Conference Coverage

Algorithm for suspected pulmonary embolism safely cut CT rate


 

ROME– A newly validated, simplified algorithm for the management of patients with suspected acute pulmonary embolism enables physicians to safely exclude the disorder in roughly half of patients without resorting to CT pulmonary angiography, Tom van der Hulle, MD, reported at the annual congress of the European Society of Cardiology.

“This is the largest study ever performed in the diagnostic management of suspected pulmonary embolism. Based on our results, I think the YEARS algorithm is ready to be used in daily clinical practice,” declared Dr. van der Hulle of the department of thrombosis and hemostasis at Leiden (the Netherlands) University Medical Center.

Bruce Jancin/Frontline Medical News

Dr. Tom van der Hulle

The YEARS prospective algorithm validation study included 2,944 consecutive patients, mean age 53 years, with suspected acute pulmonary embolism (PE) at 12 Dutch academic and nonacademic hospitals. All were managed according to the YEARS algorithm. Investigators then went back and reanalyzed their data as though participants had been managed according to the standard, guideline-recommended Wells rule in order to see how utilization of CT differed.

Using the YEARS algorithm, PE was reliably ruled out without need for CT pulmonary angiography – considered the standard in the diagnosis of PE – in 48% of patients. In contrast, adherence to the Wells rule would have meant that 62% of patients would have gotten a CT scan to rule out PE with a comparably high degree of accuracy.

But that 62% figure underestimates the actual CT rate in clinical practice. The reality is that although the guideline-recommended Wells rule and revised Geneva score have been shown to be safe and accurate, they are so complex, cumbersome, and out of sync with the flow of routine clinical practice that many physicians skip the algorithms and go straight to CT, Dr. van der Hulle said. This approach results in many unnecessary CTs, needlessly exposing patients to the risks of radiation and intravenous contrast material while driving up health care costs, he added.

Using the Wells rule or revised Geneva score, the patient evaluation begins with an assessment of the clinical probability of PE based upon a risk score involving seven or eight factors. Only patients with a low or intermediate clinical probability of PE get a D-dimer test; those with a high clinical probability go straight to CT.

The YEARS algorithm is much simpler than that, Dr. van der Hulle explained. Everyone who presents with suspected acute PE gets a D-dimer test while the physician simultaneously applies a brief, three-item clinical prediction rule. These three items were selected by the Dutch investigators because they were the three strongest predictors of PE out of the original seven in the Wells rule. They are hemoptysis, clinical signs of deep vein thrombosis such as leg swelling or hyperpigmentation, and the clinician’s global impression of PE as being the most likely diagnosis.

In the YEARS algorithm, the threshold for a positive D-dimer test warranting CT pulmonary angiography depends upon whether any of the three clinical predictors is present. If none is present, the threshold is 1,000 ng/mL or above; if one or more is present, the threshold for a positive D-dimer test drops to 500 ng/mL.

Using these criteria, PE was excluded without resort to CT in 1,306 patients with none of the three YEARS items and a D-dimer test result below 1,000 ng/mL, as well as in another 327 patients with one or more YEARS items present but a D-dimer below 500 ng/mL. Those two groups were left untreated and followed prospectively for 3 months.

The 964 patients with one or more YEARS predictors present and a D-dimer score of at least 500 ng/mL underwent CT imaging, as did the 352 with no YEARS items and a D-dimer of at least 1,000 ng/mL.

The prevalence of CT-confirmed PE in the study was 13.2%. Affected patients were treated with anticoagulants.

The primary study endpoint was the total rate of deep vein thrombosis during 3 months of follow-up after PE had been excluded. The rate was 0.61%, including a fatal PE rate of 0.20%. The rate in patients managed without CT was 0.43%, including a 0.12% rate of fatal PE. In patients managed with diagnostic CT, the deep vein thrombosis rate was 0.84%, with a fatal PE rate of 0.30%.

“I think these results are completely comparable to those in previous studies using the standard algorithms,” Dr. van der Hulle commented.

The study’s main limitation is that it wasn’t a randomized, controlled trial. But given the tiny event rates, detecting any small differences between management strategies would require an unrealistically huge sample size, he added.

Asked if he thinks physicians will actually use the new tool, Dr. van der Hulle replied that some physicians feel driven to be 100% sure that a patient doesn’t have PE, and they will probably keep overordering CT scans. But others will embrace the YEARS algorithm because it reduces wasted resources and minimizes radiation exposure, a particularly compelling consideration in young female patients.

Discussant Marion Delcroix, MD, had reservations. She said she appreciated the appeal of a simple algorithm, but she asked, “Couldn’t we do better with a bit more sophistication, perhaps by adjusting the D-dimer cutoff for age and also adding some other items, like oxygen saturation and estrogen use?

“My concern is about the applicability. The age of the study cohort is relatively young, at a mean of 53 years. The peak age of PE in a very large contemporary German database is 70-80 years. We don’t know if the YEARS score is any good in this older population,” asserted Dr. Delcroix, professor of medicine and respiratory physiology and head of the center for pulmonary vascular diseases at University Hospital in Leuven, Belgium.

“If the aim is to decrease the number of CT pulmonary angiograms for safety reasons, why not reintroduce compression ultrasound of the lower limbs in the diagnostic algorithm?” she continued. “It has been shown to effectively reduce the need for further imaging.”

Dr. Delcroix predicted that the YEARS algorithm study will prove “too optimistic” regarding the number of CT scans avoided, particularly in elderly patients.

The YEARS study was funded by the trial’s 12 participating Dutch hospitals. Dr. van der Hulle reported having no financial conflicts of interest.

Next Article: