Search

Background: There is evidence that additional internal limiting membrane (ILM) removal reduces the recurrence rate after pucker surgery with a similar functional outcome. On the other hand, morphological changes of the inner retinal layers after ILM peeling have been described. The aim of this study was to compare the long-term data after vitrectomy with and without ILM delamination in order to uncover possible differences in morphological and functional results., Methods: In a prospective study of 32 patients with idiopathic epiretinal membrane, 16 patients were randomized into each of 2 groups. Both groups underwent pars plana vitrectomy (ppV) with peeling of the epiretinal membrane. In group 1 no forced additional peeling of the ILM was performed and in group 2 the ILM or ILM residues were additionally removed after staining. The investigated parameters were visual acuity, central retinal thickness (CRT) in optical coherence tomography (OCT), metamorphopsia and surgical complications. The time points of the examinations were directly preoperative, after 1, 3 and 6 months and partly 8.4 years postoperatively., Results: In group 1 (n = 15) the preoperative mean visual acuity improved from 0.54 logMAR to 0.38 logMAR after 6 months postoperatively (n = 13). Of this group 6 patients could be examined in the long-term course and the visual acuity improved further to 0.32 logMAR after 8 years. The CRT decreased from 473 µm preoperatively to 235 µm in the long-term interval. In group 2 (n = 15) the mean visual acuity preoperatively was 0.47 logMAR and improved 6 months postoperatively (n = 13) to 0.38 logMAR and in the long-term examination (n = 5) to 0.1 logMAR. The CRT in this group decreased from 417 µm preoperatively to 278 µm in the long-term interval. In group 1 one recurrence occurred in the follow-up period, in group 2 none., Conclusion: The study showed that there was no significant difference in visual acuity and CRT between the two groups neither after 6 months nor after 8 years of follow-up. The observed recurrence in the group without ILM delamination underlines the assumption that additional ILM peeling could reduce the recurrence rate.

Background: The introduction of full fundus angiography (FAG) for retinal vein occlusion (RVO) has enriched diagnostic imaging; however, it is unclear how much of the retina has to be visualized to evaluate the risk of neovascularization. We investigated the 102° Spectralis camera to assess if imaging beyond the central 102° is necessary., Methods: In this retrospective study we assessed 80 eyes of 80 patients with RVO to analyze central and peripheral ischemia using the ultra-widefield 102° Spectralis camera. The standard viewer software was used to delineate ischemic areas (IA) and to measure the amount of pixels of IA in correlation to the total amount of pixels in the peripheral image., Results: Among the 80 eyes with RVO there were 51 eyes with ischemic areas (central and/or peripheral), of which IA could be detected in 49 eyes by analyzing a fovea-centered FAG image. The average IA in these 51 eyes was 23.5% of the total image area and 53.7% of disc areas, respectively., Discussion: A fovea-centered FAG image with a 102° angle appears to be sufficient to determine the clinically relevant extent of IA. Old recommendations and imaging methods should be rethought. The 102° Spectralis camera provides a clinically suitable, fast and exact method which might replace old methods to describe and document the extent of ischemia by disc areas.

Kranenburg's syndrome is defined as central serous detachment associated with an optic disc pit. We report the interesting and very seldom combination of Kranenburg's syndrome and iris-retinochoroidal-coloboma. Reattachment was achieved after vitrectomy, peeling of epiretinal membranes, laser coagulation adjacent and temporal to the optic disc and gas endotamponade. Coincidence of this syndrome with an iris-retinochoroidal-coloboma is extremely rare. Both coloboma and optic disc pit are based on closure defects during week 6 of embryogenesis.

A growing body of evidence supports the notion that inflammatory reactions in the central nervous system (CNS) are not only restricted to established immune mediated disorders, such as multiple sclerosis, but also contribute to the pathogenesis of Alzheimer's disease and other types of neurodegenerative disorders. The biological roles of toxic mediators, such as nitric oxide, reactive oxygen species, as well as complement and proteases in the genesis of inflammatory reactions in the CNS are reviewed within the context of demyelination and neuronal damage.

The distribution of proinsulin and insulin immunoreactivity was studied in 76 human insulinomas and in normal pancreas. One trabecular and two solid insulinomas showed the staining pattern of normal beta cells. A "near normal" staining pattern (perinuclear proinsulin and diffuse or polarized insulin staining) existed in 10 of 27 trabecular and 11 of 44 solid insulinomas. An "intermediate" staining pattern (intense perinuclear as well as weaker diffuse proinsulin staining with diffuse or polarized insulin staining) was observed in 10 of 27 trabecular and 20 of 44 solid insulinomas. Different "abnormal" staining patterns were found in 6 of 27 trabecular and 6 of 44 solid insulinomas. Of the 5 glandular insulinomas, 4 exhibited a "near normal" and one an "abnormal" staining pattern. No correlation was found between any particular staining pattern and the multihormonality or malignancy of the insulinomas. The diffuse labeling for proinsulin in about 50% of the insulinomas is suggestive of abnormal prohormone processing.

We have carried out an immunohistochemical investigation of 15 human insulinomas applying monoclonal antibodies specifically recognizing proinsulin and insulin. Our results demonstrate that the epitopes unique to proinsulin and insulin can be detected with the respective monoclonal antibodies using the protein A-gold technique after routine formaldehyde fixation and paraffin embedding of the tissues. The immunostaining pattern for proinsulin and insulin in the insulinomas was different from the observed in B cells of pancreatic islets present in the adjacent normal pancreas. Furthermore, the pattern of immunostaining was found to vary from tumor to tumor. These findings strongly suggest the possibility of a disturbed proinsulin to insulin conversion in human insulinomas.