Risk factors for pulmonary emboli after total hip or knee arthroplasty

Because it is difficult to predict which patients may sustain a pulmonary embolism after total hip or knee arthroplasty, we assessed multiple thrombophilic and hypofibrinolytic parameters to identify risk factors. Twenty-nine patients who survived a known pulmonary embolism after total knee or total hip arthroplasty were matched by age, gender, race, arthritic diagnosis, procedure, and surgery date with 29 patient-controls who had a total hip or knee arthroplasty but who did not have a symptomatic known pulmonary embolism or deep vein thrombosis. Twenty-one serologic measures and five genes associated with thrombophilia, hyporibrinolysis, or both were assessed without knowledge of group assignment. All patients with pulmonary embolism had at least one abnormality of plasminogen activator inhibitor activity, dilute Russell's viper venom time, prothrombin time, or total cholesterol versus 13 of 27 (48%) control patients. Forty-seven percent of patients who experienced pulmonary embolism had at least two abnormalities of plasminogen activator inhibitor activity, dilute Russell's viper venom time, prothrombin time, or total cholesterol, versus 7% of control patients. Preoperatively, to identify patients at high risk of pulmonary embolism, plasminogen activator inhibitor activity, dilute Russell's viper venom time, prothrombin time, and cholesterol levels were most predictive. Using at least one abnormality of these four me asures as a screening test to detect risk of pulmonary embolism, the test is sensitive (100%), and the predictive value of a negative test is high (100%). After additional prospective study, this may allow identification of patients at low risk (the majority of patients) in whom anticoagulation may not be required and a small group of patients at high risk for pulmonary embolism in whom prophylactic anticoagulation should be provided.