Liver iron measurement by MRI

Comment on St. Pierre et al, page 855

St. Pierre and colleagues have developed a magnetic resonance imaging (MRI)–based methodology for liver iron determination that is both accurate and sensitive over a wide concentration range of liver iron. The method offers the prospect of a reliable noninvasive determination of liver iron in most hospitals worldwide that have an up-to-date MRI facility.

The monitoring of iron overload and its therapy has hitherto been troublesome because the tools available have been imprecise, difficult to use, or unavailable. Serum ferritin, used since the 1970s to estimate body iron loading is flawed because factors other than iron overload, such as inflammation and liver damage, affect plasma levels. Measurement of liver iron is clinically useful because total body iron can be precisely predicted mathematically from the liver iron concentration.¹  Furthermore, sustained high levels of liver iron (> 15 mg/g dry weight) have been shown to be of prognostic value in predicting the cardiac complications of iron overload.²  A reliable and widely applicable method for measuring liver iron is therefore desirable in management of iron overload conditions.

Liver biopsy can been used successfully to measure liver iron but is invasive and inconvenient to perform frequently. Measurement of liver iron by magnetic susceptometry using a SQUID (superconducting, quantum, interface device), although accurate and precise, requires specialized and expensive facilities, which are available in only 4 centers worldwide.

The prospect of a simple measurement, which can be applied in any hospital having access to a suitable magnetic resonance imaging (MRI) scanner, is therefore highly desirable. A number of approaches to MRI estimation of tissue iron have previously been evaluated. Gradient echo sequences such as T₂*³  are better suited to estimation of iron in tissues where iron concentrations are lower than in the liver, such as in the heart. This article and others⁴  suggest that spin echo (T₂ or R₂) sequences may offer the preferred approach for liver iron determination. In this issue of Blood the method of St. Pierre and colleagues appears to give excellent sensitivity and specificity over a wide concentration range of liver iron. Data obtained using this method can be analyzed at the facility where the data is acquired or can be sent electronically for central analysis, thereby offering improved quality control and eliminating the requirement to train local radiologists in data analysis. The current method requires 20 minutes for acquisition of data, which is slower than recently published T₂* methodology using a single breath-hold.⁵  Development of a shorter acquisition time would be desirable and allow a wider range of patients to be measured, particularly young children. ▪