Improving the Diagnosis of Pulmonary Embolism in the Emergency Department FREE

Blood clots that form in deep veins (deep venous thrombosis) sometimes travel to the lungs, a condition known as pulmonary embolism. Patients who survive a pulmonary embolus need treatment with blood thinners, since having another pulmonary embolus could prove fatal. Methods commonly used to detect pulmonary embolism often fail to provide a definitive diagnosis. By using information from the medical history, physical examination, electrocardiogram, and chest x-ray, a simple clinical model has recently been developed to evaluate the probability of pulmonary embolism before other tests are performed (“pretest probability”). A blood test—the d-dimer test—that detects breakdown products of blood clots is also now available. This test is used most effectively to rule out blood clots by the absence of breakdown products. Neither the simple clinical model nor the d-dimer test had been studied in detail in emergency departments.

To evaluate the safety and reliability of using the clinical model plus the d-dimer test to rule out a diagnosis of pulmonary embolism in emergency department patients.

Nine hundred thirty patients with suspected pulmonary embolism presenting to the emergency department at one of four participating hospitals in Canada were included. According to the study plan, patients were first rated by using the clinical model as having low, moderate, or high pretest probability. Then, d-dimer testing was performed. If the patient's pretest probability was low and the d-dimer result was negative, no additional tests were performed and no blood thinners were administered. All other patients underwent other, more complex tests as needed; these test results were interpreted by physicians with no knowledge of pretest probability or d-dimer results. All patients were checked for deep venous thrombosis or pulmonary embolism at 3-month follow-up.

Pulmonary embolism was eventually diagnosed in 86 of the 930 patients. Of the 527 patients with low pretest probability, 1.3% eventually received a diagnosis of pulmonary embolism. Less than 1% of patients in whom the diagnosis was initially considered excluded developed pulmonary embolus. In some patients, more tests were done than had been called for in the study plan; pulmonary embolism was diagnosed in 7 of those patients. Only 1 patient of the 759 patients whose evaluation in the emergency department followed the study plan and in whom pulmonary embolism was not found developed pulmonary embolism or deep venous thrombosis during follow-up. The accuracy of a negative d-dimer result in ruling out clot formation was 97.3% by itself and 99.5% when combined with low pretest probability.

The study plan was not followed for all patients.

Pretest probability with d-dimer result can safely rule out pulmonary embolism and therefore decrease the need for additional testing.