Comparison of optical coherence tomography and scanning laser polarimetry for detection of localized retinal nerve fiber layer defects.

PURPOSES

To compare the ability of Stratus optical coherence tomography (Stratus OCT) and scanning laser polarimetry with variable corneal compensator (GDx VCC) in recognizing a localized retinal nerve fiber layer (RNFL) defect identified on red-free fundus photography.

MATERIALS AND METHODS

Fifty-three patients with only 1 localized RNFL defect in either eye were taken RNFL thickness analysis using Stratus OCT and GDx VCC. Thirty-nine healthy subjects were used as controls and only 1 eye per subject was considered. Using red-free photography as the standard reference test, sensitivity and specificity for photographic defects, and topographic correlation with photographic defects were compared between Stratus OCT (sector average graph) and GDx VCC (deviation from normal map). Abnormal sectors at P<5% compared with their internal normative database were evaluated.

RESULTS

After excluding eyes with unacceptable scan images, 38 healthy eyes and 47 glaucomatous eyes were finally included. Stratus OCT and GDx VCC showed moderate sensitivity (78.7%) and high specificity (94.7% and 89.5%, respectively), and there was no significant difference (P=1.00 and P=0.69, respectively). RNFL defects determined by Stratus OCT and GDx VCC were correlated well with photographic RNFL defects in terms of peripapillary localization and clock-hour size, and there was no significant difference between 2 imaging devices (P=0.20 and P=0.27, respectively).

CONCLUSIONS

In recognizing a localized RNFL defect, overall diagnostic performance of Stratus OCT and GDx VCC with regard to their internal normative database was not significantly different. As both Stratus OCT and GDx VCC showed only moderate sensitivity, these imaging devices may not substitute red-free fundus photography in clinical practice of glaucoma diagnosis.