Diagnostic discrepancies between antemortem clinical diagnosis and autopsy findings in pediatric cancer patients.

Prevalence of discrepancies between antemortem clinical diagnoses and postmortem autopsy findings is uncertain in pediatric oncology given improving diagnostic capabilities over time. Primary objective was to describe discrepancies between antemortem and postmortem diagnosis of pediatric cancer deaths. Secondary objective was to compare clinical characteristics of deaths with and without major diagnostic discrepancies. This was a retrospective study that included pediatric cancer patients diagnosed and treated in Ontario and who died from 2003 to 2012. Antemortem clinical diagnoses associated with mortality were determined by reviewing the patient's health records 2 weeks prior to death while the postmortem diagnoses were determined by the autopsy report. Discrepancies among these diagnoses were classified using the Goldman criteria where major discrepancies were directly related to the cause of death in contrast to minor discrepancies. Among the 821 patients who died, 118 (14%) had an autopsy and were included. Of these autopsies, 12 (10%) had a major diagnostic discrepancy between antemortem and postmortem diagnoses. Major discrepancies consisted of opportunistic infections (n = 5), missed cancer diagnosis (n = 3), and organ complications (n = 4). Death in a high acuity setting (12/12, 100% vs. 60/106, 57%; P = 0.003) and treatment-related mortality (12/12, 100% vs. 60/106, 57%; P = 0.003) were significantly associated with major discrepancy. Major diagnostic discrepancy was found in 10% of pediatric oncology autopsies. Missed infections and organ complications were predominant etiologies. Death in a high acuity setting and treatment-related mortality were associated with major diagnostic discrepancies. Autopsies continue to be important for improving diagnostic insight and may improve future clinical care.