Abstract
BACKGROUND: The endovascular treatment of peripheral artery disease in the below knee arteries has high rates of immediate technical failure, most commonly owing to an inability to cross the lesion with a guidewire. Noncontrast magnetic resonance imaging (MRI) is able to characterize plaque composition and may identify patients at high risk for failed endovascular crossing. This study evaluates a novel noncontrast MRI protocol in an amputated limb model to identify impenetrable plaques, offering clinicians a tool for improved patient selection and procedural planning. METHODS: Patients with chronic limb-threatening ischemia undergoing major limb amputation were included and underwent 3T noncontrast MRI. Lesions with >75% stenosis or occlusion in the below the knee arteries were evaluated and defined as either hard (>50% calcium/collagen) or soft (<50%). Hard lesions were then further delineated into two groups: minimally calcific hard (<50% calcium) and significantly calcific hard lesions (>50% calcium). Lesions were also scored using Trans-Atlantic Inter-Society Consensus Score (TASC) II and Global Limb Anatomic Severity Score (GLASS) systems. Amputated limbs then underwent percutaneous vascular intervention to cross each lesion. The primary end point was to compare crossing success in MRI-defined hard vs soft lesions. Secondary outcomes included comparing TASC C/D with TASC A/B lesions, GLASS III/IV and I/II lesions, ability to cross in minimally vs significantly calcified hard lesions, and stenotic vs total occlusions. RESULTS: The study included 27 patients with 46 arterial lesions. Twelve (26%) lesions were defined as soft and 34 (74%) were hard. MRI plaque characteristics demonstrated a strong predictive value: hard lesions were significantly less likely to be crossable compared with soft lesions (91% vs 8%; P < .001). TASC C/D lesions had significantly higher crossing failure rates compared with TASC A/B lesions (77% vs 33%; P = .039), whereas GLASS III/IV lesions did not achieve statistical significance when compared with GLASS I/II lesions (73% vs 44%; P = .129). Among hard lesions, there was no significant difference in ability to cross between minimally calcific and significantly calcific lesions (92% vs 90%; P > .99). There was also no significant difference in the ability to cross occluded vs stenotic lesions (67% vs 80%; P = .70). CONCLUSIONS: Ultrashort echo timeUTE)/steady-state free precession (SSFP) noncontrast MRI identifies critical plaque components that predict crossing failure, including dense collagen, which is not visible on conventional imaging yet appears to play a critical role in crossing failure that is independent of calcium burden. This MRI protocol could transform decision-making. Its implementation in clinical practice offers an evidence-based approach that could potentially decrease endovascular failures, ensure advanced crossing devices are available for complex lesions, and guide treatment choices for bypass surgery vs endovascular treatment.