Usefulness of the Edmonton Frail Scale in Assessing the Impact of Heart Failure on Frailty

Abstract

Background/Aim: The association of frailty with heart failure (HF) is common in the elderly, and its presence is a poor prognostic factor; it increases the risk of falls, disability, hospitalization, and mortality. The aim of this prospective study was to assess the incidence of physical frailty in patients with HF and the role of physical exercise in improving physical performance.

Patients and Methods: A total of 141 patients with musculoskeletal pathology, aged over 65 years, who followed a specific physical training program were included. The patients were assigned to two groups: HF patients –group HF (n=53) and patients without HF −group N-HF (n=88).

Results: At cohort level, mild and moderate frailty was detected in 20.56% of patients enrolled in the study (n=29). Severe form of frailty was identified in 2.83% of cases (n=4). The prevalence of mild, moderate, and severe frailty in the two groups differed significantly (p<0.05). Patients with mild frailty and moderate frailty in the HF group represented 24.52% compared to 18.18% in the N-HF group (p=0.007). Severe frailty was present in 5.66% in the HF group, not significantly different from the N-HF group (1.13%), p=0.317. The values obtained in the functional independence and functional performance domains were significantly improved at the reassessment after 6 months.

Conclusion: Exercise-based rehabilitation is a primary therapy for improving physical performance, reflected by increased independence related to daily activities and functional performance in HF patients.

Advanced age associates musculoskeletal damage, decreased bone and muscle mass, as well as multiple organ and system damage (1), leading to a gradual decline in physiological reserve and a failure of the homeostatic mechanisms collectively characterized as a biological syndrome called frailty (2). Over time, attempts have been made to define frailty. The term was introduced about 30 years ago to define vulnerability of the health status of older adults that can undergo sudden changes triggered by physiological or psychological stress (3). Currently, two studies are considered to have given the most widely accepted characterization of this syndrome: the Cardiovascular Health Study (CHS) and the Canadian Study of Health and Ageing (CSHA) (4). Clinically, the syndrome is defined (according to CHS) by self-reported decreased physical activity, decreased gait speed, weak muscle strength (measured as grip strength), self-reported exhaustion, and involuntary weight loss. Three out of these five signs are sufficient for the diagnosis (5). Loss of bone mass and muscle mass are essential factors in causing age-associated frailty (6). The CSHA proposed a cumulative deficit model, which measured frailty through an index of deficits caused by age-related disabilities and diseases (4). Physical frailty, expressed by impairments of several physical functional domains (muscle weakness, slowness of gait and movements, and low levels of activity), is associated with sarcopenia (7). The current recommendations are that all patients over 70 years of age be tested for frailty (3). The estimated incidence of cardiovascular disease is about 60% in people over 60 years of age.

Among these diseases, heart failure (HF), with an incidence ranging between 6% and 13%, increases morbidity and mortality by impairing the structure and function of the ventricles and represents the end-stage of most heart diseases. The risk of developing HF in subjects over 40 years is 1 in 5, regardless of sex; one in 9 deaths is related to heart disease with HF. The ability to work, physical exercise tolerance, sleep, and psychosocial profile of patients are all significantly altered by HF (8).

Epidemiological studies on HF began in Framingham in 1948, but until 1960 there were no clear diagnostic criteria, no known preventive measures, and no emphasis on identifying risk factors. To date, we know that the primary cause of HF is myocardial ischemia leading to decreased heart pump function. Other causes are untreated hypertension, valve diseases, and myocarditis (bacterial, parasitic, viral). In addition, metabolic syndrome, low physical activity, dyslipidemia, and smoking have all been linked to incident HF, either through coronary artery disease or conditions associated with HF, such as type 2 diabetes mellitus or overweight, which have been shown to lead to HF through several mechanisms (9).

The most commonly used classification criterion for HF is left ventricular ejection fraction (LVEF) (10). HF with preserved ejection fraction (HFpEF) is described as HF with normal LVEF (≥50%) and HF with reduced LVEF (≤40%) as HF with reduced ejection fraction (HFrEF) (11). Patients with HF with an LVEF of 40% to 49% are classified as having HF with mildly reduced ejection fraction (HFmrEF). Although HFmrEF is now classified as a distinct disease, its epidemiology, pathogenesis, therapy, and prognosis are unknown (12). The Framingham criteria for congestive HF are major and minor (Figure 1). The diagnosis is supported by one major and two minor or two major criteria.

Figure 1. Diagnostic criteria according to Framingham.

The New York Heart Association (NYHA) classification is based on the relationship between dyspnea and physical activity (13). A 4-stage classification of HF has been developed in 1928 that takes into account the clinical manifestations and limitations of physical performance, as illustrated in Figure 2A: class I: symptoms are absent; class II: minor symptoms when engaged in regular physical activity; class III: patients do not show symptoms at rest but display dyspnea during minor physical activity; and class IV: extreme shortness of breath even during rest.

Figure 2. Classification of heart failure. (A) New York Heart Association (NYHA) classes; (B) Correlation of American Heart Association and American College of Cardiology Foundation stages with NYHA classes.

Another classification proposed by the American Heart Association and the American College of Cardiology Foundation also refers to four stages, from A to D. Stage A is a pre-clinical stage, when only risk factors are identified, stage B corresponds to Class I, stage C includes Class II, III, and IV, and stage D is the terminal stage (Figure 2). The prevalence of HF is expected to double in the next 40 years (14). HF often associates frailty, with an estimated prevalence of almost 50% in patients with HF as compared to 10% in the general population (2,15). Especially in elderly patients with HF, frailty increases the risk of falls, disability, and leads to frequent and prolonged hospitalization episodes and even death (16,17).

A diversity of scales has been employed for the assessment of frailty. A systematic review published in 2016, based on consensus-based standards for the selection of health measurement instruments (COSMIN), tracked frailty assessment tools (5,063 studies) and identified 38 assessment methods; validity and reliability were demonstrated for 2 of the 38 (the frailty index-comprehensive geriatric assessment and the Tilburg frailty indicator) (4). The Edmonton frail scale (EFS) assesses frailty from a multidimensional perspective, bringing together several parameters: the mini-mental state examination (MMSE) score, the Barthel and activities of daily living (ADL) functional independence scores, the number of diagnosed illnesses and the number of medications, the mini nutritional assessment (MNA) score, the geriatric depression scale (GDS) depression score, the skeletal muscle index of sarcopenia (SMI), the osteoporosis score, and the muscle strength determined with a dynamometer (18).

A frailty-based prognostic score for patients with HF has been developed to enable risk stratification. The score takes into account the walking speed, grip strength, self-assessment of walking 400 meters (from totally distrustful to very trustful) and performance in activities of daily living (self-assessment on the performance measure for activity of daily living-8 questionnaire) (19).

The present study aimed to assess the impact of HF on frailty-related physical performance in patients with musculoskeletal conditions using the EFS and the effects of physical training on patients with frailty and HF.

Patients and Methods

Study design. The study was conducted between May 2022 and February 2023 in the Avram Iancu Clinical Hospital Oradea to assess the incidence of frailty in HF patients hospitalized on the Medical Rehabilitation ward. A total of 252 consecutive patients with various musculoskeletal conditions (neurological, degenerative, or inflammatory) were evaluated for inclusion in the study. The selected subjects were divided into two groups: 1. HF patients – group HF; 2. patients without HF – group N-HF, (control group).

Patients in the HF group had a definite diagnosis of HF for more than one year, with ultrasonographic confirmation by a certified cardiologist. All patients received medication according to current guidelines. Subjects in the control group (N-HF group) showed no signs of heart failure (assessed by the cardiologist) but had risk factors for HF (coronary heart disease, diabetes, or hypertension).

Inclusion/exclusion criteria. All patients with various musculoskeletal conditions who visited the Medical Re-habilitation Department of Avram Iancu Clinical Hospital Oradea between May 2022 and February 2023 were screened. The exclusion criteria were the following: age under 65 years, presence of a joint disorder that could significantly interfere with the evaluations, patients with cardiac arrhythmias with malignant potential, thromboembolic cardiovascular events in the prior 6 months (myocardial infarction, stroke, pulmonary thromboembolism, and deep vein thrombosis), psychiatric disorders, and unwillingness to participate in the study.

A total of 146 patients were declared eligible, but five patients were lost to follow-up because they did not complete all the required assessments; Figure 3 summarizes the information provided above.

Figure 3. CONSORT flow diagram of the study.

Sample size. The sample size of subjects included in the study was calculated based on the total number of patients who attended the outpatient clinic during the study period. To calculate the sample size, we considered the following variables: p—the probability of occurrence of the phenomenon, 0<p<1, q—counter-probability, q=1-p, t—probability factor, x—the error limit, N—the volume of the community. To determine the sample of cases, we used the formula: n=t² pq/(x² + t² pq/N). The formula is valid for studies in which the characteristic followed is an alternative (in our case healthy–sick). The value of n is maximum if the product of pq is maximum, i.e., when p=q=0.5. The probability of 95% corresponds to a value of t=1.96. A limiting error of 0.1 was set. If N is large, above 10,000 (in our case N=252), the ratio t² pq/N is neglected. The value obtained by the above formula is n=96.

Study tools. All patients were clinically assessed using ehe EFS questionnaire. The evaluation was conducted by two independent team members. The scale was translated by an authorized translator. The first assessment was performed at study entry and the second assessment was done after 6 months. The EFS assesses nine domains of frailty (cognition, general health, functional independence, social support, medication use, nutrition, mood, continence, functional performance) (20). Each domain is given a score from 0 to 2, depending on the impairment detected for each domain, with 0 being assigned for the best performance and 2 indicating significant impairment. The total score ranges between 0 and 17. Scores below 5 indicate no frailty, whereas scores above 12 associate severe frailty. Values of 6 to 7 indicate vulnerability or pre-frailty, scores of 8 and 9 indicate mild frailty, and 10-11 moderate frailty.

All patients underwent a continuous moderate-intensity physical training program up to a frequency of 70% of maximum heart rate (220 – age in years). The training sessions were performed in the physical therapy room, 5 days/week, for 2 weeks; the duration of the training was progressively increased from 20 min to 40 min, structured in three parts: warm-up period (10 min), actual training (25 min aerobic exercises), and 5 min cool-down period. At discharge, patients were recommended to continue training three times a week for six months.

Ethical approval. The study was approved by the institutional evaluation committee of the Avram Iancu Clinical Hospital of Oradea, Bihor County, Romania (CEFMF/02/19.05.2022). The research was conducted in compliance with the Declaration of the World Medical Association of Helsinki. Participation in the study was voluntary. Written informed consent was obtained from all participants.

Statistical analysis. Statistical analysis was carried out using the IBM SPSS Statistics for Windows, version 20 (IBM Corp., Armonk, NY, USA). Means and standard deviations were determined and the Student’s t-test, Wilcoxon Signed Ranks Test, Mann-Whitney, and chi-square tests were run. The statistical significance was set at p<0.05.

Results

The two groups did not differ significantly in terms of age, rural or urban background, or sex, but showed statistically significant differences in body mass index (p=0.045) past medical history (p=0.001), and the level of physical activity performed by patients (p=0.037), as shown in Table I.

Table I. Baseline characteristics of the groups.

M: Mean value; HF: heart failure; SD: standard deviation value; N: total number; BMI: body mass index, PMH: past medical history; p-values, statistical significance based on chi-square test.

The risk factors for HF assessed were: ischemic heart disease, valvular heart disease, diabetes mellitus (DM), dyslipidemia, smoking, and physical activity. Patients with HF were more likely to have coronary heart disease (p=0.001), valve diseases (p=0.001), and dyslipidemia (p=0.001), whereas non-HF patients reported more often smoking (p=0.019) and engaged in more regular physical activity (p=0.037). Significant differences were also observed between the number of patients with stage II hypertension (N=43, Group HF versus N=73, Group N-HF, p=0.001). In the HF group, 66.03% (n=35) of patients had stage II HF and 26.41% (n=14) had stage I HF.

At the cohort level, the study showed the absence of frailty in 37.58% (n=53) of the patients recruited; vulnerability was present in 39% (n=55). Mild and moderate frailty was detected in 29 patients enrolled in the study (20.56%). Severe frailty was identified in four cases (2.83%). The mean scores on the EFS obtained for the two groups indicated increased vulnerability, with significant differences between the two groups. The mean EFS score at baseline in the HF group was 7.11±2.54 compared to 6.10±1.67 in the N-HF group (p=0.001). After 6 months (Figure 4), the mean EFS score remained in the range of increased vulnerability; however, it decreased to 6.94±2.53 in the HF group and to 5.57±1.39 in the N-HF group, the difference between the two groups still reaching statistical significance (p=0.001).

Figure 4. Distribution of the Edmonton frail scale (EFS) score values across the study groups: (A) initial evaluation; (B) final evaluation. HF: Heart failure; N-HF: no HF.

Patients with HF were more likely to be associated with frailty, with significantly more patients showing moderate frailty (16.9% vs. none in the N-HF group), and three patients being identified with severe frailty as compared to a single patient in the N-HF group (although the difference did not reach statistical significance presumably due to the small number of cases). Overall, only about one quarter of HF patients were not associated with frailty compared to almost half of patients in the N-HF group, as shown in Table II.

Table II. The Edmonton Frail Scale (EFS) score values of the groups.

M: Mean value; HF: heart failure; SD: standard deviation value; N: total number; p-values, statistical significance based on chi-square test.

The physical performance of our patients was assessed by means of the Functional independence and Functional performance domains of the EFS. The Functional independence scores differed significantly between the two groups both at baseline (0.53±0.64 in the HF group vs. 0.53±0.50 in the N-HF group; p=0.012) and after 6 months of physical training (0.51±0.61 in the HF group vs. 0.53±0.50 in the N-HF group; p=0.035), but the improvement in the group of HF patients was significant. The Functional performance scores showed significant differences between the two groups both at baseline (p=0.001) and at the end of study (p=0.001).

Most patients in the N-HF group scored between 0 and 1 in Functional performance and Functional independence evaluation (Figure 5), which indicates difficulties in performing up to 4 out of the 8 activities assessed: managing money, using the phone, shopping, housekeeping, medication intake, transportation, doing the laundry, and preparing a meal. The reported difficulties were mainly related to cleaning and using transportation means, which might be explained by the associated musculoskeletal diseases. Nonetheless, 13.2% of patients in the HF group scored between 1 and 2 in both domains, pointing to difficulties in more than four activities of daily living, whereas the time needed to sit up from a chair and walking 3 meters was longer than 20 s.

Figure 5. Comparative evolution of the Edmonton frail score (EFS) score on Functional independence between the two groups: (A) initial evaluation; (B) final evaluation. *Functional independence 1-initial evaluation; Functional independence 2-final evaluation; Functional performance 1-initial evaluation; Functional performance 2-final evaluation. HF: Heart failure; N-HF: non-heart failure.

After 6 months of training, the Functional performance scores improved by 0.32±0.61 in the HF group and by 0.36±0.48 in the N-HF group, and the difference between the two groups was no longer significant (p=0.267).

To assess the benefits of physical training and functional performance between the two time points of evaluation in each group we used the Wilcoxon Signed Ranks Test. The frailty score showed significant improvement after the 6 months of physical training, as did the Functional performance score, which included shortening of the time needed to stand up from a chair and walk 3 meters (p=0.000). However, these improvements did not translate into increased functional independence, and activities of daily living remained difficult to perform (p=0.317 in the HF group and p=1.00 in the N-HF group of patients). The results of these comparisons are shown in Table III.

Table III. Comparative results of total Edmonton frail score (EFS) score and by domains for the heart failure (HF) group and the group with no HF (N-HF), respectively.

p-Values calculated with Wilcoxon Signed Ranks Test.

Discussion

Cardiovascular disease is estimated to affect 60% of people over the age of 60 and triples the incidence of frailty. Especially HF, with the associated decrease in physical fitness and marked reduction in physical activity, places the person at risk for developing frailty. By decreasing the blood supply of muscles, HF causes impaired muscle metabolism, reduced muscular performance, increased muscular catabolism, and leads to sarcopenia and physical frailty (21).

Depending on the assessment tools used, frailty has been identified in 25-62% of patients with cardiovascular disease. These patients are at risk for disability and poor outcome even following “minor” illnesses, which is why early identification and proper management of frailty can reduce morbidity, dependence rates, and improve the quality of life of these patients (2). Among the cardiovascular diseases, mainly HF is associated with frailty in about 50% of patients, regardless of age or functional class of HF, as compared to reported prevalence rates of frailty of 10.7% in the general population (2). Our study found a prevalence of frailty of 30% in HF patients and of 43% for pre-frailty in the same group, which is lower than the rates reported by other studies, possibly related to the high proportion of patients with stage I (26%) and stage II HF (66%). Nonetheless, even in our study, patients without HF had significantly lower prevalence rates of frailty (19%) and pre-frailty (36%).

The literature on frailty is rapidly expanding. A meta-analysis of 26 studies published in 2017, which included 6,896 HF patients, reports a 44.5% prevalence for frailty and 42.9% for physical frailty in HF patients, rates that were not influenced by age or functional class of HF. Another study, published in 2022 and conducted on 406 patients older than 60 years with various stages of HF, reported frailty in 28% of patients and pre-frailty in 40% (22), findings which are in line with the ones of our study. The presence of comorbidities increased the risk of frailty (22). As opposed to other reports, this study showed that the prevalence of frailty was correlated with age and functional class of HF. In hospitalized patients, the prevalence rates may be even higher. For example, Perna et al. (2017) conducted a study on 366 patients hospitalized for various diseases and found that 66.4% of these had pre-frailty, 14% had severe frailty, and only 20% did not exhibit signs of frailty regardless of sex (18). In our study, frailty was absent in 37.5% at the cohort level, and 39% were vulnerable, whereas mild and moderate frailty affected only 20% of the enrolled patients. Severe frailty was identified only in 2.8% of patients. The discrepancies may be explained by the pool of patients from which study participants were enrolled, namely patients presenting hospitalized on the rehabilitation ward, thereby excluding patients with severe, life-threatening acute illnesses.

Physical exercise plays an important role in the primary and secondary prevention of HF (17), the Heart Failure Association guidelines recommending continuous moderate-intensity training or high- and low-intensity interval training (12) for this purpose. A study led by Cattadori (12) showed a 41% lower mortality rate compared to the general male population in a study on 786 cyclists. Another study on 1,873 men supported the beneficial role of physical training on cardio-respiratory fitness and showed a 53% reduction in the risk of developing HF with regular physical exercise (23). Similarly, Kraigher-Krainer et al. showed in a study conducted on 1,142 elderly participants from the Framingham study with a follow up period extending over 10 years, that the risk of developing HF may be lowered by 15-56% through regular physical activity (24). The beneficial effects of physical training have been demonstrated in other conditions, such as spinal cord injuries, as well (25).

Exercise acts both on heart structure and at the cellular and molecular level (26). Obviously, the question as to which kind of exercise would lead to best results has been raised. The study of Ellingsen et al. (2017) showed that high-intensity interval training in chronic HF with reduced ejection fraction (HFrEF) for 12 weeks is not superior to moderate continuous exercise in terms of left ventricular remodeling (27). A study published in 2018 (12) showed that physical activity of 500 metabolic equivalent of oxygen consumption (METs)-min/work will decrease the risk of HF by 10%; activity of 2,000 METs-min/work reduces the risk of HF by 35%. Each 1 MET increase in cardio-respiratory fitness training results in a 21% reduction in the multivariable adjusted risk of HF (23). However, low intensity physical activity, such as Tai Chi training, can also positively affect the quality of life of HF patients (28).

Tools to assess frailty have been subject to criticism, some researchers claiming that unidimensional instruments are oriented mainly towards the physical domain of functioning. Nonetheless, multidimensional tools have been developed in recent years, which analyze various domains of physical, psychological, and social domains of human functioning (29,30), and demonstrated that physical frailty leads to repeated hospitalizations, elevated brain natriuretic peptide levels, the need for inotropic agents, as well as to renal impairment and cognitive decline (19).

Our study showed that 6 months of training (10 days under the supervision of a physical therapist followed by 3 training sessions/week for the remainder of time) is able to decrease the prevalence of frailty. The proportion of patients with no signs of frailty increased by 4.7%, and the percentage of vulnerable patients increased by almost 12% at the expense of patients from the mild and moderate frailty category. Unfortunately, the number of patients with severe frailty remained unchanged.

The VIVIFRAIL project proposes an individualized multicomponent exercise program targeting frail elderly patients with cognitive impairment. It consists of resistance training, gait retraining, and balance training. The intended effects relate to improving gait, balance, and strength. The exercises focus on upper and lower limb muscle strength, increasing coordination and balance to prevent falls (31). Exercises are standardized according to functional ability levels. The question as to whether this design can be applied to patients with frailty and HF and what the impact on ventricular remodeling would be remains to be answered by future studies.

Other important aspects regarding the prevention of frailty are related to the diet, which should be rich in magnesium, given the role of magnesium in myocyte homeostasis (32), regulation of vascular tone, and prevention of cardiac arrhythmias such as atrial fibrillation (33), as well as in proteins, vitamin D and group B vitamins, shown to prevent frailty-associated sarcopenia (34-36).

Strengths and limitations of the study. The novelty of this study is the multidimensional assessment (9 domains) of the impact of HF on physical performance in Romanian patients. The study was designed to assess functional performance using the chair rise, 3 m walk and chair turn test, unlike other studies that use the more user-friendly Handgrip test. This research may be a starting point for the development and implementation of optimal public health policies aimed at preventing frailty and improving physical fitness in HF patients.

The limitations are related to the single center in which these assessments were performed, as well as to the lack of long-term follow-up and lack of data on subsequent readmissions of the enrolled patients on other wards. Furthermore, sarcopenia was not assessed, although frailty often associates sarcopenia (7), which significantly influences the functional performance of an individual.

Conclusion

The prevalence of frailty in patients with HF is significantly higher than that in patients of similar age who do not have HF, and is correlated with age, functional class of HF, as well as with sex (more likely to occur in females). Exercise-based rehabilitation is a useful tool for improving physical performance, decreasing dependence in the daily activities, and improving functional performances in patients with HF.

Conflicts of Interest

The Authors declare no conflicts of interest in relation to this study.

Authors’ Contributions

Conceptualization, D.C.I. and D.C.N.C.; methodology, A.G.N.; software, C.T.G.; validation, B.G.D., A.J. and A.F.B.; formal analysis, C.T.G.; investigation, A.N.P, F.C.B.; resources, D.C.N.C., E.E.B..; data curation, A.G.N., A.J.; writing—original draft preparation, D.C.I. and D.C.N.C.; writing—review and editing, G.S.B.; visualization, B.G.D.; supervision, G.S.B.; project administration, C.D.I. All Authors have read and agreed to the published version of the manuscript.

Acknowledgements

The Authors thank the University of Oradea, considering the logistic facilities used.