Purpose:
To compare optic disc measurements between the new fourier-domain optical coherence tomography (FDOCT, RTVue-100) and the confocal scanning laser ophthalmoscopy (HRT-II) in clinical practice.
Methods:
Nonrandomized comparative study including all patients seen from August 2008 to August 2009 with good quality RTVue-100 and HRT-II images within six months of each other. The rim of the optic disc was drawn automatically by RTVue-100 and manually by a single experienced observer for HRT-II. Optic disc measurements from both imaging devices were compared using Bland-Altman plots and student t-test to assess agreement.
Results:
363 patients were included (mean age 64.7 +/-13.3). Bland-Altman plots indicate significant differences between the RTVue-100 and HRT-II for all optic nerve parameters. The HRT values were consistently larger than RTVue-100 values for optic disc area and rim area. The RTVue-100 values were larger in cup area, cup-to-disc area ratio, and vertical and horizontal cup-to-disc ratio. Student t-test showed p<0.001 for averages of all measurements between the two devices (Table 1).
Conclusions:
We conclude the disk parameters from these two imaging devices are not interchangeable. The mean RTVue-100 measurements of disc area and rim area were smaller than HRT values suggesting overall, the automated software underestimates the disc margin contours relative to HRT-II. Differences in parameters may be due to differences in scanning technology and deriving set points for defining the cup. Another possibility is that manual drawing of the disc margin in HRT-II overestimates the true margin, resulting in greater disc and rim areas. Clinically, the underestimation of RTVue-100 relative to HRT-II may result in larger estimates of cup area and cup-to-disc ratios, and thus, greater likelihood of classification as abnormal.
Keywords: imaging/image analysis: clinical • optic disc