Authors' objectives

To determine the effects of long-term oral anticoagulant (OA) therapy, stratified by intensity of anticoagulation and aspirin therapy, on outcomes in patients with CAD.

Searching

MEDLINE, EMBASE, and Current Contents Clinical Medicine were searched from 1960 to July 1999 using search terms related to OA and vascular disease. Reference lists were reviewed and experts in the field and pharmaceutical companies were contacted. Two authors of published reports were contacted for further information.

Study selection

Assessment of study quality

The authors do not state that they assessed validity.

Data extraction

Data were extracted from the published reports independently by two reviewers and a consensus was achieved for any discrepancies. Information on type, duration and method of monitoring OA therapy was recorded. The reported incidences of death, MI and stroke were recorded. Where data on fatal and non-fatal events were provided, a composite rate of death or non-fatal MI or Non-fatal stroke was calculated. Total bleeding was recorded and bleeds that required transfusion, surgical intervention or hospitalisation were identified as major bleeding events. The intensity of anticoagulation was recorded as international normalised ratio (INR). In some cases this was calculated from the reported prothrombin ratio and the type of thromboplastin used, or estimated from the period in which the study was conducted. All INR classifications were independently reviewed by a thrombosis expert.

Methods of synthesis

Results of the review

A total of 30 reports of 31 trials (n=23397) were included in the analysis. Of these, 20 were classified as high intensity (INR <2.8) (n=11692), eight were classified as moderate intensity (INR 2-3) (n=3270), and three as low intensity (INR <2) (n=8435).

High intensity OA versus control: the ORed for mortality was 22% (95% CI: 13%, 31%, p<0.001) and that for MI was 42% (95% CI: 34%, 48%, p<0.001) and for thromboembolic events was 63% (95% CI: 53%, 71%, p<0.001). For stroke specifically the ORed was 48% (95% CI: 33%, 60%, p<0.001). There was a 6-fold ()&= 6.0,95% CI: 4.4, 8.2) increase in major bleed. The risk benefit in terms of number of events prevented: number of major bleeds caused per 1000 patients treated was 98 (95% CI: 73,123): 39 (95% CI: 35, 43).

Moderate intensity INR: The ORed OA versus control for mortality was 18% (95% CI: -6%, 37%, p> 0.1), and for MI was 52% (95% CI: 37%, 64%, p< 0.001) and for stroke was 53% (95% CI: 19%, 73%, p=0.02), with an associated 7.7-fold increase in risk of bleeding (OR =7.7,95% CI: 3.3, 17.6, p<0.001). The risk benefit in terms of number of events prevented: number of major bleeds caused per 1000 patients treated was 24 (95% CI: 22,26): 35 (95% CI: 21,49).

High or moderate intensity versus aspirin (three trials): No reduction in death, MI or stroke was observed compared with aspirin and treatment was associated with a 2.4-fold (OR =2.4, 95% CI: 1.6, 3.6) increase in major bleeding. The risk benefit in terms of number of events prevented: number of major bleeds caused per 1000 patients treated was 13 (95% CI: 11, 14): 14 (95% CI: 12, 16).

Combination of OA and aspirin compared with aspirin alone had an ORed of 56% (95% CI: 17%, 77%, p-0.01), and an associated increase in major bleeding of 1.9-fold (955 CI: 0.6, 6.0, NS), but this is based on three trials with a total of only 480 patients. The risk benefit in terms of number of events prevented: number of major bleeds caused per 1000 patients treated was 54 (95% CI: 43, 65): 16 (95% CI: 10-22).

Low intensity OA plus aspirin: no reduction in death, MI or stroke was observed compared with and treatment was associated with a 1.3-fold (95% CI: 1.0, 1.8) increase in major bleeding. The risk benefit in terms of number of events prevented: number of major bleeds caused per 1000 patients treated was 7 (95% CI: 6,8): 5 (95% CI: 4,6).

Authors' conclusions

Among patients with CAD, high intensity and moderate intensity OA are effective in reducing MI and stroke, but increase the risk of bleeding. In the presence of aspirin, low-intensity OA does not appear to be superior to aspirin alone, while moderate- to high-intensity OA plus aspirin versus aspirin alone appears promising and the bleeding risk is modest, but this requires confirmation from ongoing trials.

CRD commentary

The review addresses an appropriate question with well-defined inclusion and exclusion criteria. The literature search conducted was probably adequate although it is possible that some key articles might have been missed due to the limited number of electronic databases searched. The validity of studies included in the review was not specifically addressed but 13 publications were excluded. The review presents some details of all individual trials included in the review, although the exact design (e.g level of blinding, exact nature of 'control') is omitted. The overall number of patients by treatment group in each analysis is given. The trials were checked for heterogeneity and the majority of the variation was found to stem from the older smaller trials. The summary estimates of the modern trials are consistent with the overall estimates. The studies were pooled using a formal meta-analytical method. The grouping of the studies into high, moderate and low intensity anticoagulation appears reasonable, however the decision to combine the high and moderate intensity treatment with or without aspirin when comparing with aspirin was not explained. The results presented in the review appear to support the authors conclusions, however the results presented in the tables and those given in the text are sometimes different, making interpretation difficult.

Implications of the review for practice and research

Among patients with CAD, high intensity and moderate intensity OA are effective in reducing death, MI and stroke, but increase the risk of bleeding. Moderate- to high-intensity OA plus aspirin versus aspirin alone appears promising and the bleeding risk is modest, but this requires confirmation from ongoing trials.

Bibliographic details

Anand S S, Yusuf S. Oral anticoagulant therapy in patients with coronary artery disease: a meta-analysis. JAMA 1999; 282(21): 2058-2067. [PubMed: 10591389]

Original Paper URL

http://jama.ama-assn.org/

Other publications of related interest

1. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719-48.

This additional published commentary may also be of interest. Beyrth RJ. Review: high dose and moderate dose oral anticoagulants reduce events in coronary artery disease but increase major bleeding and are no more effective than aspirin. Evid Based Med 2000;5:111.

Indexing Status

Subject indexing assigned by NLM

MeSH

Administration, Oral; Anticoagulants /administration & dosage /therapeutic use; Aspirin /administration & dosage /therapeutic use; Coronary Disease /drug therapy /physiopathology; Drug Therapy, Combination; Humans; Myocardial Infarction /epidemiology /prevention & control; Platelet Aggregation Inhibitors /administration & dosage /therapeutic use; Randomized Controlled Trials as Topic; Risk; Stroke /epidemiology /prevention & control

AccessionNumber

12000008003

Database entry date

31/03/2001

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.