Cardiovascular magnetic resonance (CMR) imaging can accurately detect left atrial (LA) and left atrial appendage (LAA) thrombi prior to pulmonary-vein isolation (PVI), potentially sidestepping the need for a transesophageal echocardiogram (TEE), new research suggests[1].

Delayed-enhancement CMR using long inversion time (long TI DE-CMR) was able to identify all nine cases of thrombi detected by the gold-standard TEE.

Long TI DE-CMR had the best diagnostic accuracy, sensitivity, and specificity (99.2%, 100%, 99.2%), followed by contrast-enhanced MR angiography (94.3%, 66.7%, 95.2%) and standard cine-CMR (91.6%, 66.7%, 92.5%), Dr Danai Kitkungvan (Houston Methodist DeBakey Heart & Vascular Center, TX) and colleagues reported in their study, published online May 25, 2016 in JACC: Cardiovascular Imaging.

“Certainly, the results of our findings would suggest you may be able to obviate the need for getting a transesophageal echo and really within the CMR test alone do both—identify the pulmonary-vein anatomy as well as look for the presence of thrombus,” senior author Dr Dipan J Shah (also from Houston Methodist DeBakey) told heartwire from Medscape.

The current paradigm is for patients get either an MR or computed tomography (CT) scan to define the pulmonary-vein anatomy and a TEE on top of that to exclude the presence of LA or LAA thrombus.

“A diagnostic study that confidently answers two clinical questions is appealing from a practical and financial point of view,” Drs Warren Manning and Aferdita Spahillari (Beth Israel Deaconess Medical Center, Boston, MA) write in an accompanying editorial[2].

In order for this approach to become incorporated into clinical practice, however, they note that the optimal interval between CMR and PVI would need to be determined, especially for patients with atrial fibrillation (AF), in whom anticoagulation is discontinued between procedures, but that CMR “may be especially attractive for those with a relative contraindication to TEE (eg, esophageal stricture, Zenker’s diverticulum) or at higher risk for conscious sedation (eg, sleep apnea).”

Single Shot

The study included 261 patients (mean age 61.8 years; 69.4% male) who underwent multicomponent CMR imaging and TEE within 7 days (mean 1.3 days). Almost two-thirds of patients (73.6%) were on anticoagulation at the time of the CMR exam.
All TEE and CMR studies were interpreted by two experienced readers blinded to clinical history and results of other imaging studies. Interobserver agreement was highest in long TI DE-CMR (κ=0.91), followed by cine-CMR (κ=0.85) and contrast-enhanced MR angiography (κ=0.83).

One of the study’s primary objectives was to see whether long TI DE-CMR, which is available on most cardiac MR scanners and does not require special software or hardware, can be performed in patients with AF or irregular heart rhythms, Shah said. It is performed with a single-shot acquisition technique 10 minutes after contrast administration and “essentially acquires each image within one single heartbeat. So the lack of being able to hold their breath or the presence of irregular heart rhythms really doesn’t compromise the image quality for this particular technique,” he added.

The diagnostic accuracy of CMR was slightly less but not significantly different in patients with AF vs those in sinus rhythm using long TI DE-CMR (98.2% vs 100%; P=0.174), contrast-enhanced MR angiography (93.6% vs 94.7%; P=0.789), and cine-CMR (88.1% vs 94.1%; P=0.113).

While an important consideration, the editorialists write, “there were very few thrombi, and the study was underpowered to detect a difference. Furthermore, the diagnostic ability of CMR to detect right atrial or right atrial appendage thrombi was not assessed.”

The investigators acknowledge that, “unlike TEE, which could provide anatomical and physiological data such as LAA-emptying velocity, the evaluated CMR components in our study could provide only anatomical data.”

The longer time required for CMR is also a common criticism, but this could be brought down—albeit not as short as for cardiac CT—if CMR protocols were tailored to assess PV anatomy and LA/LAA thrombi. CMR also provides valuable data for PVI without exposing patients to radiation, they note.

Shah agreed further studies are needed to confirm the results but observed that they currently use CMR on all patients prior to PVI.