In this randomized, open-label, superiority, multicenter, international study, researchers examined the impact of coronary revascularization through percutaneous coronary intervention (PCI) on patients with severe aortic stenosis (AS) eligible for transcatheter aortic valve implantation (TAVI) and concomitant stable coronary artery disease. Although the study was sponsored by Boston Scientific®, the company did not supply any devices and did not participate in any aspect of the study’s design, development, or publication.
Patients with severe AS selected for TAVI by the local Heart Team and presenting at least one coronary lesion in a vessel >2.5 mm, with either an angiographic stenosis >90% or an FFR <0.80, were included. Exclusion criteria comprised life expectancy <1 year, acute coronary syndrome (ACS) within 14 days prior to randomization, severe chronic renal disease with a glomerular filtration rate <20 ml/min/m², or left main coronary artery disease. Chronic occlusions did not exclude the patient, and revascularization was left to the operator’s discretion. All patients were randomized in a 1:1 ratio to receive PCI for lesions meeting inclusion criteria or to conservative treatment. Although the protocol strongly recommended revascularization prior to TAVI as sequential procedures, concurrent or delayed PCI up to 2 days post-implantation was permitted. Antiplatelet therapy and, when indicated, anticoagulation were administered according to predefined guidelines, with adaptations following the release of the AUGUSTUS study in 2019 and the POPular Trial in 2020, which influenced the antiplatelet strategy as the study progressed.
The primary endpoint was a composite event of death from any cause, myocardial infarction, or urgent revascularization, evaluated until the last included patient had been followed for at least 1 year post-TAVI. Secondary events included the individual components of the primary endpoint, as well as cardiovascular mortality, perioperative or spontaneous myocardial infarction, any need for repeat revascularization, stent thrombosis, hospitalization due to heart failure, stroke, bleeding assessed per VARC-2 criteria, acute kidney injury, and NYHA and CCS functional class at 1 and 12 months post-TAVI. Safety events were defined as any type of bleeding.
Between September 2017 and October 2022, 455 patients were enrolled across 12 hospitals (PCI group: n = 227; conservative group: n = 228), with two-thirds enrolled between 2020 and 2022. The groups were balanced in baseline characteristics, except that a higher proportion of patients in the PCI group were smokers and/or had obstructive pulmonary disease. Most patients were male (67%) and elderly (75% were over 78 years old), with median STS-PROM and SYNTAX scores of 3% (2-4%) and 9 points (6-14 points), respectively. In 12 patients, TAVI was not performed, and PCI was not completed in 8 assigned to the PCI group. Additionally, 11% of patients did not undergo complete revascularization of all lesions initially considered for PCI. In 26% of cases, PCI was performed during or after TAVI.
After a median follow-up of 2 years (1-4 years), the primary endpoint was observed in 25% of the PCI group and 35% of the conservative group (HR = 0.69; 95% CI = 0.49 – 0.97), considering only those patients who ultimately underwent TAVI (modified intention-to-treat analysis). Among secondary events, myocardial infarction and urgent revascularization were more frequent in the conservative group, while incidences of all-cause or cardiovascular death and stroke were similar. Any bleeding occurred more frequently in the PCI group (28% vs. 20%; HR = 1.51; 95% CI = 1.03-2.22), and renal failure was significantly higher in the conservative group.
In the discussion, the authors highlight differences between this study and the ACTIVATION trial, which also investigated coronary revascularization in TAVI patients. NOTION 3 included a larger and more representative population, with differences in the timing of enrollment, age, and surgical risk according to STS-PROM criteria. Additionally, a higher percentage of patients in NOTION 3 presented angina, and revascularization criteria were stricter (>90% stenosis/FFR <0.8 in NOTION 3 vs. >70% lesions in ACTIVATION).
Additionally, note that the included patients had a low SYNTAX score and little representation of patients with multivessel disease. Among the most important limitations are the exclusion of patients with ACS in the previous 14 days and of patients with left main disease, so the results cannot be extrapolated to these populations, and the long inclusion period, which among others. These things led to changes in the criteria and duration of combined anticoagulation plus antiplatelet or double antiplatelet treatments. The fact that two-thirds of patients were included in the last two years of the inclusion period has probably neutralized part of these limitations.
The conclusion of the study is that performing PCI in lesions > 90% or with FFR < 0.8 in TAVI candidate patients with coronary artery disease reduces the incidence of the combined event, subsequent revascularization and the risk of new myocardial infarction, although they advise that the decision should be individualized and must take into account the patient’s health status and comorbidities, life expectancy, complexity and severity of coronary heart disease, and patients’ risk of bleeding.
COMMENTARY:
This is an important study contributing evidence to the ongoing debate on whether revascularization in TAVI candidates with coronary artery disease results in clinical benefits beyond treating the aortic valve alone. Significant benefits were observed in the revascularized group after only a median follow-up of two years, which was expected given the significance of the lesions targeted for treatment (arteries >2.5 mm with >90% stenosis or deemed significant by functional testing). Results might have been even more striking if the rate of incomplete revascularization in the PCI group had been lower (11% despite low SYNTAX scores). It is likely that the Heart Team’s requirement to accept patients for TAVI prior to inclusion excluded those with more complex coronary disease unsuitable for PCI but eligible for valve replacement and surgical myocardial revascularization. It is worth noting that surgery would likely have achieved higher complete revascularization rates, as evidence shows that the extent of revascularization by coronary artery bypass grafting is not influenced by SYNTAX scores, even when highly elevated. Beyond the limitations identified by the authors, the 11% incomplete revascularization rate in the study group warrants an “as-treated” analysis rather than solely an “intention-to-treat” analysis.
Numerous important retrospective studies have reported reduced adverse events following surgical aortic valve replacement and coronary bypass grafting at both short- and long-term compared to non-revascularization, using thresholds of >50% and >70% stenosis for revascularization. This concept, widely endorsed by surgeons, likely explains the notable differences in associated procedures in comparative TAVI vs. surgery studies across various risk strata, with significantly higher myocardial revascularization rates in the surgical groups almost systematically. While the impact on initial risk and mortality in surgical groups is partially justified by this higher rate of associated procedures, the neutralization of benefits at mid-term likely reflects the higher degree of concomitant coronary disease treatment. Of particular concern, in the PARTNER 3 study, the component of the composite primary event that most favored TAVI was rehospitalizations. Although myocardial infarction is included among safety and efficacy events, the study protocol specifically restricts indexable rehospitalizations to those due to “aortic stenosis symptoms and/or valve implantation complications,” later defined as “hospitalizations related to the procedure, valve, or heart failure.” Finally, these are further divided into: 1) valve-related rehospitalizations, including symptoms due to acute, subacute, or late prosthetic dysfunctions like thrombosis, endocarditis, valve degeneration, prosthesis-patient mismatch, delayed coronary obstruction, coronary embolism, heart failure due to the aortic valve, or hemorrhagic complications; and 2) procedure-related rehospitalizations, including bleeding, vascular complications, stroke/TIA, arrhythmias, and acute kidney failure. The inclusion of acute coronary syndromes of non-embolic or other causes unrelated to delayed coronary occlusion as indexable rehospitalization causes within the primary event is unclear, and if not attributed to the procedure, might downplay the magnitude of this composite event.
The study results strongly suggest that patients with significant lesions in major vessels unlikely to achieve complete percutaneous revascularization should be considered surgical candidates. The study’s follow-up is still short (median of two years), and given the findings in patients with an average age of 74 years and only 25% under 78 years, expanding TAVI without revascularization to younger patients could prove highly detrimental, not only due to the increase in coronary events but also due to the challenges that may arise for revascularization in these patients with percutaneously difficult-to-access ostia and complex ascending aorta management if surgical revascularization is eventually required. It is equally crucial to conduct randomized studies with sufficient patient populations to compare outcomes of TAVI + PCI versus concomitant surgical aortic valve replacement and coronary bypass grafting in this subset of patients with aortic stenosis and ischemic heart disease.
REFERENCES:
- Lønborg J, Jabbari R, Sabbah M, Veien KT, Niemelä M, Freeman P, et al.; NOTION-3 Study Group. PCI in Patients Undergoing Transcatheter Aortic-Valve Implantation. N Engl J Med. 2024 Aug 31. doi: 10.1056/NEJMoa2401513.
- Lund O, Nielsen TT, Pilegaard HK, Magnussen K, Knudsen MA. The influence of coronary artery disease and bypass grafting on early and late survival after valve replacement for aortic stenosis. J Thorac Cardiovasc Surg. 1990 Sep;100(3):327-37.
- Thalji NM, Suri RM, Daly RC, Greason KL, Dearani JA, Stulak JM, et al. The prognostic impact of concomitant coronary artery bypass grafting during aortic valve surgery: implications for revascularization in the transcatheter era. J Thorac Cardiovasc Surg. 2015 Feb;149(2):451-60. doi: 10.1016/j.jtcvs.2014.08.073.
