Requiem for the “Imager”* in TAVR: An Irrelevant Fiction

Transcatheter aortic valve replacement (TAVR) is rapidly becoming the dominant form of aortic intervention for aortic stenosis (AS), with an expanding list of indications. Initially reserved for inoperable cases, TAVR is now proven to be safe, effective, and non-inferior to surgical aortic valve replacement (SAVR) in patients with low to moderate surgical risk; technical and procedural advances, as well as patient preference, have ensured the wide dissemination of the technique, which is tempered only by yet unresolved uncertainties about the long-term durability of TAVR valves.

It is instructive to examine the evolution of TAVR because in the 20 years since Cribier’s first ever TAVR (16/10/2002), the logistics of the procedure have evolved enormously, and a radical paradigm shift has occurred. Initially, TAVR was a procedure heavily dependent on echocardiography (echo; Table 1). At all stages of the patient’s journey, echo played a fundamental role. Following the diagnosis of AS, almost exclusively established by transthoracic echocardiography (TTE), the first hurdle involved distinguishing between the increasingly complex phenotypes of AS using standard hemodynamic Doppler indices (AVA, peak velocity, and mean gradient) but also, increasingly, stroke volume index (SVi), and flow rates, all interpreted in the context of left ventricular ejection fraction (LVEF). In cases where the severity of AS was not fully elucidated by a detailed TTE, other echo modalities such as transesophageal echocardiography (TOE) or dobutamine stress echo (DSE) were brought in to eliminate diagnostic uncertainty. TOE often served as a gatekeeper to TAVR, because a good quality 3D dataset encompassing the aortic valve (AV), the aortic root, and the left ventricular outflow tract (LVOT) would allow not only an accurate diagnosis of the severity of the AS but would also answer questions fundamental to the procedure: valve landing zone, amount and distribution of calcification, its relationship to the coronary ostia, depth and height of the sinuses of Valsalva, distance between the aortic “annulus”, and the ostium of the left main stem.

Echo really came into its own in the hybrid theater, or the catheterization lab, during TAVR procedures. With the patient under general anesthesia (GA), intubated, and ventilated, the TOE probe is passed into the esophagus and each step of the procedure is monitored and displayed in real-time to the interventional team. Among the plethora of professionals that crowded the TAVR catheterization lab/hybrid theater in the early days (structural interventionalists, vascular surgeons for arterial access and closure, cardiac surgeons for bailout SAVR, perfusionists and operating room assistants on stand-by, catheterization lab nurses, and application specialists from the valve company) the “imager” really took a prominent place, guiding the procedure, diagnosing complications early, directing their treatment, and confirming the adequacy of the final result. Such was the prominence of echo in those early days, that reports exist of TAVR procedures guided exclusively by fluoroscopy without contrast and TOE (to avoid the risk of acute kidney injury in patients with renal impairment).

A difficult intra-procedural task of the imager was to identify and quantify paravalvular leaks (PVL) immediately after valve deployment. These are consistently associated with reduced survival after TAVR, although the mechanism underpinning this association is not clear. PVLs are eccentric, often contained by solid boundaries, and not infrequently obscured by acoustic attenuation from components of the replacement valve; all of these factors make their imaging and their quantification particularly difficult, and place great pressure on the imager. Anything more than mild PVL would prompt balloon post-dilatation of the replacement valve in an effort to seal the leak, but this maneuver carries a definite risk of aortic annular rupture, an often fatal complication.

The classification of PVL severity proposed by the VARC has the advantage of simplicity but it was apparent from the outset to anyone who has ever done TOE to assess PVLs that the simple circumferential extent of the color flow map in a short axis view of the replacement valve is a crude and imprecise parameter for quantifying complex jets with unpredictable trajectories. The inadequacy of the VARC criteria for quantifying PVL has been demonstrated in elegant comparisons of TOE to CMR grading. In these comparisons, CMR consistently reclassifies PVL severity in a large proportion of patients. Quantification of PVL severity by TOE remains an unsolved problem, but fortunately, technological advances in the design and manufacture of TAVR valves, coupled with procedural refinements and increasing experience of the interventional community, has led to a significant reduction in the prevalence of significant PVL, thus removing the mandate for immediate intra-procedural assessment by TOE. Aortography is now the preferred method for diagnosing and quantifying PVL after TAVR, and recent advances with automated videodensitometric quantification of the contrast intensity in the LVOT may overcome the limitations of ultrasound in this arena.

With increasing procedural volumes comes familiarity and a desire to simplify and accelerate the procedure. This is reflected in the so-called minimalistic approach to TAVR, which emphasizes safety and speed. GA is shunned, and conscious sedation is preferred. Vascular access is performed via the Seldinger technique, with a micro-puncture kit under ultrasound guidance, using pre-closure with a suture device. Recovery time is markedly reduced, and early discharge becomes the rule. Without GA, a TOE probe cannot be tolerated for the duration of the procedure, and thus the imager and their echo scanner are suddenly booted out of the catheterization lab.

This shift of emphasis away from echo during the procedure is mirrored by a similarly ample swing in preferred imaging modalities for diagnosis and pre-procedural planning. Where echo once reigned, CT now rules. Even before achieving its current supremacy, CT was brought in as the final arbiter of the severity of AS when the echo indices were discordant, with calcium scoring recommended by guidelines as the discriminator between moderate and severe AS when echo did not allow imagers to confidently distinguish between the two. In the current era, CT almost single-handedly resolves most questions related to diagnosis and pre-procedural planning, including the diagnosis of epicardial coronary stenoses, as well as assessing the suitability of the aorto-ilio-femoral axis for device access. CT has thus become a “one-stop-shop” for TAVR assessment, almost completely obviating the need for TOE as part of the procedural planning, and relegating TTE to the position of a screening test for AS. Structural interventionalists are now perfectly capable of interpreting CT datasets and planning their procedures, completely cutting out “‘the middleman” (i.e., imager).

Intra-procedurally, increasingly slick interventional teams using improved valve designs rely almost exclusively on fluoroscopy for intra-procedural guidance and managing complications. At most, TTE is infrequently called upon for the rare case of tamponade.

Similarly, the identification of hypo-attenuation with leaflet thickening (HALT) and hypo-attenuation with abnormal motion (HAM)—new pathologies exclusively diagnosable by CT and often invisible to echo—further cements the primacy of CT in the current TAVR era.

Is there a role for the imager in TAVR? Current practice would suggest that the answer is a nuanced “no”. Of course, intra-procedural imaging is not dead, as new technologies become available for structural intervention, and TOE remains essential for most procedures that involve valves, but TAVR seems completely settled in a format where there is no space for a pure imaging specialist.