Deep venous thrombosis and pulmonary embolism are potentially life-threatening problems that present diagnostic challenges. To employ objective diagnostic tests in an efficient, safe, and cost-effective manner, the clinical probability of these disorders should be estimated before testing. A number of clinical prediction rules are available for suspected deep venous thrombosis, while there are three major prediction rules available for estimating the probability of pulmonary embolism. Recent modifications of the Wells score for deep venous thrombosis simplify its use. Although the Wells score for pulmonary embolism is commonly used, two other rules are useful for this disorder as well. This review summarizes the clinical prediction rules and gives recommendations about their application.
Key Points
* Clinical prediction rules that incorporate findings from history, examination, and simple tests allow the clinician to determine the probability of deep venous thrombosis or pulmonary embolism in patients suspected of these disorders.
* The Wells score has been validated for deep venous thrombosis and appears to function well across a range of clinical venues that include the inpatient wards and emergency room.
* Three clinical prediction rules-the Wells score, the Geneva score, and the Pisa model-are useful for predicting the probability of pulmonary embolism.
* The prediction rules for deep venous thrombosis and pulmonary embolism allow the clinician to use and interpret the D-dimer test, as well as diagnostic algorithms.
* Clinical impression appears to be helpful in modifying the clinical probability determined by these clinical prediction rules.