Validade dos critérios GLIM no diagnóstico de desnutrição para pacientes oncológicos em assistência hospitalar e ambulatorial
Resumo
Resumo: Introdução: A desnutrição é frequentemente observada em pacientes oncológicos e a Avaliação Subjetiva Global Produzida pelo Próprio Paciente (ASG-PPP) é o método de referência para avaliação nutricional nesses indivíduos, porém nota-se alguns pontos limitantes na sua prática. O Global Leadership Initiative on Malnutrition (GLIM) propôs uma ferramenta para diagnosticar desnutrição, mas até o momento não há estudos que o validem para pacientes oncológicos brasileiros, hospitalizados e em acompanhamento ambulatorial, utilizando a ASG-PPP como método comparativo. O objetivo foi validar os critérios GLIM para diagnóstico de desnutrição, por meio do método de concorrência com a ASG-PPP e validade preditiva de desfechos clínicos. Métodos: Estudo de coorte prospectivo com pacientes adultos e idosos, oncológicos, em assistência hospitalar e ambulatorial de um hospital especializado em oncologia. A desnutrição foi identificada pela ASG-PPP e pelo GLIM e os pacientes foram monitorados por meio do prontuário para verificar se houve a necessidade de internamento em Unidade de Terapia Intensiva, desenvolvimento de infecções, óbito intra-hospitalar, internamentos e mortalidade em 60 dias. Resultados: 307 pacientes hospitalizados e 208 em acompanhamento ambulatorial foram avaliados. A desnutrição foi identificada em 65,7% pela ASG-PPP e 71,6% pelo GLIM nos hospitalizados e em 67,8% e 81,2% nos ambulatoriais, respectivamente. O GLIM apresentou acurácia moderada, quando comparado a ASG-PPP, em ambas as amostras (k=0,547 e k=456, respectivamente) e não demonstrou uma validade concorrente satisfatória, por apresentar uma sensibilidade <80% em ambos os casos. Em relação aos desfechos de saúde, não foram encontradas diferenças estatísticas para as análises ajustadas. Conclusão: A validade concorrente e preditiva dos critérios GLIM não foi confirmada para pacientes oncológicos em assistência hospitalar e ambulatorial. Abstract: Introduction: Malnutrition is frequently observed in cancer patients and the Patient-Generated Subjective Global Assessment (PG-SGA) is the nutritional method of reference for cancer patients, however, there are some limiting points in practice. The Global Leadership Initiative on Malnutrition (GLIM) proposed a tool to diagnose malnutrition, but so far there are no studies that validate it for brazilian oncology patients, hospitalized and in outpatient follow-up, using the PG-SGA as a comparative method. The objective was to validate the GLIM criteria for the diagnosis of malnutrition, through the method of competition with the PG-SGA and predictive validity of clinical outcomes. Methods: Prospective cohort study with adult and elderly patients, oncology, hospitalized and in outpatient follow-up of a hospital specializing in oncology. Malnutrition was identified by the PG-SGA and the GLIM and the patients were monitored through the medical records to verify if there was a need for hospitalization in the Intensive Care Unit, development of infections, in-hospital death, hospitalizations and mortality within 60 days. Results: 307 hospitalized patients and 208 outpatients were evaluated. Malnutrition was identified in 65.7% by PG-SGA and 71.6% by GLIM in hospitalized patients and in 67.8% and 81.2% in outpatients, respectively. The GLIM showed moderate accuracy, when compared to the PG-SGA, in both samples (k=0.547 and k=456, respectively) and did not demonstrate satisfactory concurrent validity, as it presented a sensitivity <80% in both cases. Regarding health outcomes, no statistical differences were found for the adjusted analyses. Conclusion: The concurrent and predictive validity of the GLIM criteria was not confirmed for oncology patients in inpatient and outpatient care.
Collections
- Dissertações [103]