Pulmonary embolism: diagnosis, incidence and implications.

The purpose of this study was to determine the prevalence of pulmonary embolism at an academic tertiary care center, to identify the methods of diagnosis, and to characterize the outcomes for this population. A retrospective analysis of multiple data sources including records from the Medical Information System, and the Departments of Pathology, Radiology, Nuclear Medicine and Pharmacy were reviewed for all patients with positive reports of pulmonary embolism. These were verified by the medical records and a listing of all positive cases was developed. A total of 320 patients developed pulmonary embolism during the study period and 121 died before discharge. The prevalence was 3.5 cases per 1000 admissions, with a mortality of 1 per 1000 cases. Significant misclassification of patients and errors in rate estimation would have occurred if multiple sources of information were not used. Definitive diagnosis and appropriate treatment with a Greenfield filter was associated with improved early survival. By relying on ICD-9 coding, 80% of fatalities and 13% of patients with positive angiograms would have been missed.