Author + information
- Kelsey Flint, MD∗ ()
- ↵∗Reprint requests and correspondence:
Dr. Kelsey Flint, Division of Cardiology, University of Colorado, 12631 East 17th Avenue, B130, Aurora, Colorado 80045.
Frailty is the accumulation of subclinical physiologic insults across multiple organ systems, leading to a state of heightened vulnerability in the face of stress. Although initially conceived of as a geriatric syndrome most applicable in the primary care setting, over the past decade frailty has been applied to a variety of settings, including medically and surgically treated cardiovascular disease (CVD). The adverse outcomes associated with frailty and CVD are clearly demonstrated in the existing literature (1); however, little is known regarding the biological underpinnings of this association or whether the progression of one disease state affects the other. Furthermore, it is not known if frailty is reversible and, if so, what might be the optimal timing and composition of frailty treatments aimed at improving cardiovascular outcomes.
Understanding the relationship between frailty and CVD remains challenging, because the end result of both entities may clinically appear very similar. One approach to understanding this relationship is to address the age-old chicken-or-the-egg question, “Which came first?” That is, does frailty develop first, thus contributing to the development of CVD? Or does existing CVD lead to frailty? Determining if one disease state precipitates the other holds significant implications for designing interventions aimed at improving the plight of older adults suffering from both frailty and CVD.
In this issue of the Journal, Sergi et al. (2) tested the first of these hypotheses in a prospective study of 1,567 older adults without frailty or CVD at baseline. The authors report the association of pre-frailty at study entry with the development of CVD over a 4.4-year follow-up period. Frailty was measured using modified Fried criteria: slow gait speed, weak grip strength, low energy expenditure, exhaustion, and unintentional weight loss; several of the original Fried criteria (3) were modified to match the data available in the study. Pre-frailty was defined as the presence of 1 or 2 criteria, whereas frailty was defined as the presence of 3 or more. The combined primary endpoint consisted of incident coronary artery disease, peripheral artery disease, heart failure, stroke, and death from a cardiovascular cause. Physicians confirmed new cardiovascular diagnoses at study visits and deaths were adjudicated via death certificate review. The authors found that those with pre-frailty at baseline suffered an increased risk of incident CVD or death compared with their robust counterparts (i.e., met no modified Fried criteria). The association was stronger for those who met 2 modified Fried criteria rather than 1 and was independent of numerous traditional risk factors.
This study supports the hypothesis that frailty contributes to the subsequent development of CVD by demonstrating that patients are at increased risk for incident CVD even before the full frailty phenotype manifests. Interestingly, several subclinical insults existed in the pre-frail, but not robust, patients at study entry. For example, hemoglobin A1c was significantly higher in patients meeting 1 or 2 modified Fried criteria compared with those meeting 0 criteria; although not diagnostic of diabetes, this relatively higher hemoglobin A1c in pre-frail patients suggests that they may have subclinical insulin resistance not present in their robust counterparts. A similar trend was seen with white blood cell count, erythrocyte sedimentation rate, uric acid, and vitamin D levels as well as ankle brachial index and waist circumference. These subclinical changes were more pronounced in patients meeting 2 modified Fried criteria versus 1, and those meeting 1 such criterion versus none. The accumulation of these subclinical insults may manifest as pre-frailty, which, in conjunction with other comorbid conditions and advanced chronological age, likely increases risk for CVD.
Existing data suggest that the reverse is also true: clinical and subclinical CVD are associated with an increased risk of frailty (4–6). Interestingly, the current study does hold clues to support the existence of a bidirectional relationship, because pre-frail patients had lower ankle brachial index and higher prevalence of atrial fibrillation compared with robust patients. Furthermore, of the 5 modified Fried criteria studied, low energy expenditure was most strongly associated with the primary outcome. Sedentary lifestyle is a well-known risk factor for CVD; therefore, this finding may represent a common pathway through which the 2 distinct entities develop.
The evidence supporting a bidirectional causal relationship between frailty and CVD helps describe these 2 disease states as a vicious cycle, with the presence of one feeding the development of the other. The results from Sergi et al. (2) suggest that any intervention aimed at breaking this vicious cycle may be successful by targeting patients earlier in the disease course (i.e., patients with pre-frailty or subclinical CVD). Interventions more likely to break the vicious cycle between frailty and CVD may be those designed to augment overall physiologic reserve and address vulnerabilities specific to CVD risk. As suggested in the authors’ central illustration, global physiologic reserve may be augmented through multidimensional assessments designed to identify and treat problems most likely to contribute to pre-frailty. Clinicians should pay specific attention to geriatric syndromes, such as polypharmacy and poor nutrition, while ensuring patients’ other medical illnesses are optimally managed. Indeed, it makes biological and clinical sense that patients whose comorbidities are well cared for will be less likely to develop pre-frailty. Although other potential targets for augmenting global physiologic reserve are the subclinical changes associated with pre-frailty, designing interventions likely to demonstrate measurable improvement in CVD risk will be difficult given the small absolute differences between robust and pre-frail patients and the heterogeneity of subclinical insults that contribute to pre-frailty.
CVD risk factors are more prevalent in frail compared with nonfrail patients (4,5,7); therefore, an important component to breaking the vicious cycle may lie in aggressive cardiovascular risk factor management. Indeed, treatments aimed at improving low energy expenditure (the modified Fried criterion most strongly associated with incident CVD in this study) may be instrumental in treating both pre-frailty and preventing CVD. If such treatments are successful, determining whether the positive effects are mediated primarily through preventing CVD versus reversing pre-frailty will require careful study design.
Although Sergi et al. (2) provide many excellent insights into the complex relationship between frailty and CVD, their study must be interpreted with care. The authors aimed to enroll patients free from CVD at baseline; however, they did include patients with atrial fibrillation. Unfortunately, the authors did not undertake an exhaustive search for subclinical CVD at study entry. As a result, they do not address the possibility of subclinical atherosclerosis or structural heart disease contributing to the development of pre-frailty and subsequent CVD. The authors attempted to replicate the original Fried criteria (arguably the most widely recognized frailty definition) but had to make modifications on the basis of the data available in their population. Given these limitations, the insights gleaned from this study must be confirmed and built upon in large, longitudinal cohorts in whom both subclinical CVD and frailty status are clearly delineated at baseline and closely followed over time.
The existing literature increasingly characterizes frailty and CVD as intertwined processes that exacerbate one another in a vicious cycle leading to poor outcomes. The results from Sergi et al. (2) suggest that this vicious cycle has its roots long before either frailty or CVD clinically manifest. For now, it seems that the causal relationship between frailty and CVD remains as circular as the question of whether the chicken or the egg came first. Thus, to advance the study of frailty and CVD to the next level, clinicians and researchers must meticulously study the mechanisms behind their link, and rigorously test treatments aimed at mitigating the adverse consequences associated with clinical and subclinical frailty and CVD.
The author thanks Larry Allen for his insightful feedback on this manuscript.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Flint has reported that she has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Afilalo J.,
- Alexander K.P.,
- Mack M.,
- et al.
- Sergi G.,
- Veronese N.,
- Fontana L.,
- et al.
- Fried L.P.,
- Tangen C.M.,
- Walston J.,
- et al.
- Newman A.B.,
- Gottdiener J.S.,
- Mcburnie M.A.,
- et al.