The utility of non-invasive tests for detection of previous proximal-vein thrombosis.Thromb Haemost. 1995 Apr; 73(4):592-6.TH
Despite the availability of several diagnostic methods for the detection of deep-vein thrombosis (DVT), the identification of previous episodes of DVT remains a diagnostic challenge.
To assess the reliability of a combination of a standardized clinical score with three non-invasive tests: compression ultrasonography (CUS), Doppler ultrasound (DUS), and photoplethysmography (PPG), in determining the presence or the absence of previous proximal DVT.
One hundred consecutive unselected outpatients were identified, who had undergone contrast venography six to nine years previously because of the clinical suspicion of DVT (confirmed in 43). They were blindly reinvestigated by a panel of trained operators unaware of venography results. They underwent a clinical evaluation of the lower limb, by applying a standardized score to five symptoms and six signs (grading each item from 0 to 3); a PPG test to determine the venous refilling time; a DUS test to determine the venous reflux separately in the common femoral and the popliteal vein; and a CUS test to determine vein compressibility in the same regions.
An abnormal CUS test and/or the demonstration of venous reflux in the popliteal region and/or a high clinical score (> or = 8) identified twenty-four of the 43 (56%) DVT + patients with a specificity of 89%. The combination of normal CUS with the absence of venous reflux in both the common femoral and popliteal vein and a low clinical score excluded previous thrombosis in 45 (79%) of the 57 DVT-patients (negative predictive value, 78%). Abnormal venous reflux in the isolated common femoral vein did not reliably predict the presence or absence of previous DVT. However, this occurred in only 13 (13%) patients. The PPG determination of venous refilling time did not improve the results obtained with the other tests.
The combination of a standardized clinical evaluation with the results of CUS and DUS can reliably diagnose or exclude previous proximal-vein thrombosis in almost 90% of patients with previous episodes of suspected DVT.