Diagnosis of Patent Foramen Ovale Using Contrast-Enhanced Dynamic MRI: A Pilot Study

Transcription

1 Mohrs et al. MRI of Patent Foramen Ovale Oliver K. Mohrs 1,2 Steffen E. Petersen 3 Damir Erkapic 1 Christine Rubel 1 Rainer Schräder 1 Bernd Nowak 1 W. Andreas Fach 1 Hans-Ulrich Kauczor 2 Thomas Voigtlaender 1 Mohrs OK, Petersen SE, Erkapic D, et al. Received April 15, 2004; accepted after revision June 14, Department of Cardiovascular MRI, Cardiovascular Center Bethanien, Im Pruefling 17, D Frankfurt/Main, Germany. Address correspondence to O. K. Mohrs (o.mohrs@ccb.de). 2 German Cancer Research Center (DKFZ), Heidelberg, Germany. 3 University of Oxford Centre for Cardiovascular Magnetic Resonance, John Radcliffe Hospital, Oxford, England. AJR 2005;184: X/05/ American Roentgen Ray Society Diagnosis of Patent Foramen Ovale Using Contrast-Enhanced Dynamic MRI: A Pilot Study OBJECTIVE. The aim of this study was to evaluate the feasibility of dynamic contrast-enhanced MRI for detection of patent foramen ovale. SUBJECTS AND METHODS. Fifteen patients with and five patients without patent foramen ovale underwent transesophageal echocardiography and MRI, which were performed during the Valsalva maneuver. Grading results (grade 0, no patent foramen ovale and grades 1 3, minor to major enhancement due to intracardiac shunt) were assessed visually. Signal-intensity curves in the left atrium and in a pulmonary vein served to underline the diagnosis. RESULTS. The diagnoses of all patients with (15/15) and without patent foramen ovale (5/ 5) were correct compared with the findings of the reference transesophageal echocardiography. In 12 (60%) of 20 patients, the grading scores were identical, and in four (20%) of 20 patients, the scores differed by more than one grade. Overall, there was a good correlation of grading scores (r = 0.7, p < 0.05). Using signal-intensity curves, we found that the patients with patent foramen ovale showed an additional signal peak in the left atrium before the enhancement of the pulmonary vein because of an intracardiac shunt. In three of 15 patients with patent foramen ovale, an atrial septal aneurysm was correctly diagnosed. CONCLUSION. This pilot study shows that MRI is a new noninvasive method to detect patent foramen ovale and atrial septal aneurysm. A grading is possible but warrants further investigation regarding its predictive value and impact on treatment strategies. atent foramen ovale (PFO) is a P known cause of cerebral strokes or transient ischemic attacks due to paradoxical embolism. Diagnosis and treatment are required to prevent further cerebral events. In patients without cardiac symptoms, the incidence of PFO is 34% during the first three decades and declines with age, as seen in autopsy studies [1]. The annual recurrence rate of cerebral stroke or transient ischemic attacks for patients with PFO is about 3.5% [2, 3]. PFOs with large shunts have a higher risk than those with small shunts [4]. Recent studies show a higher incidence for cerebral strokes if the PFO is associated with an atrial septal aneurysm [5 8]. Therefore, an ideal method should show both diagnoses, PFO and atrial septal aneurysm. Contrast-enhanced transesophageal echocardiography (TEE) is considered the clinical reference to detect PFO and atrial septal aneurysm [9]. This method, however, is semiinvasive. Recently, MRI has been established in many cardiovascular applications, particularly because of the latest developments of hardware and ultrafast sequences [10]. The purposes of this study were threefold: to investigate the feasibility of MRI to noninvasively and visually detect PFO and atrial septal aneurysm using contrast-enhanced MRI, to evaluate the use of signal time curves to diagnose the shunt across the PFO using the Valsalva maneuver, and to correlate grading of PFO using contrast-enhanced MRI in comparison with TEE. Subjects and Methods Study Population Fifteen consecutive patients with a PFO detected on contrast-enhanced TEE were enrolled for contrast-enhanced MRI (five women; mean age, 49 ± 14 years [SD]) after written informed consent was obtained (Table 1). The study was approved by the local review board. Contrast-enhanced MRI was performed within 1 week of TEE. Twelve of these 15 patients had a history of cerebral events, including transient ischemic attacks, amaurosis fugax, or strokes. The indications for TEE in the remaining patients with PFO were recurrent peripheral emboli 234 AJR:184, January 2005

2 TABLE 1 Features Signal Time Curves Computed from Contrast-Enhanced MRI in Patients With and Without Patent Foramen Ovale (PFO) Time to Peak (sec) (one patient) and suspected PFO at routine transthoracic echocardiography. The control group consisted of five patients (one woman; mean age, 59 ± 13 years) without PFO confirmed by TEE. In these patients, TEE was performed to exclude atrial thrombi. First Peak LA First Peak PV Second Peak LA % of Baseline Signal Time to Peak (sec) % of Baseline Signal Time to Peak (sec) % of Baseline Signal PFO patient no Median 7 a 160 a Q1/Q3 6/11 142/177 13/22 372/486 15/22 458/580 Min/max 5/16 112/378 12/32 188/671 12/33 253/775 Control no n.p. n.p n.p. n.p n.p. n.p n.p. n.p n.p. n.p. Median 19 a 514 a Q1/Q3 15/19 457/ /22 456/503.5 Min/max 12/22 407/603 12/23 442/521 Note. LA = left atrium, PV = pulmonary vein, n.p. = not present, Q1 = quartile 1, Q3 = quartile 3, Min = minimum, max = maximum. a Statistically significant difference (p < 0.05) between PFO and control patients using Mann-Whitney-U test. TEE TEE was performed with a 5-MHz phased multiplane probe (Vingmed System Five, GE Healthcare). We used 0.02 mg of orally administered lidocaine (Xylocain Pumpspray, AstraZeneca) for local pharyngeal anesthesia. The contrast agent (Dgalactose, Echovist, Schering) was administered as a bolus of 10 ml into an antecubital vein during a Valsalva maneuver. Right-to-left shunting was graded in consensus by an experienced cardiologist and radiologist as follows [11]: grade 0, no contrast agent passed from the right to the left atrium; grade 1, 3 9 microbubbles passed from the right to the left atrium; grade 2, microbubbles passed from the right to the left atrium; and grade 3, greater than 30 microbubbles passed from the right to the left atrium (opacified left atrium due to bright contrast). The presence of atrial septal aneurysm was defined as a protrusion of the atrial septum into the left or right atrium of greater than 10 mm. MRI Contrast-enhanced MRI was performed on a 1.5-T MRI system (Magnetom Sonata Maestro Class, Siemens Medical Solutions). For signal detection, the combination of a six-channel body phased-array coil and a two-channel spine phased-array coil was used. ECG signal was received from an external system (Magnitude 3150, InVivo Research). An ECG-gated segmented true fast imaging with steady-state free precession (FISP) cine sequence (TR/TE, 2.7/1.2; temporal resolution, 34 msec; voxel size, mm 3 ) served for detection of atrial septal aneurysms in a stack of contiguous horizontal long-axis views and shortaxis planes. The presence of atrial septal aneurysm was assessed in consensus by an experienced cardiologist and radiologist who were blinded to the TEE diagnosis. Contrast-enhanced perfusion analysis was performed during the Valsalva maneuver using 10 ml of gadopentetate dimeglumine (Magnevist, Schering) followed by a 20-mL saline solution, administered into an antecubital vein. The infusion rate was 6 ml per second. Two slices of a saturation-recovery true FISP sequence (TE, 2.7 msec; inversion time, 217 msec; flip angle, 50 ; temporal resolution, 832 msec; matrix, ; voxel size, mm 3 ) were positioned in the horizontal long and short axis, which showed the optimal view of the fossa ovalis chosen from the cine studies. For each slice, 40 consecutive images were acquired, one per heart cycle. Contrast-enhanced perfusion studies were analyzed by reviewers who were blinded to the TEE diagnosis and grade of PFO and consisted of the following aspects: visual assessment grading of PFO and interpretation of signal time curves to diagnose PFO. Signal time curves were generated using the program Mean Curve (Siemens Medical Solutions). Two regions of interest (ROIs) were placed in a pulmonary vein and the left atrium. ROIs were manually fitted to every image without changing the size. The atrial ROIs were drawn close to the atrial septum, but inclusion of pixels from the right atrium was carefully avoided. The signal time curves were normalized to baseline signal (second image) for each ROI. The single data points represented the percentage of the baseline signal. PFO grading for contrast-enhanced MRI was performed on the basis of the grading system as follows: grade 0, no contrast enhancement in the left atrium before the contrast agent reached the pulmonary veins; grade 1, only slight contrast enhancement close to the atrial septum without enhancement of AJR:184, January

3 Mohrs et al. TABLE 2 Contrast-Enhanced MRI Crosstabulation of Transesophageal Echocardiography (TEE) and Contrast-Enhanced MRI Grading Scores the entire left atrium before the contrast agent reached the pulmonary veins; grade 2, only slight contrast enhancement in the left atrium before the contrast agent reached the pulmonary veins; and grade 3, bright contrast enhancement in the left atrium before the contrast agent reached the pulmonary veins. Statistical Analysis Data presentation and statistical analyses performed in this study were chosen to avoid the assumption of a normal distribution. Data are given as medians (quartile 1, quartile 3, minimum, maximum). TEE and contrast-enhanced MRI grading scores were correlated using the Spearman s rank correlation. The Mann-Whitney U test for unpaired variables was used to test for differences between patients with and without PFO regarding time points of peaks in the signal time curves and their signal intensities. A p value of less than 0.05 was considered statistically significant. All computations were performed with SPSS software, version 11.5 (Statistical Package for the Social Sciences) for Windows (Microsoft). TEE Grade 0 Grade 1 Grade 2 Grade 3 Grade 0 Count % within TEE Grade 1 Count % within TEE Grade 2 Count % within TEE Grade 3 Count % within TEE Total Count % within TEE Total Results Visual Diagnosis of PFO and Atrial Septal Aneurysm PFO was identified visually in all 15 patients using contrast-enhanced MRI. In all five control patients, PFO could be correctly excluded by visual contrast-enhanced MRI assessment. For all 15 patients with PFO evaluated, an early contrast enhancement due to intracardiac right-to-left shunting was present in the left atrium before the contrast agent reached the pulmonary veins (Fig. 1, Movies 1 and 2). The prevalence of atrial septal aneurysm in our study population of PFO patients was 20% (3/15) detected on TEE. Three of 15 patients were correctly identified using the contrast-enhanced MRI (Fig. 2, Movie 3). Neither false-negative nor false-positive diagnoses were made with MRI. Grading of PFO All five control patients without PFO depicted on TEE were correctly categorized as having grade 0 on contrast-enhanced MRI. Overall, seven (47%) of 15 patients with PFO (grades 1 3) were given grades identical to those of the reference TEE. The classifications of four patients differed by one grade between the diagnosis based on TEE and that based on contrast-enhanced MRI, and classifications of the remaining four patients differed by two grades. Among the patients with PFO, there was no systematic underestimation or overestimation of the grades. Taking the control patients with grade 0 into this calculation, we found that 12 (60%) of 20 patients were categorized identically and 16 (80%) of 20 were graded with a maximal grade difference of one (Table 2). The correlation for the 20 participants was good (r = 0.7, p < 0.05) but was poor without the grade 0 patients (r = 0.19, not significant). Diagnosis of PFO Using Signal Time Curves In the control group, the signal time curves showed a peak of 514% of baseline signal in the pulmonary vein 19 sec after contrast agent administration. This peak in the pulmonary vein was followed by a single peak of 484% in the left atrium at 20 sec. In contrast to the control group, all patients with PFO showed signal time curves with an early first peak in the left atrium of 160% of baseline signal at 7 sec. This was followed by the peak in the pulmonary vein of 486% at 17 sec. The initial peak and drop back to baseline in the left atrium were followed by a second higher peak of 517% at 19 sec. The time point and the percentage of baseline signal of the first peak in the left atrium were significantly different between the patients with PFO and controls (p < 0.002). Differences between the time point and the percentage of baseline signal of the peak in the pulmonary vein were not significant. On the basis of the presence or absence of an early first signal peak in the left atrium followed by a second peak in the left atrium, the diagnoses of all patients with and without PFO were correct as compared with the reference TEE diagnoses (Figs. 3 and 4). Discussion This pilot study could show the feasibility of reliably diagnosing PFO and atrial septal aneurysms using contrast-enhanced MRI. Various MRI techniques have been used to visualize atrial septal defects [12]. The combination of visualization of morphologic defects and the analysis of functional consequences (e.g., shunt detection) allows the diagnosis of an atrial septal defect [13, 14]. Until now, the diagnosis of PFO has not been possible using MRI because of insufficient spatial and temporal resolution and the absence of a measurable shunt volume. TEE is the current clinical reference for the diagnosis of PFO, which correlated well in autopsy studies [15, 16]. The major drawbacks of TEE are its semiinvasiveness and the inability to acquire more than one imaging plane during the administration of the contrast agent. The results presented in this study underline the potential of contrast-enhanced MRI to noninvasively reliably detect PFO either by visual assessment or by computation of signal time curves in the pulmonary vein and the left 236 AJR:184, January 2005

4 A B E H C F Fig year-old man with patent foramen ovale grade 3 on transesophageal echocardiography and grade 3 on MRI. A N, Figures 1A and 1H show anatomy using true fast imaging with steady-state free precession cine sequence in horizontal long axis (A) and short axis (H) at atrial level. Figures 1B 1G and 1I 1N show temporal sequence of contrast-enhanced dynamic perfusion imaging during Valsalva maneuver. Figures 1B and 1I show baseline signal without enhancement. Figures 1C and 1J show enhancement of right atrium (RA). Figures 1D and 1K show enhancement of entire left atrium (LA) due to right-to-left shunting (arrows) before enhancement of pulmonary vein (PV). Figures 1E and 1L show signal decrease in LA representing dip after first initial peak back to baseline. Figures 1F and 1M show enhancement of PV and second signal peak in LA. Figures 1G and 1N show enhancement of aorta. RV = right ventricle, LV = left ventricle, RV = right ventricle, Ao = aorta, PA = pulmonary artery. (Fig. 1 continues on next page) D G AJR:184, January

5 Mohrs et al. atrium. The advantage of signal-time-curve analysis is the clear-cut differentiation of contrast enhancement due to right-to-left shunting in the left atrium from pulmonary flow. This is not possible with TEE and may cause falsepositive results in the rare, but important, case of a pulmonary arteriovenous malformation. I L Fig. 1. (continued) 32-year-old man with patent foramen ovale grade 3 on transesophageal echocardiography and grade 3 on MRI. A N, Figures 1A and 1H show anatomy using true fast imaging with steady-state free precession cine sequence in horizontal long axis (A) and short axis (H) at atrial level. Figures 1B 1G and 1I 1N show temporal sequence of contrast-enhanced dynamic perfusion imaging during Valsalva maneuver. Figures 1B and 1I show baseline signal without enhancement. Figures 1C and 1J show enhancement of right atrium (RA). Figures 1D and 1K show enhancement of entire left atrium (LA) due to right-to-left shunting (arrows) before enhancement of pulmonary vein (PV). Figures 1E and 1L show signal decrease in LA representing dip after first initial peak back to baseline. Figures 1F and 1M show enhancement of PV and second signal peak in LA. Figures 1G and 1N show enhancement of aorta. J M PFO patients with huge shunts have been shown to have a higher risk for paradoxical embolisms than those with small shunts [17, 18]. This finding forms the rationale for grading PFO because therapeutic strategies could be based on the severity of the shunt across the PFO despite the lack of widely accepted guidelines [19]. We show a good correlation between the grading scores using TEE and contrast-enhanced MRI including all patients with and without PFO. If the control group of patients without PFO (grade 0) was excluded from the correlation analysis, then the correlation was poor. We suspect that the discrepan- K N 238 AJR:184, January 2005

6 Fig year-old man with atrial septal aneurysm (arrow) on MRI. A and B, Typical bulging of atrial septum is shown toward left atrium (LA) in horizontal long axis (A) and toward right atrium (RA) in short axis (B). cies might reflect the intraindividual variability of performing the Valsalva maneuver, which is the most effective way to induce a right-to-left atrial shunting [20]. If the patient does not perform a Valsalva maneuver correctly, a PFO can be missed [21]. Therefore, deep sedation might alter the cooperation required to perform this maneuver and thus bias the degree of shunt observed. This might pose a problem in patients who cannot tolerate the transesophageal probe or the contrast-enhanced MRI investigation without IV sedation. Baseline Signal (%) Time (sec) A Baseline Signal (%) B Another cause for the limited correlation in grading scores for both techniques in PFO patients can be sought in the limited number of imaging planes despite the much-improved temporal and spatial resolution. We used two different imaging planes in the present study. For a more reliable quantification, a 3D volume would be needed. This is currently available for neither TEE nor contrast-enhanced MRI. Because of the opposite anatomic position of the ostium of the inferior vena cava in relation to the fossa ovalis, a contrast injection into the femoral vein could potentially improve the evaluation of shunt size for both techniques [22]. The presence of atrial septal aneurysms in PFO patients has a major predictive value [6]. Diagnosis of atrial septal aneurysm and evaluation of PFO in one single investigation using contrast-enhanced MRI or TEE are major advantages over transcranial Doppler sonography, which can reliably detect PFO, but not atrial septal aneurysm [23]. In our study, atrial septal aneurysm in all three patients with this condition was correctly diagnosed using MRI. Currently, contrast-enhanced MRI cannot replace TEE to exclude other potential embolic sources such as thrombus in left atrial appendage. The feasibility of contrast-enhanced MRI to exclude this embolic source was not addressed in our study and warrants further investigation. Compared with TEE, contrastenhanced MRI examinations currently are more expensive and need a longer examination time. This pilot study was designed to show the potential of this new application to evaluate and grade PFO and to detect atrial septal aneurysms. A larger prospective study is needed to confirm our findings and to show the predictive value of PFO grading for clinical outcome. Our initial contrast-enhanced MRI experience, however, encourages the design of studies to show the feasibility of contrast Time (sec) Fig. 3. Graph of signal time curve in 32-year-old man with patent foramen ovale shows early initial signal peak (arrow) in left atrium ( ) followed by second higher peak before peak in pulmonary vein ( ). Fig. 4. Graph of typical signal time curve in 26-year-old man without patent foramen ovale shows single peak in left atrium ( ) at almost same time as peak in pulmonary vein ( ). AJR:184, January

7 Mohrs et al. enhanced MRI as a noninvasive technique to screen, for example, scuba divers [24]. In conclusion, this pilot study has shown that contrast-enhanced MRI can be used to reliably detect PFO and atrial septal aneurysm on the basis of the visual assessment. Acknowledgment This study contains data from the doctoral thesis by Damir Erkapic. References 1. Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984;59: Bogousslavsky J, Garazi S, Jeanrenaud X, Aebischer N, Van Melle G. Stroke recurrence in patients with patent foramen ovale: the Lausanne Study Lausanne Stroke with Paradoxal Embolism Study Group. Neurology 1996;46: Mas JL, Zuber M. Recurrent cerebrovascular events in patients with patent foramen ovale, atrial septal aneurysm, or both and cryptogenic stroke or transient ischemic attack: French Study Group on Patent Foramen Ovale and Atrial Septal Aneurysm. Am Heart J 1995;130: Stone DA, Godard J, Corretti MC, et al. Patent foramen ovale: association between the degree of shunt by contrast transesophageal echocardiography and the risk of future ischemic neurologic events. Am Heart J 1996;131: Agmon Y, Khandheria BK, Meissner I, et al. Frequency of atrial septal aneurysms in patients with cerebral ischemic events. Circulation 1999;99: Mattioli AV, Bonetti L, Aquilina M, Oldani A, Longhini C, Mattioli G. Association between atrial septal aneurysm and patent foramen ovale in young patients with recent stroke and normal carotid arteries. Cerebrovasc Dis 2003;15: Bruch L, Parsi A, Grad MO, et al. Transcatheter closure of interatrial communications for secondary prevention of paradoxical embolism: single-center experience. Circulation 2002;105: De Castro S, Cartoni D, Fiorelli M, et al. Patent foramen ovale and its embolic implications. Am J Cardiol 2000;86:51G 52G 9. Belkin RN, Pollack BD, Ruggiero ML, Alas LL, Tatini U. Comparison of transesophageal and transthoracic echocardiography with contrast and color flow Doppler in the detection of patent foramen ovale. Am Heart J 1994;128: Pohost GM, Hung L, Doyle M. Clinical use of cardiovascular magnetic resonance. Circulation 2003;108: Lamy C, Giannesini C, Zuber M, et al. Clinical and imaging findings in cryptogenic stroke patients with and without patent foramen ovale: the PFO-ASA Study Atrial Septal Aneurysm. Stroke 2002;33: Holmvang G, Palacios IF, Vlahakes GJ, et al. Imaging and sizing of atrial septal defects by magnetic resonance. Circulation 1995;92: Hundley WG, Li HF, Lange RA, et al. Assessment of left-to-right intracardiac shunting by velocityencoded, phase-difference magnetic resonance imaging: a comparison with oximetric and indicator dilution techniques. Circulation 1995;91: Beerbaum P, Korperich H, Barth P, Esdorn H, Gieseke J, Meyer H. Noninvasive quantification of left-to-right shunt in pediatric patients: phasecontrast cine magnetic resonance imaging compared with invasive oximetry. Circulation 2001;103: Hausmann D, Mugge A, Becht I, Daniel WG. Diagnosis of patent foramen ovale by transesophageal echocardiography and association with cerebral and peripheral embolic events. Am J Cardiol 1992;70: Chen WJ, Kuan P, Lien WP, Lin FY. Detection of patent foramen ovale by contrast transesophageal echocardiography. Chest 1992;101: Anzola GP, Zavarize P, Morandi E, Rozzini L, Parrinello G. Transcranial Doppler and risk of recurrence in patients with stroke and patent foramen ovale. Eur J Neurol 2003;10: Steiner MM, Di Tullio MR, Rundek T, et al. Patent foramen ovale size and embolic brain imaging findings among patients with ischemic stroke. Stroke 1998;29: Rodriguez CJ, Homma S. Patent foramen ovale and stroke. Curr Treat Options Cardiovasc Med 2003;5: Pfleger S, Konstantin Haase K, Stark S, et al. Haemodynamic quantification of different provocation manoeuvres by simultaneous measurement of right and left atrial pressure: implications for the echocardiographic detection of persistent foramen ovale. Eur J Echocardiogr 2001;2: Schneider B, Zienkiewicz T, Jansen V, Hofmann T, Noltenius H, Meinertz T. Diagnosis of patent foramen ovale by transesophageal echocardiography and correlation with autopsy findings. Am J Cardiol 1996;77: Schuchlenz HW, Weihs W, Beitzke A, Stein JI, Gamillscheg A, Rehak P. Transesophageal echocardiography for quantifying size of patent foramen ovale in patients with cryptogenic cerebrovascular events. Stroke 2002;33: Droste DW, Lakemeier S, Wichter T, et al. Optimizing the technique of contrast transcranial Doppler ultrasound in the detection of right-to-left shunts. Stroke 2002;33: Knauth M, Ries S, Pohimann S, et al. Cohort study of multiple brain lesions in sport divers: role of a patent foramen ovale. BMJ 1997; 314: A data supplement for this article can be viewed in the online version of this article at: AJR:184, January 2005