1. Home>
2. Specialties>
3. Emergency Medicine>
4. Summary and Comment

ED Management of DVT and PE

Pretest clinical probability scoring combined with D-dimer testing drives an algorithm for outpatient treatment of most patients with suspected deep vein thrombosis or pulmonary embolism.

The usefulness of pretest clinical probability (PCP) combined with D-dimer testing for assessing risk for deep vein thrombosis (DVT) and pulmonary embolism (PE) was evaluated in two recent studies. In a prospective, observational Italian study, Siragusa and colleagues tested an algorithm that allowed them to send home almost all patients with suspected DVT or PE. Patients on oral anticoagulant therapy were excluded. The algorithm was based on assignment of PCP according to the Wells criteria and results of subsequent D-dimer testing. Patients with low PCP (regardless of D-dimer results) and those with moderate PCP and negative D-dimer results were sent home without therapy. Patients with moderate PCP and positive D-dimer results, and those with high PCP (regardless of D-dimer results), were sent home on protective treatment with low-molecular-weight heparins, which were initiated in the emergency department. All patients were followed up with objective testing for DVT or PE within 72 hours (median time, 51.5 hours) and at 3 months.

Overall, 533 consecutive patients with suspected DVT or PE were studied. At short-term follow-up, clots were detected in approximately 6.5% of patients who did not receive anticoagulation (a prevalence consistent with that found in numerous prior studies). Half of the patients who received anticoagulation did not have clots. Only one patient (0.2%), who received protective anticoagulation, had worsening symptoms, and no patient had bleeding. At long-term follow-up, one patient with moderate PCP and negative D-dimer results and four with moderate or high PCP and positive D-dimer results developed DVT. These four patients had negative results on initial objective testing. The negative predictive value (NPV) of D-dimer testing was 98.1% in patients with low PCP and 93.4% in those with moderate PCP.

In a Swiss study, Righini and colleagues assessed whether the high NPV of D-dimer testing applies to patients with high PCP and whether raising the cutoff value for a normal D-dimer result would improve the test's positive predictive value while maintaining its high NPV. The authors retrospectively analyzed data from 1409 patients who were enrolled in two studies of ED diagnosis of PE. PCP was assessed using the Geneva classification system, which has the same predictive value as the Wells system.

During a 3-month follow-up, thrombosis was detected in 7% of patients with low PCP, 35% of those with intermediate PCP, and 77% of those with high PCP. The investigators found that in patients with high PCP and a negative D-dimer test, the 95% confidence interval (0%-23%) of 3-month thromboembolic risk was too wide for the test to be used safely as a clinical tool. Raising the cutoff level from 500 µg/L to 700 µg/L improved specificity, but at too great a sacrifice in sensitivity; in patients with intermediate PCP, the NPV dropped from 100% to 93%.

Comment: These studies add to the remarkable body of high-quality information about ED management of venous thrombosis. Results of the Italian study answer the all-important question, "Can I send this patient home?" (answer, yes, in almost every case) and the subsequent question, "Which of these patients should I treat?" (answer, those with intermediate or high PCP and positive D-dimer results). The Swiss study confirms the wisdom of this approach, suggesting that since virtually all high-risk patients are D-dimer-positive, treating all such patients might be the most efficient approach.

Two points of importance: It is wise to use a clinical prediction score and make the scoring sheet part of the permanent record. The scoring process is central to these and other similar studies. National emergency medicine organizations and hospital associations should seek increased payment for management of patients with suspected venous thrombosis in the ED because in just 36 months, the entire process has moved from the hospital (several days of inpatient care) to the ED, saving millions of dollars.

— J. Stephen Bohan, MD, MS, FACP, FACEP

Published in Journal Watch Emergency Medicine February 2, 2005

Citation(s):

Siragusa S et al. Deferment of objective assessment of deep vein thrombosis and pulmonary embolism without increased risk of thrombosis: A practical approach based on the pretest clinical model, D-dimer testing, and the use of low-molecular-weight heparins. Arch Intern Med 2004 Dec 13/27; 164:2477-82.

• Original article (Subscription may be required)
• Medline abstract (Free)

Righini M et al. Clinical usefulness of D-dimer depending on clinical probability and cutoff value in outpatients with suspected pulmonary embolism. Arch Intern Med 2004 Dec 13/27; 164:2483-7.

• Original article (Subscription may be required)
• Medline abstract (Free)