There will someday be a time when COVID is not the most likely diagnosis, and all our old decision instruments can be dusted off to risk-stratify patients. A trial that enrolled patients before COVID will once again be generalizable to patients after COVID. One day, emergency physicians will return to pre-COVID thinking; one of those pesky differentials will be tracking down acute pulmonary embolism (PE), not confounded by viral syndromes and thrombosis inflicted by the deranged coagulopathies of SARS-CoV-2 infection.

Current Procedural Considerations and Inefficiencies

Many of us have refined our use of the D-dimer assay over the past few years. Rather than a simple, immutable, dichotomous divining rod, the D-dimer has crept toward appropriate use as a continuous variable. The most prominent of these efforts have been the YEARS algorithm and PEGeD demonstrations.^1,2  Each of these studies demonstrated, in their enrolled populations, the value and safety of tweaking the D-dimer threshold. The YEARS algorithm, mixing objective findings and gestalt, safely allows the D-dimer threshold to be doubled. In a similar fashion, the PEGeD study incorporates the use of Wells criteria for pulmonary embolism as its vehicle for doubling the D-dimer threshold.

In parallel, it has been recognized the normal circulating D-dimer concentration typically increases with age. Considering this, the safe adoption of an “age-adjusted” D-dimer has been demonstrated.³ For typical assays in which the dichotomous D-dimer threshold is 500 ng/mL, this is practically applied by changing the cutoff for patients older than 50 years to Age in Years × 10.

However, it does not directly follow whether combining these two strategies remains safe. Parsing out patients deemed to be at low risk may have the effect of enriching the risk level of the remaining population such that the age-adjusted threshold no longer provides adequate negative predictive value. Then the further question remains whether the combination of these strategies reduces advanced imaging to an extent sufficient to justify a potential increased risk for missed PE.

These questions are directly addressed in a trial conducted primarily before the onset of the COVID-19 pandemic.⁴ In this prospective, multicenter trial, the authors randomized hospitals to their usual strategy, based on age-adjusted D-dimer, or to cross over to an intervention diagnostic strategy combining age adjustment with the YEARS algorithm. To be included in the trial, patients with suspected PE were those judged not to be at high risk (greater than 50 percent) but likewise unable to be excluded by the Pulmonary Embolism Rule-Out Criteria (PERC).

The answer is solely good news. In patients randomized to the intervention diagnostic strategy, only three (0.4 percent) were found to have a venous thromboembolism (VTE) of any kind during three-month follow-up. Conversely, the control strategy, with fewer patients excluded, resulted in six (0.9 percent) patients diagnosed with VTE in follow-up. No other signals of potential missed harms were observed, such as subsequent hospital admission or mortality.

Clear benefits regarding a reduction in chest imaging were observed. The control strategy, in these low- to intermediate-risk patients, resulted in 40 percent of patients being imaged for PE based on D-dimer results. The intervention strategy reduced this rate of imaging by 25 percent, down to 30.4 percent. While the trial has minor limitations as a result of its pragmatic nature and study population, it provides reasonable supporting evidence that combining the YEARS algorithm criteria with an age-adjusted threshold is safe.

News generating less enthusiasm comes along with regard to the diagnosis of subsegmental pulmonary embolism (SSPE). Generally speaking, I have been rather skeptical of new diagnoses of SSPE. With improving diagnostic resolution on modern advanced imaging, it has been suspected many SSPE are either false-positives or otherwise not clinically important. Rather, there is an idea the normal physiology of the lungs includes a small burden of thromboembolism, serving to filter clots and protect the arterial circulation. Recent CHEST guidelines have specifically addressed the uncertainty surrounding SSPE, allowing for a reasonable period of outpatient observation and investigation rather than initiation of anticoagulation.⁵

This prospective cohort study, however, indicates we ought to be concerned even for these tiny, equivocal PEs.⁶ In this multicenter study, an algorithm meant specifically to avoid anticoagulation was applied to new diagnoses of SSPE. All patients included in this study with a new diagnosis of SSPE underwent bilateral lower-extremity ultrasound at the time of diagnosis as well as one week later. If both these tests remained normal, patients were managed without anticoagulation and followed for recurrent thromboembolism within 90 days.

Unfortunately, the study was terminated early for safety by exceeding a prespecified stopping rule. After analysis of the first 292 patients out of a planned 300, the excess diagnoses of recurrent VTE triggered the end to recruitment. Overall, the cumulative incidence of recurrent VTE within 90 days was 3.1 percent, substantially greater than expected in a matched cohort. In those with single SSPE, incidence was 2.1 percent; in those with multiple SSPE, the incidence was 5.7 percent. Fortunately, no patients suffered a fatal PE during the study follow-up period.

Differences were also observed between younger and older patients. Patients younger than 65 demonstrated a recurrence rate of 1.8 percent compared to a recurrence of 5.5 percent in the older cohort. While the overall study shows a preponderance of harm to managing patients without anticoagulation, a shared decision-making conversation may still be valid in patients younger than 65.

The other interesting tidbit from this study includes those who were diagnosed with SSPE but separated from the main analysis because their lower-extremity ultrasonography demonstrated deep venous thrombosis. There were 28 patients diagnosed with lower-extremity VTE on initial or repeated ultrasonography, indicating the importance of this testing component for evaluation of those patients considered for initial management without anticoagulation.

Information Is Key

While these findings are stylistically framed as good or bad news, most important, these studies add clarity to our practice. While many have likely already been combining pretest adjusted thresholds with age-adjusted thresholds to aid in decisions regarding imaging in PE, these data solidify it as a reasonable practice. Contrariwise, we have further information regarding the risks of subsequent VTE after a diagnosis of SSPE with which to weigh against the risk of bleeding from anticoagulation. Combined, these data likely support fewer instances of advanced imaging but more initiation of anticoagulation for patients managed as outpatients with PE.

References

1. van der Hulle T, Cheung WY, Kooij S, et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet. 2017;390(10091):289-297.
2. Kearon C, de Wit K, Parpia S, et al. Diagnosis of pulmonary embolism with D-dimer adjusted to clinical probability. N Engl J Med. 2019;381(22):2125-2134.
3. Righini M, Van Es J, Den Exter PL, et al. Age-adjusted D-dimer cutoff levels to rule out pulmonary embolism: the ADJUST-PE study. JAMA. 2014;311(11):1117-1124.
4. Freund Y, Chauvin A, Jimenez S, et al. Effect of a diagnostic strategy using an elevated and age-adjusted D-dimer threshold on thromboembolic events in emergency department patients with suspected pulmonary embolism: a randomized clinical trial. JAMA. 2021;326(21):2141-2149.
5. Kearon C, Akl EA, Ornelas J, et al. Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report. Chest. 2016;149(2):315-352.
6. Le Gal G, Kovacs MJ, Bertoletti L, et al. Risk for recurrent venous thromboembolism in patients with subsegmental pulmonary embolism managed without anticoagulation: a multicenter prospective cohort study [published online ahead of print Nov. 23, 2021]. Ann Intern Med.