以鎝－99m標誌紅血球單光子射出斷層掃描三向立體處理診斷肝臟血管瘤的價值

Value of Tc-99M Labeled Red Blood Cell Three-Dimensional SPECT Images for the Detection of Hepatic Hemangiomas

肝臟血管瘤是肝臟最常見的良性腫瘤。報告指出鎝－99m標誌紅血球掃描對於診斷肝臟血管瘤有極佳的專一性。這個研究的目的是評估以鎝－99m標誌紅血球單光子射出斷層掃描三向立體處理診斷肝臟血管瘤的價值。方法：這個研究包括27位診斷為肝臟血管瘤的病患，一共有57個直徑介於0.5到10公分的病灶。所有的病患在注射鎝－99m標誌紅血球後2小時進行平面掃描及斷層掃描。斷層掃描後進行切面展示處理及三向立體處理。結果：平面掃描、切面展示處理及三向立體處理的診斷敏感度分別為26/57（45.6%）、46/57（80.7%）及46/57（80.7）。對於直徑小於2公分的病灶，平面掃描、切面展示處理及三向立體處理的診斷敏感度分別為0/10（0%），1/10（10.0%）and 1/10（10.0%）。對於直徑介於2到4公分的病灶，平面掃描、切面展示處理及三向立體處理的診斷敏感度分別為21/42（50.0%），40/42（95.2%）and 40/42 （95.2%）。對於直徑大於4公分的病灶，平面掃描、切面展示處理及三向立體處理的診斷敏感度皆為5/5（100%）。結論：單光子射出斷層掃描切面展示處理及三向立體處理對於直徑2公分以上病灶的診斷敏感度極佳，但對於2公分以下病灶的診斷敏感度都不好。然而三向立體處理對於辨識血管附近的病灶比切面展示處理方便及清楚，而且三向立體處理不須像切面展示處理所須的大量影像展示比較。

Aim: Hepatic hemangiomas (HH) are the most common benign tumors of the liver. Delayed Tc-99m labeled red blood cell (RBC) scanning is considered as a highly specific technique for detecting HH. The purpose of this study was to demonstrate the value of volume rendering of delayed Tc-99m labeled RBC three-dimensional (3D) single photo emission computed tomography (SPECT) data, for the diagnosis of HH. Methods: The study included 27 patients with a diagnosis of HH, with a total of 57 lesions ranging from 0.5 to 10 cm in diameter. Delayed planar imaging (PI) and SPECT were performed for all patients 2 hours after intravenous injection of the radiotracer. Slice display and volume rendering of 3D of SPECT (SPECT-S and SPECT-3D) data were also carried out. Result: The sensitivities of PI, SPECT-S and SPECT-3D images were 26/57 (45.6%), 46/57 (80.7%) and 46/57 (80.7%), respectively. For lesions of less than 2 cm in diameter, the sensitivities of PI, SPECT-S and SPECT-3D images were 0/10 (0.0%), 1/10 (10.0%) and 1/10 (l0.0%), respectively. For lesions of 2-4 cm in diameter, the sensitivities of PI, SPECT-S and SPECT-3D images were 21/42 (50.0%), 40/42 (95.2%) and 40/42 (95.2%), respectively. For lesions of more than 4 cm in diameter, the sensitivities of PI, SPECT-S and SPECT-3D images were all 5/5 (100%). Conclusion: SPECT-3D is also not sensitive enough for small lesions. However, in comparison with SPECT-S, SPECT-3D images may be clearer for differentiating HH from vessels as suspected lesions near vessels, in addition, 3D rendering processes can decrease the number of images that need to be viewed, while preserving all of the image features necessary for accurate interpretation.