RESUMO Objetivo: Verificar associação entre indicadores antropométricos e as escalas Avaliação Nutricional Subjetiva Global (ANSG) e Triagem de Risco para Estado Nutricional e Crescimento (STRONGkids). Métodos: Estudo transversal com pacientes de 0 a 18 anos, internados no Hospital das Clínicas, Goiânia (GO), entre agosto e novembro de 2015. Foram incluídas crianças e adolescentes admitidos em até 48 horas. Excluíram-se pacientes que requeriam instrumentos específicos para avaliar o estado nutricional e os internados em Terapia Intensiva. Coletaram-se dados de identificação, antropométricos e foram aplicadas a ANSG e a STRONGkids. Feita análise de comparação de proporções e avaliação de concordância, sendo significante p<0,05. Resultados: Avaliaram-se 71 pacientes, dos quais 9,6% com baixo ou muito baixo peso/idade, 9,7% com magreza ou magreza acentuada pelo índice peso/estatura, 16,9% com comprometimento da estatura, 7% com magreza pelo índice de massa corporal/idade e 32,4% desnutridos pela circunferência muscular do braço. A STRONGkids detectou 69% da amostra com risco nutricional moderado ou alto. Pela ANSG, a prevalência de desnutrição foi de 38,1%. Houve associação entre ANSG e índice de massa corporal/idade (p=0,022), estatura/idade (p<0,001) e circunferência muscular do braço (p=0,014). Não houve associação entre a STRONGkids e os indicadores antropométricos. As ferramentas se associaram para: risco nutricional elevado versus desnutrição grave e baixo risco nutricional x bem nutridos (p<0,001), porém a concordância foi fraca (k=0,255). Conclusões: Recomenda-se utilizar o STRONGkids como instrumento de triagem por apresentar maior sensibilidade para diagnosticar pacientes com risco nutricional. A ANSG deve ser aplicada para avaliação nutricional devido à associação com a antropometria.

ABSTRACT Objective: Verify the association between anthropometric indicators and the Subjective Global Assessment of Nutritional Status (SGA) and the Screening of Risk for Nutritional Status and Growth (STRONGkids) scales. Methods: A cross-sectional study with patients from 0 to 18 years admitted in the Hospital das Clínicas, Goiânia (GO), between August and November 2015. Children and adolescents admitted in up to 48 hours were included. Patients who required specific instruments for assessing their nutritional status and those hospitalized in Intensive Care were excluded. Identification and anthropometric data was collected and applied to the SGA and STRONGkids. We performed an analysis comparing proportions and did an agreement assessment, where p<0.05 was significant. Results: 71 patients were evaluated, of whom 9.6% had low or very low birth weight/age, 9.7% had thinness or accentuated thinness according to the weight/height index, 16.9% had a height impairment, 7% were thin according to the body mass index/age, and 32.4% were malnourished with regard to arm muscle circumference. The STRONGkids detected that 69% of the sample had a moderate or high nutritional risk. According to the SGA, malnutrition prevalence was 38.1%. There was an association between the SGA and body mass index/age (p=0.022), height/age (p<0.001) and arm muscle circumference (p=0.014). There was no association between the STRONGkids and anthropometric indicators. A correlation was found between: high nutritional risk versus severe malnutrition and low nutritional risk x the well-nourished (p<0.001), but the agreement was weak (k=0.255). Conclusions: It is recommended to use the STRONGkids as a screening instrument because it has a higher sensitivity for diagnosing patients with a nutritional risk. The SGA should be applied to nutritional assessment due to its association with anthropometry.