摘要目的　对比食管癌放射治疗中CT模拟和常规模拟定位的优劣。方法　对10例食管胸段癌病例，同时行CT模拟定位和常规模拟定位。全部设机架角为0°、120°、240°的等中心3个野，照射野长度为所见病灶上下各扩大3?cm。常规模拟野宽度分别为前野6?cm，后斜野5?cm，CT模拟野宽为大体肿瘤区（GTV）最大径外扩大10?mm。对以上二种定位方法所描述靶区大小、等中心位置、照射野大小的差异和剂量分布情况进行对比分析。结果　二种定位方法等中心点位置在X、Y、Z 3个轴上分别相差（4.2±2.6）、（3.4±2.7）、（15.4±10.6）?mm。二者的照射野大小差异有显著性意义［宽：（6.0±1.1）?cm 对（5.3±0.5）?cm，t=3.54，P<0.01；长：（12.8±3.0）?cm 对（16.4±4.1）?cm，t=6.17，P<0.01］。GTV完全被90%等剂量线所包括的计划中CT定位有10例、常规定位只有1例；脊髓最大受照剂量低于计划75%的，CT定位只有2例而常规定位有7例。结论　CT模拟定位较常规模拟定位更充分显示肿瘤外侵范围并反映其非对称生长，但在确定病灶长度时不如后者；用CT定位时常规食管吞钡X射线片仍有重要的参考价值。

abstractsObjective　To compare CT simulation (CTsim) with conventional simulation (Consim) in the treatment planning setup of esophageal carcinoma. Methods　 Ten patients with thoracic esophageal carcinoma were analyzed. All patients underwent both CTsim and Consim in the same treatment position. All treatment planning was configured with three symmetical fields in the isocenter mode. The gantry angles were constantly at 0, 120 and 240 degree for anterior-posterior (AP), right posterior oblique (RPO) and left posterior oblique (LPO) field, respectively. The radiation fields covered 3 cm beyond the tumor margin longitudinally in both CTsim and Consim plan. The fields in CTsim plan extended 1?cm beyond the gross tumor volume (GTV). Fields in Consim plan were 6 cm wide in auterior field and 5 cm wide in posterior oblique field. The field size, position of isocenter, difference in field size and dose distribution were compared between these two simulation techniques. Results　The differences in isocenter position in X, Y and Z orientation were observed between CTsim and Consim, with (4.2±2.6) mm，(3.4±2.7) mm，and (15.4±10.6) mm, respectively. The mean field size used in CTsim was significantly larger than that in Consim, with (6.0±1.1) cm versus (5.3±0.5) cm in width, and (12.8±3.0) cm versus (16.4±4.1) cm in length. GTV was totally covered by the 90% iso-dose curve in 10 of 10 patients as compared to only 1 of 10 patients in the Consim plans. The maximal dose at the spinal cord was under 75% normalization dose in 2 of 10 patients with CTsim plans and in 7 of 10 patients with Consim plans. Conclusions　CTsim has an advantage in the delineation of lateral extension of esophageal tumor than the Consim. However, CTsim is dwarfed in defining lesion length as compared to Consim. The barium esophagram is helpful for CT simulation.