Análise de sobrevida dos pacientes portadores de tumores do Sistema Nervoso Central acompanhados no Serviço de Oncologia Pediátrica do Hospital da Baleia

Abstract

OBJETIVES: To evaluate the outcome of children and adolescents with the diagnosis of Central Nervous System tumors treated and attended at the Pediatric Oncology Service of the Hospital da Baleia. To evaluate the influence of prognostic factors in the probability of survival of these patients, including: age, gender, provenience, interval between the time of onset of the symptoms and the diagnosis, histological classification of the tumors and the stage of the disease. PATIENTS E METHODS: Patients enrolled in this retrospective study were less than nineteen years old at the time of the diagnosis of a Central Nervous System tumor. All of them were treated or attended at the Pediatric Oncology Service of Hospital da Baleia between March,2003 and December,2009. Clinical data were obtained by chart review. Kaplan-Meier method was used to estimate the overall survival (OS) and event free survival (EFS). The logrank test was used to compare survival curves and Coxs method in the multivariate analysis. RESULTS: The study population consisted of 159 patients. The follow up period ranged from three days to 144 months (median: 13 months). The median of the time between onset of the symptoms and the diagnosis was 1,9 months, maximum of 42 months. Boys comprised 52% of the studied population (83 out of 159). The age at diagnosis ranged between 23 days and 18,5 years (median: 7,2 years). Most common diagnostic subtypes were low grade glioma (27%), medulloblastoma (19,5%) and brain stem tumors (17,6%). The majority of patients (59,5%) came from the central region of Minas Gerais state. Headache was the most frequent symptom (57%) and infratentorial tumors correspond to 55,3% of the cases. Only 40,9% of the patients had magnetic resonance imaging (MRI) study at the time of diagnosis, and lumbar puncture was performed in 28,3% to evaluate the presence of malignant cells. Surgery were performed in 113 patients (71,1%), 77% of them at the Hospital da Baleia. Radiotherapy in 55,4% and chemotherapy in 54,7%. Tumor progression was the main cause of death in the study. An estimated OS of 42% (CI 95%, 33% to 53%) and EFS of 32% (CI 95%, 24% to 43%) were observed for the role group. An association between the outcome and the following variables was observed: interval between the time of onset of the symptoms and the diagnosis (OS p=0,046), histological classification of the tumors (OS and EFS p<0,001), provenience (OS p=0,027 and EFS p=0,026) and the local of surgery realization (OS and EFS p<0,001). In the multivariate analysis, these variables were sustained, added of the primary location of the tumor in the Central Nervous System. The study showed that the death rate decreases when the intervals between the onset of symptoms and the diagnosis is longer. There was a significant association between histological diagnosis and outcome, although it was not possible to estimate its value because there were many histological subtypes and relative few patients in the sample. Patients who have tumors of the spine showed seven times more risk of death than the infratentorial group. Patients who came from the region II of Minas Gerais state had four times more risk of death than those of the Central region. And those patients that didnt have surgery indication had near six times more risk of death when compared with those that have been operated upon the Hospital da Baleia (p<0,001 for all). The study detected a tendency of an increased risk of death for those patients who were submitted to surgery in other hospitals. CONCLUSIONS: The overall and event free survivals detected in this study were clearly worse when compared with international centers rates, although were similar to two other Brazilian studies. The elevated amount of patients with the diagnosis of brain stem tumor, which is known to have a bad prognosis, in those three studies, could partially explain these results. The study indicates a necessity to improve the methods of diagnosis, staging, treatment and follow up of children and adolescents with Central Nervous System tumors at the institution and at the region where it was performed. It is important to develop multicentric studies to confirm the data showed in this study.