D-dimer for PE ============== þ False Positive/Negatives: - Coumadin may cause a false negative D-dimer. A study was done with 64 patients who had a positive diagnosis of PE on radionuclide lung scanning or contrast-enhanced computed tomography. Of these, 20 had had abnormal D-dimer results, 3 had had abnormal alveolar dead-space fraction, 40 had had abnormal results on both tests, and only one patient had had a normal result on both screening tests. The sensitivity for diagnosing PE using the combination of screening tests was 98.4%, according to the research team's report in the February 14th issue of The Journal of the American Medical Association. I also wanted to point out that in urban emergency departments the mean duration required to rule out PE is 8.5 hours. Date sent: Fri, 09 Mar 2001 13:38:08 -0600 From: "Ed Michelson, MD" Subject: Re: FW: Pulmonary Embolism from Genna Nosove To: EMS-L@listserv.it.northwestern.edu Send reply to: EMS-L@listserv.it.northwestern.edu Thanks for your interest in this study. The lead investigator of this multi- center trial was Jeff Kline, at Carolinas Medical Center in Charlotte. Northwestern University, and the NMH ED was one of the study sites, and I am a co-author of the manuscript mentioned below. Jeff presented this data at SAEM in San Francisco last spring. Diagnosis of PE can be very difficult, and involves invasive tests like V-Q scan, Helical CT of the chest with contrast,or pulmonary angiogram. I am sure we miss a number of patient's with PE each year. Our goal in this study was to see if we could validate a relatively non-invasive combination of bedside whole blood d-dimer and capnography to screen for P.E. Both tests are fairly inexpensive, and can be completed in under 10 minutes. Note we were not trying to diagnose P.E., rather to rule that diagnosis out, and potentially save a large number of patients the delay, expense and radiation associated with imaging studies noted above. We studied 380 adult patients presenting to one of the 6 centers in the study. 64 had PE, 316 did not. Of the 64 with PE, 40 patients had BOTH abnormal dead space (increase in deadspace as measured by volumetric capnography) and positive d-dimer. 20 had an abnormal d-dimer only, 3 had abnormal dead space only. Only one patient out of the 380 enrolled who had a P.E. came up with a normal result in both of our screens (d-dimer and deadspace) Had we relied only on d-dimer, we would have diagnosed 60/64 PE's Had we relied only on deadspace measurement we would have diagnosed 43/64 PE's The combination of the two tests, with an abnormality in either or both suggesting the possibility of PE netted us 63/64 positives out of the total population of 380. One potential problem with this study is the 'gold standard' of who did have a P.E. The most definitive test would be a pulmonary angiogram, and it is next to impossible to get 380 people to agree to undergo pulmonary angiography, not to mention the cost, and invasiveness of the test. We relied on a combination of helical CT, lung scans, and a few angios to determine who did have a P.E.. Since we all know that V-Q and CT both have false negatives, we also included 6 month phone follow-up of patients to verify that the patients called 'negative' for PE, did not have a diagnosis of DVT or PE made in the 6 months following enrollment (i.e. to reduce the liklihood that some of the patients we called negtive for P.E. actually had one at the time they were enrolled) Of note, of the 316 patients in our study who did not have PE, only 163 had both a normal d-dimer and normal dead space. Hence these tests are proposed to be used to screen for P.E, not to diagnose P.E. We have not changed our local practice yet based on these results. I am encouraged however, and hope that the study might be replicated, perhaps in an even larger population. Feel free to e.mail me if you have any questions or comments. Ed Michelson, MD