目的：探討頭頸部癌合併頸部淋巴結腫大的患者在強度調控放射治療(intensity-modulated radio-therapy，IMRT)計劃使用二階段與單一階段電腦斷層影像(CT)的差別。材料與方法：從2003年4月至2004年6月，共收集6位罹患頭頸部癌合併頸部淋巴結腫大的患者。我們對這6位病患安排第一次電腦斷層與核磁共振定位並進行影像融合。接著再由醫師定義出腫瘤(GTV，CTV)與重要器官。最後再由物理師應用電腦治療計劃系統以反算法設計出第一次IMRT計劃來進行治療。我們採用的治療方式是同步化學放射治療(CCRT)：化學治療部份是以Cisplatin 30mg/wk靜脈輸注：放射治療部份採用SIB(simultaneous intensity boost)IMRT。當患者接受到4500 cGy時（計劃A）我們立即安排第二次電腦斷層與核磁共振定位，並根據第二次獲得的影像進行第二階段治療計畫的安排（計劃B）。此外我們將計劃A的調強圖譜(fluence map)套用到第二次的CT影像上形成模擬的計劃C，最後分析計劃B和C在劑量分布之差異。結果：比較二次CT影像，我們發現病患的頸部淋巴結因接受4500 cGy而縮小甚至消失。若是未進行第二階段的IMRT計劃則會使患者的皮膚與腮腺落入高劑量區內。接著，在劑量分布上，我們發現，對CTV而言，其劑量包覆順形度上(conformity index)都是計劃B比計劃C好。對於腦幹而言，用t檢定來分析計劃B與計劃C發現沒有明顯差別(p=0.9943)，顯示計劃B與計劃C對於腦幹的保護相當。對於腮腺的淺葉與深葉而言，用t檢定來分析計劃B與C可以發現計劃B比較低(p<0.001)，顯示使用二階段電腦斷層定位IMRT治療計劃可以明顯地再降低腮腺所接受的劑量。結論：本研究顯示在執行IMRT治療時若患者體廓改變則必須調整治療計劃。若IMRT計劃未隨之調整，則其腮腺及皮膚劑量可能增加，腫瘤劑量曲線的順形度可能下降，影響治療的成效。

Purpose: To discuss the differences of dose distribution between the dual-phase IMRT and one-phase IMRT in the patients of head-and-neck (H&N) cancers with enlarged cervical lymph nodes. Materials and Methods: From April 2003 to June 2004, we collected 6 patients of H&N cancers with enlarged cervical lymph nodes. Initially, we arranged first simulation by CT-MRI fusion for one-phase IMRT awl prescribed 4500 cGy on GTV including enlarged lymph nodes (Plan A). Five weeks later, we arranged the second simulation by the same method to design dual-phase IMRT and prescribed 2700 cGy on GTV (Plan B). All 6 patients received concurrent chemoradiotherapy (CCRT) with cisplatin 30 mg/wk intravenously. The IMRT technique was simultaneous intensity boost (SIB). Furthermore, we combined two CT images and superimposed the fluence map of Plan A onto the second CT images to get the Plan C. Finally, we compared Plan B with Plan C. Result: When we compared two CT images, we found that those enlarged lymph nodes already regressed or disappeared after plan A treatment, which made the skin and parotid gland fall into the high dose area. From the comparison of both plans, B had a much better conformal index than C for CTV. Using t test to analyze the D(subscript 50%), D(supbscript 30%) and D(subscript mean) of Plan B and C for the sparing of parotid glands, B had better results than C (p<0.001). Therefore, the dual-phase IMRT treatment planning can provide another way to significantly decrease the irradiation doses to parotid glands. Conclusion This study indicate that if the patient's contour changes without modifying the prior IMRT treatment planning, the doses of parotid glands and the skin will increase and the conformity of dose distribution will decrease, which affect the result of IMRT.