CBCT引导下宫颈癌患者IMRT中影响摆位误差的相关因素研究

Related factors affecting the positioning error in IMRT of patients with cervical cancer under the guidance of CBCT

目的探讨锥形束CT(CBCT)引导下宫颈癌IMRT中3种不同的配准方式和不同腹壁厚度、腹围对配准的影响,为临床选择合理的配准模式提供科学依据。方法回顾性分析2017-06-01-2019-10-01川北医学院附属医院放疗科放疗的宫颈癌患者215例,分别采用自动骨性配准(t+r)、灰度配准(t+r)和灰度配准(t)方式进行配准。患者根据不同的腹壁厚度分为<20 mm组(57例)、20~30 mm组(107例)和>30 mm(51例)组,根据不同腹围分为<800 mm组(78例)、800~900 mm组(102例)和>900 mm组(35例)。3种配准方式之间平移误差比较采用单因素方差分析,组间比较采用LSD-t检验。应用多因素方差分析及LSD-t检验对比分析腹壁厚度及腹围不同分组分别在3种配准方式下的摆位误差。结果所有患者行CBCT引导的IMRT配准平移误差从大到小为,头脚(Y轴),腹背(Z轴),左右(X轴)。Y轴3种配准方式摆位误差,差异有统计学意义,F=4.52,P<0.05。不同腹壁厚度组患者,配准方式相同,摆位误差在X轴(F=32.98,P<0.05)、Y轴(F=96.85,P<0.05)、Z轴(F=20.75,P<0.05)差异有统计学意义。腹壁厚度同组患者,不同配准方式Y轴摆位误差差异有统计学意义,F=83.39,P<0.05。不同腹围组患者,配准方式相同,摆位误差在X轴(F=377.70,P<0.05)、Y轴(F=23.43,P<0.05)、Z轴(F=80.47,P<0.05)差异有统计学意义。腹围同组患者,不同配准方式,摆位误差在X轴(F=17.71,P<0.05)、Y轴(F=15.74,P<0.05)、Z轴(F=31.39,P<0.05)差异有统计学意义。结论灰度配准(t)是宫颈癌患者行CBCT引导的IMRT的优选配准方式。患者腹壁越厚、腹围越大摆位误差越明显,肥胖患者尤其是腹壁厚度>3 mm、腹围>900 mm应适当增加PTV的外扩边界,需行每日配准。

Objective To explore the effects of three different registration methods and different abdominal wall thickness and abdominal circumference on the registration in Cone-beam Computed Tomograghy guided intensity-modulated radiotherapy for cervical cancer,so as to provide scientific basis for clinical selection of reasonable registration mode.Methods A total of 215 patients with cervical cancer who received radiotherapy in the Department of radiotherapy,Affiliated Hospital of North Sichuan Medical College from 2017-06-01 to 2019-10-01 were retrospectively analyzed.Automatic bony registration(t+r),automatic grayscale registration(t+r)and automatic grayscale registration(t)were used for registration.At the same time,the patients were divided into groups according to different abdominal wall thickness and abdominal circumference,abdominal wall thickness<20 mm group(n=57),20-30 mm group(n=107),>30 mm group(n=51);abdominal circumference<800 mm group(n=78),800-900 mm group(n=102),and>900 mm group(n=35).The translation errors between the 3 registration methods was compared by one-way analysis of variance,and the comparison between groups was by LSD-t test.The positioning errors of different groups under three registration modes were compared and analyzed by multi-factor analysis of variance and LSD-t test.Results In the intensity modulated radiotherapy guided by CBCT,the translation error of registration was the largest in the direction of head and foot(Y axis),the second in the direction of ventral dorsal(Z axis),and the smallest in the direction of left and right(X axis).There was statistically significant difference in the positioning error of the three registration methods on the Y axis(F=4.52,P<0.05).Under the same registration mode,the positioning errors of patients with different abdominal wall thickness were significantly different on X axis(F=32.98,P<0.05),Y axis(F=96.85,P<0.05)and Z axis(F=20.75,P<0.05).When the thickness of abdominal wall was the same,the positioning errors of different registration methods were significantly different on the Y axis(F=83.39,P<0.05).In the case of the same registration mode,there were significant differences in the positioning of patients with different abdominal circumference in X axis(F=377.70,P<0.05),Y axis(F=23.43,P<0.05)and Z axis(F=80.47,P<0.05).Under the same abdominal circumference group,the differences of the position of the X axis were statistically significant(F=17.71,P<0.05),Y axis(F=15.74,P<0.05),and Z axis(F=31.39,P<0.05).Conclusions Gray-scale registration(t)is optimized in the registration mode of CBCT-guided intensity modulated radiotherapy for cervical cancer patients.The thicker the abdominal wall and the larger the abdominal circumference,the more obvious the positioning error is.For obese patients(especially the abdominal wall thickness>3 mm,abdominal circumference>900 mm),it is recommended to increase the external expansion boundary of PTV,and it is necessary to carry out daily registration.