A non-balanced spin-echo steady-state free precession (SSFP) sequence (T2FFE) is a variant of the gradient echo (GRE) sequence [1]. After the optimization of magnetic resonance imaging (MRI) parameter settings, T2FFE imaging in the hepatobiliary phase (HBP) of gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI (Gd-EOB-MRI) enables the depiction of liver hemangiomas as high signal intensity (SI) lesions and metastatic liver tumors as iso- or low SI lesions [2,3]. The T2FFE sequence may thus be useful for the differential diagnosis of liver hemangiomas and metastatic liver tumors by an interpretation of lesion SIs, but the diagnostic performance of this sequence has not been evaluated. In this study, we evaluated the diagnostic performance of T2FFE imaging in the HBP of Gd-EOB-MRI for the differential diagnosis of liver hemangiomas and metastatic liver tumors.A total of 63 patients (females:males = 27:36, mean age 62 yrs) with 133 liver lesions, including 37 hemangiomas and 96 metastatic tumors, were scanned by Gd-EOB-MRI on a 1.5T MR system (Achieva 1.5T Nova Dual, Philips Healthcare, Best, the Netherlands) using a 16-channel torso/cardiac coil. In addition to conventional MRI such as dynamic contrast-enhanced MRI with a 3D fat-suppressed gradient-echo T1WI sequence (T1FFE), T2FFE imaging was scanned in the HBP at 20 min after the intravenous administration of a total amount of Gd-EOB-DTPA based on body weight (0.1 mL/kg). The MR sequence parameters of the T2FFE imaging for the acquisition of whole-liver 3D black-blood images in a breath-holding scan were optimized referring to previous studies [2,3]. The MR sequence parameters of the T2FFE imaging were TR/TE = 9.1/4.6 msec, flip angle = 50°, field of view = 350 × 288 mm², matrix = 224 × 180, slice thickness = 3 mm, number of slices = 60, number of excitations = 1, respiratory control = breath-hold, and total scan time = 18.6 sec. The analyzed liver lesions were confirmed by characteristic imaging findings of computed tomography (CT) and/or MRI, follow-up CT and/or MRI examinations, or pathological results after surgical resections. The diagnostic performance of the T2FFE imaging in the HBP of the Gd-EOB-MRI was independently analyzed by two experienced readers in an interpretation of lesion SIs compared with the liver parenchyma SI. Based on the degree of the lesions’ SI values, the two readers evaluated all liver lesions using a four-point scale, with 1 point for bright SI lesions, 2 points for lesions that showed high SI in ≥ 50% of the lesion’s area, 3 points for lesions that showed high SI in < 50% of the area, and 4 points for iso- or low SI lesions. We considered liver lesions with 1 or 2 points liver hemangiomas and those with 3 or 4 points metastatic liver tumors. We conducted a receiver operating characteristic analysis to calculate the sensitivity, specificity and area under the curve (AUC) for the differentiation of the liver hemangiomas from metastatic liver tumors. We analyzed the inter-reader agreement using the weighted kappa statistic. The kappa values were interpreted as follows: 0.00–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial agreement; and 0.81–1.00, almost perfect agreement. P-values <0.05 were considered significant.The two readers’ evaluations of the lesions’ SI points are shown in Figure 1. The sensitivity, specificity, and AUC for the differentiation of the hemangiomas from the metastatic tumors were as follows: reader 1, 92.7%, 78.4%, and 0.86; reader 2, 90.6%, 83.8%, and 0.87, respectively. The kappa value was 0.926 (p<0.05), which indicated almost perfect agreement between the two readers. Example images of a liver hemangioma and a metastatic liver tumor on T2FFE and T1FFE images in the HBP of Gd-EOB-MRI are provided in Figures 2 and 3, respectively.On the T2FFE imaging in the HBP of the Gd-EOB-MRI, the liver hemangiomas showed positive contrast (high SI), whereas the metastatic liver tumors showed negative contrast (iso- or low SI). We speculated that an increase in the liver parenchyma SI due to the uptake of Gd-EOB-DTPA unexpectedly became a reference point of the signal level on the T2FFE images [3-5]. In addition, the T2FFE sequence could clearly show the slight differences of T2 relaxation times or T1/T2 relaxation ratios between the liver hemangiomas and metastatic liver tumors [3,4]. In conclusion, the non-balanced spin-echo SSFP (T2FFE) sequence in the HBP of Gd-EOB-MRI could provide useful information for the differential diagnosis of liver hemangiomas and metastatic liver tumors.