George R, Ramakrishnan V, Siebers J V, Chung T D, Keall P J
Department of Radiation Oncology, Virginia Commonwealth University, Richmond, VA, USA.
Phys Med Biol. 2006 Oct 21;51(20):5305-19. doi: 10.1088/0031-9155/51/20/015. Epub 2006 Oct 2.
Respiratory gating can reduce the apparent respiratory motion during imaging and treatment; however, residual motion within the gating window remains. Respiratory training can improve respiratory reproducibility and, therefore, the efficacy of respiratory-gated radiotherapy. This study was conducted to determine whether residual motion during respiratory gating is affected by patient, tumour or treatment characteristics. The specific aims of this study were to: (1) identify significant characteristics affecting residual motion, (2) investigate time trends of residual motion over a period of days (inter-session) and (3) investigate time trends of residual motion within the same day (intra-session). Twenty-four lung cancer patients were enrolled in an Institutional Review Board (IRB)-approved protocol. For approximately five sessions, 331 four-minute, respiratory motion traces were acquired with free breathing, audio instructions and audio-visual biofeedback for each patient. The residual motion was quantified by the standard deviation of the displacement within the gating window. The generalized linear model was used to obtain coefficients for each variable within the model and to evaluate the clinical and statistical significance. The statistical significance was determined by a p-value<0.05, while effect sizes of 0.1 cm (one standard deviation) were considered clinically significant. This data analysis was applied to patient, tumour and treatment variables. Inter- and intra-session variations were also investigated. The only variable that was significant for both inhale- and exhale-based gating was disease type. In addition, visual-training displacement, breathing type and Karnofsky performance status (KPS) values were significant for inhale-based gating, and dose-per-fraction was significant for exhale-based gating. Temporal respiratory variations within and between sessions were observed for individual patients. However inter- and intra-session analyses did not show significant time trends on average for any of the variables considered. The lack of significant time trends within and between sessions indicates that on average (1) there is no significant learning period for breathing training, (2) the patients did not experience training-related fatigue and (3) the margin component to account for residual motion during gated radiotherapy appears to remain constant throughout the treatment.
呼吸门控可减少成像和治疗过程中明显的呼吸运动；然而，门控窗口内仍存在残余运动。呼吸训练可提高呼吸的可重复性，从而提高呼吸门控放疗的疗效。本研究旨在确定呼吸门控期间的残余运动是否受患者、肿瘤或治疗特征的影响。本研究的具体目标是：（1）确定影响残余运动的显著特征，（2）研究数天内（不同疗程间）残余运动的时间趋势，以及（3）研究同一天内（同一疗程内）残余运动的时间趋势。24例肺癌患者纳入了一项经机构审查委员会（IRB）批准的方案。对每位患者在大约五个疗程中，在自由呼吸、音频指令和视听生物反馈情况下采集了331条四分钟的呼吸运动轨迹。残余运动通过门控窗口内位移的标准差进行量化。采用广义线性模型获得模型中每个变量的系数，并评估临床和统计学意义。统计学意义通过p值<0.05确定，而效应大小为0.1 cm（一个标准差）被认为具有临床意义。该数据分析应用于患者、肿瘤和治疗变量。还研究了不同疗程间和同一疗程内的变化。对于基于吸气和呼气的门控，唯一显著的变量是疾病类型。此外，视觉训练位移、呼吸类型和卡诺夫斯基功能状态（KPS）值对于基于吸气的门控具有显著性，而每次分割剂量对于基于呼气的门控具有显著性。观察到个体患者在不同疗程间和同一疗程内的呼吸随时间变化。然而，不同疗程间和同一疗程内的分析并未显示所考虑的任何变量平均有显著的时间趋势。不同疗程间和同一疗程内缺乏显著的时间趋势表明，平均而言：（1）呼吸训练没有显著的学习期，（2）患者没有经历与训练相关的疲劳，（3）在门控放疗期间用于考虑残余运动的边界部分在整个治疗过程中似乎保持不变。
