Análise do prognóstico e sobrevivência do osteossarcoma alto grau na paraíba

• Ano de defesa: 2023
• Autor(a) principal: Siqueira, André Luís Lopes Gomes de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso embargado
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Ciências Exatas e da Saúde; Programa de Pós-Graduação em Modelos de Decisão e Saúde; UFPB
• Palavras-chave em Português: Osteossarcoma; Regressão logística; Regressão de cox; Ortopedia; Neoplasias ósseas; Logistic regression; Logistic model; Orthopedics; Bone neoplasms; CNPQ::CIENCIAS DA SAUDE::SAUDE COLETIVA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/27073

Bone tumors are responsible for 2% of new cases of neoplasms annually in Brazil, and may be of the primary or secondary type when they originate in other locations. Osteosarcomas (OS) are part of the sarcomas that produce immature osteoid tissue, accounting for 10% of diagnoses in children and adolescents, being the most commonly affected. The prognosis is associated with the presence or absence of clinical or histopathological characteristics, such as the presence or absence of pulmonary metastasis. The purpose of the study was to analyze the prognosis and survival of osteosarcoma in association with the explanatory variables: age, sex, education, time of onset of symptoms, time until the first consultation, Huvos Index, histological type, tumor size, presence or absence of metastases, type of surgery performed and alkaline phosphatase dosage in the first consultation. For prognostic modeling, death was considered as the response variable and for the analysis of survival, a time of 5 years was defined. Data were collected in a retrospective cohort through 63 medical records at Hospital Napoleão Laureano between 2010 and 2021. Inclusion criteria were patients with OS confirmed through histopathological study. The RStudio tool was used to estimate the Binary Logistic Regression, Kaplan Meier and Cox Regression models, after excluding the multicollinearity of the variables. The death model proved to be adequate with an X(2)= 46.149; p<0.01;RNagelkerke = 0.694. The predictive variables were AF dosage (OR= 1.004; 95% CI= 1.001 – 1.006) and the presence of Lung Metastasis (OR= 26.958; 95% CI= 4.847 – 149.931). Kaplan Meier X^2(1) = 37.824; p<0.01; survival when metastatic of only 15% in up to 1800 days. The Cox Regression model X(1) = 17.137; p<0.01; variables that influenced survival were pulmonary metastasis (OR= 4.927; 95%CI= 1.756 – 13.825) and AF (OR= 1.003; 95%CI= 1.001 – 1.004). The study achieved a significant model and contribution to this rare and unknown pathology. Works carried out in the Brazilian Northeast are rare, adding the importance of knowing the local characteristics and serving as a basis for new research in the area.