TY - JOUR
T1 - Optical coherence tomography measurement of nerve fiber layer thickness and the likelihood of a visual field defect
AU - Williams, Zinaria Y.
AU - Schuman, Joel S.
AU - Gamell, Lisa
AU - Nemi, Ajit
AU - Hertzmark, Ellen
AU - Fujimoto, James G.
AU - Mattox, Cynthia
AU - Simpson, Julie
AU - Wollstein, Gadi
PY - 2002/10/1
Y1 - 2002/10/1
N2 - PURPOSE: To determine if optical coherence tomography (OCT) measurements of nerve fiber layer (NFL) thickness can be used to predict the presence of visual field defects (VFD) associated with glaucoma. DESIGN: Quota-sampled, cross-sectional study. METHODS: Retrospective study of OCT NFL thickness measurements in 276 eyes of 276 subjects. All persons received OCT NFL thickness analysis; 136 eyes underwent frequency-doubling technology (FDT) perimetry; and 140 eyes underwent Swedish interactive threshold algorithm (SITA) perimetry. We defined a parameter called NFL50, which is the NFL thickness value at which there was a 50% likelihood of a VFD with either SITA or FDT perimetry. We evaluated the use of NFL50. RESULTS: The mean NFL thickness with (n = 68) and without (n = 68) a VFD in the FDT group was 93.2 μm ± 22.6 and 108.4 μm ± 14.1, respectively. The mean NFL thickness with (n = 70) and without (n = 70) a VFD in the SITA group was 78.9 μm ± 24.8 and 103.0 μm ± 18.0, respectively. The FDT mean NFL50 value was 98.5 μm. The SITA mean NFL50 value was 87.0 μm. The area under the receiver operator characteristic (AROC) curve for mean NFL was 0.73, and the positive predictive value (PPV) for FDT mean NFL50 was 72.2%. For SITA mean NFL, the AROC was 0.79 and the PPV for NFL50 was 77.2%. CONCLUSION: Nerve fiber layer thickness analysis using OCT may be clinically useful in identifying subjects who have visual field loss. However, the PPV suggests that OCT may need higher resolution and better reproducibility to enhance its sensitivity and specificity for population screening.
AB - PURPOSE: To determine if optical coherence tomography (OCT) measurements of nerve fiber layer (NFL) thickness can be used to predict the presence of visual field defects (VFD) associated with glaucoma. DESIGN: Quota-sampled, cross-sectional study. METHODS: Retrospective study of OCT NFL thickness measurements in 276 eyes of 276 subjects. All persons received OCT NFL thickness analysis; 136 eyes underwent frequency-doubling technology (FDT) perimetry; and 140 eyes underwent Swedish interactive threshold algorithm (SITA) perimetry. We defined a parameter called NFL50, which is the NFL thickness value at which there was a 50% likelihood of a VFD with either SITA or FDT perimetry. We evaluated the use of NFL50. RESULTS: The mean NFL thickness with (n = 68) and without (n = 68) a VFD in the FDT group was 93.2 μm ± 22.6 and 108.4 μm ± 14.1, respectively. The mean NFL thickness with (n = 70) and without (n = 70) a VFD in the SITA group was 78.9 μm ± 24.8 and 103.0 μm ± 18.0, respectively. The FDT mean NFL50 value was 98.5 μm. The SITA mean NFL50 value was 87.0 μm. The area under the receiver operator characteristic (AROC) curve for mean NFL was 0.73, and the positive predictive value (PPV) for FDT mean NFL50 was 72.2%. For SITA mean NFL, the AROC was 0.79 and the PPV for NFL50 was 77.2%. CONCLUSION: Nerve fiber layer thickness analysis using OCT may be clinically useful in identifying subjects who have visual field loss. However, the PPV suggests that OCT may need higher resolution and better reproducibility to enhance its sensitivity and specificity for population screening.
UR - http://www.scopus.com/inward/record.url?scp=0036808571&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0036808571&partnerID=8YFLogxK
U2 - 10.1016/S0002-9394(02)01683-5
DO - 10.1016/S0002-9394(02)01683-5
M3 - Article
C2 - 12383810
AN - SCOPUS:0036808571
SN - 0002-9394
VL - 134
SP - 538
EP - 546
JO - American Journal of Ophthalmology
JF - American Journal of Ophthalmology
IS - 4
ER -