Which glaucoma patients benefit from 10-2 visual field testing? Proposing the functional vulnerability zone framework

Comparative analysis of central versus peripheral visual field test grids in the diagnosis of glaucoma

Accumulating evidence has recognised central visual field defects (CVFDs) as a common feature of glaucoma. Current glaucoma screening guidelines include peripherally biased perimetry (24-2 protocols), but test grids exist to test the integrity of the central visual field (10-2 protocols). However, the added benefit of incorporating central visual field assessments alongside peripheral-biased testing grids remains unclear. This scoping review aimed to compare the diagnostic accuracy of central versus peripheral visual field tests. A systematic search of six databases yielded relevant studies among glaucoma subjects. These studies were synthesised narratively, focusing on diagnostic performance indicators such as the area under the curve, sensitivity, specificity, diagnostic agreement, and structure-function concordance. Of the 1875 studies screened, 16 were included in the review. The comparative analyses demonstrated a similar diagnostic performance when comparing the ability of the 24-2 and 10-2 test grids to detect glaucoma or CVFDs. When utilising the mean deviation, the 24-2 area under the curve ranged between 0.81-0.87 and 0.74-0.84 for the 10-2, whilst the area under the curve of the pattern standard deviation was 0.95 and 0.82, respectively. The pattern standard deviation showed sensitivities reaching 0.75 for the 24-2 and 0.60 for the 10-2, with specificities as high as 0.95 for both test grids. Across all disease stages, CVFDs detected on the 24-2 demonstrated up to 88% agreement with functional damage detected on the 10-2. The agreement between structure-function damage was greatest when combining test grids with optical coherence tomography (88.7%). This review indicates that the 24-2 and 10-2 testing protocols offer comparable diagnostic performance for glaucoma, including detecting CVFDs. While targeted macula screening could provide additional diagnostic value in certain contexts, the evidence remains inconclusive. Further longitudinal studies, incorporating optical coherence tomography, are necessary to confirm these findings and consider the routine inclusion of CVFD screening in clinical practice.

The effect of macular visual field test density on central structure-function concordance in glaucoma

CLINICAL RELEVANCE

Central visual field (VF) testing often requires focussed high-density test grids. The critical number of test locations for maximising structure-function concordance in the macula is not known.

PURPOSE

The aim of this work is to determine the impact of the number of test locations in the central VF on binarized structure-function concordance in glaucoma.

METHODS

Humphrey Field Analyser (HFA) 10-2 test grid and Cirrus optical coherence tomography Ganglion Cell Analysis (GCA) results from one eye of 155 glaucoma patients were extracted. Following anatomical correction for retinal ganglion cell displacement, the pointwise results of the central 36 locations of the 10-2 pattern deviation map and their corresponding locations within the GCA deviation map were recorded. The number of test locations was systematically reduced from 36 (4 locations per step) and added from 1 (1 location per step) and binarized structure-function concordance (p < 0.05 for both) at each step was evaluated. Eleven test point subtraction and addition models were developed. Concordance rates (proportion) were plotted as a function of number of test locations, and were fitted using segmental nonlinear regression to identify the critical point of inflection at which concordance was maximised and discordance minimised.

RESULTS

Subtractive and additive approaches returned two-way estimates of the critical number, with, on average 8-14 test locations being the range at which structure-function concordance was optimised in the present cohort across all models. A randomised approach to subtracting or adding test locations returned critical numbers that were similar to systematic and empirical models, suggesting that specific test location was not as critical in optimising structure-function concordance compared to the number of test locations.

CONCLUSION

There is a potential critical number (8-14) in macular visual field testing where binarized structure-function concordance is optimised, providing a framework for guiding the development of integrated macular test locations in VF testing for glaucoma.

Predicting visual field global and local parameters from OCT measurements using explainable machine learning

Glaucoma is characterised by progressive vision loss due to retinal ganglion cell deterioration, leading to gradual visual field (VF) impairment. The standard VF test may be impractical in some cases, where optical coherence tomography (OCT) can offer predictive insights into VF for multimodal diagnoses. However, predicting VF measures from OCT data remains challenging. To address this, five regression models were developed to predict VF measures from OCT, Shapley Additive exPlanations (SHAP) analysis was performed for interpretability, and a clinical software tool called OCT to VF Predictor was developed. To evaluate the models, a total of 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) and 226 normal eyes were included. The machine learning models outperformed recent OCT-based VF prediction deep learning studies, with correlation coefficients of 0.76, 0.80 and 0.76 for mean deviation, visual field index and pattern standard deviation, respectively. Introducing the pointwise normalisation and step-size concept, a mean absolute error of 2.51 dB was obtained in pointwise sensitivity prediction, and the grayscale prediction model yielded a mean structural similarity index of 77%. The SHAP-based analysis provided critical insights into the most relevant features for glaucoma diagnosis, showing promise in assisting eye care practitioners through an explainable AI tool.