【作者】 肖泽锋；

【导师】 郭海科；

【作者基本信息】 南方医科大学 ， 眼科学（专业学位）， 2014， 博士

【摘要】 研究背景自上个世纪九十年代以来,眼科各项检查技术日新月异发展,光学相干断层扫描技术就是在这个时候迅速发展起来的一种新的医学断层成像分析技术,它融合了光学技术、半导体激光技术、数据模型计算和数字化图像处理技术等,在对人体进行非损伤性的活体形态学检测同时,获得生物组织内部微结构的横断面图像。特别是随着OCT从时域模式转变为频域模式,OCT发展也进入了新的时代。从OCT技术引入眼科伊始,其在黄斑裂孔诊断中的研究最早也最深入。在黄斑裂孔中,以特发性黄斑裂孔最为多见。研究者们发现OCT能清晰的发现视网膜内各层结构的形态学变化,对于黄斑裂孔的形成过程和分期,OCT检查能清晰的显示类似于病理切片的“解剖学”改变,因此能精确的解释患者与视网膜解剖结构变化相对应的视力改变和预后。在各种检查手段中,临床眼科医生发现OCT是优于各种传统检测手段的眼底影像学技术。随着手术技术和设备的日益完善,术后患者黄斑裂孔的裂孔封闭率日益提高。但是,术后OCT检查证实裂孔封闭的患者,视力改善的程度却不一致,有的患者术后视力甚至较术前下降,使得人们寻找影响IMH患者视力预后的各种原因。以往的时域OCT在特发性黄斑裂孔的观察上,多集中在黄斑裂孔直径、基底裂孔直径、裂孔面积、黄斑中央视网膜厚度、裂孔形态、裂孔鼻侧和颢侧边缘水肿程度等指标上。有的学者通过回顾性研究中发现裂孔大小、孔缘神经上皮层厚度与患眼视力密切相关。但是同时也有学者发现板层黄斑裂孔矫正视力与裂孔面积、裂孔基底宽度、裂孔位置均无相关性,而且与以往研究得出的“裂孔越大、视力越低,裂孔越小、视力越好”的结论并不一致,在径线选取的部位上也存在一定程度的主观性。所以,通过SD-OCT能够清晰分清视网膜各层组织结构的功能,寻找一个可以较好评估IMH患者手术视力预后的观察指标,成为学者们研究的热点。由于光感受器内外节是将光信息转化成生物电信号后向中枢神经传递的重要一环,IS/OS层的完整性,被认为是光感受器完整性的重要标志,是评估患者视功能预后的重要指标。越来越多的研究,通过频域OCT观察发现,IS/OS层连续性的变化与患者的视力预后相关。在SD-OCT对IS/OS层连续性的评估上,国内外一些学者将IS/OS层分为连续、中断和缺失的不同形态,进行了定性分析。有的研究者甚至分别测量IS/OS层水平、垂直两方向上缺失区直径后相乘得出缺失区面积后进行分析。我们的研究目的即是观察手术前后水平方向上光感受器内外节缺失区直径的变化,分析其与患者视力预后的关系。从“特发性黄斑裂孔”的命名上可见这一疾病的原因目前并不清楚,IMH的发病机制也一直是眼科领域一个探索了多年的课题。目前较为公认的是Gass提出的玻璃体的切向牵拉理论,认为玻璃体对黄斑中心凹切线方向上的牵拉是造成IMH的主要原因。有学者通过手术观察和后来出现的OCT技术检查发现,黄斑牵拉综合征的患者有半数病例最终会发展成为特发性黄斑裂孔,在完全后脱离的玻璃体后界膜上有时可见粘连的游离的孔盖组织。另一些学者通过OCT和手术发现玻璃体后脱离时对中心凹施加的向前的牵拉联合眼球转动时的反作用力是发生IMH的始动因素,以上观点均从玻璃体视网膜界面因素来考虑IMH的形成原因。但也有研究者为了了解在全层黄斑裂孔患者的手术前,其玻璃体是否仍粘连于后极部视网膜,对OCT检测未能清晰显示玻璃体后界膜的部分患者进行人工玻璃体后脱离,结果发现在玻璃体后皮质仍然与黄斑区视网膜贴合紧密的患者,依然发生了明显的黄斑裂孔,可见,除了玻璃体对黄斑切线方向牵拉力的作用之外,还有其他因素对特发性黄斑裂孔的形成产生作用。近来,国内外学者通过频域OCT的深度增强成像技术发现正常人群随着年龄的增加脉络膜厚度逐渐降低,而且女性的中心凹下脉络膜厚度值明显低于同龄男性,,而IMH多发于老年女性的特点也使学者们把目光转向脉络膜厚度因素上来探讨特发性黄斑裂孔的发生机制。采用传统的频域OCT仪作脉络膜的扫描获取图像是很困难的,因为所用的红外光很难穿透视网膜色素上皮层到达下面的脉络膜层。最近,一项被称为深度增强成像技术(EDI)的革新出现后,以上的难题迎刃而解,使脉络膜扫描成像成为可能。传统SD-OCT的眼底扫描主要显示视网膜内界膜到RPE层之间的组织结构。而脉络膜,负责眼内循环主要的血液供应,是许多脉络膜血管相关疾病的组织起源,如息肉样脉络膜血管病变、恶性脉络膜黑色素等。同样,另一类与年龄和微循环相关的疾病,如老年性黄斑变性和微小动脉粥样硬化,也同样与脉络膜有关。但是,用传统的OCT完成对脉络膜的非侵入性检查很难成功。首先,传统OCT采用的光波波长不够,难以穿透视网膜色素上皮层到达深层的脉络膜组织。光波二相性原理告诉我们：光程增加,散射越发衰减,而离零延时越远,反射光灵敏度则越低,而传统SD-OCT所采用的傅里叶转换又一定程度上制约了光波的穿透性。总结来说,传统OCT不能扫描脉络膜成像的缺点是：(1)信号的清晰度和分辨率会随着脉络膜扫描零延迟距离的增加而降低；(2)傅里叶转换造成传统SD-OCT的最大探测范围降低；(3)反射光波长和信号会在光波散射的过程中逐渐衰减；(4)图像双侧边界产生对光柱的离焦现象。普通的SD-OCT采用的是800nm左右的红外光,而波长为1060nm的红外光才能穿透达到脉络膜。因此为了使更深层的组织成像清晰,人们采用深度增强扫描成像技术,即采用1060nm左右的红外光投射被检查者,使光源聚焦于更深层的脉络膜组织,当组织反射的光信号延迟为零时,成像最清晰：相反,前面的视网膜血管和神经纤维层则时间延迟增加,成像相对模糊。而目前最新的光学相干断层扫描EDI技术避免了以上缺点,可以使传统SD-OCT获得清晰脉络膜扫描图像的同时,视网膜组织的图像也不受任何影响。第一部分频域OCT观察特发性黄斑裂孔手术前后光感受器内外节的变化目的通过频域光相干断层扫描(SD-OCT)观察光感受器内外节在特发性黄斑裂孔手术前后的变化。方法前瞻性对比分析23例(25只眼)确诊并接受玻璃体手术治疗的特发性黄斑裂孔患者的临床资料,采用Topcon公司Version3.21型频域OCT测量并对比分析手术前后水平方向上光感受器内外节缺失区直径(diameter of IS/OS absence,DIOA)的差别(配对资料的T检验)和相关性(Bivariate过程中的Pearson相关分析)。结果手术后有18只眼黄斑裂孔闭合,裂孔封闭率为72%。术前DIOA范围为(551～2900)μm,平均(1933.16±650.80)μm；术后12月DIOA范围为(412-2811)μm,平均(1329.20±780.13)μm,术后12月DIOA较术前减少了约604μm。经配对样本的t检验,手术前后差异有显著性意义(t=3.73,P=0.001)。经Pearson相关分析,术前DIOA与术后DIOA之间不存在有统计学意义的相关关系(r=0.37,P=0.069)。结论频域OCT能清晰观察视网膜IS/OS层的结构,特发性黄斑裂孔患者术后水平方向上光感受器内外节缺失区直径较术前明显减少,但两者无明显相关性。第二部分频域OCT观察特发性黄斑裂孔手术前后光感受器内外节与视力的关系目的通过频域相干光断层扫描观察光感受器内外节在特发性黄斑裂孔手术前后的变化并探讨与视力预后的关系。方法前瞻性对比分析28例(30只眼)确诊并接受玻璃体手术治疗的特发性黄斑裂孔患者的临床资料,分别记录手术前后的最佳矫正视力(转换成logMAR视力)和水平方向上光感受器内外节缺失区直径(diameter of IS/OS absence,DIOA),分析两者手术前后的差别(配对资料的T检验)和相关性(Bivariate过程中的Pearson相关分析)。结果手术后有22只眼黄斑裂孔闭合,裂孔封闭率为73.3%。术前logMAR视力为0.18-1.30,平均为0.99±0.37,术后12月logMAR视力为0.10-1.60,平均为0.75±0.46,术后12个月较术前平均下降了0.24。术前DIOA为551～2900μm,平均1945.57±609.33μm；术后12月DIOA为412～2811μm,平均1357.07±772.67μm,术后12个月较术前平均下降了589μm。经配对样本的T检验,术后12个月logMAR较术前logMAR明显减少,两者差异有显著性意义(t=2.859,P=0.008)；术后12个月DIOA较术前DIOA明显减少,两者差异有显著性意义(t=3.94,P<0,001)。经Pearson相关分析,术前logMAR与DIOA之间具有显著正相关性(r=0.895,P<0.01)；术后logMAR与DIOA之间同样也具有显著正相关性(r=0.921,P<0.01)。结论特发性黄斑裂孔患者手术后视力和光感受器内外节缺失区直径较术前明显改善；手术前后水平方向上光感受器内外节缺失区直径与logMAR视力具有明显正相关性,是评估患者视力预后的重要因素。第三部分频域OCT观察特发性黄斑裂孔手术前后光感受器内外节与中心凹下脉络膜厚度的关系目的通过频域相干光断层扫描(SD-OCT)及其深度增强成像技术(EDI)观察光感受器内外节和中心凹下脉络膜厚度在特发性黄斑裂孔手术前后的变化并探讨两者间的关系。方法记录35例单眼确诊并接受玻璃体手术治疗的特发性黄斑裂孔患者的临床资料(年龄、性别),EDI模式扫描后测量手术前后水平方向上光感受器内外节缺失区直径(diameter of IS/OS absence,DIOA)和中心凹下脉络膜厚度(subfoveal chorodial thickness,SFCT),分析两者的差别(配对资料的T检验)和相关性(Bivariate过程中的Pearson相关分析)。结果手术后有29只眼黄斑裂孔闭合,裂孔封闭率为83%。术前DIOA为386～2900μm,平均1257.49±678.39μm；术后6个月DIOA为203～1410μm,平均660.80±298.59μm,经配对资料的T检验,手术前后DIOA相比差异有显著性意义(t=6.926,P<0.001)。术前SFCT为85～268μm,平均130.80±37.09μm；术后6个月SFCT为93～201μm,平均137.17±32.64μm,经配对资料的T检验,手术前后SFCT相比差异无显著性意义(t=-1.780,P=0.084)。经Pearson相关分析,术前DIOA与SFCT之间具有显著负相关性(r=-0.797,P<0.001)；术后6个月DIOA与SFCT之间同样具有显著负相关性(r=-0.647,P<0.001)。结论特发性黄斑裂孔患者手术后6个月时水平方向上光感受器内外节缺失区直径较术前明显减小,但黄斑中心凹下脉络膜厚度无明显变化。特发性黄斑裂孔患者手术前与术后6个月时水平方向上光感受器内外节缺失区直径与黄斑中心凹下脉络膜厚度均具有显著负相关性,提示IMH的发生和严重程度与脉络膜的血流灌注状态可能存在一定关系。更多还原

【Abstract】 BackgroundIn the past more than10years,a new technique of medical tomography named optical coherence tomography was brought into the world followed after X-ray,CT,MRI and ultrasound. It encompassed the semiconductor laser technology, optical technology and computer image processing technique and developed at high speeds. OCT had emerged as an effective noninvasive method for examining the retinal architecture and for the management of macular abnormalities, including idiopathic macular holes (IMHs). In ophthalmology, OCT provided so far the best technology imaging the retinal ultrastructure including the layer of ganglion cells,photoreceptor IS/OS junctions, external limiting membrane(ELM) and RPEs. So OCT images offered a better help to describe how macular holes formed and to explain the causes of vision loss and to give a good postoperative visual acuity evaluation than other traditional checks. With the improvement of surgical techniques and equipments, the macular hole closure rate in IMH patients increased. However, in postoperative IMH patients with closed macular holes confirmed by OCT the VA recovery varied in different degrees,some even declined. And this made people look for kinds of parameters to evaluate the VA recovery in IMH. The past TD-OCT mainly concentrated in macular hole diameter,basal hole diameter, hole area, foveal retinal thickness, hole patterns, edema degrees of nasal and temporal edge of macular hole. Some scholars found that the size of macular hole and the thickness of neural fiber layer on the margin of the hole were closely linked with the VA in retrospective studies. But some other scholars found no correlations between hole area,basal hole diameter,hole location and VA in lamellar macular holes.And they also found the conclusion of "the bigger the hole,the worse the VA"was not always right.Therefore,to search for a better parameter to evaluate the VA recovery in IMH patients by SD-OCT became a focus of researchers.As the photoreceptor IS/OS junction is an important creature of light information into electrical signals to the central nervous system, the integrity of IS/OS was considered to to be a symbol of good photoreceptors and the key indicator of prognosis assessment of VA in patients with idiopathic macular hole. More and more studies showed that the changes of IS/OS associated with VA in IMH patients by SD-OCT. Some researchers divided IS/OS junctions into different forms of continuous, broken and missing in a qualitative analysis. Some researchers had even measured the horizontal and vertical diameters of IS/OS absence(DIOA) and then calculate the hole area by multiplication. To study the correlation between DIOA and logMAR before and after surgery in idiopathic macular hole(IMH) by spectral domain optical coherence tomography is our aim of the prospective research.The "idiopathic" means that the cause of a disease is still not clear. And the pathogenesis of IMH has been one of the focuses in ophthalmologists for many years. Now it is recognized that the tangential traction of vitreous on macular foveal is a major cause of IMH from Gass’ Theory. Through extensive observation and later coming OCT technology to track patients with vitreomacular traction syndrome(VMTS), some scholars found more than50%eventually developed into a full thickness macular hole. The other researchers discovered that the posterior vitreous detachment(PVD) forword against foveal traction combined with the rotation of the eyeball was the beginning cause of the disease by SD-OCT. All above were concluded from the vitreoretinal interface considerations. But still some other scholars had operated artificial posterior vitreous detachment in the full thickness IMH eyes failed to get OCT images in order to find if the vitreous was still sticking to the posterior retina. And the results showed that in the case of full thickness holes had formed, there were still some eyes without posterior vitreous detachment. This proved that the tangential traction caused by PVD maybe not the only factor. More recently, scholars at home and abroad found the SFCT decreased with increasing age in normal population and the female thinner than the male at the same age. So the eyesight of scholars was focused on the choroidal thickness for IMH happened in older female at a large proportion.But imaging the choroid with conventional commercial spectral-domain optical coherence tomography (SD-OCT) has been difficult, mainly because of difficulty in signal transmission beyond the retinal pigment epithelium. A recent modification to the standard technique, termed enhanced depth imaging optical coherence to-mography (EDI-OCT), was able to image the choroid with reasonable clarity using commercial SD-OCTs. The choroid, which accounts for most ocular blood flow, may be affected in several disease states, such as polypoidal choroidal vasculopathy and choroidal melanoma. It is also prone to suffer from age-related degeneration, microvascular artherosclerotic changes, and changes inherent to other microvascular systems.An understanding of the choroid using non-invasive imaging techniques has been limited. For instance, imaging of the choroid with OCT was not entirely possible.This was because the wavelength of the light source used to image the retina was not long enough to penetrate into the choroid.This was due to wavelength dependent light scattering and signal loss that occur in the image path, decreased sensitivity away from zero-delay, and the various inherited limitations with Fourier transformation.The common commercially available OCTs employ a wavelength of800nm, whereas those capable of choroidal imaging are in the range of1,060nm. One would have to use a light source nearer to the infrared region to penetrate into the choroid. This was not possible for most commercially available OCTs, because clarity for retinal structures would be compromised. To conclusion,the reasons why the choroid cannot be imaged clearly could be attributed to (1) decreasing sensitivity and resolution with increasing displacement from zero-delay,(2) decreased maximal dynamic range inherent in Fourier domain systems,(3) wavelength dependent light scattering and signal loss in the image path, and (4) the lateral width of the defocused imaging beam. When the instrument is moved closer to the eye to image deeper layers, the inverted image is displayed such that the choroid is shown facing up (ie, closer to zero-delay) while the inner retina is facing down. This has the effect of delivering the most closely focused portion of the illumination at the level of the choroid or the inner scleral border. Hence, the choroid can be imaged at higher sensitivity.Following the introduction of a new imaging technique, referred to as "enhanced depth imaging optical coherence tomography"(EDI-OCT), choroidal imaging with standard commercially available spectral-domain OCT (SD-OCT) was made possible.At the same time, retinal details could be seen with comparable quality.Part1. Observation of photoreceptor IS/OS junctions before and after surgery in idiopathic macular hole using spectral domain optical coherence tomographyObjective To evaluate photoreceptor inner segment/outer segment(IS/OS) junctions before and after surgery in idiopathic macular hole(IMH) using spectral domain optical coherence tomography(SD-OCT).Methods Prospective study of23patients(25eyes) who underwent macular hole surgery. Date collection included pre-and postoperative diameters of IS/OS absence(DIOA) in horizontal direction by SD-OCT(Topcon CO.Version3.21). The differences of pre-and postoperative diameters of IS/OS absence(DIOA) were analyzed with Paired-Samples T Test and the correlations were tested with Bivariate-Pearson Correlation.Results Macular holes in18eyes had successful anatomic closure and the closure rate was72%. The preoperative DIOA distributed from551μm to2900μm and the mean preoperative DIOA was1933.16±650.80μm. The postoperative DIOA distributed from412μm to2811μm and the mean postoperative DIOA was1329.20±780.3μm. Compared with the preoperative DIOA, the postoperative DIOA almost decreased604μm12months later and the differences had statistical significance by Paired-Samples T Test (t=3.73,P=0.001). Pre-and postoperative DIOA had no correlations by Bivariate-Pearson Correlation(r=0.37,P=0.069).Conclusions SD-OCT can be valuable for the observation of IS/OS junctions in patients with idiopathic macular hole. The postoperative DIOA was smaller than the preoperative significantly but without statistical correlations between them. Part2. Correction study of visual acuity and photoreceptor IS/OS junction in idiopathic macular hole by spectral domain optical coherence tomographyObjective To study the correlation between photoreceptor inner segment/outer segment(IS/OS) junction and visual acuity (VA) before and after surgery in idiopathic macular hole(IMH) on spectral domain optical coherence tomography(SD-OCT). Methods Prospective study of charts of28patients(30eyes) who underwent macular hole surgery. Date collection included pre-and postoperative best-corrected visual acuity(BCVA) which was converted to logMAR score, and diameters of IS/OS absence(DIOA) in horizontal direction. The above data was analyzed by Paired-Samples T Test and Bivariate-Pearson correlation.Results Macular holes in22eyes had successful anatomic closure and the closure rate was73.3%. The preoperative logMAR distributed from0.18to1.30and the mean preoperative logMAR was0.99±0.37. The postoperative logMAR distributed from0.10to1.60, and the mean postoperative logMAR was10.75±0.46. Compared with the preoperative logMAR, the postoperative logMAR decreased0.24and the differences had statistical significance by Paired-Samples T Test (t=2.859,P=0.008).The preoperative DIOA distributed from551μm to2900μm and the mean preoperative DIOA was1945.57±609.33μm. The postoperative DIOA distributed from412μm to2811μm and the mean postoperative DIOA12months later was1357.07±772.67μm. Compared with the preoperative DIOA, the postoperative almost decreased589μm and the differences had statistical significance by Paired-Samples T Test (t=3.94,P<0.001). Preoperative logMAR and DIOA were positive correlated significantly by Bivariate-Pearson Correlation(r=0.895,P<0.01). Also the postoperative logMAR and DIOA were positive correlated significantly by Bivariate-Pearson Correlation(r=0.921,P<0.01).Conclusions The postoperative visual ability and DIOA improved clearly. Preoperative and postoperative logMAR and DIOA were positive correlated significantly and DIOA maybe an important indicator of visual recovery in patients with idiopathic macular hole. Part3. Correction study of photoreceptor IS/OS junction and subfoveal choroidal thickness in idiopathic macular hole by spectral domain optical coherence tomographyObjective To observe the correlation between photoreceptor inner segment/outer segment(IS/OS) junction and subfoveal choroidal thickness(SFCT) before and after surgery in idiopathic macular hole(IMH) by enhanced depth imaging(EDI) of spectral domain optical coherence tomography(SD-OCT).Methods35unilateral IMH patients who underwent macular hole surgery were studed,including8males and27females. Pre-and postoperative diameters of IS/OS absence(DIOA) and SFCT in horizontal direction were determined by EDI-OCT. The above data was analyzed by Paired-Samples T Test and Bivariate-Pearson correlation.Results29eyes were anatomically closed and the closure rate was83%. The preoperative DIOA distributed from386μm to2900μm and the mean preoperative DIOA was1257.49±678.39μm. The postoperative DIOA6months later distributed from203μm to1410μm and the mean postoperative DIOA was660.80±298.59μm. Compared with the preoperative DIOA, the postoperative almost decreased597μm and the differences had statistical significance by Paired-Samples T Test it=6.926,P<0.001). The preoperative SFCT distributed from85μm to268μm and the mean preoperative SFCT was130.80±37.09μm. The postoperative SFCT6months later distributed from93μm to201μm, and the mean postoperative SFCT was137.17±32.64μm. Compared with the preoperative SFCT, the postoperative SFCT increased about6.37μm and the differences had no statistical significance by Paired-Samples T Test (t=-1.780, P=0.084). By Bivariate-Pearson correlation, a negative correlation between DIOA and SFCT was found in preoperative group (r=-0.797,P<0.001). Also the postoperative DIOA and SFCT6months later were negative correlated significantly by Bivariate-Pearson Correlation (r=-0.647,P=0.001).Conclusions The postoperative DIOA decreased significantly compared with the preoperative but SFCT had no statistical difference. Pre-and postoperative DIOA and SFCT were negative correlated significantly in IMH eyes and this may suggest a contributing role of the choroid in the pathogenesis of idiopathic macular hole.更多还原