Correlation of serum ferritin levels with hepatic MRI T2 and liver iron concentration in nontransfusion beta-thalassemia intermediate patients: A contemporary issue

ABSTRACT

Background: Beta-thalassemia intermediate is a genetic disease that is milder than beta-thalassemia major. The T2^* magnetic resonance imaging (MRI) technique is currently the gold standard for iron load detection. However, it is expensive and needs an expert radiologist to report findings. Therefore, we conducted this study to determine an optimal cut-off value of ferritin in proportion to T2 MRI of liver and measurement of liver iron concentration for early detection of hepatic iron overload in Beta-thalassemia intermediate patients. Methods: This cross-sectional study was conducted on 108 patients with Beta-thalassemia intermediate who referred to tertiary hospital, Shiraz, Iran. Serum ferritin, hepatic T2 MRI, and liver iron concentration were assessed. Receiver operator characteristic was used to determine the sensitivity and specificity of cut-off value. Results: Serum ferritin levels showed a statistically significant negative correlation with T2 hepatic MRI (r = −0.290, p value =.003) and positive correlation with liver iron concentration (r = 0.426, p value <.001) in the patients with Beta-thalassemia intermediate. According to the receiver operator characteristic, the best cut-off value for ferritin to show early diagnosis of liver iron overload was 412 ng/mL. Calculated sensitivities and specificities were 0.78 and 0.82 for T2 MRI and 0.76 and 0.86 for liver iron concentration, respectively. Conclusion: Serum ferritin levels of around 450 ng/mL might be considered as a cut-off point to evaluate hepatic iron overload before using expensive, not readily available T2 MRI. This level of serum ferritin could be considered for starting iron chelation therapy in patients with Beta-thalassemia intermediate in areas where T2 MRI is not available.

KEYWORDS:

• Beta-thalassemia intermediate
• liver iron concentration
• serum ferritin

Acknowledgments

We would like to thank Shiraz University of Medical Sciences for financial support. Also, we thank Sheryl Nikpoor for editing and improving the use of English in the manuscript and Shirin Parand for preparation of the manuscript.

Declaration of interest

All authors declare that they have no conflicts of interest.