Wood John C, Zhang Pinggao, Rienhoff Hugh, Abi-Saab Walid, Neufeld Ellis J
Department of Pediatrics and Radiology, Children's Hospital Los Angeles.
Shire Pharmaceuticals, Chesterbrook, PA.
Magn Reson Imaging. 2015 Jul;33(6):761-7. doi: 10.1016/j.mri.2015.02.016. Epub 2015 Feb 20.
Liver biopsy was long considered the reference standard for measuring liver iron concentration. However, its high sampling variability and invasive nature make it poorly suited for serial analyses. To demonstrate the fallibility of liver biopsy, we use serial estimates of iron chelation efficiency (ICE) calculated by R2 and R2* MRI liver iron concentration (LIC) estimates as well as by simulated liver biopsy (over all physically reasonable sampling variability) to compare the robustness of these three techniques.
R2, R2*, transfusional volume, and chelator compliance were obtained from 49 participants in a phase II clinical trial of deferitazole over two years. Liver biopsy LIC results were simulated using sampling errors of 0%, 10%, 20%, 30%, 40% and iron assay variability of 12%. LIC estimates by R2, R2*, and simulated biopsy were used to calculate ICE over time. Bland-Altman limits of agreement were compared across observation intervals of 12, 24, and 48 weeks.
At 48 week intervals, LIC estimates by R2, R2* and "perfect" liver biopsy had comparable accuracy in predicting ICE; both MRI methods were superior to any physically realizable liver biopsy (sampling error 10% or higher). LIC by R2* demonstrated the most robust ICE estimates at monitoring intervals of 24 and 12 weeks, but this difference did not remain significant at 48 week intervals.
MRI relaxometry is superior to liver biopsy for serial LIC observations, such as used in the care of tranfusional siderosis patients, and should also be considered the new standard of LIC determination for regulatory purposes. Among relaxometry techniques, LIC estimates by R2* are more robust for tracking changes in iron balance over intermediate time scales (<=24 weeks).
长期以来，肝活检一直被视为测量肝脏铁浓度的参考标准。然而，其高度的采样变异性和侵入性使其不太适合进行系列分析。为了证明肝活检的不可靠性，我们使用通过R2和R2*磁共振成像（MRI）肝脏铁浓度（LIC）估计值以及模拟肝活检（考虑所有物理上合理的采样变异性）计算出的铁螯合效率（ICE）系列估计值，来比较这三种技术的稳健性。
在一项为期两年的去铁唑II期临床试验中，获取了49名参与者的R2、R2*、输血体积和螯合剂依从性数据。使用0%、10%、20%、30%、40%的采样误差以及12%的铁测定变异性来模拟肝活检LIC结果。通过R2、R2*和模拟活检得到的LIC估计值用于计算随时间变化的ICE。在12周、24周和48周的观察间隔内比较布兰德-奥特曼一致性界限。
在48周的间隔时，通过R2、R2和“完美”肝活检得到的LIC估计值在预测ICE方面具有相当的准确性；两种MRI方法均优于任何实际可行的肝活检（采样误差为10%或更高）。在24周和12周的监测间隔时，R2得到的LIC显示出最稳健的ICE估计值，但在48周间隔时这种差异不再显著。
对于系列LIC观察，如用于输血性铁过载患者的护理，MRI弛豫测量法优于肝活检，并且出于监管目的，也应被视为LIC测定的新标准。在弛豫测量技术中，R2*得到的LIC估计值在跟踪中等时间尺度（<=24周）内铁平衡变化方面更稳健。
