Nutritional and inflammation factors associated with current frailty level and effect of co‐morbidities on the progression of frailty

Frailty is defined as extreme vulnerability, a syndrome that exposes the individual to a higher risk of disability. While risk factors for frailty have been gradually uncovered, the full identification of biochemical factors and co‐morbidities influencing frailty remains incomplete.


Introduction
Frailty is defined as a clinical state of decline in physiological systems resulting from age-associated disability. 1 Because individuals with disabilities caused by frailty require long-term treatment or care to maintain their daily lives, frailty poses a significant socioeconomic burden in aging societies.It increases vulnerability to internal and external stressors, thereby elevating the risk of various negative health factors such as falls, fractures, decreased quality of life, and sudden death. 2 A systematic review conducted previously indicated that the pooled prevalence of frailty was about 25% in subjects 80 years or older. 1 Numerous definitions of frailty have been proposed over the past two decades, 1,2 but the one used most extensively is the phenotype of physical frailty developed by Fried et al. using data from the Cardiovascular Health Study (CHS), referred to as the CHS index. 3In this phenotype, meeting three out of the five criteria indicates frailty. 3Recently, Chen et al. developed a simplified frailty check sheet (FRAIL-J) that demonstrates high validity and reliability (reproducibility). 4The checklist has also shown good diagnostic accuracy for assessing frailty using the CHS index.Meeting three or more checklist items has been identified as the cutoff value for frailty screening in Japanese community-dwelling older adults. 4veral previous studies have investigated various factors associated with the development of frailty, among which advanced age stands out as a strong risk factor. 1,2However, frailty is not solely an age-related irreversible process but can be a dynamic phenomenon in which frailty status changes over time. 1,2Therefore, early detection and intervention for frailty are crucial for enhancing the healthy life expectancy of older adults.Assessing the risk of frailty at the individual level is essential to prevent its progression and improve existing frailty.][7][8] Identifying present or future risk factors associated with frailty provides valuable information for the risk stratification of older adults, as such identification enables effective treatment and intervention, particularly when a modifiable risk factor for frailty is identified.
Most older adults have one or more common age-related diseases.These diseases may affect frailty in older adults; however, there are no available data extensively handling the relationship between biochemical indices, co-morbidities, and the progression of frailty.In addition, the patterns of co-morbidities in older adults display marked sex differences, particularly in skeletal disorders.[12]

Participants
1][12] The Nagano Cohort Study was established in 1993 in Nagano Prefecture, Japan.Prior to its initiation, the protocol underwent review by the institutional Ethical Committee in accordance with good clinical practice standards authorized by the Japanese Ministry of Health, Labor, and Welfare.The study was conducted in accordance with the principles outlined in the Declaration of Helsinki.Recently, the protocol of the Nagano Cohort Study underwent re-review by the Ethical Committee at the Research Institute and Practice for Involutional Diseases.Written informed consent was obtained from all participants, affirming that ethical considerations were meticulously addressed in the Nagano Cohort Study.The flowchart of the cohort study and the selection for this analysis are shown in Fig. 1.Patients were asked to confirm their willingness to participate in the Nagano Cohort Study, and a baseline survey was conducted on those who consented; thereafter, information such as frailty status, bone parameters, and incident fractures were followed up periodically.Participants who met the exclusion criteria, such as critical illness, terminal stage malignancy, or longterm steroid use, were excluded from the study.Additionally, individuals with an estimated glomerular filtration rate (eGFR) of less than 20 mL/min/1.73m 2 were also excluded.The study follow-up was concluded when participants passed away or discontinued follow-up owing to relocation, institutionalization, or referral to another hospital, whichever occurred first.The participants who were followed up for more than 1 year were included in the analysis.

Measurements
The parameters for biochemical and anthropometric measurements were recorded as follows.Body weight and height were measured, and the body mass index (BMI) was calculated using the standard method.Serum and urine samples were collected under non-fasting conditions to measure biochemical markers.Serum levels of creatinine, uric acid, albumin, triglycerides, glycated hemoglobin (HbA1c), and total, low-density, and highdensity lipoprotein cholesterol were measured in an in-house laboratory using the dry chemistry method (Fuji Dry-Chemistry Co., Tokyo, Japan).
Previous studies have indicated that incident osteoporotic fractures correlate with chronic inflammatory processes (IL-6 or hsCRP), adiposity (adiponectin or leptin), and advanced glycation end products of bone matrix protein (AGE; pentosidine).4][15][16] Additionally, branched-chain amino acids (BCAA) administration has been shown to enhance muscle performance. 17,18Therefore, measuring levels of these markers could provide valuable insights into frailty assessment.The following biochemical markers were measured by LSI Medience (Tokyo, Japan).Serum leptin levels were measured using a leptin radioimmunoassay kit (Millipore).Serum adiponectin and high-sensitivity C-reactive protein (hCRP) levels were measured using a Latex kit (LSI Medience, Tokyo, Japan) and N-Latex CRP II kit (Cardiophase HsCRP; Siemens Healthcare Diagnostics, Germany), respectively.Serum interleukin-6 (IL-6) levels were determined using an enzyme-linked immuno sorbent assay (ELISA) kit (QuantiGlo Human IL-6 ELISA Kit; Bio-Techne Co., Minneapolis, MN).Urinary excretion of pentosidine was measured by the high performance liquid chromatographymethod and standardized by the urinary creatinine concentration.Serum (BCAA, Leucine, isoleucine, and valine) and tyrosine levels were measured enzymatically using a Diacolor liquid branchedchain amino acids to tyrosine ratio kit (Toyobo Co., Ltd, Osaka, Japan).The manufacturer's data and protocol stated that all of the intra-assay coefficients of variation were <10%.

Diagnosis of co-morbidities
In this analysis, we investigated the relationship between frailty and co-morbidities that occur frequently in outpatient clinics.Lifestyle-related diseases such as diabetes, dyslipidemia, hypertension, osteoporosis, and osteoarthritis were routinely evaluated in this cohort, so these diseases were selected for this analysis.1][12] The diagnostic criteria for these co-morbidities are as follows.DM was diagnosed if the hemoglobin A1c level was 6.5% or more or if patients were actively undergoing treatment for DM.All patients who received insulin treatment were included in the study.All patients with DM in the present study were classified as having type 2 diabetes.Dyslipidemia was defined as having a low-density lipoprotein cholesterol level of 140 mg/dL or more, a high-density lipoprotein-cholesterol level of less than 40 mg/dL, or a postprandial triglycerides level exceeding 200 mg/dL.Hypertension was diagnosed when systolic blood pressure consistently exceeded 140 mmHg systolic pressure consistently exceeded 90 mmHg, or when patients were using antihypertensive medication.
To evaluate skeletal disorders in the participants, the presence or absence of vertebral osteoarthritis (OA) was determined using the Kellgren-Lawrence (KL) grading system, as described by Kellgren and Lawrence. 19To combine multiple joint assessments into a single KL grade, the patients' KL grade was determined based on both the degree and extent of pathological changes.Participants were then classified into two groups: those without vertebral OA (KL grade 0 and 1) and those with vertebral OA (KL grade 2 or higher).Osteoporosis was diagnosed according to the criteria proposed by the Japanese Osteoporosis Society in 2012 20 after assessment of bone mineral densities and prevalent bone fractures.

Frailty score
The FRAIL-J Scale was adapted from the original version, taking into consideration the social and cultural customs specific to Japan.Comparable existing items were utilized for modification. 4mong the five items, fatigue, resistance, and weight loss were obtained from the Kihon Checklist, which was developed by the Japanese Ministry of Health, Labor, and Welfare and is widely used to identify older adults at risk of requiring long-term care. 21mbulation was derived from the Kaigoyobo Checklist, another well-established index for assessing the risk of long-term care. 22railty levels were assessed using frailty scores ranging from 0 to 5. In a cross-sectional analysis, subjects were categorized into two groups: the frail group (frailty score 3 and above) and the non-frail group (frailty score <3).In contrast, for the longitudinal analysis, subjects were classified into a progression group (frailty score increased from baseline by one or more points) and a nonprogression group.

Statistical analysis
The measured background characteristics are presented as mean AE standard deviation (SD) or proportion (%), and the statistical differences between the frail and non-frail groups were assessed using analysis of variance (ANOVA) or the χ 2 test.A multiple regression model was used to estimate the odds ratio (OR) and 95% confidence interval (CI) for the prevalence of frailty, with adjustments for candidate risk factors.Candidate factors were selected from those that were significantly different (P < 0.05) between the frail and the non-frail group and were finally extracted using the stepwise method.
In the longitudinal analysis, candidate factors for frailty progression were selected from those that showed statistically significant differences between the frailty progression and nonprogression groups.A Cox proportional hazards model was used to estimate the hazard ratio (HR) and 95% CIs of frailty progression.All comparisons were two-sided, and P-values <0.05 were considered statistically significant.Data were analyzed using JMP version 16.0 (SAS Institute, Cary, NC).

Results
The characteristics of the participants are presented according to frailty status in Table 1.A total of 1035 participants underwent examination for their baseline anthropometric and biochemical parameters and the prevalence of co-morbidities.A total of 258 participants (25%) were interrupted in the follow-up study for various reasons, including death (n = 75, 7.3%), placement in nursing homes (n = 53, 5.1%), hospital referrals (n = 40, 3.9%), relocation (n = 11, 1.1%), or being lost to follow-up (n = 79, 7.6%).The follow-up data before the interruption were included in the analysis.
A total of 212 participants (20.5%) scored 3 or more points in the frailty assessment and were diagnosed with frailty.When compared with the non-frail group, the frail group exhibited higher age, adiponectin, urinary pentosidine, and serum IL-6.Additionally, the frail group had a higher incidence of hypertension, osteoporosis, and vertebral OA.In contrast, the frail group displayed significantly reduced levels of albumin, total cholesterol, estimated eGFR, and BCAAs.Furthermore, the frail group had a lower occurrence of hyperlipidemia (Table 2).
Odds ratios for the prevalence of frailty determined by multiple regression analysis are shown in Table 2.The analysis revealed that age and log IL-6 and BCAA levels were independently associated with the prevalence of frailty.Risk factors for frailty © 2024 Japan Geriatrics Society.
| 525 In the longitudinal study, the participants were followed up for 7.7 AE 5.9 years.During the observation period, 130 participants (12.6%) showed an increase in frailty scores by one or more points, with the remaining 905 participants assigned to the group without progression.The baseline characteristics of the two groups were compared (Table 3).The participants in the progression group were significantly older and had a higher prevalence of osteoporosis and vertebral OA at baseline than those in the non-progression group.Additionally, those in the progression group demonstrated a significantly lower eGFR and prevalence of hyperlipidemia.However, because the ages of both groups were significantly different, the difference in eGFR may have been reduced by the difference in age.
The HRs for the progression of frailty according to the Cox hazard model are shown in Table 4. Hyperlipidemia and baseline frailty scores were identified as phenotypes that significantly contributed to a decrease in frailty score.Conversely, age and the presence of osteoporosis at baseline contributed to an increase in future frailty scores.

Discussion
Numerous studies have reported associations between various biomarkers and frailty. 23,24The present study additionally discovered significant links between blood markers, co-morbidity, and the prevalence of frailty.In this cross-sectional study, we have presented biochemical parameters linked to baseline frailty, including age, albumin, total cholesterol, eGFR, adiponectin, pentosidine log-transformed IL-6, and BCAA (Table 1).In addition to these biochemical parameters, the presence of hyperlipidemia at baseline was lower in frail participants.In contrast, the prevalence of hypertension, osteoporosis, and OA was higher in the frail group than in the non-frail group.
Notably, consistent with other meta-analyses, in our study age emerged as a major factor associated with the development and exacerbation of frailty. 1,2Because eGFR was calculated using   T Urano et al.
serum creatinine and age, its association with frailty might be accounted for by age-related changes.In fact, eGFR was not significantly associated with frailty after adjusting for age (Table 2).The urinary excretion of pentosidine, a marker of advanced glycation end products, was significantly associated with frailty.
Because the level of pentosidine is known to increase with advancing age, 16 this marker also did not have a significant relationship with the frailty score after adjustment for age (Table 2).
In addition to age, patients with frailty had significantly higher serum adiponectin and IL-6 levels than those without frailty.Conversely, the serum levels of BCAA were lower in frail individuals than in non-frail ones (Table 1).After adjusting for age, serum levels of IL-6 (P < 0.001) and BCAA (P = 0.011) were significantly associated with frailty.Moreover, in the multivariate regression analysis (Table 2), serum adiponectin levels tended to be associated with frailty (P = 0.085).
Elevated IL-6 levels have been suggested to predict the risk of muscle weakness and disability, 25 which makes this finding noteworthy.Recently, Kistner et al. proposed that IL-6 modulates energy allocation in response to metabolic stress in various tissues, leading to disparate actions under different circumstances. 26The tight relationship between the baseline frailty score and its progression may indicate that early intervention or prevention of frailty is particularly important.
In a cross-sectional study, high plasma levels of adiponectin and IL-6 demonstrated a positive relationship with frailty after adjusting for age. 27,28Adiponectin, secreted from adipocytes, serves multiple functions, including increasing insulin sensitivity and protecting against atherosclerosis, 29 and it is associated with a higher incidence of osteoporotic fractures. 15Notably, adiponectin plays an essential role in weight reduction, which is a key phenotype of frailty. 30Although the longitudinal study indicated that the adiponectin level in frailty-progression subjects tended to be higher than in those without frailty (OR: 1.030, 95% CI: 0.996-1.065,P = 0.085), further investigation is needed to obtain conclusive evidence.
The other nutritional factor, BCAA, showed a significant negative association with frailty after adjusting for other factors (Table 2).To our knowledge, this is the first report that reveals a correlation of lower serum BCAA levels with higher frailty scores.This observation aligns with a similar trend reported in a Chinese longevity county study. 31It is important to note that higher circulating BCAA levels are strongly associated with obesity, insulin resistance, and type 2 diabetes in humans.Despite this association, BCAA supplementation in healthy individuals is linked to muscle growth and has beneficial effects on energy expenditure. 17,18Although a strong relationship between the baseline BCAA and frailty was found, the BCAA at baseline was not associated with frailty progression (Table 3).
Here, we have demonstrated for the first time that the presence of osteoporosis is a risk factor associated with the progression of frailty in a longitudinal study.A previous study demonstrated that prevalent vertebral deformities in older women with low bone mass are associated with increased risks of mortality and hospitalization. 32These data suggest that the presence of osteoporosis is an important cause of worsening pathological aging.We also demonstrated that the absence of hyperlipidemia was a significant risk factors for frailty progression.A prior study similarly identified lower serum cholesterol levels as an independent marker for frailty.Consequently, the findings of our present study align with those of the earlier research. 33oth cross-sectional and longitudinal studies have consistently highlighted the robust relationship between aging, current frailty, and frailty progression.However, it is important to recognize that frailty is not an irreversible process and that suitable interventions have the potential to modify the current and future status of frailty.Therefore, the maintenance of adequate levels of cholesterol and the prevention of osteoporosis may serve as a protective measure against frailty progression.However, further investigations are warranted to validate these findings.
This study has certain limitations.First, the participants were drawn from a subset of the Nagano Cohort Study comprising only postmenopausal women attending primary care institutions.This may introduce a potential selection bias, as the study participants may not fully represent the broader population.However, it is worth noting that a previous study reported no significant differences in propensity scores between participants and a community-dwelling sample population in Japan, 34 indicating that selection bias is likely to be minimal.Moreover, 20% of the current study population were frail at baseline, a proportion comparable to that found in a previous study involving the general population.This similarity in frailty prevalence further supports the argument that the likelihood of selection bias is low.Approximately 25% of participants in the study were disrupted in the follow-up for various reasons.Within this group, it is noteworthy that 5% of subjects transitioned to nursing homes, likely owing to disability.Consequently, the progression rate of frailty in this study may potentially lead to an underestimation of its true impact.The present study was limited by the absence of male subjects, mainly due to the anticipated differences in the proportions of co-morbidities, particularly skeletal diseases.Factors associated with frailty are known to be related not only to biological issues, but also to social factors and lifestyle habits.Although this study focused on the biological perspective, it will be necessary to include factors such as exercise habits, social participation, and polypharmacy as confounding factors in the future.Hence, there is a pressing requirement for further research in order to achieve a more comprehensive understanding of this topic.
Inflammatory and nutritional markers, IL-6 and BCAA were significantly associated with current frailty status, and hyperlipidemia and osteoporosis acted as negative and positive predictors, respectively, of future frailty progression.The present study may provide key concepts to improve the current frailty and to prevent future frailty.

Figure 1
Figure 1 Flowchart of the study and selection for this analysis.

Table 2
Multiple regression analysis for the baseline frailty score and baseline data

Table 3
Baseline data between participants with and without progression in the frailty score BMI, body mass index; HbA1c, hemoglobin A1c; eGFR, estimated glomerular filtration rate; hCRP, high-sensitivity C-reactive protein; IL, interleukin; BCAA, branched-chain amino acid.

Table 4
Cox proportional hazard model for the progression of the frailty score