【作者】 李春燕；

【导师】 黄仲奎；

【作者基本信息】 广西医科大学 ， 人体解剖与组织胚胎学， 2009， 博士

【摘要】 目的利用MRI肝脏/肌肉信号强度比值及肝脏T2值测定技术,探讨体外铁浓度及肝脏铁沉积的MRI定量诊断的准确性及临床应用价值。材料与方法1、含铁模型的MRI体外测试:选1.5升塑料瓶,内装1.45升蒸馏水作为含铁水模与动物肌肉标本放置一起,周边包裹以含蒸馏水约2.5L的密封水袋,制成含铁模型。对含铁模型每进行一次MRI扫描后,向塑料含铁水模内注入1ml含铁量为150mg/ml的右旋糖酐铁并充分摇匀,动物肌肉标本及密封水袋保持不变。高浓度含铁模型的制作:取1.5L塑料瓶,装1.3L蒸馏水后,第一次扫描前注入右旋糖酐铁20ml,之后每次扫描前注射右旋糖酐铁10ml,共注射200ml。采用GRE扫描序列测定含铁水模/肌肉的信号强度比值,采用8回波扫描及4回波扫描分别测定含铁水模T2值。2、肝脏铁超负荷动物模型MRI定量研究:成年雄性新西兰白兔20只作为实验组,体重2.0kg～2.9kg。同种白兔1只作为正常对照组。所有兔子称体重,用于计算每次注射右旋糖酐铁的剂量。实验组兔按15mg/kg剂量深部肌肉注射含铁量为150mg/ml的右旋糖酐铁,每周注射一次,左右两侧后腿轮流注射,共注射15周。注射右旋糖酐铁前,对所有兔子进行MRI检查,定量分析正常兔的肝脏/肌肉信号强度比及肝脏T2值。每次注射右旋糖酐铁一周后复查MRI检查并于第二天随机处死两只兔子取肝脏,测肝体积后取部分肝脏及脾脏等脏器进行铁沉积病理诊断及分期,剩余肝组织烘干测LIC。对照组兔第一次进行MRI检查后即处死,取肝组织进行病理检查及LIC测定。所有兔子处死前再次称体重,经耳缘静脉抽取2ml静脉血送检验科查血清铁。MRI检查序列:GRE扫描测肝脏/肌肉信号强度比,8回波及4回波扫描测肝脏T2值。3、肝铁超负荷MRI定量研究的临床应用:地中海贫血共149例,按基因检查分轻型地贫组23例,男10例,女13例,平均年龄28.3±12.9岁;中间型地贫组10例,男7例,女3例,平均年龄30.3±10.8岁;重型地贫组116例,男66例,女50例,平均年龄6.8±3.7岁。正常对照组20例,男10例,女10例,平均年龄27.1±13.8岁。重型地贫组按不同的治疗方法分为三组:规范输血并用ICL670去铁治疗23例(ICL670组),男13例,女10例,平均年龄8.7±3.4岁,治疗时间2年;规范输血并用去铁胺去铁治疗34例(去铁胺组),男20例,女14例,平均年龄8.2±3.7岁,治疗时间1～4年。仅输血治疗未经去铁治疗59例(未去铁组),男33例,女26例,平均年龄5.0±2.9岁。23例重型地贫应用ICL670治疗6个月后复查,29例重型地贫应用去铁胺治疗6个月后复查。MRI检查序列同动物实验部分。扫描设备:3.0T PHILIPS磁共振。扫描参数:(1)信号强度比值采用GRE序列横断面:TR 48ms,TE2.5ms,翻转角度60°,层厚10mm,层间距3mm,扫描层数4层,一次闭气完成扫描。(2)T2值测量:适用于轻度铁沉积的8回波横断位扫描:TR 2000ms,TE分别为8、16、24、32、40、48、56、64ms,层厚7mm,扫描1层,扫描时间约8分钟;适用于重度铁沉积4回波横断位扫描:TR 2000ms,TE分别为6、12、18、24ms,层厚7mm,扫描1层,扫描时间约7分钟。结果1、含铁模型的MRI体外测试:两位放射科医师对前15次扫描测得的含铁水模/肌肉信号强度比值及含铁水模R2(1/T2)值的差异均没有统计学意义。GRE扫描测得动物肌肉标本的信号强度的变异系数为6.08%;8回波和4回波扫描测得的动物肌肉标本T2值的变异系数分别为2.87%及3.90%。水袋中水的T2值始终为2047.00ms。含铁水模/肌肉信号强度比的自然对数与含铁水模铁浓度之间呈直线负相关(r=-0.999,P=0.000)。当含铁水模铁浓度小于等于1.74mg/ml时,即在T2值大于9.3ms时,8回波扫描测得的含铁水模R2值与含铁水模铁浓度显著直线正相关(r=0.996,P=0.000)。当含铁水模铁浓度小于等于2.64mg/ml时,即在T2值大于6.6ms时,4回波扫描测得的含铁水模R2值与含铁水模铁浓度显著直线正相关(r=0.991,P=0.000)。2、肝脏铁超负荷动物模型MRI定量研究:LIC与注射铁剂总量呈显著性直线正相关( r = 0.824,P=0.000)。肝脏含铁总量与注射铁剂总量呈显著性直线正相关( r = 0.943,P=0.000),建立直线回归方程?=37.376+0.505X(?:根据注射铁剂总量预测的肝脏含铁总量,X:注射铁剂总量)(F=153.993,P=0.000)。21只兔子在注射铁剂前MRI测量的肝脏/肌肉信号强度比值的平均值1.43±0.13,T2值(8回波扫描测得)的平均值为55.35±6.23。GRE序列所测定的肝脏/肌肉的信号强度比值的自然对数与LIC呈显著性直线负相关(r=-0.917,P=0.000)。建立直线回归方程为:?=5.640-5.044X (?:MRI预测的LIC,X: GRE序列所测的肝脏/肌肉信号强度比值的自然对数)(F=100.895,P=0.000)。8回波扫描测得肝脏R2值与LIC呈显著性直线正相关(r=0.789,P=0.000)。建立直线方程为:?=-2.93+199.64X(?:MRI预测的LIC,X:8回波扫描所测得的T2值的倒数)(F=31.403,P=0.000)。4回波扫描所测得肝脏R2值与LIC呈显著性直线正相关(r=0.958,P=0.000)。建立直线方程为: ?=-2.76+174.39X(?:MRI预测的LIC,X:4回波扫描所测得的T2值的倒数)(F=213.154,P=0.000)。病理诊断及分级:正常兔子属于0级,LIC在1.2～5.3mg/g之间的8只兔子属于Ⅰ级,而LIC在6.9～22.2mg/g的12只兔子属于Ⅲ级,实验中未发现Ⅱ级肝脏铁沉积。3、肝铁超负荷MRI定量研究的临床应用:(1)经随机设计t检验,20例正常成人与20只正常兔的之间肝脏/肌肉信号强度比值差异没有统计学意义(t=-1.416, P=0.165);肝脏T2值差异没有统计学意义(t=1.221, P=0.232)。(2)按照直线回归方程(公式9)可以计算临床诊断标准中LIC对应的4回波扫描肝脏T2值:T2值降到30ms,证明有LIC超过正常,综合铁螯合剂的毒性考虑是否去铁;T2值降低到17.5～30ms,证明LIC明显增高,需要去铁治疗;T2值降低到10～17.5ms,证明肝铁沉积严重;T2值降低到10ms以下,为极其严重的肝脏铁沉积,存在潜在的心肌毒性。(3)所有轻型地中海贫血患者和正常组的LIC均在正常值范围之内。10例中间型地中海贫血患者中:1例LIC大于15mg/g;6例LIC兼于7.0～15.0mg/g之间;1例患者LIC兼于3.2～7.0mg/g之间;2例患者LIC小于1.6mg/g。116例重型地中海贫血患者中:14例LIC大于15mg/g;91例LIC兼于7.0～15.0mg/g之间;2例年龄小于1岁的患者LIC小于1.6mg/g;其余9例患者LIC兼于3.2～7.0mg/g之间。(4)经随机设计t检验:轻型地贫组的LIC高于正常组;轻型地贫组的LIC低于中间型地贫组;中间型地贫组的LIC与重型地贫组相同。(5)经随机设计t检验:正常组的肝铁总量低于轻型地贫组;轻型地贫组的肝铁总量低于中间型地贫组;轻型地贫组的肝铁总量低于重型地贫组;中间型地贫组的肝铁总量高于重型地贫组。(6)肝脏含铁总量与年龄、输血总量之间建立二元回归方程?=269.84+429.65X1+3.29X2(公式11:?:肝脏含铁总量;X1:年龄;X2:输血总量)(7)规范输血的重型地贫中,未去铁治疗组、去铁胺组、ICL670组的LIC及肝铁总量经方差分析,差异均有统计学意义。两两比较:未去铁组LIC及肝铁总量高于去铁胺组、ICL670组;去铁胺组LIC及肝铁总量与ICL670组LIC相同。(9)去铁胺组去铁治疗半年,平均排泄铁量3013.2±1684.5mg;ICL670组去铁治疗半年,平均排泄铁量3179.5±1867.2mg。经随机设计t检验,两组半年内排泄铁量之间,差异没有统计学意义(t=-0.329,P=0.743),两种药物排泄铁的量相同。(10)应用ICL670治疗的23例患者,治疗前及治疗半年后LIC经配对t检验,差异有统计学意义(t=3.314,P=0.003),治疗后LIC减低;治疗前及治疗半年后肝脏含铁总量经配对t检验,差异没有统计学意义(t=1.679,P=0.109),治疗前后肝脏含铁总量没有差别。(11)应用去铁胺治疗的29例患者,治疗前及治疗半年后LIC经配对t检验,差异有统计学意义(t=-3.359,P=0.002),治疗后LIC增加;治疗前及治疗半年后肝脏含铁总量经配对t检验,差异没有统计学意义(t=-0.934,P=0.358),治疗前后肝脏含铁总量相同。结论1、MRI测量组织信号强度比及T2值可重复性好,准确率高,稳定性好。2、通过动物实验,证实MRI测得的肝脏/肌肉信号强度比值的自然对数、肝脏R2值与LIC有显著的直线相关,4回波扫描测得肝脏R2值与LIC的相关性最强。3、通过动物试验建立的直线回归方程,适用于人体,可以根据MRI测得肝脏/肌肉信号强度比值的自然对数、肝脏R2值计算不同类型地中海贫血的LIC。4、轻型地贫肝铁浓度在正常值范围不需要去铁治疗,但是高于正常组;中间型地贫和重型地贫肝铁浓度增高,达到一定年龄需要去铁治疗。MRI定量分析可以为地贫去铁治疗时机的选择提供详细依据。5、未去铁组的LIC及肝铁总量均高于去铁胺组及ICL670组,去铁胺组与ICL670组的LIC及肝铁总量相同。半年内,两种去铁药物增加排泄铁量相同。更多还原

【Abstract】 ObjectiveTo explore the accuracy and clinical value of the MRI quantitative diagnostic,which are liver / muscle signal intensity ratio and T2 values in vitro iron concentration and liver iron deposition. Materials and Methods1. Iron in vitro model of the MRI test: Iron model was maded with a 1.5L plastic bottle, containing 1.45L of distilled water as iron mold , placed together with an animal muscle specimens at the central, surrounding with sealed water bag contained about 2.5L of distilled water. After Each MRI scan of the iron model, injected 1ml of iron dextran with the iron content of 150mg/ml in the ferrous water mold and fully shaken, animal muscle specimens and sealed water bag remain unchanged. High concentrations of iron model: With a 1.5L plastic bottle filling with 1.30L water, did first scan after the injection of iron dextran 20ml, then before each scan injected with 10ml iron dextran, with a total of 200ml.Using GRE sequencing to measure ferrous water mold / muscle signal intensity ratio, using sequence of 8 echo and 4 echo to measure T2 value of ferrous water mold. Quantitative statistical analysis the relevance between the measured values of MRI and the concentration of ferrous iron water mode and then set up the linear regression equation.2. Quantitative MRI study of liver iron overload animal model: 20 adult male New Zealand white rabbits as the experimental group, with the weight of 2.0kg to 2.9kg. One same species rabbit as a normal control group. All rabbit weighed to calculate the dose of iron dextran injection. Deep intramuscular injected iron dextran for iron 150mg/ml with the dose of 15mg/kg weekly in the experimental group, alternate the injection position right to left hind leg. Injected a total of 15 weeks. Before the iron dextran injection ,MRI inspection was made to all rabbits, quantitative analysis of normal rabbit liver / muscle signal intensity ratio and T2 value of the liver. Every time one week after the injection of iron dextran to review the MRI,and the next day two rabbits were killed randomly to get the liver, after the measurement of liver volume, part of the liver,spleen and other organs were selected to pathological diagnosis and staging iron deposition, and the remaining liver tissue were dried for the measurement of LIC by atomic spectrophotometer. Rabbit of control group were killed after the first MRI examination, Excise the liver tissue for pathological examination and determination of LIC. All rabbits weighed again before the kill, and phlebotomized 2ml venous blood sample from marginal ear vein for the investigation of serum iron. MRI examination Sequences: GRE scan for liver / muscle signal intensity ratio, 8 and 4echo scan for the T2 value of the liver. Statistical analysis the relevance of hepatic iron content and the dose of the Injected iron dextran. Quantitative statistical analysis of MRI measured values and the LIC and set up the relevance of linear regression equation for the consideration of LIC in clinical trials,in accordance with the results of quantitative analysis of MRI calculation. 3. Clinical application of MRI quantitative research of the liver iron overload: 149 cases of thalassemia including 23 cases of thalassemia minor, 10 cases of thalassemia intermedia and 116 cases of thalassemia major.Thalassemia major group are divided into three groups according to different treatment methods: 23 cases for regular blood transfusions and chelation therapy with ICL670 (ICL670 group); 34 cases for regular blood transfusions and chelation therapy with Deferoxamine (DFO group); 59 cases for regular blood transfusions and no chelation therapy(no chelation group).Calculated liver iron concentration and liver iron content in accordance with the linear regression equation based on animal experiments. Six m onths later rechecked MRI in 57patients with thalassemia major, including 23 cases treated with ICL670, 34 cases treated with DFO. Evaluated the therapeutic efficacy of the two iron chelator taked into account of the total blood transfusions and the difference between the total liver iron within six months.Magnetic resonance imaging studies were performed using a 3.0 T Philips system (Achieva 3.0T X-series). Scan parameters: (1) Transverse GRE sequences for signal intensity ratio were acquired with TR of 48ms, TE of 2.5ms, flip angle of 60°, slice thickness of 10mm, gap of 3mm, scanning four slices, completed scan within one breath. (2) T2 value measurement with a multiple spin-echo (SE) pulse sequence: 8-echo axial scan for mild iron deposition were acquired with TR of 2000ms, TE of 8, 16, 24, 32, 40, 48, 56 and 64 ms, slice thickness of 7mm, scanning one slice , take time of about 8 minutes; 4-echo axial scan for severe iron deposition in the were acquired with TR of 2000ms, TE of 6, 12, 18 and 24 ms, slice thickness of 7mm, scanning one slice, take time of about 7 minutes. Results1. Iron in vitro model of the MRI test: The differences of the iron content of water mold / muscle signal intensity ratios and iron content of water mold R2 (1/T2) values measured by two radiologists for the first 15 scans are not statistically significant. The variation coefficient of the animal muscle specimens signal strength measured in GRE sequences was 6.08%; The variation coefficient of the animal muscle specimens signal strength measured in 8 and 4 echo sequences were 2.87% and 3.90% respectively. The T2 value of water is always 2047.00ms.With GRE sequences, there was apparent inverse linear relationship between iron concentration and the natural logarithm of model / muscle signal intensity ratio (r =-0.999, P = 0.000). With 8 echo SE sequences, R2 value of the water mold correlated closely with iron concentration (r = 0.998, P = 0.000) when the iron concentration is less than 1.74mg/ml. With 4 echo SE sequences, R2 value of the water mold correlated closely with iron concentration (r =1.000, P = 0.000) when the iron concentration is less than 2.64mg/ml.2. Quantitative MRI study of liver iron overload animal model: There was apparent linear relationship between LIC and the total iron injection (r = 0.824, P = 0.000). There was apparent linear relationship between total liver iron and total iron injection (r = 0.943, P = 0.000). Linear regression equation was seted up :? = 37.376 +0.505 X (?: total liver iron predicted based on the total dose of injected iron, X: the total dose of injected iron) (F = 153.993, P = 0.000). The average liver / muscle signal intensity ratio of the 21 rabbits measured by MRI before the injection is 1.43±0.13, and the average T2 value (measured in 8 echo sequence) is 55.35±6.23. The natural logarithm of liver / muscle signal intensity ratio measured by GRE sequence and LIC were significantly negatively correlated (r =-0.917, P = 0.000). Linear regression equation was set up: ?=5.640-5.044X (?: LIC predicted by MRI, X: the natural logarithm of liver / muscle signal intensity ratio measured by GRE sequence) (F = 100.895, P = 0.000). Liver R2 measured by 8 echo SE sequence and LIC values were significant linear correlation (r = 0.789, P = 0.000). Linear equation was set up: ? =- 2.93 +199.64 X (?: LIC predicted by MRI, X: T2 value of the liver measured by 8 echo sequence) (F = 31.403, P = 0.000). Liver R2 measured by 4 echo SE sequences and LIC values were significant linear correlation(r = 0.958, P = 0.000). Linear equation was set up: ? =- 2.76 +174.39 X (?: LIC predicted by MRI, X: T2 value of the liver measured by 4 echo sequence) (F = 213.154, P = 0.000).Pathological diagnosis and grading: the normal rabbit belonged to grade 0, 8 rabbits whose LIC was from 1.2 to 5.3mg/g belonged to gradeⅠ, and the 12 rabbits whose LIC was from 6.9 to 22.2mg/g belonged to gradeⅢ, There was no gradeⅡin the experiment.3. Clinical application of MRI quantitative research of the liver iron overload: (1) Independent-samples t test was used to determine the difference. There was no statistically significant difference of the liver / muscle signal intensity ratio between 20 normal adults and 21 normal rabbits (t =-1.416, P = 0.165); there was no statistically significant difference of the liver T2 values between 20 normal adults and 21 normal rabbits (t = 1.221, P = 0.232). (2) Liver T2 value of 4 echo sequence correspond to the clinical LIC diagnostic criteria can be calculated in accordance with the linear regression equation (equation 9): T2 value of liver more than 30ms indicates concerns forchelator toxicity; T2 value of liver of 17.5 to 30ms, optimal chelation range; T2 value of liver of 10 to 17.5ms, elevated hepatic iron levels; and T2 value of liver less than 10 ms, markedly increased iron levels and potential cardiotoxicity. (3) LIC of all the thalassemia minor and the normal group were within the normal range. Among the 10 cases of thalassemia intermedia: LIC was greater than 15mg / g in one case, from 7.0 mg / g to 15.0mg / g in 6 cases, less than 7.0 mg / g in one case and less than 1.6mg / g in 2 cases. Among the 116 cases of thalassemia major: LIC was greater than 15mg / g in 14 cases, from 7.0 mg / g to 15.0mg / g in 91 cases, from 3.2 mg / g to 7.0mg / g in 9 cases and less than 1.6mg / g in 2 cases who is less than 1-year-old. (4) Independent-samples t test was used to determine the difference. LIC of thalassemia minor is more than normal group;LIC of thalassemia minor is less than thalassemia intermedia; LIC of thalassemia intermedia is the same to thalassemia major. (5) Independent-samples t test was used to determine the difference. Total liver iron of thalassemia minor is more than normal group ;Total liver iron of thalassemia minor is less than thalassemia intermedia;Total liver iron of thalassemia intermedia is the same to thalassemia major. (6) Binary regression equation of the total liver iron , age and the volume of transfusion was set up : ? = 269.84 +429.65 X1 +3.29 X2 (?: total liver iron; X1: age; X2: the volume of transfusion ). (7) Among the thalassemia major with regular transfusions, by analysis of variance, there were significant differences in both liver iron concentration and liver iron content among the groups of no iron chelation therapy, chelation therapy with DFO, and chelation therapy with ICL670. Comparison liver iron concentration and liver iron content between the two groups, liver iron concentration and liver iron content of no iron chelation therapy group were more than groups of chelation therapy with DFO and chelation therapy with ICL670, liver iron concentration and liver iron content of chelation therapy with DFO were the same to chelation therapy with ICL670. (8)After six months of chelation therapy, the average amount of iron excretion was 3013.2±1684.5mg in DFO group,and 3179.5±1867.2mg in ICL670 group by random t test, the difference was not statistically significant (t =- 0.329, P = 0.743). (9) For the 23 cases of thalassaemia major receiving chelation therapy with ICL670 , LIC was decreased six months later, but there was no statistically significant difference of liver iron content before and after the therapy. (10) For the 29 cases of thalassaemia major receiving chelation therapy with DFO, LIC was Increased six months later, but there was no statistically significant difference of liver iron content before and after the therapy. Conclusion1. There were high repeatability, accuracy and stability of signal intensity ratio and R2 measurements on MRI.2. Both natural logarithm of liver / muscle signal intensity ratio and liver R2 values measured by MRI were significant correlated with LIC, the most significant line correlation was between liver R2 value measured by 4 echo SE sequence and LIC .3. The linear regression equation of both the natural logarithm of liver / muscle signal intensity ratio to LIC and liver R2 values to LIC set up through animal experiments applies to the human body. According to the equation,LIC of different types of thalassemia can be calculate based on liver / muscle signal intensity ratio or liver R2 value measured on MRI.4. LIC of thalassemia minor is in the normal range and needn’t chelation therapy, but is more than the normal group; LIC of thalassemia intermediate and thalassemia major are higher, and need chelation therapy. quantitative analysis of MRI can provide the best time to receive chelation therapy.5. liver iron concentration and liver iron content of no iron chelation therapy group were more than groups of chelation therapy with DFO and chelation therapy with ICL670, liver iron concentration and liver iron content of chelation therapy with DFO were the same to chelation therapy with ICL670. There is no difference between the excretion iron respond to the two chelator in Six months更多还原