Glaucoma Diagnostic Capability of Circumpapillary Retinal Nerve Fiber Layer Thickness in Circle Scans With Different Diameters.
Ghassibi Mark P, Chien Jason L, Patthanathamrongkasem Thipnapa, Abumasmah Ramiz K, Rosman Michael S, Skaat Alon, Tello Celso, Liebmann Jeffrey M, Ritch Robert, Park Sung Chul
AI Summary
This study found the 4.1mm OCT scan diameter's inferotemporal RNFLT offered superior glaucoma diagnostic capability, suggesting optimizing scan parameters can improve early detection and reduce artifacts.
Abstract
Purpose
To compare varying circumpapillary optical coherence tomographic (OCT) scan diameters for glaucoma diagnosis.
Materials and methods
Prospective, cross-sectional, observational study. Circumpapillary retinal nerve fiber layer thickness (RNFLT) was measured using spectral-domain OCT in 1 randomly selected eye. Scans with diameters of 3.5, 4.1, and 4.7 mm were obtained, each with 7 parameters: mean global (G) RNFLT and mean RNFLT for the temporal-inferior (TI), nasal-inferior (NI), temporal-superior (TS), nasal-superior (NS), nasal (N), and temporal (T) sectors. Areas under the receiver operating characteristic curve (AUCs) were calculated.
Results
Mean age was 55±18 years in 68 healthy eyes and 59±15 years in 95 glaucomatous eyes (P=0.12). Visual field mean deviation was -7.55±6.61 dB in glaucomatous eyes. In all 3 circle scans, mean TI RNFLT had the greatest AUC (0.974 to 0.983), followed by mean G RNFLT (0.949 to 0.956). The AUC of mean TI RNFLT in the 4.1-mm scan (0.983) was greater than the AUCs of mean TI RNFLTs in the 4.7- (0.978; P=0.128) and 3.5-mm (0.974; P=0.049) scans. The AUC of mean TI RNFLT in the 4.1-mm scan (0.983) was greater than the AUCs of mean G RNFLTs in the 3.5- (0.954; P=0.011), 4.1- (0.956; P=0.016), and 4.7-mm (0.949; P=0.011) scans. In 2 eyes with large parapapillary atrophy, RNFL segmentation error was noted only in the 3.5-mm scan in the area of parapapillary atrophy.
Conclusions
Further investigations to find the spectral-domain OCT circle scan diameter with the best diagnostic capability and the least artifacts are warranted, especially focusing on larger-than-conventional circle scans.
MeSH Terms
Shields Classification
Key Concepts4
In a prospective, cross-sectional, observational study of 68 healthy eyes and 95 glaucomatous eyes, the mean temporal-inferior (TI) retinal nerve fiber layer thickness (RNFLT) measured with spectral-domain OCT in the 4.1-mm scan had the greatest area under the receiver operating characteristic curve (AUC) (0.983) for glaucoma diagnosis.
In a prospective, cross-sectional, observational study of 68 healthy eyes and 95 glaucomatous eyes, the AUC of mean TI RNFLT in the 4.1-mm scan (0.983) was greater than the AUCs of mean TI RNFLTs in the 4.7-mm (0.978; P=0.128) and 3.5-mm (0.974; P=0.049) scans for glaucoma diagnosis.
In a prospective, cross-sectional, observational study of 68 healthy eyes and 95 glaucomatous eyes, the AUC of mean temporal-inferior (TI) retinal nerve fiber layer thickness (RNFLT) in the 4.1-mm scan (0.983) was greater than the AUCs of mean global (G) RNFLTs in the 3.5-mm (0.954; P=0.011), 4.1-mm (0.956; P=0.016), and 4.7-mm (0.949; P=0.011) scans for glaucoma diagnosis.
In a prospective, cross-sectional, observational study of 68 healthy eyes and 95 glaucomatous eyes, RNFL segmentation error was noted only in the 3.5-mm scan in the area of parapapillary atrophy in 2 eyes with large parapapillary atrophy.
Related Articles5
Utilization of Image-Based Deep Learning in Multimodal Glaucoma Detection Neural Network from a Primary Patient Cohort.
Observational StudyA Deep Learning Model for Glaucoma Detection Outperforms Conventional Metrics.
Observational StudyDiagnostic Performance of a Handheld Electroretinography Test for Glaucoma.
Cross-Sectional StudyNormative Percentiles of Retinal Nerve Fiber Layer Thickness and Glaucomatous Visual Field Loss.
Observational StudyDetecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions: Transformer for Improved Generalization.
Evaluation of a diagnostic technologyIs this article assigned to the wrong chapter(s)? Let us know.