Venous Thromboembolism After Acute Stroke
To the Editor:
The recent review by Kelly et al1 generally takes a balanced view of the evidence cited but surprisingly omits some very important evidence.
There is good evidence from several sources that antiplatelet agents (most of the evidence comes from trials of aspirin) are effective and safe in preventing deep venous thrombosis (DVT) and pulmonary embolism (PE) in a wide variety of patients at high risk of venous thromboembolism.2 3 A systematic review of the randomized trials available up to 1994, including a total of approximately 9000 patients, showed that antiplatelet agents (chiefly aspirin) significantly reduced the risk of DVT by 39% and PE by 64%.2 3 Some clinicians were not persuaded by the evidence from the Antiplatelet Trialists’ meta-analysis, so a large-scale trial was established to confirm or refute these results. The Pulmonary Embolism Prevention (PEP) trial sought to evaluate the effects of low-dose aspirin in the prevention of pulmonary embolism and deep vein thrombosis in patients undergoing surgery for hip fracture.4 The trial randomized 13 356 patients to aspirin or placebo. Allocation to aspirin significantly reduced the odds of PE by 43% (95% CI 18 to 60) and symptomatic DVT by 29% (95% CI 3 to 48).4
The Cochrane review by Gubitz et al5 (not cited by Kelly) identified 8 trials of antiplatelet therapy in patients with acute ischemic stroke that reported effects on venous thromboembolism. Only 2 trials (including just 136 patients) reported effects on DVT; allocation to antiplatelet therapy produced a nonsignificant 22% reduction in the odds of DVT (95% CI 64% reduction to 67% increase). However, 8 trials with 40 872 patients (chiefly Chinese Acute Stroke Trial [CAST] and International Stroke Trial [IST]) reported effects on PE; allocation to aspirin significantly reduced the odds of PE by 29% (95% CI 4 to 47%)(2P=0.03). Taken with the evidence of benefit in other categories of patient, and given the established safety of aspirin in stroke, it seems reasonable to conclude that aspirin does reduce the risk of DVT and PE after stroke. For patients with ischemic stroke, aspirin has many advantages as first-line agent for thromboprophylaxis: it is inexpensive, needs only once-daily administration, does not require injections, and is associated with a low risk of bleeding. The research question is now, can heparin add to that benefit? Any new trial would, therefore, need to compare aspirin alone with aspirin plus low-dose heparin and would need to recruit perhaps a few tens of thousands of patients to answer the question reliably (since PE appears to be uncommon after stroke these days).5
One other small omission. We agree with Kelly that the value of graded compression stockings (and the choice of full-length or below-knee) for the prevention of DVT and PE after stroke is not well established. A Cochrane review on this topic is in preparation.6 Furthermore, the Scottish Executive has just funded the Clots in Legs or TED Stockings (CLOTS) collaborative group to undertake a multicenter, randomized, controlled trial to evaluate the benefit of graded compression stockings after stroke; the study is a small “family” of 2 trials, one comparing long stockings with no stockings and the other long stockings with short stockings. Details of the protocol are available at the trial website (www.dcn.ed.ac.uk/clots).
- Copyright © 2001 by American Heart Association
Kelly J, Rudd A, Lewis R, Hunt B. Venous thromboembolism after stroke. Stroke. 2001;32:262-267.
Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy, III: reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ. 1994;308:235-246.
Collins R, Baigent C, Sandercock P, Peto R, for the Antiplatelet Trialists’ Collaboration. Antiplatelet therapy for thromboprophylaxis: the need for careful consideration of the evidence from randomised trials. BMJ. 1994;309:1215-1217.
Gubitz G, Sandercock P, Counsell C. Antiplatelet therapy for acute ischaemic stroke (Cochrane review). In: The Cochrane Library, issue 4, 2000. Oxford, UK: Update Software.
Chiodo Grandi F, Miccio M, Salvi R, Antonutti L, Mazzone C. Physical means for preventing deep vein thrombosis after stroke (protocol for Cochrane review). In: The Cochrane Library, issue 4, 2000. Oxford, UK: Update Software.
We are grateful to Prof Sandercock and Dr Dennis for their valuable comments. The currently recommended thromboprophylactic strategy after acute ischemic stroke is combined early use of aspirin and graded compression elasticated stockings (GCS).R1 Although the reduction in risk of PE seen with aspirin in the ISTR2 did not achieve significance, we do agree that accumulated data suggest that aspirin is effective in these patients, with a risk reduction of 29% demonstrated in a recent Cochrane review.R3 However, aspirin has not been shown to reduce overall mortality in surgical patients,R4 R5 and the marginal effect on early mortality after stroke has been predominantly attributable to prevention of recurrent ischemic stroke.R2 GCS reduce DVT in surgical patients by about two thirds,R6 although there are insufficient data to assess their effect on PE and total mortality. Although it seems reasonable to assume that they may also prevent DVT after stroke, this has not yet been clearly demonstrated.
By contrast, an overview of 70 trials of unfractionated heparin in some 16 000 general, orthopedic, and urological surgery patients has shown a two-thirds reduction in fatal PE and reduced total mortality,R7 and low-molecular-weight heparin is at least as effective.R8 Moreover, heparin is also commonly combined with an alternative thromboprophylactic modality, usually GCS.
The incidence of PE in the ISTR2 in aspirin-treated patients who did not receive heparin was only 0.7% at 2 weeks. However, these data do not reflect the overall importance of PE as a cause of early morbidity and mortality, because fatal pulmonary emboli rarely occur in the first week after stroke, being most common between the second and fourth weeks.R9 In addition, pulmonary emboli were not systematically sought within this period and were therefore probably underascertained, particularly given that the notorious propensity to underdiagnose or misdiagnose this condition is compounded in a population in whom dysphasia, confusion, and obtundation are common. Even in the IST, though, PE accounted for approximately 20% of early nonneurological deaths and 8% of all early deaths, and other studies have indicated that PE causes up to 25% of early deaths.R10 Clearly, although the absolute risk of fatal PE after stroke is small, it is still an important cause of early mortality.R9
Given that the prevalence of DVT after stroke is greater than that after general surgery and similar to that following hip or knee arthroplastyR11 and that venous thromboembolism (VTE) is potentially the most preventable and treatable of poststroke complications, the following question logically arises: are we doing enough to minimize VTE-associated morbidity and mortality after stroke?
Arguably, a further trial of low-dose, low-molecular-weight heparin for a period greater than 2 weeks and with a more systematic evaluation for VTE would be justified but would require large numbers of patients to demonstrate improvements in clinical end points, as Sandercock and Dennis state. In addition, the results of the forthcoming multicenter trial evaluating GCS after stroke will be awaited with great interest. However, in the absence of further data, we suggest that research might profitably be directed toward evaluating a strategy of screening for subclinical DVT, as treatment in selected cases may improve outcome. In our own unit, we are investigating the hypothesis that plasma d-dimers might be a useful discriminator of poststroke DVT status, potentially allowing identification of a patient subgroup with a high risk of proximal DVT who should be selectively imaged. We are also prospectively assessing the true prevalence of PE (clinical plus subclinical) in these patients with magnetic resonance direct thrombus imaging. While a number of studies have reported on the frequency of clinical PE, subclinical events are likely to be considerably more common,R12 and their prevalence in unselected patients is currently unknown.
Intercollegiate Working Party on Stroke. National Clinical Guidelines for Stroke. Royal College of Physicians Document. 2000.
Gubitz G, Sandercock P, Counsell C. Antiplatelet therapy for acute ischaemic stroke (Cochrane review). In: The Cochrane Library, issue 1, 2001. Oxford, UK: Update Software.
Antiplatelet Trialist’s Collaboration. Collaborative overview of randomised trials of antiplatelet therapy: reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ. 1994;308:235–246.
Pulmonary Embolism Prevention Trial Collaborative Group. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) Trial. Lancet. 2000;355:1295–1302.
Barer DH. Dysphagia in acute stroke. BMJ. 1987;295:1137–1138.
Clagett PG, Anderson Anderson FA Jr, Geerts W, Heit JA, Knudson M, Lieberman JR, Merli GJ, Wheeler HB. Prevention of venous thromboembolism. Chest. 1998;114(suppl):531S-560S.