Assessment of atrial septal defects in adults comparing cardiovascular magnetic resonance with transoesophageal echocardiography

Atrial septal defects are the most common congenital cardiac malformation first diagnosed in adults and account for approximately 10% of all congenital heart lesions [1]. Patients with a significant shunt (Qp/Qs > 1.5/1.0) experience symptoms over time with effort dyspnoea seen in about 30% of patients by the third decade and in over 75% of patients by the 5^th decade. Complications may include the development of pulmonary hypertension, supraventricular arrhythmias (atrial fibrillation and atrial flutter) and right-sided heart failure from right ventricular volume overload.

Surgical closure of atrial septal defects (ASDs) has previously been shown to have excellent results in both medium and long term studies [2], but is associated with significant morbidity and mortality [3]. Many adults with secundum ASDs are now able to have these defects closed percutaneously using septal occluder devices such as the Amplatzer Septal Occluder (ASO), a self-expanding circular double disc with a conjoint waist containing polytetrafluoroethylene (PTFE) and a nitinol mesh. This device has now become an accepted alternative to surgical repair with studies comparing ASO device closure to surgical closure showing decreased complication rates, shorter hospital stays and greater cost-effectiveness [4].

The echocardiographic morphology of the ASD and accurate assessment of the stretched diameter has been important for patient selection. Initially, it was felt that atrial septal defects up to 26 mm stretched balloon diameter could be closed with the ASO [5]. More recently, there is registry data of larger ASDs closed successfully using the 40 mm ASO [6]. Inclusion criteria for percutaneous ASD closure have included: 1) the presence of a secundum ASD <40 mm by echocardiography, 2) a left-to-right shunt with a Qp/Qs ratio of >1.5:1 or the presence of right ventricular volume overload 3) patients with minimal shunt in the presence of symptoms and 4) the presence of a distance of >5 mm from the margins of the ASD to the coronary sinus, atrioventricular valves and right upper pulmonary vein as measured by echocardiography [7‐9]. Traditionally, these assessments of ASD size, geometry and atrial septal margins have been obtained by transoesophageal echocardiography (TOE) prior to the percutaneous closure [7, 10]. This is a semi-invasive technique and all of the information could potentially be obtained by non-invasive cardiovascular magnetic resonance (CMR). Furthermore, TOE is imperfect at assessing all atrial septal margins. A large atrial septal defect that is located inferoposteriorly is difficult to both assess and close [7] and relates to the limited assessment of the posterior inferior margin by TOE [11]. Although 3-D TOE may improve this, this new technology is not yet widely available, and yet to be shown to be superior to 2-D TOE for this purpose.

The aim of this study was to compare the assessment of atrial septal defects in consecutive adult patients being considered for percutaneous ASD closure using CMR and TOE.

A total of 20 patients (M: F = 5: 15, mean age 42.8 years ± 15.7) were included in the analyses. All but 1 patient had both TOE and CMR to evaluate their suitability for percutaneous ASD closure. TOE was not performed in 1 patient due to history of an oesophageal stricture. CMR was well tolerated in all patients. Total CMR examination time was 20 to 30 minutes in all cases. Three patients were found not to be suitable for percutaneous ASD closure due to excessive stretched balloon diameter (SBD) of >40 mm at cardiac catheterisation (CMR diameters pre-balloon sizing were 32, 38 and 33 mm). All 3 patients went on to surgical closure.

Many adults with secundum atrial septal defects are now able to have these defects closed percutaneously. The size, location and margins of an atrial septal defect are major determining factors for transcatheter closure. Conventional assessment pre-closure have been with transoesophageal echocardiography (TOE) for measurement of ASD dimensions and margins[7, 10]. Inclusion criteria for closure have included ASD maximum diameter of 40 mm and rims of at least 5 mm towards the IVC, SVC, right upper pulmonary vein and mitral valve [8, 9]. However, more recently, deficiency of the margins has still allowed closure. For example, in patients with a deficient anterior superior (AS) margin, a correctly sized Amplatzer septal occluder can successfully be used with the device moulding itself around the aortic wall with minimal risk of perforation [8, 9]. There have also been small numbers of patients with small inferior and posterior defect margins (<5 mm) who have been successfully closed [9].

While TOE has been the traditional method for the evaluation and screening for patients who are candidates for transcatheter closure, it is semi-invasive and cannot easily be performed in young children and some adults. Measurements of the ASD can only be obtained from the 4-chamber, short axis and bicaval views on TOE due to the plane on TOE that generally projects the ASD in a relatively fixed direction [11]. In addition, TOE may underestimate the size of the stretched balloon diameter of the ASD, although this has been described to factors associated with choice of plane for maximal diameter in defects that are oval rather than circular in shape, displacement of the interatrial septum from enlarged right atrium and changes of the shape of the defect during the cardiac cycle [7, 14]. Three-dimensional echocardiography obtained from TOE images may provide superior anatomic detail [10, 15], but the accuracy of the reconstructed images is dependent on technical experience of the operator [8].

CMR has been shown to visualise secundum ASDs [16]. Studies comparing CMR with TOE to assess ASD and suitability for percutaneous closure have mainly been done in paediatric populations [11], although there are a limited number of studies with adult patients [17‐19]

In this study, we were able to assess maximal atrial defect size in all patients with both TOE and CMR. In the assessment of atrial septal margins, CMR could assess the posterior inferior margins in all patients compared to approximately 60% that could be assessed with TOE. One study in a paediatric population comparing TOE and MRI assessment showed that patients who had successful closure had a significantly smaller major axis of ASD and larger posterior inferior rim compared to those who were excluded from closure procedure. An adequate posterior inferior rim was also best visualised in that study with CMR and showed a better correlation of ASD diameter measurement to balloon sizing compared to TOE [11]. However, a limitation in our study of the measurement of margins using TOE relates to this study being a retrospective study and with the images that were strictly required for measurement of the margins (such as the PI margin) not being available in some patients. Hence the 60% assessibility for the PI margin is not a true reflection of the ability of TOE to define this margin.

In the patients (n = 3) that went on to surgical closure of their defects, although the pre-procedure ASD diameters on CMR were less than 40 mm (between 32-38 mm in fact), the stretched balloon diameters were >40 mm, which is beyond the largest ASD amplatzer closure devices available, at the time of invasive assessment.

Velocity encoded contrast cine imaging was not performed in this study, however, this is another CMR technique that can provide information about the size and shape of the atrial septal defect [16, 18] as well as provide accurate assessment of shunt magnitude from measuring flow in the systemic and pulmonary circulations [20, 21]. However, this technique has not been used for the assessment of margins due to its limited spatial resolution.

CMR agrees closely with TOE assessment of atrial septal defects for percutaneous closure. In addition, it is able to assess the septal margins such as the posterior inferior margin, which is known to be difficult to assess with TOE. Potentially, CMR could be used instead of TOE for the assessment for percutaneous closure.