The Dilemma of Moderate Aortic Stenosis in Patients With Heart Failure and Reduced Ejection Fraction: Are We Waiting Too Long to Intervene, or Should We Wait for More Evidence?

The cardiac valvular lesion that most frequently requires interventional therapy (via open surgery or transcatheter therapy) is aortic stenosis (AS). The natural course of AS includes, among other maladaptive changes, remodeling of the left ventricle, resulting initially in concentric myocardial hypertrophy, which occurs to overcome the intracavitary pressure increase associated with the fixed aperture created by aortic valve stenosis. The development of concentric hypertrophy takes place over time, and often remains clinically asymptomatic for extended periods, although undetected yet deleterious changes in diastolic function, systolic function, and myocardial oxygen demand almost certainly are occurring. Severe AS is associated with dismal outcomes if the lesion is uncorrected: the presence of angina, syncope, and/or congestive heart failure portends death in 50% of patients in two years or sooner after symptom onset if the AS is not relieved. Current guidelines from the American College of Cardiology/American Heart Association (ACC/AHA), Society of Thoracic Surgeons (STS), and European Society of Cardiologists/European Association of Cardio-Thoracic Surgeons (ESC/EACTS) are to recommend aortic valve replacement to patients who are diagnosed with severe symptomatic or asymptomatic AS; the only recommendation to provide aortic valve replacement (AVR) for treatment of less-than-severe AS is in the setting of other, already planned cardiac surgery (eg, coronary artery bypass grafting or ascending aortic surgery).^{,  ,  ,}  Of course there are shades of gray that color the continuum of aortic stenosis—it is unlikely that AS is only a problem when it is severe, and so, a question can be raised as to whether intervention on AS should be withheld until it is diagnosed as severe. Mortality is increased in patients with moderate AS compared to those without any AS. Some patients, for example, experience a reduction in left ventricular ejection fraction as a result of the aforementioned pathophysiologic changes related to progressive AS. Multiple reports exist describing increased mortality in patients who develop heart failure with reduced ejection fraction (HFrEF) in the setting of moderate AS.^{,  ,  ,}  It is unclear whether it is the aortic valvular obstruction or the reduced systolic function that is contributing to the poor outcomes in this population, however.

Recently, Jean et al. reported their findings in patients with both HFrEF and moderate AS in the Journal of the American College of Cardiology. This multicenter retrospective study included 262 patients diagnosed with both HFrEF (defined as left ventricular ejection fraction [LVEF] <50%) and moderate AS (defined as an aortic valve area between 1.0 and 1.5 cm² and a peak aortic valve jet velocity between 2 and 4 m/s) by echocardiography completed either at rest or after dobutamine stress echocardiography. A group of 262 patients (from three academic medical centers in Canada and Europe), diagnosed with HFrEF without echocardiographic evidence of AS (ie, no finding of aortic valve thickening that reduced leaflet mobility nor peak aortic velocity >2 m/s), was selected and matched against the HFrEF + moderate AS group with the variables of (1) sex, (2) age, (3) estimated glomerular filtration rate, (4) New York Heart Association functional class III or IV, (5) presence of diabetes, (6) LVEF(within 5%), and (7) body mass index. Exclusion criteria included patients diagnosed with hypertrophic cardiomyopathy, noncompaction cardiomyopathy, history of heart transplantation, presence of prior AVR, history of previous aortic surgery, or presence of congenital heart disease. Echocardiographic method used for assessment of LVEF included biplane Simpson method; aortic valve area (AVA) was calculated with the continuity equation indexed to body surface area, and the mean aortic gradient was determined by simplified Bernoulli equation. The study's primary endpoint was all-cause mortality after diagnosis; the secondary endpoints were composite HF hospitalization and all-cause mortality. Every patient had at least one year of follow-up.

The results of this study provided further support for the association between increased all-cause mortality and increased composite HF hospitalization when comparing patients who have moderate AS and HFrEF with patients who have HFrEF without AS. The reported details are striking: with an average follow-up time of 2.9 years (±2.2 years), 35% of patients with moderate AS and HFrEF experienced the primary endpoint of mortality, whereas that endpoint was met in only 23% of patients with HFrEF alone (HR: 2.31; 95% CI: 1.72-3.12; p < 0.0001). Moreover, when multivariate analysis was completed, moderate AS was the single most important factor associated with mortality in those who died (HR: 2.98; 95% CI 2.08-4.31; p < 0.0001). The second endpoint of composite HF hospitalizations and all-cause mortality likewise showed an association with worse outcomes in those patients with both moderate AS and HFrEF. Importantly, this study captured outcomes related to aortic valve replacement that—although not included in the preplanned investigation endpoints—revealed novel information related to the effect of intervention in this patient population. All patients with moderate AS experienced higher mortality compared to those without AS, regardless of intervention. Patients with moderate AS + HFrEF who underwent AVR were found to have reduced mortality compared to those with the condition who did not undergo valve replacement (HR: 0.59; 95% CI: 0.35-0.98; p = 0.04). The key point here is that the survival benefit was noted only in those who underwent transcatheter AVR (TAVR) but not those who underwent surgical AVR (SAVR).

Although causality cannot be implied through the aforementioned retrospective study, it adds to the considerable literature that illustrates the association between moderate AS in patients with HFrEF and increased mortality. Furthermore, this study highlighted the finding that—among several other common comorbidities and risk factors—moderate AS is the single most important variable associated with increased mortality in the moderate AS + HFrEF population. The primary and secondary outcomes reported compel an obvious (and heretofore unanswered) question: is there an intervention that can improve the outcomes in these patients? Without expressing the intention to do so at the start of their study, Jean et al. offered a glimpse of insight on a potential intervention that may help: AVR—specifically the transcatheter option. This study followed all patients for at least 12 months after study completion, but the authors reported data as far out as six years after the study was finished; the choice to follow for that long brought potentially intriguing information. A total of 44 patients (16.7%) of those patients with moderate AS + HFrEF ended up undergoing AVR at some point during the follow-up observation period. As mentioned previously, those who underwent AVR had greatly improved survival compared to those who did not have this intervention, and notably so; AVR was associated with a 40% reduction in mortality. This finding is noteworthy. But, for several reasons, this finding still does not inform us on whether or not to proceed with AVR in this subset of patients. First, the study was retrospective, with a small sample size. The oft-mentioned limitations of retrospective studies (eg, cannot determine cause and effect) are relevant here. A low sample size lowers statistical power, and, although somewhat counterintuitive, low sample size/low statistical power decreases the likelihood that a statistically significant finding represents true effect. Third, the AVR mortality association is at considerable risk for confounding. An unaddressed question in the study was why patients underwent AVR at all. The study participants came from three hospitals—one in Canada, and two in the Netherlands; presumably the relevant society guidelines for aortic valve intervention recommend AVR only for diagnosed severe AS, whether on its own or in conjunction with another indication for cardiac surgery.^,, A plausible reason for AVR to be undertaken might be due to natural progression of the moderate AS to severe AS, but this is not known. Additionally, the decision to perform AVR via either transcatheter route or as an open surgical procedure also was not delineated and was at the discretion of the treating physician; Jean et al. reported that only 34% (15/44) of those receiving AVR underwent TAVR. Perhaps this is because the remainder of the patients (who underwent SAVR) required additional cardiac interventions (eg, coronary artery bypass grafting or other valve surgery) along with the AVR. The Jean et al. study reported that only those undergoing TAVR saw survival benefit in the moderate AS + HFrEF cohort, but combined cardiac surgical procedures often carry an increased risk of complications and mortality—perhaps this was the reason that SAVR seemed not to be associated with better outcomes in this report.^,

The findings reported by Jean et al. indeed were limited by the multitude of limitations described above. And although AVR practices cannot be changed confidently based on the data provided in this one study, the fundamental question remains: should aortic valve intervention be offered to patients with moderate AS + HFrEF? The physiology/pathophysiology of these conditions suggest that patients indeed would benefit from intervention—fixed aortic valve obstruction will cause deleterious change to the left ventricle over time, related to pressure overload, and this stress is tolerated poorly in the left ventricle with impaired systolic function. Treatment of HFrEF relies heavily on reduction of afterload; removing the fixed obstruction of a stenotic aortic valve would provide considerable afterload reduction. This clinical conundrum is, in fact, undergoing investigation with a large randomized clinical trial, entitled TAVR UNLOAD (Transcatheter Aortic Valve Replacement to Unload the Left Ventricle in Patients with Advanced Heart Failure). The TAVR UNLOAD study is an international, multicenter trial that seeks to enroll 300 participants with moderate AS and LVEF <50% to undergo either TAVR with optimized guideline-directed medical therapy or optimized guideline-directed medical therapy alone. Outcomes include all-cause death, disabling stroke, hospitalizations due to heart failure, aortic valve disease symptoms, or nondisabling stroke, and assessment of quality of life. The estimated completion date is early in 2022, with data hopefully presented sometime shortly thereafter. Notably, this study will not inform on the benefit of SAVR for this particular patient population. But—importantly—it will bring information as to whether it is best to continue to wait to treat AS until it becomes severe, or if it is indeed worthwhile (and safe) to intervene earlier in those patients with moderate AS + HFrEF via TAVR.