Coronary Obstruction Following TAVR is Rare, Often Fatal, and (Sometimes) Unpredictable

A large registry of over thirteen thousand transcatheter aortic valve replacement (TAVR) procedures showed that coronary obstruction following TAVR is rare, carries a high fatality rate, and is not easily predictable based on pre-procedure imaging.¹

Coronary obstruction (CO) remains a feared complication of TAVR. Predicting is a key goal for the pre-procedure planning computed tomography (CT) scan. Markers indicating high risk for CO following TAVR include low coronary ostial height, narrow sinotubular junction, small sinuses of Valsalva, and long valve leaflets. For valve-in-valve TAVR procedures, the valve-to-coronary ostium (VTC) distance is an additional risk factor. If high likelihood of CO is anticipated, strategies to mitigate risk include: 1) “coronary protection” with a wire or undeployed stent left within the at-risk coronary; and 2) specialized procedures to modify the respective leaflet with either the BASILICA technique to lacerate the culprit leaflet (bioprosthetic or native aortic scallop intentional laceration to prevent iatrogenic coronary obstruction during TAVR) or the UNICORN technique to obliterate the leaflet (undermining Iatrogenic Coronary Obstruction With Radiofrequency Needle). However, knowledge of the true risk of CO as well as our ability to predict at-risk individuals are instrumental to guide both shared decision-making as well as deployment of mitigation strategies.

Ojeda and colleagues describe the largest experience of coronary obstruction following TAVR. They reviewed incident CO in all patients from the Spanish TAVR (Transcatheter Aortic Valve Implantation) registry augmented with additional review of imaging and clinical features for patients who suffered from CO. This database included N = 13,675 TAVR procedures across 46 Spanish centers from 2009-2021. There are several key findings.

• The incidence of coronary obstruction was low (N = 115, 0.8%), but remained unchanged across more than a decade (range, 0.3% to 1.3%), despite improvements in peri-procedural imaging, procedural techniques, and increased awareness. In cases of CO, a similar proportion had received self-expandable valves (SEV) and balloon-expandable valves (BEV) compared with the control population (TAVR without CO). However, acute obstruction was more common with BEVs whereas delayed obstruction mainly occurred with SEVs.
• The risk of coronary obstruction is hard to predict from pre-procedure CT scans, at least for native aortic valves. A combination of at least 2 anatomic risk factors based on the pre-procedure CT was present in only 31.7% of all valves, but 78.3% of prosthetic valves. A valve-in-valve (ViV) TAVR itself poses a higher likelihood of CO, especially when the prosthetic valve had externally mounted leaflets. Compared with controls, patients who had CO were far more likely to have had a ViV TAVR (21.7% vs 3.1%; P = 0.001) and 72% of these had a prosthesis with externally mounted leaflets. Interestingly, coronary protection was performed in only 18.2% of patient who eventually had CO.
• The presentation of coronary obstruction following TAVR is variable. Most cases of CO were detected immediately after valve deployment (N = 96, 83.5%), presenting as acute coronary syndrome (46.1%; 47.2% with and 52.8% without ST-segment elevation), cardiac arrest (41.7%), cardiogenic shock (7.8%), or unstable angina (4.3%). The culprit artery obstructed was the left main trunk in 68.7%, right coronary artery in 23.5%, and both vessels in 7.8% of patients.
• Revascularization following coronary obstruction had a low success rate, which was even lower for self-expandable valves. The management of CO included percutaneous coronary intervention (PCI, N = 100, 86.9%), coronary artery bypass grafting (CABG, N = 4, 3.5%), and medical management (N = 11, 9.6%). Technical success for attempted PCI was low at 78% overall, and lower for SEVs compared with BEVs (69.7% vs 88.9%, P = 0.017).
• Coronary obstruction following TAVR carries a high mortality rate. In-hospital mortality was far higher for patients with CO than without (37.4% vs 4.1%, P = 0.001) and varied according to timing of diagnosis of CO (35.4% when diagnosed during the procedure versus 66.7% when diagnosed later). Unsurprisingly, mortality after CO was higher when PCI was either unsuccessful or not attempted compared with successful PCI or CABG (78.8% vs 20.7%, P = 0.001).

These results paint a cautionary tale for TAVR operators and their patients. While the incidence of coronary obstruction is objectively low (<1%), it has not changed over time despite tremendous innovation, remains disappointingly unpredictable, and carries a sobering mortality risk. The stakes are even higher as TAVR is increasingly sought for low-risk younger patients. Time and again, we are reminded to remain humble when counseling our patients about the virtues of TAVR.

References

1. Ojeda S, González-Manzanares R, Jiménez-Quevedo P, et al. Coronary Obstruction After Transcatheter Aortic Valve Replacement: Insights From the Spanish TAVI Registry. JACC Cardiovasc Interv. 2023;16(10):1208-1217. doi:10.1016/j.jcin.2023.03.024