← Back

NINDS

Tissue Plasminogen Activator for Acute Ischemic Stroke

Share Share Copied! Compare

Year of Publication: 1995

Authors: The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group

Journal: New England Journal of Medicine

Citation: N Engl J Med 1995;333:1581–1587

Link: https://doi.org/10.1056/NEJM199512143332401

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJM199512143332401

Clinical Question

Does intravenous tissue plasminogen activator (tPA) administered within 3 hours of stroke onset improve clinical outcomes in patients with acute ischemic stroke?

Bottom Line

Intravenous tPA given within 3 hours of ischemic stroke onset significantly improves functional outcomes at 3 months, despite an increased risk of symptomatic intracranial hemorrhage.

        Major Points
        Landmark two-part trial that led to FDA approval of IV tPA for acute ischemic stroke within 3 hours (June 1996).
Part 1 (n=291): Tested whether tPA improves NIHSS by ≥4 points at 24 hours — primary endpoint not significant (47% vs 39%, P=0.21), but favorable trend.
Part 2 (n=333): Tested 90-day functional outcomes using a global statistic across 4 scales — significantly favored tPA (OR for favorable outcome 1.7, 95% CI 1.2–2.6).
Individual 90-day outcomes: mRS 0–1 (39% vs 26%, P=0.019), Barthel Index ≥95 (50% vs 38%, P=0.026), Glasgow Outcome Scale 1 (44% vs 32%, P=0.025), NIHSS ≤1 (31% vs 20%, P=0.033).
Symptomatic ICH within 36 hours: 6.4% (tPA) vs 0.6% (placebo), P<0.001. Most sICH occurred within the first 12 hours.
Mortality at 3 months: 17% (tPA) vs 21% (placebo), not significant (P=0.30). Trend favored tPA despite higher sICH rate.
Benefit was consistent across stroke subtypes (large artery, cardioembolic, small vessel, other) and was greater with earlier treatment (0–90 min showed larger effect than 91–180 min).
Protocol hemorrhage: any ICH on 24-hour CT scan was 10.6% (tPA) vs 3.2% (placebo); asymptomatic ICH was common.

      

Design

Study Type: Randomized, double-blind, placebo-controlled trial (two-part)

Randomization: 1

Blinding: Patients, treating physicians, and outcome assessors were blinded

Enrollment Period: January 1991 – October 1994

Follow-up Duration: 3 months

Centers: 43

Countries: United States

Sample Size: 624

Analysis: Intention-to-treat; chi-square tests for proportions; Cochran-Mantel-Haenszel for stratified data; significance set at P<0.05

Inclusion Criteria

Clinical diagnosis of ischemic stroke
Age ≥18 years
Able to receive treatment within 3 hours of symptom onset
Deficit measurable on NIH Stroke Scale
CT scan ruling out hemorrhage

Exclusion Criteria

Rapidly improving or minor symptoms
Symptoms suggesting subarachnoid hemorrhage
Seizure at onset of stroke
History of intracranial hemorrhage
Major surgery or serious trauma within 14 days
Stroke or serious head trauma within 3 months
Recent gastrointestinal or urinary tract hemorrhage (within 21 days)
Arterial puncture at a non-compressible site within 7 days
Systolic BP >185 mm Hg or diastolic BP >110 mm Hg
Blood glucose <50 or >400 mg/dL
Platelet count <100,000/mm³
Heparin received within 48 hours and aPTT above normal
Current use of oral anticoagulants with INR >1.7 or PT >15 sec
CT showing multilobar infarction or signs of mass effect

Baseline Characteristics

Characteristic	Control	Active
Mean Age	66.0 ± 13.2 years	67.1 ± 11.7 years
Sex - Male	52%	52%
Race - White	64%	64%
Race - Black	28%	28%
Race - Other	8%	8%
NIH Stroke Scale Score	14.1 ± 5.5	14.2 ± 5.5
Hypertension	71%	70%
Diabetes Mellitus	24%	23%
Atrial Fibrillation	19%	19%
History of Stroke or TIA	16%	16%
Coronary Artery Disease	24%	23%
Current Smoker	28%	26%
Stroke Subtype - Large Artery Atherosclerosis	~20%	~20%
Stroke Subtype - Cardioembolic	~20%	~20%
Stroke Subtype - Small Vessel Lacunar	~18%	~18%
Stroke Subtype - Other/Undetermined	~42%	~42%
Onset-to-Treatment ≤90 min	~48%	~48%
Onset-to-Treatment 91–180 min	~52%	~52%
Mean Time to Treatment	89 ± 32 minutes	90 ± 33 minutes

Arms

Field	tPA Group	Control
Intervention	Intravenous alteplase (rt-PA) 0.9 mg/kg (max 90 mg): 10% given as IV bolus over 1 minute, remaining 90% infused over 60 minutes. No anticoagulants or antiplatelet agents allowed for 24 hours after treatment. BP management per protocol: if SBP >185 or DBP >110, treat with IV labetalol before and during infusion; hold tPA if BP not controlled. 24-hour follow-up CT required before starting antithrombotics.	Intravenous saline placebo, same administration protocol
Duration	Single infusion within 3 hours of stroke onset	Single infusion within 3 hours of stroke onset

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
Part 2 primary: Global test statistic combining 4 scales at 90 days — favorable outcome defined as mRS 0–1, Barthel Index ≥95, Glasgow Outcome Scale 1, and NIHSS ≤1. Part 1 primary: Improvement of ≥4 points on NIHSS or complete resolution at 24 hours.	Primary	Part 2: 26% favorable outcome (mRS 0–1). Part 1: 21% neurologic improvement at 24h.	Part 2: 39% favorable outcome (mRS 0–1). Part 1: 33% neurologic improvement at 24h.	OR 1.7 (global statistic, Part 2)	Part 2 global: P=0.008; mRS 0-1: P=0.019; Part 1 primary (24h NIHSS): P=0.21
mRS 0–1 at 90 days (Part 2)	Secondary	26%	39%	OR 1.7	0.019
Barthel Index ≥95 at 90 days (Part 2)	Secondary	38%	50%	OR 1.6	0.026
Glasgow Outcome Scale 1 at 90 days (Part 2)	Secondary	32%	44%	OR 1.6	0.025
NIHSS ≤1 at 90 days (Part 2)	Secondary	20%	31%	OR 1.7	0.033
NIHSS improvement ≥4 points at 24 hours (Part 1 primary)	Secondary	21%	33%		0.21
Mortality at 3 months	Secondary	21%	17%		0.30
Symptomatic intracerebral hemorrhage within 36 hours	Adverse	0.6% (2/312)	6.4% (20/312)		<0.001
Any intracerebral hemorrhage on 24h CT (protocol)	Adverse	3.2%	10.6%		<0.001
Fatal ICH within 3 months	Adverse	0.3%	2.9%
Systemic bleeding requiring transfusion	Adverse	1.0%	1.0%		NS

Subgroup Analysis

Benefit of tPA was consistent across prespecified subgroups: age (<65, 65–74, ≥75), baseline NIHSS severity (≤10, 11–15, 16–20, >20), stroke subtype (large vessel, cardioembolic, small vessel, other), and time to treatment. Time-to-treatment interaction: patients treated within 0–90 minutes had a larger treatment effect (OR ~2.1) than those treated 91–180 minutes (OR ~1.5), though benefit was significant in both strata. No significant interaction by sex, race, or presence of early CT changes. 12-month follow-up (published separately) confirmed sustained benefit: absolute difference in favorable outcomes persisted at 1 year.

Criticisms

6.4% symptomatic ICH rate — acceptable risk/benefit, but led to initial reluctance in adoption.
Strict 3-hour window limited generalizability; later trials (ECASS III, IST-3) extended to 4.5 hours.
Part 1 primary endpoint (24-hour NIHSS improvement) was not significant — only Part 2 (90-day outcomes) was positive.
Excluded rapidly improving or minor stroke symptoms — debated whether these patients may also benefit (later addressed by PRISMS and TEMPO-2).
No advanced imaging selection (CTA, CTP, MRI) — all decisions based on non-contrast CT only.
Small sample size (624) by modern standards, though sufficient for the effect size observed.
Limited racial diversity — 64% White, 28% Black — limited data in Hispanic and Asian populations.
No vascular imaging required — stroke subtypes classified retrospectively, vessel occlusion status unknown.
Protocol mandated BP control <185/110 but no standardized antihypertensive protocol across sites.

Funding

National Institute of Neurological Disorders and Stroke (NINDS), National Institutes of Health (NIH)

Based on: NINDS (New England Journal of Medicine, 1995)