The impact of physical exercise on cognitive and affective functions and serum levels of brain-derived neurotrophic factor in nursing home residents: A randomized controlled trial

1 The impact of physical exercise on cognitive and affective functions and brain-derived neurotrophic factor levels in nursing home residents: a randomized controlled trial Haritz Arrietaa, Chloe Rezola-Pardoa, Maider Kortajarenab, Gotzone Hervása, Javier Gila, José J. Yanguasc, Miren Iturburuc, Susana M. Gila, Jon Irazustaa, Ana Rodriguez-Larrada aDepartment of Physiology, Faculty of Medicine and Nursing, University of the Basque Country (UPV/EHU), Spain. bDepartment of Nursing II, Faculty of Medicine and Nursing, University of the Basque Country (UPV/EHU), Spain. cMatia Instituto, Spain. Corresponding author: Haritz Arrieta Department of Physiology, Faculty of Medicine and Nursing, University of the Basque Country (UPV/EHU). Barrio Sarriena s/n, E-48940 Leioa (Bizkaia) – Spain +34 610970906 [email protected] This is the accepted manuscript of the article that appeared in final form in Maturitas 131 : 72-77 (2020), which has been published in final form at https://doi.org/10.1016/j.maturitas.2019.10.014. © 2019 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) 2 Abstract Objectives: To determine the effects of a multicomponent physical exercise program on cognitive and affective functioning among nursing home residents and to clarify whether there are any changes in serum brain-derived neurotrophic factor (BDNF) levels among participants. Study design: This was a single-blind randomized controlled trial among ten nursing homes in Gipuzkoa, Spain. The study included 112 men and women (ACTRN12616001044415). Participants in the control group participated in routine activities while the intervention group participated in a six-month individualized, progressive, multicomponent physical exercise program focused on strength, balance, and walking. Main outcome measures: Cognitive and affective functions were assessed at baseline and at six months. Serum BDNF levels were assessed via ELISA. Results: After six months, a group by time interaction in favor of the intervention group was observed on the Montreal Cognitive Assessment (MOCA), symbol search and De JongGierveld Loneliness Scale (P < 0.05). Meanwhile, the control group scored more poorly on the MOCA, WAIS-IV (coding and symbol search), verbal fluency, and semantic fluency tests after six months (P < 0.05) than they did at baseline. The intervention group showed poorer results on the coding test (P < 0.05). Loneliness perception was reduced in the intervention group (P < 0.05). No changes in serum BDNF were observed (group x time and within groups, P > 0.05). Conclusion: A six-month individualized, progressive, multicomponent physical exercise intervention is effective at maintaining cognitive function and decreasing loneliness perception among nursing home residents. Blood levels of BDNF were not affected by the intervention. Keywords: exercise training, cognition, loneliness, neurotrophin, nursing home, quality of life. Abbreviations: ANCOVA, Analysis of Covariance; BDNF, brain-derived neurotrophic factor; CG, control group; CV, co-efficient of variability; ELISA, enzyme-linked immunosorbent assay; IG, intervention group; MMSE, Mini Mental State Examination; MOCA, Montreal 3 Cognitive Assessment; MPE, multicomponent physical exercise; RAVLT, Rey Auditory-Verbal Learning Test; SPPB, Short Physical Performance Battery; WAIS-IV, Wechsler Adult Intelligence Scale, Fourth Edition; 1RM, 1 repetition maximum 4 1. Introduction Cognitive function tends to decline with aging, a phenomenon often referred to as cognitive aging [1]. Cognitive aging involves a lifelong process of gradual, ongoing, and highly variable changes in cognitive function that occur as people get older [1]. In addition, aging is related to deteriorating affective function [2]. In addition to the physical benefits of participating in an exercise program, interventional studies demonstrate that physical exercise improves certain domains of cognitive and affective functions in older adults [3,4]. However, these observations have mainly focused on community-dwelling older adults. Physical exercise interventions may be particularly important for older adults who live in nursing homes, as they typically have worryingly low levels of physical activity compared to current recommendations for community-dwelling older adults [5]. Few interventions implemented in nursing homes have assessed the effects of physical exercise programs on the cognitive and affective domains of individuals living in these settings [6,7]. The intensity of interventions is typically not reported in nursing home studies [8-10] but appears low in most cases. There is also high variability in the type of intervention [6,7] and in many studies, the length of the intervention is short, limiting the impact that the intervention might have [6-8]. Further, prior studies have not demonstrated clear effects on affective functions in nursing home residents, with some studies reporting improvements [6,7,9] that were not confirmed in others [10]. There is an urgent need for quality research involving nursing home residents, with longer interventions and larger sample sizes, to help determine the optimal physical exercise interventions needed to make significant improvements to and/or simply maintain cognitive and affective functions [8]. Brain-derived neurotrophic factor (BDNF) promotes the growth and differentiation of neurons and synapses while supporting the survival of existing neurons and is an important molecule involved in cognitive and affective functions [11,12]. BDNF also mediates improvements in 5 executive function following physical exercise [12]. However, the impact of physical exercise on serum concentrations of BDNF is equivocal [13]. While some studies describe increases in this neurotrophic factor, others do not find any changes after physical exercise programs [13]. To further explore the connection between BDNF and exercise and to generally establish guidelines for physical exercise among nursing home residents, we conducted an exploratory study evaluating the effects of a six-month individualized, progressive, multicomponent physical exercise (MPE) program on cognitive and affective functions among older adults living in nursing homes. Participants’ serum BDNF levels were analyzed before and after the intervention. 2. Methods 2.1. Study design This study was a six-month multicenter, randomized, controlled, single-blind study among ten nursing homes in Gipuzkoa, Spain (ACTRN12616001044415) [14]. This article presents the results of an exploratory secondary analysis whose main outcome was the Short Physical Performance Battery (SPPB) [15]. The intervention was conducted between October 2016 and July 2017. The trial was approved by the Committee on Ethics in Research at the University of the Basque Country, UPV/EHU (Humans Committee Code M10/2016/105; Biological Samples Committee Code M30/2016/106). Written informed consent was provided by each participant. 2.2. Participants The study included 112 men and women living in nursing homes who met the following criteria: ≥70 years old, scored ≥50 on the Barthel Index (0–100), scored ≥20 on the MEC-35 test (0–35) [an adapted and validated version of the Mini Mental State Examination (MMSE) in Spanish], and who were capable of standing up and walking independently for at least ten meters (Figure 1) [14]. Participants were randomized in a 1:1 ratio using sealed opaque envelopes to either the control (CG) or intervention group (IG) by coin-tossing sequence 6 generation. Assessments were done in each center by the research team, who were blinded to group allocation. 2.3. Control group Participants assigned to the CG participated in routine activities that the nursing homes offered to residents: memory workshops, reading, singing, and other similar activities. 2.4. Intervention group In addition to routine activities, the IG performed a previously described individualized and progressive MPE intervention [14]. The intervention consisted of a one hour supervised group session twice a week, separated by at least two days, for a six-month period. The MPE program included strength, balance, and walking recommendations. Each session began with five minutes of warm-up performing range-of-motion exercises. Strength training was individually adapted to each participant and focused on upper and lower extremity strength. The Brzycki equation was applied to calculate one repetition maximum (1RM) and adapt the adequate load progression of arm-curl, knee flexion, and knee extension exercises for every participant at baseline and every two months. Chair-stand, hip abduction, and hip adduction exercises were performed without external loads, and the intensity was tailored to the capabilities of each participant by adjusting the number of repetitions and velocity. The intensity ranged from 40% at first to 70% 1-RM in the sixth months of the program. Balance exercises were also individually adapted and included exercises progressing in difficulty, starting by decreasing arm support along with decreasing the base of support to challenge participants’ balance as they progressed. Some exercises varied throughout the period, including weight transfer from one leg to another, proprioceptive exercises, and stepping practice. Sessions ended with five minutes of deep breathing exercises and relaxation. Physical exercise sessions were conducted in a private gym or private rooms adequate for these activities so as to avoid any shared experiences with the control group. The rooms where the exercises were conducted were suitable for 7 approximately nine people at a time. Inside the room were parallel bars, firm chairs or a large table to use as a support for strength and balance exercises. All sessions were conducted by a professional instructor with a degree in physical activity and sports sciences and training in adapted physical activity in older adults. Attendance was determined by participants’ presence at physical exercise sessions. Walking recommendations were also individually tailored based on participants’ performance on the baseline six-minute walking test and started with paths that lasted five minutes per day at the beginning of the intervention, with the goal of completing 20 minutes per day (seven days/week) by the end of the six-month period. The recommendations were based on a specific number of laps on a certain track (based on the aerobic capacity of each participant as measured by the six-minute walk test). The number of laps performed by the participants in each day was collected daily the last week of the second and fourth months by each nursing home principal nurse. Those nurses monitored the completion of the recommendations checking the performance by themselves, asking the assistant nurses or the study participants according to the internal organization from each center. 2.5. Procedures All enrolled participants were assessed before the intervention period (month 0) and after the six-month intervention period. All measurements were collected by the same investigators. All outcome measurements were evaluated in the participants' place of residence. 2.6. Outcomes We previously described the benefits of the intervention on SPPB (primary outcome) in nursing homes residents [15]. Global cognition was assessed by the Montreal Cognitive Assessment (MOCA) [16], which included visuospatial/executive function, naming, attention, language, abstraction, delayed recall, orientation, and global measures of cognitive functioning. Total learning measure of the Spanish validated version of the Rey Auditory-Verbal Learning Test (RAVLT) [17] was used to evaluate verbal memory and capacity to recall and accumulate 8 words through learning trials. Trail making test A [18] was administered to evaluate executive function. The Coding and Symbol Search test was used to measure processing speed on the Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) [19]. Verbal fluency was measured by the number of words listed beginning with a given letter in 60 seconds [20]. Semantic fluency was measured through the number of words produced in the restricted "animal" category in 60 seconds [20]. Higher scores on these scales indicate better cognitive functioning, except for the Trail making test A, in which higher scores indicate worse performance. Anxiety and depression were measured with the Goldberg Anxiety and Depression Scale, in which high scores indicate more symptoms of anxiety and depression [21]. Loneliness was measured with the De Jong-Gierveld Loneliness Scale, where high scores indicate a higher perception of loneliness [22]. Quality of life was measured with the Quality of Life in Alzheimer’s Disease rating scale, where the total score ranges from 13 (worst) to 52 (best) points [23]. Blood samples were collected in the morning after an overnight fast and at least 24 hours after the last exercise session. After collection, tubes were centrifuged at 5,000 rpm for 10 min. Serum obtained for each participant was stored in aliquots at –80ºC until analysis. Serum BDNF (ng/mL) was quantified using a sandwich enzyme-linked immunosorbent assay (ELISA). Human BDNF Quantikine Immunoassay (R&D Systems, Minneapolis, MN) was performed according to the manufacturer’s instructions. Analyses, blinded for group allocation and in duplicate, were averaged. While the manufacturer reports <20 pg/mL of sensitivity, <6.2% intra-assay and <11.3% inter-assay co-efficient of variability (CV), the sensitivity of the experiments has been of <3.1% intra-assay and <9.2% inter-assay CV. 2.7. Statistical analyses 9 A sample size of 114 participants was required to detect minimal significant effects on the SPPB, the main outcome of the approved trial, accepting an alpha risk of 0.05 and a beta risk of 0.20 in a bilateral contrast assuming a 25% loss of follow-up [14]. Normal distribution of the data was checked using the Kolmogorov-Smirnov test. When not normally distributed, data were square root transformed. Statistical comparisons at baseline were performed using an unpaired student’s t test. Between-group differences were assessed using mixed design analysis of covariance (ANCOVA; two time points x two groups) controlling for baseline values. A post-hoc Bonferroni test was used to determine differences within groups. Partial η2 was calculated to estimate effect size. Values for η 2 of ≤0.02, ≤0.13, and ≥0.26 were considered small, medium, and large, respectively. The significance level for all tests was set at P < 0.05. Statistical analysis was performed using IBM SPSS Statistics 24 statistical software package (SPSS Inc., Chicago, IL). 3. Results The mean age of the participants was 84.9 years (age range = 70–102 years) and participants were predominantly women (70.5%). There were no differences between baseline values of CG and IG descriptive characteristics (Table 1). Of the 112 participants who began the study, 88 completed the six-month assessment (Figure 1). Attendance rates for the physical exercise sessions were 90.8%, and compliance with the walking recommendations was 79.0%. No adverse events associated with the physical exercise program were observed. After six months of MPE intervention, a group by time interaction in favor of the IG was observed on the MOCA (P = 0.003; η 2 = small) and symbol search tests (P = 0.049; η 2 = small) (Table 2). When differences were analyzed within groups, the CG showed poorer results for MOCA, WAIS-IV (coding and symbol search), verbal fluency, and semantic fluency tests after 16 8 L. Brett, V. Traynor, P. Stapley, Effects of physical exercise on health and well-being of individuals living with a dementia in nursing homes: A systematic review, J. Am. Med. Dir. Assoc. 17 (2) (2016) 104-116. https://doi.org/10.1016/j.jamda.2015.08.016. 9 N. Lok, S. Lok, M. Canbaz, The effect of physical activity on depressive symptoms and quality of life among elderly nursing home residents: Randomized controlled trial, Arch. Gerontol. Geriatr. 70 (2017) 92-98. https://doi.org/10.1016/j.archger.2017.01.008. 10 K. Frändin, H. Grönstedt, J.L. Helbostad, A. Bergland, M. Andresen, L. Puggaard, K. Harms-Ringdahl, R. Granbo, K. Hellström, Long-term effects of individually tailored physical training and activity on physical function, well-being and cognition in Scandinavian nursing home residents: a randomized controlled trial, Gerontol. 62 (6) (2016) 571-580. https://doi.org/10.1159/000443611. 11 A.L. Carvalho, M.V. Caldeira, S.D. Santos, C.B. Duarte, Role of the brain‐derived neurotrophic factor at glutamatergic synapses, Br. J. Pharmacol. 153 (S1) (2008) S310-324. https://doi.org/10.1038/sj.bjp.0707509. 12 R.L. Leckie, L.E. Oberlin, M.W. Voss, R.S. Prakash, A. Szabo-Reed, L. ChaddockHeyman, S.M. Philips, N.P. Gothe, E. Mailey, V.J. Vieira-Potter, S. A. Martin, B.D. Pence, M. Lin, R. Parasuraman, P.M. Greenwood, K.J. Fryxell, J.A. Woods, E. McAuley, A.F. Kramer, K.I. Erickson, BDNF mediates improvements in executive function following a 1year exercise intervention, Front. Hum. Neurosci. 8 (2014) 985. https://doi.org/10.3389/fnhum.2014.00985. 13 J.E. Ahlskog, Y.E. Geda, N.R. Graff-Radford, R.C. Petersen, Physical exercise as a preventive or disease-modifying treatment of dementia and brain aging, Mayo Clin. Proc. 86 (9) (2011) 876-884. https://doi.org/10.4065/mcp.2011.0252. 14 A. Rodriguez-Larrad, H. Arrieta, C. Rezola, M. Kortajarena, J.J. Yanguas, M. Iturburu, S.M. Gil, J. Irazusta, Effectiveness of a multicomponent exercise program in the attenuation of frailty in long-term nursing home residents: study protocol for a randomized 17 clinical controlled trial, BMC Geriatr. 17 (1) (2017) 60. https://doi.org/10.1186/s12877017-0453-0. 15 H. Arrieta, C. Rezola‐Pardo, S.M. Gil, J. Virgala, M. Iturburu, I. Antón, V. GonzálezTemplado, J. Irazusta, A. Rodriguez‐Larrad, Effects of Multicomponent Exercise on Frailty in Long‐Term Nursing Homes: A Randomized Controlled Trial, J. Am. Geriatr. Soc. (2019). https://doi.org/10.1111/jgs.15824. 16 R.F. Coen, D.A. Robertson, R.A. Kenny, B.L. King-Kallimanis, Strengths and limitations of the MoCA for assessing cognitive functioning. Findings from a large representative sample of Irish older adults, J. Geriatr. Psychiatry Neurol. 29 (1) (2015) 18–24. https://doi.org/10.1177/0891988715598236. 17 M.O. Ponton, Research and assessment issues with Hispanic populations, in: M.O. Ponton, J. Leon-Carrion (Eds.), Neuropsychology and the Hispanic patient: A clinical handbook, Lawrence Erlbaum Associates, Mahwah, 2001, pp. 39-58. 18 R.M. Reitan, D. Wolfson, The Halstead-Reitan neuropsychological test battery: Theory and clinical interpretation (Vol. 4), Neuropsychology Press, Tucson, 1985. 19 D. Wechsler, WAIS-IV UK Administration and scoring manual, Pearson, London, 2010. 20 D.J.L. Serrani, Spanish translation and validation of an Executive Battery 25 (EB25) and its shortened version (ABE12) for executive dysfunction screening in dementia, Neurología (English Edition), 28 (8) (2013) 457-476. https://doi.org/10.1016/j.nrleng.2013.10.001. 21 D. Goldberg, K. Bridges, P. Duncan-Jones, D. Grayson, Detecting anxiety and depression in general medical settings, Br. Med. J. 297 (1988) 897-899. https://doi.org/10.1136/bmj.297.6653.897. 22 J. De Jong-Gierveld, F. Kamphuls, The development of a Rasch-type loneliness scale, Appl. Psychol. Meas. 9 (3) (1985) 289-299. https://doi.org/10.1177/014662168500900307. 23 M. Gómez-Gallego, J. Gómez-Amor, J. Gómez-García, Validación de la versión española de la escala QoL-AD en pacientes con enfermedad de Alzheimer, cuidadores y 18 profesionales sanitarios, Neurología, 27 (1) (2012) 4-10. https://doi.org/10.1016/j.nrl.2011.03.006. 24 G.K.C. Wong, J.S.Y. Mak, A. Wong, V.Z.Y. Zheng, W.S. Poon, J. Abrigo, V.C.T. Mok, Minimum clinically important difference of montreal cognitive assessment in aneurysmal subarachnoid hemorrhage patients, J. Clin. Neurosci. 46 (2017) 41-44. https://doi.org/10.1016/j.jocn.2017.08.039. 25 J. Gomes-Osman, D.F. Cabral, T.P. Morris, K. McInerney, L.P. Cahalin, T. Rundek, A. Oliveira, A. Pascual-Leone, Exercise for cognitive brain health in aging: A systematic review for an evaluation of dose, Neurol. Clin. Pract. 8 (3) (2018) 1-8. https://doi.org/10.1212/CPJ.0000000000000460. 26 A.K. Welmer, D. Rizzuto, C. Qiu, B. Caracciolo, E.J. Laukka, Walking speed, processing speed, and dementia: a population-based longitudinal study, J. Gerontol. A Biol. Sci. Med. Sci. 69 (12) (2014) 1503-1510. https://doi.org/10.1093/gerona/glu047. 27 C. Milligan, C. Dowrick, S. Payne, B. Hanratty, P. Irwin, D. Neary, D. Richardson, Men's Sheds and other gendered interventions for older men: improving health and wellbeing through social activity-a systematic review and scoping of the evidence base. Lancaster University Centre for Ageing Research, Lancaster, 2013. 28 N. Stevens, Combating loneliness: a friendship enrichment programme for older women. Ageing. Soc. 21 (2) (2001) 183-202. https://doi.org/10.1017/S0144686X01008108. 29 C.M. Perissinotto, I.S. Cenzer, K.E. Covinsky, Loneliness in older persons: a predictor of functional decline and death, Ann. Intern. Med. 172 (14) (2012) 1078-1084. https://doi.org/10.1001/archinternmed.2012.1993. 30 A. Dechamps, P. Diolez, E. Thiaudière, A. Tulon, C. Onifade, T. Vuong, C. Helmer, I. Bourdel-Marchasson, Effects of exercise programs to prevent decline in health-related quality of life in highly deconditioned institutionalized elderly persons: a randomized controlled trial, Arch. Intern. Med. 170 (2) (2010) 162–169. https://doi.org/10.1001/archinternmed.2009.489. 19 31 L.N. Forti, R. Njemini, I. Beyer, E. Eelbode, R. Meeusen, T. Mets, I. Bautmans, Strength training reduces circulating interleukin-6 but not brain-derived neurotrophic factor in community-dwelling elderly individuals, Age, 36 (5) (2014) 9704. https://doi.org/10.1007/s11357-014-9704-6. 20 Figure 1 Study flow diagram. 21 Table 1 Descriptive characteristics of participants at baseline Control group (n = 55) Intervention group (n = 57) P value Age (years), mean ± SD 84.7 ± 6.1 85.1 ± 7.6 0.812 Sex, n (%) Female 37 (67.3) 42 (73.7) 0.461 Male 18 (32.7) 15 (26.3) BMI (kg/m2), mean ± SD 28.2 ± 5.3 28.2 ± 5.1 0.991 WHR, mean ± SD 0.97 ± 0.08 0.98 ± 0.07 0.697 Barthel Index, mean ± SD 82.8 ± 13.1 79.2 ± 12.9 0.153 MEC-35 test, mean ± SD 28.0 ± 3.5 27.0 ± 4.0 0.157 SD, standard deviation; BMI, body mass index; WHR, waist-to-hip ratio; MEC-35 test, adapted and validated version of the Mini Mental State Examination in Spanish 22 Table 2 Cognitive and affective function among control and intervention group participants. Control group (n = 45) Intervention group (n = 43) Baseline, mean (SD) 6 months, mean (SD) Baseline, mean (SD) 6 months, mean (SD) P a Partial η 2 b Cognitive function MOCA (0–30) 14.3 (3.7) 13.0 (5.1)* 14.2 (4.5) 14.9 (5.1) 0.003 0.121 RAVLT (total rep.) (0–75) 19.2 (8.4) 20.7 (11.5) 18.0 (8.7) 19.9 (12.1) 0.140 0.028 TMT-A 124 (54) 136 (72) 120 (45) 107 (40) 0.836 0.001 Coding (WAIS-IV) 13.8 (9.5) 10.9 (8.8)* 11.7 (7.8) 10.1 (8.4)* 0.439 0.011 Symbol search (WAIS-IV) 9.0 (4.4) 6.6 (5.4)* 7.0 (5.3) 6.7 (5.9) 0.049 0.063 Verbal fluency test 7.3 (3.9) 6.5 (4.1)* 6.7 (3.5) 6.4 (3.7) 0.489 0.006 Semantic fluency test 8.7 (3.1) 7.4 (4.2)* 7.9 (3.2) 7.2 (3.3) 0.638 0.003 Affective function Goldberg anxiety 1.3 (1.3) 1.4 (1.5) 1.5 (1.8) 1.3 (1.6) 0.817 0.001 Goldberg depression 0.8 (1.8) 1.2 (1.8) 1.0 (2.1) 0.9 (1.8) 0.478 0.007 DJGLS 5.3 (2.8) 4.3 (3.4) 5.8 (3.1) 3.2 (2.6)* 0.030 0.062 QoL-AD 34.1 (4.9) 33.1 (5.3) 34.3 (6.6) 35.4 (7.1) 0.051 0.049 BDNF (ng/mL) 33.6 (14.2) 32.8 (15.7) 34.2 (15.0) 33.5 (13.2) 0.899 <0.001 SD, standard deviation; MOCA, Montreal Cognitive Assessment; RAVLT, Rey Auditory-Verbal Learning Test; rep, repetitions; TMT-A, Trail Making Test A; WAIS-IV, Wechsler Adult Intelligence Scale, Fourth Edition; DJGLS, De Jong-Gierveld Loneliness Scale; QoL-AD, quality of life in Alzheimer’s disease; BDNF, brain-derived neurotrophic factor a P for group x time interaction. b Estimation of effect size for group x time interaction. *P < 0.05, significantly different from baseline.