Optic Disc Image Analysis Applied For Glaucoma&Regulation Of Apoptotic Genes And Protection Of Ginkgobiloba Extract In Optic Nerve Injury

Glaucoma is a leading cause of irreversible blindness. A recent report from the World Health Organization attributes 13.5% of global blindness to the disease. The detection of early glaucomatous damage is important for the treatment of disease. At present, the clinical diagnosis of glaucoma is based principally on three measure:intraocular pressures measurements, visual field results and optic nerve head analysis. The important of elevated intraocular pressures as a risk factor for glaucoma has been shown, but its diagnostic precision for glaucoma is less powerful. Intraocular pressures identified less than one half of those with glaucoma. Visual field was thought of a gold standard, but structural changes to the optic nerve and nerve fiber layer usually precede glaucomatous visual field loss as currently measured for 4-5 years. Visual field defect was measured when 30% retinal nerve fiber has lost. Sheilda think that optic disc and retinal nerve fiber layer were objective index for diagnosing early glaucoma. Therefore, optic disc image analysis applied for glaucoma is important for diagnosing early glaucoma. Chapter one: Inter-and intraobserver variation in the analysis of optic disc images with Heidelberg retina tomograph Purpose: To measure the inter-and intraobserver agreement of Heidelberg retina tomograph measured optic disc parameters. Methods: Intraobserver agreement: The optic disc parameters of 25 normal subjects were imaged three times with the HRT by author in one time per 10 minute. Interobserver agreement: The optic disc parameters of 30 subjects (16 normal subjects and 14 patients with glaucoma) were imaged three times with the HRT by three observers in same time. Three observers who were masked traced 30 images disc margin. Agreement of inter-and intraobserver was tested for each parameter using interclass correlation coefficients (ICC) and coefficient of variation (CV) and analysis of variance (ANOV). Results: ①Intraobserver agreement : CV of cup area, rim area and volume,cup shape, RNFL thickness, horizontal cup/disk ratio,vertical cup/disk ratio was less than 5%. ②ICC of cup area, rim area and volume,cup shape, RNFL thickness, horizontal cup/disk ratio was more than 0.8. There was no significant different in optic disc parameters (p>0.05). Conclusion: Cup area, rim area and volume,cup shape, RNFL thickness, horizontal cup/disk ratio were almost perfect intraobserver agreement. Those can be used for follow up.Chapter two: Identification of early primary open-angle glaucoma cases with the Heidelberg Retina Tomograph Objective: To determine which structural optic nerve head parameters measured with HRT analysis that best separate patients with early glaucoma from normal subjects, establish normal range of HRT optic nerve head parameters with normal subjects, define which affect HRT optic nerve head parameters on normal subjects and the correlation between optic nerve head parameters and visual field index. Methods: 102 normal subjects(132 eyes )and 117 patients(128eyes,age, refractive error matched)with primary open-angle glaucoma(early stage 67eyas,advanced stage 32 eyes and late stage 29 eyes)were studied . Optic nerve head parameters were measured using Heidelberg retina Tomograph, including: optic disc area, cup area, cup volume, rim area, rim volume, cup shape, mean retinal nerve fiber layer thickness, retinal nerve fiber layer cross-section area, horizontal C/D, and vertical C/D. Visual field was examined using Octopus perimeter .The range of the optic disc parameters were defined in normal subjects. The relation between optic disc area and other optic nerve head parameters by liner regression derived from the normal subjects. Compared the optic disc parameters between normal and glaucoma. Determined which structural optic nerve head parameters measured with HRT analysis that best discriminate between normal persons and those with early glaucomausing step regression discriminate analysis. The sensitivity and specificity values of the method were calculated. Receivers operating characteristic (ROC) curves of discriminate function were plotted. Comparison of the area under the ROC curves was performed with nonparametric statistics. Results ①Statistically significant difference were not found between right and left, male and female in optic nerve head parameters(p>0.05,t-test). Significant negative correlation was found between age and retinal nerve fiber layer thickness and retinal nerve fiber layer cross-section area(r=-0.234, -0.235 p<0.05). There were not significant correlation between age and other optic nerve head parameters. There were statistically significant liner correlation between optic disc area and other optic nerve head parameter except rim volume and maximum cup depth. ②Statistically significant difference were not found between normal subjects and early glaucoma in optic disc area. Cup area, cup volume, cup depth and C/D had increased; rim area, rim volume , mean retinal nerve fiber layer thickness, retinal nerve fiber layer cross-section area significantly had decreased in early glaucoma. There was statistically significant difference between normal subjects and early glaucoma in cup shape (p<0.001). ③Statistically significant difference were not found between early glaucoma , advanced glaucoma and late glaucoma in optic disc area. Cup area, cup volume, cup depth and C/D had gradually increased ; rim area, rim volume , mean retinal nervefiber layer thickness, retinal nerve fiber layer cross-section area significantly had gradually decreased from early, advanced to late glaucoma. There was statistically significant difference in cup shape in the three stages glaucoma (p<0.001). ④Significant negative correlation was found between mean deviation of visual field and rim volume, rim area, mean retinal nerve fiber layer thickness; Significant positive correlation was found between mean deviation of visual field and C/D, cup area. ⑤The highest specificity (94.2%) and sensitivity(83.3%) values to separate normal subjects and these patients with early glaucoma were obtained using 95% prediction interval from the multivariate discrimination between rim area and cup shape. The specificity and sensitivity values to separate normal subjects and these patients with early glaucoma were obtained using single optic parameter was that cup shape was 81.8 %and 82.4%, rim volume was 81.8% and 85.3%, rim area was 78.8% and 89.7%, cup/disc area ratio was 78.8 % and 94.1%, vertical C/D and horizontal were 81.8 %and 80.9%. ⑥The largest ROC curve area was found for cup shape measure (area=0.889),cup/disc area ratio(area=0.890),rim area(area=0.869),rim volume (area=0.889), cup area(area=0.850),vertical C/D(area=0.860)and horizontal C/D(0.800). Conclusion: The HRT, using multivariate discrimination, provided good separation between normal subjects and patients with early glaucoma in this population.Chapter 3: A comparative study of optic nerve damage between high-tension glaucoma and normal tension glaucoma. Abstract: Objective To investigate the characteristics of optic nerve damage in High-tension glaucoma (HTG)and normal tension glaucoma (NTG) . Methods: Using Heidelberg Retina Tomograph II measured the optic disc parameter of 47 eyes of 39 subjects with HTG and 38 eyes of with NTG and the color stereo-optic disc photographs and Octopus automated perimeter. Results The result of HRT: Rim area,rim/disc area ratios in global and three sectors( except temporal quadrant) were significantly smaller, C/D was significantly larger in patient with NTG than HTG (p<0.05). The mean RNFL thickness and RNGL cross-sectional areas were significantly smaller at temporal/inferior and temporal/superior in those with NTG than HTG (p<0.05). The global rim volume was significantly smaller and global mean cup depth was significantly larger in patients with NTG than HTG (p<0.05). 2.The type of retina nerve fiber layer (RNFL) losing: RNFL diffuse loss and local defect ratio were 53.2% and 4.3% respectably in HTG; while RNFL diffuse loss and local defect ratio were 21.5% and 55.3% respectively in NTG .The proportion of the type of RNFL defect is significant statistically in HTG and NTG (p<0.01).Conclusions: C/D and C/D area ratio were larger in patients with NTG than in those with HTG. Rim area was smaller and RNFL was thinner in NTG eyes. The type of RNFL defect is mainly diffuse defect in HTG and local defect in NTG. The results indicated that there might be different mechanisms of optic nerve damage in the two types glaucoma. Chapter4 a morphological study of neuroretinal rim for different shape of optic disk in normal eyes and early glaucoma Objective: To investigate the morphological characteristics of neuroretinal rim in different types of the normal optic disc and early glaucoma. Methods: The subjects were divided into four groups: ①41 eyes with physiologic small disc. ②40 eyes with physiologic large disc. ③42 eyes with normal size disc. ④45 eyes with early glaucoma. The optic disc area,neuroretinal rim area,cup area and a circuit(each 10°) neuroretinal rim widths were measured using computerized imaging system. The curve of sequential neuroretinal rim widths was made from temporal to superior, nasal, inferior. Results ①The morphological characteristic of the curve of sequential rim widths in normal eyes was that there are double humps in inferior and superior section, double valley in nasal and temporal section. ②Thedouble humps of inferior and superior section loss at the curve of sequential rim widths in early glaucoma . It was lower than nasal and higher than temporal section. ③Inferior rim widths was widest in physiologic large disc , superior rim widths was widest in physiologic small disc, followed by the nasal and temporal section. ④The correct discrimination ratio was 85.7% and 90.6% with rim area plus C/D and sequential neuroretinal rim width(inferior 6:20 and superior 1:00 neuroretinal rim widths were the most effective) respectively in the multivariate discrimination. Conclusion: The nasal neuroretinal rim width in one was regard as standard to compare superior and inferior with nasal neuroretinal rim widths in estimating the morphological characteristics of neuroretinal rim. This way was very effective in discrimination physiologic large optic cup and early glaucoma.Part two: Regulation of apoptotic genes and protection of ginkgo biloba extract in optic nerve injury Backgrounds: The diseases resulted in optic nerve damage are glaucoma,optic nerve injury, optic neuritis, Leber's disease and neuro-myelitis optica. The retinal ganglion cells(RGCs) difficultly recover once they died. Until now,there is little effective treatment for optic nerve damage in clinic. Recently, the research shows that there are primary optic nerve damage and secondary optic nerve damage. Primary optic nerve damage resulted from primary injury and secondary optic nerve damage is RGCs degeneration or even died because of the change of microcircumstance, which is caused by the primary optic nerve damage. The research shows that RGCs are apoptosis after the optic nerve injury,which is regulated by apoptotic gene. Caspase family plays a central role in neuronal apoptosis. However, caspase3 is the most active one. In order to further identify caspase3's effect on apoptosis and effect of ginkgo biloba extract(GBE) on protection of optic nerve damage,we perform the study as follows: Objective: A model of optic nerve injury was established successfully,which made a foundation on the research of optic nerve protection. We identify the death's style of RGCs after optic nerve crush by apoptosis and regulation of apoptotic genes. We investigate effect of GBE on RGCs after optic nerve injury.Methods: The optic nerve was crushed under microscope using calibrated cross-action forceps. Apoptosis of RGCs were observed with light microscope and TUNEL. We explored the expression of caspase3 on mRNA and protein levels by RT-PCR and immunohistochemistry and protection effect of GBE on RGCs after optic nerve injury with retrograde labeling of RGCs. Results: 1.The number of RGCs decrease severity after optic nerve injury, an initial accelerated loss of RGCs at 2 weeks after nerve injury, followed by protracted declines in RGCs densities. RGCs lose 34.4% at 4 weeks after nerve injury. 2.Immunohistochmical detection of caspase3 showed that normal rat retina showed minimum fused background staining in all layers of the retina. Three days after optic nerve injury ,scattered nuclei with positive staining in the RGCs and the inner nuclei lay(INL)were noted. At 7 days, elevated expression of caspase3 protein was observed in the RGCs and the INL, and declined expression of caspase3 at 14 days. At 28 days, the staining retina was similar to the normal. 3. Expression of caspase3 mRNA remained unchanged. 4. Ginkgo biloba extract decreased the number of TUNEL positive cells in the RGCs layer obviously. Retrograde labeling of RGCs showed that the number of RGCs in ginkgo biloba extract and brimonidine groups were more than control(p<0.05). Conclusion: The model of optic nerve injury was established successfullyin order to study optic nerve damage and protection. Apoptosis was involved in RGCs dead , the expressions of caspase3 protein elevated transiently and caspase3 mRNA unchanged after optic nerve injury. GBE inhibited apoptosis of RGCs, taken on neuroprotection.