Advances in Biomagnetic Liver Susceptometry Allow the Measurement of Liver Iron Concentration with a Room Temperature Sensor

Introduction: Iron overload is frequently observed in diverse states ranging from thalassemia, sickle cell disease, hereditary hemochromatosis, transfusion-dependent anemias, cancer chemotherapy and chronic liver disease. Management of iron overload depends on the ability to quantify and monitor the patient's iron stores with precision. Organ iron measurement by relaxometry-based MRI techniques has become the current standard. MRI is expensive and has the added limitations of multiple existing methods and reduced dynamic range with 3 Tesla scanners. Liver iron measurements by magnetic susceptometers using superconductive quantum interference device (SQUID) technology are expensive and have limited availability. In this study an improved low-cost device, the room-temperature susceptometer (RTS, Insight Magnetics, San Diego, CA), was tested against the Model 5700 Ferritometer ® 3-Channel SQUID BioSusceptometer System (Tristan Technologies, Inc. San Diego, CA). Both systems quantify liver iron concentration (LIC) using bulk magnetic susceptibility.

Where the SQUID uses an ultra-stable sensing system immersed in liquid helium, the RTS cancels the temperature and magnetic-field fluctuations inherent in an apparatus that works at room temperature. The RTS uses three main techniques to sense the very weak magnetic field produced by liver iron: (a) oscillatory magnetic fields that can be detected with high sensitivity using coils of ordinary copper wire, (b) field-producing and field-sensing coils to cancel the signal due to the applied magnetic field, and (c) movement of the sensing unit periodically toward and away from the patient so as to distinguish the patient's magnetic field response from the interfering signal caused by temperature fluctuations in the sensing system.

Methods: This study compared measurements of LIC from RTS to those by the SQUID. The RTS in this study was modified from an earlier model to make the baseline reading more stable, and to increase the accuracy of the water reference measurement to which the patient's magnetic response is compared. The magnetic-field source and sensing coils were enlarged to increase the signal of the liver compared with that of the overlying tissue.

LIC was measured once on the SQUID and once on the RTS at a single visit. Using ultrasound imaging, optimum liver measurement position was determined and marked with an x-y-positioning and z-distance sensing Locator Loop (Positronic Systemtechnik GmbH, Germany). The locator loop remained attached to the patient for measurements with both devices to preserve measurement location. LIC calculation was corrected for the susceptibility and geometry of the overlying tissue. Measurement results from SQUID and RTS were analyzed independently by two investigators.

Results: Thirteen adults (10 with transfusion-dependent thalassemia and 3 controls) with body mass index (BMI) <25 were enrolled. All measurements were completed at a single visit with no failures. LIC values (µg/g wet-liver weight), ranged from -33 to 6493 with SQUID, and -305 to 7237 with RTS. In the three controls, the LIC was -33, 144 and 427 µg/g with SQUID, and -305, 451 and 230 µg/g with RTS. The overall correlation between the two methods was excellent, yielding an r² = 0.976 and slope = 1.037 ± 0.068 (p<0.001, Fig 1). Bland-Altman analysis of percent-difference versus the average of the two methods showed bias -2.72 (95% limits of agreement -149.1 to 143.6) for all subjects, which improved to 1.94 (95% limits of agreement -41.1 to 45.3) when the average values below 350 µg/g (n=4) were excluded (Fig 2). The percent difference between the two methods was influenced by the subject's BMI (p=0.050 for all subjects; p=0.031 after excluding average LIC <350 µg/g), with the least difference observed in the BMI range of 20-23 Kg/m².

Conclusion: This study shows that, with the recent improvements in the RTS technology, LIC measurements now closely align with those using SQUID. The remaining difference between the two methods likely results from the models used to compensate for the overlying tissue. Further comparison of RTS to SQUID and MRI-based methods in diverse iron overload states is warranted in a larger study. This work may ultimately make low-cost noninvasive measurement of iron overload accessible to the large number of patients in the US, and to resource-limited countries around the world.

View largeDownload slide

View largeDownload slide

Close modal