Abstract

Objective: To identify prognostic factors of survival in patients with osteosarcoma, using a parametric survival model. Material and methods: A retrospective cohort study was carried out on 68 patients diagnosed with osteosarcoma and who were treated at Traumatology Institute from January 1 2002 to December 31 2012.In order to calculate the mean incidence of osteosarcoma, the population greater than or equal to 15 years old during the study period was considered as a denominator. The following were inclusion criteria: patients of either sex, age greater than or equal to 15 years, and positive biopsy for any kind of osteosarcoma.The following predictor variables were considered: sex, tumour site, lung metastasis, relapse, kind of osteosarcoma, histological grade, treatment, and percentage of tumour necrosis. Results: The mean incidence rate of osteosarcoma in the population studied was 0.98 cases per 1 million people per year. As a poor prognosis, the multivariate model showed: masculine gender, lung metastasis during the first year, relapse, and less than 90% of tumour necrosis. Discussion: The predictor variables of survival in patients with osteosarcoma that agree with other multivariate models are: masculine gender, lung metastasis, and less than 90% of tumour necrosis. On the other hand, the discrepancy with other authors can be explained by the wide variation in the methodology used, as well as the variability of the prognostic factors considered, rather than tangible differences in patient management in the various medical centres.