Optic disc cup slope and visual field indices in normal, ocular hypertensive and early glaucomatous eyes

Factors Associated with Large Cup-to-Disc Ratio and Blindness in the Primary Open-Angle African American Glaucoma Genetics (POAAGG) Study

BACKGROUND/AIMS

Primary open-angle glaucoma (POAG) disproportionately affects individuals of African ancestry. In these patients' eyes, a large cup-to-disc ratio (LCDR > 0.90) suggests greater retinal ganglion cell loss, though these patients often display varied visual ability. This study investigated the prevalence and risk factors associated with LCDR in African ancestry individuals with POAG and explored the differences between blind (>20/200) and not blind (≤20/200) LCDR eyes.

METHODS

A case-control methodology was used to investigate the demographic, optic disc, and genetic risk factors of subjects in the Primary Open-Angle African American Glaucoma Genetics Study. Risk factors were analyzed using univariable and multivariable logistic regression models with inter-eye correlation adjusted using generalized estimating equations.

RESULTS

Out of 5605 eyes with POAG, 1440 eyes (25.7%) had LCDR. In the multivariable analysis, LCDR was associated with previous glaucoma surgery (OR = 1.72), increased intraocular pressure (OR = 1.04), decreased mean deviation (OR = 1.08), increased pattern standard deviation (OR = 1.06), thinner retinal nerve fiber layer (OR = 1.05), nasalization of vessels (OR = 2.67), bayonetting of vessels (OR = 1.98), visible pores in the lamina cribrosa (OR = 1.68), and a bean-shaped cup (OR = 2.11). Of LCDR eyes, 30.1% were classified as blind (≤20/200). In the multivariable analysis, the statistically significant risk factors of blindness in LCDR eyes were previous glaucoma surgery (OR = 1.72), increased intraocular pressure (OR = 1.05), decreased mean deviation (OR = 1.04), and decreased pattern standard deviation (OR = 0.90).

CONCLUSIONS

These findings underscore the importance of close monitoring of intraocular pressure and visual function in African ancestry POAG patients, particularly those with LCDR, to preserve visual function.

Optic nerve head slope-based quantitative parameters for identifying open-angle glaucoma on SPECTRALIS OCT images

PURPOSE

To investigate monitoring slope-based features of the optic nerve head (ONH) cup as open-angle glaucoma (OAG) occurs.

METHOD

A dataset of 46 retrospective OCT cases was acquired from the SPECTRALIS Heidelberg Engineering OCT device. A set of five parameters, which are based on the ONH cup-incline, are measured on the OAG and normal subjects in the dataset. Then, three new ONH cup-shape indices were deduced. The ONH cup-incline parameters and ONH cup-shape indices are analyzed to estimate their clinical value.

RESULTS

The statistical difference between measurements on normal and glaucoma eyes was remarkably significant for all of the analyzed parameters and indices (p value < 0.001).

CONCLUSIONS

The geometric shape of the ONH cup can be transferred to numerical parameters and indices. The proposed ONH cup-incline parameters and ONH cup-shape indices have shown suggestive clinical value to identify the development of OAG. As OAG appears, the top ONH cup-incline parameters decrease while the bottom ONH cup-incline parameters increase. The ONH cup-shape indices suggest capability to discriminate OAG from normal eyes.

Glaucoma detection with matrix and standard achromatic perimetry

BACKGROUND

Matrix perimetry is a new iteration of frequency-doubling technology (FDT) which uses a smaller target size in the standard achromatic perimetry presentation pattern.

AIM

To compare the performance of matrix and Swedish interactive thresholding algorithm (SITA) perimetry in detecting glaucoma diagnosed by structural assessment.

DESIGN

Prospective cross-sectional study.

METHODS

76 eyes from 15 healthy subjects and 61 consecutive glaucoma suspects and patients with glaucoma were included. All patients underwent optic nerve head (ONH) photography, SITA and matrix perimetries, and optical coherence tomography (OCT) within a 6-month period. Glaucoma diagnosis was established by either glaucomatous optic neuropathy or OCT by assessing retinal nerve fibre layer (RNFL) thickness. Mean deviation (MD), pattern standard deviation (PSD), glaucoma hemifield test and cluster of abnormal testing locations were recorded from matrix and SITA perimetries.

RESULTS

Similar correlations were observed with matrix and SITA perimetry MD and PSD with either cup-to-disc ratio or OCT mean RNFL. The area under the receiver operating characteristic (AROC) curves of MD and PSD for discriminating between healthy and glaucomatous eyes ranged from 0.69 to 0.81 for matrix perimetry and from 0.75 to 0.77 for SITA perimetry. There were no significant differences among any corresponding matrix and SITA perimetry AROCs.

CONCLUSIONS

Matrix and SITA perimetries had similar capabilities for distinguishing between healthy and glaucomatous eyes regardless of whether the diagnosis was established by ONH or OCT-RNFL assessment.

Improving glaucoma diagnosis by the combination of perimetry and HRT measurements

PURPOSE

The aim of this study was to determine, whether the combination of morphologic data of the optic nerve head and visual field (VF) data would improve diagnosis of glaucoma, on the basis of the measurements alone.

PATIENTS AND METHODS

Eighty-eight perimetric glaucomatous and 88 normal optic discs from the Erlangen Glaucoma Registry were matched for age. All normals and patients were examined in a standardized manner (Slitlamp biomicroscopy, gonioscopy, 24 h-applanation tonometry, automated VF testing, 15-degree optic disc stereographs, and Heidelberg Retina Tomograph (HRT)-scanning of the optic disc). The HRT variables were calculated in 4 optic disc sectors. All variables were calculated with the software's standard reference plane. To gain the same allocation of sectors as provided by the HRT software, the VF responses were averaged within 4 sectors. Classification results of these VF responses were compared with the summarized results within 4 sectors. Six different combinations of morphologic and VF data were used to assess their suitability to diagnose the disease. HRT measurements, and the standard output of the Octopus (HRT/PERI1), HRT measurements and the summarized sectors and their standard deviations (HRT/PERI2), HRT measurements, standard output of the octopus and the summarized sectors and their standard deviations (HRT/PERI1/PERI2), standard output of the Octopus (PERI1), summarized sectors of the Octopus and their standard deviations (PERI2) and HRT measurements. To assess the diagnostic value of the different data sets machine learning classifiers, stabilized linear discriminant analysis, classification trees, bagging, and double-bagging were applied.

RESULTS

Combination of morphologic and VF data improved the automated classification rules. The accuracy to diagnose glaucoma just by VF and HRT indices was maximized for double-bagging using both diagnostic tools. An estimated misclassification probability of less than 0.07 could be achieved for the primary open angle glaucoma patients combining HRT and VF sectors by double bagging. So highest sensitivity was 95% and specificity 91%, achieved by double-bagging and combination of HRT, PERI1, and PERI2.

CONCLUSIONS

The combination of optic disc measurements and VF data could not only improve glaucoma diagnosis in future, but could also help to find an objective way to diagnose glaucomatous optic atrophy. The limitation of the topographic relationship between structure and function is the individual variability of the optic disc morphology and the subjective variability of VF testing.

Agreement and repeatability for standard automated perimetry and confocal scanning laser ophthalmoscopy in the diagnostic innovations in glaucoma study

PURPOSE

To assess agreement between structural and functional testing in classifying eyes as normal or abnormal and their repeatability on two consecutive visits by means of standard automated perimetry (SAP) and confocal scanning laser ophthalmoscopy.

DESIGN

Analysis of selected data obtained from a prospective longitudinal observational cohort study, the Diagnostic Innovations in Glaucoma Study.

METHODS

One hundred fifty-one participants with a SAP and a Heidelberg Retina Tomograph (HRT) test within a three-month window at two visits within 15 months were included. Eyes were classified by SAP and HRT at each visit. Agreement and repeatability were assessed by kappa statistics.

RESULTS

At visit 1, 33 (22%) of 151 eyes had only SAP defects, 11 (7%) eyes had only HRT defects, and 35 (23%) eyes had both SAP and HRT defects. Seventy-two eyes (48%) were classified as normal by both tests. Similar results were obtained for visit 2. The agreement between SAP and HRT in classifying eyes was 70% (kappa = 0.393) at visit one and 68% (kappa = 0.363) at visit 2. Repeatability of classification by SAP alone, HRT alone, and both SAP and HRT between visits 1 and 2 was 84% (kappa = 0.693), 89% (kappa = 0.752), and 90% (kappa = 0.723), respectively.

CONCLUSIONS

Agreement between SAP and HRT was only fair at both visits. Repeatability across visits was substantial for SAP alone, HRT alone, and for the combination of SAP and HRT. These results suggest that detection of particular features of glaucomatous damage depends on the technique used. Structural and functional tests appear complementary, and both should be used for early detection of glaucoma.

Intraobserver and Interobserver Reliability Indices for Drawing Scanning Laser Ophthalmoscope Optic Disc Contour Lines With and Without the Aid of Optic Disc Photographs

PURPOSE

To investigate the effects on topographic optic disc analysis of defining regions of interest (ROIs) by drawing contour lines with and without photographic aid.

PATIENTS AND METHODS

Forty-five patients had optic disc imaging by stereoscopic photography and confocal scanning laser ophthalmoscopy using the Heidelberg Retinal Tomograph (HRT). Two experienced observers defined ROIs with a stereoscopic optic disc photograph, with a non-stereoscopic photograph and without any photographic guide. Intraclass coefficients and 95% tolerance limits for change were calculated for each HRT optic disc parameter for the following situations: (1) Intraobserver reliability for ROIs defined with non-stereoscopic photographs and no photograph compared against ROIs defined using stereoscopic photographs; (2) interobserver reliability for ROIs defined without the aid of a photograph, with stereoscopic photographs and with non-stereoscopic photographs; and (3) intraobserver reliability for ROIs defined twice for 23 patients using non-stereoscopic photographs.

RESULTS

Intraclass correlation coefficients ranged from 0.63 to 1.00 (ie, 'substantial' to 'perfect' agreement). The 95% tolerance limits for change ranged from 3% to 34%. In general, intraobserver reliability indices were higher than interobserver reliability indices but no method of ROI definition appeared superior. The least reliable measurements were for rim area, rim volume, and disc area.

CONCLUSIONS

Non-stereoscopic optic disc photographs are an acceptable alternative to stereoscopic photographs when defining ROIs for HRT analysis. Observers with impaired stereopsis may be able to define reliable ROIs. For HRT algorithms designed to determine a 'diagnosis' of glaucoma, rim area, rim volume, and disc area data should be used with caution.

Estimating the clinical usefulness of optic disc biometry for detecting glaucomatous change over time

PURPOSE

First, to determine the absolute measurement precision of scanning laser ophthalmoscopy (SLO) parameters, by expressing them as 95% limits of agreement (LA(95%)). Second, to propose a method for mathematically estimating the clinical ability of a parameter to monitor disease progression, expressed as the Discriminating Capacity Index (DCI).

METHODS

We measured the optic disc of 14 healthy volunteers and 14 glaucoma patients. LA(95%)-values were calculated from the average standard deviation of three measurements on the same day for repeatability, and three measurements on separate days within a 6-week period for reproducibility. We then calculated the DCI by dividing the measurement range by its LA(95%) in healthy subjects and glaucoma patients separately. Thus, the DCI takes into account both the dynamic range of disease progression and the extent of measurement variance, providing an index of the possible clinical usefulness of a parameter. As the DCI is dimensionless it allows comparison across various parameters and across technologies.

RESULTS

In the glaucoma group, the SLO parameters with the highest DCIs were 'volume below' (DCI, 9.38) and 'mean contour depth' (DCI, 8.02). In the healthy group, 'Neuroretinal rim area' had the highest index (DCI, 2.15).

CONCLUSION

SLO optic disc biometry is uniformly reproducible and may prove a clinically useful method for glaucoma follow-up, due to the high DCI found for several parameters. The capacity to detect conversion from health to glaucoma is less pronounced, possibly due to a larger biological variability found in healthy volunteers.

Correlation between optic disc topography and retinal nerve fibre layer measurements in glaucoma

PURPOSE

To test the correlation between optic disc topography (as measured by confocal scanning laser ophthalmoscopy [CSLO]) and retinal nerve fibre layer (RNFL) measurements (as measured by scanning laser polarimetry [SLP]) in glaucoma.

MATERIAL AND METHODS

A total of 112 patients with primary open-angle glaucoma and 88 normal individuals were enrolled in the study. All individuals underwent a thorough ophthalmic evaluation, a 24-2 full threshold Humphrey visual field, SLP with the GDx (Version 1.0.12) and CSLO with the TOPSS. Receiver operator characteristic curves were created for each individual CSLO and SLP parameter. Linear correlations between the four best parameters from each device were calculated. Finally, linear correlations between the same variables, controlled for the severity of visual field damage, were calculated.

RESULTS

The best individual parameters in the diagnosis of glaucoma for each device were cup area, vertical cup : disc ratio, cup : disc area ratio and average cup depth for CSLO, and the Number, maximum modulation, ellipse modulation and superior nasal for SLP. Moderate to strong correlations were observed in 62% of the pairs. However, only 6% and 12% of the correlations remained moderate when we controlled for the visual field mean deviation and correct pattern standard deviation, respectively.

CONCLUSIONS

The correlation between optic disc topography and RNFL measurements in glaucoma patients is moderate and highly dependent on the level of visual field loss.

Application of Discriminant, Classification Tree and Neural Network Analysis to Differentiate between Potential Glaucoma Suspects with and without Visual Field Defects

Purpose: This study has two objectives. The first one is to investigate the question whether it is possible to discriminate between eyes with and without a glaucomateous visual field defect based on standard ophthalmologic examinations as well as optic nerve head topographic parameters. The second objective raises the question of the ability of several suggested statistical models to generalize their results to new, previously unseen patients.

Methods: To investigate the above addressed question: (a) independent, two-sidedt-tests, (b) a linear discriminant analysis with a forward stepwise variable selection algorithm, (c) four classification tree analyses and (d) three different neural network models with a forward, backward and a genetic variable selection algorithm were applied to 1020 subjects with a normal visual field and 110 subjects with a glaucomateous visual field defect. The Humphrey Visual Field Analyzer was used to test the visual fields and the TopSS®Scanning Laser Tomograph measured the optic nerve topography. A 10-fold cross-validation method was used for the models (b), (c) and (d) to compute approximative 95% confidence intervals for the specificity and sensitivity rates.

A literature study of 18 studies dealt with the question whether/how the generalization error was controlled (control of sample bias, cross-validation procedures, training net size for valid generalization). It was also looked up whether point estimators or 95% confidence intervals were used to report specificity and sensitivity rates.

Results: (a) Only few differences of the means could be found between both groups, including age, existing eye diseases, best corrected visual acuity and visual field parameters. The linear discriminant analysis (b), the classification tree analyses (c) and the neural networks (d) ended up with high specificity rates, but low sensitivity rates.

The literature study showed that three models did not apply a cross-validation procedure to report their results. Two models used a test sample cross-validation and thirteen used a v-fold cross-validation method. Although most authors reported confidence intervals for the area under the ROC, no author reported confidence intervals for the true, but unknown sensitivity and specificity rates.

Conclusions: (a) The results of this study suggest that the combination of standard ophthalmologic eye parameters and optic nerve head topographic parameters of the TopSS®instrument are not sufficient to discriminate properly among normal eyes and eyes with a glaucomateous visual field defect. (b) We follow important suggestions given in statistical learning theory for proper generalization and suggest to apply these methods to the given models or to models in future. At least three conditions should be met: (1) confidence intervals instead of point estimators to assess the classification performance of a model (control of test sample bias); (2) sensitivity and specificity rates should be estimated instead of reporting a point estimator for the area under the ROC and (3) the generalization error should be reported both in a training and a test sample and methods should be applied to select an appropriate training sample size for valid generalization.