Nagai K1, Wada Y2, Tamaki K3, Kusunoki H3, Tsuji S4, Ito M5, Sano K5, Amano M6, Shimomura S7, Shinmura K3
1Hyogo University of Health Sciences, Department of Physical Therapy, School of Rehabilitation, Kobe, Japan, 2Hyogo College of Medicine Sasayama Medical Center, Department of Rehabilitation, Sasayama, Japan, 3Hyogo College of Medicine, Department of General Medicine, Nishinomiya, Japan, 4Hyogo College of Medicine, Department of Orthopaedic Surgery, Nishinomiya, Japan, 5Hyogo University of Health Sciences, Department of Occupational Therapy, School of Rehabilitation, Kobe, Japan, 6Hyogo University of Health Sciences, School of Pharmacy, Kobe, Japan, 7Hyogo College of Medicine Sasayama Medical Center, Department of General Medicine and Community Health Science, Sasayama, Japan
Background: Sarcopenia refers to a syndrome characterized by serious loss of skeletal muscle mass and strength. Appendicular skeletal muscle mass, which is a sum of the muscle mass of both arms and legs, is generally used to calculate the skeletal muscle mass index (SMI) as an assessment parameter for sarcopenia. When evaluating muscle mass, SMI has been estimated after adjusting for body size by dividing height squared. However, because this method positively correlates with body mass index (BMI), applying this equation for subjects with a greater BMI may be inappropriate as an assessment parameter for sarcopenia.
Purpose: To compare two types of SMI adjusted by height squared and by BMI in different body types according to BMI in older adults.
Methods: The present study included 515 community-dwelling older women aged 72.6 ± 5.7 years in Japan. To evaluate muscle mass, the participants were enrolled to undergo bioelectrical impedance analysis using InBody770. We used two types of SMI defined as muscle mass divided by height squared (hSMI) or by BMI (bSMI). The participants were classified according to their BMI as follows: underweight (group I, 18.5), lower normal (group II, 18.5-21.9), higher normal (group III, 22-24.9), and overweight (group IV, >25). We investigated the correlation between each SMI and physical function such as gait speed and muscle strength. Then, receiver-operating characteristic (ROC) analysis was applied to evaluate the ability to detect low muscle strength or impaired performance according to the assessment for sarcopenia in each SMI adjustment.
Results: The coefficient correlations between SMI and normal gait speed were as follows: group I (n = 42), 0.23/0.11 (hSMI/bSMI); group II (n = 197), 0.16/0.24; group III (n = 171), 0.18/0.30; and group IV (n = 105), 0.22/0.43. The coefficient correlations with knee extensor strength were as follows: group I, 0.52/0.44 (hSMI/bSMI); group II, 0.36/0.46; group III, 0.37/0.48; and group IV, 0.48/0.55. Correlations between SMI and physical function tended to be higher when using bSMI than hSMI except for the underweight group. The increase in area under the curve (AUC) by the ROC analysis was highest in group IV (group I: 0.84/0.89 [hSMI/bSMI], group II: 0.80/0.81, group III: 0.66/0.69, and group IV: 0.59/0.75).
Conclusion(s): SMI adjusted by height squared is applicable only for older women who are underweight. Applying SMI adjusted by BMI would be more reasonable in older women with a BMI of ≥18.5 kg/m2.
Implications: SMI adjusted by height squared is commonly used in clinical settings for assessing sarcopenia. However, applying this index for all older adults with different body types should be reconsidered.
Keywords: sarcopenia, skeletal muscle index
Funding acknowledgements: This work was supported by JSPS KAKENHI [Grant number: JP16KT0012] and Mitsui Life Social Welfare Foundation.
Purpose: To compare two types of SMI adjusted by height squared and by BMI in different body types according to BMI in older adults.
Methods: The present study included 515 community-dwelling older women aged 72.6 ± 5.7 years in Japan. To evaluate muscle mass, the participants were enrolled to undergo bioelectrical impedance analysis using InBody770. We used two types of SMI defined as muscle mass divided by height squared (hSMI) or by BMI (bSMI). The participants were classified according to their BMI as follows: underweight (group I, 18.5), lower normal (group II, 18.5-21.9), higher normal (group III, 22-24.9), and overweight (group IV, >25). We investigated the correlation between each SMI and physical function such as gait speed and muscle strength. Then, receiver-operating characteristic (ROC) analysis was applied to evaluate the ability to detect low muscle strength or impaired performance according to the assessment for sarcopenia in each SMI adjustment.
Results: The coefficient correlations between SMI and normal gait speed were as follows: group I (n = 42), 0.23/0.11 (hSMI/bSMI); group II (n = 197), 0.16/0.24; group III (n = 171), 0.18/0.30; and group IV (n = 105), 0.22/0.43. The coefficient correlations with knee extensor strength were as follows: group I, 0.52/0.44 (hSMI/bSMI); group II, 0.36/0.46; group III, 0.37/0.48; and group IV, 0.48/0.55. Correlations between SMI and physical function tended to be higher when using bSMI than hSMI except for the underweight group. The increase in area under the curve (AUC) by the ROC analysis was highest in group IV (group I: 0.84/0.89 [hSMI/bSMI], group II: 0.80/0.81, group III: 0.66/0.69, and group IV: 0.59/0.75).
Conclusion(s): SMI adjusted by height squared is applicable only for older women who are underweight. Applying SMI adjusted by BMI would be more reasonable in older women with a BMI of ≥18.5 kg/m2.
Implications: SMI adjusted by height squared is commonly used in clinical settings for assessing sarcopenia. However, applying this index for all older adults with different body types should be reconsidered.
Keywords: sarcopenia, skeletal muscle index
Funding acknowledgements: This work was supported by JSPS KAKENHI [Grant number: JP16KT0012] and Mitsui Life Social Welfare Foundation.
Topic: Older people; Health promotion & wellbeing/healthy ageing; Musculoskeletal
Ethics approval required: Yes
Institution: Hyogo College of Medicine
Ethics committee: Ethics review board at Hyogo College of Medicine
Ethics number: 201705-095
All authors, affiliations and abstracts have been published as submitted.
