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


Meta-analysis: Delayed thrombolytic therapy increases the risk for cardiac rupture after acute myocardial infarction

ACP J Club. 1991 Jan-Feb;114:8. doi:10.7326/ACPJC-1991-114-1-008

Source Citation

Honan MB, Harrell FE Jr, Reimer K,A, et al. Cardiac rupture, mortality and the timing of thrombolytic therapy: a meta-analysis. J Am Coll Cardiol. 1990 Aug;16:359-67.



To evaluate the effect of time to treatment with thrombolytic therapy for acute myocardial infarction on the risk for cardiac rupture and all-cause mortality.

Data sources

Studies published before 1 March 1989 were identified using the Bibliographic Retrieval Service (BRS) program and MEDLINE searching on the keywords: fibrinolytic agents, myocardial infarction, random trials, and clinical trials. All pertinent references were included and their bibliographies were searched by hand.

Study selection

Studies were selected if they met the following criteria: randomized controlled trial of thrombolytic agents for acute myocardial infarction, report of cardiac rupture and mortality data, report of the mean time to treatment, 5 or more outcome events, and autopsies reported on at least 50% of the fatalities when cardiac rupture was being evaluated.

Data extraction

Trial results were pooled to estimate the odds ratios of cardiac rupture or death with respect to time from symptom onset to treatment. Logistic regression models were used to assess the relation of various factors to cardiac rupture or death.

Data synthesis

In the 4 trials used to assess the effect of thrombolysis on cardiac rupture that involved 1638 patients and 58 episodes of cardiac rupture, the thrombolytic agent studied was intravenous streptokinase. The odds ratio of cardiac rupture increased with delay in the mean time to treatment (P = 0.01), from 0.40 (95% CI 0.17 to 0.93) at 7 hours to 3.21 (CI 1.10 to 10.1) at 17 hours. 42 trials were used to assess the effect of thrombolytic therapy with various agents on all-cause mortality. The odds ratio for all-cause mortality was directly correlated with time to treatment (P = 0.006): at 3 hours, 0.72 (CI 0.67 to 0.77); at 14 hours, 0.88 (CI 0.77 to 1.00), and at 21 hours, 1 (CI 0.82 to 1.37).


The risk for cardiac rupture after acute myocardial infarction is directly related to the timing of thrombolytic therapy, with late treatment increasing the risk. The beneficial effect of thrombolytic therapy on survival also diminishes with treatment delay but treatment after 6 hours of onset of infarction is still effective.

Sources of funding: The National Center for Health Services Research; the National Heart, Lung, and Blood Institute; the Andrew W. Mellon Foundation; the Robert Wood Johnson Foundation.

Address for article reprint: Dr. B. Honan, 880 Montclair Road, 1st Floor, Birmingham, AL 35213, USA.


The analysis of cardiac rupture should be interpreted conservatively. Of the 42 trials selected for the mortality analysis, only 4 had sufficient data on cardiac rupture. Bias can occur in meta-analysis when data are not complete on events from all trials. This is more likely to occur when data are pooled on secondary end points that are not directly related to the main study objective.

There are other problems in the 4 trials. Overall 58 cases of rupture occurred, 29 (3.5%) among streptokinase-treated patients and 29 (3.6%) among the controls, an odds ratio of 0.98 (95% CI 0.58 to 1.65), clearly a nonsignificant result. All 4 trials were published before 1980 and used low-dose streptokinase-infusion regimens that are not in use today. 2 trials included patients up to 72 hours from the onset of symptoms, and 2 trials did not restrict inclusion to definite infarctions. In the 2 trials in which the odds ratios of cardiac rupture and death were > 1.0 (streptokinase worse), steroid use was compulsory. The use of steroids in the setting of acute myocardial infarction has been associated with cardiac rupture. Finally, the use by the authors of data from the Gruppo Italiano per lo Studio della Streptochinasi nell' Infarto miocardico (GISSI) trial as corroboration can be completely discounted because the rate of autopsy was low (16%) and the trial was not double-blind.

The mortality analysis suggests that when analyzed as mean time to treatment, a beneficial therapeutic window of less than 24 hours exists. However, results from the Second International Study of Infarct Survival (ISIS-2) trial and an overview of this subject show that treatment between 6 and 24 hours is associated with a 17% reduction (P < 0.001) in mortality (1).

David Massel, MD
Hamilton General HospitalHamilton, Ontario, Canada


1. Yusuf S, Sleight P, Held P, MacMahon S. Routine medical management of acute myocardial infarction: lessons from overviews of recent randomized controlled trials. Circulation. 1990;82:117-34.