Multiparametric Contrast-Free MRI Successfully Identifies Venous Thrombus Responsive to Lytic Therapy: From Mice to Humans.

BACKGROUND: Randomized trials of venous thrombolysis to prevent postthrombotic syndrome have produced mixed results. A method to identify patients most likely to benefit from interventional treatment is needed. This study evaluated a contrast-free, magnetic resonance-based multisequence thrombus imaging (MSTI) technique to characterize deep venous thrombi and predict susceptibility to thrombolysis. METHODS: Venous thrombosis was induced in the inferior vena cava of BALB/C mice (n=56, male), which were imaged up to 28 days postsurgery and 24 hours after systemic thrombolysis (Actilyse, 10 mg/kg, IV). The prelysis MSTI protocol included 3-dimensional T1 mapping, 3-dimensional magnetization transfer, and 2-dimensional diffusion-weighted imaging. Thrombolysis was defined as successful if inferior vena cava blood flow increased by ≥50% compared with prelysis values. In a clinical cohort, 41 patients with acute iliofemoral deep venous thrombi underwent MSTI before catheter-directed thrombolysis. Imaging parameters were analyzed against postintervention outcomes. RESULTS: MSTI identified thrombi susceptible to thrombolysis in both mice and humans. In mice, lysed thrombi showed lower T1 (723 [667-782] versus 874 [799-1000] ms; P<0.001) and higher apparent diffusion coefficient values (1.02 [0.96-1.14] versus 0.78 [0.62-0.88]×10-³ mm²/s; P<0.001) than nonlysable thrombi, with no difference in magnetization transfer. In patients, lysed thrombi demonstrated lower T1 (606 [543-656] versus 765 [630-909] ms; P<0.001), lower apparent diffusion coefficient (0.67 [0.5-1.1] versus 1.23 [0.69-1.74]×10-³ mm²/s; P=0.001), and similar magnetization transfer rates. Combining MSTI parameters optimized prediction, achieving 88% sensitivity and 97% specificity in mice, and 86% sensitivity and 91% specificity in humans. CONCLUSIONS: MSTI enables noninvasive, contrast-free characterization of thrombus composition and predicts thrombolytic susceptibility. This technique has the potential to guide patient selection for invasive therapies and should be incorporated into future trials of venous thrombosis treatment.