Search

Purpose: To assess the intraday repeatability of macular architecture measurements in glaucomatous and non-glaucomatous patients using spectral-domain optical coherence tomography (SD-OCT) and to evaluate the independence from intraindividual intraocular pressure (IOP) fluctuations., Methods: In this single-center, time-point comparison study, 88 eyes with glaucoma, 53 eyes with ocular hypertension (OHT), and 253 healthy eyes underwent two standardized SD-OCT and intraocular pressure (IOP) measurements on the same day with a 5-h time gap. Bland-Altman plots, intraclass correlation coefficients (ICC), and random-effects model were used to analyze repeatability of entire retinal thickness, retinal nerve fiber layer, ganglion cell layer, inner plexiform layer, and inner nuclear layer measurements., Results: Intraday measurements were highly reproducible in all 3 groups. ICC were greater than 0.90, respectively. The pairwise comparisons of morphometric parameters showed a statistically significant difference (P < 0.001, respectively) between groups (glaucoma vs. control, glaucoma vs. OHT) and a significant influence of time points. No correlation was found between IOP fluctuations and morphometric parameters (P > 0.05, respectively), except for a weak positive correlation with GCL (rho = 0.109, P = 0.031)., Conclusions: The evaluation of macular morphometric parameters of SD-OCT showed a high intraday repeatability and an excellent degree of agreement in glaucoma, ocular hypertension, and healthy groups. The fixed effects of time points were statistically significant. Except for a weak positive correlation of ganglion cell layer, variability did not appear to be affected by intraday IOP changes. Additional research is required to fully understand the impact of IOP fluctuations on macular morphometric parameters, considering the small observed IOP changes., (© 2024. The Author(s).)

Purpose: To assess and compare the risk for development of cystoid macula edema (CME) after glaucoma drainage device (GDD) implantation versus conventional trabeculectomy with mitomycin (trab) for glaucoma., Methods: Retrospective review of consecutive patients receiving trab or GDD implantation between 2016 and 2018. Inclusion criteria were availability of pre- and postoperative spectral domain optical coherence tomography (SD-OCT) of the macula. SD-OCT images were evaluated for presence of CME qualitatively, central subfield thickness (CST) and macular volume (MV)., Results: 73 eyes could be included, 42 received trab and 31 GDD surgery. Eyes receiving trab on average had 0.8 ± 0.8 previous intraocular operations, while eyes with GDD implantation had 3.1 ± 1.9 (p < 0.001). Occurrence of postoperative CME was significantly more frequent after GDD implantation (6 out of 31 (19.4%)) than after trab (2 out of 42 eyes = 4.8%), (p = 0.049). Mean preoperative CST as well as MV was comparable in both groups (CST before trab: 282.7 ± 23.0 µm, CST before GDD 284.2 ± 27.3 µm, p = 0.287; MV before trab: 7.8 ± 1.1 mm 3 , MV before GDD: 8.0 ± 0.8mm 3 , p = 0.305). Mean postoperative CST and MV were significantly higher after GDD (CST 338.5 ± 129.3 µm, MV 8.8 ± 2.6 mm 3 ) than after trabeculectomy (CST 290.6 ± 60.2 µm, p = 0.038; MV 7.8 ± 1.2mm 3 , p = 0.039)., Conclusions: In real-life conditions, GDD surgery seems to be associated with a higher risk to develop CME when compared to conventional trabeculectomy. This information may be helpful for glaucoma surgeons to advise the patients on postoperative risks of surgery., (© 2024. The Author(s).)

Purpose: To compare peripapillary vessel density (VD) measured by spectral domain optical coherence tomography angiography (SD-OCT-A) with morphometric parameters assessing the neuroretinal rim based on Bruch's membrane opening (BMO) by spectral domain optical coherence tomography (SD-OCT) in glaucoma., Methods: In this non-interventional cohort study, 50 eyes of 25 consecutively enrolled patients with diagnosis of glaucoma underwent SD-OCT and SD-OCT-A imaging of the optic nerve head (ONH). BMO minimum rim width (BMO-MRW) and area (BMO-MRA) as well as peripapillary retinal nerve fiber layer (RNFL) thickness were compared to peripapillary VD in the RNFL layer around the ONH., Results: Mean BMO-MRW was 221.46 ± 81.5 µm, mean BMO-MRA was 1.05 ± 0.04 mm 2 , mean RNFL thickness was 72.46 ± 23.16 µm, and mean VD was 43.8 ± 11.4%. VD was significantly lower when morphometric parameters had lower thickness values (p < 0.01). Correlation coefficients and their 95%- confidence intervals (95%-CI) with VD were r = 0.53 (95%-CI: 0.21-0.77) for BMO-MRW, r = 0.55 (95%-CI: 0.21-0.77) for BMO-MRA, and r = 0.57 (95%-CI: 0.13-0.73) for RNFL thickness. Intra-individual VD in both eyes correlated with r = 0.72 (p < 0.001), mean VD was comparable (p = 0.6). Eyes with high global RNFL thickness (>90 µm) showed less VD variance (σ 2  = 48.1) compared to eyes with highly reduced RNFL thickness (<65 µm; σ 2  = 82.0). Best corrected visual acuity, perimetric mean defect, and PSD correlated significantly with VD (95%-CI: -0.66 to -0.10, 0.16 to 0.6, and -0.65 to -0.02, respectively)., Conclusions: Peripapillary vessel density measured by SD-OCT angiography correlates significantly with Bruch's membrane opening-based parameters measured by SD-OCT in glaucoma patients.

Purpose: To characterize the two-dimensional parameter Bruch's membrane opening minimum rim area (BMO-MRA) in spectral domain optical coherence tomography (SD-OCT) of the optic nerve head (ONH) compared to minimum rim width (BMO-MRW) and retinal nerve fibre layer (RNFL) thickness in a large patient cohort., Methods: Case-control, cross-sectional study of 705 eyes of 445 participants. A total of 449 eyes with glaucoma, 67 eyes with ocular hypertension and 189 healthy controls, underwent SD-OCT and confocal laser scanning tomography (CSLT), visual field testing and clinical examination. Morphometric ONH parameters, visual field function and diagnostic power were compared. Main outcome measures were SD-OCT-derived BMO-MRA, BMO-MRW, RNFL thickness and CSLT-derived rim area (DM-RA)., Results: Mean ONH area was 2.11 ± 0.57 mm 2 ; mean BMO area was 1.89 ± 0.45 mm 2 . Correlation of mean deviation in visual field to morphometric parameters was ρ = 0.70, (p < 0.001) for RNFL thickness, ρ = 0.68 (p < 0.001) for BMO-MRA, ρ = 0.66 (p < 0.001) for BMO-MRW. These correlations were not significantly different (p > 0.05), while DM-RA correlated significantly worse (ρ = 0.55; p < 0.001). In receiver operating characteristics, the calculated area under the curve (AUC) and sensitivity at 90% specificity to differentiate glaucoma were 0.87% and 70.1% for BMO-MRA, 0.86% and 68.1% for RNFL thickness, 0.84% and 66.0% for BMO-MRW, 0.82% and 51.3% for DM-RA., Conclusions: In a heterogenous clinical cohort of glaucoma patients, all analysed SD-OCT parameters excel DM-RA of CSLT. The two-dimensional parameter BMO-MRA shows comparable levels of diagnostic power to detect glaucoma compared to established parameters BMO-MRW and RNFL thickness. Given higher comparability between ONH sizes, BMO-MRA might become an additional standard tool in SD-OCT imaging for glaucoma., (© 2018 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd.)

PurposeTo compare Bruch's membrane opening (BMO)-based spectral domain optical coherence tomography (SD-OCT) and margin based confocal scanning laser tomography (CSLT) of the optic nerve head (ONH) to visual field function in large optic discs (macrodiscs) and to assess performance for glaucoma detection.MethodsIn a case-control, cross-sectional study, 125 eyes of 125 patients with disc size >2.45 mm 2 , thereof 44 glaucoma and 11 ocular hypertension (OHT) patients and 70 healthy controls underwent SD-OCT and CSLT examination, visual field testing and clinical evaluation. Mean outcome measures BMO-based minimum rim width (BMO-MRW), retinal nerve fiber layer thickness (RNFLT) in SD-OCT, and rim area measured in CSLT were compared and correlated to visual field function.ResultsAll participants had a mean disc area of 2.91±0.38 mm 2 in CSLT and a BMO area of 2.45±0.39 mm 2 (r=0.76;P<0.001). In glaucoma patients, visual field mean deviation was -10.0±6.1 dB. Global BMO-MRW correlated better to visual field function (Spearman's Rho (ρ)=0.71; P<0.001) than RNFLT (ρ=0.52;P<0.001) and CSLT rim area (ρ=0.63; P<0.001). BMO-MRW was significantly decreased with higher visual field loss (P<0.001). In ROC analysis, diagnostic power to differentiate glaucoma patients and healthy controls was highest for BMO-MRW (Area under curve, AUC=0.96; sensitivity at 95% specificity=82%). Rim area in CSLT (AUC=0.91; sensitivity=61.0%; P=0.04) and RNFLT (AUC=0.89; sensitivity=61%; P=0.01) were significantly less powerful.ConclusionsIn macrodiscs, BMO-MRW has the best diagnostic power to discriminate glaucoma patients from normal controls compared to RNFLT and rim area in CSLT. Additionally, BMO-MRW seems to reflect the structure-function relationship better than the other two parameters.

Purpose: To assess the performance of Bruch's membrane opening (BMO)-based spectral domain optical coherence tomography (SD-OCT) of the optic nerve head for glaucoma detection in microdiscs in comparison with confocal scanning laser tomography (CSLT)., Design: Retrospective cohort study., Methods: 82 eyes of 82 patients with disc size <1.63 mm 2 underwent SD-OCT and CSLT measurements, visual field testing and clinical examination. BMO-based minimal rim width (BMO-MRW), retinal nerve fibre layer thickness (RNFLT) in SD-OCT and rim area measured in CSLT were compared and correlated with visual field defects., Results: 51 patients with glaucoma, 11 patients with ocular hypertension (OHT) and 20 healthy controls had a mean disc area of 1.36±0.19 mm 2 in CSLT, and BMO area was 1.45±0.22 mm 2 (r=0.17; p=0.12). In patients with glaucoma, visual field mean defect was -7.5±6.7 dB. Global BMO-MRW correlated better with visual field function (Spearman's r=0.65; p<0.001) than RNFLT (r=0.58; p≤0.001) and CSLT rim area (r=0.47; p=0.004). BMO-MRW significantly deteriorated with progressive visual field loss (p<0.001). In receiver operating characteristic analysis, sensitivity of BMO-MRW was 68.6% at 95% specificity (area under curve (AUC)=0.87), similar to sensitivity of RNFLT (66.4%; AUC=0.81). Performance of CSLT rim area was significantly worse (AUC=0.70, p=0.008). In healthy controls, mean BMO-MRW was 344.3±64.1 µm, mean RNFLT 78.0±11.3 µm and CSLT mean rim area 1.07±0.18 mm 2 ., Conclusions: In small optic discs, BMO-MRW and peripapillary RNFLT (OCT) have similar sensitivity to discriminate patients with glaucoma from normal controls; both exceed CSLT rim area in diagnostic power. In glaucomatous patients, BMO-MRW correlates strongest with visual field function., (Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.)

Purpose: To assess the risk of pseudophakic cystoid macular edema (PCME) following cataract surgery (bimanual phacoemulsification with posterior chamber lens implantation) combined with ab interno glaucoma surgery (trabecular aspiration or ab interno trabeculotomy) compared with solely cataract extraction., Patients and Methods: In total, 360 consecutive eyes of 180 patients with age-related cataract, and thereof 162 with coexistent glaucoma, who underwent sole cataract surgery or in combination with ab interno glaucoma surgery (trabecular aspiration or ab interno trabeculotomy) were compared retrospectively regarding PCME development. Patients with known risk factors of PCME were excluded. Macular examination was performed using spectral-domain optical coherence tomography before and at 4, 8, 12, 16, 24, 36, and 48 weeks after surgery. Main outcome measure was the incidence of PCME., Results: In total, 126 eyes (35%) underwent cataract surgery combined with ab interno glaucoma surgery (96 trabecular aspiration (76.2%), 30 ab interno trabeculotomy (23.8%), and 234 eyes (65%) received cataract surgery alone. PCME overall occurred in 23 eyes (6.4%). Following combined trabecular aspiration 5 eyes (5.2%) developed PCME, following ab interno trabeculotomy 2 eyes (6.7%), and following sole cataract surgery 16 eyes (6.8%), reaching no significant difference (P=0.676). Risk of PCME in fellow eye was significantly increased (P=0.025; odds ratio 7.177; 95% confidence interval, 5.667-8.687)., Conclusions: Additional ab interno glaucoma surgery combined with phacoemulsification does not seem to increase the risk of PCME in uncomplicated eyes. However, in patients with history of PCME a significantly increased risk of PCME in fellow eye should be taken into account.

Background: Random forests are successful classifier ensemble methods consisting of typically 100 to 1000 classification trees. Ensemble pruning techniques reduce the computational cost, especially the memory demand, of random forests by reducing the number of trees without relevant loss of performance or even with increased performance of the sub-ensemble. The application to the problem of an early detection of glaucoma, a severe eye disease with low prevalence, based on topographical measurements of the eye background faces specific challenges., Objectives: We examine the performance of ensemble pruning strategies for glaucoma detection in an unbalanced data situation., Methods: The data set consists of 102 topographical features of the eye background of 254 healthy controls and 55 glaucoma patients. We compare the area under the receiver operating characteristic curve (AUC), and the Brier score on the total data set, in the majority class, and in the minority class of pruned random forest ensembles obtained with strategies based on the prediction accuracy of greedily grown sub-ensembles, the uncertainty weighted accuracy, and the similarity between single trees. To validate the findings and to examine the influence of the prevalence of glaucoma in the data set, we additionally perform a simulation study with lower prevalences of glaucoma., Results: In glaucoma classification all three pruning strategies lead to improved AUC and smaller Brier scores on the total data set with sub-ensembles as small as 30 to 80 trees compared to the classification results obtained with the full ensemble consisting of 1000 trees. In the simulation study, we were able to show that the prevalence of glaucoma is a critical factor and lower prevalence decreases the performance of our pruning strategies., Conclusions: The memory demand for glaucoma classification in an unbalanced data situation based on random forests could effectively be reduced by the application of pruning strategies without loss of performance in a population with increased risk of glaucoma.

Purpose: The aim of this study was to determine the prevalence of glaucomatous optic nerve atrophy among a working population in Germany by secondary evaluation of a study conducted to estimate the prevalence of retinal microangiopathic abnormalities by telemedical examination of the retina., Patients and Methods: From August 2002 until January 2004 the retina and optic nerve head were examined in 19,294 Caucasians using a non-mydriatic fundus camera (Kowa, nonmyd-alpha 45), which produces colour images with 45 degrees. The images of the retina and optic nerve head were evaluated telemedically by glaucoma specialists in respect to optic nerve pathologies and microangiopathic abnormalities by a standardised procedure. Glaucomatous optic nerve atrophy was diagnosed when specific glaucomatous morphological alterations of the optic nerve head were present. A complete medical history including reported elevated intraocular pressure (IOP) and blood pressure was obtained., Results: The intra-observer and inter-observer reliability were 0.884 and 0.740, respectively. Cronbach's alpha for two evaluation cycles each of two observers was 0.870. The prevalences of glaucomatous optic nerve atrophy in the different age groups were 0.07 % (45 - 49 years), 0.40 % (50 - 54 years), 0.45 % (55 - 59 years) and 0.82 % (60 - 64 years). Age could be established as an important risk factor for glaucomatous optic nerve atrophy, while no influence of gender or family history was found., Conclusion: Telemedical evaluation of colour images of the retina and optic nerve acquired by a non-mydriatic fundus camera allows a fast and efficient screening of many subjects with medium reliability., (© Georg Thieme Verlag KG Stuttgart · New York.)

Objectives: Comparison of classification methods using data of one clinical study. The tuning of hyperparameters is assessed as part of the methods by nested-loop cross-validation., Methods: We assess the ability of 18 statistical and machine learning classifiers to detect glaucoma. The training data set is one case-control study consisting of confocal scanning laser ophthalmoscopy measurement values from 98 glaucoma patients and 98 healthy controls. We compare bootstrap estimates of the classification error by the Wilcoxon signed rank test and box-plots of a bootstrap distribution of the estimate., Results: The comparison of out-of-bag bootstrap estimators of classification errors is assessed by Spearman's rank correlation, Wilcoxon signed rank tests and box-plots of a bootstrap distribution of the estimate. The classification methods random forests 15.4%, support vector machines 15.9%, bundling 16.3% to 17.8%, and penalized discriminant analysis 16.8% show the best results., Conclusions: Using nested-loop cross-validation we account for the tuning of hyperparameters and demonstrate the assessment of different classifiers. We recommend a block design of the bootstrap simulation to allow a statistical assessment of the bootstrap estimates of the misclassification error. The results depend on the data of the clinical study and the given size of the bootstrap sample.

Objectives: The ability of various classifiers to discriminate between normal and glaucomatous eyes based on features derived from automated analysis of laser scanning images of the eye background is investigated., Methods: To compare the classifiers without over-optimization for a given dataset, we use a simulation model to create topography images. We designed three different simulation setups as model of extreme situations and medical subgroups., Results: Neither linear nor tree-based classifiers are ideal for all setups. The most robust performance is obtained by a combination of both, so-called Double-Bagging. Classification of real data from a case-control study shows best results with Double-Bagging. All results obtained with the analysis method extracting features automatically are worse than those obtained by the same classifiers but with features derived from an analysis method that requires intervention of a physician., Conclusions: Robust classification results for classification of laser scanning images obtained with the Heidelberg Retina Tomograph are achieved by combined classifiers. The examined automated procedure causes an increased misclassification error compared to the established clinical routine requiring an expert physician's intervention.