RÉSUMÉ
Venous thromboembolism is a potentially lethal disease if not properly treated. Noninvasive strategies have become an attractive clinical option for effective diagnosis. There has been controversy, however, regarding the use of standard clinical rules in a primary care setting. The objective of the present study was to validate a noninvasive diagnostic strategy in an emergency unit giving assistance to patients with primary and secondary care needs. A total of 291 outpatients (primary and secondary care needs) with suspected venous thromboembolism attending the emergency unit of a general hospital from August 2002 to 2004 were retrospectively evaluated. The diagnostic strategy included assessment of risk for venous thromboembolism and a rapid quantitative enzyme-linked immunosorbent assay D-dimer test. Venous thromboembolism was ruled out in patients with a low-probability or intermediate-probability risk (or an unlikely diagnosis) and a negative D-dimer. The prevalence of venous thromboembolism was 8.2%. Patients with an unlikely diagnosis comprised 93.8% of the evaluations for deep venous thrombosis, and those with a low probability for pulmonary embolism comprised 81.4%. Proportions of patients with venous thromboembolic events observed were 7.2% in patients with an unlikely diagnosis of deep venous thrombosis and 3.0% in those with low probability for pulmonary embolism. The percentage of patients with a thrombotic event excluded using this strategy was 37% (positive predictive value 13%, negative predictive value 100%). In conclusion, this noninvasive clinical strategy is safe for ruling out venous thromboembolism, and excludes the need for imaging tests in about one-third of the patients in the population studied.