Objective: The purpose of this study was to determine the accuracy and to assess the interobserver variability of compression sonography (CS) in detecting isolated calf deep vein thrombosis (DVT) in postoperative and symptomatic patients without clinical or contrast venographic indication of previous lower extremity DVT.
Subjects and methods: A prospective double-blinded study was undertaken on 287 consecutive legs using contrast venography (CV) as the gold standard. One hundred seventy-nine legs were from the postoperative group, and 108 legs were from the symptomatic group. The CS technique was used to diagnose DVT, and color Doppler imaging was used only as a road map. Calves were examined while the patient was sitting. Also, interobserver agreement was measured in the CS evaluation of calf veins of another group of 54 symptomatic legs.
Results: Forty (22%) CVs and 13 (7%) CS examinations for calf assessment were indeterminate in the postoperative group as compared with 33 (30%) and 10 (9%) in the symptomatic group. The sensitivity, specificity, negative predictive value, and positive predictive value of CS in detecting calf DVT with or without extension to above-the-calf veins were 92% (confidence interval [Cl], 82-100%), 100%, 98% (Cl, 95-100%), and 100%, respectively, in the postoperative group and 86% (Cl, 72-100%), 96% (Cl, 90-100%), 94% (Cl, 87-100%), and 90% (Cl, 77-100%), respectively, in the symptomatic group. The sensitivity of CS was 92% (Cl, 81-100%) in the postoperative and 88% (CL, 65-100%) in the symptomatic group with isolated calf DVT. Of all calf muscular branch DVTs, 52% were detected by CS alone as compared with 21% seen on CV alone. The kappa value for the CS assessment of isolated calf vein DVTs was 0.60.
Conclusion: Sonography is a highly accurate test with acceptable reproducibility for the detection of DVT in calf veins in both postoperative and symptomatic patients. Technically adequate calf assessment can be achieved in the majority of patients if the calf examination is performed with the patient sitting.