Evidence of the natural history of severe asymptomatic aortic stenosis (AS) in elderly individuals is lacking because the main study series have predominantly included patients younger than 70 years of age.1,2 The aged population presents special features, such as a high number of comorbidities and geriatric syndromes. In addition, the detection of early symptoms in elderly patients may be difficult, especially if their physical mobility is limited.

The presence of frailty is relatively common in patients with heart disease. This condition is a geriatric syndrome characterized by increased vulnerability to minor stressors, resulting in a decline in multiple physiologic systems.3 Frailty plays a major role in the selection of candidates for transcatheter aortic valve implantation (TAVI)4–6 and aortic valve replacement (AVR) in the geriatric population with severe AS.7–9 In this population, measures of frailty are highly predictive of poor outcomes, including death, incident disability, and hospitalization.10 In fact, in approximately 40% of patients who undergo TAVI, the procedure is futile, resulting in poor outcomes and a lack of improvement in survival and/or quality of life.11 Clinical practice guidelines recommend an evaluation to optimize the selection of patients referred for invasive treatment.12,13 However, there is a lack of consensus on the best tool to assess frailty.14 On the other hand, AS may be the cause of frailty, and there are aspects of frailty that may be reversible and susceptible to rehabilitation.15 Consequently, a clinical assessment of several geriatric and cardiological factors that influence this asymptomatic population may be vital to prevent early mortality.

Other variables related to age, such as comorbidity, functionality, cognitive status and quality of life, can influence the prognosis of these patients.16 All of these factors should be included in the comprehensive geriatric assessment. Regarding highly prevalent comorbidities in geriatric patients, some noncardiac conditions impact prognoses in the short- and long-term, irrespective of management. Therefore, the inclusion of an assessment of comorbidities along with the Charlson comorbidity index in a preoperative study17 was shown to be useful in predicting poor outcomes in patients who are candidates for invasive treatment.18

Our aim was to define the risk factors that influence early mortality in an elderly population with severe asymptomatic degenerative AS.

We define, for the first time, several predictors of mortality in asymptomatic geriatric patients with severe degenerative AS, highlighting age, frailty and a Charlson comorbidity index ≥ 5 as predictive factors.

Geriatric populations with severe AS are a major burden due to their substantial health care costs. In the aged population, severe AS may affect up to 4.6% and 8.1% of patients older than 75 and 85 years, respectively.25,26 Our study population showed a high prevalence of frailty (59.6%), moderate to severe degrees of dependence (49%), and associated comorbidities. All of these factors challenge the detection of early symptoms in the elderly population, who frequently experience severe onset of symptoms.27 According to a review by Généreux et al.,28 among populations with asymptomatic severe AS, 1-year and 5-year survival rates range from 67% to 97% and 38% to 83%, respectively. In our study, the overall survival rate was 84% at 1 year and 42% at 5 years. Our results showed a considerably lower survival rate than that reported by Taniguchi et al. (58% at 5 years) in a retrospective analysis.29 This difference can be explained by our increased prevalence of comorbidities as well as the increased age of our population (77.8±9.4 vs 83.3±8.8 years).

Comorbidities are common in elderly patients; they influence risk-benefit analyses of operative risk, late outcomes after interventions, and life expectancy, regardless of valvular disease.30,31 Similar to our results, Martínez-Sellés et al.18 demonstrated that patients with severe AS and a high comorbidity index (Charlson comorbidity index ≥ 5) had a poor short-term prognosis, with a mean survival less than 1.5 years; mortality was mainly related to noncardiac causes and the patients did not seem to benefit from interventional treatment. The mean Charlson comorbidity index in both studies was very similar; however, the number of patients with elevated comorbidities (≥ 5) was higher in the study by Martínez-Sellés et al. than in our study. In addition, it is important to mention that the Martínez-Sellés et al. population was symptomatic and that the age of inclusion was higher (80 vs 70 years) than that in our study, which could also influence the progression of these patients. This index shows an independent association with mortality and mean survival of 1 year in patients with a score ≥ 5. However, it is logical to consider that an elderly patient with 5 or more comorbidities will have a very poor prognosis. Moreover, as we observed in our population, the frail patient group had more comorbidities and greater limitations in daily living activities than the nonfrail patient group, demonstrating the interrelation of these parameters. As Fried32 described, frailty and comorbidities often overlap in elderly subjects, leading to impairment in functional status, resulting in a poor prognosis.

When the Charlson comorbidity index was assessed as a mortality discriminator for these patients, we found an area under the curve <0.75. A value ≥ 4 had a specificity of 100% but low sensitivity. This finding could be related to the limitations of this index in geriatric patients.33 The weight attributed to each of the pathologies does not correlate with the burden of disease that each causes in the geriatric population. Moreover, some prevalent and disabling diseases that cause high mortality in the elderly, such as parkinsonism, depression or ischemic heart disease without infarction, are not included in this index.

The prevalence of frailty was very high in our population; in total, 59.6% met the frailty criteria, compared with 49.3% in a study by Rodríguez et al.34 Measures of frailty, even after adjustment for age and comorbidities, are highly predictive of poor outcomes, including death, incident disability, and hospitalization in patients with cardiovascular disease.10,35 Regarding AS, it has been shown that frailty is associated with increased hospitalization due to heart failure.34 According to our results, the main factors associated with frailty are advanced age, female sex, a high burden of disease, disability, and cognitive impairment.31,36 Frail patients showed reduced low-density lipoprotein cholesterol fractions as a consequence of protein-calorie malnutrition and loss of muscle mass resulting in bed rest, sarcopenia, and prolonged hospital admission. An increased prevalence of atrial fibrillation was observed in our population, resulting in adverse impacts on morbidity and mortality.35

Our results demonstrated that frailty was an independent factor of mortality. In total, 69.4% of frail patients died, with a median survival of 2.52 years (95%CI, 1.36-3.69); the 5-year survival rate was 15%. These results are considerably lower than those of other studies carried out in the global population where frailty was not considered.28,29 The importance of frailty as a prognostic marker has been widely demonstrated, emerging as a vital cardiovascular care tool and a major component in the decision-making process. Currently, a frailty assessment is essential for patients undergoing surgery, especially those who are candidates for TAVI.12,30 Almost 40% to 50% of patients who undergo TAVI show poor health outcomes, either due to death or because their clinical status does not improve.37,38 However, frailty should not be a reason for exclusion in all cases because AS may be a frailty source and there are aspects of frailty that may be reversible. Thus, in our study population, 4 (6.5%) patients with frailty finally underwent TAVI successfully, with survival at the last follow-up. In this regard, the setting for rehabilitative procedures could prevent some episodes of CHF and improve the prognoses of invasive procedures.39

It would be interesting to consider whether the parameters currently used to assess frailty are valid in sick and debilitated patients. Perhaps other markers of advanced frailty or disability (inability to walk, low albumin, dependence in activities of daily living) should be considered for improved risk discrimination.40 Recently, the FRAILTY-AVR study investigators demonstrated that for both TAVI and AVR, the essential frailty toolset, which includes lower extremity weakness, cognitive impairment, anemia and hypoalbuminemia, had the highest predictive value for death at 1 year and was the strongest predictor of worsening disability at 1 year as well as death at 30 days.41

Consistent with other studies, elderly patients were frequently undertreated.42,43 In our population, the main reason was Heart Team rejection due to significant frailty, dependence or severe comorbidities. Only 40.6% of the patients who developed symptoms were referred for invasive treatment. Invasive therapy was carried out in 45% of the PEGASO registry cohort; this proportion is similar to our result.43 In a study by Iung et al.,42 surgery was carried out in 67% of elderly patients with severe symptomatic AS; this percentage higher than ours, but the patients were younger than out patients, and frailty measures were not included in the study. In fact, population survival was affected when patients who developed symptoms were denied invasive treatment.

In our population, patients were not considered for valve replacement due to the absence of symptoms. Notably, when they developed symptoms, they were rejected due to their frail status. The need for a frailty assessment at the time of severe AS diagnosis, regardless of the symptoms, is critical for this population. To date, frailty scoring has been carried out to decide whether to treat symptomatic patients; our results demonstrate that an early evaluation of asymptomatic AS patients could improve mortality and invasive treatment results. In this regard, the use of the FRAIL questionnaire is potentially useful for screening this population.14

We propose a more restrictive attitude regarding elderly patients requiring valvular replacement for improved mortality results. Frailty screening in nonsymptomatic severe AS patients permits the personalization of our approach, whether that involves more aggressive therapy or the application rehabilitation steps to reverse improvable frailty aspects.

Frailty may be difficult to identify.44 Weakness, decreased mobility, and limitation in performing routine physical activities are common in frail patients but could also be a consequence of AS. Therefore, the presence of frailty in patients who report being asymptomatic due to possible adaptation to their condition could be a manifestation of AS. Considering this evidence, it is crucial to develop multidisciplinary teams focused on geriatric cardiac patients. Future studies are needed based on specific programs that help to reverse frailty and improve the results of invasive treatment in this population.

Among the risk factors that influence the elderly population with asymptomatic severe AS, frailty is a common condition and was one of the main reasons for denying invasive treatment when necessary. This geriatric syndrome was a powerful independent factor of mortality and poor prognosis. Mortality was also associated with age and a Charlson comorbidity index ≥ 5, but this scale is not sensitive in geriatric patients.

This work was partially supported by grants (to M. Ramos and M. Quezada) from the VII Convocatoria del Banco de Santander and Alfonso X el Sabio University. Additionally, this work was cofinanced by the Ministry of Health of Junta de Andalucía within the framework of the Integrated Territorial Initiative, file number PI-0048-2017, and the European Regional Development Fund (ERDF) within the Operational Program of Andalusia, ERDF 2014-2020.

None declared.