https://orcid.org/0000-0001-5577-081X
-
PDF
- Split View
-
Views
-
Cite
Cite
Rocco Vergallo, Daniela Pedicino, Weekly Journal Scan: Should we treat obstructive coronary artery disease in patients undergoing transcatheter aortic valve implantation?, European Heart Journal, Volume 46, Issue 6, 7 February 2025, Pages 591–593, https://doi.org/10.1093/eurheartj/ehae761
Close - Share Icon Share
This comment refers to ‘PCI in Patients Undergoing Transcatheter Aortic-Valve Implantation’, published in the New England Journal of Medicine, https://doi.org/10.1056/NEJMoa2401513.
The Nordic Aortic Valve Intervention (NOTION)-3 trial was an investigator-initiated, industry-sponsored, international, open-label, randomized controlled trial (RCT) that enrolled patients with concomitant obstructive coronary artery disease (CAD) and symptomatic severe aortic stenosis, with the aim of testing the superiority of a revascularization strategy with percutaneous coronary intervention (PCI) in addition to transcatheter aortic valve implantation (TAVI) over a conservative strategy of TAVI alone with respect to major adverse cardiac events (MACE).1
Eligibility criteria were: (i) a diagnosis of symptomatic severe aortic stenosis; (ii) a formal indication of TAVI by the local heart team; and (iii) angiographic evidence of at least one stenosis ≥90% by visual estimation or at least one stenosis ≥50% but <90% with a positive invasive physiological test [fractional flow reserve (FFR) ≤ 0.80] in a coronary artery of at least 2.5 mm in diameter. TAVI was performed according to local practice, and the choice of transcatheter heart valve was left to the operator’s discretion based on aortic valve size and anatomy. Patients with an indication of a ‘valve-in-valve’ procedure were excluded. Additional exclusion criteria were life expectancy <1 year, severe renal failure [estimated glomerular filtration rate (eGFR) < 20 mL/min/1.73 m2], acute coronary syndrome (ACS) within 14 days before the heart team decision, and left main coronary artery stenosis. Patients randomized to PCI were intended to receive complete revascularization of all eligible lesions. The protocol strongly advised a staged PCI procedure prior to TAVI, but a PCI performed concomitantly with TAVI or within 2 days following the procedure was allowed.
The primary endpoint was a composite of all-cause death, non-fatal myocardial infarction (MI), or urgent revascularization (defined as any revascularization during an unplanned hospitalization for ACS). Secondary endpoints included the individual components of the primary endpoint; cardiovascular death; periprocedural or spontaneous MI; any revascularization; stroke; bleeding events (defined according to the Valve Academic Research Consortium-2); and acute kidney failure.
From September 2017 to October 2022, a total of 455 patients [median age: 82 years, interquartile range (IQR) 78–85; 33% female] were enrolled at 12 hospitals in Denmark, Sweden, Finland, and Latvia and were randomized to either PCI (n = 227) or conservative treatment (n = 228). The median Society of Thoracic Surgeons-Procedural Risk of Mortality (STS-PROM) score was 3% (IQR, 2% to 4%), and the median Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) score was 9 (IQR, 6–14). The median time from randomization to TAVI was 34 days (IQR, 7–62) in the PCI group and 25 days (IQR, 2–54) in the conservative treatment group. A complete revascularization was obtained in 89% of patients in the PCI group. PCI was performed at the time of or shortly after TAVI in 26% of patients. A balloon-expandable heart valve was used in 41% of cases.
After a median follow-up of 2 years (IQR, 1–4), the incidence of the primary endpoint was marginally significantly lower in the PCI group than in the conservative treatment group [26% vs. 36%, respectively; hazard ratio (HR), 0.71; 95% confidence interval (CI), 0.51–0.99; P = .04]. All-cause death occurred in 23% of patients in the PCI group and in 27% of those in the conservative treatment group (HR 0.85; 95% CI 0.59–1.23). Death was attributed to cardiovascular causes in 9% of patients in the PCI group and in 13% of those in the conservative treatment group (HR 0.67; 95% CI 0.38–1.19). Bleeding events occurred more frequently in the PCI group than in the conservative treatment group (28% vs. 20%, respectively; HR, 1.51; 95% CI, 1.03–2.22). The incidence of acute kidney failure was lower in the PCI group than in the conservative treatment group (5% vs. 11%; HR, 0.45; 95% CI, 0.23–0.89). In the PCI group, seven patients (3%) had a complication related to the PCI procedure.
Comment
CAD is a frequent finding in patients with severe aortic stenosis, being detected in approximately half of patients undergoing TAVI. The clinical relevance of CAD is substantial in TAVI recipients, as it may impact both the procedural risk and the long-term prognosis of these patients, and their concomitant management may become even more clinically relevant with the extension of TAVI indication to younger and lower-risk patients.2 A meta-analysis of observational studies including 5580 patients showed no benefit of routine PCI before TAVI on 30-day MACE or 1-year mortality.3 The randomized PercutAneous Coronary inTervention prIor to transcatheter aortic VAlve implantatION (ACTIVATION) trial failed to show non-inferiority of a PCI strategy vs. a conservative strategy in 235 patients with significant CAD and severe aortic stenosis undergoing TAVI and was interrupted early for futility.4 Yet, the results of this trial should be interpreted in the light of some fundamental methodological limitations, including the inadequate statistical power, the choice of a non-inferiority design, the short follow-up, and the predominant inclusion of asymptomatic patients with single-vessel CAD. Based on the limited available data, current European and US guidelines suggest considering PCI of significant stenoses in proximal coronary segments in patients undergoing TAVI, recommending an individualized approach with combined or staged procedures depending on clinical findings, CAD pattern, and extent of myocardium at risk.5,6
The NOTION-3 trial is the first RCT to show that PCI of significant coronary artery stenoses in TAVI recipients is marginally significantly superior to a conservative treatment strategy in reducing MACE.1 However, it should be emphasized that the true effect size of this benefit is statistically uncertain, as reflected by the wide 95% CI of MACE reduction with an upper limit close to the null effect. This difference was primarily driven by a reduction in the rate of urgent revascularizations (which included revascularization in patients admitted for unstable angina), within the limitations of an open-label design. Interestingly, the rate of any revascularization was 21% in the conservative treatment group as compared with 3% in the PCI group.
An important methodological feature of the NOTION-3 trial was the use of invasive physiological assessment (i.e. FFR) to identify target lesions when angiographic coronary artery stenosis was less than 90%, differently from the ACTIVATION trial which relied on angiographic assessment only.4 The left ventricular pressure overload caused by aortic stenosis, leading to ventricular remodelling and hypertrophy as well as to increased intracavity pressure, is known to determine a variety of changes in the anatomy and physiology of coronary arteries and microcirculation (e.g. decreased microvasculature density, reduction in resting microvascular resistance, and blunted hyperaemic response to adenosine) which may impact invasive physiological assessment, leading to an underestimation of the functional severity of a stenosis measured by FFR.7,8 Despite the promising results of NOTION-3, it remains to be determined whether CAD evaluation in patients undergoing TAVI should be based on invasive physiological measures or on angiographic assessment only (as suggested by current guidelines5,6 and position papers2).
Another important point to consider when interpreting the results of the NOTION-3 is that the CAD complexity of the trial population was relatively low, as demonstrated by the low median number of physiologically significant lesions (i.e. one per patient), the small proportion of patients with multivessel CAD (i.e. 20%), and the low median SYNTAX score (i.e. 9). These findings, along with the exclusion of patients with left main CAD, may limit the generalizability of the trial results.
The benefit of MACE observed in the PCI group was achieved at the expense of a 50% higher risk of bleeding compared to the conservative treatment group. Antithrombotic therapy following TAVI is a rapidly evolving field, and the most recent evidence suggests that single antiplatelet therapy with aspirin (or oral anticoagulant therapy in patients with an indication for anticoagulation) is associated with better safety than dual antiplatelet therapy (DAPT).2,9 When performing PCI in patients undergoing TAVI, the need for a variable period of DAPT (or triple antiplatelet therapy in patients with an indication for anticoagulation) after stenting needs to be considered carefully when giving indication for and deciding the timing of the two procedures, as TAVI patients often have multiple bleeding risk factors, including age >75 years.
The design of the NOTION-3 trial did not allow for addressing the issue of the timing of PCI with respect to TAVI, as almost three-fourth of patients received PCI prior to TAVI, while the remaining received it during or a few days after the procedure. PCI timing in patients undergoing TAVI should consider patient characteristics (including bleeding and ischaemic risk, the presence and severity of angina), the burden and severity of CAD, and the type of transcatheter heart valve.2 Performing PCI after TAVI may introduce the potential issue of challenging coronary artery re-access, particularly when self-expanding valves with supra-annular leaflet position are implanted, whereas potential advantages are a more reliable physiological assessment of angiographically intermediate lesions and a lower risk of haemodynamic instability during complex PCI. These considerations are very important with a view to a lifetime management of patients with aortic stenosis, particularly when TAVI is performed in younger patients, who will have longer life expectancy and higher chance of future coronary revascularization procedures.10
In conclusion, the NOTION-3 trial provides some evidence in support of a clinical benefit of PCI in patients with significant CAD undergoing TAVI. The results of other larger ongoing RCTs, including the TAVI PCI (NCT04310046) and the COMPLETE TAVR (NCT04634240) trials, promise to provide insights into several remaining open issues.
Declarations
Disclosure of Interest
R.V. received consulting or lecturing fees from Abbott Vascular, Abiomed, Amgen, Daiichi Sankyo, Edwards Lifesciences, Medtronic, Philips, and Novartis, outside the submitted work. D.P. received lecturing fees from Daiichi Sankyo, outside the submitted work.
