U.S. Emergency Departments Over-Test for Suspected PE

By Marilynn Larkin

February 11, 2020

NEW YORK (Reuters Health) - Overuse of computerized tomographic pulmonary angiography (CTPA) to rule out pulmonary embolism (PE) in U.S. emergency departments is costly and puts patients at risk for multiple adverse outcomes, a cross-sectional analysis suggests.

"The primary issue is, physicians in the emergency department have a duty to provide the fastest service they can to the patients in front of them," Dr. Jeffrey Kline of the Indiana University School of Medicine in Indianapolis told Reuters Health by email. "They perceive that too often, if they order a D-dimer (test to rule out thrombosis), it is going to be positive, therefore wasting time, and 'I'll have to order a CT scan anyway.' These data shows that this is actually a fallacy."

"If they're more willing to order a D-dimer, they can reduce CT scanning, and do their patients a benefit by reducing unnecessary CT scanning," he said. "In the long run, more use of D-dimer saves time, as well as radiation from low-value CT scanning."

Using health records and billing data for 27 EDs in Indiana and Texas, from 2016 to 2019, his team identified patients who had undergone D-dimer tests, CTPA, ventilation perfusion lung scanning or pulmonary angiography.

As reported in Circulation: Cardiovascular Quality and Outcomes, from among more than 1.8 million patients, 97,125 (5.3%) had a diagnostic test for PE, including 25,870 with a CTPA ordered without D-dimer.

Dr. Kline told Reuters Health that 59% of all CTs done for PE were done without D-dimer tests.

In Indiana, 30% of patients who underwent CTPA also had a D-dimer performed on the same ED visit. Rates of PE diagnosis were 1.8% in patients with CTPA and D-dimer, and 1.1% with CTPA without D-dimer.

In Dallas-Fort Worth, patients were more likely to receive D-dimer testing before CTPA (52%). The rate of PE diagnosis after CTPA was also higher in the Dallas-Fort Worth sample at 4.8% (versus 1.3% in Indiana), suggesting, the authors say, a possible cause-effect association (high D-dimer use corresponding to higher PE yield). Indeed, linear regression analysis supported this possibility.

In Indiana, 13.1% of those tested for PE had at least one repeat visit; 6.8% of patients who had CTPA later had a repeat CTPA. Of those who had a repeat CTPA, 2% had PE diagnosed on the repeat visit, but only two of the diagnoses were new.

Similarly, in Dallas-Fort Worth, 13.8% of patients tested for PE had at least one repeat visit, and 11% of patients who had CTPA had a repeat CTPA. Although 4.7% had PE diagnosed on the repeat visit, only about 10% were new PE diagnoses.

The risk of cancer from exposure to medical radiation may be the highest in women under age 45 years of age because of the risk of radiation exposure to the breast, the authors note. In this study, 59% of CTPAs done were in women; 21% were performed on women under age 45, and 1.4% on children. In Indiana, where the yield rate for PE was the lowest, only 31% of females under age 45 had D-dimer testing before CTPA.

"These data suggest the potential for implementation of D-dimer based protocols to reduce low-yield CTPA ordering," the authors conclude

Dr. Ronald Goldenberg, a pulmonologist at NYU Langone Health in New York City, commented by email to Reuters Health, "I agree that we overtest for PEs and we get too many CT scans. One reason we do that is that PEs are potentially fatal and there are no good noninvasive tests to rule out PEs."

"Several validated scoring systems are used, including Well's criteria, the Modified Geneva Score, and the PERC rule," he said. "For patients that are low risk with these scoring systems, (we) use D-dimers as an add-on to help avoid unnecessary testing. The article failed to have any scoring system calculated."

"D-dimers alone are not useful for ruling in or out PEs, but are often used in conjunction with scoring systems or lower extremity dopplers," he said.

SOURCE: http://bit.ly/2SeUsUy Circulation: Cardiovascular Quality and Outcomes, online January 20, 2020.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: