The present study involved the development of simple prediction models using traditional demographic data and anthropometric measures as a clinical tool for the quantification of skeletal muscle mass (SMM). Although several anthropometric prediction models for the estimation of SMM have been proposed in previous studies, the present investigation is a pioneering study by considering age groups in the mathematical equations [9, 20–22].
Incorporating anthropometric variables that are easily measured in clinical practice to compose a mathematical estimation of SMM using CT for comparative purposes could increase the predictive precision compared to anthropometric measures considered separately for the estimation of muscle status in patients. The SMM prediction models for adults and older people included three anthropometric variables: weight, height and calf circumference. The measures are easily obtained in public and private hospitals, requiring only a scale, stadiometer and metric tape.
Age group is an important variable in the assessment of body composition, considering the physiological changes that accompany the aging process, such as a reduction in muscle mass and increase in total adipose tissue. Thus, prediction models for SMM stratifying the sample into adults and older people can provide equations capable of predicting variations in muscle mass that occur with the increase in age [23].
The equations developed for adults and older people achieved a satisfactory degree of prediction. The relative proportion of variation (R2 value) explained by the prediction equations was higher for adults (0.73) than older people (0.67). This does not necessarily mean that the prediction equation was less precise for older people. It may simply reflect a lower relative quantity of variation that could be explained in this group. This difference between age groups likely reflects the lower muscle mass in older people and greater variability in other tissues, especially fat mass [20].
Calf circumference is a simple, effective muscle mass assessment tool that is easily implemented in the hospital setting, as the measurement that can be performed anywhere and by any person [9]. The present study demonstrated that calf circumference was strongly correlated with SMM measured by CT in adults and older people, as CC was included in the mathematical equations derived from stepwise regression in both age groups.
To the best of our knowledge, only three previous studies reported the development of a simple anthropometric equation incorporating CC to estimate SMM. However, the reference measure was DEXA and the population was composed of healthy individuals in these studies [1, 9, 24]. In the study conducted by Hwang et al. (2017), who analyzed 1,839 Taiwanese individuals < 50 years of age, the equation developed for the estimate of SMM using DEXA as the gold standard incorporated age, sex, height, weight and CC, achieving an adjusted R2 of 0.86 [24]. In the study conducted by Santos et al. (2019) for the National Health and Nutrition Examination Survey involving a sample of 15,293 individuals > 18 years of age, sex, age, ethnicity and CC were included as predictive variables and the adjusted R2 was 0.88 [1]. Kawakami et al. (2021) recruited 1262 participants > 40 years of age using age, sex, height, weight, waist circumference and CC as prediction variables. The equation had an adjusted R2 of 0.94 and the authors stated that the model had potential as a reliable, effective method for estimating SMM [9].
The predictive capacity of the equations developed in the present investigation was lower than that in the studies cited above. This may be explained by the fact that our study population was composed of hospitalized patients with diseases or catabolic conditions that affect body composition, leading to greater variability in the adipose tissue and muscle compartments. A recently published systematic review conducted to investigate literature on the prediction of muscle mass using anthropometric equations pointed out the scarcity of precise equations validated in unhealthy groups and stressed the importance of equations especially for obese individuals, malnourished individuals and older people [25].
Due to the considerable differences among the populations analyzed and methods employed, valid comparisons of published results are difficult. Besides differences in populations, most studies employed bioelectrical impedance analysis (BIA) or DEXA to estimate SMM [1, 26]. BIA is less expensive and more practical than other methods, such as CT, by has limitations and its use in clinical practice may be restricted due to different factors related to the patient. Moreover, the precision of equations to estimate SMM using BIA as the gold standard is specific to the device and the population [27, 28].
The present study has limitations that should be considered. The lower predictive power compared to equations described in the literature suggests that factors other than anthropometric measures contribute to the variation in SMM in hospitalized adults and older people. Another limitation regards the fact that the participants were recruited from a single reference center and may not be representative of all hospitalized patients.
This is the first study to develop and validate skeletal muscle mass prediction equations for hospitalized adults and older people based on anthropometric measures and using computed tomography as reference. The use of these equations could be an alternative strategy for situations in which imaging exams are not available and could constitute a screening tool for individuals at risk of sarcopenia. The equations presented here should be tested for clinical purposes and in the analysis of data from mixed hospitalized populations.