Prospective intraindividual comparison of unenhanced magnetic resonance imaging vs contrast-enhanced computed tomography for the planning of endovascular abdominal aortic aneurysm repair

Abstract

Objective: This study clarified whether unenhanced magnetic resonance imaging (MRI) is an alternative to contrast-enhanced computed tomography (CT) for aortoiliac arterial measurement before endovascular abdominal aortic aneurysm repair (EVAR).

Methods: The institutional review board approved this prospective study. Twenty patients being considered for EVAR underwent MRI using a steady-state free-precession sequence in a 1.5-T system and contrast-enhanced CT within 4 weeks of each other. Two independent observers reviewed MRI and CT in random order using vessel analysis software and measured seven diameters, four lengths, and the angle of the aortoiliac arteries. The intermodality, interobserver, and intraobserver agreements were assessed for each measurement by intraclass correlation coefficients (ICCs) and the Altman-Bland method. Additionally, the observers independently recorded the number of bilateral renal arteries, decided EVAR suitability, and selected the main endograft on each modality.

Results: Intermodality ICCs for observers A and B showed ranges of 0.83 to 0.99 and 0.70 to 0.98; interobserver ICCs for MRI and CT showed ranges of 0.73 to 0.99 and 0.65 to 0.99; and intraobserver ICCs for MRI and CT showed ranges of 0.59 to 0.99 and 0.59 to 0.99. In intermodality, interobserver, and intraobserver comparisons, mean differences in diameters were included within the range -1 to +1 mm, excluding three of seven diameters on CT in interobserver comparison and one of seven on CT in intraobserver comparison. Mean differences in lengths were included within the range -5 to +5 mm, excluding one of four lengths in observer B in intermodality comparison and one of four on MRI and CT in interobserver comparison. All mean differences in angles were included within the range -5° to +5°. Both observers detected all 40 bilateral main renal arteries on MRI and CT. Of the 13 accessory renal arteries, observers A and B detected four (31%) and nine (69%), respectively, on MRI; in contrast, both observers detected 11 (85%) on CT. The observers independently determined that the same seven patients were suitable for EVAR on MRI and CT. Of the seven selected main endografts, seven and six diameters and five and six lengths agreed exactly between MRI and CT for observers A and B, respectively.

Conclusions: Although contrast-enhanced CT remains the gold standard for preoperative EVAR planning, unenhanced MRI with steady-state free-precession sequence can be an alternative modality for patients with contraindications for CT, such as renal impairment, because the intermodality agreement for preoperative measurements is as good as interobserver and intraobserver agreement.