Off-site evaluation of liver lesion detection by Gd-BOPTA-enhanced MR imaging

The aim of this study was to determine the efficacy of Gd-BOPTA-enhanced MR I in liver lesion detection in comparison with unenhanced MRI and dynamic C T. The image sets of 148 of 151 patients enrolled in a multicenter German p hase-III trial were evaluated by two independent radiologists unaffiliated with the investigating centers. Patients underwent a routine MRI protocol c omprising T2- and T1-weighted spin-echo and T1-weighted gradient-echo (GE) sequences pre and 1 h post 0.1 mmol/kg Gd-BOPTA (Bracco-Byk Gulden, Konstan z, Germany). Additionally, a serial T1-weighted GE scan was performed after administration of the first half of the dose. All patients underwent dynam ic contrast-enhanced CT. The evaluation was performed with regard to the nu mber and size of lesions detected per patient by each modality or sequence. Furthermore, all pre CM and pre + post CM image sets were analyzed for num ber of lesions per patient. Both readers detected significantly more lesion s in the contrast-enhanced image set compared with the unenhanced image set (32 and 39 %, respectively; p < 0.0001). While contrast-enhanced CT detect ed a similar number of lesions to unenhanced MRI, it was clearly inferior t o contrast-enhanced MRI (reader 1: p = 0.0117; reader 2: p = 0.0225). Of th e T1-weighted scans performed, the dynamic and late T1-weighted GE exams co ntributed most to the increased lesion detection rate (reader 1: p = 0.0007 ; reader 2: p = 0.0037). The size of the smallest lesion detected by means of MRI was significantly larger in the pre-CM image sets than in the pre post CM image sets (reader 1: p = 0.001; reader 2: p < 0.0001). Gd-BOPTA-en hanced MRI detected significantly smaller lesions than contrast-enhanced CT (reader 1: p = 0.0117; reader 2: p = 0.0925). Gd-BOPTA-enhanced MR imaging improves liver lesion detection significantly over unenhanced MRI and dyna mic CT.