OBJECTIVE: to correlate the clinical data of necropsies in hospitalized patients of a Pediatric Rheumatology Unit. METHODS: this study is a historic cohort. In ten years, from January 1994 to December 2003, there were 57,159 hospitalizations with 1,907 (3%) deaths in Instituto da Criança. From these hospitalizations, 548 (1%) presented rheumatic or pediatric diseases and were followed by the Pediatric Rheumatology Unit, involving 348 patients. The deaths and necropsies were analyzed and the Goldman Classification was used to detect any disagreement between the clinic diagnosis and the necropsy. RESULTS: over the period of this study, 34 (10%) of patients died. The main diseases that caused deaths were: juvenile systemic erithematosus lupus (JSLE) in 18 patients (53%) and juvenile idiopathic arthritis (JIA) in 7 cases (21%). Necropsies were done in 21 patients (64% of deaths). In 18 cases of JSLE, disease activity was present in 16 cases, being associated to septicemia in 15. Differences between the clinical diagnosis and necropsy were observed in 6 cases of JSLE: 3 with fungal infection, 1 with tuberculosis, 1 with diffuse proliferative nephritis and 1 with atherosclerosis. Seven patients with JIA died: septicemia in 4 and macrophage activation syndrome in 3. In 1 case of JIA were detected only by necropsy: Hodgkin lymphoma, atherosclerosis and anterior myocardial infarction. CONCLUSIONS: the frequency of deaths was 10% and necropsy showed infection disease, atherosclerosis or malignancy not detected previously. Necropsy is important to determine events not detected or doubtful in death patients and must always be requested.
pediatric rheumatology; juvenile systemic lupus erithematosus; juvenile idiopathic arthritis; mortality; necropsy
