Background: The study was conducted to assess the rate of suspected pulmonary embolism (PE) prior to death and the diagnostic and therapeutic procedures performed.
Methods: Patients with autopsy-confirmed PE between 1998 and 2002 were included. Autopsy register and medical records were reviewed for history, diagnosis and therapy of PE. Patients were categorised into fatal and non-fatal PE according to the autopsy findings.
Results: 102 patients with fatal and 247 patients with non-fatal PE were eligible for analysis (median age 68 years; 24-95). In 58.8% with fatal and in 32% with non-fatal PE, disease was suspected pre-mortal. Clinical suspicion of PE was significantly enhanced in venous thrombosis (Odds Ratio [OR] = 12.17, p=0.004) and significantly decreased for chronic vascular disease (OR = 0.30, p=0.002). Recurrent PE was demonstrated in 31.4% fatal and in 4.5% non-fatal PE (OR = 9.81, p=0.001). 7% of all PE were localised centrally, 19% centrally and peripherally and 74% peripherally. Dyspnoea and tachycardia were the most frequent symptoms in fatal PE. About half of all patients suffered from malignancies. Suspicion of PE decreased after day 14 of hospitalisation (OR = 0.33, p=0.021).
Conclusion: PE often is not diagnosed pre-mortally. Patients with chronic vascular disease and tumours as well as those hospitalised for over 14 days are at particular risk for PE.
Georg Thieme Verlag KG Stuttgart, New York.