TY - JOUR
T1 - Diagnostic Accuracy of Spectral-Domain OCT Circumpapillary, Optic Nerve Head, and Macular Parameters in the Detection of Perimetric Glaucoma
AU - McCann, Paul
AU - Hogg, Ruth E.
AU - Wright, David M.
AU - McGuinness, Bernadette
AU - Young, Ian S.
AU - Kee, Frank
AU - Azuara-Blanco, Augusto
PY - 2019/9/7
Y1 - 2019/9/7
N2 - Purpose: To evaluate the diagnostic accuracy of circumpapillary retinal nerve fiber layer (cRNFL), optic nerve head, and macular parameters for the detection of glaucoma using Heidelberg Spectralis OCT (Heidelberg Engineering, Inc., Heidelberg, Germany). Design: Cross-sectional study. Participants: Participants of the Northern Ireland Cohort for the Longitudinal Study of Ageing with a vertical cup-to-disc ratio of 0.7 or more, vertical cup-to-disc ratio asymmetry of 0.2 or more, vertical neuroretinal rim ratio of 0.1 or less, intraocular pressure of 25 mmHg or more, or a combination thereof were invited to the study. Methods: Participants underwent clinical examination by a masked glaucoma expert and full-threshold visual field testing to define glaucoma. Five index tests were performed: (1) standard cRNFL thickness, (2) macular Early Treatment Diabetic Retinopathy Study (ETDRS) scans, (3) macular posterior pole asymmetry analysis (PPAA) scans, (4) Bruch’s membrane opening minimum rim width (BMO MRW), and (5) Glaucoma Module Premium Edition (GMPE) cRNFL Anatomic Positioning System (APS) scans. We analyzed the eye with more advanced disease per participant. Main Outcome Measures: Analysis of receiver operating characteristic (ROC) curve; area under the ROC curve (AUC); and partial AUC (pAUC) at specificity of 0.80 to 1, 0.90 to 1, and 0.95 to 1. Sensitivity at 0.95 specificity and specificity at 0.95 sensitivity were reported. Primary analysis included all available scans. Results: One hundred twenty-eight eyes from 128 participants were enrolled (52 eyes with perimetric glaucoma and 76 eyes without glaucoma). One hundred thirteen standard cRNFL thickness scans; 107 GMPE cRNFL APS 3.5-mm, 4.1-mm, and 4.7-mm scans; 107 BMO-MRW scans; 98 ETDRS scans; and 97 PPAA scans were available. Standard cRNFL mean global thickness showed highest AUC (0.869; 95% confidence interval [CI], 0.800–0.938) and the highest pAUC at specificity of 0.80 to 1 (0.815; 95% CI, 0.742–0.887), at specificity of 0.90 to 1 (0.794; 95% CI, 0.713–0.875), and at specificity of 0.95 to 1 (0.765; 95% CI, 0.696–0.861). Standard cRNFL mean global thickness scans provided the highest sensitivity at 0.95 specificity (0.630), whereas ETDRS mGCL outer inferior sector provided the highest specificity at 0.95 sensitivity (0.522). Conclusions: Macular and optic nerve head OCT parameters were not better than cRNFL measurements to diagnose glaucoma in this population.
AB - Purpose: To evaluate the diagnostic accuracy of circumpapillary retinal nerve fiber layer (cRNFL), optic nerve head, and macular parameters for the detection of glaucoma using Heidelberg Spectralis OCT (Heidelberg Engineering, Inc., Heidelberg, Germany). Design: Cross-sectional study. Participants: Participants of the Northern Ireland Cohort for the Longitudinal Study of Ageing with a vertical cup-to-disc ratio of 0.7 or more, vertical cup-to-disc ratio asymmetry of 0.2 or more, vertical neuroretinal rim ratio of 0.1 or less, intraocular pressure of 25 mmHg or more, or a combination thereof were invited to the study. Methods: Participants underwent clinical examination by a masked glaucoma expert and full-threshold visual field testing to define glaucoma. Five index tests were performed: (1) standard cRNFL thickness, (2) macular Early Treatment Diabetic Retinopathy Study (ETDRS) scans, (3) macular posterior pole asymmetry analysis (PPAA) scans, (4) Bruch’s membrane opening minimum rim width (BMO MRW), and (5) Glaucoma Module Premium Edition (GMPE) cRNFL Anatomic Positioning System (APS) scans. We analyzed the eye with more advanced disease per participant. Main Outcome Measures: Analysis of receiver operating characteristic (ROC) curve; area under the ROC curve (AUC); and partial AUC (pAUC) at specificity of 0.80 to 1, 0.90 to 1, and 0.95 to 1. Sensitivity at 0.95 specificity and specificity at 0.95 sensitivity were reported. Primary analysis included all available scans. Results: One hundred twenty-eight eyes from 128 participants were enrolled (52 eyes with perimetric glaucoma and 76 eyes without glaucoma). One hundred thirteen standard cRNFL thickness scans; 107 GMPE cRNFL APS 3.5-mm, 4.1-mm, and 4.7-mm scans; 107 BMO-MRW scans; 98 ETDRS scans; and 97 PPAA scans were available. Standard cRNFL mean global thickness showed highest AUC (0.869; 95% confidence interval [CI], 0.800–0.938) and the highest pAUC at specificity of 0.80 to 1 (0.815; 95% CI, 0.742–0.887), at specificity of 0.90 to 1 (0.794; 95% CI, 0.713–0.875), and at specificity of 0.95 to 1 (0.765; 95% CI, 0.696–0.861). Standard cRNFL mean global thickness scans provided the highest sensitivity at 0.95 specificity (0.630), whereas ETDRS mGCL outer inferior sector provided the highest specificity at 0.95 sensitivity (0.522). Conclusions: Macular and optic nerve head OCT parameters were not better than cRNFL measurements to diagnose glaucoma in this population.
U2 - 10.1016/j.ogla.2019.06.003
DO - 10.1016/j.ogla.2019.06.003
M3 - Article
SN - 2589-4196
VL - 2
SP - 336
EP - 345
JO - Ophthalmology Glaucoma
JF - Ophthalmology Glaucoma
IS - 5
ER -
