Diagnosis of lower-limb deep venous thrombosis: a prospective blinded study of magnetic resonance direct thrombus imaging

Abstract

Background: Current magnetic resonance techniques generate high signal from venous blood and show thrombi as filling defects. Magnetic resonance direct thrombus imaging (MRDTI) directly visualizes acute thrombus.

Objective: To determine the accuracy of MRDTI for diagnosis of acute symptomatic deep venous thrombosis (DVT) below and above the knee.

Design: Prospective, blinded study.

Setting: A 1355-bed university hospital.

Patients: 101 patients with suspected DVT who had had routine venography. Participants were recruited from a cohort of patients with suspected DVT. All patients with a positive venogram and one quarter of patients with a negative venogram were selected by using a random sequence.

Intervention: MRDTI was performed within 48 hours of venography and was interpreted by two reviewers.

Measurements: Diagnosis of DVT overall; isolated calf, femoropopliteal, and ileofemoral DVT; and thrombus in the calf, femoropopliteal, and iliac segments.

Results: The reports from two readers had sensitivities of 96% and 94% and specificities of 90% and 92% for diagnosis of DVT. Sensitivities were 92% and 83% for isolated calf DVT, 97% and 97% for femoropopliteal DVT, and 100% and 100% for ileofemoral DVT. Specificities were 94% and 96% for isolated calf DVT and 100% and 100% for both femoropopliteal and ileofemoral DVT. Similarly, sensitivity and specificity within each of the venous segments ranged from 91% to 100%. Interobserver variability measured by using a weighted kappa statistic ranged from 0.89 to 0.98 for these measures.

Conclusion: Magnetic resonance direct thrombus imaging is an accurate noninvasive test for diagnosis of DVT, and its accuracy is maintained below the knee. Comparison of individual venous segments showed that results of MRDTI agreed strongly with findings on venography. Scanning was well tolerated, and interpretation was highly reproducible.