Anotomic Interaction Between the Aortic Root and the Atrial Septum: A Prospective Echocardiographic Study Géraldine Bertaux, MD, Jean-Christophe Eicher, MD, Annie Petit, MD, Petr Dobšák, MD, and Jean-Eric Wolf, MD, Dijon, France and Brno, Czech Republic
Background: We recently demonstrated that patients with platypnea-orthodeoxia syndrome and an enlarged aortic root had a smaller and hypermobile atrial septum (AS) compared with those with a normal aortic root. However, this was a partly retrospective study. Methods: In all, 72 patients underwent transesophageal echocardiography and cardiac catheterization. The aortic root diameter, AS dimension, AS oscillation amplitude (ASo), and atrial pressure gradient were measured. Results: Significant correlations were found: aortic root diameter and AS dimension (r ⴝ ⴚ0.5, P <
A few case reports have raised the possibility of a link between aneurysm of the aortic root and shunt-related hypoxemia, but no documented pathophysiologic explanations have been proposed.1-10 In a recently published, partly retrospective study,11 we demonstrated that patients with platypnea-orthodeoxia syndrome and an enlarged aortic root showed a particular appearance of their atrial septum (AS), with both an apparently smaller size and greater mobility compared with patients with a normally sized aortic root. We, therefore, designed a prospective study with the aim of further analyzing the relationship between aortic root dimensions and the anatomy of the AS.
METHODS The study population consisted of 72 consecutive patients presenting with aortic valve disease, aneurysm of the ascending aorta, or both, and referred to our institution for preoperative assessment by cardiac catheterization. All From the Department of Cardiology, University Hospital of Dijon; and Clinic of Functional Diagnosis and Rehabilitation, St Ann University Hospital, Brno (P.D.). Reprint requests: Jean-Christophe Eicher, MD, Centre de Cardiologie Clinique et Interventionnelle, Hôpital du Bocage, 2 Bd Maréchal de Lattre de Tassigny, 21079 Dijon Cedex, France (E-mail: [email protected]
). 0894-7317/$32.00 Copyright 2007 by the American Society of Echocardiography. doi:10.1016/j.echo.2006.09.008
.001), aortic root diameter and ASo (r ⴝ ⴙ0.3, P ⴝ .014), AS dimension and ASo (r ⴝ ⴚ0.28, P ⴝ .02), and ASo and atrial pressure gradient (r ⴝ ⴚ0.36, P ⴝ .003). Nineteen patients presented with patent foramen ovale; those with grade 3 shunting had significantly higher mobility of the AS and larger aortic roots. Conclusion: These results confirm that an increasing aortic size affects the AS by decreasing its apparent size and increasing its mobility. In case of a patent foramen ovale, increased AS mobility is associated with greater shunting. (J Am Soc Echocardiogr 2007;20:409-414.)
the patients gave informed written consent for invasive diagnostic procedures. The protocol was approved by the institutional ethical committee. Echocardiographic Study Echocardiographic examinations were made using an ultrasound system (Sonos 5500, Philips Medical System, Andover, Mass). All of the patients underwent transesophageal echocardiography (TEE) by means of a multiplane probe. They were examined in the left lateral decubitus position, after light intravenous sedation (midazolam 0.20.5 mg/kg). Aortic study. A long-axis view of the ascending aorta was sought by gradually rotating the imaging plane until the longest sagittal image of the ascending aorta was obtained. Slight clockwise and counterclockwise movements were applied to the probe to make sure that the imaging plane ran precisely through the central axis of the vessel. Measurements of the transversal aortic root diameter (ARd) were made at 3 different levels: (1) sinuses of Valsalva; (2) sinotubular junction; and (3) maximal dimension of the ascending aorta. The number indicating the rotation angle (A) of the imaging plane was used as an estimation of the angle of the ascending aorta relatively to the horizontal plane (Figure 1). The ascending aorta was considered dilated when one ARd measured greater than 40 mm in a man and greater than 36 mm in a woman.12 Patients were divided into two groups according to their aortic size (group 1 ⫽ normal; group 2 ⫽ dilated). AS study. A view showing both the aortic root and the thin part of the AS (ie, the septum primum) was usually
410 Bertaux et al
Journal of the American Society of Echocardiography April 2007
Figure 1 Transesophageal echocardiographic measurements of transversal diameters of ascending aorta; angle of aorta relative to horizontal plane is given by angle of imaging plane (A). LA, Left atrium.
Figure 2 Transesophageal echocardiographic measurement of atrial septum apparent size (left) and oscillation amplitude (right). Ao, Aorta; LA, left atrium; RA, right atrium.
obtained at the level of the sinuses of Valsalva, with a 10to 45-degree rotation of the imaging plane. The apparent AS dimension (ASd) behind the aortic root was measured in this incidence. The AS oscillation amplitude (ASo) was measured as the distance between the maximal leftward and rightward atrial-related positions of the septum primum during spontaneous breathing (Figure 2). Patients with AS aneurysm, defined by a bulge greater than 15 mm
beyond the medium plane of the AS whatever the side,13 were excluded from further analysis. Color-flow Doppler was used to look for a shunting flow through the fossa ovalis. Contrast studies were performed using Echovist (Schering, Berlin, Germany) injected through a brachial vein during spontaneous breathing, cough, and Valsalva maneuver plus abdominal compression, to seek right-toleft shunting through a patent foramen ovale (PFO). When
Journal of the American Society of Echocardiography Volume 20 Number 4
Bertaux et al 411
present, the right-to-left shunt was categorized as grade 1 (mild, 1-5 bubbles seen in the left atrium [LA]), 2 (moderate, 6-20 bubbles), or 3 (severe, ⬎20 bubbles). All echocardiographic examinations were recorded on superVHS videotapes and were analyzed offline by a cardiologist blinded to the catheterization results. Cardiac Catheterization During right heart catheterization, a careful search for a PFO was performed by means of a Lehmann’s catheter (Medtronic Inc, Minneapolis, Minn). Traditional pressure measurements were made. The atrial pressure gradient (APg) was calculated as the difference between the mean pulmonary capillary wedge pressure and the mean right atrial (RA) pressure.
Figure 3 Relationship between aortic root diameter (ARd) and atrial septal dimension (ASd).
Statistical Analysis Qualitative variables were compared using 2 test completed, when necessary, by Fisher’s exact test. Comparison of mean values was performed using Mann-Whitney and Kruskal-Wallis tests. A Spearman correlation test was used to study the following relationships: ARd and ASd, ARd and ASo, ASd and ASo, A and ASd, A and ASo, and APg and ASo. In the group of patients with a PFO, we studied the relationship between ASo and the grade of right-to-left shunt seen at contrast echocardiography. A P value less than .05 was considered significant.
Figure 4 Relationship between aortic root diameter (ARd) and atrial septal oscillation amplitude (ASo).
Four patients with AS aneurysm were excluded from this analysis. The mean ASo was 4.9 ⫾ 4.1 mm in the whole population, 3.6 ⫾ 3.1 mm in group 1, and 5.7 ⫾ 5.1 in group 2. The difference between the two groups was near significant (P ⫽ .055). Once again, we found a significant positive correlation between ASo and ARd (Figure 4). There was also a negative correlation between ASd and ASo, and a negative correlation between ASo and APg (Figures 5 and 6). No significant correlation was found between ASo and A.
Patients were aged 66 ⫾ 9 years; there were 49 men and 23 women. The reason for catheterization was predominant aortic regurgitation in 43% of patients, predominant aortic stenosis in 40%, and isolated aortic aneurysm in 17%. The mean ARd (considering the largest measured dimension) was 43.1 ⫾ 8.9 mm (range 30-63 mm). In all, 60% of the whole population, and 52% of the valve disease group, had an enlarged aortic root. There was no significant difference between group 1 (normal ARd) and group 2 (dilated aorta) regarding age and sex ratio. Apparent ASd TEE measurements showed a mean ASd of 23 mm, ranging from 9 to 42 mm. The mean value was 26.2 ⫾ 6.1 mm in group 1 and 15.5 ⫾ 9.9 mm in group 2 (P ⬍ .0001). There was a significant negative correlation between ASd and ARd, whether ARd was measured at the level of the sinuses of Valsalva or at the level of the ascending aorta, and whether it was expressed as an absolute value or indexed to the body surface area (Figure 3). On the other hand, no significant relationship could be demonstrated between ASd and the orientation of the ascending aorta (A).
PFO Nineteen patients (26%) were found to have a PFO at catheterization. All of them were identified at TEE, but 3 were detected only with color flow Doppler that showed a small left-to-right shunt through the PFO: in these 3 patients, LA pressure was increased and no right-to-left shunt was seen at the contrast study. The characteristics of these patients compared with those without PFO are shown in Table. With regard to the prevalence of PFO, there was no difference between group 1 (24%) and group 2 (28%). The contrast study showed grade 1 shunting in 4 patients, grade 2 in 5, and grade 3 in 7. Among the 7 patients with grade 3 shunting, 6 had an enlarged aortic root (Figure 7).
412 Bertaux et al
Figure 5 Relationship between atrial septal dimension (ASd) and atrial septal oscillation amplitude (ASo).
Figure 6 Relationship between atrial pressure gradient (APg) and atrial septal oscillation amplitude (ASo).
Furthermore, these patients with the more severe shunting also had significantly increased AS mobility (ASo 19 ⫾ 5.1 mm) compared with those with grade 1 (6.2 ⫾ 5.9 mm) or grade 2 (7.2 ⫾ 5.5 mm) shunting (P ⫽ .04) (Figure 8). Finally, the APg and left ventricular end-diastolic pressure had an influence on the severity of the shunt: patients with grade 3 shunting had significantly lower APg values than other patients (2.7 ⫾ 2.8 mm Hg vs 6 ⫾ 3.3 and 7.8 ⫾ 5.1, respectively). In 15 patients who had an indication for operation, the PFO was systematically closed during the procedure.
DISCUSSION During fetal life, RA-to-LA shunting across the foramen ovale is physiologic. A PFO is present in 25% to 30% of the adult population, but the vast majority of people remain asymptomatic, because the higher LA pressure keeps the flap valve pressed against the septum secundum, thus, keeping the potential channel closed and preventing interatrial shunting. There are virtually only two situations in which a right-toleft shunt may occur: (1) when the RA pressure is increased (pressure-related shunting); and (2) when the venous flow is preferentially directed toward the foramen ovale (flow-related shunting). In both con-
Journal of the American Society of Echocardiography April 2007
ditions, hypoxemia and dyspnea are the main consequences, but in the latter situation, symptoms have been reported to vary depending on the position (platypnea-orthodeoxia syndrome). Flow abnormalities have been described in association with two categories of mechanisms. First, congenital reasons, or reminiscence of fetal circulation: persistence of embryonic remnants, such as Chiari’s network, or a prominent eustachian valve, tend to deflect blood inflow from the inferior vena cava toward the fossa ovalis.1,14 Second, acquired reasons, including right pneumonectomy, atelectasia, diaphragmatic paralysis, and kyphoscoliosis 15-17: in these situations, anatomic remodeling of the mediastinum alters the relative positions of the AS and the caval veins, thus, facilitating right-to-left shunting in the presence of a PFO. As a third potential factor, the responsibility of the aortic root has been suggested for some 15 years in about 10 case reports of shunting associated with aneurysm of the ascending aorta.1-10 Some of these reports simply mentioned the association without considering a possible cause-effect link. Others speculated on several hypotheses, mainly involving compression of the RA, or counterclockwise rotation of the heart inducing distortion of the position of the AS relative to caval inflow. These few studies reported single or dual cases and did not provide the opportunity to analyze accurately the relationship between an enlarged aortic root and the AS. Our group was the first to focus on this specific question in 19 patients with platypnea-orthodeoxia syndrome.11 The first important finding was that 63% of patients with this syndrome also had a dilated aortic root; furthermore, in 42% of them, none of the thoracopulmonary abnormalities usually involved in this syndrome were seen, so that the enlarged aortic root was the only likely causative factor. The second finding was that the patients with a dilated aortic root showed a particular morphology of their AS, which was both smaller and more mobile than in patients with a normal aortic root. This phenomenon was considered to be the consequence of compression by the aortic root and reduced tenseness of the fossa ovalis membrane, allowing a billowing movement. However, in so far as this was a selected population with platypnea-orthodeoxia syndrome and PFO, a specific texture of the AS associated with PFO, platypnea-orthodeoxia syndrome, or an enlarged aortic root could not be excluded. Therefore, a prospective study in patients not selected on the presence or absence of PFO was needed. The first fact to be noted from the this study is that PFO is not more prevalent in patients with an enlarged aortic root than in the general population, so that a common link (eg, genetic or histologic) between aortic dilatation and PFO can probably be
Journal of the American Society of Echocardiography Volume 20 Number 4
Bertaux et al 413
Table Comparison of patients with and without patent foramen ovale PFO– (n ⫽ 53) PFO⫹ (n ⫽ 19)
43.1 ⫾ 8.7 43.5 ⫾ 9.6
22.8 ⫾ 7.4 23 ⫾ 8.4
4 ⫾ 0.9 8.3 ⫾ 5.8*
9.1 ⫾ 8.2 6.1 ⫾ 3.8
122 ⫾ 13 124 ⫾ 14
A, Angle between horizontal plane and ascending aorta; APg, atrial pressure gradient; ARd, aortic root diameter; ASd, atrial septum dimension; ASo, atrial septum oscillation amplitude; PFO, patent foramen ovale. *P ⬍ .001.
Figure 7 Relationship between right-to-left shunting grade and aortic root diameter. ASo, Atrial septal oscillation amplitude.
excluded. In addition, there was no difference between patients with or without a PFO with regard to the size of the AS. However, AS mobility was significantly higher in patients with a PFO (Table), which would suggest a floppier texture of the fossa ovalis in this population. The second main finding in this study is that there is indeed a significant relationship between aortic root size, apparent AS size, and AS mobility, regardless of the presence of a PFO. This means that the more the aortic root dilates, the more it encroaches on the AS, and the more the AS billows. As hypothesized in our previous study, the enlarged aortic root reduces the distance between the aorta and the atrial posterior wall, thus, loosening the tautness of the AS: this allows the flap valve of the fossa ovalis to move more freely, somewhat in the same manner as an AS aneurysm with the difference that the septum in those cases is not redundant. Importantly, the ARd/ASo relationship is significant but loose, and it is noteworthy that we also found a negative correlation between ASo and the APg, meaning that an increased LA pressure negatively affects the mobility of the AS. Thirdly, in the presence of a PFO, mobility of the AS seems to greatly influence the degree of shunting, as demonstrated by the finding in this study that patients with the more severe shunting also had the greatest ASo. It is also noteworthy that the APg and
Figure 8 Relationship between right-to-left shunting grade and atrial septal oscillation amplitude.
the left ventricular end-diastolic pressure had an influence on shunting severity; patients with grade 3 shunting had significantly lower values of APg than other patients: this finding is the direct consequence of the aforementioned negative relationship between LA pressure and AS mobility. PFO is a valvular system usually maintained in the closed position by the higher LA pressure. It has been shown that in patients with left-sided heart disease the detection rate of PFO was lower and could be missed even by contrast TEE.18 We observed the same phenomenon in our population of predominantly valvular cases: in 3 of them, PFO could not be detected by any maneuver during contrast study, and could be identified only because of a small left-to-right shunt at color Doppler. Opening of the foramen ovale usually requires an inversion of the APg, either intermittently (cough, Valsalva maneuver), or permanently (right-sided diseases). In case of a normal or moderately increased RA pressure, a normal, taut fossa ovalis membrane cannot bulge far into the RA or LA cavity and remains in contact with the septum secundum.19 Hypermobility of the septum primum is associated with more frequent PFO,20 recurrent stroke,21 decompression illness,22 and larger shunts across PFO.23 The usually accepted explanation is that excessive septal mobility is associated with a larger PFO. However, it is important to note that type 1A AS aneurysms, which bulge continuously into the RA without phasic oscillation, are not associated with shunting,24 suggesting that the key factor for shunting is the leftward movement of the
414 Bertaux et al
fossa ovalis membrane, which unsticks it from the limbus of the fossa (ie, the septum secundum). When septal mobility is increased, the likelihood of leftward bulging (and consequently of a channel in the open position) is increased; furthermore, the flap valve may undergo a spinnaker effect with the venous flow, and a bulging to the left that maintains the foramen ovale wide open.11 Conclusion This study elicits 3 main remarks. First, the aortic root and the AS seem to be intimately related, as the size of the first influences the apparent size and the mobility of the second. Second, in case of a PFO, increased AS mobility is associated with an increased risk of right-to-left shunting, whereas, thirdly, increased left-sided pressures appear to protect against shunting. One important clinical implication is that a PFO should be systematically sought in cases of an enlarged aortic root, especially in a patient who is supposed to undergo a surgical procedure, because shunting can in some cases be the cause of dyspnea, and because unawareness of a PFO might result in postoperative hypoxemia once the LA pressure has decreased to normal. REFERENCES 1. Thomas JD, Tabakin BS, Ittleman FP. Atrial septal defect with right to left shunt despite normal pulmonary artery pressure. J Am Coll Cardiol 1987;9:221-4. 2. Siderys H, Bittles ML, Niemeier M, et al. Severe hypoxia related to uncomplicated atrial septal defect. Heart Inst J 1993;20:123-5. 3. Zuazola Martinez P, Ruano Calvo J, Martin Duran R, et al. A right-to-left shunt through a patent foramen ovale without pulmonary hypertension. Rev Esp Cardiol 1993;46:263-5. 4. Dear WE, Chen P, Barasch E, et al. Sixty-eight-year-old woman with intermittent hypoxemia. Circulation 1995;91: 2284-9. 5. Landzberg M, Sloss LJ, Fanerty CE, et al. Orthodeoxiaplatypnea due to intracardiac shunting: relief with transcatheter double umbrella closure. Catheter Cardiovasc Diagn 1995; 36:247-50. 6. Savage EB, Benckart DH, Donahue BC, et al. Intermittent hypoxia due to right atrial compression by an ascending aortic aneurysm. Ann Thorac Surg 1996;62:582-3. 7. Popp G, Melek H, Garnett AR. Platypnea-orthodeoxia related to aortic elongation. Chest 1997;112:1682-4.
Journal of the American Society of Echocardiography April 2007
8. Laybourn KA, Martin ET, Cooper RA, et al. Platypnea and orthodeoxia: shunting associated with an aortic aneurysm. J Thorac Cardiovasc Surg 1997;113:955-6. 9. Kubler P, Gibbs H, Garrahy P. Platypnea-orthodeoxia syndrome. Heart 2000;83:221-3. 10. Faller M, Kessler R, Chaouat A, et al. Platypnea-orthodeoxia syndrome related to an aortic aneurysm combined with an aneurysm of the atrial septum. Chest 2000;118:553-7. 11. Eicher JC, Bonniaud P, Baudouin N, et al. Hypoxaemia associated with an enlarged aortic root: a new syndrome? Heart 2005;91:1030-5. 12. Roman HJ, Devereux RB, Kramer-Fox T, et al. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol 1989;64:507-12. 13. Hanley PC, Jamil Tajik A, Hynes JK, et al. Diagnosis and classification of atrial septal aneurysm by two-dimensional echocardiography: report of 80 consecutive cases. J Am Coll Cardiol 1985;6:1370-82. 14. Gallaher ME, Sperling DR, Gwinn JL, et al. Functional drainage of the inferior vena cava into the left atrium–three cases. Am J Cardiol 1963;12:561-6. 15. Schnabel TG, Ratto O, Kirby CK, et al. Postural cyanosis and angina pectoris following pneumonectomy: relief by closure of an interatrial septal defect. J Thorac Surg 1956;32:246-50. 16. Mercho N, Stoller JK, White RD, et al. Right-to-left interatrial shunt causing platypnea after pneumonectomy. Chest 1994; 105:931-3. 17. Al Khouzaie T, Busser JR. A rare cause of dyspnea and arterial hypoxemia. Chest 1997;112:1681-2. 18. Siostrzonek P, Lang W, Zangeneh M, et al. Significance of left-sided heart disease for the detection of patent foramen ovale by transesophageal contrast echocardiography. J Am Coll Cardiol 1992;19:1192-6. 19. Silver MD, Dorsey JS. Aneurysms of the septum primum in adults. Arch Pathol Lab Med 1978;102:62-5. 20. Louie EK, Konstadt SN, Rao TL, Scanion PJ. Transesophageal echocardiographic diagnosis of right to left shunting across the foramen ovale in adults without prior stroke. J Am Coll Cardiol 1993;21:1231-7. 21. De Castro S, Cartoni D, Fiorelli M, et al. Morphological and functional characteristics of patent foramen ovale and their embolic implications. Stroke 2000;31:2407-13. 22. Cartoni D, De Castro S, Valente G, et al. Identification of professional scuba divers with patent foramen ovale at risk for decompression illness. Am J Cardiol 2004;94:270-3. 23. Fox ER, Picard MH, Chow CM, Levine RA, Schwamm L, Kerr AJ. Interatrial septal mobility predicts larger shunts across patent foramen ovales: an analysis with transmitral Doppler scanning. Am Heart J 2003;145:730-6. 24. Pearson AC, Nagelhout D, Castello R, et al. Atrial septal aneurysm and stroke: a transesophageal echocardiographic study. J Am Coll Cardiol 1991;18:1223-9.