Heart Failure and Moderate Aortic Stenosis: is Early TAVI the Better Strategy?

Abstract

Aortic stenosis (AS) is the most common valvular heart disease in Western countries. Current clinical guidelines recommend intervention exclusively for severe AS, given its association with poor short-term prognosis. However, accumulating evidence over recent years suggests that even moderate AS may be associated with adverse outcomes, particularly in patients with heart failure with reduced ejection fraction (HFrEF). Consequently, an early aortic valve replacement (AVR) strategy could potentially improve survival and reduce heart-failure–related hospitalizations. Moreover, treating moderate AS in patients with HFrEF may reduce left ventricular afterload and improve haemodynamic status, thereby facilitating the initiation and/or up-titration of guideline-directed medical therapy (GDMT), especially when such therapy is poorly tolerated due to hypotension or other hypoperfusion-related factors. Nevertheless, the available evidence is derived predominantly from retrospective observational studies. The only published randomized controlled trial to date (the TAVR UNLOAD trial) failed to demonstrate a prognostic benefit of transcatheter aortic valve implantation (TAVI) in patients with moderate AS and HFrEF or heart failure with mildly reduced ejection fraction (HFmrEF). Two additional ongoing randomized trials may ultimately provide more definitive evidence. In summary, although the pathophysiological rationale is sound and the available data are encouraging, an early AVR strategy in moderate AS cannot currently be regarded as evidence-based.

Epidemiology

Aortic stenosis (AS), although frequently underdiagnosed, is the most common valvular heart disease in Western countries, affecting approximately 5% of the population aged over 65 years. Within the EURObservational Research Programme (EORP) Valvular Heart Disease II (VHD II) survey, which included patients with severe native valvular disease, AS emerged as the most prevalent valvular condition, accounting for 41.2% of cases.¹

With regard to heart failure (HF), a large European registry reported a prevalence of moderate or severe AS of 3% among patients with heart failure with reduced ejection fraction (HFrEF) and heart failure with mildly reduced ejection fraction (HFmrEF), and up to 8% among those with heart failure with preserved ejection fraction (HFpEF).² Similarly, an analysis based on data from the ESC Heart Failure Long-Term Registry found an overall prevalence of moderate or severe AS of 4.3%, with rates of 4% in HFrEF, 6% in HFmrEF, and 6% in HFpEF.³

Natural history and prognosis

Historically, the scientific literature has focused almost exclusively on severe AS, which has been considered the only stage of the disease warranting invasive treatment (surgical or transcatheter) because of its poor short-term prognosis. In this context, moderate AS has traditionally been regarded as a benign, stable, or slowly progressive condition, requiring an active surveillance strategy aimed at early detection of disease progression. This approach is also endorsed by the most recent clinical guidelines, with the sole exception of a class IIa recommendation for surgical aortic valve replacement in patients with moderate AS who have a concomitant indication for other cardiac surgery.⁴

However, a growing body of recent evidence has demonstrated a substantial prognostic impact of moderate AS as well. In a recent meta-analysis, Jacquemyn et al. reported a marked reduction in 15-year survival among patients with moderate AS (23.3%) compared with those without AS or with mild AS (58.9) (hazard ratio [HR] 2.55 [95% CI, 2.46–2.64]; P < 0.001), corresponding to an approximately 2.5-fold increase risk of all-cause mortality. This difference was also confirmed in a subgroup analysis of patients with HFrEF, in whom moderate AS was associated with reduced 10-year survival (15.5% vs. 37.3%) (HR 1.83 [95% CI, 1.69–2.0]; P < 0.001). Overall, moderate AS was associated with a loss of 4.4 life-years in the general population and 1.9 life-years among patients with HFrEF.⁵

In an observational study conducted in patients with HFrEF, moderate AS was associated with all-cause mortality rates similar to those observed in severe AS over a follow-up of approximately three years (42.2% vs. 45.9%, P = 0.37), and higher than those observed in patients without AS (25.8%, P < 0.0001). Furthermore, hospitalizations for heart failure were more frequent in patients with moderate AS (43.6%) than in those without AS (30.2%, P < 0.0001) and those with severe AS (36.2%, P < 0.05), likely because aortic valve replacement in the latter group reduced HF-related hospitalizations.⁶

Several mechanisms may account for these findings. First, moderate AS is characterized by continuous haemodynamic progression, with an average annual increase in peak transvalvular velocity of 0.19 m/s, an increase in mean gradient of 4.10 mmHg, and a reduction in aortic valve area of 0.05–0.08 cm² per year.⁷ Moreover, a substantial proportion of patients exhibit accelerated progression, defined by an annual increase in peak velocity ≥ 0.3 m/s. Thus, the observed increase in mortality may be partly related to progression towards more severe disease.

Disease progression may also translate into progressive cardiac damage, with a consequent adverse impact on clinical outcomes, a phenomenon that is particularly evident in patients with HFrEF. In this population, the presence of AS—even at an early stage of its natural history—leads to increased left ventricular afterload, which in turn contributes to deterioration of left ventricular systolic function, as reflected by a reduction in ejection fraction.⁸ Nevertheless, left ventricular ejection fraction itself does not appear to modify the impact of moderate AS on survival. In other words, the detrimental effect of moderate AS is not confined solely to patients with HFrEF but also applies to individuals who have not yet developed left ventricular systolic dysfunction.⁹

In a study by Amanullah et al., evidence of cardiac damage was identified in patients with moderate AS and both reduced and preserved ejection fraction. Only 13.1% of patients showed no evidence of extra-valvular cardiac involvement (Group 0). Specifically, 26.8% had left ventricular involvement (Group 1: hypertrophy, systolic dysfunction, increased filling pressures); 42.6% exhibited left atrial dilatation, at least moderate mitral regurgitation, and/or atrial fibrillation (Group 2); 10.6% had pulmonary hypertension (systolic pulmonary artery pressure ≥ 60 mmHg) and/or at least moderate tricuspid regurgitation (Group 3); and 6.9% had developed right ventricular dysfunction (Group 4). Patients with associated cardiac damage more frequently had comorbidities such as arterial hypertension, coronary artery disease, and diabetes mellitus, which may have contributed to cardiac remodelling. Importantly, mortality increased with the extent of extra-valvular cardiac damage, with a 1.5-fold higher risk in Group 1 compared with Group 0, and up to a 2.3-fold higher risk in Groups 3 and 4.¹⁰ Similar findings have been reported in a more recent study by Harmon et al.¹¹

Although evidence regarding AS in patients with preserved ejection fraction remains limited, the relationship between AS and diastolic dysfunction is pathophysiologically close. Compensatory mechanisms in response to chronic pressure overload in AS are known to develop very early in the natural history of the disease, with the aim of reducing left ventricular wall stress and preserving systolic function. However, the resulting concentric remodelling ultimately leads to diastolic dysfunction and myocardial fibrosis, which are known to be associated with adverse prognosis in severe AS.¹² Patients with HFpEF have been shown to have smaller aortic valve areas and higher transvalvular gradients than those without diastolic dysfunction.¹² Consequently, progression of aortic valve disease in these patients represents a continuous process that fuels cardiac remodelling mechanisms and significantly impacts mortality. The presence of diastolic dysfunction may therefore, in itself, contribute to improved risk stratification, helping to identify patients with preserved ejection fraction who may genuinely benefit from early treatment of AS.¹²

Rationale for TAVI in moderate aortic stenosis

The growing body of evidence regarding the clinical and prognostic relevance of moderate AS has prompted investigation into the potential benefits of early treatment strategies, namely early aortic valve replacement (AVR). The rationale underlying this approach is to intervene in the natural history of the disease, not once stenosis has become severe, but rather before extensive cardiac damage has developed. From this perspective, a strategy of active surveillance may entail risks, as delaying treatment of AS could increase the likelihood of progression of cardiac damage and lead to poorer outcomes even after AVR.¹¹

Another potentially relevant aspect concerns the role of transcatheter aortic valve implantation (TAVI) as an enabling therapy in patients with HFrEF. Treatment of moderate AS in patients with left ventricular systolic dysfunction improves haemodynamic conditions and may allow the initiation and/or up-titration of guideline-directed medical therapy (GDMT), particularly in cases in which such therapy is poorly tolerated due to hypotension or other factors related to systemic hypoperfusion (e.g. renal dysfunction or malabsorption).²

Current evidence

The impact of treating moderate AS has been evaluated predominantly in retrospective observational studies. In a study by Ludwig et al., TAVI in patients with HFrEF and moderate AS was associated with improved survival and a reduced composite risk of all-cause mortality and heart failure hospitalizations compared with medical therapy alone.¹³ A recent meta-analysis including 2479 patients with HFrEF and moderate AS demonstrated that aortic valve replacement—either transcatheter or surgical—was associated with a significant reduction in both all-cause and cardiovascular mortality compared with an active surveillance strategy. However, the retrospective observational design of the included studies limits the strength of conclusions that can be drawn from these data.¹⁴

More recently, Van Mieghem et al. published the results of the TAVR UNLOAD trial (Transcatheter Aortic Valve Replacement to UNload the Left Ventricle in Patients with Advanced Heart Failure), the first randomized controlled trial to compare an early TAVR strategy (transfemoral TAVI using the SAPIEN 3 device, Edwards Lifesciences) plus optimized medical therapy vs. optimized medical therapy alone in patients with moderate AS and left ventricular systolic dysfunction (left ventricular ejection fraction 20–50%). This study failed to demonstrate the superiority of TAVI over active surveillance in terms of improvement in the primary composite endpoint of major cardiovascular and cerebrovascular events. Conversely, TAVI was associated with an improvement in symptoms, as assessed by the Kansas City Cardiomyopathy Questionnaire (KCCQ), at one year following the procedure.¹⁵ The key results are summarized in Figure 1.

Figure 1.

Results of the TAVR UNLOAD trial. LVEF: left ventricular ejection fraction; GDMT: guideline-directed medical therapy; KCCQ: Kansas City Cardiomyopathy Questionnaire; TAVR UNLOAD: Transcatheter Aortic Valve Replacement to UNload the Left Ventricle in Patients with ADvanced Heart Failure; TAVR: transcatheter aortic valve replacement.

Beyond the trial results themselves, several aspects warrant careful consideration to properly interpret these findings. First, reduced enrolment during the COVID-19 pandemic inevitably compromised the statistical power of the trial. In addition, extending the follow-up period from one year to 23 months may not have allowed early detection of differences in clinical outcomes between the two groups. Most importantly, the high crossover rate (43%) substantially reduced the contrast between the treatment arms, as the number of events observed in the control group was markedly lower than expected, further limiting the ability to demonstrate a benefit of early TAVI.¹¹ This finding also suggests that progression of moderate AS may occur relatively rapidly over time. Nevertheless, the study confirmed the safety of TAVI in patients with moderate AS, reporting complication rates comparable to those described in the existing literature.¹⁵

With respect to patients with moderate AS and HFpEF, two randomized controlled trials are currently ongoing: the PROGRESS trial (Prospective, Randomized, Controlled Trial to Assess the Management of Moderate Aortic Stenosis by Clinical Surveillance or Transcatheter Aortic Valve Replacement; NCT04889872) and the EXPAND II trial (Evolut EXPAND TAVR II Pivotal Trial; NCT05149755). These studies are expected to clarify the potential efficacy of transcatheter intervention in this patient population.

Therefore, a potential role for TAVI may be hypothesized only in carefully selected patients with moderate AS and HFrEF, within a personalized therapeutic approach, particularly in those with recurrent episodes of heart failure and inadequate response to optimal medical therapy. Nevertheless, it is crucial to emphasize that, in the context of early treatment of AS—more than in advanced stages of the disease—close attention must be paid to life-time management, bioprosthetic valve durability, and feasibility of potential redo TAVI.

Conclusions

Contrary to long-standing assumptions, moderate AS has a relevant prognostic impact, particularly in patients with HFrEF and HFmrEF. Evidence supporting the use of TAVI in these patients is promising but largely derived from retrospective observational studies. The only randomized controlled trial published to date, TAVR UNLOAD, failed to demonstrate a prognostic benefit of an early TAVI strategy, but showed an improvement in patients’ quality of life. Further studies are required to more clearly define the benefits, limitations, and challenges associated with early treatment of moderate AS.

Funding

None.

Data availability

No new data were generated or analysed in support of this research.

Disclaimer

This paper was originally published in the Italian language as ‘Scompenso cardiaco e stenosi aortica di grado moderato: meglio la TAVI subito?', in the Volume degli Atti del Congresso “Conoscere e Cuare il Cuore 2026”, published by Centro per la Lotta contro l’Infarto for distribution at the CCC Conference. This paper was translated by Dr. Mario Albertucci, representative of the CLI Foundation, and republished with permission.