Wide-field Specular Microscopy Research Articles

Deep learning for detection of Fuchs endothelial dystrophy from widefield specular microscopy imaging: a pilotstudy

BackgroundTo describe the diagnostic performance of a deep learning (DL) algorithm in detecting Fuchs endothelial corneal dystrophy (FECD) based on specular microscopy (SM) and to reliably detect widefield peripheral SM images with an endothelial cell density (ECD) > 1000 cells/mm2.MethodsFive hundred and forty-seven subjects had SM imaging performed for the central cornea endothelium. One hundred and seventy-three images had FECD, while 602 images had other diagnoses. Using fivefold cross-validation on the dataset containing 775 central SM images combined with ECD, coefficient of variation (CV) and hexagonal endothelial cell ratio (HEX), the first DL model was trained to discriminate FECD from other images and was further tested on an external set of 180 images. In eyes with FECD, a separate DL model was trained with 753 central/paracentral SM images to detect SM with ECD > 1000 cells/mm2 and tested on 557 peripheral SM images. Area under curve (AUC), sensitivity and specificity were evaluated.ResultsThe first model achieved an AUC of 0.96 with 0.91 sensitivity and 0.91 specificity in detecting FECD from other images. With an external validation set, the model achieved an AUC of 0.77, with a sensitivity of 0.69 and specificity of 0.68 in differentiating FECD from other diagnoses. The second model achieved an AUC of 0.88 with 0.79 sensitivity and 0.78 specificity in detecting peripheral SM images with ECD > 1000 cells/mm2.ConclusionsOur pilot study developed a DL model that could reliably detect FECD from other SM images and identify widefield SM images with ECD > 1000 cells/mm2 in eyes with FECD. This could be the foundation for future DL models to track progression of eyes with FECD and identify candidates suitable for therapies such as Descemet stripping only.

Evaluation of changes in corneal endothelium in chronic kidney disease

Purpose:Chronic kidney disease (CKD) is an emerging health problem worldwide. In CKD corneal endothelial changes also occur probably due to accumulation of inflammatory cytokines and increased multiple toxic products. The aim of this study was to analyze the effect of CKD on corneal endothelium and correlate the findings with severity of disease with help of noninvasive technique.Methods:The study comprised 75 eyes of 75 cases divided into three groups with group A comprising of CKD cases on dialysis, group B of nondialysis CKD cases, and group C of controls. Each group had 25 cases each of either sex and between 15–80 age groups. All patients were investigated for blood urea, serum creatinine, and blood sugar and underwent complete ophthalmic examination of both eyes along with wide-field specular microscopy examination.Results:The majority of patients (33.3%) belonged to age range of 61–70 years with male predominance and the most common cause of CKD was found to be diabetes with 17 (34%) cases. We found normal corneal endothelial cell density (ECD) with the mean ECD of 2364.52 ± 397.72 mm2 in the dialysis group, 2467.8 ± 352.88 mm2 in nondialysis group, and 2521.68 ± 250.26 mm2 in the control group of patients. However, we found significant increase in coefficient of variation (CV) with 36 ± 5.8% in dialysis group, 37 ± 4.5% in nondialysis group and 32 ± 0.8% in controls (P = 0.001) and decreased hexagonality (Hx) with 47 ± 7.3% in dialysis group, 46 ± 4.7% in nondialysis group and 51 ± 6.7% in the controls (P = 0.031). This showed increased tendency of pleomorphism and polymegathism in corneal endothelial cells in CKD cases. No correlation was found between blood urea or serum creatinine levels with endothelial parameters in any group.Conclusion:CKD causes morphological changes like polymegathism and pleomorphism in corneal endothelium and hence these cases are more vulnerable and special care should be taken before any intraocular surgical procedure.

A standardized methodology for longitudinal assessment of corneal endothelial morphometry in eye banked corneas.

Eye banked research-grade human donor corneas serve as principal ex vivo source for studying the mechanisms that underlie corneal endothelial cell damage/death and survival. Wide-field specular microscopy can be used for corneal endothelial visualization and allows for indirect assessment of endothelial cell function by analyzing endothelial cell density and morphometric parameters. However, a standardized approach is needed to observe corneal endothelial changes over time. This protocol describes reliable ex vivo methods for consecutive analyses of human donor corneal endothelial cell density and morphometric parameters change using a wide-field dual imaging specular microscope. This protocol involves tissue warming, acquisition and analysis of specular endothelial images, assessment of corneal layers with the new Enhance mode, optical pachymetry measurement, and qualitative image quality grading scales. This quantitative and qualitative evaluation of donor corneas allows for a systematic analysis of endothelial dynamic responses to ex vivo induced stress and can be used as a valuable tool to better elucidate specular findings and mechanisms mediating corneal endothelial cell loss in corneal disease and after transplantation.

Predicted Long-term Outcome of Corneal Transplantation

To analyze graft survival and the outcome of the corneal endothelium after corneal transplantation in a single model to predict the long-term prognosis of these grafts. Cohort study. Data were recorded prospectively and then analyzed retrospectively. One thousand one hundred forty-four consecutive eyes of 1144 patients who underwent corneal transplantation between 1992 and 2006. Penetrating keratoplasty and deep anterior lamellar keratoplasty. Slit-lamp examination and wide-field specular microscopy results. A joint analysis of endothelial cell loss and time to graft failure was undertaken. From midterm simultaneous analysis of graft survival and endothelial cell loss, long-term graft survival was predicted. The observed 5- and 10-year graft survival estimates were, respectively, 74% and 64%. The average endothelial cell density (cell loss) was 2270 cells/mm(2) before surgery, 1058 cells/mm(2) (-53%) during the sixth postoperative year, and 865 cells/mm(2) (-61%) during the 10th postoperative year. Overall, the predicted graft survival estimate was 27% at 20 years and 2% at 30 years. Both observed and predicted graft survival were higher in patients who had undergone lamellar keratoplasty than in patients who had undergone penetrating keratoplasty and had normal recipient endothelium and higher in patients who had undergone penetrating keratoplasty and had normal recipient endothelium than in patients who had undergone penetrating keratoplasty and had impaired recipient endothelium. For corneal diseases involving the endothelium, penetrating keratoplasty seems to be a good therapeutic approach in elderly patients because the graft life-span may be similar to the patient life expectancy. Conversely, for younger patients, penetrating keratoplasty is only a midterm therapeutic approach. For corneal diseases not involving the endothelium, deep anterior lamellar keratoplasty seems to be a promising therapeutic approach with higher long-term expected survival.

Evaluation of corneal endothelium and keratic precipitates by specular microscopy in anterior uveitis

BACKGROUNDThe study of corneal endothelium, by specular microscopy, in patients with anterior uveitis has largely been restricted to observations on the endothelial cells. In this prospective study “keratic precipitates” (KP)...

Effects of penetrating keratoplasty rejection on the endothelium of the donor cornea and the recipient peripheral cornea

PURPOSE: To determine the effects of penetrating keratoplasty rejection on the recipient endothelium. METHODS: Twenty transplanted corneas of 20 keratoconus patients were enrolled in this study. Ten of the corneas had undergone allograft rejection and the other 10 grafts had an uneventful postsurgical course. The endothelium of the donor cornea and the endothelium of the recipient peripheral cornea were evaluated by wide-field specular microscopy. RESULTS: The transplanted corneas that underwent allograft rejection showed a marked decrease in endothelial cell density not only in central donor cornea but also in the recipient peripheral cornea compared with that in the uneventful transplanted corneas. Additionally, the recipient endothelium had significant pleomorphism in the rejection group. There were no morphologic differences in the peripheral donor cornea between the control group and the rejection group. CONCLUSIONS: Our findings indicate that the recipient peripheral endothelium is also affected by allograft rejection, perhaps because of the contribution of recipient endothelium to the wound healing process after allograft rejection.

Corneal endothelial changes in patients with chronic renal failure

PURPOSE: To evaluate the morphologic changes of the corneal endothelium in patients with chronic renal failure. METHODS: Twenty corneas of 20 patients with chronic renal failure undergoing hemodialysis were examined by wide-field specular microscopy. Twenty normal corneas from 20 healthy age-matched subjects were enrolled as control corneas. RESULTS: Despite normal endothelial cell density, the corneal endothelium of patients with chronic renal failure showed marked polymegethism and pleomorphism. CONCLUSION: Polymegethism and pleomorphism of the corneal endothelium are found in patients with chronic renal failure.

Effects of posterior chamber lens implantation on the endothelium of transplanted corneas

AIMSThe morphological changes of the corneal endothelium after posterior chamber lens implantation in the transplanted corneas were investigated.METHODS36 patients underwent extracapsular cataract extraction with posterior chamber lens implantation. Among these,...

Structure and function of the corneal endothelium in diabetes mellitus type I and type II.

To measure and compare corneal endothelial morphologic characteristics and function in subjects with diabetes mellitus types I and II. Forty-nine patients with diabetes mellitus type I and 60 patients with diabetes mellitus type II were recruited from the active practice of the Mayo Clinic, Rochester, Minn. Thirty-one normal subjects, divided by age into two overlapping groups of 20 each, served as controls. Corneal endothelial permeability and corneal autofluorescence were measured by fluorophotometry. Central corneal endothelial photographs were taken with a wide-field specular microscope, which also measured the corneal thickness. Neither the type I nor the type II diabetics differed from their controls in endothelial permeability and endothelial cell density. The type I diabetics had polymegethism, pleomorphism, increased corneal thickness, and increased corneal autofluorescence compared with their controls. Similar measured values were found in the type II diabetics, but they did not differ significantly from those of their age-matched controls. The type II diabetics were older than the type I diabetics, and the older control group showed changes similar to those seen in the diabetics; these changes were presumably associated with aging. The severity of retinopathy was significantly correlated only with corneal autofluorescence. The corneas of patients with type I diabetes mellitus exhibit abnormalities in endothelial cell morphologic characteristics and corneal autofluorescence. The changes resemble those that occur with aging in normal subjects, making them difficult to discern as abnormal in type II diabetics, who are usually older. We found no abnormalities in endothelial permeability in either type I or type II diabetics.

Topical aldose reductase inhibitor for correcting corneal endothelial changes in diabetic patients.

Marked variations in cell size (polymegethism) and shape (pleomorphism) are characteristic of the corneal endothelium in diabetic patients and animals. Wide field specular microscopy was used to evaluate the clinical efficacy of treating the diabetic corneal endothelium with topical instillation of 0.5% aldose reductase inhibitor, CT-112. Morphological variations (polymegethism and pleomorphism) of the endothelium in eight eyes from eight patients receiving CT-112 resolved within 3 months after initiation of treatment. In contrast, no change in endothelial morphology was noted in five eyes from five patients who received placebo. These observations suggest that aldose reductase may be involved in the aetiology of corneal endothelial variations in diabetic patients.

Effects of intracameral injection of tissue plasminogen activator on corneal endothelium and intraocular pressure in dogs

Summary Contact wide-field specular microscopy was performed on eyes of 16 healthy dogs after tissue plasminogen activator at a concentration of 25 μg/100 μl (group 1, n = 8) or 50 μg/100 μl (group 2, n = 8) was injected into 1 anterior chamber of each dog. The contralateral eye served as a nontreated control. Applanation tonometry was used to measure intraocular pressure in both eyes for up to 168 hours. By use of computerized morphometric analysis and pachymetry, changes from baseline values in endothelial cell density, cell morphologic features, and corneal thickness were evaluated at postinjection hours 24, 48, and 168. Significant mean differences in intraocular pressure were not detected between treated eyes of group-1 dogs and those in group 2 at designated times, or between treated and nontreated eyes of dogs in either group. Mean corneal thickness of treated and nontreated eyes was similar in both groups through postinjection hour 168. Changes in mean percentage of endothelial cell sides were observed only in treated eyes of group-2 dogs, with the mean percentage of hexagons at postinjection hour 168 decreasing by 18%, a decrease that was significantly (P < 0.05) greater than the decrease in nontreated eyes. The mean percentage of 6-sided cells in treated eyes of group-2 dogs was significantly (P < 0.05) less than that in treated eyes of group-1 dogs at postinjection hour 168.

Effect of Gas-Permeable Contact Lenses on the Endothelium of Corneal Transplants

The effects of gas-permeable contact lens wear on the density and morphologic characteristics of corneal endothelial cells were prospectively studied by wide-field specular microscopy in 18 eyes that had undergone keratoplasty at least 14 months previously. Daily-wear rigid gas-permeable lenses were fitted to allow as much movement and tear exchange as possible. There was no evidence of morphologic instability of the endothelial cells during the course of the study, with a mean duration of lens wear of 9.8 months. Although there is concern over adverse effects associated with polymethylmethacrylate contact lens wear, this study suggests that the endothelium of a corneal transplant tolerates rigid gas-permeable lens wear without adverse effect during a limited period of follow-up.

The Effects of Denatured Sodium Hyaluronate on the Corneal Endothelium in Cats

AbstractIn an in vivo cat model, wide-field specular microscopy and scanning electron microscopy were used to observe whether heat-denatured sodium hyaluronate causes cell damage to the corneal endothelium. The endothelial cell toxicities of various drugs, including 0.01% benzalkonium chloride, 0.2% chlorhexidine digluconate, and 0.5% lidocaine, mixed with intact or untreated sodium hyaluronate were also investigated. Neither heat-denatured nor intact or untreated sodium hyaluronate alone had any adverse effect on the corneal endothelium in cats, whereas 0.01% benzalkonium chloride and 0.2% chlorhexidine digluconate, mixed with sodium hyaluronate, caused substantial endothelial morphologic changes, which resulted in corneal edema.By comparison, 0.5% lidocaine with sodium hyaluronate was found to have minimal effect on the corneal endothelium in cats. These findings indicate that some chemical contaminant with sodium hyaluronate, not heat-denatured sodium hyaluronate, induces an immediate onset of pseudophakic bullous keratopathy.

Corneal Epithelium in Diabetic Patients

The corneal epithelia of phakic control (n = 13), aphakic control (n = 10), diabetic phakic (n = 10), and diabetic aphakic patients (n = 13) were surveyed under a wide-field specular microscope with a special contact lens. In control patients, the corneal epithelium showed relatively hexagonal cells, which we divided into three groups: bright cells, medium-bright cells, and dark cells. These epithelial cells displayed a smooth-surface configuration. In diabetic phakic patients, the epithelium seemed almost normal, with slightly more irregularity. In diabetic aphakic patients, the epithelium appeared larger. The cells were irregular and easy to distinguish from normal cells. Morphometric analysis showed that the mean cell areas of phakic control, aphakic control, diabetic phakic, and diabetic aphakic patients were 648 +/- 152, 643 +/- 125, 658 +/- 146, and 821 +/- 203 microns 2, respectively. The increase in the average cell area in diabetic aphakic patients was statistically significant. Although no apparent changes of the epithelium could be detected in the diabetic corneal epithelium under biomicroscopy, abnormalities that could be evaluated by specular microscopic images and morphometric analysis were observed.

Effects of intracameral injection of viscoelastic solutions on corneal endothelium in dogs

SUMMARY Contact in vivo wide-field specular microscopy was performed on right eyes of 20 healthy dogs after sodium hyaluronate (1%, n = 5), sodium chondroitin sulfate (4%) and sodium hyaluronate (3%, n = 5), hydroxypropyl methylcellulose (2%, n = 5), or balanced salt solution (control, n = 5) was injected into the anterior chamber. Using computerized morphometric analysis and pachymetry, changes in endothelial cell density, cell morphologic features, and corneal thickness from baseline values were evaluated at postinjection hour (pih) 72 and pih 168. Changes were not seen in endothelial cell density or cell morphologic features in any treated eye. The mean corneal thickness of all treated eyes at pih 72 increased 6%, significantly greater than that of the nontreated eyes (P = 0.03). Mean corneal thickness of treated and nontreated eyes was similar at baseline and pih 168 in all treatment groups.

Protective Effect of the Anterior Lens Capsule during Extracapsular Cataract Extraction: Part II. Preliminary Results of Clinical Study

In a clinical study, 61 patients underwent phacoemulsification with subsequent intraocular lens (IOL) implantation. An intercapsular capsulotomy with phacoemulsification was performed on 23 patients and a can opener capsulotomy with phacoemulsification on 38 patients. Wide-field specular microscopy was performed preoperatively and postoperatively on all of the eyes in the study. The cell counts were not divulged to the surgeon before or at the time of surgery. Average endothelial cell loss for the intercapsular technique was 3.9%; with the can opener technique, the average cell loss was 10.1%. This difference was statistically significant (P < 0.01). Positive correlations between endothelial cell loss relating to ultrasound time and/or the hardness of the cataract existed for the can opener group only (P < 0.01). These results demonstrate that a protective effect is provided by the presence of the anterior lens capsule during lens substance removal.

The Effect of Hard Contact Lens Wear on the Keratoconic Corneal Endothelium After Penetrating Keratoplasty

Using wide-field specular microscopy, we evaluated the morphologic features of the corneal endothelium in 14 eyes of 14 patients with keratoconus who had worn a polymethylmethacrylate hard contact lens for at least four years (mean, 71 months) after penetrating keratoplasty. Comparisons were made with 14 non-lens-wearing eyes of 14 patients with keratoconus matched for age, length of postoperative period, and other variables. Examination of the endothelium of the contact lens wearers showed a significantly higher coefficient of variation in cell size (polymegethism) and a marked decrease in the percentage of hexagonal cells (pleomorphism), with a significant decrease in cell density (P less than .05). The cause of these morphologic abnormalities remains unclear but may be related to lens-induced hypoxic stress to the corneal endothelium causing ongoing cell loss.

The Correction of Aphakia in Infants with Hydrogel Extended-wear Contact Lenses Corneal Studies

Although hydrogel extended-wear contact lenses (EWCLs) have been used extensively in the correction of aphakia in neonates, little is known about the effects of these lenses on infant corneas. Recent studies have demonstrated that long-term contact lenses can induce endothelial morphometric changes, including an increased coefficient of variation (CV) of mean endothelial cell area. Using wide-field specular microscopy, the authors studied 11 eyes of 10 patients, 1 to 3 years of age who, after lensectomy for congenital cataracts, wore EWCLs for the correction of aphakia. Except for two corneas in which increased pachometric readings and CV developed after repeated episodes of lens loss and inflammation, the EWCL were well tolerated and associated with few complications in this study.

Corneal endothelial changes associated with aphakic extended contact lens wear.

Wide-field specular microscopy and computer-assisted morphometry were performed on 20 aphakic eyes in 16 patients who had worn a high-water content hydrogel contact lens on an extended wear basis for at least 12 months (mean, 28 months) following intracapsular cataract extraction. When compared with 22 non-lens-wearing aphakic eyes matched for age and postoperative period, the endothelial cell changes in the above patients included a significantly higher coefficient of variation in cell area (polymegathism) and a marked reduction in the frequency of hexagonal cells (pleomorphism) without any significant difference in cell density. Although the clinical significance of these endothelial changes remains unclear, they may indicate hypoxic stress to the corneal endothelium or be an early sign of ongoing cell loss.

Specular microscopy of vertebrate corneal endothelium: a comparative study

Central corneal endothelia in a variety of lower- and higher vertebrate animals were photographed with a widefield specular microscope and analysed with either fixed-frame or computer-assisted morphometric analysis. The endothelium of the dogfish shark, an elasmobranch, contained 2300 cells mm-2 and demonstrated a very delicate irregular 'reversal pattern'. The goldfish, a teleost, had 432 cells mm-2 and displayed a jigsaw-puzzle-like pattern. The bullfrog, an amphibian, and the gecko, a reptile, had 550- and 481 cells mm-2, respectively, and a relatively uniform polygonal endothelial pattern similar to that observed in mammals. The goose, a bird, had a cell density of 2410 cells mm-2 with a uniform hexagonal pattern (79%) which was similar to mammalian (rat, 58-76%; rabbit, 71%; dog, 78%; human, 61-75%) hexagonal patterns. The findings on the endothelial appearance in these vertebrate animals suggest that a correlation exists between endothelial morphology, vertebrate phylogeny and their respective functional and structural capacity.