Massa muscular esquelética de pacientes com câncer: comparação entre métodos de avaliação e associação com mortalidade

Abstract

Cancer impacts body composition, and loss of muscle mass (MM) is associated to worse outcomes. Therefore, it is important to investigate simple, easy and cost-effective methods of MM assessment to early identify individuals with low MM. The objective of the study was to evaluate if calf circumference (CC) may be used as a marker of MM considering skeletal muscle index (SMI) from computed tomography (CT) as a gold standard and the prognostic value of CC and SMI to predict mortality in patients with cancer. A single center prospective study was conducted with adult and elderly patients attending a reference center of oncology and had recent abdominal CT images available. Data was collected through a semistructured form and patients’ records, and included socio-demographic data (sex, age and ethnicity); clinical data (primary site and staging of tumor and treatments performed); anthropometric variables (body mass index – BMI and CC) and outcome (death). Muscle strength was accessed using handgrip strength (dynamometry). MM was assessed by CT images at the level of the third lumbar vertebra, muscle quantity was analyzed by SMI (total muscular area normalized for height) and muscle quality by muscle attenuation (MA). Spearman’s correlation, sensitivity, specificity, predictive positive and negative value and area under receiver operating characteristics (ROC) curve were analyzed. Cox proportional hazard model adjusted for age, sex and staging of disease was used to evaluate prognostic value of the MM evaluation methods to predict mortality. A total of 250 patients (52.8% female) were evaluated, with a median age of 63 years (IQR: 55-73). Colorectal cancer was the most prevalent, followed by gastric cancer. Stages III and IV together represented 64% of the patients and 48% performed combined treatment. Almost half were classified as normal BMI (44.4%), 46.4% had low CC; low SMI was observed in 29.2% of the patients, 33.6% had low MA and low muscle strength (dynapenia) was identified in 41.6% of the sample. CC was positively correlated with MM and SMI obtained by CT images, with highest correlation obtained between CC and MM not adjusted for height both in men (r = 0.667, p<0.001) and women (r = 0.392, p<0.001) (moderate and weak correlation, respectively). The area under curve for CC identifying low MM was greater in men (AUC 0.697, 78% sensitivity and 62% specificity) compared to women (AUC 0.563, 50% sensitivity and 63% specificity). Kappa test showed a fair agreement (0.366, p < 0.001) between the methods, however it was observed only for male. Death by any cause occurred in 16% of patients and significant predictor of mortality was identified only for low CC (HR = 3.02; CI 1.52-6.03, p = 0.002). In conclusion, low CC was consistent with low SMI for male patients and predicts risk of mortality in patients with cancer. The findings allow the use of CC as a simple screen tool in clinical practice to identify cancer patients in risk of death to target specific care.