Washington University Emergency Medicine Journal Club – September 2025
Dr. Brian Cohn
At this month’s journal club we will be discussing a variety of topics involving the diagnostic evaluation of PE, including the use of d-dimer in high-risk patients, pre-test clinical probabilit scores (i.e. Well’s and Geneva scores), and algorithms involving the YEARS criteria.
The PGY-1 paper will be analyzed using the Therapy Form
The PGY-2 paper will be analyzed using the Diagnostic Test form.
The PGY-3 paper will be analyzed using the Meta-Analysis form.
The PGY-4 paper will be analyzed using the Economic Analysis form.
Vignette
You are working in the D pod one afternoon when you encounter Mrs. X, a pleasant 65-year old woman with a history of hypertension and osteoporosis, who is in town visiting her grandchildren from California. She flew in 2 days earlier, and for the last 12 hours has noted some right-sided, pleuritic chest pain. She thinks she pulled a muscle picking up her 3-year old grandson, but was worried and wanted to be evaluated.
Her physical exam is unremarkable, including a heart rate of 70, a normal oxygen saturation on room air, and a normal respiratory rate. Her lungs are clear and she does seem to have pain with deep inspiration. There is no chest wall tenderness, no LE swelling, no LE cords, and no calf tenderness.
Given her recent plane trip you are concerned about a possible PE. Having read a previous journal club on reducing PE protocol CT ordering rates, you calculate her Well’s score and find that she is low risk, so order a D-dimer in addition to an ECG and chest x-ray. The ECG and CXR are normal, but the D-dimer is elevated at 600 mcg/mL FEU. Knowing that her D-dimer is below the age-adjusted cutoff, you feel confident that you have excluded PE.
This gets you thinking more about the various ways in which it possible to exclude PE without advanced imaging, including the YEARS algorithm, the use of d-dimer in high risk patients, and clinical pathways such as the one posted in Agile MD. You decide to search the literature to see what else you could be doing to limit CT scans in the evaluation of this diagnosis…
PICO Question
Population: Adult patients with suspected PE presenting to the emergency
department
Intervention: Multiple algorithms including use of the YEARS criteria with age
adjusted D-dimer, the value of D-dimer in patients with a high pretest probability,
and the use of clinical predictions rules such as Well’s score, revised Geneva score,
and PERC rule
Comparison: Use of standard work-up with D-dimer and CTPA
Outcome: Diagnostic accuracy, proportion
Article 1: Freund Y, Chauvin A, Jimenez S, Philippon AL, Curac S, Fémy F, Gorlicki J, Chouihed T, Goulet H, Montassier E, Dumont M, Lozano Polo L, Le Borgne P, Khellaf M, Bouzid D, Raynal PA, Abdessaied N, Laribi S, Guenezan J, Ganansia O, Bloom B, Miró O, Cachanado M, Simon T. 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 Dec 7;326(21):2141-2149. doi: 10.1001/jama.2021.20750. PMID: 34874418; PMCID: PMC8652602. [Answer Key]
Article 2: Bannelier H, Kapfer T, Roussel M, Freund Y, Alame K, Catoire P, Vromant A. Failure rate of D-dimer testing in patients with high clinical probability of pulmonary embolism: Ancillary analysis of three European studies. Acad Emerg Med. 2025 Feb;32(2):116-122. doi: 10.1111/acem.15032. Epub 2024 Nov 1. PMID: 39487597; PMCID: PMC11816009. [Answer Key]
Article 3: Etemadi A, Hosseini M, Rafiee H, Mahboubi A, Mahmoodi T, Kuno T, Jenab Y, Raphael CE, Aronow WS, Hosseini K, Giri J. Comparative diagnostic accuracy of pre-test clinical probability scores for the risk stratification of patients with suspected pulmonary embolism: a systematic review and Bayesian network meta-analysis. BMC Pulm Med. 2025 Apr 8;25(1):162. doi: 10.1186/s12890-025-03637-6. PMID: 40200307; PMCID: PMC11980127. [Answer Key]
Article 4: Nze Ossima A, Ngaleu Siaha BF, Mimouni M, Mezaour N, Darlington M, Berard L, Cachanado M, Simon T, Freund Y, Durand-Zaleski I. Cost-effectiveness of modified diagnostic strategy to safely rule-out pulmonary embolism in the emergency department: a non-inferiority cluster crossover randomized trial (MODIGLIA-NI). BMC Emerg Med. 2023 Nov 29;23(1):140. doi: 10.1186/s12873-023-00910-x. PMID: 38030975; PMCID: PMC10687836. [Answer Key]
Bottom Line
Diagnostic reasoning in the emergency department is frequently follows a process of Bayesian reasoning, in which disease probability is adjusted as new information is obtained until a threshold is reached at which a diagnosis is felt to be either excluded or confirmed. In this model, we begin with a pre-test probability of disease followed by acquisition of new information from history, physical examination, clinical decision rules, or diagnostic testing. This new information results in an adjustment of our level of concern for disease, leading to a new post-test probability.
This then leads to one of three outcomes: 1) the probability of disease is low enough to preclude the need for further testing, i.e. below the test threshold, 2) the probability of disease remains in a range at which further information is needed to further clarify the probability of disease, or 3) the probability of disease is high enough to necessitate the initiation of treatment (i.e. above the treatment threshold (Pauker-Kassirer 1980).
Prior knowledge → New information → Updated belief
Pre-test probability → New information → Post-test probability
The evaluation of PE has become the exemplar of this process, as various clinical decisions rules and diagnostic algorithms have arisen over the last three decades to assist clinicians. Since D-dimer testing’s early proposal as a diagnostic test for venous thrombosis in the 1980s (Rowbotham 1987), our understanding of the role of this assay in the evaluation PE has evolved, to include an age-adjusted threshold for patients over age 50 years of age and the use of the YEARS criteria to allow for an even higher threshold in those at lower risk of disease. Furthermore, multiple clinical decision rules have been studies to allow for risk stratification of patients to identify those at low enough risk for use of D-dimer testing (Well’s score and revised Geneva score [RGS]), and the development of the PERC rule to identify those whose risk is low enough that even D-dimer testing is unnecessary to exclude a diagnosis of PE (Kline 2008). We sought to explore these various rules and algorithms to gain a better understanding of their strengths and limitations.
Once the decision has been made to evaluate for PE in the ED, assessment often begins with risk-stratification via either the Well’s score or RGS. A recent meta-analysis of these pre-test clinical probability scores (pCPS) found that both scores have moderate negative likelihood ratios in terms of assigning patients to what the authors referred to as “PE-unlikely” (LR- 0.34 for Well’s [95% CI 0.25 to 0.45 ] and 0.39 for RGS [95% CI 0.27 to 0.58]). While these negative likelihood ratios would not be sufficient to exclude PE in all but the lowest pre-test probability patients, these pCPS were not devised to rule-out disease, but rather to risk stratify. Considering theoretical disease prevalence of 10%, for example, a low-risk categorization using these rules would result in post-test probabilities of 3% and 4%, respectively. The pooled negative LR for the PERC rule was 0.3 (95% CI 0.1 to 1.15); a patient stratified as low-risk by either Well’s or RGS with no PERC criteria would therefore have a post-test probability of 0.9% and 1.2%, both well-below the often-cited test threshold of 1.8% for PE (Kline 2004).
While D-dimer testing has been used in low to moderate-risk patients for many years, one recent study evaluated its use in high-risk patients. This post hoc analysis of data from three European studies (PROPER, MODIGLIANI, and TRYSPEED) patients with a high clinical probability of PE, defined as Well’s >6 or RGS >10, and a D-dimer measurement in the ED. Out of seventy patients with D-dimer levels below the age-adjusted threshold, none of them had a PE on CTPA or at three-month follow-up, indicating a failure rate of 0.0% (95% CI 0.0%–6.5%). This finding is limited by the wide confidence interval, which allowed for a possible miss rate of 6.5%, and the relatively low overall prevalence of PE in this study of only 31.3%. This is lower than the high-risk prevalence of 37.5% to 43% in validation studies of the Wells criteria (Wells 2001, Wolf 2004), and significantly lower than high-risk patients in studies on the modified Geneva score (Ceriani 2010) where the prevalence was 71%. Previous studies suggest the negative LR for an age-adjusted D-dimer is around 0.06. Using this value and a pre-test probability of 40% for high-risk patients, the post-test probability of PE would be 3.8%, which remains above the test threshold and suggests the need for additional testing (i.e. chest imaging).
YEARS Criteria
- Clinical signs of DVT
- Hemoptysis
- PE most likely diagnosis
The YEARS criteria were first proposed in 2015 to identify patients with very low risk of PE (van Es 2015). When these three components of the Well’s score are all absent (in non-pregnant patients) there is ample evidence to allow for the use of a higher D-dimer threshold (1000 μg/L) to rule-out disease and obviate the need for further testing. The MODIGLIA-NI study compared this diagnostic strategy with the routine use of an age-adjusted D-dimer. This cluster-randomized, crossover, noninferiority trial was conducted in 18 EDs from France (n = 16) and Spain (n = 2). Patients with clinical suspicion of PE and either a low subjective probability (< 15%) with one or more PERC score elements or an intermediate subjective probability of PE (16-50%) were randomized to either an intervention strategy using the YEARS algorithm or a control strategy using an age-adjusted threshold alone. The failure rate with these strategies was 0.15% (95% CI, 0.00% to 0.86%) for the intervention group (n = 1) and 0.80% (95% CI, 0.26% to 1.86%) for the control group (n = 5), for an adjusted difference of −0.64% (1-sided 97.5% CI, −∞ to 0.21%, the upper bound of which is less than the noninferiority margin of 1.35%). Chest imaging was performed less often in the intervention group (30.4%) compared to the control group (40.0%); adjusted difference -8.7%, 95% CI -13.8% to -3.5%. The median ED length of stay was lower in the intervention group (6.0 hours) compared to the control group (6.0 hours); adjusted difference -1.6 hours, 95% CI -2.4 to -0.9. A subsequent economic analysis of MODIGLIA-NI found that the incremental cost-effectiveness ratio (ICER) was 7,142 indicating that each undetected VTE averted in the intervention group is associated with cost savings of €7,142 in comparison with the control group.