Pulmonary embolism and mortality in patients with fractured hips--a prospective consecutive series.

To find out the causes of death with particular reference to venous thromboembolism all patients being operated on for hip fractures who were taking part in a trial of two methods of prophylaxis against thromboembolism were consecutively and prospectively registered. A total of 806 patients were included, 66 of whom died within three months (8%). The necropsy rate was 64%. The patients who died were significantly older than those who did not. Pulmonary emboli were diagnosed in 17 of the 42 necropsies: 3 fatal, 5 contributory, and 9 incidental. The patients with fatal and contributory emboli died a median of 31 days postoperatively. In the 24 patients who did not have necropsies the clinical cause of deaths were cardiac insufficiency (n = 11), pneumonia (n = 8), pulmonary embolism (n = 2), and myocardial infarction, cerebral infarction, and pancreatic cancer (n = 1 each). The incidence of fatal pulmonary embolism therefore varies between a minimum of 0.37% and a theoretical maximum of 3.3%. In conclusion, fatal pulmonary embolism after operations for fractured hips is low where routine thromboprophylaxis is used. Most patients who develop large pulmonary emboli are old but live independently. To study causes of death a high necropsy rate is essential.