Just this month, cardiovascular surgeons and diagnostic radiologists from the University Medical Centre in Freiburg, Germany, collaborated to publish an article in The Annals of Thoracic Surgery showing that CT is much better than ultrasound at measuring the diameter of the aortic annulus — an important matter when deciding which size of valve to implant transcutaneously.
This could both dishearten and encourage radiologist Rodrigo Salgado of Antwerp University Hospital, who spent 25 minutes persuading the audience this week both that CT is best for measuring the annulus, and that isn't circular, doesn’t have a radius, and in fact actually doesn't exist.
"Surgeons need a change of viewpoint," he said, presenting this week at ECR 2011. "I recently had a discussion with a surgeon and I said the annulus doesn't exist. He said, ‘For me it does exist. I expect that 10 years from now no surgeon will repeat that the annulus is circular.’"
What's more, he said, he is currently reviewing a prosthesis from a patient who has had the procedure. The shape of the prosthesis is no longer circular. It's oval.
Earlier, Salgado spent a great deal of time depicting and describing the vaguely teacup-shaped structure of the juncture between the ventricle and the aorta that cradles the valve. Then he showed vividly how CT is (in fact, must be) better than ultrasound at visualizing the best access route for valve insertion, the morphology and actual dimensions of its ellipsoid (not circular) base through which the valve must be threaded, the degree of occlusion of the vessels proximal to the valve, the extent of calcification, and the thickness of the ventricle.
"Finally," he said, "we're dealing with an old population, and there are likely to be significant incidental findings [on CT] that alter the procedure." Salgado published a study that appeared last May in The American Journal of Cardiology in which Washington, DC, surgeons reconsidered transcutaneous aortic valve implantation (TAVI) in 19 percent of cases due to CT findings of unexpected vascular findings.
As the population ages and as results continue to improve, he predicted, the use of TAVI will increase. Meanwhile, different valves will certainly appear on the market. Surgeons will need accurate information about them, and current measurement criteria based on ultrasound will not suffice. "They're measuring different things," he said. "We will need new criteria."
The remainder of the session was set up as a contest between the value of the two methodologies for assessing patients before percutaneous interventions (PCI). But during the discussion both "combatant" radiologists (Christian Loewe of the Medical University of Vienna and Marco Francone of Sapienza University of Rome) ended up predictably agreeing that both techniques have important roles.
About two-thirds of the time, PCIs fail, Loewe pointed out, and CT can help surgeons to avoid the risk in patients who are bad candidates, because of occlusion at the distal end of the lumen or collateral arteries branching off the proximal end. He listed the most important deciding factors for success: the length of the vessel, the age of the occlusion, vessel tortuosity, calcification, and lack of visibility. "Most of these can be demonstrated by CT," he said. "CT is the perfect tool for assessing morphology."
Late contrast enhancement also has the potential to demonstrate whether a myocardial infarction scar is subendocardial or transmural, in which case no treatment will be effective. Adenosine stress testing with CT has the potential to provide information about perfusion, and a recently described culprit lesion score may also help to predict which vessels are most suitable for revascularization. "CT might be able to provide information about morphology, function, and viability," he added. "If it could be proven in the clinical setting, it might be the only procedure able to do all three."
Francone championed the potential of MRI to predict viability of myocardium, which may rely on wall thickness, transmurality of the scar, or contractile reserve. MRI can measure them all, he said. He even suggested an algorithm for the assessment of left ventricular dysfunction. It included echocardiography, and MRI, not CT. But he also admitted that as a relatively new technique, the evidence base for cardiac MR lags that of CT, with mostly small single-center studies.
"There is no perfect test," Francone concluded. "You need a combination of them all."
"I think MR for chronic indications is well established," Loewe conceded later. "But there's been a tremendous increase in the application of CT. It's now winning the race."