Current issues of ACP Journal Club are published in Annals of Internal Medicine


Meta-analysis: t-PA and APSAC reduce short-term mortality in acute myocardial infarction

ACP J Club. 1991 Mar-April;114:36. doi:10.7326/ACPJC-1991-114-2-036

Source Citation

Held PH, Teo KK, Yusuf S. Effects of tissue-type plasminogen activator and anisoylated plasminogen streptokinase activator complex on mortality in acute myocardial infarction. Circulation. 1990;82:1668-74.



To review the effectiveness of tissue-type plasminogen activator (t-PA) and anisoylated plasminogen streptokinase activator complex (APSAC) in reducing mortality in patients with acute myocardial infarction.

Data sources

All completed, published or unpublished, randomized, controlled trials of t-PA and APSAC in acute myocardial infarction up to April 1990 were identified by formal and informal searches of the literature known to the authors and their coworkers. Bibliographies of identified articles were hand searched.

Study selection

8 randomized trials of t-PA and 10 randomized trials of APSAC were identified.

Data extraction

The 8 trials of t-PA had important similarities. All were double-blind and placebo-controlled, treatment was started within 6 hours from the onset of pain, and the dose administered most often was 100 mg of single-stranded t-PA given intravenously over 2 to 3 hours. All trials used concomitant intravenous heparin. Two trials did not require ST-segment elevation on the admission electrocardiogram (ECG). The 10 trials of APSAC included patients who received APSAC within 6 hours from the onset of pain, and all patients had ST-segment elevation on the admission ECG. The usual dose of APSAC was 30 units, and heparin use was variable. Of the 10 trials, 6 were placebo-controlled; in the other 4 trials, patients in the control group received standard therapy. For the purpose of the overview, the primary outcome of interest was mortality.

Main results

8 trials of t-PA including 6800 patients showed that early (< 1 month) death was reduced with t-PA (7% with t-PA vs 9% for controls, P < 0.001). {This absolute risk reduction (ARR) of 2% means that 39 patients would need to be treated (NNT) with t-PA (compared with placebo) to prevent 1 additional death, 95% CI 26 to 78; the relative risk reduction (RRR) was 28%, CI 15% to 38%}.* 10 trials of APSAC including 2300 patients showed that early death was reduced with APSAC (6% with APSAC vs 11% for controls, P = 0.02) {ARR 5%; NNT 28, CI 15 to 192; RRR 42%, CI 24% to 57%}*. The available data did not permit direct comparison of the efficacy of t-PA and APSAC or of these agents with streptokinase.


Short-term mortality in acute myocardial infarction is reduced by both t-PA and APSAC.

Source of funding: Not stated.

Address for article reprint: Dr. S. Yusuf, Clinical Trials Branch, DECA, NHLBI, Room 5C08, Federal Building, 7550 Wisconsin Avenue, Bethesda, MD 20892, USA.

*Numbers calculated from data in article.


Meta-analysis, or overview, is becoming increasingly accepted as a means of combining the results of clinical trials so that the ability to detect clinically important differences is enhanced. In this way a more stable estimate of treatment effects can be obtained. In this study, the authors carefully searched for all trials of t-PA and APSAC in the treatment of acute myocardial infarction. The outcome event, death, was neither subjective nor ambiguous. The results of this overview show that t-PA and APSAC reduce short-term mortality. For each thrombolytic agent one trial was large enough to independently show a statistically significant and clinically important reduction in mortality. The trials with the most outcome events (usually also the larger sample sizes) contribute most of the information in overviews. Therefore, it is important to point out that the estimate of the treatment effect from the pooled results of the smaller trials was similar to that of these larger trials.

In this overview, the odds of early death appeared to be reduced to a greater extent with APSAC than with t-PA. However, the confidence intervals overlap so that any differences observed may be due to the play of chance. The estimate of treatment effect in this overview is also similar to that for streptokinase for acute myocardial infarction (1).*

Until the results of the direct comparative mortality trials are available, the question as to which agent is the most effective will remain unanswered. Marked differences in the cost of these 3 agents must be considered when their benefits are compared.

David Massel, MD
University HospitalLondon, Ontario, Canada

*Two large randomized controlled trials (2, 3) have been published since this overview.


1. Yusuf S, Sleight P, Held P, McMahon S. Routine medical management of acute myocardial infarction. Lessons from overviews of recent randomized controlled trials. Circulation. 1990;82(Suppl. II):117-34.

2. ISIS-3 (Third International Study of Infarct Survival) Collaborative Group. ISIS-3: a randomised comparison of streptokinase vs tissue plasminogen activator vs anistreplase and of aspirin plus heparin vs aspirin alone among 41 299 cases of suspected acute myocardial infarction. Lancet. 1992 Mar 28;339:753-70.

3. The Gusto Investigators. An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. N Engl J Med. 1993 Sep 2;329:673-82.