Magnetic resonance imaging (MRI) should be the gold standard for measuring liver iron concentration (LIC) in patients receiving ongoing transfusion therapy, according to a group of researchers.
The team found evidence suggesting that liver MRI is more accurate than liver biopsy in determining total body iron balance in patients with sickle cell disease and other disorders requiring regular blood transfusions.
The findings have been published in Magnetic Resonance Imaging.
“Measuring total body iron using MRI is safer and less painful than biopsy,” said study author John Wood, MD, PhD, of Children’s Hospital Los Angeles in California.
“In this study, we’ve demonstrated that it is also more accurate. MRI should be recognized as the new gold standard for determining iron accumulation in the body.”
Dr Wood and his colleagues came to this conclusion after analyzing data from 49 patients who were undergoing treatment with deferitazole, an experimental chelating agent.
The team looked at the amount of iron the patients were receiving by transfusion and the amount of chelating agent they consumed, providing insights into the expected changes in iron levels at 12, 24, and 48 weeks after the start of deferitazole.
To compare MRI and liver biopsy, the researchers used serial estimates of iron chelation efficiency (ICE) calculated by R2 and R2* MRI LIC estimates as well as by simulated liver biopsy (over all physically reasonable sampling variability).
The estimates suggested that MRI liver iron measurements (R2 and R2*) are more accurate than a physically realizable liver biopsy (which has a sampling error of 10% or higher).
For R2, the standard deviation of ICE was 44.8% at 12 weeks, 14.8% at 24 weeks, and 7.4% at 48 weeks. For R2*, the standard deviation was 22.9% at 12 weeks, 9.3% at 24 weeks, and 5.7% at 48 weeks.
For a biopsy with a sampling error of 0%, the standard deviation of ICE was 25.9% at 12 weeks, 12.8% at 24 weeks, and 6.7% at 48 weeks. For a biopsy with a sampling error of 10%, it was 32.8%, 16.5%, and 8.7%, respectively.
For a biopsy with a sampling error of 20%, the standard deviation of ICE was 49.1% at 12 weeks, 24.9% at 24 weeks, and 13% at 48 weeks. For a biopsy with a sampling error of 30%, it was 68.1%, 34.5%, and 18%, respectively. And for a biopsy with a sampling error of 40%, it was 88.1%, 44.6%, and 23.2%, respectively.
In practice, the accuracy of MRI compared to liver biopsy is likely even greater than these estimates suggest, Dr Wood said.
