Evaluation of Possible Error Sources in Corneal Endothelial Morphometry With a Semiautomated Noncontact Specular Microscope

Discrepancies in Endothelial Cell Density Values of Human Donor Corneas Resulting From Comparison Between Specular Microscopes and Endothelial Analysis Methods

PURPOSE

The methods for specular microscopy evaluation across eye banks differ, which may result in variability in endothelial cell density (ECD) values that influence the surgeon's decision about donor tissue. A comparison of instruments and analysis methods is conducted in this study.

METHODS

Specular images were captured from 97 donor corneas using both HAI and Konan specular microscopes. A single best quality image of each cornea from each instrument was graded using the respective inherent software and analysis method (HAI: variable frame method; Konan: center method). All raw specular images were standardized for dimensions and regraded in the CellChek system in a blinded fashion. The grading variances and paired t test were performed between instruments in both inherent and standardized analyses. Correlation and Bland-Altman analyses between instruments were also performed.

RESULTS

Using the software inherent within HAI and Konan, the mean ECD readings for the 97 corneas were 2764 ± 583 and 2605 ± 517 cell/mm (P = 0.045), respectively, with a variance of 8.05% (range 0.26%-27.2%). HAI resulted in a higher ECD value in 79 corneas (81.4%). In CellChek software analysis, the mean ECD readings did not differ (2609 ± 514 and 2496 ± 507 cells/mm, respectively, P = 0.127), with a variance of 5.6% (range 0.24%-19.8%).

CONCLUSIONS

There is a slight statistically significant mean difference between the ECD values obtained from the 2 specular microscopes, which is negated by standardization to a single analysis method. Eye banks and surgeons should use caution in making decisions based only on very small differences in ECD between otherwise equivalent corneal donor tissues.

Representation of Central Endothelial Cell Density by Analysis of Single Best Specular Microscopy Image Regardless of Cell Size Variance

Purpose

The purpose of this study is to evaluate whether a single best image can represent central endothelial cell density (ECD) in corneas of differing cell size coefficient of variance (CV).

Methods

Four hundred one healthy eyes but with variant CV values were enrolled. For each eye, three nonoverlapping central cornea endothelium images were obtained with Konan NSP-9900 specular microscope. ECD and CV were evaluated by two independent graders using the well-established Center method. Only corneas with high image quality rating (IQR) and ECD >800 cell/mm² by both graders were included in the study. The study sample was stratified into five CV levels (CV ≤ 35; ≥36; ≥38; ≥40; and ≥45). In each CV level, the ECD agreement, ECD variance, and the correlation between the ECD variation and CV values were analyzed. In addition, the ECD intragrader reproducibility and interframe differences were also analyzed for all levels except CV ≤ 35.

Results

The study sample includes a total of 278 eyes. High ECD agreement for the two independent graders (intraclass correlation coefficient [ICC] > 0.99), high ECD intragrader reproducibility (ICC > 0.95), low ECD variance (2.0% ± 1.6%, overall), no correlation between the ECD variation and the CV value (P > 0.05), and no significant ECD difference among frames (P > 0.05) was found in any studied CV levels.

Conclusions

CV does not appear to be associated with ECD variance in the central cornea.

Translational Relevance

This finding highlights that in healthy corneas but with high CV values, ECD can be reliably analyzed using one single image of best quality.

Comparison of Noncontact Specular and Confocal Microscopy for Evaluation of Corneal Endothelium

PURPOSE

To compare endothelial cell analysis obtained by noncontact specular and confocal microscopy, using the Konan NSP-9900 and Nidek ConfoScan4 systems, respectively.

METHODS

Three groups including 70 healthy eyes, 49 eyes with Fuchs endothelial corneal dystrophy (FECD), and 78 eyes with glaucoma were examined with both the Konan NSP-9900 specular microscope and the Nidek ConfocScan4 confocal microscope. Certified graders at the Doheny Image Reading Center compared corneal endothelial images from both instruments side by side to assess image quality. Endothelial cell density (ECD) measurements were calculated and compared using three different modalities: (1) each instrument's fully automated analysis; (2) each instrument's semiautomatic analysis with grader input; and (3) manual grading methods by certified grader.

RESULTS

All normal eyes yielded gradable endothelial images, and most but not all glaucomatous eyes yielded images with high enough image quality to allow grading. In addition, in corneas with severe FECD, poor image quality precluded ECD grading by specular microscopy in 20 eyes (40.8%) but in only 4 (8.2%) confocal images from the same eyes. For the gradable images, the ECD values obtained using the manual grading method from either device were comparable with no statistically significant difference (P>0.05) between specular and confocal devices. Machine-generated ECD values were significantly different from manual results, measuring greater in all cases with specular microscopy. Machine-generated ECD values from confocal microscopy also differed significantly from manual determinations, but not in a consistent direction. Semiautomatic methods for both instruments obtained clinically acceptable ECD values.

CONCLUSIONS

Automatic machine-generated ECD measurements differed significantly from manual assessments of corneal endothelium by both specular and confocal microscopy, suggesting that automated results should be used with caution. But ECD values derived manually were comparable between the two devices in both normal and glaucomatous eyes, suggesting that manually graded images from the two instruments can be used interchangeably for reliable ECD measurements. Because of a higher proportion of gradable images, confocal microscopy may be superior to specular microscopy for ECD measurements in FECD.

Riding the cell jamming boundary: Geometry, topology, and phase of human corneal endothelium

It is important to assess the viability of eye-banked corneas prior to transplantation due to inherent senescence and known loss of endothelial cells during surgical manipulation. Corneal endothelial cells have a complex basal and paracellular shape making them challenging to accurately measure, particularly in oedematous ex vivo tissue. This study used calibrated centroidal Voronoi Diagrams to segment cells in images of these human corneas, in order to characterize endothelial geometry, topology, and phase. Hexagonal cells dominated the endothelia, with most comprised of five different pleomorphs exhibiting self-similar topological coarsening through most of the endothelial cell density range. There was a linear relationship between cell size and shape, though cells with greater than six sides were present in larger proportions than cells with less. Hexagonal cell regularity was stable and largely independent of density. Cell and tissue phase was also examined, using the cell shape index relative to the recently discovered 'cell jamming' phase transition boundary. Images showed fluid endothelia with a range of shape indices spanning the boundary, independent of density but dependent on hexagonal regularity. The cells showed a bimodal distribution centred at the boundary, with the largest proportion of cells on the fluid side. A shoulder at the boundary suggested phase switching via shape transformation across the energy barrier, with cells either side having distinctly different size and shape characteristics. Regular hexagonal cells were closest to the boundary. This study showed the corneal endothelium acts as a glassy viscous foam characterized by well-established physical laws. Endothelial cell death transiently and locally increases cell fluidity, which is subsequently arrested by jamming of the pleomorphically diverse cell collective, via rearrangement and shape change of a small proportion of cells, which become locked in place by their neighbours thereby maintaining structural equilibrium with little energy expenditure.

On the regional variability of averaged cell area estimates for the human corneal endothelium in relation to the extent of polymegethism

Purpose

To assess variability in the coefficient of variation (COV) in cell area estimates when using different numbers of cells for endothelial morphometry.

Methods

Using non-contact specular microscopy images of the corneal endothelium, 4 sets of 20 cases were selected that included 200 cells and had overall (global) COV values of less than 30 (group 1), 31–40 (group 2), 41–50 (group 3) and over 50% (group 4). Subjects could be normal, or had ophthalmic disease (such as diabetes), a history of rigid or soft contact lens wear or were assessed after cataract surgery. A step-wise analysis was undertaken, 20 cells at a time, of the variability in cell area estimates when using different numbers of cells for the calculations.

Results

Variability in the average cell area values was higher if only 20–60 cells were used in the calculations and then tended to decrease. The standard deviation values on these average cell area values and the calculated COV showed the same overall trends and were more than twice as large for endothelia with marked polymegethism. Using more than 100 cells/image in markedly polymegethous endothelia only increased the variability in the calculations.

Conclusions

These analyses indicate that substantial region variability in cell area values can be expected in polymegethous endothelia. The analysis further confirm that using only small numbers of cells (e.g. less than 50/image) in such cases is likely to yield far less reliable estimates of COV.

Further Analysis of the Predictability of Corneal Endothelial Cell Density Estimates When Polymegethism Is Present

PURPOSE

To assess variability in endothelial cell density (ECD) estimates when polymegethism (variance in cell areas) is present.

METHODS

Using noncontact specular microscope images of the corneal endothelium, 4 sets of 20 cases were selected, which included 200 cells and had coefficient of variation values of less than 30% (group 1), 31%-40% (group 2), 41%-50% (group 3), and over 50% (group 4). A stepwise analysis was undertaken, 20 cells at a time, of the ECD estimates when using different numbers of cells for the calculations.

RESULTS

The net differences in ECD estimates when comparing sets of 20 cells with 200 cells were 5.0% ± 3.9%, 8.1% ± 7.3%, 11.3% ± 9.4%, and 14.5% ± 12.4% for groups 1 to 4, respectively. For measures on 100 cells per image, the predicted variances in ECD values were 5.6%, 8.8%, 11.1%, and 13.7% for the 4 groups.

CONCLUSIONS

Higher values of corneal endothelial polymegethism result in predictable increases in the variability (uncertainty) in ECD estimates, thus reducing the "accuracy" of ECD values. There is no obvious utility in assessing more than 100 cells in such endothelia.

Effects of Laser Peripheral Iridotomy on Corneal Endothelial Cell Density and Cell Morphology in Primary Angle Closure Suspect Subjects

Purpose:

To evaluate the effects of prophylactic laser peripheral iridotomy on corneal endothelial cell density and cell morphology in subjects with primary angle closure suspect (PACS) within a one-year follow-up period.

Methods:

In this quasi-experimental prospective study, from June 2012 to November 2013, thirty-five PACS eyes underwent laser peripheral iridotomy at clinics affiliated to Shiraz University of Medical Sciences, Shiraz, Iran. After obtaining informed consent, specular microscopy was performed at baseline and at 3-month, 6-month and 12-month follow-up visits. Central, nasal and temporal endothelial cell counts and cell morphology were evaluated via non-contact specular microscopy.

Results:

The mean subject age was 53.4 ± 7.9 years, and the majority of subjects were women (88.2%). The mean central corneal endothelial cell count prior to laser peripheral iridotomy was 2528 ± 119.2, and this value changed to 2470 ± 175.9, 2425 ± 150.6, and 2407 ± 69.02 at the 3-month, 6-month, and 12-month follow-up visits, respectively; these differences did not reach statistical significance. Additionally, the changes in the number of cells, the hexagonality of cells, and the coefficient of variation (CV) in the central, nasal, and temporal areas were not significant.

Conclusion:

In PACS eyes, we did not find a decline in corneal endothelial cell density or a change in cell morphological characteristics, including cell hexagonality and CV, in the central, nasal, and temporal regions of the cornea in any of our subjects over a one-year follow-up period.

Connection of histological corneal endothelial cell count with endothelial cell density before penetrating keratoplasty

PURPOSE

Corneal endothelial cell density and the integrity of the monolayer are essential for maintenance of a clear cornea. In 1992, Williams et al. introduced a method to estimate the endothelial cell density in histopathologic examination. It would enable an evaluation of the corneal host endothelium, even if preoperative measurement was not possible. The goal of this study is to evaluate the accuracy of the Williams equation in corneal buttons obtained from penetrating keratoplasties.

METHODS

High power field (HPF) photographs and histological endothelial cell counts were made from the corneal endothelial cells of each corneal histopathological cross-section. We then compared the calculated endothelial cell density using the Williams equation with the preoperative measured endothelial cell density. A bivariate regression analysis of the histological HPF cell counts and the preoperative endothelial cell density count was also performed.

RESULTS

The equation of Williams et al. overestimates the endothelial density in all of our patients. Linear regression showed a strong relation between the central histological HPF count and the preoperative endothelial cell density. The regression formula for the endothelial cell density is 59.66 + (272.447 × HPF count); p < 0.001, R ² = 0.901.

CONCLUSION

This study confirms the relation between the corneal endothelial cell density, measured with specular microscopy, and the histopathological endothelial cell count in a HPF. However, the equation of Williams et al. provides an overestimation of the endothelial cell density. To proper utilize the histopathological endothelial cell count, a calibration of the equation coefficients in the local setting is necessary to prevent systematic errors.