The YEARS Algorithm for Evaluation of PE
Washington University Emergency Medicine Journal Club– February 17th, 2022
You are coming onto a shift in EM-2 one Tuesday afternoon following conference and get a patient, Mr. X, signed out to you by the off-going attending. Mr. X is a 51 year-old gentleman with a history of hypertension who presents with shortness of breath. He was reportedly sitting at his desk at work when he suddenly felt short of breath and felt like his heart was racing. He did not have any chest pain or hemoptysis and has not had any leg swelling or calf pain. He now feels completely back to normal. He endorses some recent stress (at work), denies surgery, long plane trips, or car rides, and denies personal or family history of venous thrombosis. He takes amlodipine for his hypertension but takes no other medications.
His vitals on arrival were:
HR: 89 BP: 145/79 O2 saturation: 97% Temp: 36.9 RR: 18
The remainder of physical exam was unremarkable with no signs of DVT. His ECG revealed normal sinus rhythm, rate 91, with no ST segment or T-wave changes, and no S1Q3T3. The previous attending noted low suspicion for a pulmonary embolism, but also noted that the patient was not PERC negative due to his age. She ordered basic labs, cardiac enzymes, and d-dimer. His labs are unremarkable aside from a d-dimer of 700 ng/mL. Given the abnormal age-adjusted d-dimer you order a CT angiogram to rule-out PE, which comes back negative.
Your new attending suggests that given his low-risk probability you could have applied the YEARS algorithm, which would have allowed you to forego the CT scan given that his d-dimer was < 1000 ng/mL. You go online and find a recent REBELEM post on the topic, but decide later to do a deep dive into the literature to understand more about this new algorithm…
Population: Adults patients in whom there is clinical concern for a pulmonary embolism
Intervention: Use of the YEARS algorithm to rule-out PE
Comparison: Traditional clinical decision rules (Well’s score, modified Geneva score) or clinical gestalt, in conjunction with D-dimer testing
Outcome: The miss rate for PE, mortality, proportion of patients undergoing CT scan or V/Q scan on the initial ED visit
PubMed was searched using the terms “(“years algorithm” or “years criteria” or “years score”) AND (“pulmonary embolism” or “PE”)” (https://tinyurl.com/2p8hk64j). This resulted in 38 citations from which the 4 most relevant articles were chosen.
Article 1: 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 Dec 7;326(21):2141-2149. Answer Key.
Article 2: Kabrhel C, Van Hylckama Vlieg A, Muzikanski A, et al. Multicenter Evaluation of the YEARS Criteria in Emergency Department Patients Evaluated for Pulmonary Embolism. Acad Emerg Med. 2018 Sep;25(9):987-994. Answer Key.
Article 3: Eddy M, Robert-Ebadi H, Richardson L, et al. External validation of the YEARS diagnostic algorithm for suspected pulmonary embolism. J Thromb Haemost. 2020 Dec;18(12):3289-3295. Answer Key.
Article 4: van der Hulle T, Cheung WY, Kooij S, et al; YEARS study group. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet. 2017 Jul 15;390(10091):289-297. Answer Key.
A modified clinical decision rule for the evaluation of pulmonary embolism (PE) using three of the items in the Well’s PE score and an adjusted D-dimer threshold based on the number of items positive was initially reported in 2015 (van Es 2015). This algorithm, termed the YEARS algorithm, begins with the following 3 questions:
- Clinical signs of DVT?
- PE most likely diagnosis?
If the answer to all of these questions is no (0 YEARS criteria), a D-dimer threshold of < 1000 ng/mL is used to rule-out PE without the use of advanced imaging (CTA or V/Q). If the answer to 1 or more of these questions is yes (≥ 1 YEARS criteria), a D-dimer threshold of < 500 ng/mL is used to exclude PE. In the initial derivation and validation cohorts, this algorithm had a false negative rate of 1.2% and 1.9%, respectively, and PE was excluded without CT in 36% and 46% of patients.
This YEARS algorithm has now been validated in several large, multi-center studies. A prospective, multi-center study from the Netherlands demonstrated a false negative rate for the algorithm of 0.61% (95% CI 0.36 to 0.96%) (van der Hulle 2017). Of the 18 missed cases of PE, a CT was performed on the index visit in 11 cases; the miss rate among patients managed without a CT was only 0.43% (95% CI 0.17 to 0.88%). In a similar prospective observational study conducted at 15 EDs in the US, the algorithm had a miss rate of 0.5% (95% CI 0.2 to 1.1%) (Kabrhel 2018). It should be noted that patients ruled out clinically (i.e. negative PERC criteria) and those at high risk (Well’s score > 6) were excluded, leaving a low to moderate risk group of patients and some risk of spectrum bias.
A retrospective analysis of data collected for 3 prospective studies in Belgium, France, and Switzerland demonstrated a false negative rate of 1.2% (95% CI 0.8 to 1.9%) (Eddy 2020). In this study, all 17 missed cases had 0 YEARS criteria and a D-dimer < 1000 ng/mL; a modified algorithm using an age-adjusted D-dimer in this group lowered the false negative rate to 0.0% (95% CI 0.0 to 0.4%).
Finally, there has been a single randomized study conducted to evaluate the potential clinical impact of the YEARS algorithm (Freund 2021). In this cluster-randomized, crossover study, 16 EDs in France and Spain were randomized in a 1:1 fashion to either a control strategy (using an age-adjusted D-dimer threshold for all patients) or the intervention strategy (YEARS score of 0 with a D-dimer threshold of 1000 ng/mL or YEARS score ≥ 1 with an age-adjusted D-dimer) for the initial 4 months. Following a 2-month washout, each ED then crossed over to the alternate strategy for 4 months. The failure rate was 0.15% (95% CI 0.00 to 0.86%) in the intervention group and 0.80% (95% CI 0.26 to 1.86%) in the control group and chest imaging was performed less frequently in the intervention group (30.4%) than the control group (40.0%): adjusted difference = -8.7%, 95% CI -13.8% to -3.5%).
The current evidence suggests the YEARS algorithm is effective at excluding a diagnosis of PE with a low miss rate and has the potential to significantly decrease the use of CT when PE is being considered. This clinical decision tool has not only been validated in multiple settings, but has undergone an impact analysis showing that it changes behavior and decreases cost and radiation exposure without signs of significant harm (Level 1 evidence). While clinicians may wish to avoid its use in patients felt to be truly high risk (as in the study by Kabrhel et al), there should remain a large patient population in whom its use will be of benefit.