Patent Foramen Ovale and Stroke
2005; Lippincott Williams & Wilkins; Volume: 112; Issue: 7 Linguagem: Norueguês
10.1161/circulationaha.104.524371
ISSN1524-4539
AutoresShunichi Homma, Ralph L. Sacco,
Tópico(s)Traumatic Brain Injury and Neurovascular Disturbances
ResumoHomeCirculationVol. 112, No. 7Patent Foramen Ovale and Stroke Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBPatent Foramen Ovale and Stroke Shunichi Homma, MD and Ralph L. Sacco, MD Shunichi HommaShunichi Homma From the Division of Cardiology (S.H.) and Neurological Institute (R.L.S.), Columbia University, New York, NY. and Ralph L. SaccoRalph L. Sacco From the Division of Cardiology (S.H.) and Neurological Institute (R.L.S.), Columbia University, New York, NY. Originally published16 Aug 2005https://doi.org/10.1161/CIRCULATIONAHA.104.524371Circulation. 2005;112:1063–1072This review summarizes the current state of knowledge about the relationship of patent foramen ovale (PFO) with ischemic stroke. Initial sections discuss the studies that identified this association. Subsequent sections discuss the detection techniques for PFO and other variables that may cause a PFO to be a conduit of paradoxical embolization. Finally, a section is devoted to summarizing the studies that assessed the strategies for preventing recurrent ischemic events in patients with PFO.Cryptogenic Stroke and PFOIn ≈40% of patients with acute ischemic stroke, the cause remains undefined.1 PFO is a hemodynamically insignificant interatrial communication present in >25% of the adult population. During fetal life, because the lungs do not receive blood flow, blood returning to the right atrium is shunted through a PFO to the left atrium. Postnatally, PFO closes spontaneously in ≈75% of the population. However, in a portion of adults, by maintaining a direct communication between the right- and left-sided circulation, PFO can serve as a conduit for paradoxical embolization.In 1877, Cohnheim2 described the association of PFO with stroke in a young woman with cerebral arterial embolism. However, it has been difficult to diagnose PFO in vivo until the development of echocardiography and its ability to image the interatrial shunting with an injection of saline contrast. With the use of contrast echocardiography, a strong association of cryptogenic stroke with PFO has become evident in patients <55 years of age (Table 1).3–8TABLE 1. Relationship of Cryptogenic Stroke With PFO in Younger and Older PatientsStudyPatients, nAge, yPFOPCryptogenic, % (n/total)Control, % (n/total)*Includes different stroke subtypes.Younger patients Lechat et al326<5554 (14/26)10 (10/100)<0.001 Webster et al434<4056 (19/34)15 (6/40)<0.001 Cabanes et al564<5556 (36/64)18 (9/50)<0.0001 De Belder et al6*39<5513 (5/39)3 (1/39)… Di Tullio et al721<5547 (10/21)4 (1/24)<0.001 Hausmann et al818<4050 (9/18)11 (2/18) 5520 (13/64)5 (3/56) 5538 (9/24)8 (6/77) 4015 (3/20)23 (23/98)NS Jones et al1057>5018 (10/57)16 (29/183)NS Total……21 (35/165)15 (61/414)Because stroke occurs more frequently in older population, with only 3% of cerebral infarctions occurring in patients <40 years of age, the number of stroke patients with PFO ≥40 years of age is much larger than in the younger patients.9 Several studies reported the association of PFO with cryptogenic stroke in older patient populations.6,7 However, this has not been seen in other studies (Table 1).8,10 Therefore, although the association between cryptogenic stroke and PFO is established among the younger population, it is not clearly established in the older population. This also has been substantiated in a meta-analysis of studies relating to atrial abnormalities and stroke.11In support of PFO as a conduit for paradoxical embolization, there are occasional case reports demonstrating venous thrombi trapped in a PFO in patients with central or systemic embolization.12–14 Nevertheless, other possible mechanisms of stroke cannot be excluded. Given that a PFO can be a tunnel-like structure with possibly a stagnant area of flow, in situ thrombus fromation may occur. Also, patients with PFO may be susceptible to atrial arrhythmias with possible intra-atrial thrombus formation, leading to stroke.15Detection of PFOContrast EchocardiographyTransthoracic (TT) echocardiography and transesophageal (TE) echocardiography with saline contrast injection are widely used to detect PFO. A PFO is judged to be present if any microbubble is seen in the left-sided cardiac chambers within 3 cardiac cycles from the maximum right atrial opacification. Figure 1 demonstrates the appearance of microbubbles in the left-sided cardiac chambers after the venous injection of contrast material in TT imaging. Injection is performed with and without the Valsalva maneuver. Coughing during injection may increase the sensitivity for detecting PFO over that achieved by the Valsalva maneuver.16 Use of harmonic imaging with TT echocardiography and contrast injection may also increase the sensitivity of PFO detection.17,18 Saline contrast injection can be performed while imaging the heart with a TE probe. Again, PFO is judged to be present with the visualization of microbubbles in the left atrium within 3 cardiac cycles from the right atrial opacification. Figure 2 demonstrates the passage of microbubbles from the right atrium into the left atrium through PFO as demonstrated by TE echocardiography. Download figureDownload PowerPointFigure 1. In this TT echocardiographic view, appearance of microbubbles in left atrium (LA) and left ventricle (LV) is demonstrated after saline contrast injection in patient with PFO. Arrows point to microbubbles. RA indicates right atrium; RV, right ventricle.Download figureDownload PowerPointFigure 2. Vertical TE echocardiographic view of fossa ovalis area demonstrating passage of microbubbles through PFO (arrow) from right atrium (RA) into left atrium (LA). Separation of septum primum (SP) from septum secundum (SS) is visualized and can be measured.Location of the contrast material injection can influence the chance of detecting a PFO. Contrast material injected into the lower extremities has a higher chance of crossing a PFO because the flow from the inferior vena cava is directed toward the fossa ovalis as it enters the right atrium.19 Doppler color-flow detection of a PFO is possible with TE; however, this technique may not be as sensitive as contrast injection.20Transcranial Doppler in PFO DetectionParadoxical embolization through a PFO is considered to be a mechanism for stroke associated with a PFO. In support, direct demonstration of embolism through a PFO to the cerebral circulation has been demonstrated. Figure 3 demonstrates the baseline flow pattern obtained by transcranial Doppler (TCD) in the middle cerebral artery and that seen after saline contrast injection in a patient with a PFO. However, detection of microbubbles in the cerebral circulation by TCD does not necessarily imply the presence of a PFO. Any right-to-left shunt such as that resulting from ventricular septal defect or intrapulmonary shunt may result in the detection of microbubbles in the cerebral circulation by TCD. As a result, TCD cannot identify the site of right-to-left shunt, whereas TT or TE studies provide this information.21,22 Several studies performed contrast TT, TE, and TCD imaging in the same patient group to compare the sensitivity of the techniques (Table 2).23–30 The TE contrast study is the most sensitive diagnostic test available for detecting a PFO, followed by TCD and TT contrast studies (P 10-mm protrusion beyond the plane of the septum into the left or right atrium.57,58 Although the definition varies somewhat in different series, the prevalence in the general population is estimated with TT imaging to be only 0.23% (Table 4).59–69 A considerably higher prevalence of 4.6% is noted among those referred for TE echocardiography, most likely because of the higher sensitivity of the TE technique for imaging the septal area and the selection bias for patients referred for TE echocardiography (Table 4).69–73TABLE 4. Prevalence StudiesStudyASA PatientsPrevalence, %Prevalence of ASA by TT study Hanley et al5980/36 2000.22 Gallet et al6010/48400.21 Longhini et al6123/40000.57 Bewick and Montague626/47000.12 Wolf et al6312/7241.7 Belkin et al6436/69790.5 Brand et al6535/35001.0 Roudant et al6644/62 5400.08 Katayama et al6726/20741.2 Oneglia et al6838/40310.94 Schneider et al6920/12 1370.16 Total330/141 7250.23Prevalence of ASA by TE Study Schneider et al6923/7653.0 Schreiner et al707/3402.1 Zabalgoitia et al7120/19910 Pearson et al7232/4107.8 Mirode et al7332/7514.2 Total114/24654.6PFO prevalence among patients with ASA Mügge et al57 (TE echo)106/19554 Hanley et al59 (TE echo)24/4949 Schneider et al69 (TE echo)17/2277 Zabalgoitia-Reyes et al71 (TE echo)17/2085 Pearson et al72 (TE echo)20/2969 Silver et al76 (autopsy)8/1650 Mattioli et al77 (TE echo)39/4489 Burger et al78 (TE echo)18/3256 Homma et al79 (TE echo)44/6964 Total293/47662The prevalence of ASA is greater among patients with embolic events.7,71,73–75 It is also well known that ASA is associated with PFO, with ≈60% of patients with ASA having a PFO (Table 4).57,59,69,71,72,76–79 Additionally, the PFOs seen in the presence of ASA tend to be large compared with those seen without associated ASA.79,80 Thus, the association of ASA with embolic events is likely based on the high prevalence of large PFOs. Because an ASA is usually highly mobile, protruding from right to left atrium, it is unlikely that a thrombus forms in the ASA itself. This is corroborated by a rare finding of thrombus associated with ASA in a large series of patients.57Eustachian Valve and Chiari's NetworkThe eustachian valve is a membrane-like structure in the right atrium, a remnant of the right valve of the sinus venosus that directs blood flow from the inferior vena cava to the fossa ovalis area in the fetus.81 Among adults, a eustachian valve can cause a significant right-to-left shunt in the presence of an interatrial communication by altering the blood flow pattern.82,83 Prominent eustachian valve is also more commonly found among patients with presumed paradoxical embolism than in control patients.84 The presence of Chiari's network and filamentous strands in the right atrium is also associated with the presence of PFO.79,85 Therefore, the presence of atrial anatomic variants that can promote flow from the inferior vena cava toward the PFO may increase the chance of paradoxical embolization beyond that associated with PFO size.HemodynamicsIn addition to the atrial anatomic variables, hemodynamic alterations play a major role in determining the chances of paradoxical embolization. Although transiently higher right atrial pressure can occur during normal cardiac cycle,86 cardiac lesions more consistently elevating right atrial pressure will increase the chance of right-to-left shunt. As a result, paradoxical embolization is often reported in patients with pulmonary embolism.87,88 Similarly, patients with right ventricular infarction89 or severe tricuspid regurgitation90 or those on a mechanical left ventricular assist device have an increased right-to-left shunt through a PFO.91 Although a right-sided pressure elevation can increase the flow across PFO, left-sided pressure elevation will decrease it.92Venous Thrombus and Hypercoagulable StateFor paradoxical embolization to occur, a source of thrombus is needed. A significant stroke can result from an arterial occlusion by an embolus as small as 1 mm in diameter.93 A greater prevalence of deep venous thrombus is observed in one study in patients with cryptogenic stroke compared with a control group.94 However, several other studies do not corroborate this finding.95–97 More recently, pelvic vein thrombi are reported to be found more frequently in young patients with cryptogenic stroke compared with those with more defined causes of stroke.98 Pelvic veins and abdominal veins are not studied routinely in patients with cryptogenic stroke and PFO, and these areas may harbor thrombus. Finding of a venous thrombus strengthens the possible role of PFO as a conduit for paradoxical embolization and will affect treatment strategy. Patients with a tendency toward venous thrombus formation may be exposed to a higher chance of paradoxical embolization in the presence of PFO. Several studies report a higher frequency of prothrombotic states such as G20210A and factor V Leiden mutations in patients with cryptogenic stroke and PFO.99–102 A recent study demonstrates a higher recurrent event rate in older cryptogenic stroke patients with PFO compared with similarly aged cryptogenic stroke patients without PFO.103 This may be due in part to the presence of occult thrombi in older patients compared with the younger patients.In summary, variation in PFO size, right atrial anatomy, hemodynamic conditions, and potential source of an available thrombus all play a part in influencing the chances of paradoxical embolization.Recurrent Stroke PreventionPossible treatment modalities to prevent recurrent events among stroke patients with a PFO include medical treatment with warfarin or antiplatelet agents, percutaneous PFO closure, and surgical PFO closure. In this section, recurrent event rates from the available outcome studies are summarized. Recurrent event rates are estimated from published studies in MedLine since 1990. The number of subjects, mean age, and follow-up time are obtained for each publication. If there are duplications of patient population in studies, only one is included or, to the extent possible, correction in numbers is made. Inclusion criteria are (1) presumed paradoxical embolic events without obvious cause, including cryptogenic stroke, transient ischemic attack (TIA), or other embolic arterial events such as peripheral embolism; (2) documented PFO on echocardiography, either TT or TE; and (3) original manuscript available in English.For the purpose of creating a summary table, all-cause mortality is included in the table. Thus, when discrepancies in the number exist in some of the cells compared with the published manuscript, it is due to the difference in the criteria used for end points. The number of events in each study was summed to obtain the total number of events. Similarly, the total time at risk was determined by summing the number of subjects multiplied by the mean follow-up time for each study. Event rates were calculated as the ratio of the total number of events to the total years of follow-up divided by 100 to yield event rates per 100 patient-years. The 95% CI for the pooled event rates was determined by assuming that observed events followed the Poisson distribution. For the studies of the effect of medical therapy on stroke recurrence or TIA, homogeneity of event rates was assessed using Cochran's Q test after excluding the single study with 13 subjects and no events. A significant lack of homogeneity was not detected for either recurrent stroke (Q=4.35, P=0.74) or TIA (Q=6.89, P=0.44). Similar tests of homogeneity were not performed for percutaneous closure or surgery because of the small number of events. Tables are made that include number of patients in the study, mean age, and mean follow-up in months. In terms of the end points, stroke, any-cause death, TIA, stroke or death, stroke or TIA, and stroke death or TIA are included. A similar summary, but with somewhat different criteria, has been recently published.104Medical TherapyA total of 943 patients are considered in the analysis (Table 5).33,54,105–111 The mean age of the patients is 45 years; mean duration of follow-up is 33 months. Medications used include both warfarin and aspirin. There were 15 deaths (any cause), 52 strokes, and 59 TIAs. The annual rate of stroke is 1.98% (95% CI, 1.48 to 2.60) and of stroke or death is 3.12% (95% CI, 2.32 to 4.11). Individual studies demonstrate variable recurrent event rates. This is due in part to the difference in the age of subjects; younger cryptogenic stroke patients with PFO have a significantly lower event rate compared with the older cryptogenic stroke patients with PFO.103 Only one study randomized patients to warfarin or aspirin therapy.54 In this study, there is no difference in event rates between those with and without PFO on medical therapy. When patients treated with warfarin are compared with those treated with aspirin, there is no significant difference, although the study is not adequately powered for this purpose. TABLE 5. Summary Table of Medical Therapy StudiesStudyPatients, nMean Age, yMean Follow-Up, moStroke, nDeath, nTIA, nStroke or Death, nStroke or TIA, nStroke, Death, or TIA, nHausmann et al334446591…2…3…Homma et al549855221048141822Hanna et al1051343270…0…0…Mas and Züber10610739222…3…5…Bogousslavsky et al1071294436858131621Cujec and Mainra1085238467…12…19…De Castro et al10974533155310813Mas et al11026740381219132122Nedeltchev et al11115951297…14…21…Total94345335215595011178Events per 100 patient-years………1.980.942.243.124.224.8695% CI………1.48–2.600.53–1.551.71–2.892.32–4.113.43–5.013.78–5.94Some studies identified the combination of ASA and PFO as a predisposing factor for increased recurrent event rates, whereas another has not.79,106,110 Major bleeding risk from medical therapy, particularly from the use of warfarin, is estimated at 1% to 2% annually and minor bleeding risk 10% to 20%, higher in those on warfarin compared with aspirin.112,113Percutaneous Closure of PFOBecause PFO represents a repairable lesion, interest in closing them is high. Currently, the most commonly used devices in the United States are the Amplatzer PFO Occluder (AGA Medical) and CardioSEAL (NMT Medical) devices.114,115 The Amplatzer device is made of self-expanding nickel-titanium alloy wire mesh with double disks that contain inner polyester fabric patches. The CardioSEAL device is constructed from a low-profile nickel-cobalt alloy framework shaped like an umbrella to which a knitted polyester fabric is attached. Under the Humanitarian Device Exemption (HDE) program of the US Food and Drug Administration (FDA), these devices are approved for use in patients with recurrent cryptogenic stroke caused by presumed paradoxical embolism through a PFO who have failed therapeutic dosage of oral anticoagulants.Using the same criteria as for Table 5, Table 6 shows the event rates in patients undergoing percutaneous PFO closure.53,116–126 Again, when overlap in patient population occurs, only one study appears in Table 6 or the numbers are adjusted. A total of 1430 patients are considered in the analysis. Of note, some of the studies include patients receiving a device other than an Amplatzer or a CardioSEAL, and many of studies are performed outside the United States where the devices can be clinically used. The mean age of the patients is 46 years; the mean duration of follow-up is 18 months. There is variable use of warfarin or antiplatelet agents after closure. There were 4 deaths, 4 strokes, and 32 TIAs. The annual rate of stroke is 0.19% (95% CI, 0.05 to 0.49) and of stroke or death 1.15% (95% CI, 0.46 to 2.37). TABLE 6. Summary Table of Percutaneous PFO Closure StudiesStudyPatients, nMean Age, yMean Follow-Up, moStroke, nDeath, nTIA, nStroke or Death, nStroke or TIA, nStroke, Death, or TIA, nPeripheral Embolism, n*Number of patients is adjusted.Bridges et al53363980…4…4……Ende et al1161040320…0…0……Hung et al117*284631023225…Sievert et al11828147122072990Beitzke et al11916240190…3…3……Butera et al1203548120…0…0……Wahl et al12115259201…6…7…2Martin et al122*1104728121334…Du et al1231842260…0…0……Braun et al12427645150…8…8…0Bruch et al1266648200…0…0…0Onorato et al12625648190…0…0……Total143046184432736182Events per 100 patient-years………0.190.661.521.151.622.950.2095% CI………0.05–0.490.18–1.691.04–2.150.46–2.371.13–2.241.75–4.660.02–0.72Complications from device implantation include major complications such as death, major hemorrhage, cardiac tamponade, and fatal pulmonary emboli. These occur in ≈1.5% of the patients.104 Minor complications such as atrial arrhythmias, device arm fractures, device embolization, device thrombosis, ECG changes, and AV fistula formation are reported in 7.9%.104 Thrombus formation on the device may depend largely on the device used.127 PFO represents a potential space, bordered by 2 overlapping membranes, some with tunnel-like anatomy. "PFO-specific" devices may simplify closure in the future, and there are a variety of newer devices and methods conceptualized and tested to close a PFO. These include anatomically fitting devices, staplers, and tissue glues.Surgical Closure of PFOTable 7 shows the event rates in surgically treated patients with PFO.128–132 A total of 161 patients are considered in this analysis. The mean age of the patients is 43 years, and the mean duration of follow-up is 22 months. There were 2 deaths, 1 stroke, and 11 TIAs. The annual rate of stroke is 0.34% (95% CI, 0.01 to 1.89) and of stroke or death is 0.85% (95% CI, 0.10 to 3.07). However, the number of patients in this analysis is small, and with the advent of percutaneous closure devices, the surgical approach is no longer widely used. Even with the use of a minimally invasive approach,133 it is very likely that the surgical approach will be replaced by percutaneous approaches. TABLE 7. Summary Table of Surgical PFO Closure StudiesStudyPatients, nMean Age, yMean Follow-Up, moStroke, nDeath, nTIA, nStroke or Death, nStroke or TIA, nStroke, Death, or TIA, nHarvey et al128447140…0…0…Devuyst et al1293038230…0…0…Homma et al130284119113144Giroud et al13184612000000Dearani et al132914424018189Total1614322121121213Events per 100 patient-years………0.340.853.710.854.055.5595% CI………0.01–1.890.10–3.071.8–5-6.640.10–3.072.09–7.072.96–9.49Comparison of Modalities and Current TrialsAlthough Tables 5 through 7 demonstrate summary outcome measures for the different treatment approaches, there are no direct randomized comparisons of treatment modalities. In a collective analysis, there are no convincing data to indicate that the presence of PFO increases recurrent events in medically treated patients.134 Whether PFO closure decreases the event rate further remains unanswered. Some analyses suggest possible superiority of percutaneous closure compared with medical therapy.135Using our tables, we compared medical therapy with percutaneous closure, with stroke or stroke and TIA as end points. For both comparisons, percutaneous closure gives lower event rates compared with medical therapy (P<0.0001). However, very importantly, indirect comparison of medical treatment and percutaneous closure is very difficult to interpret. Inclusion criteria for the studies reviewed are not uniform, and definitions of what constitutes a cryptogenic stroke or TIA vary widely among studies. The age of subjects is variable, which may significantly affect the observed event rates. Many of the studies are also subject to potential selection bias and do not use independent blinded adjudication of events.136 There also is prolonged time from the index event to percutaneous closure in some studies and the use of medical therapy in patients undergoing PFO closure is not accounted for in some studies. Devices may also carry a placebo effect,137 and the number of events is small, particularly for percutaneous closure, resulting in estimates with broad CIs. As such, results of ongoing randomized studies are needed to provide convincing evidence with regard to treatment options.138–141There are 3 ongoing randomized studies in the United States comparing the efficacy of percutaneous closure with medical therapy. The Randomized Evaluation of Recurrent Stroke Comparing PFO Closure to Established Current Standard of Care Treatment (RESPECT) trial (sponsored by AGA Medical) randomizes cryptogenic stroke patients with PFO to percutaneous closure with an Amplatzer device or medical therapy (antiplatelet or warfarin at the enrolling physician's discretion). CLOSURE I trial (sponsored by NMT Medical) randomizes patients with stroke or TIA thought to be due to paradoxical embolization to percutaneous closure with STARFlex septal occluder (subsequent-generation Cardio SEAL but not FDA approved under the HDE program) or medical therapy (aspirin and/or warfarin at the enrolling physician's discretion). The Cardia PFO trial (sponsored by Cardia Inc) randomizes patients to PFO closure using its PFO closure device (not FDA approved under the HDE program) or warfarin. The Amplatzer device is also used in Europe and Australia in the Patent Foramen Ovale and Cryptogenic Embolism (PC) trial (sponsored by AGA Medical). This study randomizes cryptogenic stroke patients with PFO to device closure or medical therapy (antiplatelet or warfarin at the enrolling physician's discretion). Of note, in all trials, patients randomized to the device arm also receive medical therapy for a variable period of time, in some cases for the study duration.In a subset of patients with PFO and hypercoagulable state, anticoagulation is required. In this group, the additional effect of PFO closure (in addition to anticoagulation) is undefined, and the ongoing trials do not address this issue because hypercoagulable state is a contraindication to enrollment.ConclusionsBecause PFO is commonly found in normal populations, we need to identify a subset of cryptogenic stroke patients who are likely to have experienced paradoxical embolization. Various factors need be considered such as atrial anatomic variation (PFO size, ASA, eustachian valve anatomy), hemodynamic parameters, presence of venous thrombus identified through higher-sensitivity tests such as lower extremity/abdominal/pelvic MRI, and the presence of hypercoagulable genetic variables. The presence of any of these findings increases the chance of PFO contributing to stroke. As such, tests to define these parameters are necessary. Currently, the superiority of one method over another for recurrent event prevention remains undefined. Randomized studies comparing medical and percutaneous closure approaches are underway, but large patient enrollment is necessary because of the low event rate in the younger patients. At this juncture, for those meeting enrollment criteria, participation in ongoing studies is recommended. For those requiring therapeutic decision, once the likelihood of individual PFO association with ischemic event is determine
Referência(s)