By Jeffrey Zimmet, MD, PhD
Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center
Dr. Zimmet reports no financial relationships relevant to this field of study.
SYNOPSIS: Performing planned left main percutaneous intervention before or during transcatheter aortic valve replacement does not confer increased risk of short- or intermediate-term adverse outcomes.
SOURCE: Chakravarty T, et al. Outcomes in patients with transcatheter aortic valve replacement and left main stenting: The TAVR-LM Registry. J Am Coll Cardiol 2016;67:951-960.
Severe aortic stenosis is often coincident with obstructive coronary artery disease, and cases involving significant left main (LM) coronary disease are not uncommon. For such patients, surgical aortic valve replacement with coronary artery bypass is clearly the gold standard treatment. However, for patients with advanced age or extensive comorbidities that support transcatheter aortic valve replacement (TAVR), the combination of valve replacement and bypass typically involves a more-involved procedure and prohibitive surgical risk. Although percutaneous intervention (PCI) prior to TAVR has become commonplace for branch-vessel disease, combining LM stenting with TAVR comes with some unique objections. Generally, surgeons perform revascularization by PCI in advance of TAVR to improve the safety of the valve procedure. However, the baseline higher risk of the LM PCI procedure itself is clearly elevated in the setting of severe aortic stenosis (AS), with a very real possibility of hemodynamic compromise. Additionally, the possibility exists of a negative interaction between the TAVR valve frame and a stent at the LM ostium. The initial randomized trials of TAVR did not include significant numbers of patients with LM disease. More recently, randomized trials of intermediate-risk patients, including SURTAVI and PARTNER II, have specifically excluded patients with untreated LM disease.
Recently published initial results from the TAVR-LM registry address this issue. This registry retrospectively collected information on patients who underwent TAVR and LM stenting from 11 high-volume centers in North America and Canada between January 2007 and December 2014. Full clinical and imaging data were collected for each patient, and angiograms and CT scans were evaluated by a core laboratory. Procedural, 30-day, and 1-year outcomes were collected for each patient.
From a pool of 6405 patients undergoing TAVR at the participating centers during the 8-year span, 204 patients (3.2%) were identified who also underwent LM PCI. Of these, 176 patients underwent planned LM PCI before (n = 167) or during (n = 9) TAVR. An additional 19 patients underwent emergent LM PCI due to complications from TAVR, either during the procedure itself or within 24 hours. The final nine patients underwent PCI after TAVR, due to routine progression of coronary disease. These nine patients all underwent uncomplicated PCI procedures, without reported hindrance from the valve frame. One hundred sixty-seven patients with LM stents who underwent TAVR (of whom 102 were unprotected LM stents) were compared with patients without LM revascularization, leading to 128 matched pairs for comparison and analysis.
Baseline characteristics of the matched groups were similar, including age, major comorbidities, and Society of Thoracic Surgeons risk scores. With regard to procedural outcomes, patients with LM PCI were more likely to experience major vascular complications (16.4% vs 3.9%; P < 0.01) and require permanent pacemakers (26.6% vs 14.1%; P = 0.02). Mortality at 30 days (3.1% vs 2.3%; P = 0.67) and at 1 year (9.4% vs 10.2%; P = 0.83) was not statistically different between the groups. Other 30-day and 1-year outcomes also were not statistically significant, although there was a trend toward higher target vessel revascularization in the LM PCI group (5.5% vs 1.6%; P = 0.06).
As expected, patients experiencing unplanned LM PCI due to complications during or directly after TAVR fared worse than those with planned PCI. Eighteen of 19 patients with unplanned LM PCI experienced LM occlusion during valve deployment, while one experienced LM dissection. These patients were more likely to develop cardiogenic shock (21.1% vs 3.4%; P < 0.001), to need CPR (15.8% vs 0.6%; P < 0.001), and to develop renal failure (26.3% vs 5.8%; P = 0.002). Thirty-day (15.8% vs 3.4%; P = 0.013) and 1-year (21.1% vs 8.0%; P = 0.071) mortality also greatly increased in this group.
The authors concluded performing planned LM PCI before or during TAVR does not confer increased risk of short- or intermediate-term adverse outcomes, and argue that coexisting LM disease and AS should not prevent evaluation for TAVR.
COMMENTARY
This is a significant study in several respects. Data on this important subset of TAVR patients have been in short supply, and it is essential that going forward clinical decisions be made on the basis of good evidence rather than anecdotal information. Clearly, some early fears about combining LM stenting and TAVR are unfounded, and the safety of implanting TAVR valves, even with ostial LM stents, is no longer in doubt. The data also support the ability to perform PCI following TAVR, although the numbers are much smaller.
This is not to say physicians should regard unrevascularized LM disease in potential TAVR patients as a minor issue. Rather, physicians should remember that this is a registry and not a randomized trial. Patients with highly complex LM disease and critically severe AS are undoubtedly at very high risk for intervention. Physicians have no information on patients who were deemed too high risk for intervention or who never made it to a TAVR procedure.
The issue of unplanned LM intervention during TAVR is an important one to recognize, although it addresses a different question from the main cohort of planned LM PCI. Patients with low LM ostia, effaced aortic sinuses, heavily calcified cusps, and LM stents that protrude far into the aorta are at increased risk of coronary occlusion at the time of the TAVR procedure. Mostly, ECG-gated CT scans recognize these risk factors pre-procedure. However, do not underestimate the risk for adverse outcomes in this population.
