College of American Pathologists
Printable Version

  D-dimer testing and reporting for
  the diagnosis of VTE





cap today

May 2005
Feature Story

Since its introduction, the measurement of D-dimer in plasma or whole blood has found a number of uses in clinical practice. These uses are as follows:

  • High negative predictive value for excluding venous thromboembolism, or VTE.
  • Elevated levels in patients with increased fibrinolysis and a sensitive (but not specific) diagnostic test for disseminated intravascular coagulation.
  • Elevated levels following completion of oral anticoagulation are associated with VTE.
  • Independent marker of increased risk of future myocardial infarction in patients evaluated for chest pain. Of particular importance is the use of the D-dimer in the diagnostic algorithm for deep venous thrombosis (DVT) and pulmonary embolism (PE), two forms of VTE that are common, and in the case of PE, potentially fatal. D-dimer levels become elevated in the blood of patients with VTE due to fibrinolysis of newly formed blood clots, and this pathophysiologic abnormality has been targeted for diagnostic purposes. In concept, if the D-dimer level is sufficiently low in a patient suspected of having VTE, then the diagnosis can be excluded. In practice the use of the D-dimer test in the diagnostic algorithm is usually more complex because the patient’s pretest probability of disease may also be taken into consideration before a D-dimer is ordered. Nonetheless, because of the serious nature of this clinical decision it is critical that laboratories accurately measure and report D-dimer levels.

The CAP Coagulation Resource Committee considered this to be an important patient safety issue. So to gain a better understanding of the potential variation in how laboratories are measuring and reporting D-dimer levels, the committee conducted a questionnaire survey of 4,857 CG1 and CG2 participants in the 2004-C Surveys mailing. The questions were designed to determine the D-dimer methods being used, the type of units being reported [fibrinogen equivalent units (FEU) versus D-dimer units (D-DU)], the magnitude of units being reported, whether the test was being used to exclude VTE, and the cutoff values being used to exclude VTE. These data were compared with the manufacturers’ recommended cutoff values for excluding VTE.

Of the 4,857 surveyed, 2,232 (46 percent) indicated that they report the D-dimer. Surveys were returned by 2,018 (42 percent of the total, 90 percent of those reporting the D-dimer). Of this group of 2,018 laboratories, 1,506 (75 percent) reported using the D-dimer to exclude VTE. Among these 1,506 laboratories, 11 different methods were being used (eight to 466 users per method): A quantitative method was used by 1,460 (96.8 percent), a whole blood point-of-care method by 17 (1.1 percent), and a semiquantitative method by 29 (1.9 percent). Fifty-nine percent reported FEU and 41 percent reported D-DU. Surprisingly, 126 laboratories (eight percent) did not know the type of units they were using. Eight different combinations of type and magnitude of units were reported: ng/mL FEU, ng/mL D-DU, μg/L FEU, μg/L D-DU, μg/mL FEU, μg/mL D-DU, mg/L FEU, and mg/L D-DU. All methods showed variability in the type and magnitude of the units. Another important observation: 488 laboratories (39 percent) were using a D-dimer cutoff value above that recommended by the manufacturer.

These data indicate several aspects of D-dimer testing that are of significant concern. The first and greatest concern is that many laboratories appear to be using cutoff values that are too high for their particular method. This may lead to many patients being erroneously excluded from VTE. Second, some laboratories don’t know which type of unit they are reporting (FEU versus D-DU). This makes it impossible to know with certainty what cutoff value should be communicated to clinicians. Third, a small number of laboratories are using semiquantitative methods for excluding VTE. This may be inappropriate because semiquantitative methods have not been validated in the medical literature for the exclusion of VTE.

How should laboratories respond to this problem? The Coagulation Resource Committee is recommending that laboratories that wish to use the D-dimer test for excluding VTE consider the following guidelines to help remedy these problems:

  • Verify that the D-dimer method being used has been validated for VTE screening. Determine whether the method has been validated as such by carefully reading the package insert or communicating directly with the manufacturer. If the method has been validated, the manufacturer should be able to provide a cutoff value for the exclusion of DVT, PE, or both DVT and PE. In addition, the manufacturer should state if the cutoff value can be used in all patients suspected of VTE, or only in those with certain pretest probabilities of VTE (low, moderate, or high pretest probability). If laboratories want to modify the manufacturer’s recommended cutoff, or establish their own cutoff on a method without a known value, then they should conduct their own validation study before implementation. It is also important to be aware that with some methods, the recommended cutoff for exclusion of VTE is well below the upper limit of the reference range. In these cases it is not sufficient to set the cutoff at the upper limit of normal; instead, it must be set below that level.
  • Communicate to clinicians the D-dimer cutoff value for exclusion of VTE. This can be accomplished in a number of ways. Examples are a written indication of the cutoff within the patient report or a written memo to clinicians.
  • Verify that the type and magnitude of D-dimer units are being correctly stated on patient reports. The units that the D-dimer method generates directly can be determined from the package insert for the method being used. Some laboratories may be converting the units stated in the package insert into different units for the patient report. If this is being done, one should verify that the calculations are correct. For example, a method may generate data as μg/mL FEU, and the laboratory may wish to convert this to ng/mL D-DU. In this situation one must convert both the "unit magnitude" (μg/mL to ng/mL) and the "unit type" (FEU to D-DU). A critical concept for this calculation is that the mass of one unit of FEU is twice that of the mass of one unit of D-DU (1 ng/mL D-DU equals 2 ng/mL FEU). So in this example the correct conversion factor is 500, which is calculated as follows: 1μg/mL FEU X (1000 ng/1μg) X (1D-DU/ 2FEU) = 500 ng/mL D-DU. To avoid calculation errors, it may be prudent for laboratories to use the units the manufacturer recommends rather than to make a mathematical conversion.

Dr. Cunningham, chair of the CAP Coagulation Resource Committee, is director of the hematology laboratory, Department of Pathology, University of Kansas Medical Center, Kansas City, Kan. Dr. Olson, immediate past chair of the Coagulation Resource Committee, is professor and vice chair for clinical affairs, Department of Pathology, University of Texas Health Science Center at San Antonio, and director of clinical laboratories, University Health System.