**Prevention and Treatment of Sarcopenia**

Editor **Gianluca Testa**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editor* Gianluca Testa Dipartimento di Chirurgia Generale e Specialita` Medico-Chirurgiche Universita degli Studi di Catania ` Catania Italy

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Journal of Clinical Medicine* (ISSN 2077-0383) (available at: www.mdpi.com/journal/jcm/special issues/Prevention Treatment Sarcopenia).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-1536-6 (Hbk) ISBN 978-3-0365-1535-9 (PDF)**

© 2021 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

### **Contents**


Reprinted from: *Journal of Clinical Medicine* **2021**, *10*, 1018, doi:10.3390/jcm10051018 . . . . . . . . **91**


Effects of Physical Exercises and Verbal Stimulation on the Functional Efficiency and Use of Free Time in an Older Population under Institutional Care: A Randomized Controlled Trial Reprinted from: *Journal of Clinical Medicine* **2020**, *9*, 477, doi:10.3390/jcm9020477 . . . . . . . . . . **231** **Hyuma Makizako, Yuki Nakai, Kazutoshi Tomioka, Yoshiaki Taniguchi, Nana Sato, Ayumi Wada, Ryoji Kiyama, Kota Tsutsumimoto, Mitsuru Ohishi, Yuto Kiuchi, Takuro Kubozono and Toshihiro Takenaka**

Effects of a Multicomponent Exercise Program in Physical Function and Muscle Mass in Sarcopenic/Pre-Sarcopenic Adults

Reprinted from: *Journal of Clinical Medicine* **2020**, *9*, 1386, doi:10.3390/jcm9051386 . . . . . . . . . **251**

**Takumi Kawaguchi, Sachiyo Yoshio, Yuzuru Sakamoto, Ryuki Hashida, Shunji Koya, Keisuke Hirota, Dan Nakano, Sakura Yamamura, Takashi Niizeki, Hiroo Matsuse and Takuji Torimura**

Impact of Decorin on the Physical Function and Prognosis of Patients with Hepatocellular Carcinoma

Reprinted from: *Journal of Clinical Medicine* **2020**, *9*, 936, doi:10.3390/jcm9040936 . . . . . . . . . . **263**

### **About the Editor**

#### **Gianluca Testa**

Gianluca Testa is a researcher professor at University of Catania, Italy, since January 2018. He is one of the youngest nominated researchers in Italy, a medical doctor and surgeon, and an expert in the field of Orthopaedics and Traumatology. He practices at Universitary Hospital Policlinico Rodolico - San Marco, in Catania, treating both pediatric and older people. He is a delegate for the region Sicily of the Italian Pediatric Orthopaedics and Traumatology Society (S.I.T.O.P.) and for the Italian External Fixation Society (S.I.F.E.).

### **Preface to "Prevention and Treatment of Sarcopenia"**

Sarcopenia represents the decline in skeletal muscle mass and function with age, characterized by the muscle fiber's quality, strength, muscle endurance, and metabolic ability decreasing, as well as the fat and connective tissue growing.

Reduction of muscle strength with aging leads to loss of functional capacity, causing disability, mortality, and other adverse health outcomes. Because of the increase of the proportion of elderly in the population, sarcopenia-related morbidity will become an increasing area of health care resource utilization.

Diagnostic screening consists of individuation of body composition, assessed by DEXA, anthropometry, bioelectrical impedance, MRI, or CT scan. Management is possible with resistance training exercise and vibration therapy, nutritional supplements, and pharmacological treatment.

The book includes articles from different nationalities, treating the experimental and medical applications of sarcopenia. The consequences of sarcopenia in frailty are treated in relation to other associated pathologies or lesions, as femoral neck fractures and hepatocellular carcinoma.

> **Gianluca Testa** *Editor*

### *Article* **Prevalence of Sarcopenia Employing Population-Specific Cut-Points: Cross-Sectional Data from the Geelong Osteoporosis Study, Australia**

**Sophia X. Sui 1,\*, Kara L. Holloway-Kew <sup>1</sup> , Natalie K. Hyde <sup>1</sup> , Lana J. Williams <sup>1</sup> , Monica C. Tembo <sup>1</sup> , Sarah Leach <sup>2</sup> and Julie A. Pasco 1,3,4,5**


**Abstract:** Background: Prevalence estimates for sarcopenia vary depending on the ascertainment criteria and thresholds applied. We aimed to estimate the prevalence of sarcopenia using two international definitions but employing Australian population-specific cut-points. Methods: Participants (*n* = 665; 323 women) aged 60–96 years old were from the Geelong Osteoporosis Study. Handgrip strength (HGS) was measured by dynamometers and appendicular lean mass (ALM) by whole-body dual-energy X-ray absorptiometry. Physical performance was assessed using gait speed (GS, men only) and/or the timed up-and-go (TUG) test. Using cut-points equivalent to two standard deviations (SDs) below the mean young reference range from the same population and recommendations from the European Working Group on Sarcopenia in Older People (EWGSOP), sarcopenia was identified by low ALM/height<sup>2</sup> (<5.30 kg for women; <6.94 kg for men) + low HGS (<16 kg women; <31 kg men); low ALM/height<sup>2</sup> + slow TUG (>9.3 s); low ALM/height<sup>2</sup> + slow GS (<0.8 m/s). For the Foundation for the National Institutes of Health (FNIH) equivalent, sarcopenia was identified as low ALM/BMI (<0.512 m<sup>2</sup> women, <0.827 m<sup>2</sup> men) + low HGS (<16 kg women, <31 kg men). Receiver Operating Characteristic curves were also applied to determine optimal cut-points for ALM/BMI (<0.579 m<sup>2</sup> women, <0.913 m<sup>2</sup> men) that discriminated poor physical performance. Prevalence estimates were standardized to the Australian population and compared to estimates using international thresholds. Results: Using population-specific cut-points and low ALM/height<sup>2</sup> + HGS, point-estimates for sarcopenia prevalence were 0.9% for women and 2.9% for men. Using ALM/height<sup>2</sup> + TUG, prevalence was 2.5% for women and 4.1% for men, and using ALM/height<sup>2</sup> + GS, sarcopenia was identified for 1.6% of men. Using ALM/BMI + HGS, prevalence estimates were 5.5–10.4% for women and 11.6–18.4% for men. Conclusions: This study highlights the range of prevalence estimates that result from employing different criteria for sarcopenia. While population-specific criteria could be pertinent for some populations, a consensus is needed to identify which deficits in skeletal muscle health are important for establishing an operational definition for sarcopenia.

**Keywords:** sarcopenia; skeletal muscle; prevalence; muscle strength; physical functional performance; epidemiologic studies; aging

#### **1. Introduction**

While sarcopenia is characterized by age-related declines in skeletal muscle mass, strength and function, currently, there is no unanimously agreed operational definition for sarcopenia [1–4]. Several operational definitions have been developed, notably by the

**Citation:** Sui, S.X.; Holloway-Kew, K.L.; Hyde, N.K.; Williams, L.J.; Tembo, M.C.; Leach, S.; Pasco, J.A. Prevalence of Sarcopenia Employing Population-Specific Cut-Points: Cross-Sectional Data from the Geelong Osteoporosis Study, Australia. *J. Clin. Med.* **2021**, *10*, 343. https://doi.org/10.3390/jcm10020343

Received: 1 December 2020 Accepted: 14 January 2021 Published: 18 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

European Working Group on Sarcopenia in Older People (EWGSOP1 and EWGSOP2) [3,5] and the Foundation for the National Institutes of Health (FNIH) [4]. Sarcopenia parameters usually include low muscle mass and low muscle strength or performance to identify sarcopenia, but different algorithms have been proposed. For example, the EWGSOP suggests that muscle mass be expressed relative to height, while the FNIH recommends adjustment by BMI. Such disparities contribute to poor agreement in the literature between prevalence estimates for sarcopenia [6–8]. Furthermore, the EWGSOP1, EWGSOP2 and FNIH present different cut-points for identifying low muscle mass, strength and/or performance, which have been identified on the basis of different criteria [4,5,9,10], using data largely from European or American populations [3–5]. However, in the more recent EWGSOP2, reference data have been drawn from a range of populations [5], including Australia [11]. We have recently published prevalence estimates for sarcopenia using criteria recommended by the EWGSOP1, EWGSOP2 and FNIH [8], but there remains a lack of consensus about whether or not population-specific reference data should be used to identify low muscle mass and function [9,12]. The Australian and New Zealand Society for Sarcopenia and Frailty Research (ANZSSFR) recently recommended EWGSOP1 criteria but suggested employing population-specific cut-points [12].

The aim of this study was to calculate and compare prevalence estimates of sarcopenia in a sample of older women and men using the EWGSOP and FNIH ascertainment criteria but employing cut-points derived from the same population.

#### **2. Methods**

#### *2.1. Study Design*

The Geelong Osteoporosis Study (GOS) is a population-based, prospective study in Australia. Further detailed information about the GOS is published elsewhere [13]. Participants were randomly selected from the electoral roll for the Barwon Statistical Division until there were at least 100 women and 100 men in each 5-year age group from 20 to 69 years and 200 of each sex for age groups 70–79 years and ≥80 years [13]. Inclusion criterion was a listing on the electoral roll for the Barwon Statistical Division; participants were excluded if residency in the region was less than 6 months or if they were not able to provide written informed consent. At baseline (1993–1997), an age-stratified sample of 1494 women was enrolled, with 77% response; in 2005, this sample was supplemented with a further 246 women aged 20–29 years. Baseline data for 1540 men were collected during 2001–2006 (67% response). Participants were followed-up every few years. The study was approved by the Barwon Health Human Research Ethics Committee. Written informed consent was obtained from all participants.

#### *2.2. Participants*

Cross-sectional data from the 15-year assessment waves for women and men were used in this analysis. To determine prevalence estimates of sarcopenia in older adults, we included data from the 15-year assessment for 323 women (ages 60–95 years), collected during 2010–2014, and for 342 men (ages 60–96 years), collected during 2016–2019. The sample was almost entirely Caucasian (~98%).

#### *2.3. Measures*

Weight and height were measured to the nearest ±0.1 kg and ±0.1 cm and body mass index (BMI) calculated as weight/height<sup>2</sup> (kg/m<sup>2</sup> ). Appendicular lean mass (ALM) (kg) was obtained from whole-body dual-energy X-ray absorptiometry (DXA; Lunar Prodigy-Pro, Madison, WI, USA), which provided lean mass measures for the arms and legs. Short-term precision (calculated as the coefficient of variation on repeated whole body scans) was 0.9% for ALM. ALM was expressed relative to height<sup>2</sup> (ALM/height<sup>2</sup> , kg/m<sup>2</sup> ) or relative to BMI (ALM/BMI, m<sup>2</sup> ).

Handgrip strength (HGS) was measured using a hand-held analog dynamometer (Jamar, Sammons Preston, Bolingbrook, IL, USA) for women and a digital dynamometer (Vernier, LoggerPro3) for men. The testing procedure was demonstrated to participants before the measurement trials. With the participant seated in a comfortable position and the arm holding the dynamometer flexed at the elbow to 90 degrees, the participant was asked to squeeze the device as hard as possible for several seconds and the peak reading was recorded. This procedure was repeated for each hand. For women, the readings were performed in duplicate on each hand with no time interval between trials, and for men, trials were repeated in triplicate on each hand, holding the peak for 3 s with a 5-s interval between trials. The mean of the maximum value for each hand was used in further analyses. Measures from the Vernier device were transformed to Jamar equivalent values according to the following equation: *HGSJamar (kg) = 9.50 + 0.818\*HGSVernier (kg) + 8.80\*Sex*, where sex = 1 for men, which was developed by measuring the maximum HGS on each device for 45 adults aged 21–67 years [8].

The timed up-and-go (TUG) test was used as a measure of mobility but also includes static and dynamic balance [14]. This involved timing the participant (in seconds) to rise from a chair (without armrests), walk to a marked line (3 m distance), turn around, return to the chair and sit down. For men only, usual gait speed (GS, m/s) was also determined by measuring the time taken (in seconds) to walk a distance of 4 m. All measures were collected by trained personnel.

#### *2.4. Population-Specific Cut-Points*

Table 1 presents the Australian population-specific and international cut-points for the components of sarcopenia. Population-specific cut-points were determined as equivalent to 2 standard deviations (SDs) below sex-specific mean values for young reference groups (age ≤ 49 years) generated from the same population, as previously described [11,15–17]. For women, the cut-point for low HGS was <16 kg [16]. Using the same approach, the mean ±SD for HGS among 111 men (ages 33–49 years) was 44.8 ± 6.9 kg, and thus, the cut-point for low HGS was <31 kg. Low lean mass was identified as ALM/height<sup>2</sup> <5.30 kg/m<sup>2</sup> for women and 6.94 kg/m<sup>2</sup> for men [11], and low ALM/BMI as <0.512 m<sup>2</sup> for women and 0.827 m<sup>2</sup> for men [15], corresponding to T-scores < −2 [11,15].

**Table 1.** Applied threshold values for women and men used in different definitions.


ALM: appendicular lean mass; ALM/height<sup>2</sup> : appendicular lean mass/height<sup>2</sup> ; ALM/BMI: appendicular lean mass/body mass index; HGS: handgrip strength; TUG: timed up-and-go; GS: gait speed; EWGSOP: European Working Group on Sarcopenia in Older People; FNIH: Foundation for the National Institutes of Health; ROC: receiver operating characteristic.

We used a cut-point of <0.8 m/s for GS to identify slowness (poor muscle performance) in line with extant literature [2,6,18]. The mean ±SD for TUG among women was 6.98 ± 1.14 s, and thus, slow TUG was identified as >9.3 s. We also used TUG as a proxy for GS [2,14] for men, and since the cut-points for slow GS are the same for both sexes in the literature [3–5], we used the same threshold for TUG for both women and men.

Furthermore, as the FNIH cut-points for ALM/BMI were identified on the basis of discriminating clinically significant weakness [18], we estimated cut-points for low ALM/BMI that best discriminated the presence or absence of slow TUG (>9.3 s) [2,6,9,18]. The locations of optimal cut-points were determined by the principle that the sensitivity and specificity are closest to the value of the area under the receiver operating characteristic (ROC) curve, and the absolute value of the difference between the sensitivity and specificity is the smallest [19]. The ALM/BMI that best predicted slow TUG was <0.579 m<sup>2</sup> (sensitivity 0.63, specificity 0.60) for women and <0.913 m<sup>2</sup> (sensitivity 0.73, specificity 0.57) for men (Appendix A Figure A1). The area under the ROC curve was 0.64 (95% CI 0.58–0.70) for women and 0.68 (0.63–0.74) for men (*p* < 0.001).

#### *2.5. Sarcopenia Ascertainment*

Based on EWGSOP1 [3] and EWGSOP2 [5], sarcopenia corresponds to low ALM/height<sup>2</sup> and low HGS (ALM/height<sup>2</sup> + HGS); low ALM/height<sup>2</sup> and slow GS (ALM/height<sup>2</sup> + GS); or low ALM/height<sup>2</sup> and slow TUG (ALM/height<sup>2</sup> + TUG). According to FNIH [4], sarcopenia is defined as low ALM/BMI and low HGS (ALM/BMI + HGS) (Table 1). Furthermore, severe sarcopenia was determined using a combination involving low lean mass, muscle strength and physical performance, that is, ALM/height<sup>2</sup> + HGS + TUG for EWGSOP and ALM/BMI + HGS + TUG for FNIH.

#### *2.6. Statistical Analysis*

Data for women and men were analyzed separately. Histograms were used to check the distribution of data for normality. Means and SDs were presented for normally distributed data, and medians and interquartile ranges for skewed data. Prevalence for each age decade was calculated. Age-standardized prevalence estimates (mean and 95% confidence interval (CI)) were calculated according to 2011 census data from the Australian Bureau of Statistics [20]. Age-adjusted multivariable logistic regression models were developed to examine sex differences (pooled data) in the likelihood for sarcopenia. To compare prevalence estimates obtained with different cut-points, the kappa coefficient (κ) and 95% CIs were calculated and the strength of agreement was interpreted as small (κ < 0.40), medium (κ = 0.40–0.75) or high (κ > 0.75) [7]. Analyses were performed using SPSS (v24, IBM SPSS Statistics Inc., Chicago, IL, USA) and Minitab (v18, Minitab, State College, PA, USA).

#### **3. Results**

#### *3.1. Participant Characteristics*

Table 2 shows the participant characteristics. There were 12 women (3.7%) and 23 men (6.7%) with low ALM/height<sup>2</sup> , 70 women (21.7%) and 110 men (32.2%) with low ALM/BMI and 50 women (15.5%) and 87 men (25.6%) with low HGS. A slow TUG was recorded for 143 women (44.7%) and 169 men (49.7%) and a slow GS for 102 men (30.4%). Using the cut-point values obtained from ROC curves, 162 (50.2%) women and 197 (57.6%) men were identified as having low ALM/BMIROC.

**Table 2.** Participant characteristics. Data are presented as mean (±SD) or median (IQR).


BMI: body mass index; ALM: appendicular lean mass; HGS: handgrip strength; ALM/height<sup>2</sup> : appendicular lean mass/height<sup>2</sup> ; ALM/BMI: appendicular lean mass/body mass index; TUG: timed up-and-go. Missing data: HGS *n* = 1 man; TUG *n* = 3 women, 2 men; GS *n* = 323 women, 7 men.

There was a pattern of increasing prevalence of sarcopenia with advancing age in both sexes across all the definitions (Table 3). The point estimates for men were higher than for women, especially for those aged ≥80 yr; however, 95% CIs for different age groups overlapped.

#### *3.2. Sarcopenia Prevalence in Men Compared with Women*

After adjusting for age, and according to FNIH-related definitions, men were more likely than women to have sarcopenia; for ALM/BMI + HGS, odds ratio (OR) 2.45 (95%CI 1.32–4.56; *p* = 0.005) and for ALM/BMIROC + HGS, OR 2.27 (95%CI 1.39–3.72; *p* = 0.001). When EWGSOP-related definitions were used, men appeared to be more likely than women to have sarcopenia, but differences were not significant; for ALM/height<sup>2</sup> + HGS, OR 2.8 (95%CI 0.77–10.5; *p* = 0.11), and for ALM/height<sup>2</sup> + TUG, OR 1.5 (95%CI 0.6–3.7; *p* = 0.37).

#### *3.3. Age-Standardized Estimates of Sarcopenia*

The age-standardized estimates of sarcopenia according to different definitions and population-specific cut-points are shown in Table 3 and Figure 1. Using ALM/height<sup>2</sup> + low HGS, point estimates for sarcopenia prevalence were 0.9% for women and 2.9% for men. Using ALM/height<sup>2</sup> + TUG, estimates were 2.5% for women and 4.1% for men, and using ALM/height<sup>2</sup> + GS, the estimate for men was 1.6%. Using ALM/BMI + HGS, point estimates ranged from 5.5% to 10.4% for women and from 11.6% to 18.4% for men. The prevalence estimates based on population-specific cut-points are shown in Figure 1 together with estimates based on recommended international criteria. Prevalence estimates using international cut-points (shown in Figure 1) have been published elsewhere [8].

#### *3.4. Agreement*

Table 4 shows the levels of agreement between different definitions of sarcopenia using international and population-specific cut-points. Levels of agreement ranged from poor through to high (κ = 0.1–1 for women and 0–0.8 for men). Note that the 100% agreement for women using the FNIH definition occurred because the international and population-specific thresholds were the same, even though they were obtained using different methods.



 appendicular lean mass; GS: gait speed; HGS: handgrip strength; ALM/height2: appendicular lean mass/height2; ALM/BMI: appendicular lean mass/body mass index; TUG: timed up-and-go; ROC: receiveroperatingcharacteristics.Missingdata:HGS*n*1man;TUG*n*3women,2men;GS*n*323women,7men.

 =

 =

 =

**Figure 1.** Prevalence estimates of sarcopenia for (**A**) women and (**B**) men aged 60 years and older. Error bars show 95% confidence intervals. Bars for estimates using population-specific cut-points are unshaded and those using international cut-points are shaded. EWGSOP: European Working Group on Sarcopenia in Older People; FNIH: Foundation for the National Institutes of Health; ALM: appendicular lean mass; GS: gait speed; HGS: handgrip strength; BMI: body mass index; TUG: timed up-and-go.

κ


**Table 4.** Agreement between sarcopenia prevalence estimates according to different international and population-specific cut-points. Data are presented as κ, 95% confidence intervals and *p*-values.


indicated

 as ROC (derived from receiver operating

characteristic

 curves).

#### **4. Discussion**

We have reported sarcopenia prevalence in an Australian population using several cutpoints for EWGSOP and FNIH definitions. Using these cut-points, we obtained substantial differences in prevalence estimates for sarcopenia, and the level of agreement between definitions varied widely. Using population-specific cut-points equivalent to T-scores <−2, the FNIH definition produced the greatest prevalence, while EWGSOP provided the lowest. As the cut-point for low ALM/BMIROC that discriminated slow TUG was greater than ALM/BMI T-score <−2, the prevalence estimates for sarcopenia were correspondingly higher, and this was mainly a consequence of low ALM/BMIROC among the elderly. Regardless, there was a pattern of increasing sarcopenia prevalence with advancing age across all definitions.

The higher prevalence estimates for sarcopenia for older ages was also found in a study in the Netherlands, where diagnostic criteria for sarcopenia influenced prevalence estimates in a middle-aged cohort (mean age 61.8 years for *n* = 329 women and 64.5 years for *n* = 325 men). The authors reported the prevalence of sarcopenia ranged from 0% to 15.6%, 0% to 21.8% and 0% to 25.8% in women aged <60, 60–69 and ≥70 years, respectively, and from 0% to 20.8%, 0% to 31.2% and 0% to 45.2% in men aged <60, 60–69 and ≥70 years, respectively. These results indicate an age-related increase in sarcopenia for all definitions reflecting a decline in muscle mass and performance with age [6,21–23].

For both women and men, when applying population-specific cut-points, we observed that for each age-decade, prevalence estimates were lower for EWGSOP than FNIH. The age-standardized estimates were lower according to EWGSOP than FNIH for both women and men. Dam et al. (2014) [7] of the FNIH research group reported that 2.3% of women and 1.3% of men (proportions outside the 95% CIs of our estimates) in their pooled samples from the USA were classified as having sarcopenia using FNIH, while the prevalence was 13.3% for women and 5.3% for men using EWGSOP1 (point estimates outside our 95% CIs for women, but not for men). Similarly, an Australian study by Sim et al. [2] found that FNIH diagnosed fewer women with sarcopenia than EWGSOP (9.4% vs. 24.1%). In addition, Sim et al. [2] applied Australian female population-specific definitions for FNIH (defined as ALM/BMI < 0.517 m<sup>2</sup> + HGS < 17 kg) and EWGSOP (defined as ALM/height<sup>2</sup> < 5.28 kg/m<sup>2</sup> + HGS < 17 kg). However, the percentage was similar after harmonizing the cut-points.

Our results showed that overall, the agreement between FNIH and EWGSOP was poor, regardless of the cut-points employed. The poor agreement between the original EWGSOP (EWGSOP1) and FNIH definitions is well documented in a number of studies [6–8,24]. For example, Dam et al. [7] examined the difference between FNIH definitions and EWGSOP1. The agreement between the FNIH criteria (low HGS and low lean mass) and EWGSOP was poor in women (κ = 0.14) and medium in men (κ = 0.53). However, to our knowledge, this is the first study to examine the agreement between the EWGSOP and FNIH definitions after applying population-specific cut-points in an Australian setting. Masanes et al. [25] found that small differences in cut-points for low lean mass produced substantial variations in prevalence estimates for sarcopenia, and our findings are consistent with their results.

Although the cut-points recommended by EWGSOP [3,5] were adopted from different studies, the method for identifying deficits differed; while some identified low muscle mass and poor performance using the lower portion of the population distribution, the FNIH used a Classification and Regression Tree analysis [10] to identify clinically relevant criteria [4]. In our study, the population-specific cut-points were consistently identified using the lower portion of the population distribution, with the exception of ALM/BMI, where we also used ROC curves to identify low ALM/BMI values that corresponded to poor physical performance. There is still a need to reach a consensus as to which deficits in skeletal muscle health, and the extent of these deficits, are important in defining sarcopenia. Our results highlight disparities in prevalence estimates arising from the thresholds employed, suggesting that population-specific cut-points might be useful in certain populations.

Our study has both strengths and weaknesses. The participants were selected at random from the electoral roll and represent a broad adulthood age range. Almost the entire sample was Caucasian, and this might limit the generalization of our results to other ethnic groups in Australia and beyond. Whereas in this study, we used the mean of the maximum HGS for each hand as being indicative of strength, in some other studies, the maximum irrespective of handedness has been used. In recognition that the methods reported in the literature to identify maximum HGS vary, our choice of one method over another is a potential limitation. Prevalence data for sarcopenia in this study may have been influenced by differential participation and retention bias related to muscle health. Data were also lacking for participants who had physical impairments that prevented them from performance testing. Although data for women and men were pooled to identify sex differences, prevalence estimates for women and men have otherwise been analyzed separately as they were collected at different times.

In conclusion, this study takes a step towards a response to ANZSSFR's call to investigate evidence-based cut-points for EWGSOP criteria for the populations of Australian and New Zealand [12]. We have provided population-based data which will help clinicians and researchers in the field establish new operational definitions for identifying individuals with sarcopenia in the Australian population. However, it is yet to be decided which deficits in skeletal muscle health are important in identifying sarcopenia. Until a universally agreed operational definition of sarcopenia exists internationally and in Australia, prevalence data should be reported with consideration of the ascertainment criteria used and, thus, interpreted in context.

**Author Contributions:** All authors have taken responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: S.X.S. and J.A.P. Drafting of the manuscript: S.X.S. Acquisition, analysis, and interpretation of data: all authors. Critical revision of the manuscript for intellectual content: all authors. Statistical analysis: S.X.S. Supervision: K.L.H.-K., N.K.H., L.J.W. and J.A.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** The Geelong Osteoporosis Study was funded by the National Health and Medical Research Council (NHMRC), Australia (projects 251638 and 628582). S.X.S. was supported by a Deakin Postgraduate Scholarship in conjunction with the Geelong Medical and Health Benefits Association (GMHBA). K.L.H.-K. was supported by an Alfred Deakin Postdoctoral Research Fellowship, N.K.H. by a Dean's Research Postdoctoral Fellowship (Deakin University) and L.J.W. by an NHMRC Career Development Fellowship (1064272) and an NHMRC Investigator Grant (1174060). The funding organizations played no role in the design or conduct of the study, in the collection, management, analysis and interpretation of the data, nor in the preparation, review and approval of the manuscript.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Human Research Ethics Committee at Barwon HealthInformed Consent Statement: Written informed consent was obtained from all participants in the study.

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

**Acknowledgments:** The authors acknowledge the men and women who participated in the study, and the staff who contributed to the data collection.

**Conflicts of Interest:** All authors declare that no competing interests exist.

**Ethical Approval:** All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and national research committees and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Human Research Ethics Committee at Barwon Health.

#### **Abbreviations**


#### **Appendix A**

**Figure A1.** *Cont.*

**Figure A1.** ROC analysis for optimal discrimination of slow timed up-and-go (TUG; slow TUG < 9.3 s) for ALM/BMI ((**A**,**B**) for women and men, respectively). (**A**) ALM/BMI that best predicted slow TUG (Women); (**B**) ALM/BMI that best predicted slow TUG (Men).

#### **References**


### *Review* **E**ff**ect of Sleep Quality on the Prevalence of Sarcopenia in Older Adults: A Systematic Review with Meta-Analysis**

#### **Jacobo Á. Rubio-Arias 1,\* , Raquel Rodríguez-Fernández <sup>2</sup> , Luis Andreu 3,4 , Luis M. Martínez-Aranda 4,5, Alejandro Martínez-Rodriguez <sup>6</sup> and Domingo J. Ramos-Campo <sup>4</sup>**


Received: 8 November 2019; Accepted: 3 December 2019; Published: 6 December 2019

**Abstract:** Sarcopenia is an age-related condition. However, the prevalence of sarcopenia may increase due to a range of other factors, such as sleep quality/duration. Therefore, the aim of the study is to conduct a systematic review with meta-analysis to determine the prevalence of sarcopenia in older adults based on their self-reported sleep duration. Methods: Three electronic databases were used—PubMed-Medline, Web of Science, and Cochrane Library. We included studies that measured the prevalence of sarcopenia, divided according to sleep quality and excluded studies (a) involving populations with neuromuscular pathologies, (b) not showing prevalence values (cases/control) on sarcopenia, and (c) not including classificatory models to determine sleep quality. Results: high prevalence values in older adults with both long and short sleep duration were shown. However, prevalence values were higher in those with inadequate sleep (<6–8 h or low efficiency) (OR 0.76; 95% CI (0.70–0.83); Q = 1.446; *p* = 0.695; test for overall effect, Z = 6.01, *p* < 0.00001). Likewise, higher prevalence levels were shown in men (OR 1.61; 95% CI (0.82–3.16); Q = 11.80; *p* = 0.0189) compared to women (OR 0.77; 95% CI (0.29–2.03); Q = 21.35; *p* = 0.0003). Therefore, the prevalence of sarcopenia appears to be associated with sleep quality, with higher prevalence values in older adults who have inadequate sleep.

**Keywords:** muscle-mass; sleep efficiency; sleep duration; insomnia

#### **1. Introduction**

Together with the increment of the world population and life span over the years, a parallel increase in chronic diseases [1] has been observed, such as sarcopenia. This pathology has become a serious global public health problem [2] since it can lead to a considerable increase in costs due to the frequency and duration of hospitalization, as well as an increase in the number of falls as a consequence of muscle weakness [3,4]. In addition, people who suffer a high loss of muscle mass have an increased risk of other health problems, such as heart failure, chronic obstructive pulmonary diseases, kidney failure [5] or osteoporosis [6] and, therefore, a greater risk of bone fracture, turning sarcopenia into a major health problem that should be addressed in order to determine the possible factors associated with sarcopenia.

Sarcopenia has been defined as a decrease and deterioration of muscle mass associated with aging [7]. Thus, the skeletal muscle mass is progressively lost during aging and is partially replaced by fat and connective tissue due to a reduction and leakage of type II fibers generated by a slow degenerative neurological process [8]. This decrease in muscle mass due to aging also generates a decrease in muscle strength and, therefore, a physical disability generating a functional limitation (activities of the daily life) as well as a decrease in the life quality [9,10] with an associated increase in the risk of mortality [11,12]. In addition, this muscle mass loss has a greater impact on women during menopause as a consequence of the decrease in the estrogen levels after the fifth decade of life [13]. Sex differences in body composition are well known [14], with men having a higher cross-sectional area in skeletal muscle than women and greater muscle in the upper body [15]. Additionally, women are at higher risk of developing sarcopenic obesity due to increased fat and lower muscle mass [14]. Nevertheless, the results of prevalence related to sex are inconsistent [2]. In these circumstances, efforts are required to identify the factors associated with sarcopenia and to implement interventions for the prevention or the incidence reduction of this pathology among the elderly population [16], considering sex as a modifying variable.

However, the loss of muscle mass (sarcopenia) is not only related to age and sex but also depends on a number of endogenous and exogenous factors that influence the prevalence values of sarcopenia. The most studied and validated factors that can generate an effect on the sarcopenia are age (main moderating variable), genetic factors, birth weight, early growth, diet, physical activity, other chronic diseases, and hormonal changes (secondary variables) [17,18]. In line with this, a recent systematic review with meta-analysis on the general population [2] concludes that the prevalence of sarcopenia can be modified by other factors such as race, nutrition, quality of life, and sex among others.

Nonetheless, the scientific literature shows a gap between the role that sleep quality could play and the effects on the prevalence of sarcopenia. As Buchmann et al. (2016) [19] suggest, sleep is associated with a biological and mental regeneration process. Moreover, Vitale et al. (2019) [20] reported that the maintenance of circadian rhythms can be altered by aging and the development of many chronic diseases, including sarcopenia. The preservation of circadian rhythm is very important for the sustainment of cellular physiology, metabolism, and function in the skeletal muscle. Therefore, people who have an inadequate sleeping time could have an increased risk of mortality compared to those who sleep the recommended daily hours [21]. In addition, under low sleep conditions, the cognitive abilities might be affected and can be an increment in the risk of mortality and falling in older adults [22]. In this way, sex may also play a significant role in sleep quality, due to the fact that women have a greater predisposition of insomnia found among different criteria, frequencies, and duration [23]. Nevertheless, the association between muscle mass, sleep quality, and sex is not clear yet and no studies have been found to support such affirmation.

Certainly, the lack of sleep not only leads to a deterioration of cognitive abilities but can also have a negative effect at the cellular level on muscle physiology. It impairs muscle recovery due to increased stimulation of protein degradation, which is detrimental for protein synthesis and promotes muscle atrophy [24]. In addition to the negative effect on muscle mass, it has been associated with cardiovascular disease [25], type II diabetes [26], hypertension [25], obesity [27], and colorectal cancer [28]. In this regard, public health should include sleep duration/quality as one of the risk factors associated with a large number of diseases.

Some correlational studies have determined the effect of sleep duration on muscle mass, showing that less sleep duration or quality leads to a loss of muscle mass [29]. However, no meta-analyses addressing the effect of sleep duration or quality on the prevalence of sarcopenia have been found. Therefore, the objectives of this systematic review with meta-analysis are (1) to analyze the overall

prevalence of sarcopenia in people with optimal sleep duration/quality compared to those with inadequate sleep quality, (2) to analyze whether the prevalence of sarcopenia is correlated to the sex of the participants. Our starting hypothesis is that people with poor rest show a higher prevalence of sarcopenia than those who rest in better conditions and, in addition, men will have a lower prevalence compared to women.

#### **2. Experimental Section**

#### *2.1. Study Design*

A systematic review with meta-analysis was performed following the recommendations of PRISMA (preferred reporting items for systematic review and meta-analysis) [30]. All the analyses were performed in duplicate (J.A.R.A. and L.A.), all disagreements on inclusion/exclusion were discussed and resolved by consensus. The extrinsic characteristics of the publications and the substantive characteristics—population, sex, associated pathology, habits of alcohol, tobacco, physical activity, age, and BMI—were extracted from the studies that were finally included in the quantitative analysis. Finally, the methodological characteristics—duration of sleep, quality of sleep, muscular mass and presence or not of sarcopenia—were also considered. All subjects included in the analysis were classified as cases or control differentiating sleep and sex.

#### *2.2. Search and Data Sources*

Three electronic databases were used: PubMed-Medline, Web of Science, and Cochrane Library. The search was conducted without search date restriction and ended on 28 July 2019. The key search words and strategy were "Sleep Disorders" OR "Sleep Deprivation" OR "Sleep Hygiene" OR "Sleep duration" OR insomnia OR sleep\* and "muscle mass" OR "muscular atrophy" OR sarcopenia.

#### *2.3. Data extraction and Inclusion*/*Exclusion Criteria*

The following inclusion criteria were considered: prevalence studies analyzing the effect of sleep on sarcopenia and conducted in adults (>40). Studies were excluded if they included (a) populations with neuromuscular pathologies, (b) studies that did not show prevalence values (cases/control) on sarcopenia, and (c) studies that did not include classificatory models that allowed sleep quality to be determined.

#### *2.4. Outcomes*

The variables to determine the prevalence of sarcopenia as a function of sleep were (1) the presence or absence of sarcopenia, and (2) sleep quality. Sleep can be assessed to estimate adequate or inadequate sleep in terms of quality or duration in different ways. For the questionnaires, adequate-sleep (sleep well) for those who obtained between very good or quite good in the percentage of quality and not-adequate-sleep (sleep bad), rather bad or very bad were considered [19]. Regarding the hours of sleep, they were considered inadequate (<6–8) and adequate (≥8), following the recommendations of the National Sleep Foundation [21].

#### *2.5. Assessment of Risk of Bias*

The Q-index was used to assess the methodological quality, a scale that allows us to quantify the bias, obtaining a final score between 0 (minimum quality) and 1 (maximum quality). This rescaled quality range (called Qi in MetaXL) has a monotonic relationship to ICC bias, defined as the variance of the study bias divided by the sum of the variance of bias within and between studies [31,32]. Quality analysis was performed on each study based on the method of assessing sleep quality and sarcopenia, giving higher preference to the studies that measured the sleep with instruments previously validated for this purpose, as well as sarcopenia with DXA or BIA. Therefore, the studies using both DXA

and a validated questionnaire to analyze sleep quality were scored with 1. The following criteria were conducted:

(Q1) Were the target population and the observation period well defined?: yes = 1 and no = 0;

(Q2) Diagnostic criteria, use of diagnostic system reported: sarcopenia = DXA or BIA and sleep quality = instruments validated = 1 and own system/symptoms described/no system/not specified = 0;

(Q3) Method of case ascertainment: community survey/multiple institutions = 2, inpatient/ inpatients and outpatients/case registers = 1, and not specified = 0;

(Q4) Administration of measurement protocol: administered interview = 3, systematic case-note review = 2, chart diagnosis/case records = 1 and not specified = 0;

(Q5) Catchment area: broadly representative (national or multi-site survey) = 2, small area/not representative (single community, single university) = 1, and convenience sampling/other (primary care sample/treatment group) = 0; and

(Q6) Prevalence measure: point prevalence (e.g., one month) = 2, 12-month prevalence = 1 and lifetime prevalence = 0.

In addition, the overall publication bias of the studies was analyzed using the funnel plot, dividing between older adults who slept well and those who had inadequate sleep.

#### *2.6. Data Synthesis and Statistical Analysis*

Meta-analysis and statistical analysis were performed using MetaXL software version 2.0 (Sunrise Beach, Queensland, Australia). The prevalence of sarcopenia (cases vs. control) was initially calculated in the included studies for random-effects model analysis (no transformation methods) and then recalculated under a rescaled quality of bias effects model. For the analysis, the sleep category was considered (sleep well and sleep poorly) for the calculation of the overall prevalence of sarcopenia, and this method was applied under the random-effects model and effects in quality of the rescaled bias using three possible transformations (None, Logit, and Arcsine) [31,32] to contrast the effects of the prevalence of sarcopenia. In all cases, pooled prevalence values were shown, 95% CI, heterogeneity *I* 2 , Cochran's Q, chi<sup>2</sup> , *p*, tau<sup>2</sup> . On the other hand, the grouped odds ratios (OR) and their IC95% were also calculated following a model of "quality effects" [31,33] to analyze the association between those who sleep well (control) and those who sleep poorly (effect). In addition, OR and sleep category analysis were estimated, excluding studies that did not report participants' total hours of sleep and only showed sleep quality [19,34]. Heterogeneity between studies was conducted using the *I* 2 statistic, and the variation between studies was calculated using the tau<sup>2</sup> statistic (τ 2 ) [35]. *I* <sup>2</sup> values between 30–60% were considered as moderate levels of heterogeneity, while a value of τ <sup>2</sup> > 1 suggested the presence of substantial statistical heterogeneity. The minimum level of significance was set as *p* ≤ 0.05.

#### **3. Results**

#### *3.1. General Characteristics of the Studies*

A total of 551 items were identified from the selected databases, and 0/0 items were included from other sources. After the removal of duplicated articles from the different databases, 361 titles and abstracts, as well as 106 articles, were reviewed, and 255 were removed. Finally, statistical analysis was performed on a total of 6 studies [19,34,36–39] (5 were performed in Asia and only 1 in Europe; Figure 1), with a mean age of 68.7 years (range = 44–80 years). Table 1 shows the descriptive characteristics of the studies included in our analysis. The selected studies included 6405 (990 cases and 5415 controls) older adults with adequate sleep and 12,708 (1762 cases and 10,946 controls) adults with inadequate sleep. However, only four studies contained data divided by sex (1232 men) including 142 cases with adequate sleep and 95 cases with inadequate sleep. In addition, 1381 women were enrolled in the four studies, with 109 cases in the adequate sleep group and 118 cases in the inadequate sleep group.

**Figure 1.** Flow diagram of the process of study selection.


