Deep vein thrombosis: more interpretable scans using serial ultrasound than IPG.

Clinical bottom line (level 1b)

  1. In patients with suspected deep vein thrombosis, there were no clear differences in recurrent venous thromboembolism, death or number of DVT detected by serial testing using ultrasound or impedance plethysmography.
  2. There were fewer uninterpretable scans using ultrasound (NNT = 41 at 6 months) .
Hejboer et al: New England Journal of Medicine 1993; 329 (19): 1365-1369
Expires May 2003

The study

Unblinded concealed randomised trial without intention-to-treat
Setting: outpatients clinics, Canada and Holland

1102 patients (aged range 18 to 95 years; mean 60, 57% male) clinically suspected deep vein thrombosis

Excluded if
  • <18 years old
  • received full dose of anticoagulation for >48 hours
  • history of documented DVT in same leg
  • symptoms suggestive of pulmonary embolism
  • known contrast media allergy
  • pregnant
  • unable to attend for serial testing


    • Note:
  • Diagnostic tests were independent ?blinded reference standard applied in all patients from an appropriate spectrum: reference standard was venogram or follow-up six months.


    • Control Group: (n = 494, 490 analysed): serial impedance plethysmography.
    Experimental Group: (n = 491, 490 analysed): compression ultrasound of common femoral and popliteal veins.
    If scan negative, anticoagulation withheld and scan repeated on day 2 and 8. If positive at any time, patient had a venogram. If venogram was positive or uninterpretable, patient was anticoagulated.
    100% followed for 6 months
    Outcome notes:
    • recurrent venous thromboembolism : subsequent DVT diagnosed by venogram. PE diagnosed by high probability ventilation-perfusion scan or autopsy

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    recurrent venous thromboembolism 6 months 9
    (1.84%)    		 5
    (1.02%)    		 44%
    (-65% to 81%)    		 0.82%
    (-0.67% to 2.30%)    		 120
    (NNT = 43 to infinity;
    NNH = 150 to infinity)
    death 6 months 8
    (1.63%)    		 7
    (1.43%)    		 13%
    (-139% to 68%)    		 0.20%
    (-1.33% to 1.74%)    		 490
    (NNT = 57 to infinity;
    NNH = 75 to infinity)
    DVT diagnosed by serial testing 6 months 13
    (2.65%)    		 5
    (1.02%)    		 62%
    (-7% to 86%)    		 1.63%
    (-0.05% to 3.31%)    		 61
    (NNT = 30 to infinity;
    NNH = 2200 to infinity)
    uninterpretable non-invasive test or venogram 6 months 22
    (4.49%)    		 10
    (2.04%)    		 55%
    (5% to 78%)    		 2.45%
    (0.23% to 4.67%)    		 41
    (21 to 440)

    Comments

    1. More DVTs were noted initially in the ultrasound group (P<0.02).
    2. Impedance plethysmography was not possible in 8 patients (due to massive leg swelling).
    3. Ultrasound scan was not possible in 2 patients (due to radiation ulcer).
    4. As there were significant false-positive and false-negative finding, a strategy using US alone is insufficient.Recent development to overcome these shortcomings in the management of patients with possible DVT include: clinical models, D-dimer assays, and management algorithms (see other DVT CATs).

    Citation

    1. Hejboer H, Buller HR, Lensing AWA, et al: Comparison of real-time compression ultrasonography with impedance plethysmography for the diagnosis of deep-vein thrombosis in symptomatic outpatients. New England Journal of Medicine 1993; 329 (19): 1365-1369
    Contributor: Chris Ball and Clare Wotton, May 2000
    Reviewer: Alan Forster

    Clinical Question.
    Patient DVT
    Intervention or Exposure impedance plethysmography
    Comparison ultrasound
    Outcome recurrent VTE