Iasmin Matias Sousa1, Maria Cristina Gonzalez2, Renata Moraes Bielemann3, Ilanna Marques Gomes Rocha4, Erica Roberta Barbalho5, Ana Lúcia Miranda Carvalho1, Galtieri Otávio Cunha Medeiros6, Flávia Moraes Silva7, Ana Paula Trussardi Fayh8. 1. Postgraduate Program in Nutrition, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil; Postgraduate Program in Health Sciences, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil. 2. Postgraduate Program in Health and Behavior, Catholic University of Pelotas, Pelotas, RS, Brazil. 3. Postgraduate Program in Nutrition and Food, Federal University of Pelotas, Pelotas, RS, Brazil. 4. Postgraduate Program in Nutrition, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil. 5. Postgraduate Program in Physical Education, Federal University of Rio Grande do Norte, Natal, RN, Brazil. 6. Postgraduate Program in Health Sciences, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil. 7. Department of Nutrition, Federal University of Health Sciences of Porto Alegre, Porto Alegre, RS, Brazil. 8. Postgraduate Program in Nutrition, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil; Postgraduate Program in Health Sciences, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, RN, Brazil; Postgraduate Program in Physical Education, Federal University of Rio Grande do Norte, Natal, RN, Brazil. Electronic address: ana.fayh@ufrn.br.
Abstract
BACKGROUND & AIMS: Cancer influences body composition, including a loss of muscle mass (MM), associated with worse outcomes. The study aimed to evaluate the agreement between MM estimated by calf circumference (CC) and computed tomography (CT) image as a reference method. METHODS: A cross-sectional study including patients (>20 years) diagnosed with cancer attending a reference center of oncology. Spearman's correlation was performed to verify the correlation between CC and MM by CT, including skeletal muscle area - SMA and skeletal muscle index - SMI. ROC curves, Kappa coefficient, sensitivity, specificity, positive and negative predictive values were obtained. RESULTS: The study included 219 patients, age 62.9 ± 13.1 years (mean ± standard deviation). Low CC was observed in 43.8% of the patients, and 29.2% had low SMI. CC positively correlated with SMA (rho = 0.333) and SMI (rho = 0.329), and fair agreements (K = 0.268) were observed between CC and SMI, with higher and significant values for males (K = 0.332) and patients below 60 years (K = 0.419). The area under the curve (AUC) for low CC to identifying low SMI was equal to 0.685 (CI 95% 0.606-0.765). Low CC presented fair agreement to identify low SMI in the sample; however, the negative predictive value was almost 80% for all analyses. CONCLUSIONS: Low CC is not a surrogate for low SMI in patients with cancer, but it could be an alternative, non-invasive, easy-to-perform method to pre-screen patients with cancer with adequate SMI.
BACKGROUND & AIMS: Cancer influences body composition, including a loss of muscle mass (MM), associated with worse outcomes. The study aimed to evaluate the agreement between MM estimated by calf circumference (CC) and computed tomography (CT) image as a reference method. METHODS: A cross-sectional study including patients (>20 years) diagnosed with cancer attending a reference center of oncology. Spearman's correlation was performed to verify the correlation between CC and MM by CT, including skeletal muscle area - SMA and skeletal muscle index - SMI. ROC curves, Kappa coefficient, sensitivity, specificity, positive and negative predictive values were obtained. RESULTS: The study included 219 patients, age 62.9 ± 13.1 years (mean ± standard deviation). Low CC was observed in 43.8% of the patients, and 29.2% had low SMI. CC positively correlated with SMA (rho = 0.333) and SMI (rho = 0.329), and fair agreements (K = 0.268) were observed between CC and SMI, with higher and significant values for males (K = 0.332) and patients below 60 years (K = 0.419). The area under the curve (AUC) for low CC to identifying low SMI was equal to 0.685 (CI 95% 0.606-0.765). Low CC presented fair agreement to identify low SMI in the sample; however, the negative predictive value was almost 80% for all analyses. CONCLUSIONS: Low CC is not a surrogate for low SMI in patients with cancer, but it could be an alternative, non-invasive, easy-to-perform method to pre-screen patients with cancer with adequate SMI.
Authors: Jenelle Loeliger; Lara Edbrooke; Robin M Daly; Jane Stewart; Lucy Bucci; Carmen Puskas; Marnie Fitzgerald; Brenton J Baguley; Nicole Kiss Journal: Int J Environ Res Public Health Date: 2022-03-29 Impact factor: 3.390