There have been a number of studies that have looked at using an age adjusted D-dimer for the evaluation of pulmonary embolism. In 2014 JAMA published the ADJUST-PE study. It was done in Europe (more on this later) and it supported the idea that once a patient hits 50 years old you can use AGE x 10 as the cut off for a positive d-dimer (instead of 500 µg/L). The study found that 330 out of 331 patients with an d-dimer >500 but less than their age adjusted value did not have have a PE.  For patients older than 75 it allowed them to exclude 200 of the 673 patients from further evaluation instead of just 43 patients if the 500 µg/L was used.

What about using age adjusted d-dimers for DVTs. In 2018 the PALLADIO study looked at this. The take home was that in a patient with a low pre-test probability of DVT and a negative age adjusted dimer the rate of thrombosis was <0.25% (which was actually a lot lower than a patient with a positive dimer and negative ultrasound—>around 1%).  It looks like this study was primarily done in Italy and Canada.

On a side note, because of the risk of an ultrasound having a false negative result if the if the d-dimer is positive—>  + d-Dimer and a negative Ultrasound = repeat US in one week.

Most of this conversation has been based on studies done in countries that have predominantly white populations. There have been multiple studies which have shown that African Americans have higher average d-dimer levels at baseline than white patients. It has been shown that when a person's baseline d-dimer level is high, that persons is at an increased risk of stroke and ACS. These elevated d-dimer levels also are markers of an elevated risk of thromboembolic disease. Patients with sickle cell disease also tend to have high d-dimer levels. For black patients the incidence of VTE is 30-60% higher than for whites in the United States, the mortality rates from PE are higher for black patients and black patients tend to die at younger ages from PE. The molecular causes of these increased risks are multi-factorial and probably related to higher incidence of sickle cell trait, higher baseline levels of von Willebrand factor and factor VIII (both pro-coagulants). Because of all of these factors there is no d-dimer adjustment based on race, but d-dimer still has utility in defining low risk patients.

Recently the YEARS study (which is described below in the setting of pregnancy) looked at using a binary cut off for a d-dimer of either 500 µg/L for patients with at least one of three defined risk factors (hemoptysis, unilateral leg swelling, PE most likely diagnosis), or a cut-off of 1000 µg/L for patients with none of these risk factors.  While the initial study was done in a northern European country, a validation study was done in the United States where >30% of the patients were African-American. Both studies showed promise for significantly reducing the number of CTPE scan that are ordered, while keeping the NPV of the test near 100%--but see below for why this may not mean that the YEARS algorithm is actually good to go.

• This is actually a good example of how stats can twist and turn results. In the validation study published in Academic Emergency Medicine, the NPV of the test was 99.5%, but they missed 6 out of 78 of the patients with PEs with the 1000 µg/L cut off, compared to missing 2/78 with the 500 µg/L cut-off. Honestly this confuses me. But when I sit down and think about it, one of my main take aways is that if you test a lot of low risk people, you are going to exclude most of them. That doesn't mean that the test picks up the real disease well.
• The prevalence of PE in this study was 4% (compared to 13% in the initial European study), which likely means that they considered PE in too many patients right off the bat. It would be similar to building a robot that identifies footballs, but not very well. This robot can always tell you soccer balls are not footballs, but it only tells you that 92% of footballs are actually footballs. If you just have footballs in the room, the robot will tell you that 8% of the footballs are not footballs, which isn't that good. Instead, if you fill the room with 100 footballs and 10,000 soccer balls the test will be really good at telling you what is not a football, but it still will miss 8% of the footballs when it happens upon one. I think the validation study just put in too many "soccer balls" into the study, which makes the numbers look better than they are.
•  One more quick point is that you should not combine age adjusted d-dimer with the YEARS--that has been shown to lead to more missed clots without limiting imaging to a significant extent.

And what about pregnant patients? Two respected physicians (Jeffrey A. Kline and Christopher Kabrhel) with expertise in emergent thrombosis care recommend the following steps for evaluating pregnant patient for a PE:
#1—Bilateral Lower Extremity Ultrasound
#2—Are they PERC negative except for being pregnant
#3—Are they free of any high risk features (ex: SOB with a big swollen leg)
#4—Get a D-dimer using the following adjusted thresholds

First trimester <750 ng/mL
Second trimester <1000 ng/mL
Third trimester <1250 ng/mL
If all of these are negative then there is a very low chance of PE, or as they call it, "PE can be ruled out to a reasonable degree of medical certainty.” This is expert opinion and not based on results of a study.

In 2018 a study in the Annals of Internal Medicine used the following algorithm for the eval of PE in pregnant patients, which picked up all the clots in their cohort of >400 patients:
#1: Revised Geneva Score for pre-test risk stratification
—>if low pre-test risk then D-dimer. If dimer negative (<500 ng/ml) you are done. If dimer is positive—>BLE US—> if positive treat, if negative get a CTPE.
—>If moderate or high risk pre-test risk then you go straight to BLE US—>if positive treat, if negative get a CTPE.
One note on the d-dimer utility in this study: Using the d-dimer cut off of 500 ng/ml only allowed them to exclude 25% of 1st trimester, 11% of 2nd trimester and 4% of 3rd trimester low pre-test risk patients.

Finally, in 2019 the YEARS study gave an evidenced based algorithm for evaluating pregnant patients for PE.  The algorithm is not complex but has a number of boxes, so I have attached a link to the algorithm below.  It basically is that in a very low risk patient you can use a d-dimer cutoff of 1000 ng/mL and in a patient with risk factors you still need to use the cutoff of 500 ng/mL. With this protocol 476 of 477 patients described as low risk did not develop at PE or DVT at 3 month f/u.  I think that this is the cleanest recommendation for using D-Dimers in pregnant patients. 

https://www.mdcalc.com/years-algorithm-pulmonary-embolism-pe

(Finally, if your not tired of reading at this point: there was a study called the DiPEP study, where they looked at a bunch of pregnant women and tried to correlate a number of biomarkers, including d-dimer, with pregnant patients who had DVT/PE. This could have provided great answers to the questions of what is an appropriate adjusted d-dimer in pregnant patients . . . but they drew that labs on most patients with clots after they were anticoagualted, which seems to have skewed most of their data—bummer.)

Take aways: Age adjusted D-Dimer is reasonable. As a group African Americans have higher d-dimer levels and higher risks of clotting, but d-dimer cutoffs are still reasonable.  There are varying recommendations for the use of a variable d-dimer cutoff level in pregnancy and the YEARS Algorithm has the cleanest data to back it up (but has not been properly validated).


Hunt BJ, Parmar K, Horspool K, et al. The DiPEP ( Diagnosis of PE in Pregnancy ) biomarker study : An observational cohort study augmented with additional cases to determine the diagnostic utility of biomarkers for suspected venous thromboembolism during pregnancy and puerperium. 2018;(2015):694-704. doi:10.1111/bjh.15102
Righini M, Robert-ebadi H, Elias A, Sanchez O, Moigne E Le. A Multicenter Prospective

Jeffrey A. Kline,MD*and Christopher Kabrhel,MD,MPH†TManagement Outcome Study Annals of Internal Medicine Diagnosis of Pulmonary Embolism During Pregnancy. 2019;(15). doi:10.7326/M18-1670
EMERGENCY EVALUATION FOR PULMONARY EMBOLISM, PART 2:DIAGNOSTIC APPROACH. The Journal of Emergency Medicine, Vol.-, No.-, pp. 1–14, 2015

Laura M. Raffield, Alex P. Reiner, et al. D-Dimer in African Americans Whole Genome Sequence Analysis and Relationship to Cardiovascular Disease Risk in the Jackson Heart Study. NHLBI Trans-Omics for Precision Medicine (TOPMed) Consortium, Hematology & Hemostasis TOPMed Working Group. Arterioscler Thromb Vasc Biol, November 2017

Liselotte M. van der Pol, M.D., Menno V. Huisman, M.D., et al. Pregnancy-Adapted YEARS Algorithm for Diagnosis of Suspected Pulmonary Embolism. N Engl J Med 2019; March 21, 2019. 380:1139-1149

Tyler W. Buckner, Nigel S. Key. Venous Thrombosis in Blacks. Originally published14 Feb 2012.Circulation. 2012;125:837–839

Bates SM, Jaeschke R, Stevens SM, et al. Diagnosis of DVT: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012;141:e351S-e418S.

Venous Thromboembolic Diseases: The Management of Venous Thromboembolic Diseases and the Role of Thrombophilia Testing [Internet]. NICE Clinical Guidelines, No. 144. National Clinical Guideline Centre (UK). London: Royal College of Physicians (UK); 2012 Jun.

N. Riva, W. Ageno, et al. Age‐adjusted D‐dimer to rule out deep vein thrombosis: findings from the PALLADIO algorithm. The PALLADIO Study Investigators. First published: 10 November 2017. https://doi.org/10.1111/jth.13905

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. doi:10.1001/jama.2014.2135

Van der Hulle, TomFogteloo, Anne J et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. The Lancet, Volume 390, Issue 10091, 289 - 297