Transcatheter aortic valve replacement over age 90: Risks vs benefits

Abstract As the population ages, clinicians will encounter a growing number of nonagenarians suffering from severe aortic stenosis who may be candidates for transcatheter aortic valve replacement (TAVR). By virtue of a healthy survivor effect or a referral bias, these patients may paradoxically have greater resilience and fewer comorbidities than their octogenarian counterparts. They tend to, on average, tolerate the TAVR procedure quite well with low in‐hospital and 1‐year mortality rates of 5.5% and 23%, respectively. Appropriate patient selection should consider individualized estimates of procedural risk, potential for functional recovery and for improved quantity and quality of life. Frailty is much more revealing than chronological age, and it can be measured by brief tools such as the Essential Frailty Toolset. Ultimately, the process of shared decision‐making is paramount to ensure that the course of action is patient‐centered and balances the procedure's expected risks and benefits with the nonagenarian's preferences and values.


| INTRODUCTION
Calcific aortic stenosis (AS) is the most common form of acquired valvular heart disease in older adults. 1 The pathophysiology of calcific AS is closely connected to the aging process; with inflammaging, calcification, and repetitive mechanical stress being among the driving mechanisms. 2 An estimated 17% of nonagenarians will develop at least mild AS over the course of their lifetime. 3 By 2050, the number of nonagenarians is expected to quadruple to >8 million in the United States, 4 and given that there is no proven therapy for the prevention of AS, the number of "oldest old" patients suffering from this disease is expected to mirror the demographic population trends and rise exponentially. When AS becomes severe it is often associated with debilitating symptoms, reduced functional capacity, hospitalizations, and heart failure eventually leading to death. Prognosis is poor in the absence of aortic valve replacement, and historically nonagenarians were excluded from this surgery due to the higher procedural risks and lower perceived benefits.
Transcatheter aortic valve replacement (TAVR) has emerged as a therapeutic option for severe symptomatic AS in patients at high surgical risk 5,6 as well as intermediate and low surgical risk. [7][8][9] Nonagenarians were not excluded from this procedure, to the contrary, initial TAVR trials targeted patients that were deemed too old or frail to undergo surgery. Early experiences indicated that these patients could undergo the minimally invasive TAVR procedure with acceptable risk.
Since that time, the number of TAVRs performed in nonagenarians has progressively increased, accounting for one out of seven TAVR procedures in the United States between 2011 and 2014. 10 As nonagenarians are increasingly referred for consideration of TAVR, clinicians are tasked to identify the "good 90-year-old" who will likely tolerate the procedure and derive meaningful benefits-a forecast that is imperfect but informed by objectifiable features. The goal of this article is to review the literature on TAVR in nonagenarians and to provide guidance for individualized patient-centered decision making.

| THE NONAGENARIAN PHENOTYPE
According to actuarial life tables in the United States, the average life expectancy at birth is 76.0 years for men and 80.1 years for women. 11 Having surpassed this life expectancy by more than a decade, nonagenarians have effectively overcome competing risks and declared themselves to be more resilient to stressors encountered during the course of their lives. Given this self-selection, the average remaining life expectancy at age 90 is not trivial; calculated to be 4.1 years for men and 4.9 years for women. Otherwise said, the annual probability of death at age 90 is only 16% for men and 13% for women, respectively.
One might assume that nonagenarians would have higher rates of comorbidities as compared to their relatively younger counterparts.
However, a "healthy survivor" effect has been observed whereby nonagenarians-by virtue of their achieved survival-have paradoxically lower rates of comorbidities. Still, most nonagenarians with severe AS will likely have at least one significant comorbidity. In a Spanish study of nonagenarians with severe AS managed both with TAVR and conservatively, the most common comorbidity was chronic kidney disease, which was present in 70% of the cohort. 12 Other comorbidities present in at least 10% of the cohort were diabetes mellitus (32%), myocardial infarction (16%), dementia (13%), and previous stroke (11%). The mean Charlson Comorbidity Index score was 3.2 and only 32% of the cohort had a low comorbidity burden; a high comorbidity burden was associated with increased 1-year mortality.
In particular, end-stage renal disease and oxygen-dependent lung disease have been associated with markedly increased risks of mortality and major morbidity. 13 Frailty, defined as a diminished capability to recover from pathological or iatrogenic stressors due to cumulative age-related impairments, is a key consideration in nonagenarians. Impairments to 6.9%) (Figure 1). In 12 studies encompassing 6535 nonagenarians, the pooled 1-year mortality rate was 23.0% (95% CI 20.6% to 25.5%) ( Figure 2). Interestingly, this 1-year mortality is not drastically different F I G U R E 2 Meta-analysis of 1-y mortality in nonagenarians undergoing TAVR than that of nonagenarians in the general population, which is approximately 15%. In studies comparing patients above and below age 90, the risk of mortality was slightly increased in some studies and no different in others. 10,19,21,24,[27][28][29][30]32,33,[38][39][40] This inconsistency is at least partially attributable to variable adjustment for age-related confounders.
Another reason for the inconsistent association between age and adverse outcomes is that chronological age is a flawed surrogate for biological aging and cumulative effects of clinical and subclinical impairments throughout the body. Frailty has been said to be a better indicator of biological aging and accumulated deficits. Nine studies measured frailty by at least one objective metric such as the Clinical Frailty Scale or the 5-m gait speed test. 8

| SELECTING THE RIGHT NONAGENARIAN PATIENT FOR TAVR
Selecting the appropriate nonagenarian for TAVR involves forecasting short-term risks, mid-term recovery, and long-term benefits of the procedure. In the short-term, is the patient at risk for a major procedural complication? The STS Risk calculator (http://riskcalc.sts.org/) was developed for SAVR but is still appropriately used as a benchmark to stratify candidates for TAVR. 38 The In the mid-term, is the patient likely to recover from the procedure and return home within a reasonable timeframe? Conversely, the lessfortunate patient suffers a vicious cycle of ongoing deconditioning, prolonged hospitalization, and discharge to another healthcare facility.
Attributes that may markedly impede or slow down the recovery process include: physical frailty and poor mobility, low social support, active depression. Physical frailty can be assessed with the chair rise and 5-m gait speed tests, among others. 43,44 In the long-term, is the patient likely to benefit from the implanted

| POSTPROCEDURAL CARE OF NONAGENARIANS
Despite appropriate selection and technical execution, 30%-40% of older patients report persistently poor functional status and quality of life following TAVR-especially frail older patients who were deconditioned before and increasingly-so after the procedure. Therefore, frailty should not only be viewed as a prognostic marker, but also a therapeutic target than can be improved with exercise and nutrition, ideally in the context of a structured cardiac rehabilitation program. 48 Nonagenarians have been shown to benefit from cardiac rehabilitation 49 and should not be excluded but rather preferentially referred following TAVR. Cardiac rehabilitation programs can be adapted to their abilities and progress at a suitable pace; they can even be delivered in the home-based setting as in the ongoing PERFORM-TAVR Trial (Protein and Exercise for Reversal of Frailty in OldeR Men and women undergoing TAVR; NCT 03522454).

| CONCLUSIONS
As the population ages, clinicians will encounter a growing number of nonagenarians suffering from severe AS. By virtue of a healthy survivor effect (or a referral bias), these patients can paradoxically have greater resilience and fewer comorbidities. They can, on average, tolerate the TAVR procedure quite well with low in-hospital and 1-year adverse event rates; acknowledging that this subgroup of patients is under-represented or at the very least highly selected in randomized trials, and that the evidence to-date stems mostly from observational studies. Appropriate patient selection should consider individualized estimates of procedural risk, potential for functional recovery and for improved quantity and quality of life. Age alone is inadequate to stratify patients. Frailty is much more revealing, and it can be measured by brief tools such as the EFT, and in selected cases by deeper evaluations such as that offered by a comprehensive geriatric assessment.
Ultimately, the process of shared decision-making is paramount to ensure that the course of action is patient-centered and balances the procedure's expected risks and benefits with the nonagenarian's preferences and values.