Because only 15 to 28 percent of patients who are thought to have a deep venous thrombosis (DVT) actually have one, diagnosis remains a challenge. Ultrasonography of the femoral and popliteal veins is currently the best strategy for diagnosing this condition. Schutgens and associates conducted this retrospective cohort study to determine if semiquantitative d-dimer testing along with ultrasonography was useful and what variables might limit the accuracy of d-dimer testing.

Current diagnostic strategy consists of repeating ultrasonography in one week if initial results are negative. Withholding anticoagulation is thought to be safe in patients whose repeat study is also negative. d-dimer testing measures degradation products of cross-linked fibrin and can be used in the diagnosis of DVT. A patient with a normal ultrasound examination and negative d-dimer testing can also avoid anticoagulation.

Patients thought to have DVT were included in the study. Patients had ultrasonography of the symptomatic leg and d-dimer testing. If the first ultrasound was negative, it was repeated in one week. Anticoagulants were begun if the ultrasound was positive for DVT. Demographic information and information about use of anticoagulants were recorded.

There were 704 people included in the study. Forty percent of patients (187 out of 464) had a second ultrasonography. Overall, 36 percent of patients were diagnosed with DVT. Of the 14 who were diagnosed after the initial presentation (10 at one week, two at eight weeks, one at three months, one at five months), all had positive d-dimer tests. In the whole study cohort, the sensitivity and negative predictive value of the d-dimer test were highest in patients who did not have a history of cancer or anticoagulant use. Patients taking anticoagulants had negative d-dimer tests more often than those not taking anticoagulants (46 percent and 20 percent, respectively). With or without DVT, patients with cancer were likely to have a positive d-dimer test, giving a specificity of 22 percent in this subgroup. Positive d-dimer testing also increased in prevalence as patients aged.

The authors conclude that d-dimer testing should not be used in patients taking anticoagulants because of their reduced ability to generate thrombin. In patients with a history of DVT, d-dimer testing may be more reliable than ultrasonography when evaluating a patient for recurrence. Also, d-dimer testing will probably not be helpful in patients with cancer or those who are older than 70 years.