Association of clinical parameters of sarcopenia and cognitive impairment in older people: cross-sectional study

Objective: To associate clinical parameters of sarcopenia with cognitive impairment in older people. Method: Cross-sectional study with 263 older adults (≥60 years) treated at a specialized public health facility. Sociodemographic and clinical variables were used to characterize the sample and the clinical parameters of sarcopenia (muscle strength, muscle mass and physical performance) were assessed based on handgrip strength (HGS), calf circumference (CC) and the Timed Up and Go (TUG) test. The Mini-Mental State Examination (MMSE) was used to evaluate cognitive status. Associations were analyzed by simple and multiple linear and logistic regression considering the clinical parameters of sarcopenia (independent variables) and cognitive status (dependent variable), adjusted for age, sex, years of schooling, number of medications, nutritional status and functional capacity. Results: Of participants with cognitive impairment, 59.6% exhibited low muscle strength. In simple linear regression, cognitive status was explained by muscle strength in 21.5% of cases, muscle mass in 12.3% and physical performance in 7.6%, with muscle strength and muscle mass as explanatory variables for cognitive status in non-adjusted multiple regression and muscle strength alone for adjusted analyses. Only muscle strength remained significantly associated with cognitive status in adjusted multiple logistic regression (OR=0.846; [95%CI: 0.774 – 0.924] p <0.001). Conclusion: Low muscle strength was the sarcopenia parameter independently associated with cognitive impairment. This information is useful in highlighting the likelihood of cognitive impairment when poor muscle strength is identified in older people.


INTRODUCTION
Sarcopenia and cognitive impairment are worrisome issues related to aging and public health due to the high risk of functional disability, hospitalization and death 1,2 . In community-dwelling older people, sarcopenia has a global prevalence ranging from 10 to 27%, with a progressive increase with advancing age and in rehabilitation units 3 .
Cognitive impairment, on the other hand, has been experiencing an exponential increase, with estimates that 65 million older people will have dementia worldwide by 2030, also with higher prevalence in advanced age 4 .
Sarcopenia is a disease that causes progressive loss of strength and muscle mass in older people 5,6 . It can be explained by the interaction of multiple risk factors, in particular, aging itself, with a reduction in cell metabolism and hormones that participate in myogenesis 5 . Furthermore, the presence of comorbidities, sedentary lifestyle, poor diet and bad habits are risk factors that are involved in the production of high concentrations of inflammatory cytokines, causing apoptosis 6 and reduction in the structure and function of systems involved in both sarcopenia and cognitive impairment 7 , with the possibility of coexistence of both 8,9 . Sarcopenia has been shown to be associated with cognitive impairment [8][9][10] . A systematic review identified a higher prevalence of sarcopenia in participants with cognitive impairment in most of the included studies 8 . It has also been noted that the concomitant presence of sarcopenia and cognitive impairment is a common finding 9 . In addition, researchers have specifically pointed out physical performance and muscle strength as clinical parameters of sarcopenia independently associated with cognitive impairment 11,12 . However, results for muscle mass measurements so far are inconsistent [12][13][14][15] . Additionally, there are still uncertainties about this interaction due to the great methodological heterogeneity of the studies 8,12-16 . In this context, the objective of the study was to associate the clinical parameters of sarcopenia with cognitive impairment in older people who use a public specialized care service. The results of this study will contribute to elucidate which clinical parameters of sarcopenia are associated with cognitive impairment. Considering that the clinical parameters of sarcopenia are modifiable outcomes 17 , with this information, the entire multidisciplinary team will be able to better direct the investigation for the screening of older people at risk of cognitive impairment and implement interventions aimed at its primary and secondary prevention 8 . Participants were 281 older people selected for convenience and assessed in a public specialized care service in the western health region of the Federal District (DF) between the years 2020 and 2021. This specialized care service is composed of a multidisciplinary team that performs a multidimensional assessment of older people who require geriatric care referred by primary care services. Older people who are 80 years old or older are assisted, regardless of complaint or health condition, and older people under 80 years old who have at least one of the following criteria: dependence on basic activities of daily living; cognitive disability; parkinsonism; urinary or fecal incontinence; partial or total immobility; postural instability, falls or lowimpact fractures; polypathology; polypharmacy and clinical decompensations or frequent hospitalizations. In the present study, older people were included and those with missing data on the Mini-Mental State Examination (MMSE) and/or on the three assessments of the clinical parameters of sarcopenia (muscle strength, muscle mass and physical performance) were excluded.

METHODS
The sample size required for analyzing the variables in this study was estimated by performing a sample calculation based on the odds ratio (OR) value found in a meta-analysis of the association between sarcopenia and cognitive status (OR=2.926 [2.297-3.728]) 9 . Using the Logistic Regression Test and considering an OR of 2.926, a power of 80% and an alpha error of 0.05, it was estimated that a sample size of 138 older people would be sufficient to identify the investigated associations.
Initially, the older people were assessed to collect sociodemographic variables such as age (in complete years), gender (female or male) and education (in years of study). These data were collected through a form prepared by the researchers.
Then, the following clinical data were collected: nutritional status (by means of the Body Mass Index -BMI), amount of continuous use medication (checked by means of a medical prescription), practice of physical exercise, depressive symptoms and functional capacity. Based on the BMI data, the participants were grouped into underweight (BMI<22 Kg/m²), eutrophic (BMI 22-27 Kg/ m²) and overweight (BMI>27 Kg/m²) 18 . Regular physical exercise was considered to be those lasting at least 150 minutes per week of moderate-intensity activity 19 , with participants categorized as active or inactive. Depressive symptoms were assessed using the Geriatric Depression Scale (GDS- 15), and participants were categorized into severe depression (≥11 points), with depressive symptoms (from 6 to 10 points) or without depressive symptoms (<6 points) 20 . Functional capacity was assessed using the Pfeffer questionnaire 21 for the older people with cognitive impairment and the Lawton and Brody scale 22 for the older people without cognitive impairment. Older people who scored between 6 and 30 on the Pfeffer questionnaire 21 and between 7 and 20 on the Lawton and Brody scale 22 were considered dependent. This information was self-reported by the older person and confirmed by the companion.
Cognitive status was assessed using the MMSE and cognitive impairment defined as a score below the recommended level, according to education level. Participants with more than 7 years of schooling who totaled <28 points, between 4 and 7 years of schooling who totaled <24 points, between 1 and 3 years of schooling who totaled <23 points and illiterates who totaled <19 points were classified as having cognitive impairment 23 .
The clinical parameters of sarcopenia were evaluated and defined according to Cruz-Jentoft et al. 5 . A Saehan® manual hydraulic dynamometer (Saehan Corporation, 973, Yangdeok-Dong, Masan, Korea) was used to obtain muscle strength through handgrip strength (HGS). It is a valid instrument with excellent test-retest reliability for use in older people with questionable to moderate dementia 24 . The collection took place in the dominant upper limb, with the older person sitting, elbow flexed at 90°, forearm in a neutral position, thumb up and feet flat on the floor. Considering the average of three attempts 25 , muscle weakness was identified for values <27 Kgf for men and <16 Kgf for women 5 .
Muscle mass was obtained by measuring the calf circumference (CC), using a non-elastic measuring tape, with the older person sitting, legs and ankles positioned at 90º, measuring the circumference of the largest diameter of the dominant leg. Measurements smaller than 31 centimeters (cm) characterized low muscle mass 5,13 .
The Timed Up and Go Test (TUG) was used as a measure of physical performance. The participant got up from an armless chair, walked a distance of three meters at their usual pace, turned 180 degrees and returned the same way until they sat down again. The execution time of the test was timed and those who performed the TUG in ≥20 seconds were considered to have low physical performance 5 .
Descriptive analyzes (mean, median, standard deviation, 25 and 75 percentiles, absolute frequency and percentage) were performed with data on sample characteristics and sarcopenia parameters. Data distribution was investigated using the Kolmogorov Smirnov Test. Independent Student's t-test (parametric numerical data), U Mann Whitney (non-parametric numerical data) or chi-square test (categorical data) were used to compare sociodemographic, anthropometric, clinical and functional capacity variables including sarcopenia parameters between groups with and without cognitive impairment.
Quantitative measures of sarcopenia parameters were included in simple linear regression analysis with the aim of investigating the existence of a relationship with the output variable (cognitive status -MMSE score). Additionally, multiple linear regression analysis was performed including the three sarcopenia parameters as independent variables and the MMSE score as a dependent variable. This analysis was carried out with the aim of determining whether the individual importance of these parameters was maintained to explain possible variations in the MMSE score when combined with the others. Next, a multiple linear regression analysis was performed, adjusted for possible confounding variables: sex, years of study, number of medications, nutritional status, practice of physical exercise and functional capacity.
Simple binary logistic regressions were performed between each of the quantitative sarcopenia parameters (independent variables) and cognitive status (dependent variable). Additionally, a multiple logistic regression analysis including the three quantitative sarcopenia parameters was performed to investigate the joint association of these factors with cognitive status. Then, a multiple logistic regression analysis was performed, adjusted for possible confounding variables: age, sex, years of study, number of medications, nutritional status, practice of physical exercise and functional capacity.
In multiple linear regression analyses, variables not identified as predictors were removed and the model with the highest adjusted R² value or that explained a greater percentage of the output variable was presented. Odds Ratios (ORs) with 95% confidence intervals and Beta were calculated for each independent variable. For each linear and logistic regression analysis, the principles of independence between residuals were respected (Durbin-Watson), normality of the residuals, presence of homoscedasticity, absence of multicollinearity between the variables (VIF <10 and Tolerance >0.1), minimum number of cases in each variable and, therefore, guaranteed the assumptions for carrying out the regression by the stepwise-forward method. No imputations were performed for missing data. In cases of participants with missing data, the data were analyzed using pairwise exclusion, so that available data could be included in the analyses.
Cohen f values were calculated as a measure of effect size from linear regression and results were interpreted as small (>0.02), medium (>0.15), and large (>0.35) for f 26 . A significance level of 5% was considered.

RESULTS
In total, 263 older people were included in the study, of which 234 (89%) had cognitive impairment, as shown in Figure 1.
Study participants were aged between 60 and 98 years, mostly women, with low education, inactive, overweight, depressive symptoms and functional dependence. The characterization of the sample is represented in Table 1.
The comparison of sarcopenia parameters and the diagnosis between older people with and without cognitive impairment is shown in    It was observed that the cognitive status was explained by muscle strength in 21.5%, muscle mass in 12.3% and physical performance in 7.6%. Multiple analysis including the three sarcopenia parameters resulted in a statistically significant model  Table 3.
The results of the simple and multiple binary logistic regression analyzes are presented in Table 4. The simple analyzes showed that muscle strength (in KgF), muscle mass (in cm) and physical performance (in seconds) were associated with the cognitive state. In the multiple analysis including the three quantitative sarcopenia parameters, muscle strength (in KgF) and muscle mass maintained an association with cognitive status ( p=0.005 and p=0.038, respectively). In the multiple logistic regression analysis with the three quantitative parameters of sarcopenia adjusted for covariates, only muscle strength, adjusted for gender ( p=0.018) remained significantly associated with cognitive status.

DISCUSSION
This study associated the clinical parameters of sarcopenia (strength, muscle mass and physical performance) with cognitive impairment in older people who use a public specialized care service. Although the three parameters were associated with cognitive status, in the multiple adjusted analyses, only low muscle strength remained independently associated with cognitive impairment.
It was observed that 59.6% of older people in the group with cognitive impairment had low muscle strength, with a significant difference in HGS between groups. HGS explained by 21.5% the cognitive status presented by the older people in the MMSE and 1 KgF of HGS more reduced by 15.4% the chance of the older person to present cognitive impairment at the time of assessment. Our results are consistent with previous studies that identified an association between muscle strength and cognitive status, which demonstrated that low muscle strength almost doubles the risk for cognitive impairment 12,[27][28][29] and that HGS suffers a greater reduction in the simultaneous presence of cognitive and physical impairment 27 . The main mechanism that explains this relationship is the sharing of pathophysiological pathways, involving oxidative stress and chronic inflammation, resulting from aging, physical inactivity, increased visceral fat and chronic diseases 6,7,30 . These factors lead to a metabolic imbalance with the activation of inflammatory pathways, which produce oxidative damage on muscle cells and brain structures 30 . However, it has been discussed that the strength of this association depends on the cognitive tool used 27 , the cutoff points to identify muscle weakness and the different HGS measurement protocols, which interfere with its reproducibility 8,25 .
In the adjusted analyses, we found no association between low muscle mass, represented by calf circumference, and cognitive impairment. Previous investigations also showed that the individual effect of low muscle mass was not significant for cognitive impairment, even when measured by more specific measurement instruments, corroborating our findings 11,27,31 . However, on the contrary, data indicate an association of low muscle mass (measured by bioimpedance) with specific cognitive domains 31 and that calf circumference was a significant predictor of cognitive impairment using several cognitive tests in a cross-sectional analysis 15 . Meta-analyses 8,9 have argued that the inconsistencies in the association between muscle mass and cognitive impairment may be related to different body composition measurement devices. It has been suggested that not muscle mass, but adipose tissue may be directly related to cognitive impairment 32 . This is because metabolically adipocytes actively participate in the central nervous system, altering insulin sensitivity, responsible for synaptic failure, brain atrophy and cognitive decline, so that infiltration of adipose tissue macrophages causes the activation of a network of inflammatory pathways that results in in apoptosis 30,32 .
Physical performance is a third parameter described by Cruz-Jentoft et al. 5 and classifies the severity of sarcopenia. Analyzing the execution time (in seconds) of the TUG, we found that the older people with cognitive impairment took longer to execute the TUG. However, in the adjusted analyses, physical performance was not associated with cognitive impairment. Evaluating physical performance as a sarcopenia parameter also lacks standardization of the tool used and measurement protocols for better reproducibility8. We used the TUG, but the physical performance measure most used in the literature was gait speed, which has been shown to contribute to more than doubling the risk of cognitive impairment 12,16 . Kubicki 33 justifies this predilection for gait speed due to the fact that the test execution commands are simpler than those of the TUG and, in addition, are subject to less measurement bias. Previous studies have found an association between physical performance when assessed using gait speed 12,16,30 and the Short Physical Performance Battery tool 11 , noting that cognitive impairment is not only associated with, but is preceded by, a reduction in physical function 29,34 . This relationship is not so clear, but it is known that inflammatory markers, hormones, insulin resistance and oxidative stress are negatively correlated with muscle strength, physical performance and cognitive function 11,30,35 .
As a strong point, this study used instruments and cutoff points recommended by consensus to measure muscle strength, muscle mass and physical performance of the older people 5 , thus facilitating its reproducibility. However, some limitations can be listed. Due to the cross-sectional design of the study, it was not possible to identify a causal relationship between sarcopenia parameters and cognitive impairment. Considering that the studied population had a medical condition that could affect their ability to self-report, sociodemographic and clinical information was confirmed with the respective caregivers, usually a family member or a trained professional. There was a high prevalence of cognitive impairment in our sample, because the participants were older people referred by primary care services, most of them with complaints of functional dependence and cognitive impairment, and, because of this, this limitation could not have been avoided. We also use calf circumference to measure muscle mass and this tool has been questioned. However, a strong correlation between calf circumference and skeletal muscle mass index was previously observed in both men (r=0.78) and women (r=0.75) and circumference measurement was inversely associated with sarcopenia in both genders (men: OR= 0.62; 95%CI: 0.56 -0.69 and women: OR= 0.71; 95%CI: 0.65-0.78) 13 . In addition, this measure is considered low-cost and easy to measure, and can be used in environments with limited resources, making it a viable measure for use in older people with cognitive impairment 5 .

CONCLUSION
Low muscle strength was the sarcopenia parameter independently associated with cognitive impairment. This information is useful for the team involved in the multidisciplinary care of the older person to pay attention to the probability of cognitive impairment when low muscle strength is identified. In clinical practice, the information from this study reinforces the importance of monitoring the muscle strength of older people in order to prevent adverse outcomes such as sarcopenia and cognitive impairment. When faced with a probable sarcopenic older person, the multidisciplinary team must be attentive to the possibility of cognitive impairment and, when faced with an older person with cognitive impairment, one cannot fail to evaluate the sarcopenia parameters aiming at preventive intervention and control of the pathophysiological mechanisms shared between both illnesses.