Comparison between clinical diagnoses and autopsy findings in a pediatric intensive care unit in São Paulo, Brazil

Abstract

Objective: To verify the frequency of discrepancies between clinical diagnoses and autopsy findings in patients from a pediatric intensive care unit and to look for predictive factors of the discrepancies.

Design: Prospective evaluation performed between September 1996 and December 1998.

Setting: Eight-bed pediatric intensive care unit of a university hospital.

Patients: One hundred and two autopsies.

Interventions: None.

Measurements and main results: Disagreements between autopsy and antemortem diagnoses were classified as proposed by Goldman. Patient age, presence of underlying disease, and length of stay were studied as possible predictive factors for diagnosis discrepancies. During the 28 months of study there were 779 admissions to the pediatric intensive care unit; the death rate was 26% and the autopsy rate was 55%. One hundred and two of 114 (89.5%) autopsies were evaluated. The median age of the patients was 21 months, and 85% of them had a previous underlying disease. One third of patients died before 24 hrs of admission to the pediatric intensive care unit. The autopsy revealed unexpected findings in 73 study patients (72%), 33 of which were related to "major diagnoses" (Goldman's classes I or II), either causes of death or main underlying disease. In 12 patients (12%), the correct diagnosis, if known before death, might have led to a change in the patient's therapy or outcome (class I). Unexpected findings in this group included viral or fungal infection and pulmonary embolism. None of the possible predictive factors that we studied showed significant statistical association between clinical and autopsy discrepant diagnoses in the univariate analysis.

Conclusions: Although diagnoses of both cause of death and underlying disease were accurate in most cases before death, some autopsies revealed findings that would have changed intensive care unit therapy. Nonbacterial infections and pulmonary thromboembolism should always be considered when managing critically ill patients with underlying disease. Autopsy examinations continue to provide important information, especially in the pediatric intensive care unit setting, despite the advances in diagnostic technology.