Purchase this article with an account.
Adeleh Yarmohammadi, Sasan Moghimi, Linda Zangwill, Huiyuan Hou, James Proudfoot, Christopher Bowd, Fazio Massimo, Haft Michael, De Moraes Gustavo Carlos, Christopher Girkin, Jeffrey M Leibmann, Robert N Weinreb; Racial differences in detection of glaucoma using retinal nerve fiber layer thickness and neuroretinal
minimum rim width. Invest. Ophthalmol. Vis. Sci. 2020;61(9):PB0078.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare the sensitivities and specificities of the Bruch’s membrane opening minimum rim width(BMO-MRW) and retinal nerve fiber layer thickness (RNFLT) reference database based criteria for detection ofglaucoma in subjects of European Decent (ED) and African Descent (AD).
382 eyes of 255 glaucoma patients (ED=170, AD=85) and 94 eyes of 50 healthy individuals(ED=30, AD=20) with axial length ≤26mm with global and sectoral BMO-MRW and RNFLT measurements withSpectralis Optical Coherence Tomography were included. Six diagnostic criteria were evaluated: globalmeasurement below the 5 or the 1 percentile; ≥1 sector measurement below the 5 or the 1 percentile;and superotemporal (ST) and/or inferotemporal (IT) measurement below the 5 or the 1 percentile. Thesensitivities and specificities of these measurements for detection of glaucoma were compared usingbootstrapping methods.
Sensitivities of the 6 RNFLT based criteria were higher than the corresponding MRW definition(P<0.023 for all). While MRW specificities tended to be higher than for RNFLT (p values ranged from <0.001 to0.170). ST and/or IT RNFLT below the 5 percentile has the best performance for detection of glaucomaamong RNFLT classifications with a sensitivity [95% CI] of 89.5% (86.1, 92.5), which was significantly higherthan that of ST and/or IT BMO-MRW below the 5 percentile (83.0% [78.2, 87.3]) (p=0.004) but at lowerspecificity (87.2% and 94.7%, respectively, p=0.079). In subjects of AD, sensitivities of both RNFLT and MRWusing ST and IT RNFLT and MRW below the 5 percentile criteria were significantly lower than in ED subjects(83.7% [79.6, 83.6] vs 92.5% [91.0, 96.8], and 72.1% [61.5, 80.8] vs 88.5% [83.6, 92.7], respectively), withlower specificities in AD subjects (AD: 73.7% and 89.5% vs ED: 96.4% and 98.2%, respectively). Similarly,sensitivities [95% CI] for MRW classifications to detect mild glaucoma were lower in subjects of AD (57.1%[69.4, 69.4]) compared to ED (82.1% [72.2, 88.3]) with similar specificities (89.5% and 94.1%, respectively).
RNFLT reference database criteria for detecting glaucoma had higher sensitivity but lowerspecificity than MRW classifications both in ED and AD subjects. Caution should be exercised when utilizingRNFLT and specially MRW criteria in patients of AD, as the diagnostic performance is lower than in individualsof ED.
This is a 2020 Imaging in the Eye Conference abstract.
This PDF is available to Subscribers Only