Diagnostic strategy for pulmonary embolism could reduce use of CT
The combination of a D-dimer level below 1,000 ng/mL and low clinical pretest probability correctly identified a group of patients at low risk for pulmonary embolism during follow-up, according to results of a prospective study published in The New England Journal of Medicine.
PE is usually diagnosed through chest imaging with CT scans. However, chest imaging results in exposure to radiation, contrast reactions and high costs. It also is time-consuming.
“The primary goal of diagnostic testing for pulmonary embolism is to identify which patients should be treated with anticoagulant agents and which should not,” Clive Kearon, PhD, MB, professor of medicine at McMaster University and a thrombosis specialist with Hamilton Health Services in Canada, said in a press release. “When a physician is concerned that pulmonary embolism may be present, chest imaging with CT pulmonary angiography is usually done in half of these patients. We wanted to find a way to reduce the number of CT scans that need to be done.”
Kearon and colleagues tested whether they could rule out PE in patients with a low clinical pretest probability and a D-dimer level below 1,000 ng/mL and in patients with a moderate clinical pretest probability and a D-dimer level below 500 ng/mL.
Researchers analyzed 2,017 patients (mean age, 52 years; 66.2% women) with signs or symptoms of PE. Physicians used Wells clinical prediction rule scores to classify patients’ clinical pretest probability as low, moderate or high. Most patients (86.9%) had a low clinical pretest probability, whereas 10.8% had moderate probability and 2.3% had high probability.
Patients with a low clinical pretest probability and negative D-dimer test (< 1,000 ng/mL) or a moderate clinical pretest probability and negative D-dimer test (< 500 ng/mL) received no further testing for PE nor anticoagulant therapy. The other patients underwent chest imaging with CT pulmonary angiography or ventilation-perfusion lung scanning.
Among all patients, 7.4% had PE on initial diagnostic testing. None of the patients with a low (n = 1,285) or moderate (n = 40) clinical pretest probability and a negative D-dimer test (0%; 95% CI, 0-0.29) had venous thromboembolism during follow-up. These included 315 patients who had a low clinical pretest probability and a D-dimer level of 500 ng/mL to 999 ng/mL (0%; 95% CI, 0-1.2).
Among 1,863 study participants not diagnosed initially with PE and who did not receive anticoagulant therapy, one (0.05%; 95% CI, 0.01-0.3) had VTE.
With the diagnostic strategy used in the study, 34.3% of patients underwent chest imaging. A strategy that would rule out PE among those with a low clinical pretest probability and a D-dimer level below 500 ng/mL would result in the use of chest imaging for 51.9% of patients, for a percentage-point difference of 17.6 (95% CI, 19.2 to 15.9).
All but one of the study participants were outpatients, so the findings may not apply to inpatients, which served as the study’s primary limitation.
“Our analyses show that PE is ruled out by a D-dimer level of less than 1,000 ng/mL in patients with a low probability, and by a D-dimer level of less than 500 ng/mL in patients with a moderate probability,” Kearon said in a press release. “This way of using D-dimer testing and clinical assessment reduced the need for CT scanning by one-third.” – by John DeRosier
Disclosures: Kearon reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.