【作者】 朱珍；

【导师】 邵肖梅； 蒋泽栋；

【作者基本信息】 复旦大学 ， 儿科学， 2011， 博士

【摘要】 目的：观察正常新生儿脑在磁敏感加权成像上的表现,并探讨纠正胎龄足月的早产儿和出生28天内的足月新生儿脑在SWI的图像上是否有差异。材料和方法：10例临床上无脑损伤的高危因素及临床表现的新生儿,为排除其他疾病可能引起的脑损伤而进行常规MRI检查,结果显示未见明显异常。10例新生儿中5例为纠正胎龄足月的早产儿,5例为出生28天内的足月儿。应用西门子Avanto 1.5T超导型MR扫描系统的fast SWI序列,进行SWI检查。扫描时间为3分29秒。观察SWI图上正常新生儿脑的表现。结果：10例新生儿成功进行SWI序列。SWI图上新生儿脑实质呈现比较均匀的等信号,灰白质分界不清,脑实质内除了连续的呈低信号的静脉,其余区域未见异常低信号影；脑室系统呈现较正常脑实质略低的中等信号。MinIP图可清晰显示新生儿主要的静脉和脑表面的细小静脉。对SWI图进行不同层厚的MinIP重建,可提供更全面的关于静脉和解剖部位的信息。结论：纠正足月龄的早产儿和28天内的足月儿的脑SWI图像未见显著不同,正常新生儿SWI图除了静脉呈低信号,其余脑实质内未见低信号影。SWI图MinIP图能有助于显示新生儿脑静脉。第二部分磁敏感加权成像在新生儿颅内出血的应用目的：探讨SWI在新生儿颅内出血的应用价值。材料和方法：对596例临床怀疑脑损伤的新生儿同时进行常规MRI和SWI扫描,常规MRI和/或SWI上诊断为脑出血的病例共134例,其中早产儿81例,足月儿53例。分析不同部位颅内出血在SWI上的表现；记录常规MRI和SWI检测的出血病例数和出血灶数目,并记录最大出血灶的面积。结果：134例中19例仅显示颅内脑外出血(17例为硬膜下出血,1例为硬膜下出血伴蛛网膜下腔出血,1例为硬膜外出血),19例均在SWI图像中显示,但其中1例硬膜下出血在常规MRI上未显示。115例新生儿在SWI图像上显示存在脑内出血,其中16例常规MRI未显示。98例新生儿颅内出血(包括生发基质-脑室内出血、大脑和小脑实质出血)最大病灶的面积在SWI上大于常规MRI的T2WI序列,两者之间的差异有统计学意义(Z=-8.337,P<0.01)。生发基质-脑室内出血(6个部位)中,除了中脑导水管出血外,其它5个部位的检出阳性数均是SWI明显优于常规MRI,两者之间的差异有统计学意义(P<0.05)。SWI对蛛网膜下腔出血的检出阳性数明显优于常规MRI,两者之间的差异有统计学意义(P<0.05)；对硬膜下出血和硬膜外出血的检出阳性数上,两者之间的差异无统计学意义(P>0.05)。结论：SWI较常规MRI在检出新生儿各种类型的脑内出血和蛛网膜下腔出血的阳性率、病灶数目、大小方面有着明显的优势,可作为常规MRI的有力的补充。第三部分磁敏感加权成像在新生儿脑白质损伤中的应用目的：探讨磁敏感加权成像在新生儿脑白质损伤病变(斑点状白质病变和囊性脑室周围白质软化)中的应用价值。方法：对2008年8月-2010年4月中我院常规MRI上显示为斑点状白质病变(punctate white matter lesions, PWML)和囊性脑室周围白质软化(Periventricular leukomalacia, PVL)的新生儿进行SWI检查。其中PWML64例,PVL29例。64例1PWML病例中34例早产儿,30例足月儿,根据T2WI上信号特征将PWML病变分为两组：第一组病变在T2WI上呈等信号；第二组病变在T2WI上呈低信号,分别观察两组PWML在SWI上的表现。29例PVL病例中24例早产儿,5例足月儿观察PVL的囊性病变区有无出血和侧枝静脉形成。结果：64例PWML中仅6例(9.4%)在SWI可见斑点状低信号出血。第一组共17例,仅1例SWI示相应PWML区域显示出血,其余16例SWI示相应PWML区域未显示出血；第二组共47例,其中42例SWI上相应PWML区域未显示出血,仅5例SWI示相应PWML区域显示出血。29例PVL病例常规MRI未显示囊性病变区存在出血和侧枝静脉形成,在SWI图像上,29例PVL中6例(20.7%)囊性病变区域见斑点状低信号的出血,23例(79.3%)SWI上相应区域未见呈低信号的出血灶；29例PVL中18例(62.1%)在囊性病变区域见侧枝静脉形成,其余11例(37.9%)SWI上PVL区域未见明显的侧枝静脉形成。结论：SWI能鉴别新生儿脑T1高信号而T2等或低信号的PWML内是否存在出血,新生儿大部分PWML在SWI上显示无出血。SWI能显示常规MRI上难以显示的PVL囊性病变区域是否存在出血和侧枝静脉形成,可作为常规MRI的补充。更多还原

【Abstract】 PartⅠThe application of MR susceptibility weighted imaging in normal neonatal brainObjective:To investigate the the appearance of normal neonatal brain on susceptibility weighted imaging.Methods:Ten neonates without high risks of brain injury and the manifestation of neonatal brain injury were examined with a 1.5T MR imaging system (Avanto; Siemens, Erlangen, Germany). All of 10 neonates were performed with conventional MRI sequences and SWI. The routine MR imaging protocol included sagittal and transverse T1-weighted, transverse T2-weighted, transverse Flair(fluid-attenuated inversion recovery) and transverse diffusion-weighted imaging. There were no abnormal appearances on conventional MRI. Among of 10 neonates,5 neonates were preterm at term-equivalent age, another 5 neonates were term neonates. The acquisition time of SWI was 3 minutes and 29 seconds. To assess appearance and signal-intensity changes of the healthy neonatal brain on SWI.Results:All of 10 neonates were successfully performed with conventional MRI sequences and SWI. The cerebral parenchyma has intermediate signal intensity with loss of gray/white matter differentiation. There is partial suppression of cerebrospinal fluid (CSF) signal with the ventricles of intermediate signal, slightly hypointense to brain. The small cortical veins are seen as linear hypointensities. The MinIP images improve visualisation of cortical veins and demonstrates their tubular nature enabling differentiation from focal hemorrhagic lesions. Conclusions:There was no evidence difference on SWI between preterm at term-equivalent age and term neonates. The cortical veins are hypointense on SWI, and there no other hypointense in pararenchyma of neonatal brain.The MinIP images improve visualisation of veins. PartⅡThe application of MR susceptibility weighted imaging in intracranial hemorrhage of neonateObjective To evaluate neonatal intracranial hemorrhage with susceptibility weighted imaging (SWI), and to explore the value and limitation of SWI compared with conventional magnetic resonance imageing.Materials and Methods We conducted a prospective study including 596 neonates who were suspected injury of brin. All MR examinations were performed at 1.5 Tesla unit including conventional MR(T1, T2 and Flair sequences), DWI and SWI. Among 596 neonates there were 134 neonates with intracranial hemorrhage detected on routine MRI sequence and/or SWI.Among neonates with intracranial hemorrhage there were 81 preterm and 53 term neonates. To access neonatal intracranial hemorrhage appearance on SWI, and compare the numbers and areas of hemorrhage on conventional MRI sequences and SWI.Results Among 134 neonates with intracranial hemorrhage, there were only 19 case showed extracerebral hemorrhage (17 cases subdural hemorrhage, and 1 case extradural hemorrhage with subarachnoid hemorrhage, and 1 case extradural hemorrhage), all 19 cases were showed on SWI, and 1 case with subdural hemorrhage did not showed on conventional MRI sequences.115 case were detected with parenchymal hemorrhage, and among them 16 cases did not be detected on conventional MRI sequences.98 cases with parenchymal hemorrhage were measured the maximal areas, the maximal areas of hemorrhage was larger on SWI than conventional,the statistical differences were considered as statistically significant (Z =-8.337, P<0.01). Six areas of germinal matrix hemorrhage/intraventricular hemorrhage, except the hemorrhage of aqueduct of mesencephalon, other 5 areas with hemorrhage were deteced more cases on SWI than conventional MRI sequences, the statistical differences were considered as statistically significant (P<0.05).SWI detected more cases with subarachnoid hemorrhage than conventional MRI sequences. There was no significant differences when compared subdural hemorrhage and extradural hemorrhage detected on SWI and conventional MRI.Conclusions:SWI is a potent examination for cerebral bleeding displaying due to the higher sensitivity compared with the other conventional MRI sequences, and has predominant advantage over conventional MRI sequences in detecting intracranial hemorrhage of neonatePartⅢThe application of MR susceptibility weighted imaging in neonatal white matter injuryObjective:To evaluate neonatal white matter injuty (punctate white matter lesions and periventricular leukomalacia) in neonatal brain injury with susceptibility weighted imaging,and to explore the value and limitation of SWI compared with conventional magnetic resonance imageing.Materials and Methods:We conducted a prospective study including 64 neonates presenting with PWML and 29 neonates presenting with periventricular leukomalacia at first MRI. All MR examinations were performed at 1.5 Tesla unit including conventional MR (T1, T2 and Flair sequences), DW1 and SWI. PWML were defined as punctuate lesions with T1 hyperintensity and T2 isointensity or hypointensity in the white matter. Among neonates with PWML there were 34 preterm and 30 term neonates. PWML were classified into two groups:1) T1 hyperintensity and T2 isointensity; 2) T1 hyperintensity and T2 hypointensity. To analyze whether two groups of PWML contains hemorrhage on SWI. Among neonates with PVL there were 24 preterm and 5 term neonates. To accesse whether PVL contains hemorrhage and/or collateral vein on SWI.Results:Among all cases of PWML, there are only 6 cases (9.4%) showed evidence of hemorrhage on SWI. There were 17 cases in group 1, only 1 case showed punctate hypointensity in the areas of PWML on SWI, and other 16 cases showed no hemorrhage on SWI.47 cases were in group 2, only 5 cases showed evidence of hemorrhage on SWI, and other 42 cases were absent hemorrhage on SWI. Among 29 cases with PVL, there were 6 (20.7%) cases showed punctate hypointensity in the areas of PVL on SWI, and other 23 cases were abent hemorrhage on SWI. There were 18 (62.1%) cases showed collateral vein in the areas of PVL on SWI, and other 11 cases were abent collateral vein in the areas of PVL on SWI.Conclusion:Most areas of PWML in neonatal brain showed no hemorrhage on SWI. SWI can help identify whether PWML of neonates contains hemorrhage. SWI also can help identify whether PVL of neonates contains hemorrhage and collateral vein.更多还原