Comprehensive geriatric assessment as a morbimortality predictor in patients vulnerable to the chemotherapy treatment/Avaliacao geriatrica ampla como preditor de morbimortalidade nos pacientes oncologicos vulneraveis submetidos ao tratamento quimioterapico

Objectives: to determine the predictive value of the Comprehensive geriatric assessment (CGA) regarding toxicity and mortality in elderly cancer patients classified as vulnerable who were submitted to chemotherapy. Method: A prospective cohort study was performed with data gathered from the database already built by the anchor project and available online. Tables of the frequency of distribution of the studied variables were created. The chi-square test was used in the univariate analysis, and the Fisher test when indicated. The level of significance of 5% were considered and all the tests were applied with a confidence of 95%. Results: The frequency of patients classified as vulnerable by CGA was 40.2%, and the mortality rate in the total number of patients was 37%. The association between nutritional status and mortality (p=0.003) was observed, where most of the denuded patients (51.4%) died during follow-up, and most patients (46.5%) had some type of complication. Conclusion: There was an association between the increase in mortality in patients vulnerable to chemotherapy (p=0,046), when compared to healthy patients. The results of this study follow the literature on the use of CGA can help oncologists and geriatricians in their decision making. Keywords: Medical oncology; Geriatrics; Geriatric assessment; Predictive value of tests; Indicators of morbidity and mortality; Drug therapy; Patients. Objetivos: O objetivo desse estudo foi determinar o valor preditivo da Avaliacao Geriatrica Ampla (AGA) em relacao a toxicidade e mortalidade nos pacientes oncologicos idosos classificados como vulneraveis, submetidos a quimioterapia. Metodo: Foi realizado um estudo de coorte prospectivo com coleta de informacoes extraidas de banco de dados ja construido do projeto ANCORA e disponivel online. Foram construidas tabelas de distribuicao de frequencia das variaveis estudadas e na analise univariada foi usado o chi-quadrado e Teste de Fisher, quando indicado. Por fim, considerou-se o nivel de significancia de 5% e todos os testes foram aplicados com 95% de confianca. Resultados: Em uma amostra total de 254 pacientes, o percentual de mortalidade foi de 37% e de intercorrencias foi de 46,5%, sendo infeccao a mais frequente. Houve associacao entre o deficit nutricional e o aumento da mortalidade (p=0,003). Conclusao: Observou-se associacao estatistica entre o aumento da mortalidade em pacientes vulneraveis que realizaram quimioterapia (p=0,046), quando comparados aos pac