Correlation of liver iron concentration determined by R2 magnetic resonance imaging with serum ferritin in patients with thalassemia intermedia

Thalassemia intermedia is a highly diverse group of thalassemia syndromes associated with anemia and a range of specific complications, such as extramedullary hematopoiesis, leg ulcers, gallstones and a hypercoagulable state, which are uncommon in patients with thalassemia major.1 The degree of anemia present in patients with thalassemia intermedia is typically mild and generally does not require regular blood transfusion therapy. However, patients can still be at risk of the clinical sequelae of iron overload (as commonly seen in regularly transfused thalassemia major patients) due to increased intestinal iron absorption triggered by chronic anemia, ineffective erythropoiesis and, possibly, decreased serum hepcidin.2,3 The principal methods of determining body iron levels are measurement of serum ferritin levels and assessment of liver iron concentration from biopsy tissue. Non-invasive approaches for determining liver iron concentration are increasingly used as an alternative to biopsy, although R2 magnetic resonance imaging (MRI) is currently the only validated approach.4,5 A significant correlation between serum ferritin and liver iron concentration has been established in regularly transfused patients with thalassemia major.6,7 Data of patients with thalassemia intermedia are limited, but recent studies have highlighted differences compared with the studies performed in thalassemia major patients.3,8 In these studies, serum ferritin levels were seen to be significantly lower ni patients with thalassaemia intermedia than in those with thalassemia major, despite comparable liver iron concentration (as evaluated by biopsy or superconducting quantum interference device). The aim of our study was to investigate the correlation between liver iron concentration determined by R2 MRI and serum ferritin levels in patients with thalassemia intermedia. The data reported here represent the largest investigation of this correlation in thalassemia intermedia using R2 MRI and, therefore, provide valuable information on the relationship between these parameters in this specific patient population.

Table 1.Patients’ characteristics.

This was a cross-sectional study of randomly selected thalassemia intermedia patients treated at a chronic care center in Hazmieh, Lebanon. The sampling frame consisted of 120 thalassemia intermedia patients ≥2 years of age. We were able to contact 109 of these patients by telephone and 74 agreed to participate. Patient charts were reviewed and a medical history compiled, which included details of drug history, co-morbid illnesses and transfusional history. Data from a randomly selected population of patients with thalassemia major treated at the center were also obtained for comparative evaluation. Blood samples were obtained for assessment of serum ferritin levels. Direct determination of iron burden was performed using R2 MRI to obtain liver iron concentration values, using established methodology.9 The reading of MRI results was performed by Dr. Tim St Pierre. Written informed consent was provided by all patients. Data from 74 thalassemia intermedia patients were included in the analysis (Table 1). Transfusion-naïve patients had significantly lower iron levels compared to those with a history of transfusion therapy (p=0.003). None of the patients were receiving iron chelation therapy at the time of data collection and had not received chelation therapy for at least two years prior to study entry. In addition, none of the patients involved were positive for hepatitis B or C virus or had elevated alanine aminotransferase levels. Iron levels in the study population demonstrate that many patients with thalassemia intermedia have serum ferritin and liver iron concentration levels above the recommended threshold levels identified in patients with thalassemia major, indicating a risk of significant morbidity and mortality. A positive correlation between age and serum ferritin levels (R=0.29; p=0.01), and between age and liver iron concentration (R=0.35; p=0.002) was observed. After standardizing for serum ferritin level, a near significant correlation between age and liver iron concentration was observed (R=0.23; p=0.055). Serum ferritin levels were, therefore, seen to increase with age, reflecting increased iron accumulation over time, even in the absence of transfusion therapy. While this observation is in accordance with one previous study,10 others have not reported increasing iron levels.8,11 This highlights the variability in iron loading in patients with thalassemia intermedia and the need for a more accurate assessment of iron burden in these patients.

Figure 1.Serum ferritin versus liver iron concentration as determined by R2 MRI (line equation for thalassemia intermedia: serum ferritin = 67.8*liver iron concentration; thalassemia major: serum ferritin=106.6*liver iron concentration).

A significant positive correlation between mean serum ferritin and liver iron concentration values was observed in our study (R=0.64; p<0.001). Previous observations are limited, but two studies have been performed, one demonstrating a statistically significant correlation,8 the other showing no significant correlation.3 The relationship between liver iron concentration and serum ferritin has previously been shown to lack significance in splenectomized thalassemia intermedia patients,3 but data from our study show a positive correlation in this subset (R=0.62; p<0.001), thereby supporting the use of liver iron concentration assessment in splenectomized thalassemia intermedia patients.

Comparative data from the patients with thalassemia major (n=65; 36 male, 29 female; mean age 18.35 [SD ± 7.33; range 7–44]) showed mean liver iron concentration values of 9.2 mg Fe/g dry weight, (SD ± 8.57; range 0.8–47.3) and mean serum ferritin values of 2292 ng/mL (SD ± 1461; range 146–6,320). Therefore, while the mean liver iron concentration values were similar in both the thalassemia intermedia and major groups, serum ferritin levels were statistically different, with serum ferritin being significantly lower (p=0.003) in the thalassemia intermedia group (Figure 1). Therefore, as demonstrated in previous studies, evaluation of serum ferritin levels appears to underestimate the extent of iron overload in the thalassemia intermedia population.3,8 It has been suggested that, in transfused patients, iron is preferentially distributed to the reticuloendothelial system and that ferritin synthesis and release is responsible for higher serum ferritin levels.8 In contrast, in non-transfused patients, iron accumulated as a result of hyperabsorption is accumulated in hepatocytes and, therefore, lower serum ferritin levels are seen.8 Due to limited resources, assessment of liver iron concentration is often not possible in our region. Where serum ferritin levels provide the only available indication of iron levels, we would suggest that levels below 1,000 ng/mL should not be used as a negative predictor of significant iron overload in patients with thalassemia intermedia.

In conclusion, the data presented confirm that serum ferritin levels do not accurately reflect the level of iron overload in patients with thalassemia intermedia. These observations have important implications for patient management, as assessment of serum ferritin alone may result in a delay in initiating chelation therapy and may, therefore, prolong patient exposure to high iron levels and the associated morbidity and mortality risks. Unlike thalassemia major, in which transfusion history can be a useful indicator as to whether iron chelation therapy is required, patients with thalassemia intermedia will require accurate assessment of body iron levels in order to guide therapy. As assessment of serum ferritin is evidently inappropriate in these patients, disease-specific recommendations for the management of patients with thalassemia intermedia should include direct assessment of liver iron concentration by biopsy or non-invasive imaging methods.