Abstract
CONTEXT AND OBJECTIVE: Malnutrition in cancer patients has many causes. Nutritional status is usually assessed from weight/height indices. These present limitations for the nutritional assessment of cancer patients: their weights include tumor mass, and lean mass changes are not reflected in weight/height indices. The objective was to evaluate differences between two anthropometric methods and compare deficits, in non-hematological tumor patients and hematological disease patients. DESIGN AND SETTING: Cross-sectional study at Instituto de Oncologia Pediátrica, Universidade Federal de São Paulo. METHODS: Children and adolescents were evaluated between March 1998 and January 2000. Traditional anthropometric measurements were obtained in the first month of treatment (induction therapy), by weight-for-height (W/H) using z-scores index for children and body mass index (BMI) for adolescents. Body composition evaluations consisted of specific anthropometric measurements: triceps skinfold thickness (TSFT), mid-upper arm circumference (MUAC) and arm muscle circumference (AMC). Data were analyzed to compare nutritional assessment methods for diagnosing malnutrition prevalence. The chi-squared test was used for comparative analyses between tumor patients and hematological disease patients. RESULTS: Analysis was done on 127 patients with complete data. Higher percentages of deficits were found among tumor patients, by W/H z-scores or BMI and by MUAC and AMC. Higher percentages of deficits were shown by TSFT (40.2%) and MUAC (35.4%) than by W/H z-scores or BMI (18.9%). CONCLUSION: Non-hematological tumor patients presented higher malnutrition prevalence than did hematological disease patients. Body composition measurements by TSFT and MUAC detected more patients with malnutrition than did W/H or BMI.