Send to

Choose Destination
See comment in PubMed Commons below
J Stroke Cerebrovasc Dis. 2009 Sep-Oct;18(5):343-8. doi: 10.1016/j.jstrokecerebrovasdis.2008.12.001.

Transcranial Doppler and transesophageal echocardiography: comparison of both techniques and prospective clinical relevance of transcranial Doppler in patent foramen ovale detection.

Author information

Fondazione IRCCS (Scientific Institute for Research, Hospitalization and Healthcare) Istituto Neurologico C. Besta Milan, Italy.



Patent foramen ovale (PFO) has been investigated in several conditions apart from cryptogenic ischemic stroke. Contrast transesophageal echocardiography (cTEE) is the gold standard for the diagnosis, although it has some known limitations. Contrast transcranial Doppler (cTCD) allows a semiquantitative estimation of right-to-left shunt (RLS) volume. The aims of our study were to confirm the diagnostic accuracy of cTCD in PFO diagnosis and to compare the abilities of cTCD and cTEE to detect a RLS and PFO, respectively, under normal breathing. The latter could represent an important feature for its clinical significance.


A total of 100 consecutive patients (59 women and 41 men, age 46 +/- 12 years) were evaluated after stabilized ischemic stroke/transient ischemic attack, migraine, and lacunae, and before neurosurgery in sitting position. All patients undertook cTEE and cTCD, at rest and under Valsalva maneuver (VM). cTEE under VM was the reference standard. A categorization of patients and a semiquantitative cTCD classification were proposed.


In all, 63 of 100 patients had PFO diagnosed by cTEE. A general concordance of up to 90% between both techniques was found. cTCD sensitivity and specificity were 96.8% and 78.4%, respectively. In 17 of 100 patients with cTEE-proven PFO under VM, cTCD and cTEE detected RLS at rest in 75% (95% confidence interval [CI] 62%-85%) and 48% (95% CI 35%-61%) of cases, respectively (P < .001). cTEE disclosed RLS at rest in about 71% (95% CI 9%-42%) of cTCDs showing a "shower-curtain" pattern and only in about 22% (95% CI 52%-85%) of those cTCDs without that pattern.


In diagnosing PFO, cTCD has a good accuracy compared with cTEE. To detect a RLS at rest, cTCD appears to be more sensitive than cTEE. The latter resulted positive under normal breathing, mostly in cases of significant RLS at cTCD. Our results point out the impact of cTCD in the evaluation of RLS volume, thus aiding, in association with the anatomic details by cTEE, in the prevention of the occurrence or recurrence of paradoxical embolism in individuals with and without cerebrovascular diseases. The combination of cTEE and cTCD could be considered the real gold standard for PFO in the near future.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Support Center