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


Etiology

Review: Impaired fibrinolytic activity is associated with postoperative but not symptomatic venous thromboembolism

ACP J Club. 1992 Jan-Feb;116:28. doi:10.7326/ACPJC-1992-116-1-028


Source Citation

Prins MH, Hirsh J. A critical review of the evidence supporting a relationship between impaired fibrinolytic activity and venous thromboembolism. Arch Intern Med. 1991 Sep;151:1721-31.


Abstract

Objective

To review the evidence for an association between fibrinolytic activity (FA) and symptomatic or postoperative venous thromboembolism.

Data sources

MEDLINE and Current Contents were searched for publications presenting clinical data on the relation between FA and venous thromboembolic disease. Reference lists and conference proceedings were scanned for pertinent articles.

Study selection

Articles that included both an assessment of FA and of the presence of symptomatic or postoperative thrombosis were selected.

Data extraction

Details of the study design, patients, diagnosis of thrombosis, FA assays used and time of assay, and study results were extracted. Studies were classified by their strength of design.

Main results

Symptomatic thrombosis—No controlled prospective studies relating FA to the occurrence of symptomatic venous thrombosis were found. In a study measuring impaired FA 3 to 90 months after venous thrombosis (n = 272), rates of recurrent deep venous thrombosis were 6.3% in patients with impaired FA compared with 2.3% for those with normal FA (P < 0.1). In the only case-control study with an acceptable control group, relative odds for increased plasminogen activator antigen levels > 14 weeks after venography were 0.96 (95% CI 0.24 to 3.84) for patients with venous thrombosis confirmed by venography. Cross-sectional studies of FA and thrombosis found no consistent association.

Postoperative thrombosis—4 randomized, controlled trials of the effects of pharmacologic interventions intended to increase FA in surgical patients found no reduction in the risk for postoperative deep venous thrombosis in treated patients. In a fifth trial, although intermittent compression of the arms did not increase FA, the treated patients had a reduced risk for deep venous thrombosis (relative risk 0.43, P < 0.03). 28 prospective cohort studies and 1 case-control study of the relation between FA and postoperative thrombosis were found. Preoperative FA in patients with postoperative deep venous thrombosis was lower in 14 studies, not different in 7, and higher in 5 (P < 0.05, sign test, 2-sided). Postoperative FA was reduced in patients with postoperative thrombosis in most of these trials (P < 0.01, sign test, 2-sided).

Conclusions

The evidence of an association between fibrinolytic activity and symptomatic venous thrombosis is weak and inconclusive. Good evidence exists to show an association between impaired fibrinolytic activity measured either preoperatively or postoperatively and postoperative thrombosis, but it is not clear that this association is causal.

Source of funding: Not stated.

Article reprints unavailable.


Commentary

When new diagnostic tests and therapies become available, it is often tempting to view their use as "progress" and quickly incorporate them into practice. This is especially tempting when the tests and interventions are biologically plausible and when there are studies and experts supporting their use. The authors of this review identified such a situation in the possible relationship between impaired fibrinolytic activity and venous thromboembolism. After an informal review by the authors that suggested that the acceptance of this relationship might be premature, they embarked on a systematic review of the literature pertinent to this subject.

This review is an excellent example of the current standard for review articles (1, 2). Prins and Hirsh determined the study methods required to answer the relevant question; assembled the literature relevant to the subject; evaluated the methods of published studies to assess the ability of each to answer the relevant questions; organized and synthesized the results of the studies with emphasis on those with methodologic strengths; and offered recommendations for practice based on available evidence as well as making suggestions for further research about unanswered questions.

The review shows convincingly that the evidence for a relationship between impaired fibrinolysis and venous thromboembolism is inconclusive and that the use of measures of fibrinolytic activity should be limited to a research setting. Practitioners, take note!

John T. Philbrick, MD
University of VirginiaCharlottesville, Virginia, USA

John T. Philbrick, MD
University of Virginia
Charlottesville, Virginia, USA


References

1. Mulrow C. The medical review article: state of the science. Ann Intern Med. 1987;106:484-7.

2. Oxman AD, Guyatt GH. Guidelines for reading literature reviews. Can Med Assoc J. 1988;138:697-703.