Article Text

Download PDFPDF

Embolic stroke: the heart–brain connection reassessed
  1. M Brainin1,
  2. C Stöllberger2
  1. 1Department Neurology Donauklinikum and Centre Neurosciences, Donau-Universität, Maria Gugging, Austria
  2. 22nd Medical Department, Krankenanstalt Rudolfstiftung, Vienna, Austria
  1. Correspondence to:
 Professor Michael Brainin
 Donauklinikum and Donau-Universität, Hauptstrasse 2, A-3400 Maria Gugging, Austria;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Transesophageal echocardiography possibly underutilised in stroke?

Autopsy series published a century ago, showed that cerebral emboli arising from the major vessels and the heart are an important pathophysiological cause of stroke.1 Since then not much has changed. According to one recent and comprehensive textbook2 embolism accounts for up to 60% of all ischemic strokes. Cardiac conditions associated with such cerebral emboli are atrial fibrillation in 45% of the cases, followed by acute myocardial infarction, ventricular aneurysm, rheumatic heart disease, prosthetic cardiac valves, aortic atheroma, and other causes.

With the availability of transesophageal echocardiography (TEE) in many stroke centres, intracardiac emboli resting in the left atrial appendage or left atrium can now be readily identified and treatment with anticoagulation can be installed.3 In this issue Sen et al 4 (see page 1421) have reassessed the cardiac risk factors in a systematic study using TEE within a week of stroke onset in a largely unselected cohort of stroke patients. Their aim was to set up criteria for refinement of selection for TEE in stroke patients due to clinical or laboratory characteristics. Analysis of their consecutive 151 stroke patients showed that demographic factors, clinically assessable risk factors, as well as blood coagulation parameters did not differ between patients with and without a cardiac thrombus, whereas several other investigational findings were in favour of finding a cardiac thrombus by means of TEE—stroke appearance on CT/MRI compatible with a large vessel stroke, cardiac ischemia visible on ECG, and systolic left ventricular dysfunction.

What the authors found was only partly what they were looking for. Firstly, they found a 26% prevalence rate (40 out of 151 patients) of intracardiac thrombus (mostly in the left atrial appendage), which is higher than in most previous reports. The authors attribute this high rate to the routinely used multiplanar TEE technique (whereas the lower prevalence reported in previous studies had mostly used monoplanar or biplanar probes).

Secondly, Sen et al found a similarly high prevalence in lacunar type infarcts (24%, 16 out of 66 patients), which implies that the presence of small artery infarcts on neuroimaging does not exclude the necessity to perform TEE in a considerable proportion of lacunar stroke patients.

One other frequently encountered clinical uncertainty comprises the necessity to perform TEE in stroke patients with atrial fibrillation. It is commonly held that the presence of atrial fibrillation obviates TEE due to the fact that anticoagulation would have been the therapy of choice anyway. Thus, it is argued that it is not necessary to add additional costs, discomfort, or a small but notable risk using a semi-invasive technique such as TEE. Furthermore, clinical findings of increased age and hypertension have been shown to be more useful than TEE findings to assess embolic risk in atrial fibrillation patients.5 Others might argue that it is simply important to localize, specify, and treat the cause of the stroke. The demonstration of an underlying pathology such as an intracardiac thrombus, the documentation of its size and location, plus any additional finding of cardiac structural abnormality or associated intra-cardiac pathology, might give additional weight to initiate anticoagulant therapy that otherwise might not have been indicated in such a mandatory way. The latter argument is supported by other reports, among them one by the same group6 that accumulates evidence to show the beneficial effect of anticoagulation on thrombus size reduction. Up until now, however, it has not been proven by any prospective study that stroke patients with intracardiac thrombi diagnosed by TEE have a higher recurrence rate than stroke patients without thrombi.

In spite of increasing awareness of the benefits of TEE in treating stroke patients there is still a need to do what this study has set out to achieve: to develop a clinically reliable set of criteria for investigating patients by means of TEE and to compare the results by looking for markers for urgency, intensity, or duration of anticoagulation. To do this properly a predefined and prospective series will have to be performed to validate these criteria and give some predictive rates. Until then the impression prevails that TEE is underused in acute stroke.

Transesophageal echocardiography possibly underutilised in stroke?


Linked Articles