Transcatheter Aortic Valve Replacement After Heart Failure Hospitalization: Too Little, too Late?

This article refers to ‘Impact of recent heart failure hospitalization on clinical outcomes in patients with severe aortic stenosis undergoing transcatheter aortic valve replacement: an analysis from the PARTNER 2 trial and registries’ by S. Chen et al., published in this issue on pages 1866–1874. Among adults with severe aortic stenosis (AS) who are scheduled to undergo surgical (SAVR) or transcatheter aortic valve replacement (TAVR), a substantial proportion is hospitalized with heart failure (HF) in the weeks preceding the procedure.1, 2 In Ontario, Canada, 12% of patients referred for TAVR were hospitalized for HF within a median of 80 days wait time.1 Similarly, in the FinnValve registry, 11% of patients waiting for TAVR or SAVR developed acute HF within 60 days prior to the procedure.2 Patients with AS who develop HF have suboptimal outcomes after TAVR or SAVR compared to their non-HF counterparts.3-5 Acute HF at presentation in TAVR or SAVR candidates is associated with even worse prognosis despite correction of AS.2, 6, 7 In the FinnValve registry, recent acute HF was associated with lower 30-day and 5-year survival, more post-procedural complications (including acute renal failure), and higher rates of healthcare resource utilization in both the SAVR and TAVR groups.2 Similarly, in a retrospective cohort from France, TAVR recipients presenting with acute HF had more post-procedural acute renal failure and requirement for inotropic support, and significantly higher mortality both at 30 days (10% vs. 5%, P = 0.02) and at 1 year (27% vs. 15%, P < 0.001) compared to stable patients.7 Importantly, acute HF was widely prevalent (>50%) in patients with AS on the TAVR waitlist who were untreated.7 This finding is consistent with contemporary reports of high rates of hospitalization for HF among untreated patients with AS.8 In this issue of the Journal, Chen and colleagues provide additional insights into the impact of acute HF on the clinical outcomes of intermediate-to-high or prohibitive surgical risk patients undergoing TAVR for severe symptomatic AS.9 The investigators stratified patients from the PARTNER 2 studies, including the PARTNER 2A and 2B randomized cohorts and the PARTNER 2 S3 high-risk and intermediate-risk registries, according to recent (i.e. within 6 months prior to TAVR) hospitalization for HF. Patients had severe AS by echocardiography [i.e. mean gradient >40 mmHg or jet velocity >4.0 m/s and an aortic valve area (AVA) of <0.8 cm2 or AVA index of <0.5 cm2/m2] and New York Heart Association functional class ≥II symptoms. Of 3988 patients, 1622 (40.7%) were hospitalized for HF within the prior 6 months — an alarming finding.9 In the propensity-matched subgroups, 2-year all-cause mortality was 25% higher in patients with recent acute HF [27.3% vs. 22.3%; hazard ratio (HR) 1.25; 95% confidence interval (CI) 1.07–1.47; P = 0.005], driven by cardiovascular mortality (19.3% vs. 14.5%, HR 1.37; 95% CI 1.13–1.67; P = 0.002). Importantly, patients with recent acute HF had also higher rates of infective endocarditis at 2 years vs. their matched counterparts (3.1% vs. 1.1%; HR 2.69, 95% CI 1.42–5.09, P = 0.002).9 Several findings in this study merit further discussion. First, the study by Chen and colleagues emphatically reminds us of the risk of de novo HF or decompensation of pre-existing HF as long as severe AS remains untreated. Although this is not a ‘waitlist’ study, the 40% rate of HF hospitalization within the preceding 6 months is still an alarming finding from a public health and healthcare system perspective. On the positive side, in a cost-effectiveness analysis of more than 15 000 Medicare beneficiaries in the United States, patients with severe left ventricular (LV) dysfunction at baseline derived the most benefit in terms of all-cause and HF hospitalizations with TAVR.10 Second, from a disease biology perspective, this study brings to the forefront hospitalization as a turning point in the course of HF regardless of aetiology. Hospitalization for HF is associated with worse prognosis among patients with HF and is accompanied by myocardial damage, worsening disease biology, and a period of increased event rates.11 In patients with severe AS, development of HF has been traditionally considered as a marker of dismal prognosis, necessitating prompt intervention. However, not all structural damage is reversible with late intervention. The pre-procedural echocardiographic characteristics of patients with recent HF hospitalization in the study by Chen and colleagues indicated more advanced structural damage.9 The extent to which immediate LV afterload reduction with TAVR ameliorates LV remodelling remains unclear. In Chen et al.,9 the worse echocardiographic characteristics of patients with recent HF persisted in the 30-day post-TAVR echocardiogram. Of note, in a previous study with longer-term echocardiographic follow-up, impaired LV systolic function improved but did not normalize at 1 year post-TAVR,3 suggesting that severe baseline structural disease limits the degree of LV reverse remodelling after TAVR. Third, Chen and colleagues report a more than 2.5-fold increase in the 2-year incidence of post-TAVR infective endocarditis (IE) in the group with recent HF hospitalization.9 History of HF has been associated with post-TAVR IE.12 To complicate things, recent HF hospitalization was also associated with prolonged length of stay during the index hospitalization for TAVR in Chen et al.,9 as well as with higher rates of readmission post-TAVR. Therefore, it is unclear which of these extra days in the hospital contributed the most to the higher incidence of IE in this group. Of note, while no significant difference in access route was noted in patients with vs. without recent HF, patients with recent HF hospitalization were more likely to receive a second-generation SAPIEN XT transcatheter heart valve (THV) vs. the more recent SAPIEN 3 THV (Edwards Lifesciences, Irvine, CA, USA). The latter has a lower profile and improved design, offering multiple clinical benefits, including lower rates of paravalvular leak, vascular complications, and bleeding.13 Of note, paravalvular leak and vascular complications have been identified as independent predictors of IE after TAVR.14 However, in the PARTNER 2 data, rates of post-TAVR paravalvular leak and vascular complications were not more frequent in patients with recent HF hospitalization, implying that prolonged and multiple hospitalizations in the recent HF group was probably the main culprit for the higher incidence of HF in these patients.9 Taken together, these considerations lead to the ultimate question: would earlier intervention improve outcomes for these patients? Currently enrolling randomized clinical trials aiming to assess the benefit of earlier intervention include the Transcatheter Aortic Valve Replacement to Unload the Left Ventricle in Patients with Advanced Heart Failure (TAVR UNLOAD)15 and the Evaluation of Transcatheter Aortic Valve Replacement Compared to Surveillance for Patients with Asymptomatic Severe Aortic Stenosis (EARLY TAVR) studies.16 TAVR UNLOAD randomizes adults with moderate AS and evidence of HF (hospitalization for HF within 2 years or N-terminal pro-B-type natriuretic peptide >900 pg/mL or B-type natriuretic peptide >200 pg/mL) to transfemoral TAVR with a SAPIEN 3 THV or optimal medical management.15 The primary endpoint is the hierarchical occurrence of all-cause death, disabling stroke, HF hospitalization, symptomatic aortic valve disease or non-disabling stroke, and the change in the Kansas City Cardiomyopathy Questionnaire at 1 year. In EARLY TAVR, adults with severe asymptomatic AS are randomized to receive either a SAPIEN 3 (or SAPIEN 3 Ultra) THV or clinical surveillance.16 The primary outcome is freedom from a composite endpoint of all-cause death, stroke, and unplanned cardiovascular hospitalization for 2 years. Ultimately, we hope the optimal timing of aortic valve intervention in patients with severe AS will soon be answered by these studies. In the meantime, the data by Chen et al. provide further support to prioritizing TAVR for patients with ongoing or recent HF symptoms. This is even more important in constrained healthcare systems where TAVR is not incentivized and access to this procedure may be limited.17 In this light, we believe that the recent position statement from the Canadian Cardiovascular Society appropriately considers inpatient HF status or recent hospitalization with ongoing New York Heart Association III–IV HF symptoms an indication for urgent wait listing, with a corresponding goal of performing TAVR within 2 weeks.18 Nevertheless, it is important to keep in mind that underlying HF in patients with severe AS may also be attributed to or aggravated by other aetiologies, including concomitant coronary artery disease, other valvular heart disease (especially mitral regurgitation), or e.g. amyloid-related disease (especially in the elderly), conditions that would not be resolved by TAVR. In conclusion, Chen et al.9 provide valuable insights into the prognostic and healthcare utilization impact of acute HF in pre-TAVR patients. These data, coming from a large, well-characterized cohort of patients, suggest that action should be potentially prioritized for this high-risk group of TAVR candidates. Conflict of interest: P.B.P. serves as a consultant for Medtronic, Inc and has served on the Scientific Advisory Board for AstraZeneca Pharmaceuticals. G.T. has nothing to disclose. A.P.K. has received research support from the National Heart, Lung, and Blood Institute, American Heart Association, and Critical Diagnostics.