Hi George, I am wondering if anyone is using the age-related D-dimer cutoff for the emergency room based on the ADJUST-PE study? We use Innovance D-dimer reagents from Siemens on the BCS XP analyzer. Our ER physicians are very impatient for us to allow them to use this age-adjusted cut-off, but we have reservations due to CAP and CLIA regulations. Thanks, Sylvia Stacy, Concord Hospital, Concord, NH.
Hello, Sylvia, and thank you for your question. Judging by the number of recent publications, the use of the age-adjusted D-dimer cutoff to rule out pulmonary embolism (PE) — defined as age X 10 for patients over 50, and reported as ng/mL or ug/L Fibrinogen Equivalent Units–is gathering momentum. The article that reports the ADJUST-PE results is Righini M, Van Es J, Den Exter P, et al. Age-adjusted D-dimer cutoff levels to rule out pulmonary embolism. JAMA 2014;311:1117–24, and is attached below. Another compelling article that generalizes age-related cutoffs to all venous thromboembolic disease is Lippi G, Favaloro EJ, Cervellin G. A review of the value of D-dimer testing for prediction of recurrent venous thromboembolism with increasing age. Semin Thrombos Hemostas 2014;40:634–9.
I’ll be interested in learning from our participants whether they are currently using age-adjusted cutoffs to rule out PE. I read one criticism of the Righini study whose author suggested the patients who were ruled out on the basis of the age-related cutoff were not confirmed as PE-negative using tomography. I would add that there is considerable variation in the way D-dimer results are reported, generating some confusion about reporting units and laboratory-generated cutoffs. The study reported variation in the outcomes based on the use of various D-dimer testing platforms. Nevertheless, if the age-related cutoff were judiciously applied, a number of patients could be spared the effort of tomography and perhaps unnecessary therapy.
Here is the JAMA report: joi140021
I’ve a similar request from
I’ve a similar request from our emergency room physicians as well. We are also using the Innovance D-dimer reagents from Siemens on the BCS XP analyzer. I did find a white paper online from Siemens addressing this:
I’m not sure that it would be enough to satisfy CAP and CLIA requirements if it not included in the package insert though:
CAP Checklist HEM.37925: D-dimer–Exclusion of VTE: If a D-dimer method is used in the exclusion of venous thromboembolism, the method is valid for this purpose.
The FDA has two levels of clearance for D-dimer methods to evaluate venous thromboembolism (VTE). The level of FDA approval/clearance can be found in the manufacturer’s package insert under Intended Use. FDA-cleared/approved tests for exclusion of VTE can be used to exclude VTE with non-high pretest probability (i.e. low or low/moderate pretest probability; see manufacturer’s instructions for appropriate pretest categories). In an exclusion strategy, a D-dimer below the established threshold in a non-high pretest probability patient does not require further testing to exclude VTE.
FDA-cleared/approved tests for use as an aid in the diagnosis of VTE do not require determination of pretest probability, but the test results must be used with one or more additional tests (usually objective imaging studies) to evaluate VTE. Therefore, FDA tests cleared/approved as an aid in diagnosis should not be used in an exclusion strategy with pretest probability alone.
For D-dimer methods that are FDA-cleared/approved for exclusion of VTE, the package insert includes the approved cut-off value and this value should be provided in the report. It is not feasible for most laboratories to perform a sufficient clinical validation of a D-dimer cut-off for use in the evaluation of VTE (i.e. either exclusion or aid in diagnosis), including separate validation of the cut-off for deep vein thrombosis and pulmonary embolism. Therefore using the cutoff supplied from the manufacturer is strongly recommended.
Evidence of Compliance:
✓ Package insert stating an Intended Use for the exclusion of VTE OR an extensive clinical validation study including cut-off values appropriate to the use of the assay.
What have other CAP accredited labs in the US been doing with such request from clinicians?
Since this seems a relevant issue to me, I have
Dear all, since this seems a relevant issue to me, I have just posted an on-line survey, that can be accessed here:
It would be nice to collect a large number of feedbacks around the globe, to have a complete picture of the situation. Giuseppe Lippi
From Emmanuel Favaloro, PhD. Thanks George
From Emmanuel Favaloro, PhD. Thanks George for citing one of our STH papers above. The obvious question is do the preachers practice what they preach? I think the data that show that D-dimer levels increase with increasing age is undeniable, as is the data that shows that thrombosis risk increases with increasing age, as well as aging being associated with a higher prothrombotic state in general (examples: Favaloro, Franchini, Lippi. Aging hemostasis: changes to laboratory markers of hemostasis as we age–a narrative review. Semin Thromb Hemost 2014;40:621-33. and Raskob et al. Thrombosis: a major contributor to global disease burden. Semin Thromb Hemost 2014;40:724-35). Evidence for the potential utility of age-adjusted D-dimer levels to improve diagnostic utility in VTE exclusion is also growing, although all the confounders make this difficult to prove, and the pertinent observations made by the first responder regarding statistical validation of the approach, which is beyond the ability of most labs, including ours, adds to the complexity. Of course, additional issues such as preanalytical and analytical problems may lead to false negative or positive D-dimer results in occasional individuals that further confounds the evidence/utility. And in the end, the D-dimer is fairly meaningless if our clinical colleagues do not apply a clinical pretest probability assessment first. That all outlined, we are planning to more formally evaluate the the situation with a view to incorporate age-adjusted D-dimer ranges as part of our ongoing evidence based approach. However, whilst this could once be undertaken with relative ease (review of literature, approval of lead scientist and lab haematologist) the reality nowadays is far more complex. Our own institution, being in Australia, is not burdened by considerations related to CAP and CLIA; however, we do have alternative ‘bureaucratic’ considerations, and as our own Pathology Network comprises 27 laboratories, we can no longer make individual lab decisions. Such a decision would now need the further approval of our haematology discipline committee, then be placed on a laboratory information system change list for further approval and action, and then finally tested/applied at all 27 labs. Might just be easier for the ER clinicians to use the current test result we issue them plus a calculator if they are so keen to adopt!
From Melissa Bethel: A clarification is nee
From Melissa Bethel: A clarification is needed. The IL DDHS 500 assay reports in FEU. The DDHS assay uses D-dimer units. From Geo: thanks for your help on this, Melissa.
From John D. Olson, MD. The studies cited a
From John D. Olson, MD. The studies cited and others have correctly pointed out the rise in D-dimer levels that are seen with increasing age. This limits the value of the D-dimer level for exclusion of VTE in elderly patients. The thresholds published in the package inserts are safe, but there will be many “false positive” tests because of the increased upper limit of the reference interval in elderly patients. Thus, adjusting the threshold for exclusion of VTE in the elderly makes good sense. The caveat is that making such an adjustment requires a study with a large number of patients in order to reach adequate statistical power, a number beyond the ability of most laboratories. Extracting the threshold from published reports is problematic because of variability among methods. If there is a study that is done with the same method (reagent and instrument) that reports a threshold that demonstrates adequate statistical power, one might consider using the published threshold. This needs to be done with care. I would point out that the ADJUST-PE report in JAMA does not appear to mention whether they are measuring/reporting Fibrinogen Equivalent or D-dimer units. As examples, included in the study are VIDAS D-Dimer Ex-clusion test (bioMérieux), STA-Liatest D-Dimer (Stago) both of which measure FEU and the D-Dimer HS 500 (IL Diagnostics) which measures DDU. The authors do not address how adjustment was made for this discrepancy. This is just an example of the care that one needs to take when applying data in the literature to a single lab.