Prevalence of sarcopenia defined using the Asia Working Group for Sarcopenia criteria in Japanese community-dwelling older adults: A systematic review and meta-analysis

Abstract

The aim of this systematic review and meta-analyses was to calculate the pooled prevalence of sarcopenia based on the Asia Working Group for Sarcopenia (AWGS) criteria among Japanese community-dwelling older adults. Data from 8 studies were used to determine the prevalence of sarcopenia in the overall population and in men; data from 9 studies were used to determine that of women. The pooled prevalence rates of sarcopenia using random-effects models were 9.9% (95% confidence interval [CI], 6.2%-15.4%) overall; 9.8% (95% CI, 6.2%-15.2%) among men; and 10.1% (95% CI, 6.4%-15.5%) among women. These findings would be useful to inform community-based strategies and advanced research addressing sarcopenia prevention.

The term “sarcopenia” was coined by Rosenberg in 1989¹⁾ to draw attention to the age-related loss of muscle mass. Sarcopenia increases the risk of negative health outcomes, such as falls, fractures, dependency, use of hospital services, institutionalization, poor quality of life, and mortality²⁾.

Several different clinical diagnostic criteria for sarcopenia have been reported worldwide^3-8). Among these, the 2010 European Working Group on Sarcopenia in Older People (EWGSOP)³⁾ consensus diagnostic criteria have been most widely adopted and used in many cohort studies involving community-dwelling older adults. Based on the EWGSOP criteria, slow walking speed (≤0.8 m/sec) or low grip strength (<30 kg for men and <20 kg for women) are indications for muscle mass measurement; sarcopenia is diagnosed if muscle mass is reduced based on sex-specific criteria.

In Japan, the most widely utilized criteria for determining sarcopenia are based on the Asia Working Group for Sarcopenia (AWGS) consensus, which was published in 2014⁴⁾. The AWGS criteria are consistent with those of the EWGSOP; however, the cut-off values for muscle strength (<26 kg for men and <18 kg for women), walking speed (≤0.8 m/sec), and muscle mass (7.0 kg/m² for men and 5.4 kg/m² for women measured using dual X-ray absorptiometry and 7.0 kg/m² for men and 5.7 kg/m² for women measured using bioimpedance analysis) have been revised according to data from regional cohort studies among Asian populations.

In a recent review article, the estimated prevalence of sarcopenia based on the AWGS criteria ranged between 4.1% and 11.5% among the general older population⁹⁾. However, few Japanese cohort studies were included in that review, and a meta-analysis was not performed. In addition, the prevalence of sarcopenia among Japanese community-dwelling older adults was not evaluated.

Therefore, this systematic review and meta-analyses aimed to calculate the pooled prevalence of sarcopenia based on the AWGS criteria among Japanese community-dwelling older adults.

Methods

Protocol

The AWGS published their consensus report for recommended diagnostic algorithm in Asian population. We reviewed the literature on sarcopenia published by researchers from Japan, including articles represented by the AWGS, from February 2014 to October 2018. We searched PubMed (United States National Library of Medicine, National Institutes of Health) database records on October 31, 2018 using the following search terms: “sarcopenia” AND “prevalence” AND “Japan.” Inclusion criteria were studies that had enrolled community-dwelling participants aged ≥60 years among the general population and those that were published in English or Japanese. The titles and abstracts of all retrieved records were screened. Studies were excluded if they were conference abstracts or commentaries, diagnosed sarcopenia using criteria other than those of the AWGS, or included patients with confirmed illnesses (e.g., Alzheimer's disease and diabetes mellitus). The remaining full-text articles were retrieved and reviewed. Those that reported prevalence data for sarcopenia based on AWGS criteria among community-dwelling Japanese older adults were retained for analyses.

Statistical analyses

The Cochran's Q test was used to assess the presence of heterogeneity across studies, which was indicated by p < 0.05. I² statistic values of 25%, 50%, and 75% indicated low, moderate, and high degrees of heterogeneity, respectively¹⁰⁾. Pooled prevalence and 95% confidence intervals (CI) of sarcopenia were calculated using a random-effects model if heterogeneity was present and a fixed-effects model if heterogeneity was absent¹¹⁾. Prevalence values were calculated among the overall, male, and female participants. Statistical analyses were completed using Comprehensive Meta-Analysis (Version 3; Biostat, Englewood, NJ, USA) software.

Results

Literature review results

The literature selection process is shown in Figure 1. From among 101 potential citations identified through PubMed and 4 additional citations found through other sources, 72 were excluded on the basis of title and abstract screening. Full-text review of the remaining 33 citations was conducted, among which 24 were excluded as follows: studies that included patients aged <60 years (n = 3); or patients with Alzheimer's disease (AD), diabetes mellitus (DM), or other illnesses (n = 11); studies that used criteria other than those of the AWGS (n = 5); studies that did not define sarcopenia (n = 1); reviews and other article types (n = 3); and duplicates (n = 1). The remaining 9 studies were included in the meta-analysis^12-21), among which 1 included only older women. The total number of participants was 7,974 (3,723 men and 4,367 women).

Figure 1.

Systematic literature review protocol

Footnote: AD, Alzheimer's Disease; AWGS, Asian Working Group for Sarcopenia; DM, diabetes mellitus.

Prevalence of sarcopenia

Data from 8 studies were synthesized to determine the prevalence of sarcopenia in the overall and male participants, and data from 9 studies were synthesized to determine that in female participants. The prevalence of sarcopenia in the overall, male, and female participants based on individual studies ranged from 4.7% to 25.7%, 4.9% to 25.0%, and 4.5% to 26.1%, respectively. Among all 3 groups, high degrees of heterogeneity were found across studies (I² = 92.6% to 97.0%, p < 0.01); therefore, random-effects models were used. The pooled prevalence of sarcopenia based on AWGS criteria in Japanese community-dwelling older adults was 9.9% (95% CI, 6.2% to 15.4%) overall, 9.8% (95% CI, 6.2% to 15.2%) among men, and 10.1% (95% CI, 6.4% to 15.5%) among women (Figure 2, 3).

Figure 2.

Forest plots of the prevalence of sarcopenia among the overall participants (n = 7,974)

Footnote: CI, confidence interval.

Figure 3.

Forest plots of the prevalence of sarcopenia stratified by sex

Footnote: A: men, B: women. CI, confidence interval.

Discussion

This systematic review and meta-analysis identified 9 studies including 7,974 Japanese community-dwelling older adults, among which 8 were used to calculate the overall prevalence of sarcopenia. The pooled prevalence of sarcopenia based on the AWGS criteria among Japanese community-dwelling older adults was 9.9% overall, 9.8% among men, and 10.1% among women.

The prevalence of sarcopenia using the EWGSOP definition in a previous systematic review ranged from 1% to 29% in community-dwelling populations, 14% to 33% in long-term care populations, and 10% in an acute hospital-care population, with regional and age-related variations²²⁾. In addition, a high prevalence of sarcopenia was reported in nursing home residents (more than 40%) based on EWGSOP criteria²³⁾. The estimated prevalences of sarcopenia based on EWGSOP, AWGS, or the International Working Group on Sarcopenia (IWGS) among Asian individuals was 10% in men and 11% in women²⁴⁾, which are similar to the findings of this study. A previous study including Japanese community-dwelling older adults demonstrated a higher prevalence of sarcopenia in women than men among the young old, whereas prevalence was lower in women than men among those aged above 85 years²⁵⁾. Although additional analyses are required, such as those stratified by age, the pooled prevalence rate among Japanese community-dwelling older adults in this meta-analysis are acceptable by comparison with previous results from studies of populations in Europe and Asia^9,22,24).

Recently, a corrected definition of sarcopenia was proposed by EWGSOP2²⁶⁾, which included recommendations aimed to increase awareness of sarcopenia and its risk at earlier ages. Muscle quantity and quality, and low muscle strength, were emphasized as key characteristics, and the clinical algorithm that can be used for case-finding, diagnosis and confirmation, and severity determination of sarcopenia was updated. EWGSOP2 recommended that poor physical performance with clear cut-off points was a useful indicator of severe sarcopenia²⁶⁾. Measures of physical functioning, such as walking speed and the timed up and go test²⁷⁾, assessments of mobility, such as the 400 m walking test²⁸⁾, and comprehensive measures of physical function, such as the Short Physical Performance Battery^29,30), which includes walking speed, chair standing, and standing balance, are recommended. Considering the updated EWGSOP2 recommendations, the Asian criteria for defining sarcopenia could presumably be updated soon. Future studies that analyze the pooled prevalence of sarcopenia based on updated definitions and recommendations among community-dwelling older adults and other populations, such as those in long-term care and hospital-care, will be needed.

The results of this meta-analysis indicated high heterogeneity, which may be related to various factors, such as differences in age³¹⁾, living environment³²⁾, and muscle mass measurement methods²⁴⁾ among included studies. Studies included in this meta-analysis reported different muscle mass measurement methods (dual X-ray absorptiometry and bioimpedance analysis). In addition, differences in the prevalence rates of sarcopenia have been observed in rural (13.1%) and urban (7.0%) areas³²⁾; however, this variable was not investigated in the present study.

The findings of this meta-analysis should be interpreted with caution. The pooled prevalence of sarcopenia among community-dwelling older adults calculated in this study may be underestimated. Data from several cohort studies were analyzed, which included participants with access to community-setting such as community centers, public halls, and hospital outpatient units. In addition, there were high degrees of heterogeneity among the included studies. Although the participant recruitment processes differed between each cohort study, most included participants had high levels of physical function and could live independently in the community. Therefore, community-dwelling older adults at greater risk of sarcopenia, such as those who require care or support, might not be represented in our findings, contributing to an underestimated prevalence.

In summary, this review and meta-analysis found a pooled prevalence of sarcopenia based on AWGS criteria among Japanese community-dwelling older people of 9.9%. Similar prevalence rates in older men (9.8%) and women (10.1%) were found. These findings could be used to inform community-based strategies and advance research addressing sarcopenia prevention.

Conflict of Interest

There is no conflict of interest to disclose.