Diluted gadolinium works well at minimizing ghosting artifacts on liver MRI

Comparing two methods of administering gadolinium-based contrast for MRI exams head to head, researchers have found that dilution with saline is better than a reduced injection rate at minimizing arterial-phase artifacts.

In a study published Dec. 15 in Radiology, Young Kon Kim, MD, PhD, of UCLA and colleagues describe their work reviewing the cases of 318 consecutive patients undergoing gadoxetic acid-enhanced MR imaging to diagnose hepatic malignancies such as liver cancer and related metastases.

In a dilution subcohort of 159 patients, gadoxetic acid was diluted 1:1 with saline and injected at a rate of 2.0 milliliters per second.

In a slow-injection subcohort, also of 159 patients, undiluted gadoxetic acid was injected at a rate of 1.0 milliliter per second.  

Some 89 patients in the dilution subcohort received follow-up MRI with the slow-injection method.

Likewise, 34 patients in the slow injection subcohort received follow-up MRI with the dilution method.

The researchers used both patient- and image-based analyses, as well as intraindividual analysis, to compare two parameters—mean artifact score rated by two observers using a five-point scale and frequency of severe artifact—between the dilution and slow injection subcohorts.

In a patient-by-patient analysis, they found a significant difference in the mean artifact score of arterial phase images between patients with slow injection of contrast material at a rate of 1.0 mL/sec and patients with a 50 percent contrast material dilution injected at a rate of 2.0 mL/sec.

Most notable among their comparison findings:

• The frequency of severe arterial phase ghosting artifact was significantly higher in patients who received gadoxetic acid via slow injection compared with patients who received the contrast agent via the dilution method (15.1 percent [24 of 159] vs. 3.8 percent [6 of 159]) (P < .001).
• In an image-based analysis, the estimated log odds of mean artifact score would decrease by 0.605 with the dilution method when compared with that attained with the slow injection method (P < .001).
• For the frequency of severe artifact, the estimated log odds of severe artifact would decrease by 1.780 with the dilution method compared with that attained with the slow injection method (P < .001).

From these findings the authors conclude that dilution of gadoxetic acid with saline is beneficial in minimizing the ghosting artifact seen on arterial phase gadoxetic acid-enhanced MR images.

In their discussion, Kim et al. mention an intriguing finding: In 18 MR scans of 14 patients with chronic obstructive pulmonary disease (COPD), a severe artifact turned up in three (23 percent) of the 13 MR scans with the undiluted method—but was not seen in the five MR imaging studies with dilution.

“Because of the limited statistical power of this small sample size, we could not determine any association between the individual factors that could predispose patients to respiratory motion and artifacts,” they write.