Dealing with D-dimer debacles

D-dimers are specific cross-linked fibrin derivatives that are the product of plasmin-mediated fibrinolytic degradation. They are an excellent marker of fibrinolytic activity.

The D-dimer assay can be used as a “rule out” test in the emergency department when the pre-test probability of venous thromboembolism (VTE) is low.

The pre-test probability is best assessed using a clinical prediction guide (CPG), such as the Wells criteria or the revised Geneva score for pulmonary embolism (PE) and the Wells criteria for deep vein thrombosis (DVT). For instance, a ‘negative’ D-dimer test result when using a highly sensitive assay can ‘rule out’ VTE using the Wells’ clinical prediction guides if:

• Low/intermediate probability of PE (score 0 to 6) — Wells criteria for PE
• DVT unlikely (score 0 or 1) — Wells criteria for DVT

The D-dimer assay can also be used to help diagnose fibrinolytic disorders such as venom-induced consumptive coagulopathy (VICC) and disseminated intravascular coagulation (DIC).

This is my approach to the use of the d-dimer assay in the emergency department:

How to use the D-dimer assay in the Emergency Department. Note that highly sensitive D-dimer assays can be used to rule out PE in patients with low or intermediate Wells scores (0 to 6), less sensitive assays should only be used for low Wells scores (0 to 2).

[see Pulmonary Puzzle 010 for a discussion of the PERC rule]

No

The systematic review of 10 studies by Sutherland et al (2008) showed that D-dimer cannot be used as a sole screening tool for acute aortic dissection. Problems with the studies include:

• 4 of the 10 studies had sensitivities <95%, and 3 of the 5 studies with sensitivities of 100% had the lower limit of the 9% confidence interval <85%. Thus the sensitivty of the D-dimer is insufficient for ruling out a ‘must not miss’ diagnosis like acute aortic dissection
• inclusion bias and poorly defined inclusion criteria
• poorly defined duration of symptoms prior to performing the D-dimer test (up to 120 hours)
• the studies do not assess how clinical suspicion or other investigations affected the use of the D-dimer test
• heterogeneity of D-dimer assays used and the cut-off values used
• the non-specificity of the D-dimer test might lead to excessive imaging
• use of the D-dimer in high risk cases may delay definitive imaging and treatment

No — early studies are promising but the test is not yet ready for ‘prime-time’ for this use.

A Best Bets review (updated in 2007) suggests that a negative D-dimer test may be able to rival echocardiography in ruling out an atrial thrombus in atrial fibrilation (negative predictive value ~98%), thus allowing cardioversion to be safely performed in the emergency department without the need for further investigation. However many of the studies included were in Japanese and only the English language abstracts were appraised. A large well-designed study is needed to confirm these promising findings.

VTE can occur in the presence of a negative D-dimer test

D-dimer assays can only be used to rule out VTE if the pre-test probability is sufficiently low. Negative results in the presence of VTE can occur if:

• there is a small clot load (eg. small calf only DVTs)
• the thrombus has matured over time
• the patient has defective fibrinolysis (which can contribute to thrombus formation in the first place)

Non-pathological conditions associated with elevated D-dimer titres include:

• Age (healthy elderly people)
• Cigarette smoking
• Functional impairment
• Post-operatively
• Pregnancy
• Race (e.g. African Americans)

Pathological conditions associated with elevated D-dimer titres include

• Acute coronary syndromes
• Acute upper gastrointestinal haemorrhage
• Aortic dissection
• Arterial or venous thromboembolism
• Atrial fibrillation
• Consumptive coagulopathy – DIC, VICC
• Infection
• Malignancy
• Pre-eclampsia
• Sickle cell disease
• Stroke
• Superficial thrombophlebitis
• Trauma

Yes, with caveats.

The D-dimer assay is more likely to give a positive result in the elderly, limiting the usefulness of the test. However, if the D-dimer is negative it can still be used to rule out VTE in elderly patients with low pre-test probabilities. A recent retrospective study by Douma et al (2010) suggests that higher cut-offs might be safely used in patients over 55 years of age.

Yes, with caveats.

As in the elderly, the D-dimer test result is more likely to be positive. The D-dimer result is nearly always positive in the third trimester, and less so the earlier the pregnancy. A negative D-dimer test in a patient with a low pre-test probability  still ‘rules out’ VTE.

There are many different D-dimer assays (qualitative and quantitative) available.

Different types of assay are not standardized and have varying sensitivities and specificities, but they generally have high sensitivity and negative predictive value for the presence of thrombus in patients with low pre-test probabilities. Furthermore, different laboratories may use different ‘cut-offs’ for what constitutes a positive or negative result.

The performance characteristics of the available D-dimer assay can affect its use in clinical decision making.

I am most familiar with the HemosIL DD HS assay, which is a rapid highly sensitive ‘second-generation’ latex agglutination D-dimer assay.  This has similar performance characteristic to the best performing ELISA assays, unlike ‘first-generation’ latex agglutination assays. The published data for the HemosIL DD HS assay’s performance characteristics (using a cut-off of 230 ng/mL) are:

• Sensitivity: 100% (95%CI 95.4-100%)
• Negative predictive value: 100% (95%CI 96.5-100%)
• Specificity: 46.8% (95%CI 40.1-53.6%)

As you would say to any patient, something that encapsulates this information:

• The D-dimer test is a marker of blood clotting activity and is not diagnostic of VTE
• When used appropriately the D-dimer test helps “rule out” VTE if the test is negative and the chance of the patient having a VTE is relatively low
• If the D-dimer test is positive further investigation may or may not be necessary depending on the likely cause of the elevated D-dimer result, of which there are many
• Diagnosis of VTE is based on probabilities and balances the risks of further investigation against the risk of missing the diagnosis. No investigation can reduce the chance of having VTE to ‘absolute zero’.
• D-dimer negative patients can still develop VTE and should be advised to return if symptoms worsen or new symptoms develop (e.g. worsening of leg pain and swelling, shortness of breath, chest pain or collapse)

• Body R, Allie B. Towards evidence based emergency medicine: best BETs from the Manchester Royal Infirmary. Negative D-dimer may allow safe early cardioversion of atrial fibrillation. Emerg Med J. Jun 2007;24(6):432-435 PMID: 17513547 [D-dimer in AF]
• Chunilal SD, et al. (2003) Does this patient have pulmonary embolism?. JAMA, 290(21), 2849-58. PMID: 14657070
• De Moerloose P,  et al. Performances of the HemosIL D-dimer HS assay for the exclusion of venous thromboembolism. J Thromb Haemost. 2005;3:2361-2363 PMID: 16194215
  [The highly sensitive D-dimer assay mentioned in Q10]
• Douma RA et al. Potential of an age adjusted D-dimer cut-off value to improve the exclusion of pulmonary embolism in older patients: a retrospective analysis of three large cohorts. BMJ May 2010;340:c1475 PMID: 20354012
• Fields JM, & Goyal M (2008). Venothromboembolism. Emerg Med Clin North Am, 26 (3) PMID: 18655939
  [Highly recommended review article includes the VTE CPGs used in this Q&A]
• Ho WK, et al. Venous thromboembolism: diagnosis and management of deep venous thrombosis. Med J Aust. May 2 2005;182(9):476-481 PMID: 15865594
  [Australian DVT Guidelines]
• Lee CH, et al. Venous thromboembolism: diagnosis and management of pulmonary embolus. Med J Aust. 2005; 182 (11): 569-574 PMID: 15938684
  [Australian PE Guidelines]
• Le Gal G, et al. Prediction of pulmonary embolism in the emergency department: the revised Geneva Score. Ann Intern Med 2006;144:168 PMID: 16461960
  [Revised Geneva Score]
• Sutherland A, et al. D-dimer as the sole screening test for acute aortic dissection: a review of the literature. Ann Emerg Med. Oct 2008;52(4):339-343 PMID: 18819176
  [D-dimer in acute aortic dissection]
• Wakai A, et al. Role of fibrin D-dimer testing in emergency medicine. Emerg Med J. 2003;20:319-325 PMID: 12835339
  [Good review of the use of the D-dimer in the emergency department]
• Wells PS, et al. Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med. Sep 25 2003;349(13):1227-1235 PMID: 14507948
• Wells PS, et al. Derivation of a simple clinical model to categorize patient’s probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000;83:416–20
• Wells PS, et al. Does this patient have deep vein thrombosis? JAMA. Jan 11 2006;295(2):199-207 PMID: 10744147