PARIS – Advances in CT scanner technology make cardiac CT a viable alternative to transesophageal echocardiography for preprocedural detection of left atrial appendage thrombus in candidates for transcatheter aortic valve replacement, Dr. Paul D. Williams said at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.
This strategy helps simplify transcatheter aortic valve replacement (TAVR). That has become a major goal for the field now that TAVR’s safety and effectiveness have been established, said Dr. Williams of James Cook University Hospital in Middlesbrough, England.
“With the use of CT as the preferred method for annular sizing prior to the procedure and the increasing use of conscious sedation, transesophageal echo can be avoided altogether in many patients. And identification of left atrial appendage thrombus [LAAT] on CT may lead to changes in management, including optimization of oral anticoagulation, the use of an embolic protection device, and possibly obtaining consent from patients for a higher risk of procedural stroke,” he added.
Transesophageal echocardiography (TEE) has long been considered the gold standard method for detecting LAAT. But it has several disadvantages: It’s invasive, requires heavy sedation, and poses a small risk of serious complications.
Dr. Williams presented a single-center prospective study involving 198 consecutive patients who underwent dual source CT scanning with retrospective gated acquisition and flash angiographic acquisition during their pre-TAVR workup. The study showed that atrial fibrillation (AF) is very common in TAVR candidates, that LAAT is far more common in the TAVR population with AF than in the general AF population, and that while TAVR can still be performed in patients with LAAT, the periprocedural stroke risk appears to be higher.
Of the 198 TAVR candidates, 32% had AF. Two independent cardiologists with CT expertise rated 11.1% of TAVR candidates as having definite LAAT on the basis of a filling defect in both phases of imaging. Another 83.8% were deemed to definitely not have LAAT, while in 5.1% of cases the CT image quality wasn’t sufficient to render a judgment.
Dr. Paul D. Williams
“The literature would suggest only about 5% of patients in the general AF population have LAAT. The rate is much higher in a TAVR population,” the cardiologist observed.
As expected, AF was a strong predictor of LAAT being found on CT, with a 32% prevalence in the AF subgroup, compared with just 1.6% in patients without AF.
Ninety-eight patients with a diagnostic CT also had a TEE. Six of the eight with LAAT on CT also showed LAAT on TEE. Two patients had LAAT on CT but not TEE. Thus, CT had 100% sensitivity, 97.8% specificity, a 75% positive predictive value, and a 100% negative predictive value, Dr. Williams continued.
Of the 198 patients evaluated by CT, 124 actually underwent TAVR. AF was present in 34% of these patients, whose mean CHA2DS2-VASc score was 3.7. CT showed that 8.1% of the patients who had TAVR had definite LAAT, and 84.7% definitely did not.
Six of the 124 patients (4.8%) had a stroke during their hospital stay for TAVR. Two of the six had LAAT on their preprocedural CT; both were being anticoagulated with warfarin at the time. The other four patients with periprocedural stroke didn’t have AF, were negative for LAAT on preprocedural CT, and weren’t being anticoagulated.
“Importantly, in the overall TAVR cohort, 8 of the 10 patients with LAAT on CT did not have a clinically evident periprocedural stroke,” Dr. Williams noted.
Session chair Dr. Rajendra Makkar, director of interventional cardiology at the Cedars-Sinai Heart Institute in Los Angeles, commented, “Your study has very, very important implications for how we actually change the management of some of our patients. You’ve shown LAAT is much more important in TAVR patients than I’d thought. A stroke risk of 20% with LAAT versus 3.8% in patients without LAAT is an impressive difference.”
His take away lesson from the study, Dr. Makkar added, is that if a patient has a preprocedural CT scan that’s negative for LAAT, there’s no need to do a TEE. If CT is positive or nondefinitive, it makes sense to review the patient’s anticoagulation regimen, then bring the patient back a few weeks later for a TEE to see if the LAAT has resolved.
Dr. Williams replied that’s exactly the practice now being followed at his hospital. If CT shows LAAT in a patient with AF who’s already on warfarin, physicians will consider aiming for a higher target INR [international normalized ratio], or they’ll switch to one of the novel oral anticoagulants if there’s a warfarin compliance issue. Then they’ll bring the patient back for a TEE.