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


Review: Heterogeneous studies show a reduction in stroke after anticoagulation but not after thrombolysis in patients with acute myocardial infarction

ACP J Club. 1993 Mar-April;118:43. doi:10.7326/ACPJC-1993-118-2-043

Source Citation

Vaitkus PT, Berlin JA, Schwartz JS, Barnathan ES. Stroke complicating acute myocardial infarction. A meta-analysis of risk modification by anticoagulation and thrombolytic therapy. Arch Intern Med. 1992 Oct;152:2020-4.



To assess the effect of anticoagulation and thrombolytic therapy on the incidence of stroke after acute myocardial infarction, using meta-analysis.

Data sources

Citations were retrieved from BRS Colleague and Index Medicus for the years 1948 to 1990. Key words included myocardial infarction, anticoagulation, thrombolysis, streptokinase, tissue plasminogen activator, and heparin. Bibliographies of review and original articles were also scanned.

Study selection

Clinical trials investigating anticoagulants or thrombolytic agents that reported the total incidence of strokes in treated and control patients in the hospital phase of an acute myocardial infarction were included. 9 studies of anticoagulants and 6 studies of thrombolytic agents met the criteria.

Data extraction

Data on the method used to assign patients to treatment groups, the therapeutic agent used, and the incidence of total strokes in both treatment groups were recorded. Results were pooled using the Mantel-Haenszel estimate of the common odds ratio.

Main results

2 of the 9 studies investigating anticoagulants randomly allocated patients to the treatment groups. For randomized, nonrandomized, and all studies, the incidence of stroke was reduced in the anticoagulated group (odds ratio [OR] 0.41, 95% CI 0.22 to 0.78; OR 0.49, CI 0.32 to 0.77; OR 0.46, CI 0.30 to 0.64, respectively). For each of these analyses a substantial heterogeneity was found among study results. 3 of the 6 studies investigating thrombolytic therapy used streptokinase, 2 used tissue plasminogen activator (t-PA), and 1 used anisoylated plasminogen streptokinase activator complex. All were randomized controlled trials. The ORs for all thrombolytic, t-PA, and streptokinase trials were 1.08 (CI 0.87 to 1.35), 1.28 (CI 0.76 to 2.17), and 1.02 (CI 0.80 to 1.30), respectively, suggesting no difference in the incidence of stroke when comparing treated and control groups. The OR for the 3 studies directly comparing streptokinase and t-PA was 0.73 (CI 0.61 to 0.86), indicating an excess of strokes among patients treated with t-PA.


Anticoagulants appear to reduce the incidence of stroke after myocardial infarction, but this effect is difficult to interpret because of the heterogeneity among studies. Thrombolytic therapy does not appear to affect the incidence of stroke, but a direct comparison of tissue plasminogen activator and streptokinase showed an excess of strokes with tissue plasminogen activator.

Source of funding: Not stated.

For article reprint: Dr. P.T. Vaitkus, Cardiology Unit, Medical Center Hospital of Vermont, Burlington, VT 05401, USA. FAX 802-656-3637.


Meta-analysis is most useful when the results of studies with relatively comparable design, but inadequate sample size, are pooled. Unfortunately, few trials with considerable heterogeneity of design have data that can be pooled to assess the effect of anticoagulation on the incidence of stroke and myocardial infarction. Despite the heterogeneity in study design, anticoagulation appears to reduce the incidence of stroke. These results are consistent with those of recent studies where patients with anterior wall myocardial infarction and mural thrombosis had a decreased incidence of systemic embolization when treated with anticoagulation. On the basis of the available data, the recent Third ACCP Consensus Conference on Antithrombotic Therapy (1) recommended that every patient with a myocardial infarction receive at least a prophylactic dose of heparin with consideration given to full anticoagulation depending on risk status and presence or absence of thrombolysis. Another important finding of this and other studies is that the incidence of stroke from myocardial infarction is not increased in patients treated with fibrinolytic agents. Thrombolysis, however, appears to increase the incidence of hemorrhagic stroke while decreasing embolic strokes. Whether differences exist between the incidence of hemorrhagic strokes with t-PA and streptokinase continues to be questioned because the substantial difference in hemorrhagic strokes between the 2 agents with pooled data analysis primarily reflects the increased incidence of strokes observed in the ISIS-3 trial (2) with a high-dose 2-chain t-PA (Duteplase). Nonetheless, with currently approved fibrinolytic regimens for acute myocardial infarction, the overall incidence of stroke is extremely low and no greater than that which occurs in patients not treated with thrombolysis.

Paul R. Eisenberg, MD, MPH
Washington University School of Medicine St. Louis, Missouri, USA