van Nimwegen Frederika A, Cutter David J, Schaapveld Michael, Rutten Annemarieke, Kooijman Karen, Krol Augustinus D G, Janus Cécile P M, Darby Sarah C, van Leeuwen Flora E, Aleman Berthe M P
Department of Psychosocial Research, Epidemiology, and Biostatistics, The Netherlands Cancer Institute, Amsterdam, The Netherlands.
Clinical Trial Service Unit, University of Oxford, Oxford, United Kingdom; Oxford Cancer Centre, Oxford University Hospitals NHS Trust, Oxford, United Kingdom.
Int J Radiat Oncol Biol Phys. 2015 May 1;92(1):153-60. doi: 10.1016/j.ijrobp.2015.02.019.
To describe a new method to estimate the mean heart dose for Hodgkin lymphoma patients treated several decades ago, using delineation of the heart on radiation therapy simulation X-rays. Mean heart dose is an important predictor for late cardiovascular complications after Hodgkin lymphoma (HL) treatment. For patients treated before the era of computed tomography (CT)-based radiotherapy planning, retrospective estimation of radiation dose to the heart can be labor intensive.
Patients for whom cardiac radiation doses had previously been estimated by reconstruction of individual treatments on representative CT data sets were selected at random from a case-control study of 5-year Hodgkin lymphoma survivors (n=289). For 42 patients, cardiac contours were outlined on each patient's simulation X-ray by 4 different raters, and the mean heart dose was estimated as the percentage of the cardiac contour within the radiation field multiplied by the prescribed mediastinal dose and divided by a correction factor obtained by comparison with individual CT-based dosimetry.
According to the simulation X-ray method, the medians of the mean heart doses obtained from the cardiac contours outlined by the 4 raters were 30 Gy, 30 Gy, 31 Gy, and 31 Gy, respectively, following prescribed mediastinal doses of 25-42 Gy. The absolute-agreement intraclass correlation coefficient was 0.93 (95% confidence interval 0.85-0.97), indicating excellent agreement. Mean heart dose was 30.4 Gy with the simulation X-ray method, versus 30.2 Gy with the representative CT-based dosimetry, and the between-method absolute-agreement intraclass correlation coefficient was 0.87 (95% confidence interval 0.80-0.95), indicating good agreement between the two methods.
Estimating mean heart dose from radiation therapy simulation X-rays is reproducible and fast, takes individual anatomy into account, and yields results comparable to the labor-intensive representative CT-based method. This simpler method may produce a meaningful measure of mean heart dose for use in studies of late cardiac complications.
描述一种新方法，通过在放射治疗模拟X射线上勾勒心脏轮廓，来估算几十年前接受治疗的霍奇金淋巴瘤患者的平均心脏剂量。平均心脏剂量是霍奇金淋巴瘤（HL）治疗后晚期心血管并发症的重要预测指标。对于在基于计算机断层扫描（CT）的放射治疗计划时代之前接受治疗的患者，回顾性估算心脏的辐射剂量可能需要大量人力。
从一项针对5年霍奇金淋巴瘤幸存者的病例对照研究（n = 289）中随机选择患者，这些患者之前已通过在代表性CT数据集上重建个体治疗来估算心脏辐射剂量。对于42名患者，由4名不同的评估者在每位患者的模拟X射线上勾勒出心脏轮廓，并将平均心脏剂量估算为心脏轮廓在辐射野内的百分比乘以规定的纵隔剂量，再除以通过与基于个体CT的剂量测定法比较获得的校正因子。
根据模拟X射线方法，在规定的纵隔剂量为25 - 42 Gy的情况下，4名评估者勾勒出的心脏轮廓所获得的平均心脏剂量中位数分别为30 Gy、30 Gy、31 Gy和31 Gy。组内绝对一致性相关系数为0.93（95%置信区间0.85 - 0.97），表明一致性极佳。模拟X射线方法得出的平均心脏剂量为30.4 Gy，而基于代表性CT的剂量测定法得出的为30.2 Gy，两种方法之间的组内绝对一致性相关系数为0.87（95%置信区间0.80 - 0.95），表明两种方法之间一致性良好。
从放射治疗模拟X射线估算平均心脏剂量具有可重复性且快速，考虑了个体解剖结构，并且产生的结果与需要大量人力的基于代表性CT的方法相当。这种更简单的方法可能会得出用于晚期心脏并发症研究的有意义的平均心脏剂量测量值。
Zotero 插件