnickova, P Kos, P Stepanek, P Stepanek, J Malek, P Sabl, J Vesely, G Marcinek. France (209 patients)—J-P Auffray, M Aussedat, J-J Bauchart, P Beaufils, P Benatia, P Benoit, A Bonneau, J-L Bonnet, J Boschat, C Boureux, P Carli, J Cassagnes, S Cattan, N Danchin, J-E De la Coussaye, F Dissait, J-L Dubois-Rande, F Dupas, E Ferrari, F Funck, Y Garcia, P Goldstein, Y Gottwalles, J-M Haegy, K Khalife, G Kirkorian, F Lapostolle, D L’Azou, F Leclercq, M Lopez, P Marcollet, A Margenet-Baudry, P Messner-Pellenc, H Milon, J-M Montely, M-C Morice, F Penit, P Petit, J Ponsonnaille, C Pouges, V Raphaël, P Rioux, P Sauval, L Soulat, C Spaulding, X Tabone, A Vahanian. Germany (80 patients)—HR Arntz, S Behrens, FL Bertschat, M Borggrefe, J vom Dahl, HJ Ehrens, K Ellinger, HR Figulla, R Füller, O Göing, M Gross, J Grossmann, M Haass, T Kolbe, H Krieter, W Lengfelder, J Münch, H Nebelsieck, D Nowzohour, G Oltmanns, A Rottmann, A Schmidt, G Schuler, W Segiet, M Sigmund, R Simon, J Stachowitz, G Steinbeck, R Strasser, H Thiele. Greece (120 patients)—JN Nanas, J Kanakakis, S Rokas, N Agrios, G Maroulidis, E Tsolakis, J Venetsanakos, AS Manolis, S Koulouris, V Tsagou, J Zarifis, GE Kochiadakis, G Mertzanos, P Harbis, E Scoubourdis, K Kalaitzaki, G Lappas, P Sinnis, D Papapetrou, A Dalianis, G Papaioannou. Hungary (22 patients)—B Merkely, D Becker, M Keltai, R Kiss, K Zamolyi, B Merkely, I Edes, G Lupkovics, T Forster, I Horvath, G Veress, A Kollar, C András Dézsi, G Csokor, M Sereg, K Heltai, A Rónaszéki, L Jobbagy, L Regos, C Dioszeghy, Z Pecsvarady, J Dinnyes, I Szakal, B Oze, T Berenyi, P Polgar, E Kalo, C Toth, A Matoltsy, J Dobos, A Kovács, A Nagy, B Nyúzó, A Pálinkás, M Józan-Jilling, F Magel, G Kutor, M Bela, D Becker. Ireland (7 patients)—P Crean, J Crowley. Italy (83 patients)—F Mascia, A Vetrano, U Bugatti, A Gigantino, P Lisanti, R Spirito, C Fresco, R Copetti, G Zur, A Fiscella, J Salerno Uriarte, E Varani, G Andrea Amadei, S Bongo, S Iliceto, D Marchese, R Zanini, M Flavio Pasqualini, D Antoniucci, G Maria Santoro, N De Cesare, S Pirelli, O Ferrari, C Bonifazi. Mexico (24 patients)—LA Elizondo-Sifuentes, R De la Cruz-Obregón, R Arriaga-Nava, A Castro-Gutierrez, I Hernández-Santamaría. Netherlands (44 patients)—RJ de Winter, JGM Tans, AWJ van ’t Hof, JG Engbers, HR Michels, JG Meeder, CJPJ Werter, HMP Wilbers-van Rens, FWA Verheugt. New Zealand (1 patient)—G Wilkins. Norway (25 patients)—K Endresen, M Ljosland, A Mangschau, E Cesilie Knudsen, O Jonas Rolstad, K Andersen, H Schartum-Hansen, JE Nordrehaug, HO Fadnes, F Bergo. Poland (272 patients)—M Goszczynska, R Sciborski, W. Ruzyllo, R Gil, G Opolski, K Loboz-Grudzien, T Widomska-Czekajska, J.D. Kasprzak, A Cieslinski, M Tendera, M Trusz-Gluza, L Polonski, K Zmudka, L Bryniarski, W Pluta, Z Gasior, A Kania, E Jankowski, E Dziuba-Tuszynska, M Ogorek, R Szyda, M Kazmierski, W Wozniak, M Hamankiewicz, B Pankiewicz, Z Bryndal, J Dosiak, J Frycz, M Krauze-Wielicka, A Wojtas, K Bujak, P Maslany, M Markowska-Kuzniak, W Koczubik, A Jankowski, M Kuczera, T Reginek, A Straszak, G Muskala, G Okularczyk, P Krawczuk, I Babiak, E Kluczewska, D Wojciechowski, S Niemczyk, M Regulski, P Kolodziej, K Makowiecki, E Komorowska, B Zbyszynski, M Stopinski, J Socha, R Malycha, T Sebzda, J Dubejko, M Skrzynska, F Monies, Z Kalarus, M Karcz, M Wojtowicz, L Sokalski, M Sledziona, M Gasior, P Trzeciak, M Krzeminska-Pakula, M Kisielczyck, K Wilczek, M Lesiak, S Grajek, K Kawecka-Jaszcz, J Dragan, A Wnuk-Lipinska. Portugal (6 patients)—H Horta Pereira, R Menezes Caria, A Leitão Marques, L Santarém Semedo. South Africa (39 patients)—AJ Dalby, V Singh, J King, D Kettles, JJ Patel, A Horak, TA Mabin, C Corbett, M Sender, R Routier, R Lamparelli, C Schamroth, G Ignjatovic, J Roos, W Bodenstein. www.thelancet.com Vol 367 February 18, 2006

Articles

South Korea (45 patients)—Seung-Yun Cho, Joo Young Yang, Tae Hoon Ahn, Joo Young Yang, Moo-Yong Lee, Hyeon Cheol Gwon, Jung Han Yoon, Si Hoon Park, Myung Ho Jeong, Young Joon Hong. Taiwan (2 patients)—Chiou Herng-Cheng. Turkey (76 patients)—S Güneri, Ö Kırımlı, O Badak, A Narin, G Hobiko¨glu, T Norgaz, E Tutar, Ç Özdöl, O Ulaz Özcan. UK (8 patients)—J Adgey, M DeBelder. USA (75 patients)—R Clark, M Dorogy, L Garza, S Khanal, V Marques, G McKendall, B McLaurin, A Niederman, E Rivera, S Sharma, V Singh. Contributors All members of the executive committee of the ASSENT-4 PCI trial were involved in design of the study, enrolment and clinical follow-up of patients, analysis of data, and writing of the report. The steering committee overlooked the progress of the trial. Conflict of interest statement F Van de Werf has received research grants and speaking support from Boehringer Ingelheim, Genentech, Hoffman-LaRoche, Procter and Gamble, Alexion, Arginox, Novartis, and Sanofi Aventis. A M Ross has received research and speaking support from Boehringer Ingelheim, Genentech, and Hoffman-LaRoche. P W Armstrong has received research grants from Boehringer Ingelheim, Sanofi Aventis, Schering, and Hoffman-LaRoche, and honoraria from Hoffman-LaRoche, and Sanofi Aventis. Christopher B Granger has received research grants from AstraZeneca, Procter and Gamble, Sanofi Aventis, Alexion, Novartis, Boehringer Ingelheim, Genentech, and Berlex. Acknowledgments We thank the 1667 patients who agreed to participate; the study coordinators, nurses, and monitors in the participating countries; and Krista Bogaert for secretarial assistance in preparing the report. The study was funded by Boehringer Ingelheim, Ingelheim, Germany, and Genentech, South San Francisco, USA References 1 Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003; 361: 13–20. 2 Van de Werf F, Ardissino D, Betriu A, et al. Management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2003; 24: 28–66. 3 Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients with Acute Myocardial Infarction). Circulation 2004; 110: e82–292. 4 Nallamothu BK, Bates ER, Herrin J, et al. Times to treatment in transfer patients undergoing primary percutaneous coronary intervention in the United States: National Registry of Myocardial Infarction (NRMI)-3/4 analysis. Circulation 2005; 111: 761–67. 5 Steg PG, Bonnefoy E, Chabaud S, et al. Impact of time to treatment on mortality after prehospital fibrinolysis or primary angioplasty: data from the CAPTIM randomized clinical trial. Circulation 2003; 108: 2851–56. 6 Danchin N, Blanchard D, Steg PG, et al. Impact of prehospital thrombolysis for acute myocardial infarction on 1-year outcome: results from the French Nationwide USIC 2000 Registry. Circulation 2004; 110: 1909–15. 7 Widimsky P, Budesinsky T, Vorac D, et al. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction: final results of the randomized national multicentre trial—PRAGUE-2. Eur Heart J 2003; 24: 94–104. 8 Stone GW, Cox D, Garcia E, et al. Normal flow (TIMI-3) before mechanical reperfusion therapy is an independent determinant of survival in acute myocardial infarction: analysis from the primary angioplasty in myocardial infarction trials. Circulation 2001; 104: 636–41. 9 O’Neill WW, Weintraub R, Grines CL, et al. A prospective, placebocontrolled, randomized trial of intravenous streptokinase and angioplasty versus lone angioplasty therapy of acute myocardial infarction. Circulation 1992; 86: 1710–17.

www.thelancet.com Vol 367 February 18, 2006

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

Widimsky P, Groch L, Zelizko M, et al. Multicentre randomized trial comparing transport to primary angioplasty vs immediate thrombolysis vs combined strategy for patients with acute myocardial infarction presenting to a community hospital without a catheterisation laboratory: the PRAGUE study. Eur Heart J 2000; 21: 823–31. Vermeer F, Oude Ophuis AJ, vd Berg EJ, et al. Prospective randomised comparison between thrombolysis, rescue PTCA, and primary PTCA in patients with extensive myocardial infarction admitted to a hospital without PTCA facilities: a safety and feasibility study. Heart 1999; 82: 426–31. Ross AM, Coyne KS, Reiner JS, et al. A randomized trial comparing primary angioplasty with a strategy of short-acting thrombolysis and immediate planned rescue angioplasty in acute myocardial infarction: the PACT trial. J Am Coll Cardiol 1999; 34: 1954–62. Fernandez-Aviles F, Alonso JJ, Castro-Beiras A, et al. Primary versus facilitated percutaneous coronary intervention (tenecteplase plus stenting) in patients with ST-elevated myocardial infarction: the final results of the GRACIA-2 randomized trial. Eur Heart J 2004; 25: 33 (abstr). Le May MR, Wells GA, Labinaz M, et al. Combined angioplasty and pharmacological intervention versus thrombolysis alone in acute myocardial infarction (CAPITAL AMI study). J Am Coll Cardiol 2005; 46: 417–24. van’t Hof AW, Ernst N, de Boer MJ, et al. Facilitation of primary coronary angioplasty by early start of a glycoprotein 2b/3a inhibitor: results of the ongoing tirofiban in myocardial infarction evaluation (On-TIME) trial. Eur Heart J 2004; 25: 837–46. Lee DP, Herity NA, Hiatt BL, et al. Adjunctive platelet glycoprotein IIb/IIIa receptor inhibition with tirofiban before primary angioplasty improves angiographic outcomes: results of the TIrofiban Given in the Emergency Room before Primary Angioplasty (TIGER-PA) pilot trial. Circulation 2003; 107: 1497–501. Mesquita Gabriel H, Oliveira J, Canas da Silva P, et al. Early administration of abciximab bolus in the emergency room improves microperfusion after primary percutaneous intervention as assessed by TIMI frame count: results of the ERAMI trial. Eur Heart J 2003; 24: 543 (abstr). Arntz HR, Schroder JF, Pels K, et al. Prehospital versus periprocedural administration of abcizimab in STEMI: early and late results from the randomised REOMOBILE study. Eur Heart J 2003; 24: 268 (abstr). Zorman S, Zorman D, Noc M. Effects of abciximab pretreatment in patients with acute myocardial infarction undergoing primary angioplasty. Am J Cardiol 2002; 90: 533–36. Cutlip DE, Ricciardi MJ, Ling FS, et al. Effect of tirofiban before primary angioplasty on initial coronary flow and early ST-segment resolution in patients with acute myocardial infarction. Am J Cardiol 2003; 92: 977–80. Martinez-Rios MA, Rosas M, Gonzalez H, et al. Comparison of reperfusion regimens with or without tirofiban in ST-elevation acute myocardial infarction. Am J Cardiol 2004; 93: 280–87. Zeymer U, Zahn R, Schiele R, et al. Early eptifibatide improves TIMI 3 patency before primary percutaneous coronary intervention for acute ST elevation myocardial infarction: results of the randomized integrilin in acute myocardial infarction (INTAMI) pilot trial. Eur Heart J 2005; 26: 1971–77. Giugliano RP, Roe MT, Harrington RA, et al. Combination eptifibatide and reduced-dose tenecteplase versus eptifibatide monotherapy preceding primary PCI: final results of the ADVANCE-MI trial. Circulation 2004; 110: 340 (abstr). Strategies for Patency Enhancement in the Emergency Department (SPEED) Group. Trial of abciximab with and without low-dose reteplase for acute myocardial infarction. Circulation 2000; 101: 2788–94. Kastrati A, Mehilli J, Schlotterbeck K, et al. Early administration of reteplase plus abciximab vs abciximab alone in patients with acute myocardial infarction referred for percutaneous coronary intervention: a randomized controlled trial. JAMA 2004; 291: 947–54. Efficacy and safety of tenecteplase in combination with enoxaparin, abciximab, or unfractionated heparin: the ASSENT-3 randomised trial in acute myocardial infarction. Lancet 2001; 358: 605–13.

577

Articles

27

28

29

30

578

Assessment of the Safety and Efficacy of a New Thrombolytic Investigators. Single-bolus tenecteplase compared with frontloaded alteplase in acute myocardial infarction: the ASSENT-2 double-blind randomised trial. Lancet 1999; 354: 716–22. Eisenberg PR, Sobel BE, Jaffe AS. Activation of prothrombin accompanying thrombolysis with recombinant tissue-type plasminogen activator. J Am Coll Cardiol 1992; 19: 1065–69. Topol EJ. Reperfusion therapy for acute myocardial infarction with fibrinolytic therapy or combination reduced fibrinolytic therapy and platelet glycoprotein IIb/IIIa inhibition: the GUSTO V randomised trial. Lancet 2001; 357: 1905–14. Wallentin L, Goldstein P, Armstrong PW, et al. Efficacy and safety of tenecteplase in combination with the low-molecular-weight heparin enoxaparin or unfractionated heparin in the prehospital setting: the Assessment of the Safety and Efficacy of a New Thrombolytic Regimen (ASSENT)-3 PLUS randomized trial in acute myocardial infarction. Circulation 2003; 108: 135–42.

31

32

33

Gersh BJ, Stone GW, White HD, Holmes DR Jr. Pharmacological facilitation of primary percutaneous coronary intervention for acute myocardial infarction: is the slope of the curve the shape of the future? JAMA 2005; 293: 979–86. Huber K, De Caterina R, Kristensen SD, et al. Pre-hospital reperfusion therapy: a strategy to improve therapeutic outcome in patients with ST-elevation myocardial infarction. Eur Heart J 2005; 26: 2063–74. Ellis SG, Armstrong P, Betriu A, et al. Facilitated percutaneous coronary intervention versus primary percutaneous coronary intervention: design and rationale of the Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events (FINESSE) trial. Am Heart J 2004; 147: E16.

www.thelancet.com Vol 367 February 18, 2006