不同全脑全脊髓放疗方式的剂量学比较及摆位误差对靶区的影响

目的 比较不同全脑全脊髓放射治疗（CSI）方式的剂量学差异及摆位误差对靶区剂量的影响。方法 选取2011年7月至2012年10月间接受CSI的9例患者的CT图像，分别完成常规二维（2D）、单野三维（3D-1）、三野三维（3D-3）、逆向调强（IMRT）及电子线的CSI计划，比较不同计划间靶区的覆盖度（V₉5）、高量（V₁₀₇）、最大剂量（D_max）、适形指数（CI）、剂量均一性指数（HI）；观察甲状腺、心脏、双肺、小肠、肾脏及全身正常组织5、15、25 Gy的受照体积。取患者每周的3个治疗中心（头部、上段脊髓、下段脊髓）的摆位误差值，将该周5次放疗计划的治疗中心分别按照此值移动得到新的放疗计划，比较不同计划方式新计划的靶区剂量与原计划的差异。结果 电子线V₉₅略低于其余各组（q=11.2~11.7，P<0.05）。IMRT具有最小的V₁₀₇（q=4.3~11.6，P<0.05），其次为3D-3（q=4.3~7.1，P<0.05）；2D具有最大的D_max（q=2.4~2.7，P<0.05）；各组HI的差异无统计学意义；靶区CI从高至低依次为，IMRT>3D-3>2D、3D-1及电子线（q=7.1~14.3、7.1~9.6、0.00~0.01，P<0.05）。IMRT及电子线可以显著降低各器官及全身组织接受的15 Gy及25 Gy剂量；3D-3次之。但与2D及3D-1相比，3D-3及IMRT均不同程度增加了5 Gy的照射体积。引入摆位误差后，3D-1及3D-3靶区剂量与原计划差别小于其余各组（q=2.8~4.1，P<0.05）。结论 对于脊髓深度<4.5 cm者，电子线有可能是一种安全、可靠的治疗方式。3D-1虽适形度略差，但有减少摆位误差的影响、降低低剂量体积的趋势，仍为可考虑治疗方式之一。3D-3及IMRT显示了较好的靶区剂量分布，但其大范围低剂量体积需引起重视。

The dosimetric comparison of different craniospinal irradiation plans and impact of setup error on target dose distribution

Objective To make a dosimetric comparison of different craniospinal irradiation (CSI) plans and to explore the impacts of setup error on target dose distribution.Methods Five radiotherapy plans[2D-conventional,single filed 3D(3D-1),three fields 3D(3D-3),IMRT and electron] were made from 9 patients' CT images, who received CSI from July 2011 to October 2012. The target coverage (V₉₅), hot spot(V₁₀₇), maximum dose (D_max), conformal index (CI) and dose homogeneity index (HI) of the target of each plan were compared. The volume of thyroid, heart, lungs, small intestine, kidney and normal tissue irradiated by 5,15,25 Gy were observed. The setup errors of the three treatment centers(head, upper spinal, lower spinal) per week were recorded and the treatment centers of the five plan in the week were moved according to the collected data to make a new plan, then the comparison of the target doses with the original plans were performed. Results The V₉₅ of electron was lower than the other groups (q=11.2-11.7,P<0.05), the difference between the rest of the group was not statistically significant (q=0.00-0.01,P> 0.05). IMRT had the smallest V₁₀₇ (q=4.3-11.6,P<0.05), followed by the 3D-3 (q=4.3-7.1,P<0.05), 2D had the largest D_max (q=2.4-2.7,P<0.05). The differences from HI in each group was not statistically significant.CIs from high to low were IMRT> 3D-3> 2D, 3D-1 and electron (q=7.1-14.3, 7.1-9.6, 0.00-0.01, P<0.05). IMRT and electron could significantly reduce the volume irradiated by 15 Gy and 25 Gy and 3D-3 was the second. However, compared with 2D and 3D-1, there were 5 Gy irradiation volume increased in 3D-3 and IMRT. When setup errors were taken into account, there was the smallest difference between the dose distribution of 3D-1, 3D-3 and the original plan(q=2.8-4.1,P<0.05). Conclusions For those spinal cord depth<4.5 cm, electron could be regarded as a kind of safe and reliable way of CSI. Although the CI of 3D-1 is slightly less, it still could be used as the choice of CSI with the trend of less being affected by setup errors and reducing low dose volume. While the 3D-3 and IMRT have better target dose distribution, the large volume of low dose region is a thought-provoking problem.