- Email: [email protected]
Longitudinal association between brain volume change and gait speed in a general population
Experimental Gerontology 118 (2019) 26–30
Contents lists available at ScienceDirect
Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero
Longitudinal association between brain volume change and gait speed in a general population Sunghee Leea, Eun Young Kimb, Chol Shinb,c,
T
⁎
a
Department of Food and Nutrition, College of Health Science, Kangwon National University, Samcheok, Republic of Korea Institute of Human Genomic Study, School of Medicine, Korea University Ansan Hospital, Ansan, Republic of Korea c Division of Pulmonary, Sleep and Critical Care Medicine, Department of Internal Medicine, Korea University, Ansan Hospital, Ansan, Republic of Korea b
ARTICLE INFO
ABSTRACT
Keywords: Gait speed Brain volume General population
Objective: To determine the association between brain structural changes and gait speed in a four-year longitudinal prospective cohort study. Measurements: A total of 767 well-functioning community-dwelling participants, free of arthritis, silent infarct, stroke, dementia, head injury, and cancer, completed baseline brain magnetic resonance imaging scan and gait speed tests between 2011 and 2014, and follow-up tests between 2015 and 2017. The gait test consisted of measuring the elapsed time to walk four meters at usual speed. To estimate whether brain volume changes predict gait speed decline at follow-up, a generalized linear regression model was used after adjusting for potential confounding factors including gait speed at baseline. Results: Participants who experienced ≥0.05 m/s gait speed decline, previously defined as a clinically meaningful decline, were more likely to be women, less likely to be smokers, and had lower physical activity scores (p = 0.003, p = 0.025, and p = 0.006, respectively), as compared to those who did not experience the decline. Also, they demonstrated smaller volumes of hippocampus, total gray matter, parietal gray matter, temporal gray matter, and temporal white matter (p = 0.004, p = 0.042, p = 0.021, p = 0.001, and p = 0.004, respectively). Even after correcting the significance level due to multiple comparisons, overall gray matter and overall white matter volume changes during four-year follow-up period showed significant associations with gait speed at follow-up (p < 0.001 and p = 0.002). Regarding region-specific volumes, frontal white matter and parietal gray matter volume changes demonstrated significant associations with gait speed (p = 0.002, p = 0.004, respectively). Conclusion: In a four-year longitudinal study among 767 well-functioning community-dwelling healthy participants from a general population, we observed a significant association between brain volume changes and gait speed.
1. Introduction Most individuals experience gait speed decline with advanced aging. Slow gait speed has been reported to increase the risk for fall (Doi et al., 2015), cognitive decline (Marquis et al., 2002), dementia (Dumurgier et al., 2017), and death (Studenski et al., 2011) among functionally healthy adults. A gait speed test has been validated as a reliable and simple geriatric assessment tool (Munoz-Mendoza et al., 2011; NIH, 2018; Peel et al., 2013). Longitudinal studies have confirmed a time sequence between slow gait speed and cognitive functional decline as one of causal criteria (Mielke et al., 2013; Waite et al., 2005). The development of a functional decline of the brain, such as
mild cognitive impairment, is preceded by a decline in gait speed by approximately 12 years (Buracchio et al., 2010). Attempts to explain the cognitive functional decline include age-related changes, mainly small vessel lesions in the brain structure of the prefrontal area responsible for executive functions (Rosano et al., 2012; Taylor et al., 2017; Watson et al., 2010). The functional decline of the brain is preceded by its structural changes including volume atrophy with aging (Jack Jr. et al., 2013). Thus, early identification of brain structural change preceding cognitive functional decline may provide a wider window of opportunity to implement an active intervention to decrease the risk of cognitive impairment. Particularly, age-related brain structural changes of the gray
⁎ Corresponding author at: Institute of Human Genomic Study, Ansan Hospital Korea University, 516 Gojan-1-dong, Danwon-gu, Ansan-si, Gyeonggi-do 15355, Republic of Korea. E-mail address: [email protected] (C. Shin).
https://doi.org/10.1016/j.exger.2019.01.004 Received 17 August 2018; Received in revised form 11 November 2018; Accepted 3 January 2019 Available online 04 January 2019 0531-5565/ © 2019 Elsevier Inc. All rights reserved.
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
2.2. Brain volume measurement
matter have been attributed to neural cell loss (Giorgio et al., 2010). Additionally, the frontal, temporal, and hippocampal regions have been reported to be linked with motor performance such as gait speed and other physical activities (Arenaza-Urquijo et al., 2017). Few studies have examined the longitudinal relationship between brain structural region-specific volume change and gait speed decline among well-functioning adults from a general population. Therefore, we investigated an association between brain structural volume change and gait speed. We have previously reported a cross-sectional association between slow gait speed and cognitive functional decline as well as an interaction with obstructive sleep apnea among functionally healthy middle-aged and older adults (Lee and Shin, 2017). Here, we further aimed to investigate an association between brain structural volume changes over a four-year follow-up period and gait speed at follow-up among 767 well-functioning community-dwelling adults aged 49 to 79 years. We hypothesized that individuals with brain atrophy would be more likely to exhibit slow gait speed in a four-year longitudinal study. To test this hypothesis, we used data from the Korean Genome and Epidemiology Study (KoGES).
Using a GE Signa HDxt 1.5 Tesla MRI scanner (GE Healthcare, USA), three-dimensional T1-weighted MRI data were collected between 2011 and 2014 as a baseline measurement. After the first sequence between March 1, 2011 and April 18, 2011, MRI protocols were changed to improve the image quality for the second sequence from April 19, 2011 to April 18, 2013, while retaining the main characteristics of the sequence. More detailed procedure has been published (Kim et al., 2018). Analysis of all images was performed at the Korea University in collaboration with the University of Iowa (Iowa City, IA, USA). Sub-regions of interest, including frontal, parietal, temporal, and occipital regions, were merged from 215 independent sub-regions of the brain. 2.3. Gait speed The gait speed test has been used as a simple and reliable geriatric assessment tool as introduced by the National Institutes of Health [Bethesda, MD, USA (NIH, 2018)]. Its reliability has been validated (Munoz-Mendoza et al., 2011). All participants were asked to walk a four-meter course at usual gait speed from the start point to the end point along a hallway. At the same time, a research staff measured the time to pass the end point of the course using a stopwatch. The gait speed was calculated by dividing four meters by the time in seconds to finish the four-meter walk (meters per seconds, m/s).
2. Methods 2.1. Study population The KoGES is an ongoing longitudinal study initiated in 2001 to investigate the prevalence rates and risk factors of chronic diseases in a general population. A detailed report has been published (Lee and Shin, 2017). At enrollment from 2001 to 2002, 5012 adults aged 40 to 69 years were randomly selected by telephone contacts from a general population in Ansan City, located south of Seoul, the capital of South Korea (Lee and Shin, 2017). On biennial follow-up visits, participants underwent biochemical and physical examinations, and completed an interviewer-administrated questionnaire. Professionally trained research staff carried out these tasks according to standardized study protocols. The study participants underwent height and weight measurements wearing only light clothing and no shoes. On each visit, venipuncture was performed early in the morning after more than eight hours of fasting. Blood samples were sent to the Seoul Clinical Laboratory (Seoul, South Korea). To focus on brain aging, geriatric measurements including brain magnetic resonance imaging (MRI) scan, depression score, cognitive and gait tests, were performed every four years since 2011. At baseline of the brain aging study, 2587 participants underwent an MRI and completed a gait test between 2011 and 2014. From 2015 to 2017, 934 individuals completed a follow-up MRI and a follow-up gait test. The follow-up data were limited because the follow-up MRI examinations have been still ongoing and will be completed in December 2018. Nonetheless, we confirmed no significant difference in the primary outcome of interest between participants who were lost to follow up and those who were successfully followed up (p-value = 0.539, Supplemental Table 1). Of 934 participants with both baseline and follow-up data, 159 individuals were excluded due to silent infarct (n = 68), stroke (n = 8), arthritis (n = 52), or cancer (n = 31). No participant met the additional exclusion criteria head injury or dementia diagnosis. Additionally, eight participants were excluded due to missing covariates such as depression (n = 2) and diabetes mellitus (n = 6). Therefore, 767 participants (375 men and 392 women) were included in the final analyses. A study flow diagram is presented in Supplemental Fig. 1. All study participants were informed about any potential risks and benefits prior to signing an informed written consent form. The current study adhered to the ethics rules for human research described in the Declaration of Helsinki. The Human Subjects Review Committee and the Institutional Review Board reviewed and approved this study (IRB No. 2018AS0015).
2.4. Cognitive measurement To examine cognitive function of the brain, the Mini-Mental State Examination (MMSE), the most widely used global cognitive screening test, was used (Folstein et al., 1975). Its maximum possible score was 30. A Korean version of the MMSE (K-MMSE), developed and validated for the Korean language, was employed (Kang et al., 1997). 2.5. Statistical analysis To present general characteristics of the study participants between at baseline and at follow-up, paired t-test or Wilcoxon signed-rank test were utilized. For categorical variables, McNemar's test was used. Also, baseline characteristics were compared according to either individuals who experienced a clinically meaningful decline in gait speed such as ≥0.05 m/s (Perera et al., 2006) or individuals who did not experience. To estimate the association between brain region-specific volume changes over a four-year period and gait speed at follow-up, generalized linear regression models were utilized, after adjusting for age, body mass index (BMI), education, sex, physical activity (metabolic equivalent task [MET], hours/week), smoking, alcohol use, hypertension, diabetes, high-sensitivity C-reactive protein (hs-CRP), depression (Beck's Depression Index [BDI]), total cholesterol, total intracranial volume, and gait speed at baseline. A two-sided p-value < 0.05 was considered to be statistically significant. A corrected significance level (0.05/9 = 0.0056) was considered due to multiple comparisons among hippocampus, frontal gray matter, frontal white matter, parietal gray matter, parietal white matter, occipital gray matter, occipital white matter, temporal gray matter, and temporal white matter volumes. SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses. 3. Results 3.1. General characteristics of the study participants The general characteristics of the 767 participants are summarized in Table 1. The baseline of the brain aging study from 2011 to 2014 and the follow-up of the study from 2015 to 2017 were completed. At baseline, the average age of the study participants was 58.97 years; at 27
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
Table 1 General characteristics of the study participants (n = 767).
Age, year Female, % Education, % None Elementary Middle school High school University Body mass index, kg/m2 Total cholesterol, mg/dL Smoking, pack-year⁎ Current drinker, % Hypertension, % Diabetes, % hs-CRP, mg/L⁎ Physical activity, MET (h/ week)⁎ Depression, BDI⁎ MMSE Time to walk four meters Usual gait (s) Usual gait speed (m/s) Brain volume, ml Gray matter White matter Hippocampus Frontal Frontal gray matter Frontal white matter Parietal Parietal gray matter Parietal white matter Occipital Occipital gray matter Occipital white matter Temporal Temporal gray matter Temporal white matter Intracranial volume, ml
Baseline Year 2011–2014
Follow-up Year 2015–2017
p-value
58.97 ± 6.44 51.11
62.97 ± 6.43
< 0.001 0.539
1.17 1.83 12.52 61.67 22.82 24.50 ± 3.01 193.78 ± 34.95 0 (0, 15.00) 45.89 34.42 19.69 0.64 (0.35, 1.22) 13.95 (0, 33.00)
24.41 ± 3.04 185.95 ± 35.55 0 (0, 13.50) 43.68 35.85 20.47 0.62 (0.35, 1.11) 13.50 (0, 30.00)
0.022 < 0.001 0.113 0.002 < 0.001 < 0.001 0.572 0.174
7.00 (3.00, 11.00) 27.68 ± 1.89
5.00 (2.00, 9.00) 27.66 ± 1.99
< 0.001 0.852
3.91 ± 0.74 1.06 ± 0.19
3.91 ± 0.70 1.05 ± 0.18
0.993 0.553
707.50 ± 59.29 454.47 ± 43.32 8.34 ± 0.83
698.11 ± 58.02 455.32 ± 44.30 8.35 ± 0.85
< 0.001 < 0.001 0.115
217.42 ± 20.85 164.24 ± 17.50
214.76 ± 20.24 165.33 ± 17.98
< 0.001 < 0.001
132.84 ± 12.48 99.34 ± 10.21
131.28 ± 12.37 99.72 ± 10.45
< 0.001 < 0.001
79.34 ± 8.08 32.00 ± 3.96
76.86 ± 7.85 32.29 ± 4.11
< 0.001 < 0.001
133.51 ± 13.21 55.63 ± 5.83 1375.56 ± 120.59
133.13 ± 12.96 55.15 ± 5.93
< 0.001 < 0.001
Table 2 Baseline characteristics of the study participants according to a clinically meaningful change in usual gait speed (n = 767).
Age, year Female, % Education, % None Elementary Middle school High school University Body mass index, kg/m2 Total cholesterol, mg/dL Smoking, pack-year⁎ Current drinker, % Hypertension, % Diabetes, % hs-CRP, mg/L⁎ Physical activity, MET (h/week)⁎ Depression, BDI⁎ MMSE Brain volume, ml Hippocampus Gray matter White matter Frontal Frontal gray matter Frontal white matter Parietal Parietal gray matter Parietal white matter Occipital Occipital gray matter Occipital white matter Temporal Temporal gray matter Temporal white matter
< 0.001
Mean ± SD; paired t-test; chi-square test; McNemar test; hs-CRP, high sensitivity C-reactive protein; MET, metabolic equivalent task. ⁎ Median (interquartile range) and p-values with Wilcoxon signed-rank test.
follow-up, their average age was 62.97 years. The proportion of women was 51.11%. More than half of these participants graduated from high school (61.67%). Compared to those at baseline, participants at followup had a slightly lower BMI (p = 0.022), lower levels of total cholesterol (p < 0.001), and a smaller proportion of current drinkers (p = 0.002). However, at follow-up, the study participants presented higher prevalence rates of hypertension (34.42% vs. 35.85%; p < 0.001) as well as diabetes mellitus (19.69% vs. 20.47%; p < 0.001) compared to the rates at baseline. As for depression, participants had lower BDI scores at follow-up compared to those at baseline (p < 0.001). However, < 10 BDI scores were not considered as clinically meaningful. The brain volume of the gray matter decreased over the four-year follow-up (p < 0.001), whereas the volume of the white matter increased (p < 0.001). Regarding brain sub-regional volume changes, the gray matter volumes in all four investigated areas including the frontal, parietal, occipital, and temporal region were reduced at follow-up (all p < 0.001). Conversely, white matter volumes were not reduced, except for the temporal area.
Participants with a clinically meaningful decline⁎⁎ (n = 310)
Participants without a meaningful decline (n = 457)
p-value
58.83 ± 6.78 57.74
59.06 ± 6.20 46.61
0.632 0.003
1.29 2.26 12.90 62.90 20.65 24.42 ± 3.21
1.09 1.53 12.25 60.83 24.29 24.55 ± 2.86
0.762
0.580
193.78 ± 34.20
193.77 ± 35.49
0.997
0 (0, 10.00) 42.90 33.23 18.06 0.65 (0.35, 1.26) 11.30 (0, 27.90)
0 (0, 18.00) 47.92 35.23 20.79 0.64 (0.35, 1.20) 16.00 (0, 36.00)
0.025 0.171 0.567 0.352 0.753 0.006
7.00 (3.00, 12.00) 27.75 ± 1.88
6.00 (3.00, 11.00) 27.63 ± 1.89
0.426 0.416
8.23 ± 0.84 702.21 ± 59.96 450.77 ± 44.43 378.86 ± 37.27 215.83 ± 20.71
8.41 ± 0.82 711.08 ± 58.63 456.98 ± 42.41 383.56 ± 36.73 218.49 ± 20.90
0.004 0.042 0.052
163.03 ± 17.72
165.07 ± 17.32
0.115
229.94 ± 22.00 131.58 ± 12.44
233.71 ± 21.69 133.70 ± 12.44
0.021
98.36 ± 10.27
100.01 ± 10.13
0.028
110.54 ± 12.21 78.81 ± 8.38
111.88 ± 11.38 79.70 ± 7.85
0.134
31.73 ± 4.11
32.18 ± 3.84
0.122
186.56 ± 18.48 131.66 ± 13.05
190.88 ± 18.45 134.76 ± 13.18
0.001
54.89 ± 5.89
56.12 ± 5.75
0.004
0.082
Mean ± SD; paired t-test; chi-square test; hs-CRP, high sensitivity C-reactive protein; McNemar's test; MET, metabolic equivalent task. ⁎ Median (interquartile range) with p-values from Wilcoxon signed-rank test. ⁎⁎ The clinically meaningful change was defined as 0.05 m/s (Perera et al., 2006).
2006) of > 0.05 m/s in gait speed or who did not experienced that decline are presented in Table 2. Individuals with the clinically meaningful decline in gait speed were more likely to be women, less likely to be a smoker, and had a lower physical activity score (p = 0.003, p = 0.025, and p = 0.006, respectively). Regarding the brain volumes, participants who experienced a clinically meaningful change in gait speed showed smaller volumes in the hippocampus, the total gray matter, the parietal gray matter, the temporal gray matter, and the temporal white matter (p = 0.004, p = 0.042, p = 0.021, p = 0.001, and p = 0.004, respectively).
3.2. Baseline characteristics of the study participants according to a clinically meaningful change in usual gait speed
3.3. Significant association of brain structural change on slow gait speed Table 3 shows the association between brain region-specific structural volume changes and gait speed at follow-up among the 767 study participants. Total gray matter and total white matter volume changes
Comparisons of general characteristics according to participants either who experienced a clinically meaningful decline (Perera et al., 28
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
for incident physical impairment including slow gait speed with brain abnormalities including enlarged ventricular, subclinical brain infarcts, and white matter hyperintensities (Rosano et al., 2005). There have also been several cross-sectional studies examining the association between brain volume and gait speed (Blumen et al., 2018; Callisaya et al., 2014; Rosano et al., 2012). One of these studies involving 214 individuals reported a significant association between a small prefrontal area volume and a slow gait speed (Rosano et al., 2012). Using voxel-based morphometry, another cross-sectional study among 305 adults aged > 60 years demonstrated a significant association between smaller total gray matter volume in brain regions such as frontal, parietal, occipital, temporal, cingulate, insula, parahippocampal, and cerebellar areas, and slower gait speed (Callisaya et al., 2014). Using voxel-based morphometry and multivariate covariate patterns, another cross-sectional study among 352 non-demented community-dwelling older adults from three different cohorts investigated the relationships between gray matter volume and gait speed, explaining with age-related changes in shared neural regions attributed to processing speed and gait speed (Blumen et al., 2018). The underlying mechanism on how the brain structure affects slow gait speed has not been fully elucidated. However, previous studies suggested that both brain structural atrophy and slow gait speed share the same neural regions vulnerable to age-related changes (Blumen et al., 2018). While brain atrophy of the gray matter is known to consistently progress from early stages of the adult life, white matter volume atrophy has been recognized to start at a much later stage of the adult life with a greater rate of atrophic degeneration (Ge et al., 2002). Further studies are warranted to disentangle the underlying mechanisms. We observed that participants who experienced a clinically meaningful decline ≥0.05 m/s were less likely to smoke. This is because they had higher proportion of women (57.74%, 0 pack-year), as compared to those who did not experience the decline had lower proportion of women (46.61%, 0 pack-year). Additionally, we tried adjusting for ‘total change in brain volume’ as a covariate, although this procedure might cause over-adjustment because each regional volume change was already included in the total change of the brain volume. No significant association was observed, when considering the corrected significance level (Supplemental Table 2). The present study has several strengths. First, we utilized a fouryear longitudinal prospective cohort study design that elucidates the time sequence between structural brain volume changes and gait speed decline to investigate a causal relationship. Second, our study population consisted of well-functioning, community-dwelling, middle-aged and older adults from a general population, which has the advantage of generalizability. Third, our participants were relatively young to exhibit brain functional cognitive impairment, which turned out to be a favorable time for investigating brain structural change. These strengths enable us to investigate the early stages of brain structural changes before cognitive functional decline starts to occur. Finally, we measured region-specific brain volumes including the frontal, parietal, occipital, and temporal regions. However, this study also had a limitation that needs to be considered when interpreting our findings. We did not investigate all brain regions; our study did not include the cerebellum or thalamus. In conclusion, in this four-year longitudinal study among 767 wellfunctioning community-dwelling healthy participants from a general population, we observed significant associations between brain volume changes and gait speed.
Table 3 Association between brain volume changes and usual gait speed at follow-up. Brain volume change, ml
ΔGray matter volume ΔWhite matter volume ΔHippocampus ΔFrontal Frontal gray matter Frontal white matter ΔParietal Parietal gray matter Parietal white matter ΔOccipital Occipital gray matter Occipital white matter ΔTemporal Temporal gray matter Temporal white matter
Time to walk four meters at usual gait speed, seconds (n = 767) β
p-value
0.01 0.01 −0.21
< 0.001 0.002 0.167
0.01 0.03
0.021 0.002
0.03 0.04
0.004 0.006
0.03 0.06
0.009 0.063
0.06 0.02
0.063 0.455
Adjusted for age, sex, body mass index, education, smoking, drinking, physical activity, hypertension, diabetes, total cholesterol, depression score, total intracranial volume in ml, hs-CRP, and gait speed at baseline; Δ, the change in the brain volume was defined as the volume at baseline subtracted by the volume at follow-up.
over the four-year follow up indicated significant associations with gait speed (p < 0.001 and p = 0.002, respectively). Regarding region-specific volumes, frontal gray matter, frontal white matter, parietal gray matter, parietal white matter, and occipital gray matter volume changes demonstrated significant associations with gait speed at follow-up (p = 0.021, p = 0.002, p = 0.004, p = 0.006, and p = 0.009, respectively). Even after considering the corrected significance level due to multiple comparisons (0.05/9 = 0.0056), the associations with overall gray matter, overall white matter, frontal white matter, and parietal gray matter volume changes, on gait speed remained statistically significant (p < 0.001, p = 0.002, p = 0.002, and p = 0.004). 4. Discussion In this four-year longitudinal prospective cohort study, we found that gray matter and white matter volume changes over the four-year follow-up period indicated significant associations on usual gait speed. At the level of brain sub-regions, frontal white matter and parietal gray matter volume changes demonstrated significant associations with gait speed at follow-up. Our current findings are unique in that we investigated the association between brain region-specific volume changes and gait speed among well-functioning community-dwelling middle-aged and older adults from the general population in a four-year longitudinal follow-up study. Our analysis of the follow-up MRI data from a general population helps us better understand the association between brain atrophy and slow gait speed in aging. Previous studies examined gait characteristics associated with brain abnormalities. A study examined the association of brain volume changes on slow gait speed using MRI data (Callisaya et al., 2013). This study reported a longitudinal association of brain structural changes such as white matter atrophy, hippocampal atrophy, and white matter lesion progression on gait speed decline over approximately two and a half years among 225 adults aged 60 to 86 year (Callisaya et al., 2013). Compared with that study, the current study used region-specific MRI analysis in a larger study population of 767 community-dwelling adults from a general population over a longer follow-up period. Another four-year longitudinal study among 2450 adults with an age ≥ 60 years did not use follow-up MRI data to evaluate brain volume changes but determined the risk of one MRI examination on incident physical impairment events. They showed a significantly increased risk
Acknowledgements We thank all the study participants as well as the research staff at the Institute of Human Genomic Study at the Ansan Hospital of Korea University and Central Hospital for their contributions to the data collection. 29
S. Lee et al.
Experimental Gerontology 118 (2019) 26–30
Funding
predictors of incident dementia. J. Gerontol. A Biol. Sci. Med. Sci. 72, 655–661. Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res. 12, 189–198. Ge, Y., Grossman, R.I., Babb, J.S., Rabin, M.L., Mannon, L.J., Kolson, D.L., 2002. Agerelated total gray matter and white matter changes in normal adult brain. Part I: volumetric MR imaging analysis. AJNR Am. J. Neuroradiol. 23, 1327–1333. Giorgio, A., Santelli, L., Tomassini, V., Bosnell, R., Smith, S., De Stefano, N., JohansenBerg, H., 2010. Age-related changes in grey and white matter structure throughout adulthood. NeuroImage 51, 943–951. Jack Jr., C.R., Knopman, D.S., Jagust, W.J., Petersen, R.C., Weiner, M.W., Aisen, P.S., Shaw, L.M., Vemuri, P., Wiste, H.J., Weigand, S.D., Lesnick, T.G., Pankratz, V.S., Donohue, M.C., Trojanowski, J.Q., 2013. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 12, 207–216. Kang, Y., Na, D.L., Hahn, S., 1997. A validity study on the Korean mini-mental state examination (K-MMSE) in dementia patients. J. Korean Neurol. Assoc. 15, 300–308. Kim, R.E., Yun, C.H., Thomas, R.J., Oh, J.H., Johnson, H.J., Kim, S., Lee, S., Seo, H.S., Shin, C., 2018. Lifestyle-dependent brain change: a longitudinal cohort MRI study. Neurobiol. Aging 69, 48–57. Lee, S., Shin, C., 2017. Interaction of obstructive sleep apnoea and cognitive impairment with slow gait speed in middle-aged and older adults. Age Ageing 46, 653–659. Marquis, S., Moore, M.M., Howieson, D.B., Sexton, G., Payami, H., Kaye, J.A., Camicioli, R., 2002. Independent predictors of cognitive decline in healthy elderly persons. Arch. Neurol. 59, 601–606. Mielke, M.M., Roberts, R.O., Savica, R., Cha, R., Drubach, D.I., Christianson, T., Pankratz, V.S., Geda, Y.E., Machulda, M.M., Ivnik, R.J., Knopman, D.S., Boeve, B.F., Rocca, W.A., Petersen, R.C., 2013. Assessing the temporal relationship between cognition and gait: slow gait predicts cognitive decline in the Mayo Clinic study of aging. J. Gerontol. A Biol. Sci. Med. Sci. 68, 929–937. Munoz-Mendoza, C.L., Cabanero-Martinez, M.J., Millan-Calenti, J.C., Cabrero-Garcia, J., Lopez-Sanchez, R., Maseda-Rodriguez, A., 2011. Reliability of 4-m and 6-m walking speed tests in elderly people with cognitive impairment. Arch. Gerontol. Geriatr. 52, e67–e70. National Institutes of Health (NIH), 2018. Toolbox 4-meter Walk Gait Speed Test. [Cited: May 9, 2018] Available from: http://www.nihtoolbox.org/WhatAndWhy/Motor/ Locomotion/Pages/NIH-Toolbox-4–Meter-Walk-Gait-Speed-Test.aspx. Peel, N.M., Kuys, S.S., Klein, K., 2013. Gait speed as a measure in geriatric assessment in clinical settings: a systematic review. J. Gerontol. A Biol. Sci. Med. Sci. 68, 39–46. Perera, S., Mody, S.H., Woodman, R.C., Studenski, S.A., 2006. Meaningful change and responsiveness in common physical performance measures in older adults. J. Am. Geriatr. Soc. 54, 743–749. Rosano, C., Kuller, L.H., Chung, H., Arnold, A.M., Longstreth Jr., W.T., Newman, A.B., 2005. Subclinical brain magnetic resonance imaging abnormalities predict physical functional decline in high-functioning older adults. J. Am. Geriatr. Soc. 53, 649–654. Rosano, C., Studenski, S.A., Aizenstein, H.J., Boudreau, R.M., Longstreth Jr., W.T., Newman, A.B., 2012. Slower gait, slower information processing and smaller prefrontal area in older adults. Age Ageing 41, 58–64. Studenski, S., Perera, S., Patel, K., Rosano, C., Faulkner, K., Inzitari, M., Brach, J., Chandler, J., Cawthon, P., Connor, E.B., Nevitt, M., Visser, M., Kritchevsky, S., Badinelli, S., Harris, T., Newman, A.B., Cauley, J., Ferrucci, L., Guralnik, J., 2011. Gait speed and survival in older adults. JAMA 305, 50–58. Taylor, M.E., Lasschuit, D.A., Lord, S.R., Delbaere, K., Kurrle, S.E., Mikolaizak, A.S., Kvelde, T., Close, J.C.T., 2017. Slow gait speed is associated with executive function decline in older people with mild to moderate dementia: a one year longitudinal study. Arch. Gerontol. Geriatr. 73, 148–153. Waite, L.M., Grayson, D.A., Piguet, O., Creasey, H., Bennett, H.P., Broe, G.A., 2005. Gait slowing as a predictor of incident dementia: 6-year longitudinal data from the Sydney older persons study. J. Neurol. Sci. 229–230, 89–93. Watson, N.L., Rosano, C., Boudreau, R.M., Simonsick, E.M., Ferrucci, L., Sutton-Tyrrell, K., Hardy, S.E., Atkinson, H.H., Yaffe, K., Satterfield, S., Harris, T.B., Newman, A.B., Health, A.B.C.S., 2010. Executive function, memory, and gait speed decline in wellfunctioning older adults. J. Gerontol. A Biol. Sci. Med. Sci. 65, 1093–1100.
This work was supported by a fund from the Korea Centers for Disease Control and Prevention (2011-E71004-00, 2012-E71005-00, 2013-E71005-00, 2014-E71003-00, 2015-P71001-00, 2016-E71003-00, 2017-E71001-00) and was provided with bioresources from the National Biobank of Korea, the Centers for Disease Control and Prevention, Republic of Korea (KBP-2018-002). This study was supported by a National Research Foundation of Korean (NRF) grant funded by the Korean government (NRF-2017R1D1A1B03036232). This study was also supported by 2018 Research Grant from Kangwon National University. Author contributions SL contributed to the concept and design of the study, the analysis and interpretation of data, draft writing, and preparation of the manuscript. EYK contributed to the data acquisition. CS contributed to the conception, acquisition and interpretation of the data, and the critical review of the manuscript. Declaration of interest None. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.exger.2019.01.004. References Arenaza-Urquijo, E.M., de Flores, R., Gonneaud, J., Wirth, M., Ourry, V., Callewaert, W., Landeau, B., Egret, S., Mezenge, F., Desgranges, B., Chetelat, G., 2017. Distinct effects of late adulthood cognitive and physical activities on gray matter volume. Brain Imaging Behav. 11, 346–356. Blumen, H.M., Brown, L.L., Habeck, C., Allali, G., Ayers, E., Beauchet, O., Callisaya, M., Lipton, R.B., Mathuranath, P.S., Phan, T.G., Pradeep Kumar, V.G., Srikanth, V., Verghese, J., 2018. Gray matter volume covariance patterns associated with gait speed in older adults: a multi-cohort MRI study. Brain Imaging Behav. http://lps3. doi.org.libproxy.snu.ac.kr/10.1007/s11682-018-9871-7. Buracchio, T., Dodge, H.H., Howieson, D., Wasserman, D., Kaye, J., 2010. The trajectory of gait speed preceding mild cognitive impairment. Arch. Neurol. 67, 980–986. Callisaya, M.L., Beare, R., Phan, T.G., Blizzard, L., Thrift, A.G., Chen, J., Srikanth, V.K., 2013. Brain structural change and gait decline: a longitudinal population-based study. J. Am. Geriatr. Soc. 61, 1074–1079. Callisaya, M.L., Beare, R., Phan, T.G., Chen, J., Srikanth, V.K., 2014. Global and regional associations of smaller cerebral gray and white matter volumes with gait in older people. PLoS One 9, e84909. Doi, T., Shimada, H., Park, H., Makizako, H., Tsutsumimoto, K., Uemura, K., Nakakubo, S., Hotta, R., Suzuki, T., 2015. Cognitive function and falling among older adults with mild cognitive impairment and slow gait. Geriatr Gerontol Int 15, 1073–1078. Dumurgier, J., Artaud, F., Touraine, C., Rouaud, O., Tavernier, B., Dufouil, C., SinghManoux, A., Tzourio, C., Elbaz, A., 2017. Gait speed and decline in gait speed as
30
Contents lists available at ScienceDirect
Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero
Longitudinal association between brain volume change and gait speed in a general population Sunghee Leea, Eun Young Kimb, Chol Shinb,c,
T
⁎
a
Department of Food and Nutrition, College of Health Science, Kangwon National University, Samcheok, Republic of Korea Institute of Human Genomic Study, School of Medicine, Korea University Ansan Hospital, Ansan, Republic of Korea c Division of Pulmonary, Sleep and Critical Care Medicine, Department of Internal Medicine, Korea University, Ansan Hospital, Ansan, Republic of Korea b
ARTICLE INFO
ABSTRACT
Keywords: Gait speed Brain volume General population
Objective: To determine the association between brain structural changes and gait speed in a four-year longitudinal prospective cohort study. Measurements: A total of 767 well-functioning community-dwelling participants, free of arthritis, silent infarct, stroke, dementia, head injury, and cancer, completed baseline brain magnetic resonance imaging scan and gait speed tests between 2011 and 2014, and follow-up tests between 2015 and 2017. The gait test consisted of measuring the elapsed time to walk four meters at usual speed. To estimate whether brain volume changes predict gait speed decline at follow-up, a generalized linear regression model was used after adjusting for potential confounding factors including gait speed at baseline. Results: Participants who experienced ≥0.05 m/s gait speed decline, previously defined as a clinically meaningful decline, were more likely to be women, less likely to be smokers, and had lower physical activity scores (p = 0.003, p = 0.025, and p = 0.006, respectively), as compared to those who did not experience the decline. Also, they demonstrated smaller volumes of hippocampus, total gray matter, parietal gray matter, temporal gray matter, and temporal white matter (p = 0.004, p = 0.042, p = 0.021, p = 0.001, and p = 0.004, respectively). Even after correcting the significance level due to multiple comparisons, overall gray matter and overall white matter volume changes during four-year follow-up period showed significant associations with gait speed at follow-up (p < 0.001 and p = 0.002). Regarding region-specific volumes, frontal white matter and parietal gray matter volume changes demonstrated significant associations with gait speed (p = 0.002, p = 0.004, respectively). Conclusion: In a four-year longitudinal study among 767 well-functioning community-dwelling healthy participants from a general population, we observed a significant association between brain volume changes and gait speed.
1. Introduction Most individuals experience gait speed decline with advanced aging. Slow gait speed has been reported to increase the risk for fall (Doi et al., 2015), cognitive decline (Marquis et al., 2002), dementia (Dumurgier et al., 2017), and death (Studenski et al., 2011) among functionally healthy adults. A gait speed test has been validated as a reliable and simple geriatric assessment tool (Munoz-Mendoza et al., 2011; NIH, 2018; Peel et al., 2013). Longitudinal studies have confirmed a time sequence between slow gait speed and cognitive functional decline as one of causal criteria (Mielke et al., 2013; Waite et al., 2005). The development of a functional decline of the brain, such as
mild cognitive impairment, is preceded by a decline in gait speed by approximately 12 years (Buracchio et al., 2010). Attempts to explain the cognitive functional decline include age-related changes, mainly small vessel lesions in the brain structure of the prefrontal area responsible for executive functions (Rosano et al., 2012; Taylor et al., 2017; Watson et al., 2010). The functional decline of the brain is preceded by its structural changes including volume atrophy with aging (Jack Jr. et al., 2013). Thus, early identification of brain structural change preceding cognitive functional decline may provide a wider window of opportunity to implement an active intervention to decrease the risk of cognitive impairment. Particularly, age-related brain structural changes of the gray
⁎ Corresponding author at: Institute of Human Genomic Study, Ansan Hospital Korea University, 516 Gojan-1-dong, Danwon-gu, Ansan-si, Gyeonggi-do 15355, Republic of Korea. E-mail address: [email protected] (C. Shin).
https://doi.org/10.1016/j.exger.2019.01.004 Received 17 August 2018; Received in revised form 11 November 2018; Accepted 3 January 2019 Available online 04 January 2019 0531-5565/ © 2019 Elsevier Inc. All rights reserved.
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
2.2. Brain volume measurement
matter have been attributed to neural cell loss (Giorgio et al., 2010). Additionally, the frontal, temporal, and hippocampal regions have been reported to be linked with motor performance such as gait speed and other physical activities (Arenaza-Urquijo et al., 2017). Few studies have examined the longitudinal relationship between brain structural region-specific volume change and gait speed decline among well-functioning adults from a general population. Therefore, we investigated an association between brain structural volume change and gait speed. We have previously reported a cross-sectional association between slow gait speed and cognitive functional decline as well as an interaction with obstructive sleep apnea among functionally healthy middle-aged and older adults (Lee and Shin, 2017). Here, we further aimed to investigate an association between brain structural volume changes over a four-year follow-up period and gait speed at follow-up among 767 well-functioning community-dwelling adults aged 49 to 79 years. We hypothesized that individuals with brain atrophy would be more likely to exhibit slow gait speed in a four-year longitudinal study. To test this hypothesis, we used data from the Korean Genome and Epidemiology Study (KoGES).
Using a GE Signa HDxt 1.5 Tesla MRI scanner (GE Healthcare, USA), three-dimensional T1-weighted MRI data were collected between 2011 and 2014 as a baseline measurement. After the first sequence between March 1, 2011 and April 18, 2011, MRI protocols were changed to improve the image quality for the second sequence from April 19, 2011 to April 18, 2013, while retaining the main characteristics of the sequence. More detailed procedure has been published (Kim et al., 2018). Analysis of all images was performed at the Korea University in collaboration with the University of Iowa (Iowa City, IA, USA). Sub-regions of interest, including frontal, parietal, temporal, and occipital regions, were merged from 215 independent sub-regions of the brain. 2.3. Gait speed The gait speed test has been used as a simple and reliable geriatric assessment tool as introduced by the National Institutes of Health [Bethesda, MD, USA (NIH, 2018)]. Its reliability has been validated (Munoz-Mendoza et al., 2011). All participants were asked to walk a four-meter course at usual gait speed from the start point to the end point along a hallway. At the same time, a research staff measured the time to pass the end point of the course using a stopwatch. The gait speed was calculated by dividing four meters by the time in seconds to finish the four-meter walk (meters per seconds, m/s).
2. Methods 2.1. Study population The KoGES is an ongoing longitudinal study initiated in 2001 to investigate the prevalence rates and risk factors of chronic diseases in a general population. A detailed report has been published (Lee and Shin, 2017). At enrollment from 2001 to 2002, 5012 adults aged 40 to 69 years were randomly selected by telephone contacts from a general population in Ansan City, located south of Seoul, the capital of South Korea (Lee and Shin, 2017). On biennial follow-up visits, participants underwent biochemical and physical examinations, and completed an interviewer-administrated questionnaire. Professionally trained research staff carried out these tasks according to standardized study protocols. The study participants underwent height and weight measurements wearing only light clothing and no shoes. On each visit, venipuncture was performed early in the morning after more than eight hours of fasting. Blood samples were sent to the Seoul Clinical Laboratory (Seoul, South Korea). To focus on brain aging, geriatric measurements including brain magnetic resonance imaging (MRI) scan, depression score, cognitive and gait tests, were performed every four years since 2011. At baseline of the brain aging study, 2587 participants underwent an MRI and completed a gait test between 2011 and 2014. From 2015 to 2017, 934 individuals completed a follow-up MRI and a follow-up gait test. The follow-up data were limited because the follow-up MRI examinations have been still ongoing and will be completed in December 2018. Nonetheless, we confirmed no significant difference in the primary outcome of interest between participants who were lost to follow up and those who were successfully followed up (p-value = 0.539, Supplemental Table 1). Of 934 participants with both baseline and follow-up data, 159 individuals were excluded due to silent infarct (n = 68), stroke (n = 8), arthritis (n = 52), or cancer (n = 31). No participant met the additional exclusion criteria head injury or dementia diagnosis. Additionally, eight participants were excluded due to missing covariates such as depression (n = 2) and diabetes mellitus (n = 6). Therefore, 767 participants (375 men and 392 women) were included in the final analyses. A study flow diagram is presented in Supplemental Fig. 1. All study participants were informed about any potential risks and benefits prior to signing an informed written consent form. The current study adhered to the ethics rules for human research described in the Declaration of Helsinki. The Human Subjects Review Committee and the Institutional Review Board reviewed and approved this study (IRB No. 2018AS0015).
2.4. Cognitive measurement To examine cognitive function of the brain, the Mini-Mental State Examination (MMSE), the most widely used global cognitive screening test, was used (Folstein et al., 1975). Its maximum possible score was 30. A Korean version of the MMSE (K-MMSE), developed and validated for the Korean language, was employed (Kang et al., 1997). 2.5. Statistical analysis To present general characteristics of the study participants between at baseline and at follow-up, paired t-test or Wilcoxon signed-rank test were utilized. For categorical variables, McNemar's test was used. Also, baseline characteristics were compared according to either individuals who experienced a clinically meaningful decline in gait speed such as ≥0.05 m/s (Perera et al., 2006) or individuals who did not experience. To estimate the association between brain region-specific volume changes over a four-year period and gait speed at follow-up, generalized linear regression models were utilized, after adjusting for age, body mass index (BMI), education, sex, physical activity (metabolic equivalent task [MET], hours/week), smoking, alcohol use, hypertension, diabetes, high-sensitivity C-reactive protein (hs-CRP), depression (Beck's Depression Index [BDI]), total cholesterol, total intracranial volume, and gait speed at baseline. A two-sided p-value < 0.05 was considered to be statistically significant. A corrected significance level (0.05/9 = 0.0056) was considered due to multiple comparisons among hippocampus, frontal gray matter, frontal white matter, parietal gray matter, parietal white matter, occipital gray matter, occipital white matter, temporal gray matter, and temporal white matter volumes. SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses. 3. Results 3.1. General characteristics of the study participants The general characteristics of the 767 participants are summarized in Table 1. The baseline of the brain aging study from 2011 to 2014 and the follow-up of the study from 2015 to 2017 were completed. At baseline, the average age of the study participants was 58.97 years; at 27
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
Table 1 General characteristics of the study participants (n = 767).
Age, year Female, % Education, % None Elementary Middle school High school University Body mass index, kg/m2 Total cholesterol, mg/dL Smoking, pack-year⁎ Current drinker, % Hypertension, % Diabetes, % hs-CRP, mg/L⁎ Physical activity, MET (h/ week)⁎ Depression, BDI⁎ MMSE Time to walk four meters Usual gait (s) Usual gait speed (m/s) Brain volume, ml Gray matter White matter Hippocampus Frontal Frontal gray matter Frontal white matter Parietal Parietal gray matter Parietal white matter Occipital Occipital gray matter Occipital white matter Temporal Temporal gray matter Temporal white matter Intracranial volume, ml
Baseline Year 2011–2014
Follow-up Year 2015–2017
p-value
58.97 ± 6.44 51.11
62.97 ± 6.43
< 0.001 0.539
1.17 1.83 12.52 61.67 22.82 24.50 ± 3.01 193.78 ± 34.95 0 (0, 15.00) 45.89 34.42 19.69 0.64 (0.35, 1.22) 13.95 (0, 33.00)
24.41 ± 3.04 185.95 ± 35.55 0 (0, 13.50) 43.68 35.85 20.47 0.62 (0.35, 1.11) 13.50 (0, 30.00)
0.022 < 0.001 0.113 0.002 < 0.001 < 0.001 0.572 0.174
7.00 (3.00, 11.00) 27.68 ± 1.89
5.00 (2.00, 9.00) 27.66 ± 1.99
< 0.001 0.852
3.91 ± 0.74 1.06 ± 0.19
3.91 ± 0.70 1.05 ± 0.18
0.993 0.553
707.50 ± 59.29 454.47 ± 43.32 8.34 ± 0.83
698.11 ± 58.02 455.32 ± 44.30 8.35 ± 0.85
< 0.001 < 0.001 0.115
217.42 ± 20.85 164.24 ± 17.50
214.76 ± 20.24 165.33 ± 17.98
< 0.001 < 0.001
132.84 ± 12.48 99.34 ± 10.21
131.28 ± 12.37 99.72 ± 10.45
< 0.001 < 0.001
79.34 ± 8.08 32.00 ± 3.96
76.86 ± 7.85 32.29 ± 4.11
< 0.001 < 0.001
133.51 ± 13.21 55.63 ± 5.83 1375.56 ± 120.59
133.13 ± 12.96 55.15 ± 5.93
< 0.001 < 0.001
Table 2 Baseline characteristics of the study participants according to a clinically meaningful change in usual gait speed (n = 767).
Age, year Female, % Education, % None Elementary Middle school High school University Body mass index, kg/m2 Total cholesterol, mg/dL Smoking, pack-year⁎ Current drinker, % Hypertension, % Diabetes, % hs-CRP, mg/L⁎ Physical activity, MET (h/week)⁎ Depression, BDI⁎ MMSE Brain volume, ml Hippocampus Gray matter White matter Frontal Frontal gray matter Frontal white matter Parietal Parietal gray matter Parietal white matter Occipital Occipital gray matter Occipital white matter Temporal Temporal gray matter Temporal white matter
< 0.001
Mean ± SD; paired t-test; chi-square test; McNemar test; hs-CRP, high sensitivity C-reactive protein; MET, metabolic equivalent task. ⁎ Median (interquartile range) and p-values with Wilcoxon signed-rank test.
follow-up, their average age was 62.97 years. The proportion of women was 51.11%. More than half of these participants graduated from high school (61.67%). Compared to those at baseline, participants at followup had a slightly lower BMI (p = 0.022), lower levels of total cholesterol (p < 0.001), and a smaller proportion of current drinkers (p = 0.002). However, at follow-up, the study participants presented higher prevalence rates of hypertension (34.42% vs. 35.85%; p < 0.001) as well as diabetes mellitus (19.69% vs. 20.47%; p < 0.001) compared to the rates at baseline. As for depression, participants had lower BDI scores at follow-up compared to those at baseline (p < 0.001). However, < 10 BDI scores were not considered as clinically meaningful. The brain volume of the gray matter decreased over the four-year follow-up (p < 0.001), whereas the volume of the white matter increased (p < 0.001). Regarding brain sub-regional volume changes, the gray matter volumes in all four investigated areas including the frontal, parietal, occipital, and temporal region were reduced at follow-up (all p < 0.001). Conversely, white matter volumes were not reduced, except for the temporal area.
Participants with a clinically meaningful decline⁎⁎ (n = 310)
Participants without a meaningful decline (n = 457)
p-value
58.83 ± 6.78 57.74
59.06 ± 6.20 46.61
0.632 0.003
1.29 2.26 12.90 62.90 20.65 24.42 ± 3.21
1.09 1.53 12.25 60.83 24.29 24.55 ± 2.86
0.762
0.580
193.78 ± 34.20
193.77 ± 35.49
0.997
0 (0, 10.00) 42.90 33.23 18.06 0.65 (0.35, 1.26) 11.30 (0, 27.90)
0 (0, 18.00) 47.92 35.23 20.79 0.64 (0.35, 1.20) 16.00 (0, 36.00)
0.025 0.171 0.567 0.352 0.753 0.006
7.00 (3.00, 12.00) 27.75 ± 1.88
6.00 (3.00, 11.00) 27.63 ± 1.89
0.426 0.416
8.23 ± 0.84 702.21 ± 59.96 450.77 ± 44.43 378.86 ± 37.27 215.83 ± 20.71
8.41 ± 0.82 711.08 ± 58.63 456.98 ± 42.41 383.56 ± 36.73 218.49 ± 20.90
0.004 0.042 0.052
163.03 ± 17.72
165.07 ± 17.32
0.115
229.94 ± 22.00 131.58 ± 12.44
233.71 ± 21.69 133.70 ± 12.44
0.021
98.36 ± 10.27
100.01 ± 10.13
0.028
110.54 ± 12.21 78.81 ± 8.38
111.88 ± 11.38 79.70 ± 7.85
0.134
31.73 ± 4.11
32.18 ± 3.84
0.122
186.56 ± 18.48 131.66 ± 13.05
190.88 ± 18.45 134.76 ± 13.18
0.001
54.89 ± 5.89
56.12 ± 5.75
0.004
0.082
Mean ± SD; paired t-test; chi-square test; hs-CRP, high sensitivity C-reactive protein; McNemar's test; MET, metabolic equivalent task. ⁎ Median (interquartile range) with p-values from Wilcoxon signed-rank test. ⁎⁎ The clinically meaningful change was defined as 0.05 m/s (Perera et al., 2006).
2006) of > 0.05 m/s in gait speed or who did not experienced that decline are presented in Table 2. Individuals with the clinically meaningful decline in gait speed were more likely to be women, less likely to be a smoker, and had a lower physical activity score (p = 0.003, p = 0.025, and p = 0.006, respectively). Regarding the brain volumes, participants who experienced a clinically meaningful change in gait speed showed smaller volumes in the hippocampus, the total gray matter, the parietal gray matter, the temporal gray matter, and the temporal white matter (p = 0.004, p = 0.042, p = 0.021, p = 0.001, and p = 0.004, respectively).
3.2. Baseline characteristics of the study participants according to a clinically meaningful change in usual gait speed
3.3. Significant association of brain structural change on slow gait speed Table 3 shows the association between brain region-specific structural volume changes and gait speed at follow-up among the 767 study participants. Total gray matter and total white matter volume changes
Comparisons of general characteristics according to participants either who experienced a clinically meaningful decline (Perera et al., 28
Experimental Gerontology 118 (2019) 26–30
S. Lee et al.
for incident physical impairment including slow gait speed with brain abnormalities including enlarged ventricular, subclinical brain infarcts, and white matter hyperintensities (Rosano et al., 2005). There have also been several cross-sectional studies examining the association between brain volume and gait speed (Blumen et al., 2018; Callisaya et al., 2014; Rosano et al., 2012). One of these studies involving 214 individuals reported a significant association between a small prefrontal area volume and a slow gait speed (Rosano et al., 2012). Using voxel-based morphometry, another cross-sectional study among 305 adults aged > 60 years demonstrated a significant association between smaller total gray matter volume in brain regions such as frontal, parietal, occipital, temporal, cingulate, insula, parahippocampal, and cerebellar areas, and slower gait speed (Callisaya et al., 2014). Using voxel-based morphometry and multivariate covariate patterns, another cross-sectional study among 352 non-demented community-dwelling older adults from three different cohorts investigated the relationships between gray matter volume and gait speed, explaining with age-related changes in shared neural regions attributed to processing speed and gait speed (Blumen et al., 2018). The underlying mechanism on how the brain structure affects slow gait speed has not been fully elucidated. However, previous studies suggested that both brain structural atrophy and slow gait speed share the same neural regions vulnerable to age-related changes (Blumen et al., 2018). While brain atrophy of the gray matter is known to consistently progress from early stages of the adult life, white matter volume atrophy has been recognized to start at a much later stage of the adult life with a greater rate of atrophic degeneration (Ge et al., 2002). Further studies are warranted to disentangle the underlying mechanisms. We observed that participants who experienced a clinically meaningful decline ≥0.05 m/s were less likely to smoke. This is because they had higher proportion of women (57.74%, 0 pack-year), as compared to those who did not experience the decline had lower proportion of women (46.61%, 0 pack-year). Additionally, we tried adjusting for ‘total change in brain volume’ as a covariate, although this procedure might cause over-adjustment because each regional volume change was already included in the total change of the brain volume. No significant association was observed, when considering the corrected significance level (Supplemental Table 2). The present study has several strengths. First, we utilized a fouryear longitudinal prospective cohort study design that elucidates the time sequence between structural brain volume changes and gait speed decline to investigate a causal relationship. Second, our study population consisted of well-functioning, community-dwelling, middle-aged and older adults from a general population, which has the advantage of generalizability. Third, our participants were relatively young to exhibit brain functional cognitive impairment, which turned out to be a favorable time for investigating brain structural change. These strengths enable us to investigate the early stages of brain structural changes before cognitive functional decline starts to occur. Finally, we measured region-specific brain volumes including the frontal, parietal, occipital, and temporal regions. However, this study also had a limitation that needs to be considered when interpreting our findings. We did not investigate all brain regions; our study did not include the cerebellum or thalamus. In conclusion, in this four-year longitudinal study among 767 wellfunctioning community-dwelling healthy participants from a general population, we observed significant associations between brain volume changes and gait speed.
Table 3 Association between brain volume changes and usual gait speed at follow-up. Brain volume change, ml
ΔGray matter volume ΔWhite matter volume ΔHippocampus ΔFrontal Frontal gray matter Frontal white matter ΔParietal Parietal gray matter Parietal white matter ΔOccipital Occipital gray matter Occipital white matter ΔTemporal Temporal gray matter Temporal white matter
Time to walk four meters at usual gait speed, seconds (n = 767) β
p-value
0.01 0.01 −0.21
< 0.001 0.002 0.167
0.01 0.03
0.021 0.002
0.03 0.04
0.004 0.006
0.03 0.06
0.009 0.063
0.06 0.02
0.063 0.455
Adjusted for age, sex, body mass index, education, smoking, drinking, physical activity, hypertension, diabetes, total cholesterol, depression score, total intracranial volume in ml, hs-CRP, and gait speed at baseline; Δ, the change in the brain volume was defined as the volume at baseline subtracted by the volume at follow-up.
over the four-year follow up indicated significant associations with gait speed (p < 0.001 and p = 0.002, respectively). Regarding region-specific volumes, frontal gray matter, frontal white matter, parietal gray matter, parietal white matter, and occipital gray matter volume changes demonstrated significant associations with gait speed at follow-up (p = 0.021, p = 0.002, p = 0.004, p = 0.006, and p = 0.009, respectively). Even after considering the corrected significance level due to multiple comparisons (0.05/9 = 0.0056), the associations with overall gray matter, overall white matter, frontal white matter, and parietal gray matter volume changes, on gait speed remained statistically significant (p < 0.001, p = 0.002, p = 0.002, and p = 0.004). 4. Discussion In this four-year longitudinal prospective cohort study, we found that gray matter and white matter volume changes over the four-year follow-up period indicated significant associations on usual gait speed. At the level of brain sub-regions, frontal white matter and parietal gray matter volume changes demonstrated significant associations with gait speed at follow-up. Our current findings are unique in that we investigated the association between brain region-specific volume changes and gait speed among well-functioning community-dwelling middle-aged and older adults from the general population in a four-year longitudinal follow-up study. Our analysis of the follow-up MRI data from a general population helps us better understand the association between brain atrophy and slow gait speed in aging. Previous studies examined gait characteristics associated with brain abnormalities. A study examined the association of brain volume changes on slow gait speed using MRI data (Callisaya et al., 2013). This study reported a longitudinal association of brain structural changes such as white matter atrophy, hippocampal atrophy, and white matter lesion progression on gait speed decline over approximately two and a half years among 225 adults aged 60 to 86 year (Callisaya et al., 2013). Compared with that study, the current study used region-specific MRI analysis in a larger study population of 767 community-dwelling adults from a general population over a longer follow-up period. Another four-year longitudinal study among 2450 adults with an age ≥ 60 years did not use follow-up MRI data to evaluate brain volume changes but determined the risk of one MRI examination on incident physical impairment events. They showed a significantly increased risk
Acknowledgements We thank all the study participants as well as the research staff at the Institute of Human Genomic Study at the Ansan Hospital of Korea University and Central Hospital for their contributions to the data collection. 29
S. Lee et al.
Experimental Gerontology 118 (2019) 26–30
Funding
predictors of incident dementia. J. Gerontol. A Biol. Sci. Med. Sci. 72, 655–661. Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res. 12, 189–198. Ge, Y., Grossman, R.I., Babb, J.S., Rabin, M.L., Mannon, L.J., Kolson, D.L., 2002. Agerelated total gray matter and white matter changes in normal adult brain. Part I: volumetric MR imaging analysis. AJNR Am. J. Neuroradiol. 23, 1327–1333. Giorgio, A., Santelli, L., Tomassini, V., Bosnell, R., Smith, S., De Stefano, N., JohansenBerg, H., 2010. Age-related changes in grey and white matter structure throughout adulthood. NeuroImage 51, 943–951. Jack Jr., C.R., Knopman, D.S., Jagust, W.J., Petersen, R.C., Weiner, M.W., Aisen, P.S., Shaw, L.M., Vemuri, P., Wiste, H.J., Weigand, S.D., Lesnick, T.G., Pankratz, V.S., Donohue, M.C., Trojanowski, J.Q., 2013. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 12, 207–216. Kang, Y., Na, D.L., Hahn, S., 1997. A validity study on the Korean mini-mental state examination (K-MMSE) in dementia patients. J. Korean Neurol. Assoc. 15, 300–308. Kim, R.E., Yun, C.H., Thomas, R.J., Oh, J.H., Johnson, H.J., Kim, S., Lee, S., Seo, H.S., Shin, C., 2018. Lifestyle-dependent brain change: a longitudinal cohort MRI study. Neurobiol. Aging 69, 48–57. Lee, S., Shin, C., 2017. Interaction of obstructive sleep apnoea and cognitive impairment with slow gait speed in middle-aged and older adults. Age Ageing 46, 653–659. Marquis, S., Moore, M.M., Howieson, D.B., Sexton, G., Payami, H., Kaye, J.A., Camicioli, R., 2002. Independent predictors of cognitive decline in healthy elderly persons. Arch. Neurol. 59, 601–606. Mielke, M.M., Roberts, R.O., Savica, R., Cha, R., Drubach, D.I., Christianson, T., Pankratz, V.S., Geda, Y.E., Machulda, M.M., Ivnik, R.J., Knopman, D.S., Boeve, B.F., Rocca, W.A., Petersen, R.C., 2013. Assessing the temporal relationship between cognition and gait: slow gait predicts cognitive decline in the Mayo Clinic study of aging. J. Gerontol. A Biol. Sci. Med. Sci. 68, 929–937. Munoz-Mendoza, C.L., Cabanero-Martinez, M.J., Millan-Calenti, J.C., Cabrero-Garcia, J., Lopez-Sanchez, R., Maseda-Rodriguez, A., 2011. Reliability of 4-m and 6-m walking speed tests in elderly people with cognitive impairment. Arch. Gerontol. Geriatr. 52, e67–e70. National Institutes of Health (NIH), 2018. Toolbox 4-meter Walk Gait Speed Test. [Cited: May 9, 2018] Available from: http://www.nihtoolbox.org/WhatAndWhy/Motor/ Locomotion/Pages/NIH-Toolbox-4–Meter-Walk-Gait-Speed-Test.aspx. Peel, N.M., Kuys, S.S., Klein, K., 2013. Gait speed as a measure in geriatric assessment in clinical settings: a systematic review. J. Gerontol. A Biol. Sci. Med. Sci. 68, 39–46. Perera, S., Mody, S.H., Woodman, R.C., Studenski, S.A., 2006. Meaningful change and responsiveness in common physical performance measures in older adults. J. Am. Geriatr. Soc. 54, 743–749. Rosano, C., Kuller, L.H., Chung, H., Arnold, A.M., Longstreth Jr., W.T., Newman, A.B., 2005. Subclinical brain magnetic resonance imaging abnormalities predict physical functional decline in high-functioning older adults. J. Am. Geriatr. Soc. 53, 649–654. Rosano, C., Studenski, S.A., Aizenstein, H.J., Boudreau, R.M., Longstreth Jr., W.T., Newman, A.B., 2012. Slower gait, slower information processing and smaller prefrontal area in older adults. Age Ageing 41, 58–64. Studenski, S., Perera, S., Patel, K., Rosano, C., Faulkner, K., Inzitari, M., Brach, J., Chandler, J., Cawthon, P., Connor, E.B., Nevitt, M., Visser, M., Kritchevsky, S., Badinelli, S., Harris, T., Newman, A.B., Cauley, J., Ferrucci, L., Guralnik, J., 2011. Gait speed and survival in older adults. JAMA 305, 50–58. Taylor, M.E., Lasschuit, D.A., Lord, S.R., Delbaere, K., Kurrle, S.E., Mikolaizak, A.S., Kvelde, T., Close, J.C.T., 2017. Slow gait speed is associated with executive function decline in older people with mild to moderate dementia: a one year longitudinal study. Arch. Gerontol. Geriatr. 73, 148–153. Waite, L.M., Grayson, D.A., Piguet, O., Creasey, H., Bennett, H.P., Broe, G.A., 2005. Gait slowing as a predictor of incident dementia: 6-year longitudinal data from the Sydney older persons study. J. Neurol. Sci. 229–230, 89–93. Watson, N.L., Rosano, C., Boudreau, R.M., Simonsick, E.M., Ferrucci, L., Sutton-Tyrrell, K., Hardy, S.E., Atkinson, H.H., Yaffe, K., Satterfield, S., Harris, T.B., Newman, A.B., Health, A.B.C.S., 2010. Executive function, memory, and gait speed decline in wellfunctioning older adults. J. Gerontol. A Biol. Sci. Med. Sci. 65, 1093–1100.
This work was supported by a fund from the Korea Centers for Disease Control and Prevention (2011-E71004-00, 2012-E71005-00, 2013-E71005-00, 2014-E71003-00, 2015-P71001-00, 2016-E71003-00, 2017-E71001-00) and was provided with bioresources from the National Biobank of Korea, the Centers for Disease Control and Prevention, Republic of Korea (KBP-2018-002). This study was supported by a National Research Foundation of Korean (NRF) grant funded by the Korean government (NRF-2017R1D1A1B03036232). This study was also supported by 2018 Research Grant from Kangwon National University. Author contributions SL contributed to the concept and design of the study, the analysis and interpretation of data, draft writing, and preparation of the manuscript. EYK contributed to the data acquisition. CS contributed to the conception, acquisition and interpretation of the data, and the critical review of the manuscript. Declaration of interest None. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.exger.2019.01.004. References Arenaza-Urquijo, E.M., de Flores, R., Gonneaud, J., Wirth, M., Ourry, V., Callewaert, W., Landeau, B., Egret, S., Mezenge, F., Desgranges, B., Chetelat, G., 2017. Distinct effects of late adulthood cognitive and physical activities on gray matter volume. Brain Imaging Behav. 11, 346–356. Blumen, H.M., Brown, L.L., Habeck, C., Allali, G., Ayers, E., Beauchet, O., Callisaya, M., Lipton, R.B., Mathuranath, P.S., Phan, T.G., Pradeep Kumar, V.G., Srikanth, V., Verghese, J., 2018. Gray matter volume covariance patterns associated with gait speed in older adults: a multi-cohort MRI study. Brain Imaging Behav. http://lps3. doi.org.libproxy.snu.ac.kr/10.1007/s11682-018-9871-7. Buracchio, T., Dodge, H.H., Howieson, D., Wasserman, D., Kaye, J., 2010. The trajectory of gait speed preceding mild cognitive impairment. Arch. Neurol. 67, 980–986. Callisaya, M.L., Beare, R., Phan, T.G., Blizzard, L., Thrift, A.G., Chen, J., Srikanth, V.K., 2013. Brain structural change and gait decline: a longitudinal population-based study. J. Am. Geriatr. Soc. 61, 1074–1079. Callisaya, M.L., Beare, R., Phan, T.G., Chen, J., Srikanth, V.K., 2014. Global and regional associations of smaller cerebral gray and white matter volumes with gait in older people. PLoS One 9, e84909. Doi, T., Shimada, H., Park, H., Makizako, H., Tsutsumimoto, K., Uemura, K., Nakakubo, S., Hotta, R., Suzuki, T., 2015. Cognitive function and falling among older adults with mild cognitive impairment and slow gait. Geriatr Gerontol Int 15, 1073–1078. Dumurgier, J., Artaud, F., Touraine, C., Rouaud, O., Tavernier, B., Dufouil, C., SinghManoux, A., Tzourio, C., Elbaz, A., 2017. Gait speed and decline in gait speed as
30