以磁振造影技術評估椎體壓迫性骨折

Evaluation of Vertebral Compression Fracture using Magnetic Resonance Imaging

謝佳宏(Chia-Hong Hsieh)

指導教授 ： 李境和

共同指導教授 ： 陳博洲(Po-Chou Chen)

義守大學/電機資訊學院/資訊工程學系/碩士(2015年)

https://doi.org/10.6343/ISU.2015.00230

摘要

臨床上常使用磁振造影(Magnetic Resonance Imaging, MRI) 作為椎體壓迫性骨折 (Vertebral compression fracture) 的第一線診斷工具。急性椎體壓迫性骨折在脂肪抑制T1與T2加權影像中，椎體訊號因水腫與出血現象而產生訊號差異。本研究的目的為使用三種不同脂肪抑制技術之T1與T2加權影像，進行統計量化以區分比較急性與慢性椎體壓迫性骨折。其中脂肪抑制技術包括頻率選擇脂肪抑制 (Spectral fat saturation, Fat Sat)、短反轉時間反轉回復 (Short TI Inversion Recovery, STIR) 及頻率選擇絕熱反轉回復 (Spectrally Adiabatic Inversion Recovery, SPAIR) 等三種。研究收集34例急性椎體壓迫性骨折或慢性椎體壓迫性骨折病例之MRI影像；掃描波序則包括T1 快速自旋回訊(Turbo Spin Echo, TSE)、T1-STIR TSE、T1-Fat Sat TSE、T1-SPAIR TSE、T2 TSE、T2-STIR TSE、T2-Fat Sat TSE及T2- SPAIR TSE等八種。取得影像後使用INFINITT G3 PACS工作站圈選正常與骨折椎體感興趣區域(Region of Interest, ROI)，比較分析不同波序影像所量測的訊號強度比(Signal Intensity Ratio, SIR)與對比雜訊比(Contrast-to-Noise Ratio, CNR)。最後，針對有無脂肪抑制量測的訊號強度差異進行統計分析。SIR量測結果發現T1 TSE及T2 TSE，對區分急性與慢性椎體壓迫性骨折之準確率分別為85%及59%；而CNR量測結果之準確率則分別為85%及47%。脂肪抑制技術中以T2-STIR TSE(SIR)之診斷準確率最高為100%，與醫師診斷報告完全吻合，研究結果可提供醫師診斷急性椎體壓迫性骨折交叉參考的診斷資訊。

關鍵字

椎體壓迫性骨折 ； 水腫 ； 頻率選擇脂肪抑制 ； 短反轉時間反轉回復 ； 頻率選擇絕熱反轉回復

並列摘要

MRI is often used as the primary tool for diagnosing vertebral compression fracture in clinical usage. The signal difference on fat suppressed T1 and T2-weighted images of acute vertebral compression fracture comes from vertebral body edema and bleeding. This study was aimed to statistically quantify to distinguish and compare between acute and chronic vertebral compression fractures using three different fat-suppressed T1 and T2-weighted images techniques. Fat suppression techniques used were Spectral fat saturation (Fat Sat), Short TI Inversion Recovery (STIR), and Spectrally Adiabatic Inversion Recovery (SPAIR). Thirty-four MR images with acute or chronic vertebral compression fracture cases were collected. Eight scanning pulse sequences including T1 turbo spin echo (TSE), T1-STIR TSE, T1-Fat Sat TSE, T1-SPAIR TSE, T2 TSE, T2-STIR TSE, T2-Fat Sat TSE and T2- SPAIR TSE were used. Then, INFINITT G3 PACS workstation was used to obtain normal and fractures vertebral region of interest (ROI) values, followed by comparative analysis of signal intensity ratio (SIR) and contrast to noise ratio (CNR) on MR images obtained from various pulse sequences. Finally, statistical analysis was performed according to the signal intensity difference between fat suppression and non- fat suppression measurements. SIR results showed that the accuracy for differentiating acute and chronic vertebral compression fractures on T1 and T2 TSE images was 85% and 59%, individually, while CNR results showed that the accuracy was 85% and 47%, individually. The accuracy of SIR (T2-STIR TSE) was 100% which was the best among all fat suppression techniques and totally consistent with the physician diagnostic reports. This study might provide physician a cross diagnostic reference for diagnosis of acute vertebral compression fracture.

並列關鍵字

Vertebral Compression Fracture ； Edema ； Frequency Selective Fat Suppression ； Short TI Inversion Recovery ； Spectrally Adiabatic Inversion Recovery

參考文獻

(8)李欣岱、蘇千田，林茂榮。 (2008) 骨質疏鬆症性骨折之預防篩檢. 台灣老年醫學暨老年學雜，3(2), 64-77。

(15)WW, O., JD, L., JA, S., MF, H. (1995). Functional Brain Mapping： Mosby-Year Book, Inc. 198-201.

(2)Baker, L. L., Goodman, S. B., Perkash, I., Lane, B., Enzmann, D. R. (1990). Benign versus pathologic compression fractures of vertebral bodies： assessment with conventional spin-echo, chemical-shift, and STIR MR imaging. Radiology, 174(2), 495-502.

(3)Tender, G. C., Serban, D. (2012). Traumatic spinal fracture treated by vertebroplasty： a case report. J Med Case Rep, 6, 390.

(4)Mussurakis, S., Buckley, D., Drew, P. (1993). Consensus development conference： diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med, 94(6), 646-650.