Wavelength independence and interdevice variability of optical coherence tomography

BACKGROUND AND OBJECTIVE: Comparison of nerve fiber layer (NFL) thickness and macular retinal thickness measurements in two different commercial optical coherence tomography (OCT) machines with the same and different super luminescent diode lasers (SLDs). PATIENTS AND METHODS: Thirty eyes of 30 subjects were studied, with each eye scanned on two different OCT machines on the same day. Three 3.4-mm diameter circumpapillary NFL scans and six 6-mm radial scans centered on the fovea were obtained. Macular volumes were calculated from the central 3.45 mm of the 6 radial scans. RESULTS: The first study (identical SLDs of 850 nm) included 10 eyes of 10 subjects for mean NFL thickness and 6 eyes of 6 subjects for macular volume. There was no systematic difference between the measurements on the two machines for NFL (difference = 5.6 μm, standard error [SE] = 3.8, P = .18) or macular volume (difference = -0.003 mm³, SE = 0.064, P = .96). The second study (SLDs of 850 and 820 nm) included 20 eyes of 20 subjects. There was no significant difference between the two machines for mean NFL (difference = 3.4 μm, SE = 2.9, P = 0.26) or macular volume (difference = -.017 mm³, SE = 0.017, P = .33). For the smaller areas of the macular scan, several (including the outer ring) showed a significant systematic difference between measurements at the two wavelengths, but even these had at most 12% of the total variance attributable to the different wavelengths. CONCLUSION: Our study indicates that OCT measurements of NFL thickness do not differ much when the device or the wavelength of the SLD is changed. Some measurements of the macular thickness in specific areas may differ systematically for different wavelengths, but in most cases the difference is small.