Purchase this article with an account.
Francisco Folgar, Cynthia Toth, Gui-Shuang Ying, Maureen Maguire, Glenn Jaffe, ; Comparison of spectral domain and time domain OCT for qualitative and quantitative fluid assessments in the Comparison of AMD Treatments Trials (CATT). Invest. Ophthalmol. Vis. Sci. 2013;54(15):6291.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine agreement of certified readers assessing spectral domain (SD) or time domain (TD) optical coherence tomography (OCT) for the presence of fluid and for thickness measurements of retinal layers in eyes with neovascular age-related macular degeneration (AMD) treated with ranibizumab or bevacizumab.
Paired SDOCT and TDOCT scans were obtained in a subset of patients during year 2 of the Comparison of AMD Treatments Trials (CATT). Two masked readers graded each scan and performed manual measurements independently. A senior reader arbitrated discrepant scores. Agreement on presence of fluid types was evaluated with kappa coefficients with 95% confidence limits (CL), and McNemar tests. Agreement on central foveal thickness measurements was evaluated with mean ± standard deviation (SD), mean difference (Δ) ± 2 SD (95% agreement limits), and intraclass correlation coefficients (ICC) with 95% CL.
A total of 1213 pairs of OCT scans were assessed for 384 eyes. Agreement on readable fluid status was excellent for all fluid types: range 93-99%. Agreement on presence or absence of fluid varied among fluid types. Intraretinal fluid had 73% agreement (k=0.47, CI: 0.42-0.52) and was more frequent with TDOCT (p<0.001). Subretinal and sub-retinal pigment epithelium (RPE) fluid had 87% and 80% agreement, respectively, and were more frequent with SDOCT (both p<0.001). Presence of any fluid had 82% agreement (k=0.46, CI: 0.40-0.52) and was more frequent with SDOCT (p<0.001). Retinal thickness was 154 ±66 μm with SDOCT vs. 158 ±58 μm with TDOCT (Δ 5 ±67 μm), subretinal thickness was 11 ±37 μm with SDOCT vs. 10 ±35 μm with TDOCT (Δ 2 ±35 μm), and sub-RPE thickness was 132 ±105 μm with SDOCT vs. 126 ±97 μm with TDOCT (Δ 5 ±86 μm). Agreement on paired measurements was excellent for all layers: retinal thickness ICC=0.84 (CL: 0.83-0.86), subretinal thickness ICC=0.88 (CL: 0.86-0.89), and sub-RPE tissue thickness ICC=0.91 (CL: 0.89-0.92).
In CATT, manual thickness measurements were clinically similar between SDOCT and TDOCT, and measurements among paired scans had excellent agreement. Subretinal and sub-RPE fluid were detected more frequently with SDOCT. Lower-resolution imaging, with interpretation of dark areas as cystoid macular edema, may explain the greater frequency of intraretinal fluid detected with TDOCT.
This PDF is available to Subscribers Only