Accuracy of Very Low Pretest Probability Estimates for Pulmonary Embolism Using the Method of Attribute Matching Compared with the Wells Score

Abstract

Objectives: Attribute matching matches an explicit clinical profile of a patient to a reference database to estimate the numeric value for the pretest probability of an acute disease. The authors tested the accuracy of this method for forecasting a very low probability of venous thromboembolism (VTE) in symptomatic emergency department (ED) patients. Methods:The authors performed a secondary analysis of five data sets from 15 hospitals in three countries. All patients had data collected at the time of clinical evaluation for suspected pulmonary embolism (PE). The criterion standard to exclude VTE required no evidence of PE or deep venous thrombosis (DVT) within 45 days of enrollment. To estimate pretest probabilities, a computer program selected, from a large reference database of patients previously evaluated for PE, patients who matched 10 predictor variables recorded for each current test patient. The authors compared the outcome frequency of having VTE [VTE(+)] in patients with a pretest probability estimate of <2.5% by attribute matching, compared with a value of 0 from the Wells score.Results: The five data sets included 10,734 patients, and 747 (7.0%, 95% confidence interval [CI] = 6.5% to 7.5%) were VTE(+) within 45 days. The pretest probability estimate for PE was <2.5% in 2,975 of 10,734 (27.7%) patients, and within this subset, the observed frequency of VTE(+) was 48 of 2,975 (1.6%, 95% CI = 1.2% to 2.1%). The lowest possible Wells score (0) was observed in 3,412 (31.7%) patients, and within this subset, the observed frequency of VTE(+) was 79 of 3,412 (2.3%, 95% CI = 1.8% to 2.9%) patients.Conclusions: Attribute matching categorizes over one-quarter of patients tested for PE as having a pretest probability of <2.5%, and the observed rate of VTE within 45 days in this subset was <2.5%.ACADEMIC EMERGENCY MEDICINE 2010; 17:133-141 ª

Effectiveness of D-dimer as a screening test for venous thromboembolism: An update

“…[22] A missed diagnosis of VTE can lead to sudden death, chronic cardiopulmonary dysfunction, and impaired quality-of-life. [72] Among patients with suspected VTE, it is important to identify those who do not have the disease, in order to obviate the need for anticoagulation therapy and associated bleeding risk. [73] Simultaneously, over-testing patients for PE can be costly and expose them to harm from radiation, intravenous contrast, a false-positive diagnosis, and anticoagulation treatment.…”

Age and sex-dependent trends in pulmonary embolism testing and derivation of a clinical decision rule for young patients

“…More recent work, including the Pulmonary Embolism Rule-out Criteria (PERC), has focused on identifying very low risk patients in whom testing can be avoided entirely 12 13. Age has been included as a factor in these investigations, but the sample populations all have mean age above 50 years and analyses have not focused on the youngest patients.…”

Personalized risk stratification through attribute matching for clinical decision making in clinical conditions with aspecific symptoms: The example of syncope

“…Attribute matching (AM) is a prediction approach that differs considerably from the regression models and has shown promising results in ruling out acute coronary syndrome and pulmonary embolism in patients with chest pain [13][14][15]. Instead of considering each clinical characteristic as an individual predictor and deriving a risk estimate based on the sum of their regression coefficients, each individual patient is matched to a group of patients with the same combination of the relevant clinical characteristics (or attributes) from a large reference database.…”