|Home | About | Journals | Submit | Contact Us | Français|
The characterization of iron stores is important to prevent and treat iron overload. Serum markers such as ferritin, serum iron, iron binding capacity, transferrin saturation, and nontransferrin-bound iron can be used to follow trends in iron status; however, variability in these markers limits predictive power for any given individual. Liver iron represents the best single marker of total iron balance. Measures of liver iron include biopsy, superconducting quantum interference device, computer tomography, and magnetic resonance imaging (MRI). MRI is the most accurate and widely available noninvasive tool to assess liver iron. The main advantages of MRI include a low-rate of variability between measurements and the ability to assess iron loading in endocrine tissues, the heart and the liver. This manuscript describes the principles, validation, and clinical utility of MRI for tissue iron estimation.
Iron overload can occur among patients with hereditary hemochromatosis, thalassemia, sickle cell disease, aplastic anemia, myelodysplasia, and other diseases. Excess iron absorption and transfusional iron intake cause iron accumulation in the liver, endocrine organs, heart, and other tissues with severe, life-threatening consequences. Iron cardiomyopathy is of particular concern, and remains the leading cause of death in patients with thalassemia major [1–4]. The characterization of iron stores is, therefore, important to prevent and treat iron overload in these patients. Imaging modalities and other techniques available for this purpose are discussed in this article.
Iron overload represents an imbalance of iron intake and iron elimination. To appropriately tailor iron removal therapies, the amount of iron entering a patient’s body must also be considered. In thalassemia intermedia and hereditary hemochromatosis, the progression of iron overload is modest and easily managed by phlebotomy or short-term chelation therapy. In contrast, chronic transfusion therapy delivers between 0.4 and 0.5 mg/kg/day of iron. Chronic transfusion required in patients with severe anemia syndromes including thalassemia major, myelodysplastic syndromes, Diamond-Blackfan. Routine transfusion therapy is also being used extensively in patients with sickle cell disease to prevent neurologic complications. Chronically transfused patients will become iron overloaded within 1 year of therapy and need iron chelation therapy to prevent deleterious consequences of iron overload. Iron chelators currently available include deferoxamine, which is administered subcutaneously or intravenously, and the oral chelators deferiprone and deferasirox.
Variations in transfusional requirements will also help determine appropriate chelator dosing. This was illustrated by a recent study in which patients treated with deferasirox or deferiprone were grouped into three transfusion regimens (<7 mL/kg/month, 7–14 mL/kg/month, and >14 mL/kg/month of red blood cells). Dose-responsiveness was observed in all three categories, but heavily transfused patients required nearly twice as much chelator as lightly transfused patients in order to maintain iron balance .
Several serum markers can be used to follow trends in a patient’s iron status over time. These include ferritin, serum iron, and nontransferrin-bound iron (NTBI) (Table I), as well as total iron binding capacity and transferrin saturation (TSAT). Ferritin is the most frequently used measure as it is inexpensive, widely available, and reliable, with extensive clinical validation in monitoring iron status. Ferritin measurements have prognostic value, as demonstrated by recent studies, one of which identified cardiac-related mortality greater than 80% over 15 years among patients with thalassemia in whom more than 67% of ferritin measurements exceeded 2,500 ng/mL [6,7]. However, in individuals the predictive value of ferritin is limited by inflammation and vitamin C deficiency. As a result, even patients with low ferritin levels experience elevated rates of heart disease as they age. Furthermore, the predictive value of ferritin is not documented in diseases other than thalassemia, such as myelodysplastic syndromes and sickle cell disease.
NTBI appears in the blood when transferrin is highly saturated so its presence can be predicted by TSAT values . NTBI is toxic to the liver, heart, and other endocrine tissues and increased blood levels may indicate developing organ toxicity in iron-overloaded patients. Quantification of TSAT is readily available, however, interlaboratory assay variability, rapid physiologic modulation by inflammation, and nonlinearity with respect to total body iron levels limit the practical usefulness of this measure.
NTBI is an intuitively appealing biomarker because it is responsible for parenchymal iron loading and toxicity. NTBI assayed periodically as labile plasma iron (LPI) can be used to monitor iron overload and a patient’s response to chelation therapy. This is demonstrated by a decrease in LPI over time in thalassemia intermedia patients . However, data regarding the use of LPI in transfused patients are limited, there is no standardized assay, and risky levels of LPI have yet to be identified.
Overall, NTBI levels differ considerably between assay methods. One study identified coefficients of variation ranging from 4 to 193% between tests, with differences due primarily to procedural variations, iron contamination, and variation in NTBI isoforms .
Liver iron levels accurately reflect total body iron stores because the liver is the dominant iron storage organ [10,11]. Liver iron levels have also been used to estimate risk, and predict outcomes such as liver failure, diabetes, heart failure, and death [7,12]. Methods for quantifying liver iron include biopsy, use of the superconducting quantum interference device (SQUID), computed tomography (CT), and magnetic resonance imaging (MRI).
Liver biopsy is the only direct assessment of liver iron and remains the standard of care in institutions without access to noninvasive surrogate techniques for iron measurement. It allows the assessment of liver histology, which is important in staging liver fibrosis, particularly in hepatitis C positive patients. In adults, liver biopsy can be performed as an outpatient procedure. However, liver biopsy is expensive and carries a 0.5% risk for serious bleeding . Post-procedure discomfort limits patient acceptance and sampling variability of the procedure is relatively high at 12–15% overall, and up to 40% among patients with cirrhosis [14,15].
SQUID was among the first noninvasive techniques used to measure body iron loading . Its limitations include high installation costs, limited availability, and limited utility as it measures only liver and spleen iron content. In addition, validation data for SQUID are limited.
CT is well-tolerated by patients and relatively inexpensive, so it has the potential for wide clinical use. However, application of this technique has been critically limited by lack of validation in humans, poor sensitivity in patients with low iron loads, and exposure to ionizing radiation.
Although the principles of using MRI for detection of iron have been known for the past 20 years, this technology has only recently become reproducible and routine for the measurement of iron in the liver and heart. MRI measures iron content in all organs, is widely available, and has been validated for measuring liver iron content. Recently, a specialized application of MRI (Ferriscan®, Resonance Health, Australia) was approved by the Food and Drug Administration for the measurement of liver iron. MRI limitations include expense, the need for trained personnel for acquisition and postprocessing, and the necessity of standardizing the technique prior to its implementation.
To measure hepatic iron concentration, an MRI scanner transmits a radio stimulus that excites water protons in hepatic tissue. Free diffusing water protons experience variations in the magnetic field produced by iron particles within the liver. Iron causes MRI images to darken at a rate proportional to the hepatic iron load, with the half-life of this darkening defined as T2*. The rate of darkening, designated as R2*, is the reciprocal of T2* and is proportional to the iron content of the tissues. MRI scanning estimates tissue iron concentration both by gradient echo imaging, which provides T2*, and spin echo imaging, which provides T2, the reciprocal of R2 .
Two studies have compared the amount of iron measured at liver biopsy to measurements of R2 and R2* with MRI. The first study measured R2 and identified a strong nonlinear relationship between R2 and hepatic iron concentration (Fig. 1) . In the second study, MRI measurements of both R2*and R2 with MRI showed good correlation with liver biopsy and interexam reproducibility was acceptable at 3–8% .
An important advantage of MRI is the low-rate of variability (typically 5–7%) between hepatic iron measurements, which is not attainable by liver biopsy. Therefore, MRI is a good noninvasive method for long-term monitoring of liver iron concentrations during chelation therapy. Another advantage of MRI is its ability to quantify iron loading in the heart and other endocrine tissues, which may occur in clinically significant amounts even during adequate chelation.
Cardiomyopathy is the most harmful manifestation of transfusional iron overload . Once cardiac symptoms develop, decompensation and death occur rapidly unless chelation therapy is intensified [1,4]. Cardiac iron is removed slowly from the myocardium, a factor that contributes to the high-mortality of patients with cardiomyopathy, even in the presence of intensive chelation . MRI evaluation of cardiac iron can help predict cardiac risk . A study showed that patients with T2* > 20 msec (no detectable cardiac iron) do not typically develop heart dysfunction, while patients with T2* < 10 msec are at a proportionally higher risk for cardiac dysfunction .
R2* exhibits a linear relationship to heart iron according to data from animal studies . In addition, although storage patterns of iron in the heart and in the liver were different, the MRI calibration for an R2* technique does not differ significantly . While human data on the correlation between R2/R2* and cardiac iron are limited, existing findings show that both R2* and R2 demonstrate a linear correlation of R2*, with the iron in cardiac tissue .
Liver iron concentrations correlate poorly with those in cardiac tissue because the mechanisms of iron uptake and clearance differ between organs. In particular, iron is deposited and removed more quickly from the liver than from cardiac tissue, creating hysteresis between measured iron levels in these tissues. Many patients, particularly adolescents, can have high liver iron without detectable cardiac iron. If this situation exists long-term, cardiac iron begins to accumulate, even in the absence of additional hepatic iron loading. Conversely, intensive chelation can clear iron from the liver fivefold more quickly than the heart. Therefore, a patient may have high cardiac iron despite a lower total body iron burden following chelation therapy.
While high liver iron is clearly a risk factor for cardiac iron accumulation, it is important to recognize that there is no hepatic iron concentration at which cardiac iron deposition does not occur. Two patient cases illustrate this point, both of whom were compliant with chelation therapy and presented initially without cardiac iron by MRI (Fig. 2) . The first patient had an initial liver iron level of 5 mg/g dry weight that increased to 7 mg/g dry weight, while the patient’s cardiac iron concentration tripled over the same time period. The second patient had an initial liver iron of 2 mg/g dry weight that decreased to 1 mg/g dry weight but had detectable cardiac iron 1 year later despite chelation therapy. These cases emphasize that there are mechanisms other than overwhelming iron saturation of the liver, that lead to cardiac iron loading.
All patients on chronic transfusion therapy require serial monitoring of iron stores. A logical paradigm for monitoring body iron loading in these patients is to measure ferritin levels at least quarterly, and iron panels once each year (Table II). Liver iron should be measured annually (either by biopsy or noninvasively), plus every 3–6 months in patients who are intensively chelated for heart failure. If MRI techniques are available, cardiac iron and cardiac function should also be measured by MRI yearly, plus every 6 months in patients chelated intensively.
Blood transfusion burden is an important measure of total body iron balance. Ferritin is a relatively inexpensive and widely-available measure, useful in monitoring chelation therapy. LPI measurements show promise for predicting endocrine and cardiac iron toxicity, although existing LPI assays require more refinement, standardization, and clinical validation. Liver iron concentration reflects total body iron stores, but incompletely stratifies the risks of iron overload complications. MRI offers the most accurate and widely available noninvasive tool for assessing liver iron concentration. As barriers to broad implementation of MRI are overcome, comprehensive MRI assessment of liver, and cardiac iron and cardiac function is likely to become the standard of care in iron overload.
The author is fully responsible for contents and editorial decisions for this manuscript.
Contract grant sponsor: National Heart Lung and Blood Institute; Contract grant number: 1 RO1 HL075592-01A1; Contract grant sponsor: General Clinical Research Center at Childrens Hospital Los Angeles; Contract grant number: RR000043-43; Contract grant sponsor: Center for Disease Control; Contract grant number: U27/CCU922106 (Thalassemia Center Grant); Contract grant sponsors: Novartis Pharma; Department of Pediatrics at Childrens Hospital Los Angeles; Novartis Oncology.