Extracranial Arterial Dissection
Anticoagulation Is the Treatment of Choice: Against
Carotid artery dissection (CAD) was first recognized as a cause of ischemic stroke in the mid-1950s,1 although there were pathological reports on the condition dating from 1872.2 The incidence of extracranial internal CAD is ≈2 to 3 per 100 000 per year.3 For extracranial vertebral artery dissection (VAD), community-based epidemiological data are not available. Data from hospital-based case series show the extracranial VAD incidence to be estimated ≈1 to 1.5 per 100 000 per year.3 The recurrence rate of stroke in CAD is estimated to be <1% per year.4
The pathogenesis of ischemic strokes in CAD is still a matter of debate. Former studies showed a high frequency of angiographic findings suggesting embolism with cerebral arterial branch occlusions in ≈15%,5 and cases with free-floating intraarterial thrombi were occasionally reported. Recently, transcranial doppler monitoring studies revealed microembolic signals downstream of the dissected arteries.6 In extracranial internal CAD, high-intensity transient signals suggesting embolism were detected in the middle cerebral artery in up to 60% of cases. Together with brain imaging features of stroke lesions, one may assume that extracranial arterial dissections cause embolic stroke. However, most data are based on small, presumably highly selected case series. Hence, from these impressions many authors suggested that immediate anticoagulation may be the right thing to do.7
Although CAD has been known to cause ischemic stroke for almost 50 years, at present no data based on randomized controlled trials are available.8 But there are arguments against routine anticoagulation. First: extracranial CAD is caused by intramural arterial bleeding. At least theoretically, anticoagulation may promote enlargement of the mural bleed. This may lead to worsening of the hemodynamic condition. It may also cause intracranial bleeding, mostly feared in cases with VAD. In extracranial VAD, there are single case reports about subarachnoid hemorrhage as the presenting feature in which any antithrombotic treatment, and particularly anticoagulation, is considered deleterious. Delayed occlusion of the internal carotid artery during heparin therapy have recently been reported in 5 of 20 patients with extracranial internal CAD. Thus, the likelihood for delayed ICA occlusion seems to increase with higher degrees of anticoagulation.9 Second: recurrent stroke does not very often occur in CAD and no data are available for VAD. The magnitude of the potential therapeutic benefit of anticoagulation in reducing the occurrence of ischemic strokes may be very limited and moreover may be counteracted by bleeding complications. Third: at least some strokes may be due to hemodynamic impairment, rather than thromboembolism, especially when collaterals from the circle of Willis are lacking.9 Stroke occurrence or recurrence was reported in extracranial internal CAD patients treated with antiplatelets7 as well as in patients with sufficient anticoagulation,10 indicating that protection even with anticoagulation—not surprisingly—is not absolute. For other stroke pathogeneses it was shown by means of a systematic meta-analysis that immediate anticoagulant therapy in patients with acute ischemic stroke is not associated with net short- or long-term benefit.11 Forth: a recent Cochrane meta-analysis targeting the effect of antithrombotic drugs for extracranial internal CAD did not show any data from controlled trials. Further anticoagulants were not superior for primary outcomes, such as death or dependency. Twenty-six studies including 327 patients (who either received antiplatelets or anticoagulants) were eligible. Two of 109 patients (1.8%) treated with antiplatelets were reported dead, and 4 of 218 (1.8%) treated with anticoagulants were reported dead. The weighted estimates across studies showed that the likelihood of death within the follow-up period does not differ between both treatment groups as indicated by a Peto odds ratio of 1.59, with a 95% confidence interval of 0.22 to 11.59.8
To conclude, there is no evidence from any controlled data that anticoagulation for extracranial internal CAD might be helpful. It could be that, in a subset of patients with proven embolism, immediate anticoagulation may protect from first or recurrent strokes. But this has not been shown yet. Furthermore, there is even less data for VAD. As there are safety concerns and the benefit–risk ratio is not established, patients with CAD are unlikely to be harmed by antiplatelets, and there seems little justification for giving anticoagulants as a first line therapy in all patients. Therefore it is more appropriate not to recommend anticoagulation on a routine basis for CAD or VAD.
As a consequence, a large randomized controlled trial comparing anticoagulants with antiplatelets is desirable and ethically justified. Its protocol should include a stringent definition of carotid dissection, a standardized diagnostic protocol, strictly random allocation to different types of antithrombotic treatment, as well as accurate and unbiased assessment of outcome.
- Received April 1, 2005.
- Accepted May 9, 2005.