Functional Mobility and postural balance in older adults with Alzheimer's disease: comparative study between mild and moderate stages

Objective : This study aimed to compare the functional mobility and postural balance of older adults among smild and moderate stages of Alzheimer's disease using the Timed Up and Go test (TUGT) and the Clinical Test of Sensory Interaction and Balance (CTSIB). Methodology : Forty elderly people were divided into two groups according to the mild (CDR1; n = 26) and moderate (CDR2; n = 14) stages of the disease. The Clinical Dementia Rating Scale (CDR) was used for staging the disease, which allows classifying the different degrees of dementia, assessing cognition and behavior. The scale allows classification into CDR 0 (normal or no alteration); 0.5 (questionable or mild cognitive impairment); 1 (mild dementia); 2 (moderate dementia) and 3 (severe dementia). In this study, only subjects classified as CDR 1 or CDR 2 were included. For the assessment of functional mobility, the Timed Up and Go Test (TUGT) was used in the conditions of single task, dual cognitive task and dual motor task, and the Clinical Test of Sensory (CTSIB) to assess postural balance. Data were compared between groups. Results : Performance on the TUGT single task, cognitive dual task, and motor dual task was significantly worse in the CDR2 group compared to the CDR1 group (p < 0.05). The CTSIB was not significantly different between the groups in the four conditions. Conclusion : Functional mobility during tasks involving cognition differs between older adults with mild and moderate dementia, and this commitment is more accentuated in dual-task situations. Postural balance did not differ between the stages of the disease.

AD is characterized by a progressive cognitive decline affecting the frontal cortex and impairing memory, language, and executive functions (Nietzsche, et al., 2015;Silva, et al., 2018).
Executive function is the cognitive capacity to plan, initiate, maintain, and inhibit behaviors (Cristofori, et al., 2019).
Studies showed that cognitive decline was associated with postural balance deficits and prevalence of falls, with greater impairment in older adults with AD than in those without it (Bortoli, et al., 2015;Cruz, et al., 2015). In addition, the fall risk was higher in older adults with executive dysfunction than in cognitively healthy older adults (Allali & Verghese, 2020;Allali, et al., 2017). In this context, understanding changes in functional mobility and postural balance among different AD stages allows health professionals to identify triggers for altered motor function and develop early interventions. Therefore, the present study aimed to compare functional mobility and postural balance of older adults among different AD stages using the Timed Up and Go test (TUGT) and Clinical Sensory of Test Interaction and Balance (CTSIB).

Methodology
This cross-sectional study was approved by the research ethics committee of the Federal University of Rio Grande do Norte (number 2.772.429) and conducted at the Clinical Center Dr. José Carlos Passos (Natal, Brazil) according to the Declaration of Helsinki. We selected patients of both sexes diagnosed with AD (Frota et al., 2011), over 60 years old, and presenting mild (CDR1) or moderate (CDR2) dementia according to the Clinical Dementia Rating scale (CDR). All volunteers and caregivers signed an informed consent form.
Patients without associated neurological diseases (e.g., dementia of other etiology, Parkinson's disease, and previous brain stroke), complaints, vertigo signs, or vestibular syndrome were included in the study. Patients were excluded if they showed physical and cognitive limitations during the tests, practiced regular physical activity, performed physical therapy for balance in the last six months, refused to perform the tests, or had MMSE score lower than nine or CDR different than 1 or 2. Limitations considered for exclusion were severely diminished visual and auditory acuity not improved with corrective devices and inability to reproduce movements or understand and respond to verbal commands.
Variables collected were classified in sociodemographic and clinical data (sex, age, monthly income, educational level, body mass index, number of diseases, number of medications, and AD diagnosis time) and postural balance (falls in the last six months, fear of falling, use of a walking aid device, dizziness).
Functional mobility was assessed using TUGT and the postural balance using CTSIB, and rest intervals were given to patients during and between tests according to their needs. The TUGT is a reliable and comprehensive test to assess functional mobility, including dynamic postural balance. It consists of getting up from a chair, walking three meters, turning around a cone, walking back to the chair, and sitting down (Podsiadlo & Richardson, 1991). The TUGT single-task was performed first, followed by dual-cognitive (animal names were dictated during the test) and dual-motor tasks (patients had to hold a glass of water with the dominant limb during the test) (Fatori, et al., 2015). The time to complete the test, number of steps, and cadence (steps/minute) were registered. The test identified which sensory system (visual, somatosensorial, or vestibular) affected postural control. Patients were instructed to remain for 30 seconds in each sensory condition without taking steps to compensate for instability or moving upper limbs, heels, and feet; they wore a blindfold for visual occlusion. We reduced the support by requesting older adults to stand with feet together (Romberg position). One trial was performed for each CTSIB condition. The time (seconds) during each condition was included for analysis, and the test was categorized as "normal" or "abnormal" (i.e., the patient could not maintain stability).
Data were expressed as absolute and relative frequency, mean ± standard deviation, median, minimum and maximum values, and 95% confidence intervals. The Kolmogorov-Smirnov test verified data normality. Time and number of steps to complete the TUGT and time in the CTSIB were compared between patients with CDR1 and CDR2 (groups) using the Mann-Whitney test; the paired t-test compared cadence during TUGT. Inferential analysis was performed using SPSS (IBM Corp.®, USA; version 17.0), and significance was set at p < 0.05 (two-tailed).
CTSIB was not different between groups in the four conditions (Table 3).

Discussion
Functional mobility was more affected during cognitive-motor tasks and worse in CDR2 than in CDR1. Fujisawa et al. (2017) observed that patients with moderate AD spent more time to complete the TUGT single-task than patients with normal cognition, mild cognitive decline, and mild AD. We also observed differences between groups in time to complete the test and number of steps in the TUGT single-task. Older adults with AD have a high fall risk, which can be reliably screened using TUGT (Allali & Verghese, 2020;Allali, et al., 2017;Ansai, Andrade & Rossi, et al., 2017;Borges, et al., 2015;Cruz, et al., 2015;Dyer, et al., 2020;Kato-Narita & Radanovic, 2009). We found a higher time to complete the test than previous studies (Alexandre, et al., 2012;Ansai, et al., 2019). Alexandre et al. (2012) and Ansai et al. (2019) presented a TUGT cutoff point of 12.47 and 17.56 seconds, respectively, for Brazilian older adults with AD. Patients with CDR2 completed the TUGT with median times above these cutoff points Bortoli, et al., 2015;Podsiadlo & Richardson, 1991).
Dual tasks require more planning, increasing the cognitive demand in different brain areas. A systematic review (Muir-Hunter & Wittwer, 2016) showed strong associations between gait in dual tasks and fall risk in older adults, although prospective studies did not observe this association. Data on dual tasks corroborate literature (Montero-Odasso, et al., 2012) and demonstrate how they may influence gait speed and cadence of older adults. Our data also indicated different functional mobility among CDR stages due to the influence of cognition on motor processes.  observed that patients with AD performed more stops and mistakes during dual-task TUGT and had a shorter cadence and poorer motor and cognitive performance than other groups. Other studies found that dual tasks predict falls in older adults with mild cognitive decline but not with mild AD (Ansai, Andrade & Masse, et al., 2017;Gonçalves, et al., 2018).
Using the Berg Balance Scale, Kato-Narita et al. (2011) identified a worse postural balance in older adults with moderate AD than in control groups. This corroborated our findings since the loss of functional capacity was associated with deficits in postural balance and falls. Unlike Kato-Narita et al. (2011), we assessed postural balance using the CTSIB, which required less cognitive demands. However, it failed to distinguish postural balance deficits between groups, probably due to the similar motor dysfunction among AD stages. Differences in TUGT may be related to the required cognitive demand since it is reduced in patients with AD (Nietzsche, et al., 2015). To our knowledge, no prospective studies analyzed postural balance in this population using CTSIB, which may be less influenced by cognitive biases than the Berg Balance Scale.
The primary motor cortex and mesencephalic locomotor regions contribute to gait and postural control (Demain, et al., 2013); thus, cognitive processes are involved in the control of static and dynamic postural balance and functional mobility (Maki & Mclroy, 2007). Also, the neurodegenerative process of AD and advanced age may slow postural correction, which is needed to maintain the stability during dynamic tasks and anticipate postural adjustments after changes in surfaces (Borel & Alescio-Lautier, 2014;Maki & Mclroy, 2000). In addition, the division of attention needed during dual tasks generates greater impairment than in static postural balance.
Cadence refers to the number and regularity of steps per minute. De Melo et al. (2019) observed significant differences in cadence between older adults with AD and mild cognitive decline, suggesting the interference of executive functions and cognition in gait performance and mobility. Dyer et al. (2020) found that slow gait speed was associated with cognitive decline and increased fall risk. Their results confirmed the reduced cadence found in CDR2 and reinforced the importance of gait analysis to predict falls in patients with AD.
This study is not free of limitations. The relatively small number of patients and the study design hampered analyzing whether functional mobility deficit was the main contributor to TUGT performance, which involves cognition (e.g., executive functions) and requires patients to process instructions associated with the tasks. Nevertheless, this is the first study comparing functional mobility and postural balance between patients with mild and moderate AD using dynamic and static postural balance tests. Longitudinal studies are recommended to understand the influence of balance postural and functional mobility on TUGT performance.

Conclusion
Functional mobility differed between older adults with mild and moderate dementia; the difference was more evident during dual tasks, probably due to high cognitive demand. Thus, major cognitive decline (i.e., CDR2) may affect gait and functional mobility due to executive dysfunction.