- Email: [email protected]
Daily muscle activity and quiescence in non-frail, pre-frail, and frail older women
Experimental Gerontology 45 (2010) 909–917
Contents lists available at ScienceDirect
Experimental Gerontology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ex p g e r o
Daily muscle activity and quiescence in non-frail, pre-frail, and frail older women Olga Theou a, Gareth R. Jones b, Anthony A. Vandervoort a, Jennifer M. Jakobi b,⁎ a b
Graduate Program in Health and Rehabilitation Sciences, University of Western Ontario, London, Ontario, Canada Human Kinetics — Faculty of Health and Social Development, University of British Columbia Okanagan, Kelowna, British Columbia, Canada
a r t i c l e
i n f o
Article history: Received 13 February 2010 Received in revised form 16 August 2010 Accepted 16 August 2010 Available online 22 August 2010 Section Editor: Dr. Christiaan Leeuwenburgh Keywords: Frailty Pre-frailty Aging Older adults Muscle activity Muscle quiescence EMG Long-term EMG Burst analysis Gap analysis
a b s t r a c t Reduced muscle mass and strength are likely fundamental components of frailty. The purpose of this study was to measure daily muscle activity and quiescence in non-frail, pre-frail, and frail older women using portable electromyography (EMG). Thirty-three community-dwelling older Greek women were categorized as non-frail (n = 10, 74 ± 4 years), pre-frail (n = 11, 75 ± 4 years), and frail (n = 12, 81 ± 6 years) based upon the frailty phenotype. Surface EMG over a 9-hour typical day was recorded from the biceps brachii, triceps brachii, vastus lateralis, and biceps femoris of the dominant side. Burst and gap analysis was used to quantify muscle activity and quiescence. The total duration of the muscles that were active (~ 2.5 h) and quiescent (~ 4 h) did not differ across frailty groups. However, the number of bursts was 28% fewer and the mean burst duration was 26% longer in frail women compared with the non-frail women. In addition, muscle activity was greater in the arm muscles than the thigh muscles across all groups (e.g. ~ 60% greater burst percentage). Burst number and duration indicate that muscle activity differs across stages of frailty in older women and is greater in the upper compared with lower limbs. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Research into frailty has recently become a key focus of gerontology research. Knowledge surrounding this geriatric syndrome within the past decade has increased exponentially (Bauer and Sieber, 2008). The criteria used for identification of frailty continue to be a matter of debate; however, the most commonly used approaches to qualify this syndrome are the Frailty Phenotype (Fried et al., 2001) and the Frailty Index (Rockwood and Mitnitski, 2007). The development of frailty progresses across a spectrum of defined stages described as non-frail, pre-frail, and frail (Bandeen-Roche et al., 2006; Fried et al., 2001).While frailty often culminates in the need for institutional care (Fried et al., 2004), many frail older adults still remain in the community despite impairments in one or more activities of daily living (ADL) (Chandler et al., 1998). This syndrome is more common in women than men and transitioning toward frailty likely occurs earlier in women because of greater age-related decline of muscle mass and strength (Cesari et al., 2006; Fried et al., 2001). Greece has one of the oldest populations in Europe [19.2% of population over 65 years of age (Central Intelligence Agency, 2009)] ⁎ Corresponding author. Faculty of Health and Social Development, University of British Columbia — Okanagan, 3333 University Way, Kelowna, BC Canada. V1V 1V7. Tel.: + 1 250 807 9884; fax: + 1 250 807 8085. E-mail address: [email protected] (J.M. Jakobi). 0531-5565/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.exger.2010.08.008
where the prevalence of frailty and pre-frailty in community-dwelling older adults is reported to be 15% and 45%, respectively (SantosEggimann et al., 2009). In North America, in those over the age of 65 years 17% are considered frail and over 46% may already exhibit pre-frailty characteristics (Fried et al., 2001). Fried et al. (2001) utilized low isometric handgrip strength as an indicator of frailty. While handgrip has been proposed as a good predictor of health related events (Rantanen et al., 2000), it only measures upper limb strength and may not entirely capture lower limb strength (Cesari et al., 2006). It is well established that agerelated decline in strength is not similar across all muscles (Candow and Chillibeck, 2005; Dalton et al., 2010; Frontera et al., 2000; Vandervoort, 2002), therefore when examining frailty various muscles need to be considered. To our knowledge, no investigation has yet examined if frailty is associated with changes in muscle activity based upon anatomical location (upper or lower limb) or functional movement (flexion, extension). In addition, physical activity levels should also be considered. Physical activity interacts with the natural process of aging and is known to alter the rate of age-related progressive decline in muscle function (Paterson et al., 2007). Muscle activity and quiescence, termed low-threshold electromyography (EMG), was successfully used to compare muscle function in healthy community-living older adults relative to young adults (Harwood et al., 2008), as well as with working adults to explore
910
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
job-related injuries (Blangsted et al., 2003, 2004; Holtermann et al., 2008; Laursen et al., 2001; Madeleine et al., 2006). No studies have yet examined EMG during daily life in frail persons. Daily activities are often reported to be “hard work” for most frail adults, and often this ‘work’ results in falls and injuries. Measuring muscle activity during daily life might contribute to understanding how older adults use their upper limbs relative to their lower limbs. Kern et al. (2001) reported in young adults that the upper limb muscles were more active than the lower limb muscles during daily activities. In addition, daily upper and lower limb muscle activity and quiescence is a result of an interaction of several systems (e.g. muscular and nervous system) and may be a more complete indicator of frailty than handgrip strength. Laboratory and functional performance measures are important, but limited for this population. Most laboratory tests developed for older adults are not applicable to frail adults and often frail older adults are unable to attend a laboratory for testing due to their impaired health. Measurements that will capture the daily life of the frail person in the home environment are needed. Recent studies measuring muscle activity during daily life and discrete tasks reported muscle activity was greater in non-frail older adults relative to young adults and these age-related differences were greater in women compared with men (Harwood et al., 2008; Jakobi et al., 2008). Muscle activity and quiescence recorded during daily life in older women across stages of frailty is unknown, but could contribute to our understanding of the progression of this syndrome. The purpose of this study was to determine whether muscle activity and quiescence recorded over a 9-hour typical day in upper and lower limb muscles differs between non-frail, pre-frail and frail older women. Women were investigated since the development of frailty is sex-specific (Fried et al., 2001). It was hypothesized that community-dwelling, older women would have greater muscle activity in upper limbs compared to lower limbs. As a result of reduced muscle strength, a higher proportion of the muscle is needed to execute a movement therefore muscle activity would increase across stages of frailty.
2. Methods Thirty-three community-dwelling women aged 68–90 years were recruited within the rural prefecture of Thessaloniki, Greece. Initially, participants were recruited through word-of-mouth, these individuals then recommended additional participants from their acquaintances. Individuals, who wore the equipment, also generated additional participant's interest in the study. The study was approved by the University of Western Ontario ethics board, in accordance with the declaration of Helsinki, and informed consent was received prior to participation. Inclusion criteria were women older than 65 years of age who were living in the local community. During weekdays the researcher visited the home of the participants on two separate and subsequent days. On the first day a health history questionnaire was administered, the stage of frailty phenotype determined (Fried et al., 2001), and muscle strength measured. On day two, the same researcher arrived approximately 1 h after the participant awoke. An EMG device and accelerometer were attached to the participant and Maximal Voluntary Exertions (MVE) for each muscle of interest were performed. Participants were then instructed to proceed with their normal daily activities while wearing the portable EMG and accelerometer. Participants were also asked not to bath or exercise vigorously in order to prevent dislodging the electrodes and damage to the recording device. The researcher encouraged participants to disregard the equipment and go about their normal typical day. Approximately 9–10 h later, on day two, the researcher returned to the participant's home to remove the equipment.
2.1. Frailty definition Physical frailty was defined in accordance with the Frailty Phenotype (Fried et al. 2001), as a clinical syndrome in which the participant expressed three or more of the five criteria below. Participants with one or two criteria were classified as pre-frail, whereas those exhibiting no criteria were considered non-frail. 1) Weight loss: A positive response to the question “In the last year, have you lost more than 5 kg unintentionally (i.e., not due to dieting or exercise)”. 2) Muscle strength: The highest of three consecutive maximal handgrip strength measures of the dominant hand using a Jamar® hand-held dynamometer. Cutoff scores were applied based upon body mass index (BMI ≤ 23, cutoff strength ≤ 17 kg; BMI 23.1–26, cutoff strength ≤ 17.3 kg; BMI 26.1–29 cutoff strength ≤ 18 kg; BMI N29 cutoff strength ≤ 21 kg). 3) Walking speed: Time to walk 15 ft at usual pace. The cutoff scores were applied based upon height (Height ≤ 159 cm, cutoff time ≥ 7 s; Height N 159 cm, cutoff time ≥ 6 s). 4) Physical activity: A weighted score of kilocalories expended per week was calculated based on participant's response to the Short version of the Minnesota Leisure Time Activity Questionnaire (cutoff b270 Kcals per week). 5) Subjective fatigue: Responding to the questions “How often in the last week did you feel that everything you did was an effort?” or “How often in the last week did you feel that you could not get going?” either moderate amount of the time or most of the time. 2.2. Electromyography Muscle activity and quiescence were measured with a portable surface EMG device (Biometrics DataLOG P3X8, Gwent, UK). Details of the EMG data collection and analysis are described elsewhere (Harwood et al., 2008; Jakobi et al., 2008). Briefly, surface electrodes were placed mid-belly of two major arm muscles [biceps brachii (BB), triceps brachii (TB)] and two major thigh muscles [vastus lateralis (VL), and biceps femoris (BF)] on the self-reported dominant side. A common ground electrode was placed on the lateral malleolous of the fibula. The inter-electrode distance was fixed at 20 mm and the cables from the electrodes were taped to the skin and placed into the EMG data logger (9.5 × 15.8 × 3.3 cm; 380 g) which was secured to a belt worn at the waist. The signal from the electrodes were sampled at 1000 Hz, amplified (1000×), band-pass filtered (20–450 Hz), and stored on a 512 MB MMC flashcard. Subsequent to EMG electrode placement and setup of the recording unit isometric maximal voluntary exertions (MVE) were performed for the four muscles (VL, BF, BB, and TB) in order to normalize the 9-hour EMG recordings to a percentage of the participant's maximum. The MVE were recorded in the seated position during isometric knee and elbow extension and flexion against resistance provided manually by the same researcher. The knee and elbow joint were bent to ~ 90° during the MVE of the thigh and arm muscles, respectively. Each muscle was tested in a randomized order three times with 60 s rest between trials; however, failure to maintain proper position warranted additional attempts until correct position was maintained throughout their maximal effort. The greatest of the three trials was used for normalization of the 9-hour EMG data. Verbal encouragement was provided by the same researcher to ensure maximal effort. All EMG data during the MVE and the 9-hours testing were imported into Biometrics software (Biometrics DataLog version 3, Gwent, UK) for preliminary visual inspection and subsequently into Spike 2 Version 5 (Cambridge Electronics Design, Cambridge, UK) for analysis in custom script software. Data artefacts (~ 5% of the total time) which arise from contact with the electrodes or device worn at
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917 Table 1 Subject Characteristics. Non-frail (n = 10)
Pre-frail (n = 11)
Frail (n = 12)
Age (years) 74 ± 4 75 ± 4 81 ± 6⁎ Height (cm) 155 ± 6.0 156 ± 4.8 150 ± 7.5 Weight (kg) 65.7 ± 8.4 77.8 ± 16.2 71.7 ± 13.4 Number of self-reported comorbidities 1.8 ± 1.3 2.4 ± 1.8 3.4 ± 2.0 Number of prescription medication(s) 4.4 ± 3.0 3.6 ± 2.8 6.2 ± 3.1 Number of fall in the past year 0.9 ± 1.7 1.7 ± 3.1 1.9 ± 2.4 Number of steps in 9 h 3147 ± 2031 2094 ± 1087 481 ± 394⁎ Isotonic leg extension Strength (kg) 10.0 ± 2.3 8.7 ± 2.7 6.2 ± 1.5⁎ ⁎ Significantly different from non-frail and pre-frail (p b 0.05).
the hip were manually removed across all four channels in a timelocked fashion. Signals were rectified, smoothed at a time constant of 0.01 s and down-sampled by a factor of 100. Bursts and gaps in the EMG signal were computed to quantify muscle activity and quiescence during the 9-hour testing period. Bursts, which represent muscle activation, were defined as a period of EMG activity greater than 2% of MVE for duration longer than 0.1 s. Burst characteristics examined were: number of bursts, mean duration (seconds), burst percentage (% of total recording time occupied by bursts), peak amplitude (average peak amplitude of all bursts, %MVE), and mean amplitude (average mean amplitude of all bursts, %MVE). Gaps, which represent muscle quiescence, were quantified as a period of EMG less than 1% of MVE for a duration longer than 0.1 s. Gap characteristics examined were: number of gaps, mean duration (seconds), and gap percentage (% of total recording time occupied by gaps). Previous research has used burst and gap analysis to quantify muscle activity
911
and quiescence (Blangsted et al., 2003; Harwood et al., 2008; Howe and Rafferty, 2009; Jakobi et al., 2008; Kern et al., 2001; Laursen et al., 2001; Mork and Westgaard, 2006). 2.3. Mobility Mobility during the 9 h of testing was measured using the ActiTrainer accelerometer (Actigraph, LLC, Fort Walton Beach, FL). The ActiTrainer (8.6 × 3.3 × 1.5 cm; 51 g) is a uniaxial accelerometer that was programmed to record data in 1-min epochs. It was secured in a holster, attached to a belt, which was worn at the waist on the dominant side parallel to the mid-axillary line. ActiTrainer data was downloaded into the ActiLife software (Actigraph, LLC, Fort Walton Beach, FL) and step-counts per minute were used to calculate the number of steps completed by the participants during the 9-hour testing. 2.4. Muscle strength Maximal isotonic knee extension strength of the dominant leg was measured using the adjustable Recordman® foot weights and one repetition maximum with the participant seated in a chair with the knee bent to ~90°. Initially, participants performed three submaximal knee extensions with a light load foot weight (~2–3 kg) to warm up. Participants performed a series of single repetition lifts with increasing weight loads until a 1RM (one repetition maximum) lift was achieved in approximately 3–5 attempts. 1RM was defined as the maximal weight that the participants could lift through the full range of motion (Latham et al., 2003).
Fig. 1. Burst activity for non-frail, pre-frail, and frail women. (A) Burst percentage; (B) Number of bursts; (C) Mean burst duration; and (D) Mean burst amplitude. %, percentage; s, seconds; MVE, maximal voluntary exertion.⁎Significantly different from non-frail women; #Significantly different from frail women (p b 0.05).
912
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
2.5. Statistical analysis
3.2. Muscles
The Statistical Package for the Social Sciences (SPSS, Chicago, IL) for windows version 16.0 was used for statistical analysis. A one-way analysis of variance (ANOVA) was used to examine participants' characteristics (age, height, weight, number of comorbidities and medications, history of falls, muscle strength, and mobility) between the three frailty groups. A 3 × 4 multivariate analysis of variance (MANOVA) was performed with the between subjects factor frailty (non-frail, pre-frail, and frail) and the within subjects factor muscle (VL, BF, BB, and TB). The dependent variables were the five burst (number of bursts, mean duration, burst percentage, peak amplitude, and mean amplitude) and three gap characteristics (number of gaps, mean duration, and gap percentage) which indicate muscle activity and quiescence, respectively. Pearson's product-moment correlations were computed between number of steps and burst characteristics of the lower limb muscles, as well as number of steps and gap characteristics of the lower limb muscles. A significance level of p b 0.05 was accepted. Data in the text and table are reported as values ± standard deviation of the mean, whereas figures are presented as values ± standard error of the mean.
A significant (p b 0.001) multivariate main effect of muscles on burst activity across all frailty groups was found. Univariate tests demonstrated a significant (p b 0.05) main effect of muscles on all burst characteristics. Pair-wise comparisons revealed that the burst percentage was greater (p≤ 0.01) in both of the arm muscles (BB and TB) compared with the two thigh muscles (VL and BF). In addition, burst percentage was greater (p b 0.001) in the TB than the BB; however, there was no difference (p = 0.19) between the VL and BF (Fig. 3A). The number of bursts was greater (p b 0.001) in the arm muscles (BB and TB) than the thigh muscles (VL and BF); however, there was no difference between the VL and BF and between the BB and TB (Fig. 3B). Mean burst duration was greater (p b 0.05) in the TB compared with the
3. Results Ten women were categorized as non-frail, 11 as pre-frail, and 12 as frail (Table 1). Height, weight, number of self-reported comorbidities, number of medications, and number of falls within the past year were similar among the three groups (p N 0.05). Frail women were older, had weaker leg extension strength, and completed fewer steps during the 9-hour testing compared with the pre-frail and non-frail women (p b 0.05). In contrast, no differences were found in age, leg extension muscle strength, and mobility between non-frail and pre-frail women (p N 0.05) (Table 1). The two way interaction of frailty by muscle was non-significant for both burst (p = 0.06) and gap activity (p = 0.96).
3.1. Frailty A significant (p = 0.001) multivariate main effect of frailty on burst activity across all muscles was found. Univariate tests demonstrated a significant (p b 0.05) main effect of frailty on three of the burst characteristics (number of bursts, mean burst duration, and mean burst amplitude). Burst percentage (Fig. 1A) and peak amplitude (non-frail 8.8 ± 3.0, pre-frail 8.8 ± 2.1, frail 7.7 ± 2.4% MVE) were similar (p N 0.05) across all frailty groups. Post-hoc testing revealed that the number of bursts was less (p = 0.01) in the frail than the nonfrail women, and pre-frail women did not differ from non-frail or frail (Fig. 1B). Mean duration of bursts was greater (p b 0.05) in frail and pre-frail than in the non-frail women; however, there was no difference (p = 0.73) between the pre-frail and frail women (Fig. 1C). Mean burst amplitude was greater (p b 0.05) in the prefrail women than the non-frail and frail women. No difference (p = 0.54) was found in the mean burst amplitude between the frail and non-frail women (Fig. 1D). A significant (p = 0.01) multivariate main effect of frailty on gap activity across all muscles was found. Univariate tests demonstrated a significant (p b 0.05) main effect of frailty on number and mean duration of gaps. The gap percentage was similar (p = 0.91) across all frailty groups (Fig. 2A). Post-hoc testing revealed that the number of gaps was greater (p b 0.01) in the frail than the non-frail and pre-frail women; however, there was no difference between the non-frail and pre-frail women (Fig. 2B). Mean gap duration was less (p = 0.01) in the frail than the pre-frail women. In contrast, no difference (p N 0.05) was found in the mean gap duration of the non-frail women compared with the pre-frail and frail women (Fig. 2C).
Fig. 2. Gap activity for non-frail, pre-frail, and frail women. (A) Gap percentage; (B) Number of gaps; and (C) Mean gap duration. %, percentage; s, seconds. ⁎Significantly different from non-frail women; $Significantly different from pre-frail women (p b 0.05).
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
913
Fig. 3. Burst activity for the vastus lateralis, biceps femoris, biceps brachii, and triceps brachii. (A) Burst percentage; (B) Number of bursts; (C) Mean burst duration; (D) Peak burst amplitude; and (E) Mean burst amplitude. %, percentage; s, seconds; MVE, maximal voluntary exertion; VL, vastus lateralis; BF, biceps femoris; BB, biceps brachii; TB, triceps brachii. ⁎Significantly different from VL; #Significantly different from BF; $Significantly different from BB; €Significantly different from TB (p b 0.05).
VL and BB (Fig. 3C). Peak burst amplitude was less (p b 0.01) in the BB compared with the other three muscles, and the BF was greater (p b 0.05) than the VL and TB (Fig. 3D). Mean burst amplitude was greater (p b 0.05) in the TB compared with the other 3 muscles (Fig. 3E). A significant (p b 0.001) multivariate main effect of muscles on gap activity across all frailty groups was found. Univariate tests demonstrated a significant (p b 0.001) main effect of muscles on all gap characteristics. Pair-wise comparisons revealed that the gap percentage was less (p b 0.01) in the TB compared with the other three muscles and in the BB compared with the BF (Fig. 4A). The number of gaps was less (p b 0.01) in the BB compared with the other three muscles and in the TB compared with the BF (Fig. 4B). Mean gap duration was greater (p b 0.05) in the BB compared with the other three muscles and in the VL compared with the TB (Fig. 4C).
3.3. Relation between mobility and burst and gap characteristics The number of steps was significantly related to all burst characteristics of the two thigh muscles (Fig. 5). Pearson's correlations between the number of steps and the gap characteristics of the two thigh muscles were not significant (r = 0.01–0.22; p N 0.05). 4. Discussion Daily muscle activity and quiescence of the two upper limb (BB, TB) and two lower limb (VL, BF) muscles, quantified by burst and gap activity, was compared between non-frail, pre-frail, and frail older women. To our knowledge this is the only study that has examined the association of EMG with frailty during daily life. The percentage of
914
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
44% of the total recording time, which is similar to the total duration of muscle activity that we found for the arm muscles (30–39%). Kern et al. (2001) found that VL and BB of women were active 12% and 23% of the recording time; however, these participants were young women which likely accounts for the shorter total duration of muscle activity compared with our study of older women. Howe and Rafferty (2009) reported that the VL of women (mean age 64 years) was active 10% of the recording time which is less than the duration observed in these groups of non-frail, pre-frail, and frail women. These observed differences from previously published studies indicate quantifying bursts and gaps enables dissociation of muscle activity between healthy young and old; the current study further suggests that frail adults can be distinguished from non-frail. 4.1. Effect of frailty on muscle activity and quiescence
Fig. 4. Gap activity for the vastus lateralis, biceps femoris, biceps brachii, and triceps brachii. (A) Gap percentage; (B) Number of gaps; and (C) Mean gap duration.%, percentage; s, seconds; VL, vastus lateralis; BF, biceps femoris; BB, biceps brachii; TB, triceps brachii. ⁎Significantly different from VL; #Significantly different from BF; $ Significantly different from BB; €Significantly different from TB (p b 0.05).
time the muscles were active and quiescent were similar across all frailty groups (Figs. 1A; and 2A); however, the characteristics of muscle activity (bursts and gaps) were different. In frail women there were fewer muscle bursts but these bursts were for a longer duration compared with non-frail women. Moreover, the number of gaps was greater in frail women, but the duration was much shorter (~28%; Fig. 2C). Thus, the pattern of muscle activity is dissimilar across the frailty spectrum and likely offers a measure to dissociate stages of frailty. Mork and Westgaard (2005) measured long-term EMG activity in the trapezius and low back muscles of sedentary young and middle aged women and found that the duration of muscle activity was 29–
The number of bursts observed in frail women was 28% fewer compared with the non-frail women, whereas there were 29% and 25% more gaps in frail women than non-frail and pre-frail women, respectively. This suggests that muscle activity was less in frail women relative to non-frail and pre-frail women. This study showed that muscle activity differs across stages of frailty; however, it is not possible to determine whether frailty or changes in muscle activity comes first. Future longitudinal studies may address this question. Frail women were older and 77–85% less mobile (accelerometry) than non-frail and pre-frail women. Mobility was highly correlated with the number of bursts across all frailty groups. Chronological age and mobility likely account for the limited muscle activity observed in these frail older women compared to non-frail. The number of steps completed in this study was much lower than those found in other studies even for the non-frail older adults (Harris et al., 2009; Webber and Porter, 2009). However, we recorded steps only for 9 h compared with other studies that examine steps over a 24-hour period. In addition, physical activity participation measured by questionnaire in Greek older adults, is low compared with other European countries (Tzorbatzakis and Sleap, 2007). No studies have examined physical activity in Greek older adults using accelerometers but our data on the non-frail older women suggest that physical activity is likely quite low. Actigraph accelerometer steps counts have been shown to be accurate for walking speeds above 0.9 m/s but not for lower speeds (Webber and Porter, 2009), thus the steps recorded in our study may have been underestimated for frail older women. Previous work in healthy non-frail older adults has indicated that a reduction in the number of times that muscles were active is associated with an increase in the duration of each muscle burst (Harwood et al., 2008; Jakobi et al., 2008). These studies suggest that muscles of older adults burst fewer times but for longer duration relative to younger adults. This change is progressive as the oldest and most frail in this study had fewer bursts of longer duration relative to the non-frail. Longer duration bursts might be attributed to rate of movement. Fast velocity movements are more affected by aging than slow velocity movements (Candow and Chillibeck, 2005; Petrella et al., 2005). Also, burst amplitude was 36% less in frail women compared with the pre-frail women. These differences might be related to frail women requiring greater force generation for daily activities. Bursts and gaps discriminated later stage frailty, but only bursts discriminated early stage frailty (differences in pre-frail women compared with the non-frail women). Each burst in pre-frail women was approximately 30% longer with a 40% greater amplitude than the muscles of non-frail women. These two groups were of similar age, anthropometric and health characteristics, strength, and mobility. Factors such as impaired motor control and increased subjective fatigue, which are both outcomes of frailty, likely contribute to the differences in muscle activity (Dedering et al., 1999; Ferucci et al.,
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
915
Fig. 5. The relationship between number of steps and burst activity of the two thigh muscles. (A) Bursts percentage; (B) Number of bursts; (C) Mean bursts duration; (D) Peak bursts amplitude; and (E) Mean burst amplitude. r, Pearson's correlation coefficient; %, percentage; s, seconds; MVE, maximal voluntary exertion.
2004; Laursen et al., 2001). Other physiological factors (e.g. muscle fiber-type proportion, motor unit firing rate, nerve conduction velocity, and muscle fatigue) may be related to differences between non-frail and pre-frail, but studies have yet to examine the effect of frailty on these characteristics relative to bursts and gaps. Future studies should examine underlying mechanisms that contribute to changes in burst and gap activity across the stages of frailty. 4.2. Differences between muscles on activity and quiescence This study demonstrated that across all frailty groups lower limb muscles were active 22% (~ 2 h) and quiescent 50% (~4.5 h) of the
time. However, the percentages of bursts were greater in the upper limb muscles compared to the lower limb muscles. These results are consistent with a previous study in younger adults which reported that arm muscles are more active relative to thigh muscles (Kern et al., 2001). The greater activity of the upper limb muscles compared with the lower limb muscles (e.g. ~ 60% greater total duration of muscle activity) is likely because they are needed extensively by older adults to execute activities of daily living (ADL) (Hortobagyi et al., 2003). Older women may spend much of their day seated or standing performing a variety of household activities that utilize the upper limbs. These EMG recordings of daily life suggest that upper limb movement is maintained relative to lower body ambulation. A
916
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
previous study (Harwood et al., 2008) has shown that when the same task was performed there were age- and sex-related differences in muscle activity. Future studies should examine how older women activate arm and thigh muscles to perform ADL (e.g. video analysis of ADL). Muscle activity (except peak amplitude) and quiescence were similar in the two thigh muscles (VL and BF) across all three frailty groups. In contrast, differences in muscle activity and quiescence were found between the two arm muscles (BB and TB). Triceps brachii was active 39% of the time and quiescent 27% of the time; whereas BB was active 30% of the time and quiescent 41% of the time. Previous work in young adults indicated similarities in muscle activity between the lower limb muscles but not the upper limb muscles (Kern et al., 2001). In addition, mean duration, mean amplitude, and peak amplitude of muscle activity were greater in the TB, whereas mean duration of muscle quiescence was greater in the BB. In contrast, the number of times that the muscles were active was similar between the two arm muscles and the number of times that the muscles were quiescent was greater in the TB. The observed differences between the two arm muscles in our study may be related to the greater strength of the BB compared with the TB and the greater usage of the TB for daily activities. A longitudinal study of older men described the rate of decline in muscle strength per year to be 1.6% and 1.8% for the TB and BB, respectively (Frontera et al., 2000). In addition, TB muscle mass (Candow and Chillibeck, 2005) and its ability to activate (Jakobi and Rice, 2002) is better maintained with age than the BB and it may be relied on more to complete daily activities. This would result in greater muscle activity and less muscle quiescence, as observed. Alternatively, little is understood about fiber types between these muscles and thus differences in young adults as well as unique age-related changes within each muscle might contribute to the observed patterns of muscle activation between these arm muscles (Dalton et al., 2010). Muscle activity was measured over one day, and for a time period limited to approximately 10 h because many of the participants were frail and the battery capacity of the recording unit. Future research should examine the day to day reliability in measuring daily muscle activity using portable EMG in older adults across levels of frailty, and the minimal recording period to represent daily muscle activity patterns. In addition, the older women commented during the posttesting discussion that the portable EMG device was minimally obtrusive; however, future studies should examine whether participants alter personal daily habits when portable EMG devices are worn. 5. Conclusions Frailty alters the individual characteristics of muscle activity (bursts) and quiescence (gaps) which may assist in dissociation between stages of frailty. Muscle activity discriminated both early and later stage frailty; however, muscle quiescence discriminated later stage frailty. Mobility was associated with the differences observed in muscle activity across the stages of frailty. Although mobility and strength, which are two known frailty indicators, were statistically similar between the non-frail and pre-frail women these groups differed in muscle activity. Moreover, measures of bursts and gaps in EMG provided identification of muscle patterning between the upper and lower limbs, and these measures might be useful in quantifying agonist and antagonist coordination, as well as functional use. References Bandeen-Roche, K., Xue, Q.-L., Ferruci, L., Walston, J., Guralnik, J.M., Chaves, P., Zeger, S.L., Fried, L.P., 2006. Phenotype of frailty: characterization in the women's health and aging studies. J. Gerontol. A Biol. Sci. Med. Sci. 61 (3), 262–266.
Bauer, J.M., Sieber, C.C., 2008. Sarcopenia and frailty: a clinician's controversial point of view. Exp. Gerontol. 43, 674–678. Blangsted, A.K., Hansen, K., Jensen, C., 2003. Muscle activity during computer-based office work in relation to self-reported job demands and gender. Eur. J. Appl. Physiol. 89, 352–358. Blangsted, A.K., Sogaard, K., Christensen, H., Sjogaard, G., 2004. The effect of physical and psychological loads on the trapezius muscle activity during computer keying tasks and rest periods. Eur. J. Appl. Physiol. 91, 253–258. Candow, D.G., Chillibeck, P.D., 2005. Differences in size, strength, and power of upper and lower body muscle groups in young and older men. J. Gerontol. A Biol. Sci. Med. Sci. 60, 148–156. Central Intelligence Agency, 2009. The world factbook 2009. ISSN 1553-8133, Washington, DC, viewed February 10, 2010 from https://www.cia.gov/library/ publications/the-world-factbook/index.html. Cesari, M., Leeuwenburgh, C., Lauretani, F., Onder, G., Bandinelli, S., Maraldi, C., Guralnik, J.M., Pahor, M., Ferrucci, L., 2006. Frailty syndrome and skeletal muscle: results from the Invecchiare in Chianti study. Am. J. Clin. Nutr. 83, 1142–1148. Chandler, J.M., Duncan, P.W., Kochersberger, G., Studenski, S., 1998. Is lower extremity strength gain associated with improvement in physical performance and disability in frail, community-dwelling elders? Arch. Phys. Med. Rehabil. 79, 24–30. Dalton, B.H., Jakobi, J.M., Allman, B.L., Rice, C.L., 2010. Differential age-related changes in motor unit properties between elbow flexors and extensors. Acta Physiol. 200, 45–55. Dedering, A., Nemeth, G., Harms-Ringdahl, K., 1999. Correlation between electromyographic spectral changes and subjective assessment of lumbar muscle fatigue in subjects without pain from the lower back. Clin. Biomech. 14, 103–111. Ferucci, L., Guralnik, J.M., Studenski, S., Fried, L.P., Cutler, G.B., Walston, J.D., 2004. Designing randomized, controlled trials aimed at preventing or delaying functional decline and disability in frail, older persons: a consensus report. J. Am. Geriatr. Soc. 52, 625–634. Fried, L.P., Tangen, C.M., Walston, J., Newman, A.B., Hirsch, C., Gottdiener, J., Seeman, T., Tracy, R., Kop, W.J., Burke, G., McBurnie, M.A., for the Cardiovascular Health Study Collaborative Research Group, 2001. Frailty in older adults: evidence for a phenotype. J. Gerontol. A Biol. Sci. Med. Sci. 56 (3), M146–M156. Fried, L.P., Ferrucci, L., Darer, J., Williamson, J.D., Anderson, G., 2004. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J. Gerontol. A Biol. Sci. Med. Sci. 59 (3), 255–263. Frontera, W.R., Hughes, V.A., Fielding, R.A., Fiatarone, M.A., Evans, W.J., Roubenoff, R., 2000. Aging of skeletal muscle: a 12-yr longitudinal study. J. Appl. Physiol. 88, 1321–1326. Harris, T.J., Owen, C.G., Victor, C.R., Adams, R., Cook, D.G., 2009. What factors are associated with physical activity in older people, assessed objectively by accelerometry? Br. J. Sports Med. 43, 442–450. Harwood, B., Edwards, D.L., Jakobi, J.M., 2008. Age- and sex-related differences in muscle activation for a discrete functional task. Eur. J. Appl. Physiol. 103, 677–686. Holtermann, A., Sogaard, K., Christensen, H., Dahl, B., Blangsted, A.K., 2008. The influence of biofeedback training on trapezius activity and rest during occupational computer work: a randomized controlled trial. Eur. J. Appl. Physiol. 104, 983–989. Hortobagyi, T., Mizelle, C., Beam, S., DeVita, P., 2003. Old adults perform activities of daily living near their maximal capabilities. J. Gerontol. A Biol. Sci. Med. Sci. 58 (5), M453–M460. Howe, T.E., Rafferty, D., 2009. Quadriceps activity and physical activity profiles over long durations in patients with osteoarthritis of the knee and controls. J. Electromyogr. Kinesiol. 19 (2), e78–e83. Jakobi, J.M., Rice, C.L., 2002. Voluntary muscle activation varies with age and muscle group. J. Appl. Physiol. 93, 457–462. Jakobi, J.M., Edwards, D.L., Connelly, D.M., 2008. Utility of portable electromyography for quantifying muscle activity during daily use. Gerontology 54, 324–331. Kern, D.S., Semmler, J.G., Enoka, R.M., 2001. Long-term activity in upper- and lowerlimb muscles of humans. J. Appl. Physiol. 91, 2224–2232. Latham, N.K., Anderson, C.S., Lee, A., Bennett, D.A., Moseley, A., Cameron, I.D., Fitness Collaborative Group, 2003. A randomized, controlled trial of quadriceps resistance exercise and vitamin D in frail older people: the Frailty Interventions Trial in Elderly Subjects (FITNESS). J. Am. Geriatr. Soc. 51 (3), 291–299. Laursen, B., Jensen, B.R., Ratkevicius, A., 2001. Performance and muscle activity during computer mouse tasks in young and elderly adults. Eur. J. Appl. Physiol. 84, 329–336. Madeleine, P., Vedsted, P., Blangsted, A.K., Sjogaard, G., Sogaard, K., 2006. Effects of electromyographic and mechanomyographic biofeedback on upper trapezius muscle activity during standardized computer work. Ergonomics 49 (10), 921–933. Mork, P.J., Westgaard, R.H., 2005. Long-term electromyographic activity in upper trapezius and low back muscles of women with moderate physical activity. J. Appl. Physiol. 99, 570–578. Mork, P.J., Westgaard, R.H., 2006. Low-amplitude trapezius activity in work and leisure and the relation to shoulder and neck pain. J. Appl. Physiol. 100, 1142–1149. Paterson, D.H., Jones, G.R., Rice, C.L., 2007. Ageing and physical activity: evidence to develop exercise recommendations for older adults. Can. J. Public Health 98, S69–S108. Petrella, J.K., Kim, J., Tuggle, S.C., Hall, S.R., Bamman, M.M., 2005. Age differences in knee extension power, contractile velocity, and fatigability. J. Appl. Physiol. 98, 211–220. Rantanen, T., Harris, T., Leveille, S.G., Visser, M., Foley, D., Masaki, K., Guralnik, J.M., 2000. Muscle strength and body mass index as long-term predictors of mortality in initially health men. J. Gerontol. A Biol. Sci. Med. Sci. 55A (3), M168–M173. Rockwood, K., Mitnitski, A., 2007. Frailty in relation to the accumulation of deficits. J. Gerontol. A Biol. Sci. Med. Sci. 62A (7), 722–727.
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917 Santos-Eggimann, B., Cuenoud, P., Spagnoli, J., Junod, J., 2009. Prevalence of frailty in middle-aged and older community-dwelling Europeans living in 10 countries. J. Gerontol. A Biol. Sci. Med. Sci. 64A (6), 675–681. Tzorbatzakis, N., Sleap, M., 2007. Participation in physical activity and exercise in Greece: a systematic literature review. Int. J. Public Health 52, 360–371.
917
Vandervoort, A.A., 2002. Aging of the human neuromuscular system. Muscle Nerve 25, 17–25. Webber, S.C., Porter, M.M., 2009. Monitoring mobility in older adults using global positioning system (GPS) watches and accelerometers: a feasibility study. J. Aging Phys. Act. 17, 455–467.
Contents lists available at ScienceDirect
Experimental Gerontology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / ex p g e r o
Daily muscle activity and quiescence in non-frail, pre-frail, and frail older women Olga Theou a, Gareth R. Jones b, Anthony A. Vandervoort a, Jennifer M. Jakobi b,⁎ a b
Graduate Program in Health and Rehabilitation Sciences, University of Western Ontario, London, Ontario, Canada Human Kinetics — Faculty of Health and Social Development, University of British Columbia Okanagan, Kelowna, British Columbia, Canada
a r t i c l e
i n f o
Article history: Received 13 February 2010 Received in revised form 16 August 2010 Accepted 16 August 2010 Available online 22 August 2010 Section Editor: Dr. Christiaan Leeuwenburgh Keywords: Frailty Pre-frailty Aging Older adults Muscle activity Muscle quiescence EMG Long-term EMG Burst analysis Gap analysis
a b s t r a c t Reduced muscle mass and strength are likely fundamental components of frailty. The purpose of this study was to measure daily muscle activity and quiescence in non-frail, pre-frail, and frail older women using portable electromyography (EMG). Thirty-three community-dwelling older Greek women were categorized as non-frail (n = 10, 74 ± 4 years), pre-frail (n = 11, 75 ± 4 years), and frail (n = 12, 81 ± 6 years) based upon the frailty phenotype. Surface EMG over a 9-hour typical day was recorded from the biceps brachii, triceps brachii, vastus lateralis, and biceps femoris of the dominant side. Burst and gap analysis was used to quantify muscle activity and quiescence. The total duration of the muscles that were active (~ 2.5 h) and quiescent (~ 4 h) did not differ across frailty groups. However, the number of bursts was 28% fewer and the mean burst duration was 26% longer in frail women compared with the non-frail women. In addition, muscle activity was greater in the arm muscles than the thigh muscles across all groups (e.g. ~ 60% greater burst percentage). Burst number and duration indicate that muscle activity differs across stages of frailty in older women and is greater in the upper compared with lower limbs. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Research into frailty has recently become a key focus of gerontology research. Knowledge surrounding this geriatric syndrome within the past decade has increased exponentially (Bauer and Sieber, 2008). The criteria used for identification of frailty continue to be a matter of debate; however, the most commonly used approaches to qualify this syndrome are the Frailty Phenotype (Fried et al., 2001) and the Frailty Index (Rockwood and Mitnitski, 2007). The development of frailty progresses across a spectrum of defined stages described as non-frail, pre-frail, and frail (Bandeen-Roche et al., 2006; Fried et al., 2001).While frailty often culminates in the need for institutional care (Fried et al., 2004), many frail older adults still remain in the community despite impairments in one or more activities of daily living (ADL) (Chandler et al., 1998). This syndrome is more common in women than men and transitioning toward frailty likely occurs earlier in women because of greater age-related decline of muscle mass and strength (Cesari et al., 2006; Fried et al., 2001). Greece has one of the oldest populations in Europe [19.2% of population over 65 years of age (Central Intelligence Agency, 2009)] ⁎ Corresponding author. Faculty of Health and Social Development, University of British Columbia — Okanagan, 3333 University Way, Kelowna, BC Canada. V1V 1V7. Tel.: + 1 250 807 9884; fax: + 1 250 807 8085. E-mail address: [email protected] (J.M. Jakobi). 0531-5565/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.exger.2010.08.008
where the prevalence of frailty and pre-frailty in community-dwelling older adults is reported to be 15% and 45%, respectively (SantosEggimann et al., 2009). In North America, in those over the age of 65 years 17% are considered frail and over 46% may already exhibit pre-frailty characteristics (Fried et al., 2001). Fried et al. (2001) utilized low isometric handgrip strength as an indicator of frailty. While handgrip has been proposed as a good predictor of health related events (Rantanen et al., 2000), it only measures upper limb strength and may not entirely capture lower limb strength (Cesari et al., 2006). It is well established that agerelated decline in strength is not similar across all muscles (Candow and Chillibeck, 2005; Dalton et al., 2010; Frontera et al., 2000; Vandervoort, 2002), therefore when examining frailty various muscles need to be considered. To our knowledge, no investigation has yet examined if frailty is associated with changes in muscle activity based upon anatomical location (upper or lower limb) or functional movement (flexion, extension). In addition, physical activity levels should also be considered. Physical activity interacts with the natural process of aging and is known to alter the rate of age-related progressive decline in muscle function (Paterson et al., 2007). Muscle activity and quiescence, termed low-threshold electromyography (EMG), was successfully used to compare muscle function in healthy community-living older adults relative to young adults (Harwood et al., 2008), as well as with working adults to explore
910
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
job-related injuries (Blangsted et al., 2003, 2004; Holtermann et al., 2008; Laursen et al., 2001; Madeleine et al., 2006). No studies have yet examined EMG during daily life in frail persons. Daily activities are often reported to be “hard work” for most frail adults, and often this ‘work’ results in falls and injuries. Measuring muscle activity during daily life might contribute to understanding how older adults use their upper limbs relative to their lower limbs. Kern et al. (2001) reported in young adults that the upper limb muscles were more active than the lower limb muscles during daily activities. In addition, daily upper and lower limb muscle activity and quiescence is a result of an interaction of several systems (e.g. muscular and nervous system) and may be a more complete indicator of frailty than handgrip strength. Laboratory and functional performance measures are important, but limited for this population. Most laboratory tests developed for older adults are not applicable to frail adults and often frail older adults are unable to attend a laboratory for testing due to their impaired health. Measurements that will capture the daily life of the frail person in the home environment are needed. Recent studies measuring muscle activity during daily life and discrete tasks reported muscle activity was greater in non-frail older adults relative to young adults and these age-related differences were greater in women compared with men (Harwood et al., 2008; Jakobi et al., 2008). Muscle activity and quiescence recorded during daily life in older women across stages of frailty is unknown, but could contribute to our understanding of the progression of this syndrome. The purpose of this study was to determine whether muscle activity and quiescence recorded over a 9-hour typical day in upper and lower limb muscles differs between non-frail, pre-frail and frail older women. Women were investigated since the development of frailty is sex-specific (Fried et al., 2001). It was hypothesized that community-dwelling, older women would have greater muscle activity in upper limbs compared to lower limbs. As a result of reduced muscle strength, a higher proportion of the muscle is needed to execute a movement therefore muscle activity would increase across stages of frailty.
2. Methods Thirty-three community-dwelling women aged 68–90 years were recruited within the rural prefecture of Thessaloniki, Greece. Initially, participants were recruited through word-of-mouth, these individuals then recommended additional participants from their acquaintances. Individuals, who wore the equipment, also generated additional participant's interest in the study. The study was approved by the University of Western Ontario ethics board, in accordance with the declaration of Helsinki, and informed consent was received prior to participation. Inclusion criteria were women older than 65 years of age who were living in the local community. During weekdays the researcher visited the home of the participants on two separate and subsequent days. On the first day a health history questionnaire was administered, the stage of frailty phenotype determined (Fried et al., 2001), and muscle strength measured. On day two, the same researcher arrived approximately 1 h after the participant awoke. An EMG device and accelerometer were attached to the participant and Maximal Voluntary Exertions (MVE) for each muscle of interest were performed. Participants were then instructed to proceed with their normal daily activities while wearing the portable EMG and accelerometer. Participants were also asked not to bath or exercise vigorously in order to prevent dislodging the electrodes and damage to the recording device. The researcher encouraged participants to disregard the equipment and go about their normal typical day. Approximately 9–10 h later, on day two, the researcher returned to the participant's home to remove the equipment.
2.1. Frailty definition Physical frailty was defined in accordance with the Frailty Phenotype (Fried et al. 2001), as a clinical syndrome in which the participant expressed three or more of the five criteria below. Participants with one or two criteria were classified as pre-frail, whereas those exhibiting no criteria were considered non-frail. 1) Weight loss: A positive response to the question “In the last year, have you lost more than 5 kg unintentionally (i.e., not due to dieting or exercise)”. 2) Muscle strength: The highest of three consecutive maximal handgrip strength measures of the dominant hand using a Jamar® hand-held dynamometer. Cutoff scores were applied based upon body mass index (BMI ≤ 23, cutoff strength ≤ 17 kg; BMI 23.1–26, cutoff strength ≤ 17.3 kg; BMI 26.1–29 cutoff strength ≤ 18 kg; BMI N29 cutoff strength ≤ 21 kg). 3) Walking speed: Time to walk 15 ft at usual pace. The cutoff scores were applied based upon height (Height ≤ 159 cm, cutoff time ≥ 7 s; Height N 159 cm, cutoff time ≥ 6 s). 4) Physical activity: A weighted score of kilocalories expended per week was calculated based on participant's response to the Short version of the Minnesota Leisure Time Activity Questionnaire (cutoff b270 Kcals per week). 5) Subjective fatigue: Responding to the questions “How often in the last week did you feel that everything you did was an effort?” or “How often in the last week did you feel that you could not get going?” either moderate amount of the time or most of the time. 2.2. Electromyography Muscle activity and quiescence were measured with a portable surface EMG device (Biometrics DataLOG P3X8, Gwent, UK). Details of the EMG data collection and analysis are described elsewhere (Harwood et al., 2008; Jakobi et al., 2008). Briefly, surface electrodes were placed mid-belly of two major arm muscles [biceps brachii (BB), triceps brachii (TB)] and two major thigh muscles [vastus lateralis (VL), and biceps femoris (BF)] on the self-reported dominant side. A common ground electrode was placed on the lateral malleolous of the fibula. The inter-electrode distance was fixed at 20 mm and the cables from the electrodes were taped to the skin and placed into the EMG data logger (9.5 × 15.8 × 3.3 cm; 380 g) which was secured to a belt worn at the waist. The signal from the electrodes were sampled at 1000 Hz, amplified (1000×), band-pass filtered (20–450 Hz), and stored on a 512 MB MMC flashcard. Subsequent to EMG electrode placement and setup of the recording unit isometric maximal voluntary exertions (MVE) were performed for the four muscles (VL, BF, BB, and TB) in order to normalize the 9-hour EMG recordings to a percentage of the participant's maximum. The MVE were recorded in the seated position during isometric knee and elbow extension and flexion against resistance provided manually by the same researcher. The knee and elbow joint were bent to ~ 90° during the MVE of the thigh and arm muscles, respectively. Each muscle was tested in a randomized order three times with 60 s rest between trials; however, failure to maintain proper position warranted additional attempts until correct position was maintained throughout their maximal effort. The greatest of the three trials was used for normalization of the 9-hour EMG data. Verbal encouragement was provided by the same researcher to ensure maximal effort. All EMG data during the MVE and the 9-hours testing were imported into Biometrics software (Biometrics DataLog version 3, Gwent, UK) for preliminary visual inspection and subsequently into Spike 2 Version 5 (Cambridge Electronics Design, Cambridge, UK) for analysis in custom script software. Data artefacts (~ 5% of the total time) which arise from contact with the electrodes or device worn at
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917 Table 1 Subject Characteristics. Non-frail (n = 10)
Pre-frail (n = 11)
Frail (n = 12)
Age (years) 74 ± 4 75 ± 4 81 ± 6⁎ Height (cm) 155 ± 6.0 156 ± 4.8 150 ± 7.5 Weight (kg) 65.7 ± 8.4 77.8 ± 16.2 71.7 ± 13.4 Number of self-reported comorbidities 1.8 ± 1.3 2.4 ± 1.8 3.4 ± 2.0 Number of prescription medication(s) 4.4 ± 3.0 3.6 ± 2.8 6.2 ± 3.1 Number of fall in the past year 0.9 ± 1.7 1.7 ± 3.1 1.9 ± 2.4 Number of steps in 9 h 3147 ± 2031 2094 ± 1087 481 ± 394⁎ Isotonic leg extension Strength (kg) 10.0 ± 2.3 8.7 ± 2.7 6.2 ± 1.5⁎ ⁎ Significantly different from non-frail and pre-frail (p b 0.05).
the hip were manually removed across all four channels in a timelocked fashion. Signals were rectified, smoothed at a time constant of 0.01 s and down-sampled by a factor of 100. Bursts and gaps in the EMG signal were computed to quantify muscle activity and quiescence during the 9-hour testing period. Bursts, which represent muscle activation, were defined as a period of EMG activity greater than 2% of MVE for duration longer than 0.1 s. Burst characteristics examined were: number of bursts, mean duration (seconds), burst percentage (% of total recording time occupied by bursts), peak amplitude (average peak amplitude of all bursts, %MVE), and mean amplitude (average mean amplitude of all bursts, %MVE). Gaps, which represent muscle quiescence, were quantified as a period of EMG less than 1% of MVE for a duration longer than 0.1 s. Gap characteristics examined were: number of gaps, mean duration (seconds), and gap percentage (% of total recording time occupied by gaps). Previous research has used burst and gap analysis to quantify muscle activity
911
and quiescence (Blangsted et al., 2003; Harwood et al., 2008; Howe and Rafferty, 2009; Jakobi et al., 2008; Kern et al., 2001; Laursen et al., 2001; Mork and Westgaard, 2006). 2.3. Mobility Mobility during the 9 h of testing was measured using the ActiTrainer accelerometer (Actigraph, LLC, Fort Walton Beach, FL). The ActiTrainer (8.6 × 3.3 × 1.5 cm; 51 g) is a uniaxial accelerometer that was programmed to record data in 1-min epochs. It was secured in a holster, attached to a belt, which was worn at the waist on the dominant side parallel to the mid-axillary line. ActiTrainer data was downloaded into the ActiLife software (Actigraph, LLC, Fort Walton Beach, FL) and step-counts per minute were used to calculate the number of steps completed by the participants during the 9-hour testing. 2.4. Muscle strength Maximal isotonic knee extension strength of the dominant leg was measured using the adjustable Recordman® foot weights and one repetition maximum with the participant seated in a chair with the knee bent to ~90°. Initially, participants performed three submaximal knee extensions with a light load foot weight (~2–3 kg) to warm up. Participants performed a series of single repetition lifts with increasing weight loads until a 1RM (one repetition maximum) lift was achieved in approximately 3–5 attempts. 1RM was defined as the maximal weight that the participants could lift through the full range of motion (Latham et al., 2003).
Fig. 1. Burst activity for non-frail, pre-frail, and frail women. (A) Burst percentage; (B) Number of bursts; (C) Mean burst duration; and (D) Mean burst amplitude. %, percentage; s, seconds; MVE, maximal voluntary exertion.⁎Significantly different from non-frail women; #Significantly different from frail women (p b 0.05).
912
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
2.5. Statistical analysis
3.2. Muscles
The Statistical Package for the Social Sciences (SPSS, Chicago, IL) for windows version 16.0 was used for statistical analysis. A one-way analysis of variance (ANOVA) was used to examine participants' characteristics (age, height, weight, number of comorbidities and medications, history of falls, muscle strength, and mobility) between the three frailty groups. A 3 × 4 multivariate analysis of variance (MANOVA) was performed with the between subjects factor frailty (non-frail, pre-frail, and frail) and the within subjects factor muscle (VL, BF, BB, and TB). The dependent variables were the five burst (number of bursts, mean duration, burst percentage, peak amplitude, and mean amplitude) and three gap characteristics (number of gaps, mean duration, and gap percentage) which indicate muscle activity and quiescence, respectively. Pearson's product-moment correlations were computed between number of steps and burst characteristics of the lower limb muscles, as well as number of steps and gap characteristics of the lower limb muscles. A significance level of p b 0.05 was accepted. Data in the text and table are reported as values ± standard deviation of the mean, whereas figures are presented as values ± standard error of the mean.
A significant (p b 0.001) multivariate main effect of muscles on burst activity across all frailty groups was found. Univariate tests demonstrated a significant (p b 0.05) main effect of muscles on all burst characteristics. Pair-wise comparisons revealed that the burst percentage was greater (p≤ 0.01) in both of the arm muscles (BB and TB) compared with the two thigh muscles (VL and BF). In addition, burst percentage was greater (p b 0.001) in the TB than the BB; however, there was no difference (p = 0.19) between the VL and BF (Fig. 3A). The number of bursts was greater (p b 0.001) in the arm muscles (BB and TB) than the thigh muscles (VL and BF); however, there was no difference between the VL and BF and between the BB and TB (Fig. 3B). Mean burst duration was greater (p b 0.05) in the TB compared with the
3. Results Ten women were categorized as non-frail, 11 as pre-frail, and 12 as frail (Table 1). Height, weight, number of self-reported comorbidities, number of medications, and number of falls within the past year were similar among the three groups (p N 0.05). Frail women were older, had weaker leg extension strength, and completed fewer steps during the 9-hour testing compared with the pre-frail and non-frail women (p b 0.05). In contrast, no differences were found in age, leg extension muscle strength, and mobility between non-frail and pre-frail women (p N 0.05) (Table 1). The two way interaction of frailty by muscle was non-significant for both burst (p = 0.06) and gap activity (p = 0.96).
3.1. Frailty A significant (p = 0.001) multivariate main effect of frailty on burst activity across all muscles was found. Univariate tests demonstrated a significant (p b 0.05) main effect of frailty on three of the burst characteristics (number of bursts, mean burst duration, and mean burst amplitude). Burst percentage (Fig. 1A) and peak amplitude (non-frail 8.8 ± 3.0, pre-frail 8.8 ± 2.1, frail 7.7 ± 2.4% MVE) were similar (p N 0.05) across all frailty groups. Post-hoc testing revealed that the number of bursts was less (p = 0.01) in the frail than the nonfrail women, and pre-frail women did not differ from non-frail or frail (Fig. 1B). Mean duration of bursts was greater (p b 0.05) in frail and pre-frail than in the non-frail women; however, there was no difference (p = 0.73) between the pre-frail and frail women (Fig. 1C). Mean burst amplitude was greater (p b 0.05) in the prefrail women than the non-frail and frail women. No difference (p = 0.54) was found in the mean burst amplitude between the frail and non-frail women (Fig. 1D). A significant (p = 0.01) multivariate main effect of frailty on gap activity across all muscles was found. Univariate tests demonstrated a significant (p b 0.05) main effect of frailty on number and mean duration of gaps. The gap percentage was similar (p = 0.91) across all frailty groups (Fig. 2A). Post-hoc testing revealed that the number of gaps was greater (p b 0.01) in the frail than the non-frail and pre-frail women; however, there was no difference between the non-frail and pre-frail women (Fig. 2B). Mean gap duration was less (p = 0.01) in the frail than the pre-frail women. In contrast, no difference (p N 0.05) was found in the mean gap duration of the non-frail women compared with the pre-frail and frail women (Fig. 2C).
Fig. 2. Gap activity for non-frail, pre-frail, and frail women. (A) Gap percentage; (B) Number of gaps; and (C) Mean gap duration. %, percentage; s, seconds. ⁎Significantly different from non-frail women; $Significantly different from pre-frail women (p b 0.05).
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
913
Fig. 3. Burst activity for the vastus lateralis, biceps femoris, biceps brachii, and triceps brachii. (A) Burst percentage; (B) Number of bursts; (C) Mean burst duration; (D) Peak burst amplitude; and (E) Mean burst amplitude. %, percentage; s, seconds; MVE, maximal voluntary exertion; VL, vastus lateralis; BF, biceps femoris; BB, biceps brachii; TB, triceps brachii. ⁎Significantly different from VL; #Significantly different from BF; $Significantly different from BB; €Significantly different from TB (p b 0.05).
VL and BB (Fig. 3C). Peak burst amplitude was less (p b 0.01) in the BB compared with the other three muscles, and the BF was greater (p b 0.05) than the VL and TB (Fig. 3D). Mean burst amplitude was greater (p b 0.05) in the TB compared with the other 3 muscles (Fig. 3E). A significant (p b 0.001) multivariate main effect of muscles on gap activity across all frailty groups was found. Univariate tests demonstrated a significant (p b 0.001) main effect of muscles on all gap characteristics. Pair-wise comparisons revealed that the gap percentage was less (p b 0.01) in the TB compared with the other three muscles and in the BB compared with the BF (Fig. 4A). The number of gaps was less (p b 0.01) in the BB compared with the other three muscles and in the TB compared with the BF (Fig. 4B). Mean gap duration was greater (p b 0.05) in the BB compared with the other three muscles and in the VL compared with the TB (Fig. 4C).
3.3. Relation between mobility and burst and gap characteristics The number of steps was significantly related to all burst characteristics of the two thigh muscles (Fig. 5). Pearson's correlations between the number of steps and the gap characteristics of the two thigh muscles were not significant (r = 0.01–0.22; p N 0.05). 4. Discussion Daily muscle activity and quiescence of the two upper limb (BB, TB) and two lower limb (VL, BF) muscles, quantified by burst and gap activity, was compared between non-frail, pre-frail, and frail older women. To our knowledge this is the only study that has examined the association of EMG with frailty during daily life. The percentage of
914
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
44% of the total recording time, which is similar to the total duration of muscle activity that we found for the arm muscles (30–39%). Kern et al. (2001) found that VL and BB of women were active 12% and 23% of the recording time; however, these participants were young women which likely accounts for the shorter total duration of muscle activity compared with our study of older women. Howe and Rafferty (2009) reported that the VL of women (mean age 64 years) was active 10% of the recording time which is less than the duration observed in these groups of non-frail, pre-frail, and frail women. These observed differences from previously published studies indicate quantifying bursts and gaps enables dissociation of muscle activity between healthy young and old; the current study further suggests that frail adults can be distinguished from non-frail. 4.1. Effect of frailty on muscle activity and quiescence
Fig. 4. Gap activity for the vastus lateralis, biceps femoris, biceps brachii, and triceps brachii. (A) Gap percentage; (B) Number of gaps; and (C) Mean gap duration.%, percentage; s, seconds; VL, vastus lateralis; BF, biceps femoris; BB, biceps brachii; TB, triceps brachii. ⁎Significantly different from VL; #Significantly different from BF; $ Significantly different from BB; €Significantly different from TB (p b 0.05).
time the muscles were active and quiescent were similar across all frailty groups (Figs. 1A; and 2A); however, the characteristics of muscle activity (bursts and gaps) were different. In frail women there were fewer muscle bursts but these bursts were for a longer duration compared with non-frail women. Moreover, the number of gaps was greater in frail women, but the duration was much shorter (~28%; Fig. 2C). Thus, the pattern of muscle activity is dissimilar across the frailty spectrum and likely offers a measure to dissociate stages of frailty. Mork and Westgaard (2005) measured long-term EMG activity in the trapezius and low back muscles of sedentary young and middle aged women and found that the duration of muscle activity was 29–
The number of bursts observed in frail women was 28% fewer compared with the non-frail women, whereas there were 29% and 25% more gaps in frail women than non-frail and pre-frail women, respectively. This suggests that muscle activity was less in frail women relative to non-frail and pre-frail women. This study showed that muscle activity differs across stages of frailty; however, it is not possible to determine whether frailty or changes in muscle activity comes first. Future longitudinal studies may address this question. Frail women were older and 77–85% less mobile (accelerometry) than non-frail and pre-frail women. Mobility was highly correlated with the number of bursts across all frailty groups. Chronological age and mobility likely account for the limited muscle activity observed in these frail older women compared to non-frail. The number of steps completed in this study was much lower than those found in other studies even for the non-frail older adults (Harris et al., 2009; Webber and Porter, 2009). However, we recorded steps only for 9 h compared with other studies that examine steps over a 24-hour period. In addition, physical activity participation measured by questionnaire in Greek older adults, is low compared with other European countries (Tzorbatzakis and Sleap, 2007). No studies have examined physical activity in Greek older adults using accelerometers but our data on the non-frail older women suggest that physical activity is likely quite low. Actigraph accelerometer steps counts have been shown to be accurate for walking speeds above 0.9 m/s but not for lower speeds (Webber and Porter, 2009), thus the steps recorded in our study may have been underestimated for frail older women. Previous work in healthy non-frail older adults has indicated that a reduction in the number of times that muscles were active is associated with an increase in the duration of each muscle burst (Harwood et al., 2008; Jakobi et al., 2008). These studies suggest that muscles of older adults burst fewer times but for longer duration relative to younger adults. This change is progressive as the oldest and most frail in this study had fewer bursts of longer duration relative to the non-frail. Longer duration bursts might be attributed to rate of movement. Fast velocity movements are more affected by aging than slow velocity movements (Candow and Chillibeck, 2005; Petrella et al., 2005). Also, burst amplitude was 36% less in frail women compared with the pre-frail women. These differences might be related to frail women requiring greater force generation for daily activities. Bursts and gaps discriminated later stage frailty, but only bursts discriminated early stage frailty (differences in pre-frail women compared with the non-frail women). Each burst in pre-frail women was approximately 30% longer with a 40% greater amplitude than the muscles of non-frail women. These two groups were of similar age, anthropometric and health characteristics, strength, and mobility. Factors such as impaired motor control and increased subjective fatigue, which are both outcomes of frailty, likely contribute to the differences in muscle activity (Dedering et al., 1999; Ferucci et al.,
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
915
Fig. 5. The relationship between number of steps and burst activity of the two thigh muscles. (A) Bursts percentage; (B) Number of bursts; (C) Mean bursts duration; (D) Peak bursts amplitude; and (E) Mean burst amplitude. r, Pearson's correlation coefficient; %, percentage; s, seconds; MVE, maximal voluntary exertion.
2004; Laursen et al., 2001). Other physiological factors (e.g. muscle fiber-type proportion, motor unit firing rate, nerve conduction velocity, and muscle fatigue) may be related to differences between non-frail and pre-frail, but studies have yet to examine the effect of frailty on these characteristics relative to bursts and gaps. Future studies should examine underlying mechanisms that contribute to changes in burst and gap activity across the stages of frailty. 4.2. Differences between muscles on activity and quiescence This study demonstrated that across all frailty groups lower limb muscles were active 22% (~ 2 h) and quiescent 50% (~4.5 h) of the
time. However, the percentages of bursts were greater in the upper limb muscles compared to the lower limb muscles. These results are consistent with a previous study in younger adults which reported that arm muscles are more active relative to thigh muscles (Kern et al., 2001). The greater activity of the upper limb muscles compared with the lower limb muscles (e.g. ~ 60% greater total duration of muscle activity) is likely because they are needed extensively by older adults to execute activities of daily living (ADL) (Hortobagyi et al., 2003). Older women may spend much of their day seated or standing performing a variety of household activities that utilize the upper limbs. These EMG recordings of daily life suggest that upper limb movement is maintained relative to lower body ambulation. A
916
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917
previous study (Harwood et al., 2008) has shown that when the same task was performed there were age- and sex-related differences in muscle activity. Future studies should examine how older women activate arm and thigh muscles to perform ADL (e.g. video analysis of ADL). Muscle activity (except peak amplitude) and quiescence were similar in the two thigh muscles (VL and BF) across all three frailty groups. In contrast, differences in muscle activity and quiescence were found between the two arm muscles (BB and TB). Triceps brachii was active 39% of the time and quiescent 27% of the time; whereas BB was active 30% of the time and quiescent 41% of the time. Previous work in young adults indicated similarities in muscle activity between the lower limb muscles but not the upper limb muscles (Kern et al., 2001). In addition, mean duration, mean amplitude, and peak amplitude of muscle activity were greater in the TB, whereas mean duration of muscle quiescence was greater in the BB. In contrast, the number of times that the muscles were active was similar between the two arm muscles and the number of times that the muscles were quiescent was greater in the TB. The observed differences between the two arm muscles in our study may be related to the greater strength of the BB compared with the TB and the greater usage of the TB for daily activities. A longitudinal study of older men described the rate of decline in muscle strength per year to be 1.6% and 1.8% for the TB and BB, respectively (Frontera et al., 2000). In addition, TB muscle mass (Candow and Chillibeck, 2005) and its ability to activate (Jakobi and Rice, 2002) is better maintained with age than the BB and it may be relied on more to complete daily activities. This would result in greater muscle activity and less muscle quiescence, as observed. Alternatively, little is understood about fiber types between these muscles and thus differences in young adults as well as unique age-related changes within each muscle might contribute to the observed patterns of muscle activation between these arm muscles (Dalton et al., 2010). Muscle activity was measured over one day, and for a time period limited to approximately 10 h because many of the participants were frail and the battery capacity of the recording unit. Future research should examine the day to day reliability in measuring daily muscle activity using portable EMG in older adults across levels of frailty, and the minimal recording period to represent daily muscle activity patterns. In addition, the older women commented during the posttesting discussion that the portable EMG device was minimally obtrusive; however, future studies should examine whether participants alter personal daily habits when portable EMG devices are worn. 5. Conclusions Frailty alters the individual characteristics of muscle activity (bursts) and quiescence (gaps) which may assist in dissociation between stages of frailty. Muscle activity discriminated both early and later stage frailty; however, muscle quiescence discriminated later stage frailty. Mobility was associated with the differences observed in muscle activity across the stages of frailty. Although mobility and strength, which are two known frailty indicators, were statistically similar between the non-frail and pre-frail women these groups differed in muscle activity. Moreover, measures of bursts and gaps in EMG provided identification of muscle patterning between the upper and lower limbs, and these measures might be useful in quantifying agonist and antagonist coordination, as well as functional use. References Bandeen-Roche, K., Xue, Q.-L., Ferruci, L., Walston, J., Guralnik, J.M., Chaves, P., Zeger, S.L., Fried, L.P., 2006. Phenotype of frailty: characterization in the women's health and aging studies. J. Gerontol. A Biol. Sci. Med. Sci. 61 (3), 262–266.
Bauer, J.M., Sieber, C.C., 2008. Sarcopenia and frailty: a clinician's controversial point of view. Exp. Gerontol. 43, 674–678. Blangsted, A.K., Hansen, K., Jensen, C., 2003. Muscle activity during computer-based office work in relation to self-reported job demands and gender. Eur. J. Appl. Physiol. 89, 352–358. Blangsted, A.K., Sogaard, K., Christensen, H., Sjogaard, G., 2004. The effect of physical and psychological loads on the trapezius muscle activity during computer keying tasks and rest periods. Eur. J. Appl. Physiol. 91, 253–258. Candow, D.G., Chillibeck, P.D., 2005. Differences in size, strength, and power of upper and lower body muscle groups in young and older men. J. Gerontol. A Biol. Sci. Med. Sci. 60, 148–156. Central Intelligence Agency, 2009. The world factbook 2009. ISSN 1553-8133, Washington, DC, viewed February 10, 2010 from https://www.cia.gov/library/ publications/the-world-factbook/index.html. Cesari, M., Leeuwenburgh, C., Lauretani, F., Onder, G., Bandinelli, S., Maraldi, C., Guralnik, J.M., Pahor, M., Ferrucci, L., 2006. Frailty syndrome and skeletal muscle: results from the Invecchiare in Chianti study. Am. J. Clin. Nutr. 83, 1142–1148. Chandler, J.M., Duncan, P.W., Kochersberger, G., Studenski, S., 1998. Is lower extremity strength gain associated with improvement in physical performance and disability in frail, community-dwelling elders? Arch. Phys. Med. Rehabil. 79, 24–30. Dalton, B.H., Jakobi, J.M., Allman, B.L., Rice, C.L., 2010. Differential age-related changes in motor unit properties between elbow flexors and extensors. Acta Physiol. 200, 45–55. Dedering, A., Nemeth, G., Harms-Ringdahl, K., 1999. Correlation between electromyographic spectral changes and subjective assessment of lumbar muscle fatigue in subjects without pain from the lower back. Clin. Biomech. 14, 103–111. Ferucci, L., Guralnik, J.M., Studenski, S., Fried, L.P., Cutler, G.B., Walston, J.D., 2004. Designing randomized, controlled trials aimed at preventing or delaying functional decline and disability in frail, older persons: a consensus report. J. Am. Geriatr. Soc. 52, 625–634. Fried, L.P., Tangen, C.M., Walston, J., Newman, A.B., Hirsch, C., Gottdiener, J., Seeman, T., Tracy, R., Kop, W.J., Burke, G., McBurnie, M.A., for the Cardiovascular Health Study Collaborative Research Group, 2001. Frailty in older adults: evidence for a phenotype. J. Gerontol. A Biol. Sci. Med. Sci. 56 (3), M146–M156. Fried, L.P., Ferrucci, L., Darer, J., Williamson, J.D., Anderson, G., 2004. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J. Gerontol. A Biol. Sci. Med. Sci. 59 (3), 255–263. Frontera, W.R., Hughes, V.A., Fielding, R.A., Fiatarone, M.A., Evans, W.J., Roubenoff, R., 2000. Aging of skeletal muscle: a 12-yr longitudinal study. J. Appl. Physiol. 88, 1321–1326. Harris, T.J., Owen, C.G., Victor, C.R., Adams, R., Cook, D.G., 2009. What factors are associated with physical activity in older people, assessed objectively by accelerometry? Br. J. Sports Med. 43, 442–450. Harwood, B., Edwards, D.L., Jakobi, J.M., 2008. Age- and sex-related differences in muscle activation for a discrete functional task. Eur. J. Appl. Physiol. 103, 677–686. Holtermann, A., Sogaard, K., Christensen, H., Dahl, B., Blangsted, A.K., 2008. The influence of biofeedback training on trapezius activity and rest during occupational computer work: a randomized controlled trial. Eur. J. Appl. Physiol. 104, 983–989. Hortobagyi, T., Mizelle, C., Beam, S., DeVita, P., 2003. Old adults perform activities of daily living near their maximal capabilities. J. Gerontol. A Biol. Sci. Med. Sci. 58 (5), M453–M460. Howe, T.E., Rafferty, D., 2009. Quadriceps activity and physical activity profiles over long durations in patients with osteoarthritis of the knee and controls. J. Electromyogr. Kinesiol. 19 (2), e78–e83. Jakobi, J.M., Rice, C.L., 2002. Voluntary muscle activation varies with age and muscle group. J. Appl. Physiol. 93, 457–462. Jakobi, J.M., Edwards, D.L., Connelly, D.M., 2008. Utility of portable electromyography for quantifying muscle activity during daily use. Gerontology 54, 324–331. Kern, D.S., Semmler, J.G., Enoka, R.M., 2001. Long-term activity in upper- and lowerlimb muscles of humans. J. Appl. Physiol. 91, 2224–2232. Latham, N.K., Anderson, C.S., Lee, A., Bennett, D.A., Moseley, A., Cameron, I.D., Fitness Collaborative Group, 2003. A randomized, controlled trial of quadriceps resistance exercise and vitamin D in frail older people: the Frailty Interventions Trial in Elderly Subjects (FITNESS). J. Am. Geriatr. Soc. 51 (3), 291–299. Laursen, B., Jensen, B.R., Ratkevicius, A., 2001. Performance and muscle activity during computer mouse tasks in young and elderly adults. Eur. J. Appl. Physiol. 84, 329–336. Madeleine, P., Vedsted, P., Blangsted, A.K., Sjogaard, G., Sogaard, K., 2006. Effects of electromyographic and mechanomyographic biofeedback on upper trapezius muscle activity during standardized computer work. Ergonomics 49 (10), 921–933. Mork, P.J., Westgaard, R.H., 2005. Long-term electromyographic activity in upper trapezius and low back muscles of women with moderate physical activity. J. Appl. Physiol. 99, 570–578. Mork, P.J., Westgaard, R.H., 2006. Low-amplitude trapezius activity in work and leisure and the relation to shoulder and neck pain. J. Appl. Physiol. 100, 1142–1149. Paterson, D.H., Jones, G.R., Rice, C.L., 2007. Ageing and physical activity: evidence to develop exercise recommendations for older adults. Can. J. Public Health 98, S69–S108. Petrella, J.K., Kim, J., Tuggle, S.C., Hall, S.R., Bamman, M.M., 2005. Age differences in knee extension power, contractile velocity, and fatigability. J. Appl. Physiol. 98, 211–220. Rantanen, T., Harris, T., Leveille, S.G., Visser, M., Foley, D., Masaki, K., Guralnik, J.M., 2000. Muscle strength and body mass index as long-term predictors of mortality in initially health men. J. Gerontol. A Biol. Sci. Med. Sci. 55A (3), M168–M173. Rockwood, K., Mitnitski, A., 2007. Frailty in relation to the accumulation of deficits. J. Gerontol. A Biol. Sci. Med. Sci. 62A (7), 722–727.
O. Theou et al. / Experimental Gerontology 45 (2010) 909–917 Santos-Eggimann, B., Cuenoud, P., Spagnoli, J., Junod, J., 2009. Prevalence of frailty in middle-aged and older community-dwelling Europeans living in 10 countries. J. Gerontol. A Biol. Sci. Med. Sci. 64A (6), 675–681. Tzorbatzakis, N., Sleap, M., 2007. Participation in physical activity and exercise in Greece: a systematic literature review. Int. J. Public Health 52, 360–371.
917
Vandervoort, A.A., 2002. Aging of the human neuromuscular system. Muscle Nerve 25, 17–25. Webber, S.C., Porter, M.M., 2009. Monitoring mobility in older adults using global positioning system (GPS) watches and accelerometers: a feasibility study. J. Aging Phys. Act. 17, 455–467.