Evaluation of standard and high-dose contrast-enhanced breath-hold liver magnetic resonance for visualization of metastatic disease in an animal model.

Abstract

The ability of dynamic contrast-enhanced magnetic resonance imaging to improve detection of liver metastases was evaluated in an experimental rabbit model. Scans using two different contrast doses, 0.1 and 0.3 mmol/kg, of a gadolinium chelate with extracellular distribution were compared to precontrast T2- and T1-weighted scans. Seven New Zealand White rabbits with VX-2 adenocarcinoma metastases to the liver were imaged at 1.5 tesla. Each animal was studied twice, on different days, to evaluate both contrast doses. Precontrast T2- and T1-weighted scans were compared to dynamic postcontrast T1-weighted scans obtained at 1, 2, 3, 4, and 5 minutes after intravenous injection. All scans were acquired during suspended respiration. The contrast agent, Gd HP-DO3A (gadoteridol or ProHance), was administered as a bolus. Images were analyzed by region of interest measurements. Injection of 0.3 mmol/kg Gd HP-DO3A produced liver enhancement which was statistically superior to 0.1 mmol/kg at all time points postcontrast. Enhancement of the paraspinous musculature at the higher dose was also statistically superior at all time points, with one exception (5 minutes postcontrast). Lesion detectability, evaluated by the signal difference over noise ratio, peaked at one minute postcontrast and was substantially greater at the higher contrast dose (31.4 +/- 8.3 at 0.3 mmol/kg versus 16.8 +/- 4.2 at 0.1 mmol/kg, P = 0.02). Using a rabbit model and breath-hold imaging technique, metastatic lesions in the liver were best visualized on early (1 minute) dynamic high dose (0.3 mmol/kg) postcontrast scans. Contrast dose and timing of image acquisition are critical issues for optimal liver lesion detection on magnetic resonance imaging.