Benefits of Resistance Training in Early and Late Stages of Frailty and Sarcopenia: A Systematic Review and Meta-Analysis of Randomized Controlled Studies

Sarcopenia and frailty are age-related syndromes with negative effects on the quality of life of older people and on public health costs. Although extensive research has been carried out on the effects of physical exercise and physical syndromes, there is a knowledge gap when it comes to the effect of resistance training on muscular strength, physical performance, and body composition at early (prevention) and late (treatment) stages in both syndromes combined. We conducted this systematic review and meta-analysis (CRD42019138253) to gather the evidence of randomized controlled trials examining the effects of resistance training programs lasting ≥8 weeks on strength, physical function, and body composition of adults ≥65 years old diagnosed with pre-sarcopenia, sarcopenia, pre-frailty, or frailty. A search from the earliest record up to and including December 2020 was carried out using the PubMed, Scopus, Web of Science, and Cochrane Library databases. A total of 25 studies (n = 2267 participants) were included. Meta-analysis showed significant changes in favour of resistance training for handgrip (ES = 0.51, p = 0.001) and lower-limb strength (ES = 0.93, p < 0.001), agility (ES = 0.78, p = 0.003), gait speed (ES = 0.75, p < 0.001), postural stability (ES = 0.68, p = 0.007), functional performance (ES = 0.76, p < 0.001), fat mass (ES = 0.41, p = 0.001), and muscle mass (ES = 0.29, p = 0.002). Resistance training during early stages had positive effects in all variables during early stages (ES > 0.12), being particularly effective in improving gait speed (ES = 0.63, p = 0.016) and functional strength (ES = 0.53, p = 0.011). Based on these results, resistance training should be considered as a highly effective preventive strategy to delay and attenuate the negative effects of sarcopenia and frailty in both early and late stages.


Introduction
Life expectancy is the highest to date, and world aging has increased at a staggering rate [1]. Despite the fact that people live longer than ever, human aging produces various syndromes that reduce their quality of life, contribute to their dependence, and increase public care costs [2]. Prominent among these syndromes are sarcopenia and frailty. A recent estimate from 28 European countries suggests increments of 60-70% in the prevalence of sarcopenia by 2045, resulting in 12.9 to 22.3% of people over 65 years old being affected [3]. Sarcopenia is generated by a severe loss of muscle mass as a consequence of diverse factors such as nutritional status, physical activity, genetic heritability, or hormonal changes [4]. This fact, together with a decline in the tendon proprieties [5] and neural patterns [6], results in a loss of muscular strength and mobility (i.e., functional status). Frailty, for its

Methodological Quality
As described in our PROSPERO protocol, we initially intended to use the GRADE scale as a widely recommended system for observational studies and randomized controlled trials [29]. However, subsequent to the protocol registration, we decided to use the Physiotherapy Evidence Database (PEDro) scale to assess the methodological quality of included studies [30]. PEDro provides an assessment of the quality of randomized controlled trials, especially in evidence-based physical-therapy [31]. The PEDro scale has demonstrated reliability with score range from 0 to 11, where scores ≤ 3 represent poor study quality, scores of 4-5 indicate fair quality, and scores ≥ 6 represent good to excellent quality [32]. This change to the protocol has been registered in PROSPERO.

Statistical Analysis
The effect sizes (ES) were calculated as the standardized mean differences between the resistance training group and the control group. Sub-group analyses were conducted to examine ES for early stages of sarcopenia and frailty. As traditional meta-regression methods do not allow for multiple dependent outcomes from the same study to be included in one analysis, we used a meta-analytic method for dealing with dependent effect sizes named robust variance estimation (RVE). RVE is a form of random-effects meta-regression for multilevel data structures, which allows for multiple effect sizes from the same study to be included in a meta-analysis, even when information on the covariance of these effect sizes is unavailable. Instead, RVE estimates the variance of meta-regression coefficient estimates using the observed residuals. It does not require distributional assumptions and does not make any requirements on the weights [33,34]. A study was used as the clustering variable to account for correlated effects within studies. Observations were weighted by the inverse of the sampling variance. A sensitivity analysis using alternative correlational values to calculate the standard error revealed that the choice of correlational value did not impact the overall results of the meta-analysis. I 2 was used to evaluate between-study heterogeneity. Values of I 2 more than 25%, 50%, and 75% were selected to reflect low, moderate, and high heterogeneity, respectively [35]. All analyses were performed using packages robumeta (version 2.0) and metafor (version 2.0-0) in R version 3.4.4 (The R Foundation for Statistical Computing, Vienna, Austria).

Study Selection
The database search yielded 1468 articles. Of those, 155 full texts were retrieved, and 26 deemed eligible . As one study reported its results in two separate articles, 25 studies were included in this review, as shown in the PRISMA flow diagram (Figure 1). regression for multilevel data structures, which allows for multiple effect sizes from the same study to be included in a meta-analysis, even when information on the covariance of these effect sizes is unavailable. Instead, RVE estimates the variance of meta-regression coefficient estimates using the observed residuals. It does not require distributional assumptions and does not make any requirements on the weights [33,34]. A study was used as the clustering variable to account for correlated effects within studies. Observations were weighted by the inverse of the sampling variance. A sensitivity analysis using alternative correlational values to calculate the standard error revealed that the choice of correlational value did not impact the overall results of the meta-analysis. I 2 was used to evaluate between-study heterogeneity. Values of I 2 more than 25%, 50%, and 75% were selected to reflect low, moderate, and high heterogeneity, respectively [35]. All analyses were performed using packages robumeta (version 2.0) and metafor (version 2.0-0) in R version 3.4.4 (The R Foundation for Statistical Computing, Vienna, Austria).

Study Selection
The database search yielded 1468 articles. Of those, 155 full texts were retrieved, and 26 deemed eligible . As one study reported its results in two separate articles, 25 studies were included in this review, as shown in the PRISMA flow diagram (Figure 1).

Methodological Quality
The methodological quality of the studies is detailed in Table 1. Since all the studies obtained the predefined minimum score of 6 points, they were all included in the qualitative and quantitative syntheses. The minimum, maximum, and mean scores of the quality analysis were 6, 11, and 8.76 (±1.26) points, respectively.

Study Characteristics
In total, there were 2267 participants (1484 women). The mean age ranged from 62 to 98 years. The mean duration of resistance training programs was approximately 23 weeks (range 10-48 weeks), and the most common training frequency was 2-3 times per week (in 21 studies). Tables 2 and 3 show the characteristics and the overall effect of the 25 studies included in the review. Table A2 presents the diagnostic criteria for frailty or sarcopenia with the prevalence of participants for each study.
Usual care, no exercises. Participants had access to one standard care from health and aged care services that were normally available to older people, including primary and secondary level care from government or private clinics and hospitals, and community-based social, recreational, and day-care rehabilitation services. 90 min group chair-based RT using Thera-bands, and 20 min aerobic exercises, one-home session weekly, gait speed, twice per week.

Discussion
This systematic review found that resistance training is a highly effective strategy to improve muscular strength, physical function, and body composition parameters in older adults with pre-frailty, frailty, pre-sarcopenia, or sarcopenia. Besides, resistance training during early stages had positive effects in all variables, being particularly effective in improving physical function. These findings reinforce the use of strength training interventions to delay and attenuate negative effects related to both physical syndromes.

Muscular Strength
Muscular strength is considered the primary determinant of sarcopenia [62]. To date, the handgrip evaluation represents the most common test used to measure this physical capacity [63] due to its high affordability, portability, simplicity, and test-retest repeatability [64]. Nevertheless, to obtain an overall indicator of strength, some investigations have suggested the need to complement this test with specific evaluations of the lowerlimb muscles (e.g., isometric knee extension) [65][66][67]. The present study found that individuals suffering from (pre-) sarcopenia or (pre-) frailty significantly improved both handgrip (ES = 0.51, p = 0.001) and lower-limb (ES = 0.93, p < 0.001) strength after a training intervention based on resistance exercises. Indeed, except for one study in each analysis (handgrip [57] and lower-limb [60]), all investigations reported effects in favor of the resistance training group (Figures 2 and 3). Moreover, specifically to lower-limb strength, our results revealed that these strength enhancements were detected both isometrically [37,41,43,49,51,61] and dynamically [38][39][40]46].

Discussion
This systematic review found that resistance training is a highly effective strategy to improve muscular strength, physical function, and body composition parameters in older adults with pre-frailty, frailty, pre-sarcopenia, or sarcopenia. Besides, resistance training during early stages had positive effects in all variables, being particularly effective in improving physical function. These findings reinforce the use of strength training interventions to delay and attenuate negative effects related to both physical syndromes.

Muscular Strength
Muscular strength is considered the primary determinant of sarcopenia [62]. To date, the handgrip evaluation represents the most common test used to measure this physical capacity [63] due to its high affordability, portability, simplicity, and test-retest repeatability [64]. Nevertheless, to obtain an overall indicator of strength, some investigations have suggested the need to complement this test with specific evaluations of the lowerlimb muscles (e.g., isometric knee extension) [65][66][67]. The present study found that individuals suffering from (pre-) sarcopenia or (pre-) frailty significantly improved both handgrip (ES = 0.51, p = 0.001) and lower-limb (ES = 0.93, p < 0.001) strength after a training intervention based on resistance exercises. Indeed, except for one study in each analysis (handgrip [57] and lower-limb [60]), all investigations reported effects in favor of the resistance training group (Figures 2 and 3). Moreover, specifically to lower-limb strength, our results revealed that these strength enhancements were detected both isometrically [37,41,43,49,51,61] and dynamically [38][39][40]46].

Discussion
This systematic review found that resistance training is a highly effective strategy to improve muscular strength, physical function, and body composition parameters in older adults with pre-frailty, frailty, pre-sarcopenia, or sarcopenia. Besides, resistance training during early stages had positive effects in all variables, being particularly effective in improving physical function. These findings reinforce the use of strength training interventions to delay and attenuate negative effects related to both physical syndromes.

Muscular Strength
Muscular strength is considered the primary determinant of sarcopenia [62]. To date, the handgrip evaluation represents the most common test used to measure this physical capacity [63] due to its high affordability, portability, simplicity, and test-retest repeatability [64]. Nevertheless, to obtain an overall indicator of strength, some investigations have suggested the need to complement this test with specific evaluations of the lower-limb muscles (e.g., isometric knee extension) [65][66][67]. The present study found that individuals suffering from (pre-) sarcopenia or (pre-) frailty significantly improved both handgrip (ES = 0.51, p = 0.001) and lower-limb (ES = 0.93, p < 0.001) strength after a training intervention based on resistance exercises. Indeed, except for one study in each analysis (handgrip [57] and lower-limb [60]), all investigations reported effects in favor of the resistance training group (Figures 2 and 3). Moreover, specifically to lower-limb strength, our results revealed that these strength enhancements were detected both isometrically [37,43,47,49,51,61] and dynamically [38][39][40]46].

Physical Function
We found that all of the analyzed functional capacities were significantly improved by the implementation of a resistance training intervention (ES from 0.68 to 0.78). With the exception of one study for the agility [56] and gait [41] tasks (Figures 6 and 8), and two investigations for the balance task [58,59] (Figure 7), all studies found effects in favor of the resistance training group. Furthermore, all investigations reported superior effects for the resistance training group in relation to functional tasks (Figure 9). These findings could be strongly related to the significant lower-limb strength gains (Figure 3). Since the lower-limb muscles (e.g., knee extensors) are mainly responsible for actions such as chair rising or walking [68][69][70], the increment of strength in these structures could have been positively transferred into the physical function. In turn, these improvements in physical function can potentially reduce the dependency situation of older adults, thus increasing their quality of life [71] and decreasing the public health costs [72,73].

Body Composition
Our results revealed a positive effect of resistance training on the reduction of fat mass (ES = 0.41, p = 0.001, Figure 8) and increases in muscle mass (ES = 0.29, p = 0.002, Figure 9). Since the muscle mass can explain approximately 60-70% of strength capacity [74], these muscle mass enhancements are strongly related to the strength gains described above (Figures 2 and 3). Similarly, the increases in muscle mass could have generated the decreases in fat mass as a result of the rise in the energy expenditure of the individuals [75,76]. Together, these positive changes in body composition parameters could reduce the risk of other common diseases in older adults, such as metabolic syndrome [77][78][79].
Generally, exercise interventions can decrease the prevalence of frailty and sarcopenia and are also effective in reducing the severity of these syndromes [12]. Our results are consistent with previous studies supporting that resistance training is beneficial for the muscular strength and physical function in older adults with frailty or sarcopenia [10,22,24], but they do not combine both syndromes (sarcopenia and frailty) and if they do, they examine the effect of exercise overall [23]. These new results support evidence that resistance training is the most effective exercise type of intervention to improve muscle strength and physical performance in older people compared to endurance training or whole-body vibration training [18]. More specifically, it seems preferable to perform multi-component exercise programs combining a power-oriented resistance training regime with endurance and balance exercises [11,80].
Great emphasis should also be placed on the issue of financial sustainability of healthcare. It has been observed that frailty and sarcopenia lead to the increase of public health costs [81][82][83][84][85]. According to Bock et al., 2016 the mean total 3-month costs of frail participants in Saarland, Germany were € 3659 and non-frail older adults € 642, thus more than 80% of costs could be easily saved [8]. For this reason, preventing, postponing or even reducing frailty could potentially decrease total healthcare costs in many countries.

Limitations
This study is not exempt from limitations. Firstly, except for the functional strength and fat mass, most of the meta-analyses indicated moderate to high levels of heterogeneity. This fact could be explained mainly by the different variables included in the quantitative analysis (i.e., clinical diversity), as well as by the different methodologies (e.g., volume, intensity, exercise, program duration) used in each study (i.e., methodological diversity). Secondly, although the mean duration of training interventions included in the present review (~20 weeks) allows us to suggest that resistance training is an effective short/medium-term strategy, more evidence including longer resistance training programs is needed to confirm the long-term benefits, in particular, whether they are effective for reducing prevalence of sarcopenia and frailty. Thirdly, future systematic reviews are encouraged to examine the effects of resistance training on other physiological parameters, such as the neural drive, muscle architecture, or tendon proprieties, among individuals with pre-sarcopenia, sarcopenia, pre-frailty, or frailty.

Conclusions
Based on these results, resistance training should be considered as a highly effective preventive strategy to delay and attenuate the negative effects of sarcopenia and frailty in both early and late stages.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.
Appendix A   Table A1. The search terms used in the review to identify resistance exercise intervention designed to improve (pre-) sarcopenic and (pre-)frail older adults' strength, physical function and body composition. These search terms were used in PubMed, Web of Science, Cochrane Library and Scopus databases. PubMed Filters: Randomized controlled trial, Full text, 1975-2020, Aged 65+, Humans, English.

Variable Terms
Outcome term ("sarcopenia" OR "presarcopenia" OR "*sarcopenia" OR "pre-sarcopenia" OR "presarcopenic" OR "pre-sarcopenic" OR "sarcopenic" AND "frailty" OR "*frailty" OR "prefrailty" OR "pre-frailty" OR "frail" OR "frail*" OR "prefrail" OR "pre-frail") Measurement parameters ("muscle strength" OR "muscular strength" OR "muscle mass" OR "fat mass" OR "FFM" OR "body composition" OR "gait speed" OR "gait" OR "walking speed" OR "balance" OR "appendicular skeletal muscle mass" OR "body fat mass" OR "balance test" OR "HGS" OR "hand strength" OR "grip strength" OR "lower limb strength" OR "chair stand time" OR "knee extension 1-RM" OR "TUG" OR "agility" OR "SPPB" OR "physical function" OR "one leg stand time" OR "body fat percentage" OR "total body fat mass" OR "upper limb fat mass" OR "lower limb fat mass" OR "walking speed" OR "knee flexion" OR "knee extension isokinetic" OR "knee extension isometric" OR "leg lean mass" OR "thigh muscle area" OR "muscle mass" OR "fat mass" OR "fat-free mass" OR "calf circumference" OR "arm lean mass" OR "total lean mass" OR "appendicular lean mass" OR "upper limb muscle mass" OR "lower limb muscle mass") Exercise intervention ("resistance training" OR "resistance exercise" OR "resistance exercises" OR "resistance" OR "strength training" OR "strength exercise" OR "strength exercises" OR "physical strength" OR "physical activity" OR "strength" OR "physical strength" OR "weight training" OR "weight" OR "exercises" OR "exercise" OR "training" OR "physical activity" OR "physical activities" OR "physical training" OR "physical fitness" OR "weight exercises" OR "weight exercise" OR "weight-bearing training" OR "weight-bearing exercises" OR "weight bearing exercise" OR "weight bearing training") Study design ("randomized controlled trial*" OR "controlled" OR "RCT" OR "clinical trial" OR "controlled clinical trial" OR "randomized controlled trials" OR "random allocation" OR "double blind method" OR "single blind method" OR "clinical trials" OR "single" OR "double" OR "triple" OR "placebos" OR "research design" OR "follow-up stud*" OR "placebo" OR "random" OR "comparative study" OR "evaluation studies" OR "prospective stud*" OR "control" OR "prospective*" OR "volunteer" OR "research design" OR "control") Population ("old" OR "age" OR "old age" OR "older adult*" OR "older people" OR "elderly" OR "elder*" OR "people" OR "aging adults" OR "ageing adults" OR "geriatric*" OR "senile people" OR "senile person" OR "older person*" OR "old-age" OR "older" OR "person*" OR "adult*" OR "elderly person*" OR "senior*" OR "aging" OR "ageing" OR "aged" OR "old man" OR "old men" OR "old woman" OR "old women" OR "older woman" OR "older women") AWGS criteria: ASM/height 2(ASM/Ht2) measured using DXA of less than 7.0 kg/m 2 for men and 5.4 kg/m 2 for women; low HGS (less than 26 kg for men and 18 kg for women) and/or low usual GS (less than or equal to 0.