Goals of Care in Patients With Severe Aortic Stenosis

Christopher R. Burke; James N. Kirkpatrick; Catherine M. Otto


Eur Heart J. 2020;41(8):929-932. 

Over the past two decades, there has been a dramatic increase in the number of adults with severe valvular aortic stenosis (AS) undergoing surgical aortic valve replacement (SAVR) or transcatheter valve implantation (TAVI). Based on data from the US Nationwide Inpatient Sample, in this issue of the European Heart Journal Alkhouli and colleagues estimated that the number of interventions for AS in adults over age 60 years increased from 96 to 137 per 100 000 population between 2003 and 2016.[1] Even more impressively, the proportion of patients treated with TAVI, rather than SAVR, increased from 11.9% in 2012 to 43.6% in 2016. Over this time period, early mortality declined both for SAVR (5.4% to 3.3% from 2003 to 2016) and for TAVI (4.7% to 2.2% from 2012 to 2016). The rates of stroke and new renal failure remained similar with both procedures over this time period. Despite a decline in the need for a permanent pacer after SAVR, rates of pacer implantation after TAVI remained high (9.6%) and vascular complications continued to be an issue with TAVI (4.6%).

The increase in the total number of aortic valve interventions was largely driven by an increase in TAVI numbers, with little change in SAVR volumes, suggesting that much of this apparent increase probably does not reflect an 'epidemic' of calcific AS but instead is due to referral of patients who previously would not have received treatment due to age, comorbidities, high surgical risk, or inappropriate failure to refer patients to a Heart Valve Centre.[2] The data also reveal some disturbing disparities in treatment, with fewer AS interventions per 100 000 population in females compared with males and in black or Hispanic compared with white patients, raising concerns about access to care and bias in diagnosis and treatment.

Remarkably, the increase in AS interventions over this 13-year time period appears to be almost entirely related to patient age. There was only a slight increase in AS interventions for patients aged 61–70 and 71–80. In contrast, the number of AS interventions increased substantially in those aged ≥81, most dramatically in those over 90 years old. TAVI was chosen in only 27% of those under age 80, 72% of those aged 80–90, and in nearly all patients over 90 years old. These observations raise important questions about whether we are making the right decisions in our patients with severe AS facing the choice of SAVR, TAVI, or palliative measures instead of a procedure.[3] Although it now is clear that TAVI is a reasonable option for many patients regardless of surgical risk,[4,5] questions still remain about long-term outcomes, which patients are better treated with SAVR, and whether we are fully integrating patient goals of care into the decision-making process.[6] (Table 1)

Younger adults (age <60 years) were not included in the study by Alkhouli et al.[1] and we need to remember that there are several important considerations in deciding between SAVR and TAVI in this age group. The first issue is valve durability given that a 60-year-old American man can expect to live an additional 22 years and a woman an additional 25 years. Thus, based on life expectancy and valve durability, both European Society of Cardiology (ESC) and American Heart Association/American College of Cardiology (AHA/ACC) guidelines recommend a mechanical aortic valve prosthesis for patients under 60 years old, unless anticoagulation in not desired or is contraindicated.[7,8] Mechanical valves will become an even better option once we have a more effective and safer oral anticoagulant medication. There also are extensive data on durability of SAVR bioprosthetic valves, whereas there is uncertain durability of TAVI, particularly in younger patients given the known inverse association between age and bioprosthetic valve durability.[9–11] It is hoped that valve-in-valve transcatheter procedures will be a viable option for younger patients who choose an initial surgical bioprosthetic valve, but there is little evidence to support this approach so far.

The second important consideration in younger patients is valve and aortic anatomy. Most young adults with severe AS have a congenital bicuspid aortic valve (BAV). Not only is this anatomy less favourable for TAVI but many patients require concurrent replacement of the aortic sinuses or ascending aorta due to the associated aortopathy. In addition, patients with a BAV often have an annulus size larger than currently available TAVI valves. Conversely, some patients with congenital aortic valve disease have a small outflow tract, annulus, and aortic sinuses which requires a surgical enlarging procedure to allow placement of an adequate size prosthetic valve and avoid patient–prosthesis mismatch. Finally, a lower risk of permanent pacer implantation with SAVR is an important consideration given the long-term risks of heart failure, infection, and tricuspid valve dysfunction with a permanent pacer. Surgical AVR therefore remains the most appropriate choice in the majority of young adults. Mechanical SAVR is probably underutilized in this patient group; only time will tell the long-term impact of the changing trend towards bioprosthetic valves in younger patients

The decision about SAVR vs. TAVI is challenging in patients who are not young or very old, loosely defined as patients 60–80 years old, with severe AS. There is a continuum of balance between benefits and risks across this age range, with individual patient factors tilting this balance one way or the other, as detailed in Table 1. Expected remaining years of life range from 24 years for the average 61-year-old female to 8 years for the average 80-year-old male, but may be greater or less than this estimate in a specific patient depending on comorbid conditions.[12] At the younger end of this age spectrum, valve durability, risk of pacer implantation, and the presence of a BAV remain important considerations. Procedure-specific impediments such as previous cardiac surgery with bypass grafts vulnerable to damage during repeat sternotomy, prior chest irradiation, and severe aortic calcification shift the risk–benefit balance towards TAVI. On the other hand, the balance shifts towards SAVR in patients with concurrent conditions such as severe primary mitral regurgitation needing mitral valve surgery, aortic dilation meeting criteria for graft replacement, severe coronary disease requiring bypass grafting, septal hypertrophy with subaortic obstruction that would respond to myectomy, and atrial fibrillation in whom a maze procedure would be beneficial. Physicians need improved guidance from the ESC, AHA/ACC, and BMJ Rapid Recs to inform decision-making in this patient group.[7,8,13] Patients need better decision aids to actively participate in informed shared decision-making that considers patient goals and expectations, as well as medical considerations in choosing between SAVR and TAVI.[14,15] Some patients will place most value on a shorter hospital stay, less pain, and faster recovery, while accepting the risks of a pacer or vascular complications and uncertain TAVI durability. Other patients will have the opposite point of view. It is the responsibility of the Heart Team to balance individual preferences with accurate expectations for long-term treatment success.[2]

In the oldest patients with severe AS, TAVI nearly always will be preferred over SAVR, depending on individual factors such as comorbidities, estimated remaining years of life, patient preferences, and values. TAVI is at least equivalent, and possibly superior, in terms of mortality and symptom relief over short and intermediate (5 year) follow-up. The obvious advantages of TAVI over SAVR in terms of a shorter hospital stay, less pain, and a faster recovery are important to most older patients. However, it is also essential to discuss patient expectations and goals of care. Life prolongation, which will be a smaller absolute length of time than in younger patients, may be less important than avoiding complications such as a stroke or a permanent pacer. Severe AS is not just a disease of the valve; these patients have irreversible changes in left ventricular function and the systemic vasculature that continue to contribute to persistent symptoms, and heart failure remains the primary cause of death after TAVI, to say nothing of the burdens of multimorbidity. Symptoms that are not due to AS will not improve after TAVI; this seems obvious to physicians, but patient expectations may be unrealistic unless these issues are explicitly discussed. We all have had the patient with successful resolution of heart failure symptoms after TAVI who then switches from focusing on her heart to focusing on her hip pain and is as dissatisfied with life after TAVI as before the procedure. Rather than thinking about decision-making in terms of standard 'informed consent', which is very short term oriented and generally not nuanced enough, we need to consider the alternative of a palliative approach and should resist the urge reflexively to treat a stenotic valve simply because 'it is there'.[3]

The other observation in the study by Alkhouli et al.[1] that should give us pause is the almost doubling in cost for AS interventions in just over a decade. At some point, we will need to consider whether this cost is affordable, especially when considered in light of other healthcare gaps and suboptimal health outcomes in the USA.