Is Doppler ultrasound sufficient as the sole investigation before carotid endarterectomy?

Abstract

Background and aims: Doppler ultrasound (DU) of the extracranial carotid arteries has been advocated as the sole imaging modality in carotid endarterectomy (CE) candidates. However this approach fails to identify patients with potentially significant intracranial disease, at high risk of stroke and death. Therefore, many stroke clinicians recommend angiography after screening DU. We aimed to identify the proportion of cases referred for CE in whom the identification of intracranial disease could have altered management. Methods: Two neuroradiologists, blinded to the clinical history, reviewed the films of 111 CE candidates, predominantly of Caucasian background, who had undergone carotid angiography after screening DU. Intracranial stenoses >50% luminal diameter, incidental aneurysms and non-atherosclerotic lesions were documented. Demographic and epidemiological data were collected. Results: Of the 111 patients, 87 had >50% extracranial stenoses although two thirds were asymptomatic. Intracranial stenotic lesions were recorded in 29% of patients. Over half of these were tandem lesions, distal to an extracranial stenosis. Aneurysms were found in 4.5% of patients. Conclusions: DU alone would have failed to detect significant intracranial disease in nearly a third of cases. These patients are at high risk of stroke. The identification of this group allows more aggressive stroke prevention therapy.

Introduction

In Australia, it is common practice to perform a carotid endarterectomy (CE) solely on the result of a Doppler ultrasound (DU) of the extracranial carotid artery.¹ This approach has become popular because this non-invasive investigation is nearly as accurate as conventional angiography in the evaluation of the extracranial portion of the artery,[2], [3] and avoids the approximately 1% risk of stroke or transient ischaemic attack (TIA) associated with catheter angiography.⁴ The disadvantage is that the intracranial circulation is not imaged.

Intracranial atherosclerosis is receiving more attention in reviews of carotid disease. In some ethnic groups, particularly Asians, it is the most common vascular lesion identified in patients presenting with either strokes or TIAs.[5], [6] It is also more prevalent amongst those with hypertension or diabetes.[7], [8], [9]

The prognosis of patients with intracranial stenoses is extremely poor. Their annual stroke and mortality rate approaches 25%, twice the rate in patients with severe symptomatic extracranial stenoses.[10], [11], [12] Although this poor natural history of intracranial disease might be expected to influence the outcome of CE, the results of studies have been contradictory. The North American Carotid Endarterectomy Trial (NASCET) indicated that patients with symptomatic tandem disease probably benefit from CE in the long term.¹³ However their immediate risk of stroke or death from surgery may also be greater.¹⁴ Ideally, a randomized prospective study is needed to better define the risk benefit ratio of surgery for patients with tandem intracranial and extracranial lesions, but such a trial is most unlikely to be organised. A further argument for intracranial imaging before CE is the possibility of other distal pathologies, such as intracranial aneurysms.¹⁵

At our institution, it has been the policy to investigate patients with cerebral angiography prior to CE, in those with moderate or severe extracranial carotid disease identified by a screening DU. Because of the controversy concerning the need to image the intracerebral circulation before CE, we aimed to assess the prevalence of significant intracranial disease, tandem disease and cerebral aneurysms in CE candidates, over a 12 month period.