Introduction

Frailty is defined as a state of increased vulnerability resulting from age-associated declines in reserve and the functions of multiple physiological systems, such that the ability to cope with every day or acute stressors is compromised [1]. The morbidity of frailty of Japanese community-dwelling elderly aged 65 years or older is 10–30% [2, 3], and the prevalence increases with increasing age [4, 5]. Moreover, frailty independently predicts incident falls, worsening mobility or activities of daily living, disability, hospitalization, and death [4, 5]. In Japan, a super aging society, more than one-third of the population is considered elderly, and thus, the increase in frailty is not only a personal issue but also a public issue. In community-dwelling elderly people, cardiovascular (CV) disease is associated with an odds ratio (OR) of 3–4 for prevalent frailty and 1.5 for incident frailty [6]. Moreover, patients with severe coronary artery disease or heart failure are likely to have comorbid frailty, which is associated with a high risk of all-cause mortality [6].

Management of the CV risk factors in these patients is crucial to improve prognosis. Our understanding of the association between frailty and CV risk factors or risk clustering, however, is limited. Population-based studies indicate that frailty is associated with an increased risk of CV risk factors independent of established CV disease [7,8,9,10]. Among hospitalized elderly people, the association between frailty and CV risk factors is equivalent [11]. A community-based study of subjects aged 90 years or over demonstrated that frailty, but not metabolic syndrome, increases mortality risk [12]. In very old patients, on the other hand, low body mass index (BMI), low blood pressure (BP), and low total and high-density lipoprotein cholesterol strongly predict total mortality [13]. There are currently no guidelines for the management of CV risk factors in this spectrum of subjects, and no optimal levels have been established. Given this background, we evaluated the CV risk profile of patients with frailty in an outpatient setting.

Methods

Study patients

The Nambu Cohort Study is a prospective cohort study of elderly patients in an outpatient setting in the southern area of Okinawa, Japan, that began in 2017. The aim of the study is to collect data regarding the prolongation of a healthy life expectancy to evaluate the clinical significance of frailty. Patients 65 years of age or older treated for CV disease, such as hypertension, diabetes, dyslipidemia, coronary artery disease (CAD), cerebrovascular disease, and heart failure, were registered at each institute. Patients who could complete the interview by the medical assistant to provide information regarding frailty were continuously registered. Among them, patients <65 years old and those with Kihon Checklist score of 5 (the worst) in the activities of daily living and physical function section were excluded. Finally, a total of 599 patients were analyzed (Fig. 1). We performed a cross-sectional analysis to evaluate the CV risk profile and comorbidities in these patients and analyzed the effect of frailty using baseline data from the Nambu Cohort Study.

Fig. 1
figure 1

Diagram of the Nambu Cohort Study

Full size image

Data collection

Blood pressure and resting heart rate were recorded using an automatic blood pressure monitor (HBP-9020, Omron Corp. Kyoto, Japan) after having the subject sit for 10 min. Body weight and height were measured to the nearest 0.1 kg and 0.1 cm, respectively, with the subjects wearing light indoor clothing and no shoes. Body mass index was calculated as the weight in kilograms divided by the square of the height in meters (kg/m2). All blood samples were obtained from the antecubital vein after an overnight fast. Fasting plasma glucose was measured using the hexokinase/glucose-6-phosphate dehydrogenase method. Serum creatinine, total cholesterol, triglycerides, and low-density lipoprotein cholesterol (LDL-C) levels were measured by enzymatic methods. High-density lipoprotein cholesterol was measured by a direct method. The hemoglobin A1c level was measured by high-performance liquid chromatography. Leukocyte counts, hematocrit, hemoglobin level, and platelet count were quantified using an automated blood cell counter. Grip strength was measured with a Smedley-type (mechanical) handgrip dynamometer (TTM, Tokyo, Japan).

Hypertension was defined as systolic blood pressure (SBP) ≥ 140 mmHg, diastolic blood pressure ≥90 mmHg, or antihypertensive drug use. Obesity was defined as BMI ≥ 25 kg/m2. Diabetes mellitus was defined as a fasting plasma glucose concentration ≥ 126 mg/dl (7 mmol/L), hemoglobin A1c ≥ 6.5%, or taking an antidiabetic drug. Dyslipidemia was defined as LDL-C ≥ 140 mg/dl (5.69 mmol/L) if patients did not have CAD and ≥100 mg/dl if patients did have CAD or used statins. CAD was defined as angiographically documented significant coronary stenosis, history of myocardial infarction, history of percutaneous coronary intervention, or coronary artery bypass graft. Stroke was defined as cerebral thromboembolism and cerebral hemorrhage documented by computed tomography or magnetic resonance imaging. Chronic kidney disease was defined as an estimated glomerular filtration rate <60 ml/min/1.73m2. Chronic heart failure was defined as a history of congestive heart failure or brain natriuretic peptide levels of 100 pg/ml or more [14]. Peripheral artery disease was defined as an ankle brachial index < 0.9 [15] or patients undergoing percutaneous transarterial angioplasty. Cardiovascular risk factors included hypertension, diabetes, dyslipidemia, and obesity. Optimal CV risk control levels were defined as BP < 140/90 mmHg, BMI < 25 kg/m2, and LDL-C < 140 if the patients did not have CAD and <100 mg/dl if the patients had CAD or used statins.

Frailty was diagnosed using the Kihon Checklist. The Kihon Checklist is a simple self-reporting yes/no survey comprising 25 questions regarding instrumental (3 questions) and social (4 questions) activities of daily living, physical functions (5 questions), nutritional status (2 questions), oral function (3 questions), cognitive function (3 questions), and depressive mood (5 questions). This comprehensive questionnaire assesses the physical, psychological, functional, and social status of nondisabled older adults in multiple domains. The usefulness of the Kihon Checklist as an index of frailty was verified by Satake et al. [16]. A Kihon Checklist score of 0–3 was considered to indicate non-frailty, 4–7 pre-frailty, and ≥8 was considered to indicate frailty [16]. Well-trained medical assistants asked the patients about the content of the Kihon Checklist before the medical examination and recorded the patients’ responses.

Statistical analysis and ethics

The clinical characteristics of the study population stratified by frailty were compared using a one-way analysis of variance or Kruskal–Wallis test for continuous variables according to the normality of the distributions and chi-square test for categorical variables. The distribution normality was evaluated using the Kolmogorov–Smirnov test. Continuous data are presented as medians (interquartile ranges), and categorical data are presented as frequencies. The odds ratios and 95% confidence intervals for frailty were calculated after adjusting for confounding variables using multiple logistic regression analysis. Statistical analyses were performed using JMP 9.0.2 (SAS Institute Inc., Cary, NC, USA). All statistical tests were two-sided, and p < 0.05 was considered statistically significant. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and approved by the Institutional Review Board of Social Medical Corporation Yuaikai, Okinawa, Japan.

Results

A total of 599 patients with a median age of 78 (70–83) years, 50% of whom were men, were evaluated. A total of 178 patients (30%) were considered non-frail, 197 patients (33%) were diagnosed as pre-frail, and 224 patients (37%) were diagnosed as frail. The baseline characteristics of the patients stratified by frailty status are shown in Table 1. Patients with frailty were significantly older, had a lower level of handgrip strength, and were more likely to have CV disease. The frequency of patients with multiple CV risk factors tended to decrease in frail patients, but the effect was not statistically significant. The CV risk levels of these patients according to frailty level, however, were in the opposite direction. Frail patients were likely to have lower BP, BMI, and serum lipid levels. Plasma glucose and hemoglobin A1c levels were not significantly different regardless of frailty. Logistic regression analysis was performed to evaluate the effect of CV risk factors on frailty. Frailty was significantly associated with SBP, marginally associated with LDL-C, and not associated with BMI. The odds ratio (OR; 95% confidence interval [CI]) of a unit increases in each risk factor was as follows: SBP (each 10 mmHg increase) 0.83 (0.72–0.95, p = 0.008); LDL-C (each 10 mg/dl increase) 0.96 (0.86–1.05, p = 0.301); and BMI (each 1 kg/m2 increase) 1.03 (0.97–1.10, p = 0.306) (Fig. 2). Finally, we evaluated the association between the number of CV risk factors within the optimal level according to the clinical guidelines and frailty (Table 2). The ORs (95% CIs) for frailty of patients with one, two, and three CV risk factors within the optimal level were 2.30 (0.75–8.69, p = 0.153), 3.22 (1.07–11.79, p = 0.038), and 4.79 (1.56–18.05, p = 0.005), respectively, compared with patients without a CV risk factor within the optimal level after adjusting for confounding factors. We also performed a sensitivity analysis using zero and one CV risk factor within the optimal level (n = 182) as the reference group and showed similar results (adjusted OR 1.54, 95% CI 0.95–2.51, p = 0.077 for two and OR 2.30, 95% CI 1.36–3.92, p = 0.002 for three CV risk factors within the optimal level).

Table 1 Baseline characteristics of 599 patients in the Nambu Cohort Study stratified by frailty
Full size table
Fig. 2
figure 2

Adjusted odds ratio and 95% confidence interval for frailty of 599 patients in the Nambu Cohort Study. Data were adjusted for age, sex, handgrip, blood pressure, and lipid-lowering medication, such as calcium channel blockers, angiotensin receptor blockers, and statin use, and comorbidity, such as coronary artery disease, stroke, chronic kidney disease, and atrial fibrillation

Full size image
Table 2 Odds ratios and 95% confidence intervals for frailty stratified by the number of cardiometabolic risk factors within optimal levels in 599 patients of Nambu Cohort Study
Full size table

Discussion

The results of the present analysis demonstrated that frailty is significantly associated with lower BP and a tendency toward a lower LDL-C but not with BMI levels. More interestingly, a higher number of CV risk factors within the optimal level accelerates the risk of frailty. To the best of our knowledge, these findings are the first to demonstrate an association between CV risk level and the accumulation of CV risk factors with frailty.

Cardiovascular risk factors in frailty

The evidence demonstrating an association between CV risk factors and frailty is limited. A population-based study demonstrated an association between frailty and CV risk factors independent of established CV disease [7], especially diabetes [8]. Moreover, the prevalence of frailty was higher in those with prior CV disease, and combined CV risk factors were linearly and positively associated with frailty [9]. An association between CV risk and frailty was also found in hospital patients [11]. Subjects with frailty are at risk for all-cause death or a CV event [17, 18]. The difference between the previous results and ours might derive from the characteristics of the cohort. Our cohort was significantly older, had a lower BMI, and, more importantly, had more CV comorbidities, especially a higher prevalence of stroke (Table 1). The higher prevalence of stroke in our cohort might be the main reason for the higher prevalence of frailty compared with the other cohorts. These characteristics of our subjects accelerate biological aging and lead to the catabolic syndrome that is often observed in very old subjects [19].

Lower SBP levels are associated with poor prognosis in community-dwelling elderly persons [20, 21] and persons with frailty [22,23,24]. Both cholesterol and BMI levels are also inversely associated with prognosis in elderly subjects [25, 26], and a 1 mmol/L (38.6 mg/dl) increase in total cholesterol is associated with an 18% reduction in mortality [26]. Elderly subjects with a BMI > 30 had the lowest mortality during a 5-year follow-up [27]. The relative risk of death associated with excess adiposity is lower for older adults than for younger adults [28]. Not only the baseline levels of CV risk factors but also a serial decline in these factors is crucial for patient prognosis. The trajectory of BP, cholesterol, and BMI in elderly subjects gradually declines by the time of death [29,30,31]. These time-dependent changes in CV risk factor levels might be associated with a physical status change from robust to frail, to end of life, and finally death and indicate that homeostasis is disrupted in frail subjects, continuing to the end of life [32]. In elderly patients, these CV risk factors may still be hazardous, but the direction is totally the opposite, a phenomenon known as “reverse epidemiology” or “reverse metabolic syndrome” [19, 33].

Mechanisms of reverse metabolic syndrome

Aging, usually referring to chronological age, is a powerful and independent risk factor for prognosis. Frailty is a state, in which homeostasis against stress is impaired and is considered to accelerate biological aging [32]. Frailty might bridge the gap between chronological aging and biological aging. A crucial threshold of age-related cumulative decline, beyond which frailty becomes evident, exists in many physiological systems. Among several possible mechanisms that contribute to the progression of frailty, age-related low-grade chronic inflammation may contribute to muscle wasting in sarcopenia through the upregulation of proinflammatory cytokines [34]. Subjects with frailty may have an imbalance in muscle homeostasis leading to accelerated muscle breakdown. Inflammatory cytokines activate muscle breakdown to generate amino acids for energy and cleave antigenic peptides [34], referred to as catabolic syndrome [19]. This circumstance also leads to hemodynamic instability, resulting in decreased BP, as found in our study. BMI may not reflect muscle loss caused by frailty because the muscle mass loss observed in frail elderly individuals could be masked by age-related changes in body composition [35]. In fact, our results indicated no association between BMI and frailty, which supports this phenomenon.

A strength of the present study is that it confirms the association between the favorable CV risk profile and frailty. Frailty is a systemic phenomenon in elderly patients with disrupted homeostasis. Physicians should pay more attention to the levels of CV risk factors—not only their increase but also their decline. An unexpected decline in the risk factor level might also indicate frailty. At the present time, the optimal BP/lipid level for subjects with frailty is not clear. In elderly patients, frailty may be a useful indicator for screening subjects for whom “lower is better” and subjects for whom excessive control is a risk factor.

Limitations

This study has some limitations. First, patients who required assistance to walk and nursing home residents were part of the population included in the original database used for this study. To classify all the patients as ambulatory, we excluded those patients with a Kihon Checklist score of 5 (the worst) in the physical function section. Simply excluding those with a score of 5, however, might not provide an accurate representation of ambulatory patients. Second, patients in this study were recruited from a hospital or clinic, and most had multiple comorbidities. For these reasons, the patients in this study were quite different from the general population. Third, the patients were residents of a relatively limited region—the southern part of Okinawa, Japan. Although the results of the present analysis are consistent with previous findings, caution should be taken in generalizing the results. Lastly, this was a cross-sectional observational study with the inherent limitations and biases of such studies. It is unclear whether the optimal risk level leads to undesired outcomes. A prospective study to evaluate these issues is warranted.

Conclusion

A favorable CV risk profile was associated with frailty in an outpatient setting.