Phase angle as a screening method for sarcopenia in community-dwelling older adults

Objective We investigated the utility of the phase angle as a screening tool for sarcopenia. We performed a cross-sectional study that included 169 active community-dwelling elderly women. The phase angle was determined using tetrapolar bioelectrical impedance, and sarcopenia was diagnosed based on skeletal muscle mass, muscle strength, and physical performance using bioelectrical impedance analysis, a handheld dynamometer, and the gait speed test, respectively. Receiver operating characteristic curve analysis was performed to investigate the role of the phase angle as a predictor of sarcopenia. The prevalence of sarcopenia was 12.4%. The median phase angle was 5.30°; elderly women with sarcopenia had lower phase angles than those without sarcopenia ( p =0.006). The phase angle cutoff for the detection of sarcopenia was ≤ 5.15°, with an area under the curve of 0.685, sensitivity 81.0%, specificity 60.8%, and accuracy 63.31%. Elderly women with a low phase angle show a high risk of presenting with reduced muscle mass. The phase angle was shown to be a useful screening tool in elderly women with sarcopenia.


A B S T R A C T Objective
We investigated the utility of the phase angle as a screening tool for sarcopenia.

Methods
We performed a cross-sectional study that included 169 active community-dwelling elderly women. The phase angle was determined using tetrapolar bioelectrical impedance, and sarcopenia was diagnosed based on skeletal muscle mass, muscle strength, and physical performance using bioelectrical impedance analysis, a handheld dynamometer, and https://doi.org/10.1590/1678-9865202235200243 2 LB SANTIAGO et al. the gait speed test, respectively. Receiver operating characteristic curve analysis was performed to investigate the role of the phase angle as a predictor of sarcopenia.

Results
The prevalence of sarcopenia was 12.4%. The median phase angle was 5.30°; elderly women with sarcopenia had lower phase angles than those without sarcopenia (p=0.006). The phase angle cutoff for the detection of sarcopenia was ≤5.15°, with an area under the curve of 0.685, sensitivity 81.0%, specificity 60.8%, and accuracy 63.31%. Elderly women with a low phase angle show a high risk of presenting with reduced muscle mass.

Conclusion
The phase angle was shown to be a useful screening tool in elderly women with sarcopenia.

I N T R O D U T I O N
Sarcopenia is a muscle disease that is associated with an increased risk of falls, fractures, and mortality, which compromise the quality of life in elderly individuals [1][2][3]. Prevention and early diagnosis enable prompt interventions, which can improve survival of elderly individuals affected by sarcopenia [4].
Sarcopenia is diagnosed based on tests that evaluate muscle strength, muscle mass, and physical performance [1]. Therefore, preserved functional capacity is important to enable appropriate testing for accurate diagnosis of sarcopenia in elderly individuals [5]. Furthermore, modalities and equipment including computed tomography, magnetic resonance imaging, and dual-energy X-ray absorptiometry that are necessary to measure muscle mass and a dynamometer to measure muscle strength may not be widely available in routine clinical practice [1].
Estimation of the phase angle is a linear method used to determine the association between reactance (affected by tissue hydration) and resistance (affected by cell membrane integrity) measured using Bioelectrical Impedance Analysis (BIA) [6]. The phase angle is known to be correlated with many health conditions. It is considered a predictor of negative outcomes and is associated with muscle mass and strength and is shown to be an indicator of sarcopenia in institutionalized and hospitalized elderly patients [5,[7][8][9]. However, its role in active community-dwelling elderly women remains unclear.
It is important to identify simple and cost-effective indicators of sarcopenia to facilitate early diagnosis and prompt intervention. Therefore, in this study, we investigated the utility of the phase angle as a screening tool for sarcopenia in active community-dwelling elderly women.

M E T H O D S
This cross-sectional study included elderly women recruited from the three Open Universities of the Third Age of Salvador, Bahia, Brazil, which represents 75% of the universities of the city.
Elderly women (≥60 years) who enrolled the universities and could undergo the necessary testing for diagnosis of sarcopenia were enrolled in the study. We excluded patients with implantation of cardiac pacemakers, aneurysm clips, or metallic devices of any type to avoid interference with the results of BIA.
With regard to sample calculation, the total number of elderly women who visited the universities was 408; the prevalence of sarcopenia was 15.4%, the sampling error was 5%, confidence level 95%, and we added 10% for possible losses to follow-up or refusal to participate in the study [10]. Therefore, the estimated minimum sample size was 149 elderly women. Owing to the small numbers (n=7) who enrolled these universities, we did not include men in the study.
The study was approved by the Research Ethics Committee of the Nutrition School of the Federal University of Bahia, assent 1.159.885/2015/12. All participants signed an informed consent form.
The phase angle was calculated following BIA using the Biodynamics ® tetrapolar device, model 450 (Biodynamics Corp. Seattle, WA, USA). We used the following formula proposed by Baumgartner, Chumlea, and Roche [11]: phase angle (degrees)=tangent arc of reactance/resistance×180/π. Bioelectrical impedance analysis was performed based on the protocol recommended by Kyle et al. [12].
Sarcopenia was diagnosed based on the algorithm proposed by the Sarcopenia: revised European consensus on definition and diagnosis, which measures muscle strength, Skeletal Muscle Mass (SMM), and physical performance [1].
Using a portable hand-held dynamometer (Sammons Preston Smedley) with a gradation scale from 0 to 100kg/strength (kg/f), we measured handgrip strength as an indicator of muscle strength. Two maximum attempts were made with each hand (dominant and non-dominant) to produce maximum voluntary handgrip strength, with a one-minute rest between them. The highest mean value was considered for analysis [13].
Physical performance was measured using the gait speed in m/s. This test involved walking a distance of 4 m over a flat and straight surface at the usual walking speed. The time required to cover the distance was measured [15]. diagnosed as reduced muscle strength and mass (handgrip strength ≤18.75kg/f and SMI ≤6.41kg/m 2 ), and severe sarcopenia was defined as a reduction in the three components (handgrip strength ≤18.75kg/f, SMI≤6.41kg/m 2 , and gait speed ≤0.77m/s).
The body mass index was calculated using the following equation: weight/height 2 and was analyzed as a continuous variable. The physical activity level was determined using a structured questionnaire; the variable used was categorically analyzed (yes or no).
Data normality was analyzed using the Kolmogorov-Smirnov test and homogeneity of variance using Levene's test. Descriptive analyses were presented as measures of central tendency, dispersion, absolute numbers, and frequencies.
The Kruskal-Wallis test was used to determine the various phase angle values across different categories of sarcopenia. The Mann-Whitney test was used for pair-wise comparison of categories in cases of a statistically significant difference.
A Receiver Operating Characteristic (ROC) curve was constructed using a positive diagnosis of sarcopenia as a reference; we estimated the sensitivity (sens), specificity (spec), and Area Under the Curve (AUC) with a confidence interval of 95%. The following diagnostic accuracy measures were also used to investigate the utility of the phase angle as a diagnostic tool for sarcopenia: negative and positive predictive values, the Youden index, and accuracy. The cutoff value that produced the best combination of sens and spec was selected as the most appropriate value of the phase angle as an indicator of sarcopenia.
Logistic regression models (crude and adjusted for age and physical activity) were used to confirm the association between muscle mass, muscle strength, physical performance, and sarcopenia with the phase angle cutoff point determined using the ROC curve.
All statistical analyses were performed using the SPSS software, Windows version 21.0 (SPSS, Inc., Chicago, IL, USA). A p-value<0.05 was considered statistically significant for all statistical analysis.

R E S U L T S
Our study included 169 active community-dwelling elderly women (median age 69 years). The prevalence of sarcopenia and severe sarcopenia was 12.4% and 6.5%, respectively. The median phase angle in this study was 5.30° (Table 1).
The phase angles were lower in elderly women with sarcopenia than in those without sarcopenia (p=0.006). Subcategory analysis showed that the differences remained statistically significant in the non-sarcopenia and severe sarcopenia groups (p=0.001) and between elderly women with probable and severe sarcopenia (p=0.001) ( Table 2).
The cutoff value that showed good diagnostic accuracy for sarcopenia in elderly women was ≤5.15° (sens 81.0%, spec 60.8%) based on the area under the ROC curve ( Figure 1). The cutoff point showed negative and positive predictive values and accuracy of 95.74%, 22.67%, and 63.31%, respectively. The Youden index was 0.418.  Logistic regression analysis after adjustment for age and physical activity showed that only muscle mass and sarcopenia were associated with a reduced phase angle (≤5.15°). We observed that the risk of reduced muscle mass was approximately three-fold higher and that of sarcopenia was approximately five-fold higher in elderly women with a reduced phase angle than in elderly women with a phase angle >the reference value for the detection of sarcopenia (Table 3).

D I S C U S S I O N
In this study, the best phase angle cutoff point that showed maximal diagnostic accuracy for sarcopenia was ≤5.15° (sens 84.2%, spec 60.4%). This value was higher than that reported by other authors. In a study that included hospitalized and community-dwelling elderly Turkish individuals, Kilic et al. [5] observed a phase angle cutoff value ≤4.55° (area under the ROC curve 0.703, sens 70%, spec 65.9%). Yamada et al. [16] reported a phase angle cutoff value ≤3.55° among community-dwelling elderly Japanese women. These variations may be attributable to differences in ethnicity, age, body mass characteristics, and health status of the populations investigated [6].
The phase angle decreases with aging [6,8] owing to changes in body composition (reduced muscle mass and increased adipose tissue), which occur secondary to changes in body water distribution between tissues, which decreases reactance and increases resistance [17]. Therefore, a reduced phase angle, which indicates reduction in muscle mass and strength (which characterize sarcopenia), serves as a good marker of sarcopenia.
Sarcopenia is diagnosed based on tests that require preserved physical capacity. Therefore, the phase angle can be considered a useful tool to diagnose sarcopenia in elderly patients, particularly in those who cannot perform the tests necessary to evaluate handgrip strength or walking ability. In addition to its role as a determinant of sarcopenia, phase angle measurement serves as a noninvasive and inexpensive BIA-derived method that can be used in clinical practice and epidemiological studies [5,18].
Our study showed that the phase angle was lower in elderly women with sarcopenia than in those without sarcopenia, which is consistent with findings reported by other studies [5,9,16]. This observation may be attributable to the association between the phase angle and the parameters that define sarcopenia, specifically the muscle mass, which is positively correlated with the phase angle [5,9,16,19].
Randomized studies that included elderly women who underwent physical training have shown that resistance training performed for muscle development increases the phase angle; this observation reaffirms the correlation between muscle mass and the phase angle [20][21][22]. In a study performed by Ribeiro et al. [21], physically independent elderly women underwent two types of resistance training; the authors observed an increase in the muscle mass (20.0-20.4kg and 20.2-20.5kg, p<0.05) and phase angles (5.60-5.76° and 5.41-5.63°, p<0.05) after 8-week training. Cunha et al. [22] observed an increased phase angle (5.87-6.12° and 5.50-5.90°, respectively, p<0.05) and improved muscle mass quality, following 12-week resistance training in elderly women who underwent resistance training compared with elderly women who did not receive this training.
Although we did not observe an association between the phase angle and muscle strength and physical performance in our study, other studies have reported a correlation between the phase angle and these aforementioned parameters of sarcopenia [5,9,16,[23][24][25][26].
The convenience nature of the sample and the fact that it is not representative of the Brazilian population are the limitations of our study. Furthermore, BIA is a relatively expensive method, and the results are affected by an individual's hydration status and also pacemaker implantation. Following are the strengths of our study: (a) We used reference values from our study sample itself to diagnose sarcopenia and, (b) we confirmed the utility of the phase angle as a relatively simple and inexpensive screening tool for the diagnosis of sarcopenia.

C O N C L U S I O N
This study highlights the role of the phase angle as a good predictor of sarcopenia in the population described in this report. The association observed between the phase angle and skeletal muscle mass and sarcopenia strengthens the recommendation for its application as a simple screening tool for diagnosis of sarcopenia to enable early intervention and avoid complications that compromise functional capacity and the quality of life in an aging population.

A C K N O W L E D G E M E N T S
The authors thank the study participants and the research team.

C O N T R I B U T O R S
LB SANTIAGO and LB RAMOS participated in all activities/stages of this research. CC OLIVEIRA contributed to data analysis and interpretation. AKC RORIZ and TM OLIVEIRA contributed to the conception and design of the study, and MEPC MACHADO was actively involved with data interpretation. All authors participated in the review and approval of the final version of the manuscript.