Comparison of spectral domain optical coherence tomography and color photographic imaging of the optic nerve head in management of glaucoma

Glaucoma Specialist Detection of Optical Coherence Tomography Suspicious Rim Tissue in Glaucoma and Glaucoma Suspect Eyes

PURPOSE

To assess glaucoma specialists' detection of optic nerve head (ONH) rim tissue that is thin by optical coherence tomography (OCT) criteria.

DESIGN

Reliability analysis.

METHODS

Five clinicians marked the disc margin (DM) and rim margin (RM) on stereophotographs of 151 glaucoma or glaucoma suspect eyes obtained within 3 months of OCT imaging. The photo and OCT infrared image for each eye were co-localized and regionalized into 12 sectors relative to the axis between the Bruch membrane opening (BMO) centroid and the fovea. For each clinician, the distance from BMO centroid to their DM (DM radius) and RM (RM radius) was used to generate sectoral rim width (RW) (DM radius-RM radius) and cup-to-disc ratio (CDR) (RM radius/DM radius) estimates. OCT minimum rim width (MRW) was determined by sector. Among all eyes, for each OCT MRW suspicious sector (<5% of OCT normative database), we determined each clinician's detection (clinician CDR ≥ 0.7).

RESULTS

Clinicians most commonly failed to detect OCT suspicious rim tissue in the nasal sectors. Among 502 sectors with suspicious OCT MRW, all 5 clinicians rated CDR ≥ 0.7 in only 29.5% and all 5 clinicians rated CDR < 0.7 in 21%. OCT suspicious rim thickness was most common (32% of eyes) in the nasal and inferior sectors. MRW vs clinician RW discordance was greatest nasally, while BMO vs clinician DM discordance was greatest temporally.

CONCLUSIONS

Clinicians most commonly failed to detect OCT suspicious rim thickness nasally where suspicious rim tissues were also most common.

Differences in Optical Coherence Tomography Assessment of Bruch Membrane Opening Compared to Stereoscopic Photography for Estimating Cup-to-Disc Ratio

PURPOSE

To compare the vertical and horizontal cup-to-disc ratio (VCDR, HCDR) by an updated optical coherence tomography (OCT) Bruch membrane opening (BMO) algorithm and stereoscopic optic disc photograph readings by glaucoma specialists.

DESIGN

Reliability analysis.

METHODS

A total of 195 eyes (116 glaucoma and 79 glaucoma suspect) of 99 patients with stereoscopic photographs and OCT scans of the optic discs taken during the same visit were compared. Optic disc photographs were read by 2 masked glaucoma specialists for VCDR and HCDR estimation. Intraclass correlation coefficient (ICC) and Bland-Altman plots were used to assess the agreement between photograph reading and OCT in estimating CDR.

RESULTS

OCT images computed significantly larger VCDR and HCDR than photograph reading before and after stratifying eyes based on disc size (P < .001). The difference in CDR estimates between the 2 methods was equal to or greater than 0.2 in 29% and 35% of the eyes for VCDR and HCDR, respectively, with a mean difference of 0.3 in each case. The ICCs between the readers and OCT ranged between 0.50 and 0.63. The size of disagreement in VCDR correlated weakly with cup area in eyes with medium (r² = 0.10, P = .008) and large (r² = 0.09, P = .007) discs.

CONCLUSIONS

OCT and photograph reading by clinicians agree poorly in CDR assessment. The difference in VCDR between the 2 methods was depended on cup area in medium and large discs. These differences should be considered when making conclusions regarding CDRs in clinical practice.

Cup-to-Disc Ratio From Heidelberg Retina Tomograph 3 and High-Definition Optical Coherence Tomography Agrees Poorly With Clinical Assessment

PURPOSE

To investigate the agreement of vertical cup-to-disc ratio (VCDR) measured from Heidelberg Retina Tomograph 3 (HRT-3), high-definition optical coherence tomography (HD-OCT), and clinical grading.

METHODS

A total of 933 consecutive subjects underwent optic nerve head imaging with HRT-3 and HD-OCT during a single visit. The vertical dimensions of the disc and cup were measured by slit-lamp examination using an eyepiece graticule. Bland-Altman plots evaluated the agreement in the VCDR obtained by the 2 instruments and clinical grading.

RESULTS

We found a significant trend for the difference in VCDR measured clinically and with HRT and HD-OCT (P<0.001 for both). The mean VCDR differed significantly between clinical measurement and that evaluated by HRT and HD-OCT, respectively (0.40 ± 0.12 vs. 0.37 ± 0.21 and 0.40 ± 0.12 vs. 0.50 ± 0.14, P<0.001 for both), with significantly poor correlation (r(2)=0.35 and 0.34, P<0.001 for both).

CONCLUSIONS

In general, there is poor correlation and a lack of agreement of VCDR measured using HRT-3, HD-OCT, and clinical grading. HD-OCT has somewhat better agreement with clinical measurements of VCDR. Compared with clinical grading, HD-OCT tended to overestimate VCDR, whereas HRT-3 tended to underestimate it.

Agreement between spectral domain optical coherence tomography and retinal nerve fiber layer photography in chinese

PURPOSE

To compare spectral domain optical coherence tomography (OCT) and fundus photography to identify localized retinal nerve fiber layer defects (RNFLD) in Chinese patients with early-to-medium advanced glaucoma.

METHODS

Thirty-nine patients (48 eyes) with localized RNFLDs by fundus photography and 48 age-matched control individuals were included into the study. The individuals underwent spectral domain OCT of the retinal nerve layer. In OCT, a localized RNFLD was defined as a dipping of the retinal nerve fiber layer thickness curve into the red-colored band of the graph, measured at a peripapillary circle with a diameter of 3.46 mm.

RESULTS

In the 48 eyes of the study group, 63 localized RNFLDs were seen on the fundus photographs. On the OCTs, 58 of these 63 localized RNFLDs were detected, whereas 5 defects were not detected. Two localized RNFLDs seen on the OCTs were not found on the corresponding fundus photographs. The resulting sensitivity and specificity of OCT for detecting localized RNFLDs were 92% and 96%, respectively. The overall agreement rate between both methods was 94% (90/96), and the κ value was 0.90 (P<0.001). The results of both techniques correlated with each other for the determination of the location (Pearson correlation coefficient (r)=0.99; P<0.001) and the width of the localized RNFLDs (201 ± 123 degrees vs. 207 ± 115 degrees; r=0.93; P<0.001).

CONCLUSIONS

Spectral domain OCT as compared with examiner-performed assessment of conventional fundus photographs is capable of detecting and measuring localized retinal nerve fiber layer defects with a relatively high diagnostic precision.

Comparison of clinical and spectral domain optical coherence tomography optic disc margin anatomy

PURPOSE

To investigate spectral domain optical coherence tomography (SD-OCT)-detected optic disc margin anatomy in the monkey eye by colocalizing disc photographs to SD-OCT scans acquired from the same eyes.

METHODS

The neural canal opening (NCO) was delineated within 40 digital radial sections generated from SD-OCT volumes acquired from 33 normal monkey eyes (15 degrees, 290 x 768 horizontal grid pattern). Each volume was colocalized to its disc photograph by matching the retinal vessels within each photograph to vessel outlines visible within en face SD-OCT images. Border tissue was delineated where it extended internally to the NCO. A clinician (masked to delineated points) marked the disc margin onto each photograph while viewing the relevant stereophotograph pair. Alignment of the clinician-ascribed disc margin to the NCO and border tissue delineation was assessed. The process was repeated in a single myopic human eye.

RESULTS

In 23 eyes, the NCO aligned to the disc margin. In 10 eyes, externally oblique border tissue was detectable in the temporal disc. In these regions of the disc, the termination of border tissue was the disc margin. An exaggerated form of this phenotype was observed in the myopic human eye. In this case, temporal border tissue terminated at the anterior scleral canal opening, which was detected as the disc margin.

CONCLUSIONS

The termination of Bruch's membrane, border tissue, and the anterior scleral canal opening may constitute the disc margin within the same eye, depending on the border tissue architecture; this anatomy is consistently visualized by SD-OCT.