Author + information
- Received January 14, 1999
- Revision received February 23, 1999
- Accepted March 24, 1999
- Published online July 1, 1999.
- John K French, MB, PhDa,* (, )
- Thomas A Hyde, MBa,
- Hitesh Patel, MBa,
- David J Amos, MBa,
- Stephanie C McLaughlin, PhDa,
- Bruce J Webber, DSR (Diagnostic)a and
- Harvey D White, MB, DSca
- ↵*Reprint requests and correspondence: Dr. John French, Cardiology Department, Green Lane Hospital, Private Bag 92 189, Auckland 1030, New Zealand
The purpose of this study was to determine whether the mortality benefit of intravenous streptokinase administered within 4 h of the onset of acute myocardial infarction is maintained at 12 years, and whether Thrombolysis in Myocardial Infarction (TIMI) flow grades independently influence late survival.
Treatment with reperfusion therapies and achievement of TIMI 3 flow are associated with increased short- and medium-term survival after infarction. Whether infarct artery flow independently influences survival more than five years after infarction is unknown.
The late survival of patients randomized to receive either streptokinase (1,500,000 IU over 30 to 60 min) or a matching placebo within 4 h of symptom onset in 1984–1986 was determined. Angiography was performed in surviving patients at three to four weeks, and TIMI flow grades were assessed blind to randomization and outcomes. The late vital status was determined in 99% of patients.
Patients randomized to receive streptokinase (n = 107) had improved survival compared with those randomized to placebo (n = 112) at five years (84% vs. 70%; p = 0.023) and 12 years (66% vs. 51%; p = 0.022). At five years 94% of patients with TIMI grade 3 flow, 81% of those with TIMI grade 2 flow and 72% of those with TIMI grade 0–1 flow survived (p = 0.005). At 12 years 72% of patients with TIMI 3, 67% of those with TIMI 2 and 54% of those with TIMI 0–1 flow survived (p = 0.023). Multivariate analysis identified the ejection fraction (p = 0.014), exercise duration (p = 0.013) and TIMI 3 flow (p = 0.04 compared with TIMI 0–2 flow) as important factors for five-year survival. At 12 years multivariate predictors of late survival were the ejection fraction (p = 0.006), exercise duration (p = 0.003) and myocardial score (p = 0.013). The end-systolic volume index was similar to the ejection fraction as a predictor of survival at five and 12 years.
The survival benefits of streptokinase persist for 12 years after infarction. TIMI flow at three to four weeks is an independent predictor of five-year survival.
The rationale for the use of thrombolytic therapy in acute myocardial infarction is that rapid reestablishment of coronary blood flow distal to the culprit lesion reduces infarct size, preserves left ventricular function and hence increases survival (1–4). The efficacy of reperfusion achieved by different thrombolytic strategies at 90 min varies (5–8), though by 3 h most thrombolytic regimens achieve 70% to 80% patency of infarct-related arteries, and this degree of patency is maintained at one and three weeks (3,5,9,10).
The completeness of reperfusion of the infarct-related artery has prognostic significance, and Thrombolysis in Myocardial Infarction (TIMI) grade 3 flow (normal flow) in the infarct-related artery (9)has been shown to be associated with a better prognosis at 30 days than TIMI 0–2 flow (no or slow filling) (11). At five years after infarction, patients who had TIMI 2 flow at hospital discharge have been reported to have a similar prognosis to that of patients with an occluded infarct-related artery (12).
The Second International Study of Infarct Survival has reported parallel survival curves at 10 years (13)in patients randomized to receive either streptokinase or placebo, whereas the Western Washington study appeared to show convergence at eight years (14). These studies did not determine the influence of TIMI flow on late survival. We have previously reported the important prognostic influence of infarct-related artery patency for up to five years after a first infarction in patients treated with thrombolytic therapy (15), 23% of whom are included in this study.
In this study we aimed to determine whether improved survival persisted for up to 12 years after randomization to treatment with intravenous streptokinase versus placebo, and whether the TIMI flow grade at three to four weeks influenced late survival.
Consecutive consenting patients aged ≤70 years who presented within 4 h after the onset of chest pain lasting ≥30 min were randomized to receive either intravenous streptokinase (1,500,000 IU over 30 to 60 min) or a matching placebo from August 1984 through August 1986 (3). Eligible patients required ST-segment elevation of ≥1 mm on the admission electrocardiogram in at least two contiguous leads (limb leads or V4–V6), or ≥2 mm in leads V1–V3(16). The exclusion criteria have been previously reported (3). The trial was approved by the local ethics committee, which required written informed consent.
A 48-h infusion of intravenous heparin (1,000 IU/h) was commenced after the infusion of streptokinase or placebo, with adjustment of the activated partial thromboplastin time to 2–2.5 × normal. Aspirin (50 mg) and dipyridamole (400 mg daily) were administered until angiography. Intravenous propranolol (0.1 mg/kg over 5 min) was recommended and was received by 47% of patients. Oral beta-adrenergic blocking agents were commenced on days 2–3 in the absence of contraindications; 66% were receiving this therapy at angiography. Calcium antagonists and nitrates were prescribed for postinfarction angina. After angiography, medication was at the physician’s discretion. Patients were advised to stop smoking, to exercise regularly, to reduce weight and to consume a low fat diet. Exercise testing on the Bruce protocol was performed at three to four weeks.
Biplane ventriculography and coronary angiography were performed three to four weeks after infarction (median 25 days [interquartile range 21 to 32]) unless refractory ischemia necessitated earlier investigation. Ventricular volumes and ejection fractions were calculated and regional wall motion was assessed as previously described (3,17). Infarct-related artery flow was assessed in our core angiographic laboratory (8), blind to treatment and clinical outcome, according to TIMI flow grading (0 = occluded, 1 = penetration of contrast, 2 = slow flow or contrast clearance, 3 = normal filling) (9)and corrected TIMI frame counting (18). The severity of coronary artery disease was assessed according to the Coronary Artery Surgery Study criteria (19)and the myocardial score, which measures the amount of myocardium supplied as well as the severity of coronary stenoses (20). Revascularization was performed for left main coronary artery disease or refractory angina despite medical therapy.
Late follow-up was done either by recall to the clinic, by the family doctor or by telephone or mail. The mortality status of patients was determined in 99.1% at five years and 94.0% at 10 years. End points were verified by source documents. Deaths were classified as cardiac or noncardiac (21). Reinfarction was defined as two of the following: prolonged ischemic chest pain, creatine kinase >2 times normal or development of new Q waves on the electrocardiogram. Late percutaneous or surgical revascularization and hospitalization for other cardiovascular causes were also recorded.
Baseline characteristics are expressed as mean ± SD or percentages. Survival curves were constructed using the Kaplan-Meier method and compared using Mantel and Haenszel log-rank tests, with no imputations for deaths before angiography. Multivariate analysis of survival was performed using Cox proportional hazards regression methods, and the assumption of proportional hazards was satisfied for variables in final models. All factors in Table 4(univariate analysis) were included in the initial models. Continuous variables, including exercise duration, were compared by one-way analysis of variance, and discrete variables were compared by the chi-square test with Yates’ correction where appropriate, with the level of significance being p = 0.05.
Patients were randomized to receive either streptokinase (n = 107) or placebo (n = 112); their baseline characteristics are shown in Table 1. Angiography was performed in 194 patients and 186 underwent exercise testing (Table 2). Eighteen patients (14 placebo) died before angiography, seven patients refused angiography (two for medical reasons) and three patients were not catheterized because of stroke, pneumonia and esophageal rupture, respectively.
During follow-up for a median of 9.9 years (interquartile range 4.7 to 10.8 years), there were 67 cardiac deaths and 45 nonfatal myocardial infarctions. There were 24 noncardiac deaths (13 streptokinase and 11 placebo), of which eight were due to stroke (one placebo and seven streptokinase). Between five and 12 years there were 24 cardiac deaths (14 placebo and 10 streptokinase) and 15 noncardiac deaths (seven placebo and eight streptokinase). At five years 84% of the streptokinase group and 70% of the placebo group survived (p = 0.023), and at 12 years 66% of the streptokinase group and 51% of the placebo group survived (p = 0.022) (Fig. 1). During follow-up, 47 patients (22%) underwent coronary revascularization (40 coronary artery surgery and seven angioplasty). The coronary revascularization rates at 1 and 5 years were 10% (8% streptokinase and 11% placebo) and 14% (13% streptokinase and 14% placebo), respectively. The event-free survival at 12 years (freedom from death, recurrent infarction or revascularization) was 38% (streptokinase 45% and placebo 31%, p = 0.032).
At follow-up there were no differences between the streptokinase and placebo groups in terms of angina class (0.9 ± 1.1 vs. 0.8 ± 1.0; p = 0.58), aspirin therapy (66% vs. 63%; p = 0.8) or beta-blocker therapy (48% vs. 45%; p = 0.9). There were no differences in these variables with respect to TIMI flow grades at three to four weeks (not shown).
Thrombolysis in myocardial infarction flow and late survival
At five years the survival rate among patients who had undergone angiography was 84%, (TIMI 3 flow 94%, TIMI 2 flow 81%, TIMI 0–1 flow 72%; p = 0.005 [three-way log-rank comparison]), and at 12 years the survival rate was 65% (TIMI 3 flow 72%, TIMI 2 flow 67%, TIMI 0–1 flow 54%; p = 0.024) (Fig. 2). The two-way log-rank comparisons of TIMI 3 versus TIMI 2 flow at five and 12 years were p = 0.045 and p = 0.36, respectively. Of 82 patients with TIMI 3 flow, 36 had corrected TIMI frame counts of ≤27 and 18 had frame counts of >40 (Table 3). Of 49 patients with TIMI 2 flow, 8 had frame counts of ≤27 and 34 had frame counts of >40. Corrected TIMI frame counts at cut points of either 27 or 40 frames did not influence survival in the 131 patients with patent arteries.
Predictors of late death
Univariate predictors of survival at five and 12 years included the end-systolic volume index, ejection fraction, treadmill exercise time (all p < 0.001), TIMI flow grade, myocardial score, randomization to receive streptokinase and no diabetes mellitus (all p < 0.05) (Table 4). On multivariate analysis using Cox proportional hazards models, independent predictors of five-year survival were the ejection fraction (p = 0.014), treadmill exercise time (p = 0.013) and TIMI 3 flow (p = 0.04) (Table 5). At 12 years after infarction the ejection fraction (p = 0.006), treadmill exercise time (p = 0.003) and myocardial score (p = 0.013) were multivariate factors predictive of survival; TIMI flow grade was not an independent predictive factor. The end-systolic volume index was similar to the ejection fraction as a predictor of survival at five and 12 years.
We observed that survival after acute myocardial infarction was improved up to 12 years after randomization to receive intravenous streptokinase compared with placebo. The TIMI flow grade, assessed at three to four weeks, was an independent prognostic factor for five-year survival, and patients with TIMI 2 flow had a survival rate intermediate between that of patients with TIMI 3 and those with TIMI 0–1 flow. In 1994 we reported the prognostic importance at 39 months of a patent infarct-related artery (15)in patients randomized to receive thrombolytic therapy with streptokinase versus either placebo (3)or tissue plasminogen activator (22); 23% of those patients are included in the current study.
Effect of thrombolytic therapy on late survival
The survival curves remained parallel at 12 years, indicating a persisting survival benefit among patients randomized to receive streptokinase (p = 0.02). Other studies have reported a sustained benefit of intravenous streptokinase at up to five years (2,13,14,23), though survival was not improved at eight years in the Western Washington studies of intravenous and intracoronary streptokinase (p = 0.16) (14). The Second International Study of Infarct Survival and Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico studies, which included patients with non–ST-elevation infarction, have shown parallel survival curves to 10 years (24,25), as have data from the Dutch Interuniversity study (26), extending to 10 to 14 years.
Infarct artery flow and late outcome
The five-year mortality benefit of a patent infarct-related artery (15), particularly with TIMI 3 flow, has been extended by the observations reported here. The benefit of TIMI 3 flow was maintained after adjustment for ventricular function (either the end-systolic volume index or ejection fraction) and treadmill exercise time. TIMI 3 flow was associated with increased survival at 12 years on univariate analysis, but not after adjustment for other independent prognostic factors, which may be due to beta error. Also, progression of coronary artery disease in the non–infarct-related arteries (evidenced by the independent influence of the myocardial score on 12-year survival), the effects of revascularization and noncardiac disease in an aging population may all have influenced the prognostic effects of infarct artery flow on survival at 12 years.
In this study, survival at five years for patients with TIMI 3 flow was 94% compared with 91% reported by the European Cooperative Study Group (ECSG) (12). In patients with occluded infarct-related arteries (TIMI 0–1 flow), the five-year survival was 72% compared with 84% in the ECSG study (12), which may in part reflect a more conservative approach to revascularization in New Zealand —10% at one year and 14% at five years, compared with approximately 20% at one year reported by the ECSG (12).
In our study, TIMI 2 flow, assessed in a core laboratory setting, was associated with a survival rate at five and 12 years that was intermediate between that of TIMI 3 and TIMI 0–1 flow. These findings contrast with those reported by the ECSG (12)showing five-year survival benefits of TIMI 3 flow assessed 10 to 22 days after infarction, but not TIMI 2 flow, which tracked with TIMI 0–1 flow. There may be significant differences in TIMI 2 flow assessment between local and core angiographic laboratories (18). Also, the two-year survival of patients who had had TIMI 2 flow at 90 ± 45 min in the Global Use of Strategies to Open Occluded Coronary Arteries trial (27)was not significantly better than that of patients with TIMI 0–1 flow, but those data were not adjusted for baseline risk.
Corrected TIMI frame counting has been reported to provide a more reproducible assessment of infarct-related artery flow than TIMI flow grading does (18). There were insufficient patients with patent arteries (n = 131) to undertake statistical analysis in order to determine whether frame counting was superior to TIMI flow grading.
Half the patients reported here were randomized to receive a placebo infusion on a background of aspirin and heparin therapy. At three to four weeks the patency rate (TIMI 2–3 flow) in these patients was 61%, similar to the 73% patency rate in patients receiving streptokinase. In patients not treated with thrombolytic therapy the infarct-related artery is usually occluded in the first few hours after myocardial infarction (28), but in the subsequent 24 h recanalization may occur either spontaneously or as a result of antithrombotic and antiplatelet therapy (29). The rate of reocclusion of patent arteries is 10% to 15% in the two weeks after myocardial infarction despite regular aspirin therapy (29).
Ventricular function and survival
We have previously shown in non–thrombolytic-treated patients that end-systolic volume was a more powerful predictor of outcome than ejection fraction. Here the end-systolic volume index and the ejection fraction were similar in predicting five- and 12-year survival (3,30). Migrino et al. (31)highlighted the importance of an end-systolic volume index of 40 ml/m2at 90 or 180 min for predicting one-year mortality. Also, there was an interaction between ventricular volume and the frequency of patients achieving TIMI 3 flow, with only about 30% of patients having TIMI 3 flow with volumes above this figure (31). Late revascularization of occluded infarct-related arteries may have beneficial effects on remodeling and end-systolic volume, although this has not been evaluated in randomized trials (32).
The Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico group (33)has reported the influence of exercise test performance on mortality at six months. Here we report that exercise time is independently associated with both five- and 12-year outcomes; the ECSG (12)indicated similar findings at five years. We did not detect any influence of baseline total or high density lipoprotein cholesterol on late outcome. Although all patients received dietary advice, the general use of lipid-modifying drugs after infarction was not widespread during the follow-up period of 1984–1996. The Scandinavian Simvastatin Survival Study (34)was published toward the end of our follow-up, and thus any resulting changes to clinical practice would have had minimal impact.
This study did not determine the effect on late mortality of TIMI flow grades assessed early after symptom onset. The TIMI flow results reported at three to four weeks represent the net effects of reperfusion and reocclusion in patients treated with 50 mg of aspirin and 400 mg of dipyridamole, intravenous heparin for 48 h and intravenous streptokinase or placebo. In this study streptokinase was administered for 30 to 60 min to patients aged ≤70 years presenting within 4 h of symptom onset, though this regimen has not been formally tested in large clinical trials. Streptokinase continued to be used widely even when thrombolytics were provided free in a large clinical trial (35). Aspirin doses of <150 mg daily have not been tested in large clinical trials of myocardial infarction, though antiplatelet therapy with aspirin (50 mg) and dipyridamole (400 mg), as used in this study, has been shown to be beneficial in other manifestations of cardiovascular disease, for example, in reducing stroke and preventing saphenous vein graft occlusion (36,37). However, the Coumadin Aspirin Reinfarction Study (38), which used 80 mg of aspirin in combination with 1 or 3 mg of warfarin, or 160 mg of aspirin alone for five years, found no difference in the subsequent rates of reinfarction, nonfatal ischemic stroke or cardiovascular death between these groups.
Additionally, beta-blocker therapy could be used more frequently than the rates of 47% intravenously and 66% orally used in our patients. There have been changes to medical care after infarction in the last decade, including new thrombolytic therapies, the use of lipid-modifying drugs, angiotensin-converting enzyme inhibition and different rates of revascularization, which may also have influenced the results of this study.
The survival and event-free survival benefits of thrombolytic therapy with intravenous streptokinase are sustained at 12 years. The TIMI flow grade at three to four weeks after myocardial infarction is an important predictor of late outcome, as are ventricular function and exercise duration. Whether outcomes are improved by revascularization or adjunctive medical therapies, such as direct thrombin inhibitors, low molecular weight heparins and glycoprotein IIb/IIIa receptor antagonists, in patients without TIMI 3 flow in the infarct-related artery is unknown.
We thank Patricia Wilson, Mary Denton, Barbara Williams, Loretta Bush and Josie Slack for patient follow-up and data entry, Clarissa Gould-Thorpe and Edie Scadden for secretarial assistance and Dr. Toby Whitlock and Heather Smith for statistical advice.
☆ Supported in part by a research grant from the Health Research Council of New Zealand.
Presented in part at the 71st Annual Scientific Sessions of the American Heart Association, Dallas, Texas, November 1998.
- European Cooperative Study Group
- Thrombolysis in Myocardial Infarction
- Received January 14, 1999.
- Revision received February 23, 1999.
- Accepted March 24, 1999.
- American College of Cardiology
- French J.K,
- Hyde T.A,
- White H.D
- GUSTO Angiographic Investigators
- Smalling R.W,
- Bode C,
- Kalbfleisch J,
- et al.
- Bode C,
- Smalling R.W,
- Berg G,
- et al.
- White H.D,
- Aylward P.E,
- Frey M.J,
- et al.
- Chesebro J.H,
- Knatterud G,
- Roberts R,
- et al.
- Simes R.J,
- Topol E.J,
- Holmes D.R Jr.,
- et al.
- Lenderink T,
- Simoons M.L,
- van Es G.-A,
- et al.
- ISIS Collaborative Group,
- Baigent C,
- Collins R
- Cerqueira M.D,
- Maynard C,
- Ritchie J.L,
- Davis K.B,
- Kennedy J.W
- White H.D,
- Cross D.B,
- Elliott J.M,
- Norris R.M,
- Yee T.W
- French J.K,
- Williams B.F,
- Hart H.H,
- et al.
- Whitlock R.M.L,
- Bass N.M
- Gibson C.M,
- Cannon C.P,
- Daley W.L,
- et al.
- Principal Investigators of CASS and Their Associates
- LIPID Study Group
- Baigent C,
- Collins R,
- Appleby P,
- et al.
- GISSI Investigators,
- Franzosi M.G,
- Santoro E,
- De Vita C,
- et al.
- ↵Maas ACP, van Domburg RT, Deckers JW, et al. Sustained benefit at 10–14 years follow-up after thrombolytic therapy in myocardial infarction. Eur Heart J 1999 (in press).
- Ross A.M,
- Coyne K.S,
- Moreyra E,
- et al.
- Roux S,
- Christeller S,
- Ludin E
- White H.D,
- Norris R.M,
- Brown M.A,
- Brandt P.W.T,
- Whitlock R.M.L,
- Wild C.J
- Migrino R.Q,
- Young J.B,
- Ellis S.G,
- et al.
- White H.D