Here is a response from Dr. John Olson differentiating the D-dimer reference interval from the venous thromboembolism (VTE) exclusion threshold. Dr. Olson responded to my question sent last week, excerpted here:
Dr. Olson, I hope all is well with you. I am having trouble answering Kim Kinney’s question, and it brings me back to your D-dimer post from last summer. Ms. Kinney has a locally established reference interval, 110-290 D-dimer units,and a limit she uses to exclude VTE in the ED, 230 D-dimer units. Here at UAB we publish a normal range of 110 to 240 ng/mL but a PE/DVT exclusion limit of 500 ng/mL. Others establish a normal range and just use the upper limit of normal as the exclusion cutoff. What is CAP looking for when requesting a normal range and a separate VTE cutoff?
Dr. Olson’s answer…
George:
Sorry to be delayed replying to your query. It is good to hear from you. Please publish my response below in FritsmaFactor if you like.
The issue that is raised by Ms Kinney is an important one. I cannot speak for the CAP, but will give you my opinion about the problem. The D-dimer reference interval and the threshold for VTE exclusion are two very different concepts and need to be developed independently. Although the upper limit of the reference interval may be the same as the threshold for exclusion of VTE, for most methods it is not. It is usually a little higher but, in some cases, the threshold is below the upper limit of the reference range.
It would be most fortunate if the study for the threshold for VTE resulted in a value that was the same at the upper limit of the reference interval. This is the least confusing for the clinician, if the test is normal, VTE can be excluded.
However, the threshold usually is not the same as the upper limit of the reference interval. This is problematic because clinicians are accustomed to thinking of tests as positive or negative. This is particularly worrisome in the case of a threshold that is within the reference interval. In this case the assumption that a “normal” value excludes VTE could exclude individuals who are actually above the threshold and at risk for VTE even though the value is normal. The contrary situation is not as risky. If the threshold for VTE is above the reference interval and the clinician assumes exclusion is at the upper limit of the reference interval, then some individuals will be called at risk when they are not. This is not risky for the patient, but reduces the power of the test and increased cost for unnecessary studies.
I believe that it is for these reasons that the CAP now requires that both the reference interval and the threshold for the exclusion of VTE be reported with the result of the test.
–
Thanks, John
No comments here.