林苑，查云飛，2*，邢棟，胡磊，王嬌，曾菲菲，陸雪松

釓劑對(duì)MR IDEAL-IQ骨髓脂肪定量影響的實(shí)驗(yàn)研究

林苑1，查云飛1，2*，邢棟1，胡磊1，王嬌1，曾菲菲1，陸雪松2，3

骨髓；脂肪類；磁共振成像；對(duì)比劑

骨髓脂肪組織作為骨質(zhì)疏松、糖尿病等代謝性疾病、神經(jīng)厭食癥、血液系統(tǒng)疾病及良惡性骨腫瘤等的重要生物學(xué)標(biāo)記物，與骨髓的各種生理性、病理性改變都密切相關(guān)[1-3]，迭代最小二乘法六回波梯度回波非對(duì)稱采集水脂分離磁共振成像(iterative decomposition of water and fat with echoasymmetry and least square estimationironquantification，IDEAL-IQ)是在IDEAL技術(shù)上改良的MR三維脂肪定量技術(shù)，其在肝臟、骨髓組織脂肪定量的準(zhǔn)確性已得到充分肯定[4-6]。

常規(guī)MRI增強(qiáng)和動(dòng)態(tài)對(duì)比磁共振成像(DCEMRI)對(duì)鑒別骨髓良惡性病變、評(píng)估惡性腫瘤預(yù)后、監(jiān)測(cè)腫瘤對(duì)放化療和抗血管基因治療的反應(yīng)以及評(píng)價(jià)骨髓病變血流灌注具有重要意義[7-9]。IDEAL-IQ序列定量評(píng)估骨髓脂肪分?jǐn)?shù)(FF)是否受到釓對(duì)比劑增強(qiáng)掃描的影響尚未見專題研究報(bào)道，本研究旨在探討靜脈注射釓對(duì)比劑對(duì)IDEALIQ定量兔腰椎體骨髓脂肪含量的影響，以評(píng)估增強(qiáng)掃描后應(yīng)用IDEAL-IQ技術(shù)定量骨髓脂肪含量的可行性。

1 材料與方法

1.1 研究對(duì)象

于武漢大學(xué)動(dòng)物實(shí)驗(yàn)中心購進(jìn)新西蘭大白兔8只，雌雄不限，3～5月齡。所有實(shí)驗(yàn)兔在武漢大學(xué)實(shí)驗(yàn)動(dòng)物中心單籠飼養(yǎng)，予以標(biāo)準(zhǔn)兔飼料及衛(wèi)生水喂養(yǎng)，室溫控制在25℃。本實(shí)驗(yàn)遵循武漢大學(xué)關(guān)于保護(hù)和使用實(shí)驗(yàn)動(dòng)物的相關(guān)規(guī)定。

1.2 MRI成像方法

將實(shí)驗(yàn)兔固定后，用留置針(20 G)穿刺耳緣靜脈，注射3%戊巴比妥鈉溶液1.5 ml/kg進(jìn)行麻醉。取仰臥位、足先進(jìn)置于8通道膝關(guān)節(jié)專用相控陣線圈，運(yùn)用3.0 T超導(dǎo)MR機(jī)(Discovery MR 750 Plus，GE Healthcare)行三平面定位掃描后，對(duì)每只兔常規(guī)行腰椎矢狀面FSE-T1WI、FSE-T2WI和IDEALIQ掃描，掃描范圍包括第3～7腰椎。隨后采用MEDRAD Spectris Solaris EP磁共振壓力注射器，經(jīng)兔耳緣靜脈留置針注射釓貝葡胺(Gd-BOPTA，商品名：莫迪司)3 ml，用6 ml生理鹽水以相同流率(0.5 ml/s)進(jìn)行沖洗。注射后5 min、30 min分別行腰椎矢狀面IDEAL-IQ序列掃描，注射釓對(duì)比劑前后IDEAL-IQ掃描參數(shù)保持一致，掃描參數(shù)如下。

矢狀面T1WI掃描參數(shù)：TR 400 ms，TE 13 ms，掃描層厚3 mm，視野16 cm×16 cm，矩陣512×512，激勵(lì)次數(shù)為1，掃描時(shí)間2 min 20 s。

矢狀面T2WI掃描參數(shù)： TR 2500 ms，TE 102.9 ms，掃描層厚3 mm，視野16 cm×16 cm，矩陣512×284，激勵(lì)次數(shù)為1，掃描時(shí)間3 min 25 s。

矢狀面IDEAL-IQ掃描參數(shù)：反轉(zhuǎn)角6°，TR 19.6 ms，TE1 1.2 ms，ΔTE 2 ms，帶寬125 kHz，掃描層厚3 mm，視野16 cm×12.8 cm，矩陣288×288，激勵(lì)次數(shù)為2，掃描時(shí)間4 min 22 s。

1.3 組織病理學(xué)檢查

于MR掃描后當(dāng)日采用空氣栓塞法將實(shí)驗(yàn)兔處死，取出第3～7腰椎，4%多聚甲醛固定24 h后，用EDTA脫鈣液進(jìn)行脫鈣3周，隨后脫水、石蠟包埋、切片，沿每個(gè)椎體短軸切4 μm厚薄片兩張，分別行HE染色和普魯士藍(lán)鐵染色。

1.4 數(shù)據(jù)分析

1.4.1 IDEAL-IQ骨髓脂肪分?jǐn)?shù)(FF)及R2*測(cè)定

1.4.2 HE染色骨髓脂肪含量

在Image-Pro Plus 6.0圖像分析系統(tǒng)對(duì)骨髓脂肪含量進(jìn)行定量分析，按照下列公式計(jì)算脂肪含量：FCHIS=選中空泡區(qū)域的面積/整體面積。每張HE病理切片隨機(jī)選擇5個(gè)200倍光鏡視野計(jì)算平均值。

1.4.3 普魯士藍(lán)染色椎體骨髓鐵沉積

在顯微鏡(OLYMPUS BX51)下觀察腰椎普魯士藍(lán)染色切片，確定腰椎骨髓內(nèi)有無鐵沉積。

1.5 統(tǒng)計(jì)學(xué)分析

2 結(jié)果

注射釓對(duì)比劑前IDEAL-IQ骨髓脂肪分?jǐn)?shù)值(FF0min)與注射釓對(duì)比劑5 min后骨髓脂肪分?jǐn)?shù)值(FF5min)、30 min后骨髓脂肪分?jǐn)?shù)值(FF30min)均服從正態(tài)分布，且組間方差齊，應(yīng)用單變量重復(fù)測(cè)量的方差分析，各時(shí)間點(diǎn)骨髓脂肪分?jǐn)?shù)(FF)無明顯統(tǒng)計(jì)學(xué)差異(F=3.118，P＞0.05，組間比較結(jié)果分別為P=0.835、0.916和0.754)

Pearson相關(guān)分析結(jié)果顯示HE染色計(jì)算的脂肪含量(FCHIS)與注射釓劑前FF0min(r=0.813)存在高度正相關(guān)性(P＜0.05)(圖3)。

Bland-Altman分析顯示注射釓劑前FF0min與注射釓劑5 min后FF5min(0.90、1.10)及注射釓劑30 min后FF30min(0.94、1.07)均具有良好的一致性(測(cè)量值比率95%置信區(qū)間)，提示95%置信區(qū)間內(nèi)骨髓脂肪分?jǐn)?shù)測(cè)量值最大差值分別為2.89%、2.42%(圖4，5)。

圖1兔腰椎矢狀面IDEAL-IQ圖像。A、B分別為注射釓劑前FF圖、R2*圖；C、D分別為注射釓劑5 min后FF圖、R2*圖；E、F分別為注射釓劑30 min后FF圖、R2*圖圖2A：兔腰椎骨髓HE染色(HE ×200)。紅色箭頭所示為脂肪空泡，黑色箭頭所示為骨髓細(xì)胞；B：兔腰椎骨髓普魯士藍(lán)染色( ×400)光鏡下細(xì)胞排列整齊，形態(tài)規(guī)則，未見明顯鐵顆粒Fig. 1The fat fraction (FF) and R2*mappings of the spleen. A, B: The spine fat fraction (FF) and R2*mappings before administration of Gd-BOPTA. C, D: The spine fat fraction (FF) and R2*mappings 5 min after administration of Gd-BOPTA. E, F: The spine fat fraction (FF) and R2*mappings 30 min after administration of Gd-BOPTA.Fig. 2A: Bone marrow HE staining (HE ×200), the fat cell (red arrow) and the bone marrow cells (black arrow). B: Bone marrow (BM) iron deposits were assessed by Prussian blue staining( ×400).

表1注射釓劑前后兔腰椎骨髓FF值和R2*值時(shí)序性變化Tab. 1Fat fractionfor the spine before and after gadolinium-based contrast media injection

表1注射釓劑前后兔腰椎骨髓FF值和R2*值時(shí)序性變化Tab. 1Fat fractionfor the spine before and after gadolinium-based contrast media injection

Notes: 40 vertebrae, FF measurements of revealed no signif i cant systematic bias between the three measurements(P＞0.05 for all), R2*increased (P=0.046, P=0.024) after administration of gadolinium.

±s (s-1) Pre-contrast After-contrast 5 min After-contrast 30 min F value P valuePre-contrastAfter-contrast 5 min After-contrast 30 min F value P value 40 31.77±2.66 31.89±2.54 31.71±2.520.0510.950121.64±20.28125.70±19.15125.21±19.18 3.1180.048 FF x n ±s (%) R2*x

圖3注射釓劑前IDEAL-IQ定量的骨髓脂肪分?jǐn)?shù)(FF0min)與HE染色計(jì)算的骨髓脂肪含量(FCHIS)的相關(guān)性圖4IDEAL-IQ定量的注射釓劑前及注射釓劑5 min后兔腰椎FF值Bland-Altman分析結(jié)果圖5IDEAL-IQ定量的注射釓劑前及注射釓劑30 min后兔腰椎FF值Bland-Altman分析結(jié)果Fig. 3Correlation between the IDEAL-IQ fat fraction (FF0min) and the historic fat content(FCHIS).Fig. 4Pre- and post-contrast (5 min) Bland-Altman plots.Fig. 5Pre- and post-contrast (30 min) Bland-Altman plots.

3 討論

本實(shí)驗(yàn)中根據(jù)釓對(duì)比劑在體內(nèi)的代謝時(shí)間曲線特點(diǎn)[10]，設(shè)定在注射釓劑前及注射后5 min、30 min 三個(gè)時(shí)間點(diǎn)分別采用IDEAL-IQ序列對(duì)兔腰椎進(jìn)行掃描，將相應(yīng)時(shí)間點(diǎn)椎體骨髓脂肪分?jǐn)?shù)進(jìn)行分析且將增強(qiáng)前脂肪分?jǐn)?shù)與組織病理學(xué)脂肪含量進(jìn)行對(duì)比，結(jié)果顯示注射釓劑前后各時(shí)間段IDEAL-IQ測(cè)得的椎體骨髓脂肪分?jǐn)?shù)無明顯差異，且增強(qiáng)前脂肪分?jǐn)?shù)與組織病理學(xué)脂肪含量具有高度相關(guān)性。Bland-Altman一致性分析顯示注射釓劑前IDEAL-IQ測(cè)得的脂肪分?jǐn)?shù)與注射后各時(shí)間段測(cè)得的脂肪分?jǐn)?shù)一致性良好。這證明了IDEAL-IQ序列定量骨髓脂肪含量準(zhǔn)確度高，穩(wěn)定性及可重復(fù)性好，且不受釓對(duì)比劑增強(qiáng)掃描的影響，即在靜脈注射釓對(duì)比劑后采用IDEAL-IQ序列定量骨髓脂肪含量具有可行性。

DIXON方法可利用水脂分離技術(shù)測(cè)得組織脂肪含量，傳統(tǒng)的兩點(diǎn)式DIXON[11]技術(shù)通過調(diào)節(jié)回波時(shí)間TE做兩次采集，得到同相位及反相位圖像，對(duì)兩幅圖像進(jìn)行加減，可得到水像和脂像，進(jìn)一步算出脂肪分?jǐn)?shù)，然而傳統(tǒng)的兩點(diǎn)式DIXON方法受T2*效應(yīng)影響，在進(jìn)行水脂分離時(shí)產(chǎn)生相位誤差，導(dǎo)致測(cè)量結(jié)果誤差[12]。IDEAL-IQ采用小角度激發(fā)降低T1偏倚，采集多個(gè)(≥6)梯度回波擬合T2*衰減曲線，將生成的T2*值用來校正源數(shù)據(jù)，從而修正T2*效應(yīng)的影響[13]。并用fly-back方法進(jìn)行k空間填充，生成水像、脂像以及脂肪比像等六幅圖像，在脂肪比圖像上及弛豫率圖像放置ROI可以直接得到脂肪分?jǐn)?shù)(FF)及R2*值而無需進(jìn)一步計(jì)算。本實(shí)驗(yàn)將骨髓脂肪分?jǐn)?shù)(FF)與病理學(xué)定量的骨髓脂肪細(xì)胞面積比(FCHIS)進(jìn)行相關(guān)性分析，進(jìn)一步驗(yàn)證IDEAL-IQ技術(shù)定量骨髓脂肪含量準(zhǔn)確度高。

MR釓類對(duì)比劑具有一定程度的T1弛豫時(shí)間和T2、T2*弛豫時(shí)間縮短效應(yīng)。橫向弛豫率R2*是T2*的倒數(shù)，釓劑能縮短T2*弛豫時(shí)間，相對(duì)應(yīng)增加R2*值，本實(shí)驗(yàn)注射釓劑后骨髓R2*值升高，亦予以證實(shí)。Ge等的研究[14]顯示IDEAL-IQ技術(shù)定量的肝臟脂肪分?jǐn)?shù)不受靜脈注射釓對(duì)比劑帶來的R2*值變化的影響而保持較高的穩(wěn)定性，但是無法排除鐵沉積對(duì)R2*定量參數(shù)測(cè)量的影響。Liau等[15]通過實(shí)驗(yàn)證實(shí)IDEAL-IQ定量的肝臟脂肪分?jǐn)?shù)不受鐵劑的影響，但是Hines等[16]通過對(duì)不同濃度的水-脂-鐵試劑進(jìn)行定量分析，發(fā)現(xiàn)隨著鐵劑濃度的增高，IDEAL-IQ定量的脂肪分?jǐn)?shù)與實(shí)際脂肪含量差異增大，其原因可能是鐵劑對(duì)水組織的T2*弛豫的影響大于對(duì)脂肪組織的影響。本實(shí)驗(yàn)中，對(duì)兔腰椎椎體切片后進(jìn)行普魯士藍(lán)染色，在400倍顯微鏡下觀察骨髓內(nèi)鐵沉積情況，見細(xì)胞排列整齊，形態(tài)規(guī)則，未見明顯鐵顆粒，參照 Sehgal等[17]鐵沉積半定量分析評(píng)價(jià)標(biāo)準(zhǔn)可歸為0級(jí)即無鐵沉積。因此，可基本排除椎體骨髓內(nèi)鐵沉積對(duì)R2*值的影響，在本實(shí)驗(yàn)中R2*值的變化主要代表釓劑對(duì)骨髓組織T2*弛豫的影響。

本實(shí)驗(yàn)結(jié)果顯示由于釓劑可通過加快骨髓的T2*弛豫率而導(dǎo)致R2*值的升高，然而進(jìn)行骨髓脂肪定量時(shí)，通過采集多個(gè)回波擬合T2*衰減曲線，能修正T2*效應(yīng)的影響，因此在注射釓劑后，IDEAL-IQ序列定量的脂肪分?jǐn)?shù)值(FF)依然能保持一定的穩(wěn)定性，這與Ge等[14]證實(shí)的IDEAL-IQ對(duì)靜脈注射釓噴酸葡胺保持良好穩(wěn)定性的結(jié)果一致，進(jìn)一步提示磁共振釓類對(duì)比劑等影響T2*弛豫率的物質(zhì)對(duì)IDEAL-IQ定量組織脂肪含量影響較小。

本研究的局限性：(1)本研究試驗(yàn)樣本量較小，增大樣本量及變異程度的改變可能會(huì)得到不同的結(jié)果，所以仍需后續(xù)大樣本實(shí)驗(yàn)研究證實(shí)；(2)本實(shí)驗(yàn)未對(duì)IDEAL-IQ技術(shù)T2*校準(zhǔn)效應(yīng)對(duì)骨髓脂肪分?jǐn)?shù)(FF)的影響進(jìn)行評(píng)估，即未比較經(jīng)T2*校準(zhǔn)的FF值和未經(jīng)T2*校準(zhǔn)的FF值的差異。然而根據(jù)Meisamy等[18]的報(bào)道，在未注射釓劑的情況下，肝臟經(jīng)T2*校準(zhǔn)的FF值和未經(jīng)T2*校準(zhǔn)的FF值有統(tǒng)計(jì)學(xué)差異，筆者推斷IDEAL-IQ可修正骨髓T2*效應(yīng)而保持FF測(cè)量結(jié)果的穩(wěn)定性。

總之，在靜脈注射釓劑引起的R2*值增高的情況下，IDEAL-IQ定量椎體骨髓脂肪分?jǐn)?shù)的準(zhǔn)確度高，且仍維持較高的穩(wěn)定性和一致性。在靜脈注射釓對(duì)比劑后采用IDEAL-IQ序列定量骨髓脂肪含量具有可行性。

[References]

[1] Rosen CJ, Ackert-Bicknell C, Rodriguez JP, et al. Marrow fat and the bone microenvironment: developmental, functional, and pathological implications. Crit Rev Eukaryot Gene Expr, 2009, 19(2): 109-124.

[2] Krings A, Rahman S, Huang S, et al. Bone marrow fat has brown adipose tissue characteristics, which are attenuated with aging and diabetes. Bone, 2012, 50(2): 546-552.

[3] Shen W, Chen J, Gantz M, et al. MRI-measured pelvic bone marrow adipose tissue is inversely related to DXA-measured bone mineral in younger and older adults. Eur J Clin Nutr, 2012, 66(9): 983-988.

[4] Tang A, Tan J, Sun M, et al. Nonalcoholic fatty liver disease: MR imaging of liver proton density fat fraction to assess hepatic steatosis. Radiology, 2013, 267(2): 422-431.

[5] Hu L, Zha YF, Lin Y, et al. The feasibility of IDEAL-IQ quantitative evaluation of vertebral fat fraction content in rabbit models of diabetes mellitus. Chin J Magn Reson Imaging, 2015, 6(12): 941-946.胡磊, 查云飛, 林苑, 等. IDEAL-IQ定量評(píng)價(jià)兔糖尿病模型椎體骨髓脂肪含量的可行性研究. 磁共振成像, 2015, 6(12): 941-946.

[6] Idilman IS, Aniktar H, Idilman R, et al. Hepatic steatosis: quantif i cation by proton density fat fraction with MR imaging versus liver biopsy. Radiology, 2013, 267(3): 767-775.

[7] Hawighorst H, Libicher M, Knopp MV, et al. Evaluation of angiogenesis and perfusion of bone marrow lesions: role of semiquantitative and quantitative dynamic MRI. J Magn Reson Imaging, 1999, 10(3): 286-294.

[8] Bluemke DA, Petri M, Zerhouni EA. Femoral head perfusion and composition: MR imaging and spectroscopic evaluation of patients with systemic lupus erythematosus and at risk for avascular necrosis. Radiology, 1995, 197(2): 433-438.

[9] Bollow M, Knauf W, Korfel A, et al. Initial experience with dynamic MR imaging in evaluation of normal bone marrow versus malignant bone marrow inf i ltrations in humans. J Magn Reson Imaging, 1997, 7(1): 241-250.

[10] Men WW, Li N, Yu JH, et al. Accurate depicting metabolic process of macromolecule gadolinium magnetic resonance contrast agent. Chin J Inter Imag Ther, 2010, 7(3): 320-324.門衛(wèi)偉, 李娜, 余家會(huì), 等. 大分子釓磁共振對(duì)比劑代謝曲線的精確描繪. 中國介入影像與治療學(xué), 2010, 7 (3): 320-324.

[11] Dixon WT. Simple proton spectroscopic imaging. Radiology, 1984, 153(1): 189-194.

[12] Ma J. Dixon techniques for water and fat imaging. J Magn Reson Imaging, 2008, 28(3): 543-558.

[13] Yu H, Shimakawa A, Mckenzie CA, et al. Multiecho water-fat separation and simultaneous R2*estimation with multifrequency fat spectrum modeling. Magn Reson Med, 2008, 60(5): 1122-1134.

[14] Ge M, Zhang J, Wu B, et al. Effect of gadolinium on hepatic fat quantification using multi-echo reconstruction technique with T2*correction and estimation. Eur Radiol, 2016, 26(6): 1913-1920.

[15] Liau J, Shiehmorteza M, Girard O M, et al. Evaluation of MRI fat fraction in the liver and spine pre and post SPIO infusion. Magn Reson Imaging, 2013, 31(6): 1012-1016.

[16] Hines CD, Yu H, Shimakawa A, et al. T1 independent, T2*corrected MRI with accurate spectral modeling for quantification of fat: validation in a fat-water-SPIO phantom. J Magn Reson Imaging, 2009, 30(5): 1215-1222.

[17] Sehgal V, Delproposto Z, Haddar D, et al. Susceptibility-weighted imaging to visualize blood products and improve tumor contrast in the study of brain masses. J Magn Reson Imaging, 2006, 24(1): 41-51.

[18] Meisamy S, Hines CD, Hamilton G, et al. Quantif i cation of hepatic steatosis with T1-independent, T2-corrected MR imaging with spectral modeling of fat: blinded comparison with MR spectroscopy. Radiology, 2011, 258(3): 767-775.

Effect of gadolinium on vertebral fat fraction content using IDEAL-IQ technique

LIN Yuan1, ZHA Yun-fei1,2*, XING Dong1, HU Lei1, WANG Jiao1, ZENG Fei-fei1, LU Xue-song2,3

1Department of Radiology, Renmin Hospital of Wuhan University, Wuhan 430060, China

2Hubei Key Laboratory of Medical Information Analysis and Tumor Diagnosis &Treatment, Wuhan 430060, China

3Department of Biological Engineering, School of Biomedical Engineering, South-Central University for Nationalities, Wuhan 430060, China

ACKNOWLEDGMENTSThis thesis is sponsored by the Opening Foundation of Hubei Key Laboratory of Medical Information Analysis and Tumor Diagnosis & Treatment (No. PJS140011511). Scientif i c Research Funding Project of Hubei Provincial Department of Health (No. JX6B68).

Objective:This study evaluates the robustness of a magnetic resonance (MR) fat quantification method to changes in R2*caused by an intravenous infusion of gadolinium.Materials and Methods:The R2*and fat fraction content (FF) were measured in vertebral marrow in eight New Zealand white rabbits using an investigational sequence (IDEAL IQ) provided by the MR scanner vendor. Measurements were made once before and twice after Gd-BOPTA infusion (5 min, 30 min). Then HE stain was performed for calculating vertebral fat content (FCHIS), Prussian blue stain was performed for showing iron in vertebral.Results:Vertebral marrow FF measurements revealed no significant systematic bias between the three measurements (P＞0.05 for all). Good agreement (95% confidence interval) of FF measurements were demonstrated between FF0minand FF5min(0.90, 1.10) and FF0minand FF30min(0.94, 1.07). A signif i cant positive correlation is found between FF0minand FCHIS(r=0.813, P＜0.05). R2*increased after administration of gadolinium.Conclusion:Although under the impact of an increased R2*in vertebral marrow post-contrast, the investigational sequence can still obtain accurate and stable fat fraction content. the IDEAL IQ method of fat quantif i cation is robust to changes in R2*.

Bone marrow; Fats; Magnetic resonance imaging; Contrast media

Zha YF, E-mail: zhayunfei@hotmail.com

Received 26 Sep 2016, Accepted 20 Oct 2016

醫(yī)學(xué)信息分析及腫瘤診療湖北省重點(diǎn)實(shí)驗(yàn)室開放課題基金項(xiàng)目(編號(hào)：PJS140011511)；湖北省衛(wèi)生廳科研資助項(xiàng)目(編號(hào)：JX6B68)

1. 武漢大學(xué)人民醫(yī)院放射科，武漢430060

2. 醫(yī)學(xué)信息分析及腫瘤診療湖北省重點(diǎn)實(shí)驗(yàn)室，武漢 430060

3. 中南民族大學(xué)生物醫(yī)學(xué)工程學(xué)院，武漢 430060

查云飛，E-mail：zhayunfei@hotmail. com

2016-09-26

接受日期：2016-10-20

R445.2

A

10.12015/issn.1674-8034.2016.11.012

林苑, 查云飛, 邢棟. 等. 釓劑對(duì)MR IDEAL-IQ骨髓脂肪定量影響的實(shí)驗(yàn)研究. 磁共振成像, 2016, 7(11): 856-860.*