We conducted a retrospective observational study including consecutive patients who underwent TAVI at our center in a 9-year period (from October 2010 to December 2019).

According to the standard practice of our center, all patients were referred for TAVI on the basis of formal, multidisciplinary, Heart Team discussion. Clinical data and procedure details were prospectively entered into the TAVI-dedicated section of an electronic database that allowed previous assessment of the impact of EuroSCORE on coronary interventions10 and the safety of transradial procedures.11

Patients’ surgical risk was graded according to the logistic EuroSCORE and the Society of Thoracic Surgeons (STS) predicted operative mortality at the time of Heart Team consultation. TAVI risk was graded according to the STS/American College of Cardiology Transcatheter Valve Therapy using the on-line TAVI in-hospital mortality risk calculator.12

All patients signed a dedicated informed consent form to undergo the scheduled procedure which includes authorization for database insertion and clinical follow-up assessment.

For the present study, all patients who underwent a TAVI procedure in our institution between October 2010 and December 2019 were retrospectively screened.

All patients underwent coronary evaluation before TAVI. The vast majority of the patients received coronary angiography (not performed only in the case of nonobstructive coronary arteries on computed tomography coronary angiography, 80 patients). Patients were classified as having CAD if they fulfilled 1 of the following conditions: coronary stenosis> 70% in at least 1 major epicardial coronary artery or prior coronary revascularization by either PCI or CABG. For each patient with concomitant CAD, the jeopardized myocardium before the TAVI was graded using the British Cardiovascular Intervention Society (BCIS) Jeopardy Score (JS).13 The extent of coronary revascularization and the use of fractional flow reserve to guide the revascularization were left at the operator's discretion.

For all enrolled patients, clinical records were carefully evaluated, and clinical follow-up was obtained (as in-person visits, telephone interviews, and medical notes from any hospital admission or outpatient visits). Nonfatal myocardial infarction (MI) during follow-up was defined as the rise and fall of cardiac enzymes (usually serum high-sensitivity troponin I) in the presence of electrocardiogram signs or symptoms compatible with myocardial ischemia, as described in the fourth universal definition of MI14 as enrollment started before the publication of the fourth universal definition. Stroke was defined as any new, permanent, global, or focal neurological deficit ascertained by a standard neurological examination, lasting longer than 24hours or less if evidence of cerebral infarction was obtained by imaging. Coronary revascularization was defined as any coronary revascularization (either PCI or CABG) performed after TAVI regardless of the fact that it was related to a previously treated or untreated segment.

The primary study endpoint was major adverse cardiovascular and cerebrovascular events (MACCE) defined as the composite of death, nonfatal MI, nonfatal stroke, and coronary revascularization. The individual components of the primary endpoint constituted the secondary endpoints.

Continuous variables are presented as mean± standard deviation and categorical variables as numbers and percentages. Comparisons of continuous variables across different groups were performed using the Student t test or ANOVA test (as appropriate). Categorical variables were evaluated using the chi-square test or Fisher exact test, as appropriate. A Cox regression analysis was performed to identify the independent predictors of the primary endpoint among the main baseline characteristics (age, sex, cardiovascular risk factors, renal failure, prior MI, prior PCI, prior cardiac surgery, mean aortic gradient, aortic valve area, ejection fraction, NYHA class III-IV, STS mortality, TAVI score, coronary anatomy, and BCIS-JS). Adjusted hazard ratios (HR) with associated 95% confidence interval (95%CI) were calculated for the significant primary endpoint predictors and corresponding adjusted survival curves were determined. The subdistributions of a competing risk with the Fine-Gray test were calculated for the competing risk events included in the secondary endpoints.15 A 2-tailed, P-value <.05 was established as the level of statistical significance. All statistical analyses were performed using SPSS software v22.0 (IBM Corporation, United States) and SAS software 9 (SAS Institute, United States).

Out of 421 patients who underwent TAVI in our hospital from October 2010 to December 2019, 18 patients were excluded (10 died in hospital and 8 patients were lost to follow-up). Thus, 403 patients (180 with concomitant CAD) were enrolled in the study. The median BCIS-JS before TAVI was 4.0 [IQR=0.7]. Thus, the study population was divided in 3 groups: patients without concomitant CAD (no-CAD), patients with CAD and BCIS-JS before TAVI ≤ 4 (CAD BCIS-JS ≤ 4), and patients with concomitant CAD and BCIS-JS before TAVI> 4 (CAD BCIS-JS> 4). By combining the presence or absence of CAD and BCI-JS before TAVI (≤ or> 4), the study population was divided in the following 3 groups: 223 patients with no-CAD, 94 patients with ER-CAD, and 86 patients with IR-CAD (figure 1).

Figure 1.

Study flow chart. CAD, coronary artery disease; BCIS-JS, British Cardiovascular Intervention Society Jeopardy Score; ER-CD, extensively revascularized coronary artery disease; IR-CAD, incompletely revascularized coronary artery disease; TAVI, transcatheter aortic valve implantation.

(0.3MB).

The baseline characteristics of the patients are listed in table 1, showing marked differences between the 3 study subgroups. Among these, in CAD BCIS-JS> 4, male sex was less prevalent (29% vs 48% vs 70%, P <.0001) and the incidence of peripheral artery disease was higher (35% vs 22% vs 23%, P=.004) compared with CAD BCIS-JS ≤ 4 and no-CAD. As expected, the incidence of all cardiovascular risk factors was more frequent in the CAD BCIS-JS> 4 and CAD BCIS-JS ≤ 4 groups compared with no-CAD group (diabetes 41% vs 39% vs 23%, P <.0001; dyslipidemia 77% vs 61% vs 45%, P <.0001; current smoker 7% vs 0.7% vs 2%, P=.042; hypertension 93% vs 91% vs 87%, P=.003; family history 30% vs 27% vs 15%, P=.001). The incidence of atrial fibrillation was higher in the CAD BCIS-JS> 4 and CAD BCIS-JS ≤ 4 groups compared with the no-CAD group (38% vs 37% vs 16%, P <.0001). Similarly, the STS mortality score was higher in patients with concomitant CAD (CAD BCIS-JS> 4 and CAD BCIS-JS ≤ 4) compared with no-CAD patients (6.8±3.1 vs 6.2±3.6 vs 5.8±3.5, P=.004). In contrast, TAVI score did not differ between the 3 study subgroups. Of note, among patients with concomitant CAD, there were no differences in the distribution of coronary disease. Echocardiography characteristics are reported in table 2. In no-CAD patients, the mean aortic gradient was higher (54±15 vs 50±15 vs 47±17, P=.003) and the aortic valve area was smaller (0.6±0.2 vs 0.7±0.2 vs 0.7±0.2, P=.006) compared with patients with concomitant CAD (both CAD BCIS-JS> 4 and CAD BCIS-JS ≤ 4).

All patients had a follow-up of> 3 months [range 104-3 296 days] and the mean length of follow-up was ∼5 years. figure 2 reports the survival curves (table 3) in the 3 study groups showing that CAD BCIS-JS ≤ 4 had event-rates comparable to the no-CAD group and significantly better compared with CAD BCIS-JS> 4 patients (P=.049). As shown in figure 3, this result was driven by a statistically significantly higher occurrence of death in patients with CAD BCIS-JS> 4 compared with those with CAD BCIS-JS ≤ 4 and no-CAD (SHR, 1.29; 95%CI, 0.96-1.74; P=.043). Of note, the incidence of nonfatal MI was higher in patients with concomitant CAD independently of the extent of jeopardized myocardium achieved before TAVI compared with no-CAD patients (SHR, 1.74; 95%CI,1.28-2.50; P=.032). The Cox regression analysis identified BJS <4 (HR, 0.43; 95%CI, 0.21-0.90; P=.02) and NYHA III-IV class (HR, 0.31; 95%CI, 0.12-0.78; P=.03) as independent predictors of the primary endpoint among the main baseline characteristics (table 3). On multivariate analysis (table 4) including all the main baseline characteristics, BCIS-JS ≤ 4 (HR, 0.53; 95%CI, 0.19-0.87; P=.02) and NYHA class III-IV (HR, 0.19; 95%CI, 0.04-0.60; P=.04) were independent predictors of the primary endpoint.

Figure 2.

Primary endpoint according to extent of jeopardized myocardium before transcatheter aortic valve implantation. The figure shows the event-free survival curves in the study population. The coronary artery disease population was stratified according to myocardial burden, assessed by the British Cardiovascular Intervention Society Jeopardy Score (BCIS-JS) before TAVI. BCIS-JS, British Cardiovascular Intervention Society Jeopardy Score; CAD, coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary interventions; TAVI, transcatheter aortic valve implantation.

(0.1MB).

Figure 3.

Single component of the primary endpoint according to extent of jeopardized myocardium before transcatheter aortic valve implantation. The figure shows the cumulative incidence function curve using Fine-Grey competing risk model of the individual components of the primary endpoint in the study population. A, death; B, nonfatal myocardial infarction; C, new-percutaneous coronary interventions; D, nonfatal stroke). The coronary artery disease population was stratified according to myocardial burden, assessed by the British Cardiovascular Intervention Society Jeopardy Score (BCIS-JS) before TAVI. CAD, coronary artery disease; PCI, percutaneous coronary interventions.

(0.24MB).

In 180 patients with concomitant CAD, 88 patients were underwent coronary PCI before TAVI and 92 patients underwent coronary revascularization prior to TAVI indication, considering all previous coronary revascularizations. No patients underwent a staged PCI after TAVI. In these patients, the extent of myocardial revascularization was graded using the BCIS-JS revascularization-index (RI) (range: 0-1) according to the following formula: (BCIS-JS pre -BCIS-JS post)/BCIS-JS pre.13 The median BCIS-JS revascularization index (RI) was 0.67 [0.7]. Thus, the study population was divided into 3 groups: 223 patients without CAD (no-CAD), 99 patients with extensive revascularization (RI> 0.67, ER-CAD), and 81 patients with less extensive revascularization (RI ≥ 0.67, LER-CAD). The baseline characteristics of patients are listed in table 1 and table 2 of the supplementary data. Patients without CAD and patients with more complete revascularization (ER-CAD, RI> 0.67-1.0) had better MACCE-free survival compared with those with less extensive revascularization (LER-CAD, RI ≤ 0.67; P=.042, figure 1 of the supplementary data). This result was driven by a significant mortality reduction ([SHR, 1.31; 95%CI, 0.97-1.54; P=.041], figure 2 of the supplementary data). Of note, the incidence of nonfatal MI was higher in patients with concomitant CAD independently of the revascularization extent achieved before TAVI compared with no-CAD patients ([SHR, 1.74; 95%CI, 1.28-2.50; P=.046], figure 2 of the supplementary data). In the multivariate analysis, BCIS-JS RI ≤ 0.67 (HR, 2.10; 95%CI, 1.35-5.83; P=.002), BCIS-JS post (HR, 0.17; 95%CI, 0.24-1.34; P=.024) and NYHA class III-IV NYHA class III-IV (HR, 1.19; 95%CI 1.02-1.39; P=.018) were independent predictors of MACCE.

First, although this is a large single-center registry on the subject of the impact of CAD on TAVI outcome with simultaneous evaluation of its complexity and management, the sample was too small to comprehensively address and evaluate the clinical impact. Second, coronary revascularization was performed without a protocol and the extent of coronary revascularization and the use of fractional flow reserve to guide revascularization were left at the operator's discretion. Finally, the safety and efficacy of PCI prior to TAVI compared with isolated TAVI can only be best tested in randomized trials.