[Skip to Content]
[Skip to Content Landing]
February 1979

Risk Assessment of Pulmonary Embolism by Multivariate Analysis

Author Affiliations

From the Departments of Surgery (Drs Sigel, Justin, and Felix), Radiology (Drs Popky and Parker), and Community Medicine (Dr Ipsen), Medical College of Pennsylvania, Philadelphia, and the Franklin Institute Research Laboratories (Mr Gibson), Philadelphia. Dr Sigel is presently with the Abraham Lincoln School of Medicine, University of Illinois, Chicago; Dr Felix, the West Roxbury Veterans Administration Hospital, Boston; and Dr Ipsen, the Arhus University, Denmark.

Arch Surg. 1979;114(2):188-192. doi:10.1001/archsurg.1979.01370260078013

• A prospective study of 6,527 hospitalized patients was performed to evaluate the effectiveness of multivariate analysis of risk factors to correctly designate risk for pulmonary embolism. History of previous pulmonary embolism was the single most important factor. In patients without prior history, five factors emerged: inactivity, congestive heart failure, Doppler ultrasound evidence of deep-vein occlusion, female sex, and black race. Used together, these factors permitted a discrimination of risk such that 68.7% of pulmonary embolization was found to occur in 32.2% of the population designated as showing increased risk. Multivariate analysis of clinical factors improved assessment of risk, compared to the use of lower extremity findings alone, and proved to be useful in identifying individuals at increased risk for pulmonary embolism.

(Arch Surg 114:188-192, 1979)