The Splitting of Leaflets to Prevent Coronary Occlusion During TAVR^∗Free Access

Creativity Energy and Scientific Discipline

The line between impulsive madness and creative ethical scientific investigation must be drawn. On the one hand, the frontiers of medicine are advanced with new and creative ideas versus, on the other hand, the back alleys of tinkering-dabbling may lead to a serious compromise of a patient’s well-being when the physician attempts something novel with little to no planning. The early days of cardiac surgery are worth remembering as they too were intense and required physician-mavericks to think “out-of-the-box” (1).

The inventors and investigators of the bioprosthetic or native aortic scallop intentional laceration to prevent iatrogenic coronary artery obstruction during TAVR (BASILICA) technique present their initial results in this issue of JACC: Cardiovascular Interventions (2). The steps they took to produce these data involve much more than meets the eye. They developed this technique starting with animal studies, using National Heart, Lung, and Blood Institute funding, eliciting Food and Drug Administration guidance with first-in-human study protocols, partnering with industry, gathering data, refining techniques, and subsequently managing limited dissemination.

A Critical but Supportive Review of BASILICA

There are many unanswered questions regarding this technique that must be addressed before it can be broadly applied to patients and outside of a protocol with approval of an institutional review board to protect human subjects. This report of 30 patients is only a start and truly is an early feasibility study.

How rare is coronary obstruction during TAVR in 2019? Previous reports may over-represent its current frequency. In our health care system, there has been 1 case in over 1,500 TAVRs and then the computed tomography angiography (CTA) was of suboptimal quality to predict obstruction. One additional patient deemed at risk by CTA was referred out of state for TAVR with BASILICA.

Is stroke risk increased with BASILICA? We await further evidence of whether or not the liberation of debris that results in stroke is greater when TAVR is accompanied by BASILICA. Perhaps examining cerebral protection filters for valve fragments will provide some clues, but more rigorous study of the risk of stroke with BASILICA is needed.

The additional time BASILICA adds to the TAVR procedure is substantial, and it also remains unclear whether this requires transesophageal echocardiography guidance. If coronary obstruction is highly likely and surgical valve replacement is not an option, then the time, increased procedure complexity, and even a small increased risk of stroke may justify BASILICA.

Splitting an aortic valve cusp to prevent coronary obstruction first requires identification of those patients at risk. Predictive algorithms, perhaps more of an informed gestalt, were used in this study, and others have identified predictive features from the cardiac CTA (3,4). These have not risen to the level of having a predictive value with high sensitivity and specificity to allow accurate identification of patients who will have coronary obstruction. Using a low threshold will lead to exposing patients to the risks of BASILCA without benefit. Using a high threshold could lead to underutilization and coronary obstruction occurring.

When does the technique fail, how does it fail, and does the splitting last forever? The investigators have anecdotal evidence that confluent heavy leaflet calcification at the nadir of the target aortic leaflet may predict failure to traverse the leaflet. Can this technique be effectively applied to the bulky leaflets encountered in bicuspid aortic stenosis? What quality of CTA is needed to predict obstruction, and how can the leaflet be split exactly at the location of the coronary ostium at risk?

Will this ever become a technique ready to be used by hundreds of interventionalists? There are many technique details that are now being refined, characterized, and performed by a select group of cardiologists. In this study, 30 procedures were performed at 4 to 5 institutions; it is difficult to apply the word “experienced” to even the investigators.

Who should be performing BASILICA? Let us assume the frequency of coronary obstruction in native valve TAVR is 1 in 500. Between 2015 and 2017, the median annualized U.S. hospital TAVR volume was 54 (interquartile range: 36 to 86), and operator TAVR volume was 27 (interquartile range: 17 to 43) (5). Is it worth our effort to learn BASILICA to perform it once every 5 years? After spending time to learn the technique, would our threshold to perform it be inappropriately lowered? Furthermore, the results of a recent analysis of over 100,000 patients entered into the STS-ACC TVT (Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy) registry definitively reaffirms an inverse relationship between the volume of TAVR and the risk of death following the procedure (5). All of these facts argue that only selective centers should be offering BASILICA. How many are needed in the United States? A system of valve care that triages complex patients to high-volume and more comprehensive valve centers is proposed, but for BASILICA, do we need more than a handful of centers in the United States (6)?

Practical aspects of triaging patients with potentially some risk of coronary obstruction need to be considered. Pre-TAVR balloon valvuloplasty does allow an aortogram to be performed with the aortic leaflets held in an open position as would occur after TAVR. If performed in a left anterior oblique projection, it is possible to see the leaflets, their proximity to coronary ostia, and whether coronary flow appears compromised. Use of a fully retrievable TAVR system would also provide a bail-out if coronary obstruction appears imminent. These scenarios may result in an aborted procedure. Preparation for coronary stenting with the snorkeling technique is an option, but this technique has many potential problems in both efficacy and safety. Should such patients be referred to a BASILICA center, or should the recommendation be that they should have surgical AVR unless they have prohibitive risk? For a patient at low or intermediate risk for surgical AVR and high risk of potentially fatal coronary obstruction from TAVR, I would currently recommend surgery.

What is the next step to answer the key questions now that an early feasibility study has been completed? Is a randomized trial next? If so, what would you compare BASILICA to—snorkeling or surgical AVR? Would a large registry be adequate, or could it be misleading if a substantial number of patients were entered who would not have had coronary obstruction without BASILICA?

The investigators raise some additional potential applications of BASILICA that may impact how this technique is further studied. They report the absence of hypoattenuated leaflet thickening in TAVR leaflets adjacent to lacerated aortic leaflets. Could access to coronaries for future percutaneous coronary intervention be enhanced by proactive splitting of leaflets pre-TAVR?

Other issues that must be clarified include the regulatory pathway, the evidence needed for Centers for Medicare & Medicaid Services coverage, the role of the TAVR companies in supporting cases involving BASILICA, and the interest and liability of companies supplying the equipment. Furthermore, there have been more cases performed outside of this investigation, by new operators, and potentially outside of an institutional review board protocol. Bad results are much less likely to be published than successes. The encouraging preliminary efforts presented by the investigators could experience irreparable damage if BASILICA-related deaths or major strokes emerge by premature dispersion of the technique and hasty reclassification that it is no longer an investigative technique.

We will capture this real-world experience, hopefully performed under local institutional review board protocols, by adding BASILICA to a dropdown list of adjunctive procedures in the data collection form of the TVT Registry, as we did in January 2018 for cerebral protection. Gathering data on new developments in transcatheter valve therapies is a major goal of the TVT registry (7).

In conclusion, these early results are encouraging enough, in my mind, to cheer on the champions of this technique. What the next steps are and how the questions posed (vide supra) will be answered are critical to define and organize.

We appreciate and congratulate the creativity, skill, and courage that the investigators have shown to get us to this place.