D-Dimer focused testing for PE

Jul 23, 2024

Here is a common scenario: The resident presents a case of a 58 yo patient who has presented with shortness of breath. They go over the examination, the ECG findings and then say ".... and the D-Dimer was...." Your next question may be, "Why did you do a D-dimer? Were you suspicious for a pulmonary embolism(PE)? What's the Well's Score, what's your pretest probability?........" (15 minutes reading)

Does just doing a D-dimer on its own help make or exclude the diagnosis and does it result in ...

Become an EM Lectures member

and have access to all the content

_{Already registered?}_{Log in here.}

Find Out More and Become a Member

Does just doing a D-dimer on its own help make or exclude the diagnosis and does it result in less imaging?

This study(1) looked at a D-Dimer-only pathway (ie., no clinical prediction rules), with the objective of reducing the number of patients receiving imaging.

What They Did

This was a prospective, multicenter implementation study, conducted in three public Canadian Emergency Departments. One of the three departments was used as the control.

The PE testing pathway used, consisted of D-dimer testing. If the D-dimer level was > 500 ng/ mL, a CTPA or VQ scan was performed. If the D-dimer level was less than 500 ng/mL, PE was excluded. Clinicians were permitted to make alternative diagnoses and not perform imaging, even if the d-dimer was high.

Primary Outcome: This included adherence to the D-dimer only testing pathway and the proportion of patients tested for PE who underwent imaging.

Secondary Outcomes:

PE prevalence among all patients.
The proportion of PE scans that were positive
The proportion of imaging ordered without D-dimer or with a negative D-dimer.
The proportion of imaging not ordered when the D-dimer was positive.
Length of stay in the ED.
Time between physician assessment and leaving the ED.

The primary safety outcome was patients who tested negative for PE, who then had a PE diagnosed in the next 30 days.

N = 10,354

What Did They Find?

10,354 patients were tested for PE

3,852 had imaging
PE was diagnosed in 494 (4.8%) of patients tested.

Primary Outcomes

There was significantly increased adherence to the pathway in the intervention sites and as would be expected an increase in the numbers of patients who had D-dimer testing ie., from 6.5% to 10.6%.
Although the proportion of patients tested for PE who underwent imaging was decreased, there was no association with the D-dimer testing only pathway.
Of the 5,153 patients who tested negative for PE on D-dimer testing, only 2 (0.04%) tested positive for PE in the next 30 days. In comparison, at the control site 3 of 2,886 (0.1%) who tested negative were found to have a PE in the next 30 days.

Secondary Outcomes

There was a significant association between introduction of the D-dimer-only PE testing pathway and increased diagnoses of PE in the overall population.
There was an increased positive yield of pulmonary imaging (aOR 4.89; 95% CI 1.17 to 20.53) at the intervention site.
There was a significant reduction in patients having PE imaging following a negative D-dimer result.
There was no effect on the length of stay in the ED, or the length of stay after being assessed by a physician.
At the control site, there was a significant reduction in the length of stay in the ED.

The authors' conclusion was:

"In this Canadian ED study, the uptake of a D-dimer- only PE testing strategy was sustained and high among emergency physicians. Implementation was associated with higher imaging yield, and a D-dimer result of <500 ng/mL safely excluded PE."

My Take on This

This study tells me a few things:

D-dimer testing alone is safe, with only a 0.04% miss rate.
In 96 cases in the study, there was deviation from the protocol, where imaging was performed without D-dimer testing. We are unsure how D-dimer would have performed. Our assumption is that these were considered to be high risk cases and the prevalence of PE in these patients was nearly 22% higher than those of the general study population.
There may be confounding involved as this is a before-after comparison.
There were some differences between the intervention and control sites in terms of patient age.
There needs to be external validation, especially as the rates of PE were low in this study.

Will This study change my practice?

Here are some questions I ask:

Will I perform a D-dimer if I have a high pre-test probability for a PE and then not image if that's normal. Probably not.
Will I age adjust? Yes.
Did this study result in less imaging? Not really.

It won't affect my practice, but it will make me feel better about using the YEARS Study results and knowing that the miss rate with a normal D-dimer is low.

References

Germini F et al. Implementation, Clinical Benefit and Safety of a D-Dimer-Focused Pulmonary Embolism testing Pathway in the Emergency Department. Ann Emerg Med. 2024 Apr 23:S0196-0644(24)00156-2. doi: 10.1016/j.annemergmed.2024.03.010. Epub ahead of print. PMID: 38661619.