Influence of coronary artery disease and percutaneous coronary intervention on mid‐term outcomes in patients with aortic valve stenosis treated with transcatheter aortic valve implantation

Abstract Background A high frequency of coronary artery disease (CAD) is reported in patients with severe aortic valve stenosis (AS) who undergo transcatheter aortic valve implantation (TAVI). However, the optimal management of CAD in these patients remains unknown. Hypothesis We hypothesis that AS patients with TAVI complicated by CAD have poor prognosis. His study evaluates the prognoses of patients with CAD and severe AS after TAVI. Methods We divided 186 patients with severe AS undergoing TAVI into three groups: those with CAD involving the left main coronary (LM) or proximal left anterior descending artery (LAD) lesion (the CAD[LADp] group), those with CAD not involving the LM or a LAD proximal lesion (the CAD[non‐LADp] group), and those without CAD (Non‐CAD group). Clinical outcomes were compared among the three groups. Results The CAD[LADp] group showed a higher incidence of major adverse cardiovascular and cerebrovascular events (MACCEs) and all‐cause mortality than the other two groups (log‐rank p = .001 and p = .008, respectively). Even after adjustment for STS score and percutaneous coronary intervention (PCI) before TAVI, CAD[LADp] remained associated with MACCE and all‐cause mortality. However, PCI for an LM or LAD proximal lesion pre‐TAVI did not reduce the risk of these outcomes. Conclusions CAD with an LM or LAD proximal lesion is a strong independent predictor of mid‐term MACCEs and all‐cause mortality in patients with severe AS treated with TAVI. PCI before TAVI did not influence the outcomes.

Conclusions: CAD with an LM or LAD proximal lesion is a strong independent predictor of mid-term MACCEs and all-cause mortality in patients with severe AS treated with TAVI. PCI before TAVI did not influence the outcomes.

K E Y W O R D S
coronary artery disease, severe aortic valve stenosis, transcatheter aortic valve implantation

| INTRODUCTION
Moderate or severe mitral and aortic valvular diseases are extremely common in elderly populations, and their incidence increases with age. 1 In Japan and worldwide, the population is aging rapidly and the problems presented by valvular diseases have become a serious public health issue. Transcatheter aortic valve implantation (TAVI) is one of the treatment options for severe aortic valve stenosis (AS), especially for older patients. In 2019, the Placement of AoRTic traNscathetER valves (PARTNER) 3 trial 2 and the Evolut Low Risk Trial 3 revealed the effectiveness of TAVI for even low-risk patients. It is expected that the number of TAVI cases will continue to grow. In severe AS patients, the reported prevalence of coronary artery disease (CAD) as a comorbid disorder has ranged widely from 11% 4 to 63%. 5 Among the previous studies, the PARTNER trials, which are worldwide randomized clinical trials of TAVI, reported a comparably higher rate of concomitant CAD with severe AS. 6,7 However, the optimal management method for patients with both CAD and severe AS has not been established. In the era of TAVI, it is important to determine whether TAVI can and should be performed before percutaneous coronary intervention (PCI) for elderly patients with CAD and severe AS. It is not yet known whether or not concomitant CAD in severe AS patients increases the risks posed by the TAVI procedure, and fewer data have been published from relevant studies in Asia compared to those from Europe and the U.S. Left main coronary (LM) artery CAD and proximal left anterior descending (LAD) artery CAD are considered high-risk features because these arteries supply the large area of myocardium.
However, the strategy of conducting a PCI before TAVI for coronary lesions has not been established. We therefore retrospectively analyzed the prognoses of patients with CAD and severe AS who underwent TAVI.

| Study population and study design
This study was a retrospective single-center cohort study. We analyzed the cases of the series of 199 consecutive symptomatic patients with severe AS who were treated with TAVI at St. Marianna University School of Medicine, Kawasaki, Japan since January 2016. All patients' cases and options were discussed by our cardiology team for appropriate adaptation. Before undergoing the TAVI, the patients were examined by coronary angiography (CAG) or coronary computed tomography angiography (CTA). We excluded 13 patients who underwent coronary artery bypass grafting (CABG) (Figure 1). the patients in whom CAD was detected at sites other than the LM and proximal LAD by CAG or CTA. The non-CAD group (n = 108) was the patients in whom CAD was not detected by CAG or CTA. CAD was defined as >75% stenosis of ≥1 major branch on CAG/CTA or 50% stenosis only in the LM based on the AHA classification derived from the visual assessment of CAG or CTA results. The angiographic assessment of CAD was based on the consensus of three experienced cardiologists. If PCI had been performed previously and there was no re-stenosis at the time of coronary screening, the patient was assigned to the non-CAD group. When deciding whether to perform PCI for the patients with CAD, we took into consideration the ischemic burden, the technical complexity, the risk of in-stent restenosis, and the patient's general condition (e.g., the presence of chronic F I G U R E 1 Flowchart of this study. AS, aortic stenosis; CABG, coronary artery bypass grafting; CAD, coronary artery disease, LAD, left anterior descending artery; CAG, coronary angiography; PCI, percutaneous coronary intervention; TAVI, transcatheter aortic valve implantation kidney disease). PCI for CAD detected by this coronary screening was basically implemented before TAVI as an independent hospitalization.
After discharge, the outpatient follow-up examinations were performed at 1 month, 12 months, and then annually. We conducted a follow-up survey of the patients' outcomes by checking their clinical records and consulting the patients' referring physicians.

| TAVI and PCI procedures
Before each TAVI procedure, we generally added on antiplatelet therapy (100 mg of aspirin or 75 mg of clopidogrel) or anticoagulant therapy (warfarin or direct oral anticoagulants). Procedural anticoagulation was achieved by heparinization, and the target activated clotting time was 250-300 msec. We used two TAVI systems: the Medtronic CoreValve™ or Evolut R™/PRO™ (Medtronic, Minneapolis, MN) and the Edwards SAPIEN XT™/3™ (Edwards Lifesciences, Irvine, CA). The choice of valve type was discussed by a heart team conference made up of cardiologists, cardiac surgeons, anesthesiologists, nurses, and clinical engineers. Balloon expandable valves were generally chosen for CAD patients indicated for PCI, so that the coronary artery could be easily accessed after TAVI. However, consideration was also made for factors other than the coronary artery itself (e.g., the status of the access route, the annulus size, patient frailty, the presence or absence of atrioventricular block) in order to make a comprehensive decision. All patients were initially screened for a transfemoral approach. In cases in which transfemoral access was not possible, the patients were evaluated for alternative access sites in the following order of preference: subclavian arteries, left ventricular apex, ascending aorta.
The PCI procedure was initiated following the administration of unfractionated heparin (100 IU/kg) into the peripheral vein. The decision to choose either a drug-eluting stent or a drug-coated balloon was left to the operators' discretion. The patients generally continued dual antiplatelet therapy for 1 year after the PCI.

| Outcomes
The primary outcome was the major adverse cardiovascular and cerebrovascular events (MACCEs) defined as non-lethal myocardial infarction, unstable angina pectoris, heart failure requiring hospitalization, stroke, and cardiovascular death. The secondary outcome was all-cause mortality.

| Statistical analyses
All statistical analyses were carried out with SPSS/Windows, ver. 22.0 (SPSS, Chicago, IL). Data are presented as the mean (±SD) or as frequencies (percentages). We used the chi-squared test, an analysis of variance, and the Kruskal-Wallis test to examine differences among three groups, and we performed a Kaplan-Meier analysis and a multivariate Cox regression analysis to examine changes over time to the endpoints. Bonferroni corrected post-hoc tests were conducted to determine where these differences occurred. A two-tailed p value <.05 was considered significant.

| RESULTS
A total of 186 consecutive patients underwent a TAVI, with a followup duration of 405 ± 236 days. The baseline characteristics of the patients of three groups in this study are summarized in Table 1.
Among the 186 patients, the mean age was 84 years old; 85% were hypertensive; 50% had dyslipidemia; 26% had diabetes. The 8% had a past medical history of angina pectoris; 14% had peripheral artery disease; and 12% had cerebrovascular disease. The prevalence of CAD in the patients was 42%. PCI was successfully performed without any complications in 29 patients. The univariate analysis demonstrated that CAD in an LM or LAD proximal lesion was significantly associated with increased incidences  Figure S1).

| DISCUSSION
The main findings of this study were as follows. (1)    Note: Values are expressed as the mean ± SD or the median (Q1-Q3) or number (percentage) unless otherwise indicated. Bonferroni correction in which p < .0167 is significant.
a Syntax score was calculated only for patients performed coronary angiography.
b N-terminal pro-brain natriuretic peptide (NT-proBNP) was adjusted by 6th the level of BNP in patients who did not have NT-proBNP data. 19 c Renin-angiotensin system (RAS) inhibitors include either angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers.

CAD[LADp] group had a higher incidence of MACCE and a higher
incidence of all-cause mortality than the other two patient groups.
(3) The multivariate Cox regression adjusted for the STS score and the performance of a PCI before TAVI revealed that CAD involving an LM or LAD lesion was independently associated with increased rates of MACCEs and all-cause mortality. (4) PCI before TAVI did not affect the patients' outcomes.
In this series of patients with severe AS, the prevalence of CAD was 42%. In previous TAVI studies the prevalences of CAD ranged from 11% 4 to 63%, 5 and our present results were roughly intermediate between these findings.
The impact of coexisting CAD on clinical outcomes in TAVI patients is very controversial. Dewey et al. reported that coexisting CAD negatively influenced procedural outcomes and long-term survival in TAVI patients. 8 On the other hand, Ussia et al. reported that CAD coexisting with severe AS did not impact procedural outcomes and mid-term incidence of MACCE and survival in elderly patients undergoing TAVI. 9 However, a multivariate analysis performed in a large number of TAVI patients in the UK reported that CAD was one of the independent predictors of 2-year mortality. 10 We classified the present TAVI candidates into three groups, and our analyses revealed that (1) the CAD[LADp] group had higher incidences of MACCEs and all-cause mortality than the other two groups, and (2) CAD involving an LM or LAD proximal lesion was an independent prognostic factor for MACCEs and all-cause mortality. Our present study indicates that CAD influences the prognosis of those TAVI candidates with lesions in the LM or proximal LAD, which supply the largest area of myocardium among the coronary arteries.
We also found that the performance of a PCI before the TAVI did not influence the incidence of MACCEs or all-cause mortality in the multivariate Cox regression analysis. A previous report showed that there was no difference in 2-year mortality between PCI plus TAVI showed that an addition of PCI to optimal medical therapy (OMT) did not reduce the risk of death, myocardial infarction, or other major cardiovascular events in patients with stable angina pectoris, 12 which may support our present findings. Another study demonstrated that performing a PCI along with a TAVI during the patient's same hospital admission is associated with a higher mortality compared to performing TAVI alone. 13 Unlike the results of that investigation, our present study did not find that PCI is harmful. This difference may be attributable to the fact that the timing of the PCI, the coronary artery anatomy, and the SYNTAX score were more thoroughly analyzed in our present investigation At least two studies have described that coronary revascularization before TAVI seemed to be safe, and the outcomes of this strategy were the same as those observed in non-CAD TAVI patients. 14,15 Our present findings also suggest that coronary revascularization before TAVI is safe, as supported by other studies.
On the other hand, it is reported that the reduction in ischemic burden achieved by PCI reduces the risk for death or myocardial infarction.
A COURAGE trial nuclear substudy suggested that a treatment target of more than 5% ischemia reduction is suitable for adding PCI to OMT. 16 However, this finding is not conclusive since the results obtained using the current method of measuring the ischemic burden in severe AS patients are inconsistent. Fractional flow reserve (FFR) guided PCI may change the prognosis, 17 but it can be challenging to define the optimal timing to assess the functional severity of coronary artery lesions using the flow wire method in patients with AS. A recent report showed that coronary hemodynamics are influenced by TAVI and AS removal. 18 In addition, the instantaneous wave-free ratio (iFR) is also a useful tool to evaluate the coronary physiology. Recently, using a cut-off value of 0.83 instead of 0.89 as a standard threshold was shown to increase the accuracy of coronary hemodynamics measurement in patients with severe AS and CAD. 19  More evidence will be needed before reaching any definitive conclusion.

| CONCLUSIONS
A lesion in the LM and a LAD proximal lesion are both a strong independent prognostic factor for patients with severe AS treated with a TAVI, and the performance of a PCI before the TAVI did not influence the present patients' mid-term MACCEs or all-cause mortality. For AS patients who undergo a TAVI and are complicated by CAD involving an LM or LAD proximal lesion, more intensive management may be necessary to improve clinical outcomes.