Detectability of small liver metastases with gadolinium BOPTA.

RATIONALE AND OBJECTIVES

The ability to detect small liver metastases was evaluated with both gadolinium Gd BOPTA and Gd HP-DO3A on high-field (1.5 tesla [T]) magnetic resonance (MR) imaging using a rabbit tumor model.

METHODS

Five New Zealand White rabbits with metastatic liver disease (VX-2 adenocarcinoma) were imaged on a 1.5 T Siemens Vision MR system. Magnetic resonance studies were obtained in each animal on days 8 and 9 after tumor implantation. Each animal was studied twice, once after injection of 0.3 mmol/kg Gd HP-DO3A (gadoteridol or ProHance) and once after injection of 0.1 mmol/kg Gd BOPTA (gadobenate dimeglumine or MultiHance). The order of injection for the two agents was randomized with the two studies in any one animal separated by 24 hours to allow for clearance. Magnetic resonance image acquisition was performed in all cases with suspended respiration. Baseline two-dimensional FLASH T1-weighted and turbo-spin echo T2-weighted scans were acquired first. The contrast was then administered as an intravenous bolus. T1-weighted scans were acquired at 1, 5, 15, 30, 45, and 60 minutes after administration of Gd BOPTA and 1, 5, and 15 minutes after administration of Gd HP-DO3A. Each rabbit was killed after completion of imaging, their liver removed and taken to the veterinarian at the University's animal disease diagnostic laboratory for lesion confirmation.

RESULTS

Despite acquisition of precontrast T2-weighted scans, lesions could not be identified with certainty in four of five animals in the Gd HP-DO3A study. Normal liver signal intensity increased from 895 +/- 17 to a peak of 1384 +/- 50 at 1 minute after Gd HP-DO3A administration. After Gd BOPTA administration, normal liver signal intensity increased from 899 +/- 105 to a peak of 1433 +/- 76 at 15 minutes. Liver enhancement thereafter decreased gradually to 1297 +/- 84 at 60 minutes. The injection of 0.3 mmol/kg Gd HP-DO3A resulted in parenchymal enhancement, which was statistically superior (P < 0.01) to an injection of 0.1 mmol/kg Gd BOPTA at 1 minute, not statistically different at 5 minutes, and inferior (P < 0.02) at 15 minutes. From region of interest measurements, lesion detectability was statistically superior on scans at 15 to 60 minutes after Gd BOPTA administration compared with precontrast T1- and T2-weighted scans (P values: < 0.03- < 0.005). Lesion detectability was maximum at 30 minutes postcontrast (15.2 +/- 4.5), markedly superior to that precontrast on both T1- (5.7 +/- 5.0) and T2-weighted scans (7.2 +/- 1.5). On masked film review of the Gd BOPTA case set, no lesions were noted prospectively on T2-weighted scans. Lesions in all five animals were well visualized on scans 45 to 60 minutes after Gd BOPTA administration. The Gd HP-DO3A case set was not read masked, as lesions could be identified only in one of the five animals with all films available for inspection. An additional feature of scans with Gd BOPTA (used at a dose of 0.1 mmol/kg), in distinction to those with Gd HP-DO3A (used at a dose of 0.3 mmol/kg), was the diminished enhancement of hepatic vessels.

CONCLUSIONS

Using a rabbit model, small metastatic lesions (diameter, 2-4 mm) were well visualized on delayed postcontrast Gd BOPTA scans. These lesions could not be diagnosed prospectively on T2-weighted images. In only one of five animals were lesions detected on early dynamic post-contrast high-dose Gd HP-DO3A scans.