Abstract
Abstract: :
Purpose: Recent version of OCT3 software(ver. 2.0) compares the patient’s RNFL thickness with those of normal database. The RNFL photograph and OCT3 results with color coded normal distribution percentiles were analyzed in the eyes with localized nerve fiber layer defects to evaluate the ability of OCT3 to detect localized nerve fiber layer defects. Methods: The medical records of glaucoma patients who were taken both RNFL photography and OCT3 and having localized RNFL defects in either eye by RNFL photography were reviewed. The localized RNFL defect was defined as a defect that connected to the disc margin and gradual widening of the defect periperally. In OCT3 results the normal distribution percentiles less than 5% were considered as a significant defect. For each disc, superotemporal (75°) and inferotemporal (75°) portions were evaluated. Total 144 locations of 36 patients (29 normal–tension glaucoma and 7 primary open–angle glaucoma patients) were analyzed. The mean age of the patients was 56.9 years old and male to female ratio was 25:11. The diagnostic abilities, such as the sensitivity, specificity, positive predictive value and negative predictive value of OCT3 to detect RNFL defect were calculated. Results: The overall sensitivity, specificity, positive predictive value and negative predictive value of OCT3 to detect RNFL defect were 82.14%, 95.00%, 95.83% and 79.16%, respectively. OCT3 showed good diagnostic agreement with the RNFL photography (kappa = 75.00).
Conclusion: OCT3 with normal database can detect RNFL defect with high sensitivity and specificity, and showed good diagnostic agreement with the RNFL photography.
Keywords: nerve fiber layer • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • imaging/image analysis: clinical