The New England Journal of Medicine September 02, 1993

An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction

The GUSTO Investigators

Bottom Line

The GUSTO-I trial demonstrated that an accelerated infusion of tissue plasminogen activator (t-PA) combined with intravenous heparin significantly reduced 30-day mortality in acute myocardial infarction compared to standard streptokinase regimens.

Key Findings

1. 30-day mortality was significantly lower in the accelerated t-PA group (6.3%) compared to the streptokinase plus subcutaneous heparin (7.2%), streptokinase plus IV heparin (7.4%), and combination therapy (7.0%) groups [4.1.1].
2. Accelerated t-PA provided a 14% relative reduction (95% CI, 5.9 to 21.3%) in mortality compared to the pooled streptokinase-only strategies (P = 0.001).
3. The rate of hemorrhagic stroke was higher with accelerated t-PA (0.72%) and combination therapy (0.94%) compared to streptokinase with subcutaneous (0.49%) or intravenous heparin (0.54%).
4. The combined endpoint of death or disabling stroke remained significantly lower in the accelerated t-PA group than in the streptokinase-only groups (6.9% vs. 7.8%, P = 0.006).

Study Design

Design
RCT
Open-Label
Sample
41,021
Patients
Duration
30 days
Median
Setting
15 countries
Population Patients presenting within 6 hours of acute myocardial infarction symptom onset with ST-segment elevation
Intervention Accelerated tissue plasminogen activator (t-PA) and intravenous heparin
Comparator Streptokinase with subcutaneous heparin, streptokinase with intravenous heparin, or a combination of streptokinase plus t-PA with intravenous heparin
Outcome 30-day mortality

Study Limitations

The open-label design may have introduced bias in the administration of adjunctive treatments, although the primary endpoint of 30-day mortality is highly objective.
The absolute mortality benefit of t-PA over streptokinase was approximately 1%, requiring the treatment of 100 patients to save one life, which raised cost-effectiveness debates due to the higher price of t-PA.
Accelerated t-PA carried a significantly higher risk of hemorrhagic stroke compared to standard streptokinase regimens [6.1.1].
The widespread adoption of primary percutaneous coronary intervention (PCI) has since superseded thrombolysis as the standard of care for STEMI in regions with capable facilities, limiting the contemporary relevance of these specific pharmacological regimens.

Clinical Significance

GUSTO-I established accelerated t-PA with intravenous heparin as the superior pharmacological reperfusion strategy for acute myocardial infarction compared to standard streptokinase. Despite a slight increase in hemorrhagic stroke risk, the significant survival advantage cemented accelerated t-PA as the standard thrombolytic regimen worldwide until the advent and widespread implementation of primary PCI.

Historical Context

Prior to GUSTO-I, mega-trials such as GISSI-1 and ISIS-2 had established the life-saving benefits of streptokinase in acute myocardial infarction. However, the comparative efficacy of more fibrin-specific agents like t-PA remained intensely debated. GUSTO-I was designed to definitively test the 'open-artery' hypothesis—that early, rapid, and sustained restoration of normal coronary blood flow translates directly to improved survival. The trial successfully validated this hypothesis and profoundly shaped cardiovascular medicine in the 1990s.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How do the mechanisms of action differ between tissue plasminogen activator (t-PA) and streptokinase, and how does this difference explain the requirement for concurrent intravenous heparin with t-PA but not necessarily with streptokinase?

Key Response

t-PA is fibrin-specific and rapidly degrades the clot while leaving most circulating fibrinogen intact, which can lead to a highly prothrombotic state post-lysis that requires IV heparin to prevent reocclusion. In contrast, streptokinase is non-fibrin-specific and causes systemic fibrinogen depletion (a systemic lytic state), which inherently acts as a continuous anticoagulant, making concurrent IV heparin less strictly necessary for maintaining vessel patency.

Resident
Resident

In the GUSTO-I trial, accelerated t-PA showed a mortality benefit over streptokinase but was associated with a specific severe adverse event. What is this adverse event, and how does its incidence influence patient selection and absolute contraindications for thrombolytic therapy today?

Key Response

t-PA carries a slightly higher risk of hemorrhagic stroke (intracranial hemorrhage) compared to streptokinase (0.72% vs 0.49% in GUSTO-I). Residents must recognize this critical trade-off: a net mortality benefit but an increased ICH risk. This directly informs the absolute contraindications checklist for fibrinolysis in STEMI (e.g., prior hemorrhagic stroke, recent major surgery, active internal bleeding) when evaluating a patient in a non-PCI capable center.

Fellow
Fellow

The GUSTO-I angiographic substudy introduced the concept of TIMI flow grades to explain the clinical differences between the treatment arms. How did the differences in TIMI 3 flow rates at 90 minutes between the accelerated t-PA and streptokinase groups validate the 'open-artery hypothesis' in acute myocardial infarction?

Key Response

The GUSTO angiographic substudy demonstrated that accelerated t-PA achieved TIMI 3 (normal) flow in about 54% of patients at 90 minutes, compared to approximately 30% for streptokinase. This early, complete reperfusion directly correlated with improved survival and preserved left ventricular function, definitively establishing that early and complete patency (TIMI 3, rather than just partial TIMI 2 flow) is the primary determinant of the mortality benefit in STEMI.

Attending
Attending

Given that primary percutaneous coronary intervention (PCI) is now the standard of care for STEMI, how do the core principles established by the GUSTO-I trial regarding 'time to reperfusion' and 'accelerated dosing regimens' continue to inform our modern pharmacoinvasive strategies?

Key Response

GUSTO-I established that 'time is muscle' and that rapid, complete reperfusion saves lives, proving the superiority of a fast-acting, fibrin-specific agent. This paved the way for modern pharmacoinvasive strategies (using tenecteplase, a modified t-PA) for patients presenting to non-PCI centers who cannot be transferred within 120 minutes. The trial teaches us that if PCI is delayed, rapid administration of a fibrin-specific lytic followed by transfer for rescue or routine PCI remains a highly effective, evidence-based paradigm.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The GUSTO-I trial enrolled over 41,000 patients to detect a 1% absolute risk reduction in 30-day mortality. How does this massive sample size requirement reflect the statistical challenges of 'mega-trials' comparing active treatments, and what are the implications for designing modern cardiovascular outcome trials?

Key Response

Detecting small absolute differences between two effective therapies (active-control trials) requires massive statistical power. As baseline standard-of-care improves, expected event rates drop, and the delta between therapies shrinks. Researchers must understand how this mathematical reality drives modern trials toward composite endpoints (e.g., MACE) rather than isolated all-cause mortality, as powering a contemporary trial for a hard mortality endpoint alone has become logistically and financially prohibitive.

Journal Editor
Journal Editor

GUSTO-I was an open-label trial, largely due to the differing administration protocols of the four regimens. As an editor evaluating this paper, how would you assess the risk of bias introduced by the open-label design, particularly regarding the use of adjunctive therapies and the assessment of the primary mortality endpoint?

Key Response

While all-cause mortality is a hard, objective endpoint relatively resistant to ascertainment bias, an open-label design introduces a significant risk of performance bias. Physicians knowing the patient received a 'superior' or 'newer' drug (t-PA) might treat them more aggressively with other adjunctive measures (e.g., rescue PCI, intra-aortic balloon pumps, blood transfusions). A rigorous reviewer would scrutinize the supplementary data to ensure balanced utilization of downstream non-protocolized interventions across all arms to validate that the mortality benefit was solely due to the lytic strategy.

Guideline Committee
Guideline Committee

Although GUSTO-I established accelerated t-PA as superior to streptokinase, current ACC/AHA STEMI guidelines preferentially recommend fibrin-specific agents with single-bolus administration (like tenecteplase) when primary PCI is unavailable. How did the findings and practical limitations of the GUSTO-I t-PA regimen drive the evolution of these current Class I recommendations?

Key Response

GUSTO-I proved the superiority of fibrin-specific agents (t-PA) over non-specific ones. However, the complex weight-based accelerated infusion of t-PA was prone to dosing errors and delays in preparation. This directly drove the biochemical development of single-bolus mutant t-PAs (tenecteplase/TNK), which match or slightly exceed t-PA's efficacy but eliminate infusion complexity. Consequently, current guidelines give single-bolus fibrin-specific agents a Class I recommendation (Level of Evidence A) over standard t-PA infusions to minimize door-to-needle times and reduce medication errors in high-stress emergency settings.

Clinical Landscape

Noteworthy Related Trials

1986

GISSI-1 Trial

n = 11,712 · Lancet

Tested

Intravenous streptokinase

Population

Patients with acute MI within 12 hours of symptom onset

Comparator

Standard medical care

Endpoint

21-day overall mortality

Key result: Streptokinase significantly reduced 21-day mortality compared to standard care, with the greatest benefit seen when administered within 3 hours of symptom onset.
1988

ISIS-2 Trial

n = 17,187 · Lancet

Tested

Streptokinase, aspirin, or both

Population

Patients with suspected acute MI within 24 hours of symptom onset

Comparator

Placebo infusions and tablets

Endpoint

5-week vascular mortality

Key result: The combination of streptokinase and aspirin produced significantly greater reductions in 5-week vascular mortality than either agent alone.
1999

ASSENT-2 Trial

n = 16,949 · Lancet

Tested

Single-bolus tenecteplase (TNK-tPA)

Population

Patients with acute STEMI within 6 hours of symptom onset

Comparator

Accelerated infusion of alteplase (tPA)

Endpoint

30-day mortality

Key result: Tenecteplase and alteplase had equivalent 30-day mortality rates, but tenecteplase was easier to administer and associated with fewer non-cerebral bleeding complications.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis