Quiescent-Interval Slice-Selective MRA Accurately Estimates Intravascular Stent Dimensions Prior to Intervention in Patients With Peripheral Artery Disease

Background Quiescent-interval slice-selective (QISS) magnetic resonance angiography (MRA) is a non-contrast alternative for the pre-procedural assessment of patients with peripheral artery disease (PAD). However, the feasibility of pre-procedural stent size estimation using QISS MRA would merit investigation. Purpose To evaluate the feasibility of QISS MRA for pre-procedural stent size estimation in PAD patients compared to computed tomography angiography (CTA). Study Type Retrospective. Subjects Thirty-three PAD patients (68 +/- 9 years, 18 men, 15 women). Field Strength/Sequence Two-dimensional balanced steady-state free precession QISS MRA at 1.5 T and 3 T. Assessment All patients received QISS MRA and CTA of the lower extremity run-off followed by interventional digital subtraction angiography (DSA). Stenotic lesion length and diameter were quantified (AMF and AVS with 3 and 13 years of experience in cardiovascular imaging, respectively) to estimate the dimensions of the stent necessary to restore blood flow in the treated arteries. Measured dimensions were adjusted to the closest stent size available. Statistical Tests The Friedman test with subsequent pairwise Wilcoxon signed-rank test was used to compare the estimated stent dimensions between QISS MRA, CTA, and the physical stent size used for intervention. Intra-class correlation (ICC) analysis was performed to assess inter-reader agreement. Significant differences were considered at P < 0.05. Results No significant difference was observed between estimated stent diameter by QISS MRA or CTA compared to physical stent diameter (8.9 +/- 2.9 mm, 8.8 +/- 3.0 mm, and 8.8 +/- 3.8 mm, respectively; chi(2) = 1.45, P = 0.483). There was a significant underestimation of stent length for both QISS MRA and CTA, compared to physical stent length (45.8 +/- 27.8 mm, 46.4 +/- 29.3 mm, and 50.4 +/- 34.0 mm, respectively; chi(2) = 11.96) which could be corrected when measurements were adjusted to the next available stent length (chi(2) = 2.38, P = 0.303). Inter-reader assessment showed good to excellent agreement between the readers (all ICC >= 0.81). Data Conclusion QISS MRA represents a reliable method for pre-procedural lesion assessment and stent diameter and length estimation in PAD patients. Level of Evidence 3 Technical Efficacy Stage 2