Provoked Right-to-Left Shunt in Patent Foramen Ovale Associates With Ischemic Stroke in Posterior Circulation
Background and Purpose—Right-to-left shunt (RLS) via the patent foramen ovale is an important cause of cryptogenic stroke. The Valsalva maneuver provokes or enhances RLS, but RLS can also occur during normal respiration. This study examined whether the ischemic lesion pattern differs depending on the character of RLS.
Methods—All consecutive patients with a patent foramen ovale (diagnosed by transesophageal echocardiography) who had a cryptogenic stroke and underwent transcranial Doppler–patent foramen ovale test (monitoring of microbubbles in the right middle cerebral artery by transcranial Doppler after hand-agitated saline injection) were divided according to whether RLS was constant (microbubbles detected both at baseline and after the Valsalva maneuver) or provoked (microbubbles detected only after the Valsalva maneuver). The groups were compared in terms of clinical and imaging characteristics.
Results—Seventy-six patients met the eligibility criteria: 50 had constant RLS and 26 had provoked RLS. Provoked RLS patients were significantly younger. The ischemic lesions in provoked RLS patients were located predominantly in the vertebro-basilar circulation (73.1% versus 28.0%; P=0.002), whereas constant RLS patients were more likely to have multicirculatory lesions (16.0% versus 0.0%; P=0.045). After adjusting for confounders, provoked RLS associated independently with a vertebro-basilar lesion location (OR=3.306; P=0.03).
Conclusions—The predominance of posterior-circulatory infarction in provoked RLS patients suggests that the Valsalva maneuver may promote RLS and paradoxical embolization to the posterior circulation.
Paradoxical embolization through the patent foramen ovale (PFO) is considered to be one of the main mechanisms of cryptogenic stroke in patients with PFO (PFO-stroke). The mechanism relates to the incomplete closure of the intra-atrial septum, which allows right-to-left shunt (RLS). Because RLS increases when the pressure gradient between the 2 atria is increased,1 the detection of PFO by transesophageal echocardiography (TEE) or transcranial Doppler (TCD) studies is enhanced when the Valsalva maneuver is performed.2
Detection of microbubbles by TCD after agitated saline injection (TCD-PFO test) is a useful tool to detect PFO. In some patients, the microbubbles are detected only after the Valsalva maneuver, whereas in others, they are also detected during normal respiration. The response of RLS to the Valsalva maneuver may contribute to this difference in microbubble detection. Notably, it has also been shown that the Valsalva maneuver increases the blood flow in the vertebro-basilar circulation and that the predominance of posterior circulatory infarction in PFO-stroke may be explained, at least in part, by this mechanism.3
These 2 observations led us to hypothesize that patients with PFO-stroke whose RLS persists in normal respiration (ie, constant-RLS) will differ from patients with PFO-stroke whose RLS manifests only on the Valsalva maneuver (ie, provoked-RLS) in terms of ischemic lesion distribution.
This study was performed in parallel with a previously published study with a slight modification of the inclusion criteria and study period.4 Briefly, all consecutive patients with PFO-stroke who were admitted to the Asan Medical Center between January 2005 and June 2014 were identified retrospectively. PFO-stroke was tentatively defined as cryptogenic stroke in patients with a PFO who lacked coexisting potential embolic sources, namely, significant stenosis at the corresponding artery, high-risk cardioembolism, or evidence of other arteriopathy. Patients classified as undetermined etiology – negative, according to the Trials of Org 10 172 in Acute Stroke Treatment classification or patients highly suspicious of cryptogenic embolic source, were regarded as cryptogenic stroke patients. Only the patients with PFO definitely observed from TEE and underwent a TCD-PFO test were included. PFO was diagnosed on the basis of TEE findings and PFO >2 mm in diameter combined with or without atrial septal aneurysm or hypermobile septum was regarded as medium- or high-risk PFO, respectively.4 The demographic and risk factors were obtained from a prospectively collected stroke registry. RoPE score which represents the possibility of PFO-stroke was also calculated.5 This study was approved by the Institutional Review Board of our center. Informed consent was not obtained because of the retrospective design of the study.
TCD-PFO Test and Neuroimaging
The TCD-PFO test, diffusion-weighted image, and magnetic resonance angiography were performed within 2 weeks of symptom onset. Normal saline that had been agitated with 1 cc of room air was injected into the antecubital vein and the microbubbles in the right middle cerebral artery were monitored by the M-Mode of TCD (ST3 TCD; Spencer Technologies, WA). The numbers of microbubbles at baseline and after the Valsalva maneuver (40 mm Hg calibrated respiratory strain for 10 seconds) were counted. A patient was deemed to have constant RLS or provoked RLS if microbubbles were detected both at baseline and after the Valsalva maneuver or only after the Valsalva maneuver, respectively. The diffusion-weighted image lesion pattern was classified similarly as in the previous study.4 The pertinent vascular territory of the ischemic lesions was classified as right-carotid, left-carotid, vertebro-basilar, or multicirculatory considering the results of magnetic resonance angiography.
The constant RLS and provoked RLS groups were compared in terms of clinical and imaging variables by chi-square, Fisher exact test, or Student’s t test, depending on the nature of the variable. Binary logistic multivariate analysis was performed to test whether provoked RLS associated independently with lesions in the vertebro-basilar circulatory after adjustment for potential confounders (P<0.20). Statistical analyses were performed by using SPSS for Windows, Version 17.0 (SPSS Inc, Chicago, IL).
During the study period, 10 668 patients were admitted to our center because of ischemic stroke. Of these, 1470 patients (13.8%) were diagnosed with cryptogenic stroke. Both TEE and the TCD-PFO test were performed on 245 patients (16.6%). PFO was observed in 76 of these patients (31.0%). The characteristics of those 76 patients were similar to those of patients who received only TEE and were diagnosed as PFO-stroke (Table I in the online-only Data Supplement). Of these, 50 (65.8%) had constant RLS and 26 (34.2%) had provoked RLS.
The patients with provoked RLS were younger on average than the patients with constant RLS (P=0.03; Table 1). The 2 groups did not differ in terms of risk factors or lesion patterns, but the provoked RLS group had a significantly higher mean RoPE score than the constant RLS group (P=0.03).
The 2 groups also differed significantly in terms of lesion location (P=0.004): the provoked RLS group was more likely to have ischemic lesions located in the vertebro-basilar circulation (73.1% versus 28.0%; P=0.002), whereas the constant RLS group was more likely to have multicirculatory lesions (16.0% versus 0.0%; P=0.045; Figure). Number of microbubbles detected at baseline was higher in patients with multicirculatory lesions (Table II in the online-only Data Supplement). Multivariate analysis adjusted for potential confounders revealed that provoked RLS (OR=3.306; P=0.03) and age (OR=0.955; P=0.03) associated independently with lesion location in the vertebro-basilar circulation (Table 2).
In the present study, 34% of the PFO-stroke patients had provoked RLS. The ischemic lesions in these patients were located predominantly in the vertebro-basilar circulation, whereas multicirculatory lesions occurred more frequently in patients with constant RLS, who had high-grade shunting at baseline.
Old asymptomatic juxtacortical lesions associate with RLS and are frequently observed in the anterior circulation.6 However, the index strokes in patients with PFO occur more frequently in the vertebro-basilar circulation.4 The high prevalence of posterior circulatory infarction has been explained by the lower innervation of the sympathetic nervous system and the increased blood flow to the posterior circulatory area after the Valsalva maneuver.3 Therefore, the Valsalva maneuver may provoke RLS and simultaneously increase the blood flow to the vertebro-basilar circulation, thereby causing PFO-stroke to predominate in the posterior circulation (Figure).
Small PFOs are hard to detect by TEE, and their provocation by the Valsalva maneuver is critical for detecting them.7 However, it can be difficult to perform the Valsalva maneuver during TEE, especially in elderly patients with severe neurological deficits. Because provoked RLS is common (in our cohort, a third of the patients had provoked RLS), the TCD-PFO test plays an important complementary role to TEE and should be considered when there is suspicion of a PFO-stroke, especially if the patient performs the Valsalva maneuver poorly.
Our study has several limitations that stem from its small sample size and its retrospective design. Despite these limitations, the present study suggests that the Valsalva maneuver may associate with the high prevalence of posterior circulation infarction in PFO-stroke and that the TCD-PFO test may complement TEE.
Sources of Funding
This study is supported by a grant from the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare Republic of Korea (HI10C2020).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.007453/-/DC1.
- Received September 17, 2014.
- Revision received September 17, 2014.
- Accepted September 24, 2014.
- © 2014 American Heart Association, Inc.
- Homma S,
- Sacco RL
- Hayashida K,
- Fukuchi K,
- Inubushi M,
- Fukushima K,
- Imakita S,
- Kimura K
- Kim BJ,
- Sohn H,
- Sun BJ,
- Song JK,
- Kang DW,
- Kim JS,
- et al