Pulmonary embolism can be a difficult diagnosis in the ED. How are you using d-dimer and spiral CT for the work-up of pulmonary embolism?

by
Jonathan A. Edlow, MD
Vice chairman,
Department of Emergency Medicine
Beth Israel Deaconess Medical Center
Assistant Professor of Medicine
Harvard Medical School Boston, MA

D-dimer is a substance formed in the blood when fibrin is degraded. Theoretically, its absence implies that there is no active clotting occurring in the body. This fact has been exploited by the following logic: if a patient being ruled out for DVT or PE has a negative "D-dimer", then the diagnosis is ruled out. The theory has not neatly translated into reality. First, one must consider pre-test probability. Second, one must be exceptionally careful about how D-dimer is being measured (latex agglutination, rapid latex agglutination, ELISA). Even within one method, differences exist between manufacturers. The literature suggests that the new rapid ELISA D-dimer made by Bio-merieux is sufficiently sensitive and rapid to be clinically useful, in an ED population (from Canada and Switzerland)[1]. I would personally be comfortable using a negative D-dimer (using this particular method) in excluding the disease in low risk patients. I would not use any other kind at this time nor would I feel comfortable using it on high-risk patients. Other products have their own proponents as well [2,3].

Regarding spiral CT angiography (CTA), at our institution, we use CTA frequently as the test of choice. As with any test, one must be very aware of its limitations [4]. It is not a great test for small emboli. Despite the fact that some argue that small emboli are clinically unimportant, I would not feel comfortable in a patient whom I felt was medium to high probability in excluding PE with a CTA. As with any test, one must factor in the pre-test probability of the disease being tested for. Hardware and software packages, abilities of different radiologists differ and I don't think one can automatically extrapolate data from one center to another. As well, there is no large PIOPED-like study yet published that shows the performance characteristics of CTA (as PIOPED did for lung scanning). Those caveats said, CTA has many advantages (alternative diagnoses, ease of performance, increasing experience with the technique) and it is being increasingly used.

The bottom line is that with these newer diagnostic tests (as well as our current old ones), emergency physicians must be very aware of their limitations.

 

Send our editor your comments on this question with your name, title and affiliation.

References

1. Perrier et al; Lancet; 353: 190-195; 1999.
2. Ginsburg et al; Annals of Internal Medicine; 129: 1006-1011; 1998.
3. Bates et al; Archives of Internal Medicine; 161: 447-453; 2001.
4. Bloomgarden and Rosen; Emergency Clinics of North America; 19: 975-994; 2001.