Long-term risk of unplanned percutaneous coronary intervention after transcatheter aortic valve replacement

Introduction

A majority of patients with severe aortic stenosis (AS) can be safely and effectively treated with both transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) based on periprocedural risk and anatomic criteria12. However, the lifetime management of patients with AS introduces an additional level of complexity to the Heart Team’s decision. The selection of the optimal treatment strategy for younger patients with a longer life expectancy requires the integration of downstream cardiac risk due to concomitant coronary artery disease (CAD), combined valvulopathies, and repeat and future cardiac interventions3.

Coronary access after TAVR is a concern with important repercussions on lifetime management of patients with severe AS. Previous studies have detailed the challenges of selective coronary access for percutaneous coronary intervention (PCI) after TAVR, particularly in the presence of supra-annular devices featuring tall stent frames12. The need for repeat TAVR in younger patients is anticipated to further exacerbate the complexity of coronary access, particularly in the setting of low coronary offtake in relation to the neoskirt and misalignment of the two stent frames345. The complexity of coronary access directly translates into adverse clinical outcomes in patients presenting with acute coronary syndrome (ACS)6 and needs to be anticipated before the implantation of the first transcatheter heart valve (THV). Thus, in order to tailor lifetime management to an individual context, it is crucial to estimate during treatment for AS the probability that PCI will be required in the future.

In the present study, we aimed to evaluate the incidence, characteristics, and predictors of unplanned PCI after TAVR in a prospective TAVR registry.

Methods

Study population

The Bern TAVR registry is a prospective registry enrolling consecutive patients undergoing TAVR at Bern University Hospital, which forms part of the nationwide SwissTAVI Registry (ClinicalTrials.gov: NCT01368250). The registry was approved by the ethics committee, and all patients provided written informed consent for participation. The study was conducted in compliance with the Declaration of Helsinki.

Patients undergoing TAVR were systematically screened for concomitant CAD by means of coronary angiography prior to TAVR. Concomitant CAD was treated with PCI before, during, or after TAVR based on the Heart Team's decision, taking into account myocardium at risk, lesion complexity, and symptom status7. Functional ischaemia testing was not routinely performed during the study period.

Data collection and outcome measures

Baseline clinical characteristics, procedural, and follow-up data were prospectively recorded in a web-based database. Computed tomographic imaging data were independently re-evaluated by dedicated imaging specialists, as previously described, and integrated into the database8. The presence of CAD was defined by a history of surgical and/or percutaneous coronary revascularisation, previous myocardial infarction (MI), and/or at least 1 significant lesion (diameter stenosis ≥50%) in a major native coronary artery by visual assessment with coronary angiography prior to TAVR9.

Regular clinical follow-up was scheduled at 30 days, 1 year, 5 years, and 10 years after TAVR, and the data were obtained by standardised interviews, documentation from referring physicians, and hospital discharge summaries. All adverse events, including unplanned PCI, were systematically collected and adjudicated by a dedicated clinical event committee. The Clinical Trials Unit Bern was responsible for central data monitoring to verify the completeness and accuracy of data, and to perform independent statistical analysis. In the registry, unplanned PCI included all PCI following TAVR, excluding staged PCI planned at the time of TAVR. PCI was performed in accordance with guidelines at the corresponding time of intervention7. For the purpose of the present study, coronary revascularisation resulting from mechanical coronary obstruction complicating TAVR was excluded.

Statistical analysis

Categorical variables are reported as frequencies and percentages. Continuous variables are presented as mean values±standard deviation (SD) or median values with interquartile ranges (IQR). Univariable Cox proportional hazard models were used to calculate hazard ratios (HR), 95% confidence intervals (CI), and p-values. The cumulative rate of PCI over time was represented through a cumulative incidence curve. Multivariable Fine and Gray regressions (with all-cause death as a competing risk) were used to build a prediction model for the incidence of unplanned PCI. All variables potentially related to unplanned PCI were tested in a univariable model, and variables with a p-value <0.2 were subsequently entered into a multivariable model. All p-values were 2-sided, and a p-value <0.05 was considered significant for all tests. All statistical analyses were performed with Stata 17.0 (StataCorp).

Results

Incidence of unplanned PCI

Among 3,015 patients undergoing TAVR between August 2007 and December 2020, 109 patients (3.6%) underwent staged PCI after TAVR and 67 patients (2.2%) underwent unplanned PCI during follow-up. The median follow-up time available for living patients was 1,095 (IQR 366-1,824) days, and the median time to death was 799 (IQR 268-1,485) days. The cumulative incidence rates of unplanned PCI were 0.4%, 1.6%, and 2.3% at 1, 5, and 10 years, respectively (Figure 1). The median interval from TAVR to unplanned PCI was 605 (IQR 292-1,340) days.

Figure 1. Lifetime risk of unplanned PCI after TAVR. Cumulative incidence of unplanned PCI after TAVR considering competing risk with death is shown. PCI: percutaneous coronary intervention; TAVR: transcatheter aortic valve replacement

Baseline and procedural characteristics

Baseline characteristics of patients with and without unplanned PCI after TAVR are shown in Table 1. Patients who underwent unplanned PCI after TAVR were younger (80.2±6.5 years vs 81.9±6.4 years; p=0.028), more likely to be male (75% vs 50%; p<0.001), and more frequently had peripheral artery disease (21% vs 13%; p=0.048) than those without unplanned PCI. There were no significant differences in traditional risk factors (hypertension, diabetes mellitus, and dyslipidaemia), chronic kidney disease, or previous cerebrovascular events between the 2 groups. Pre-TAVR coronary angiographies revealed that 94% of patients who underwent unplanned PCI had CAD at baseline, compared with 59% of patients who did not undergo unplanned PCI (p<0.001). CAD was more likely to be multivessel disease in patients with unplanned PCI than in those without unplanned PCI (p=0.025). Patients with unplanned PCI were more likely to have had previous myocardial infarction (30% vs 14%; p<0.001) and a history of PCI (52% vs 26%; p<0.001).

Table 1. Baseline characteristics.

Details of TAVR procedures are shown in Table 1. TAVR was performed via transfemoral access in 90% of patients, without a significant difference between groups. There was no significant difference in the type of THV used between groups.

Predictors of unplanned PCI

In a multivariable model, the number of diseased vessels, male sex, and younger age were independently associated with an increased risk of unplanned PCI after TAVR, while traditional risk factors, previous peripheral artery disease, previous myocardial infarction and a history of PCI were not (Table 2, Supplementary Figure 1, Supplementary Figure 2). The cumulative incidence curves stratified by the presence of CAD and number of diseased vessels prior to TAVR are shown in the Central illustration. The cumulative incidence rates of unplanned PCI at 1, 5, and 10 years were 0.1%, 0.4%, and 0.6%, respectively, in patients with no CAD. In patients with single-vessel disease, the unplanned PCI rates were 0.7%, 2.5%, and 3.4%, respectively. In patients with multivessel disease, unplanned PCI was performed in 1.5%, 5.4%, and 7.4% at 1, 5 and 10 years, respectively. The effect of CAD was largely consistent after excluding patients with a history of coronary artery bypass grafting (Supplementary Figure 1).

Table 2. Predictors of unplanned PCI under competing risk with death.

Central Illustration. Long-term risk of unplanned PCI after TAVR.

Cumulative incidence curve of unplanned PCI after TAVR stratified by baseline CAD. The red line denotes multivessel disease, the green line denotes single-vessel disease, and the blue line denotes no CAD. Clinical indications and types of lesions of unplanned PCI after TAVR. ACS: acute coronary syndrome; CAD: coronary artery disease; CCS: chronic coronary syndrome; NSTEMI: non-ST-elevation myocardial infarction; PCI: percutaneous coronary intervention; STEMI: ST-elevation myocardial infarction; TAVR: transcatheter aortic valve replacement; UAP: unstable angina pectoris; VD: vessel disease

Clinical indications and procedural characteristics of unplanned PCI

The clinical indications and procedural characteristics of unplanned PCI are shown in Table 3. More than half of the patients underwent unplanned PCI for treatment of ACS (56%). Non-ST-elevation myocardial infarction (NSTEMI) was the most frequent indication (30%), followed by ST-elevation myocardial infarction (STEMI; 16%), and unstable angina pectoris (UAP; 10%) (Central illustration). The most frequent target vessels were, in descending order, the left anterior descending artery (LAD; 51%), followed by the right coronary artery (RCA; 30%), the left circumflex artery (LCx; 18%), the left main (LM; 16%), and coronary artery bypass grafts (9%). Sixty-seven percent of lesions were de novo lesions, 33% were restenotic lesions, and 3% of lesions occurred in the setting of stent thromboses (Central illustration). Femoral access was preferentially used over radial access, and PCI of the target vessel was successful in 99% of cases.

Table 3. Clinical indication for unplanned PCI.

Supplementary data

To read the full content of this article, please download the PDF.