Purchase this article with an account.
William Best, Ryan Man, Jonathan Edward Noonan, Jing Xie, Sukhpal Singh Sandhu, Ecosse Luc Lamoureux, ; Choroidal Thickness Measurement with Varying Frame Number in Patients with Diabetic Macular Edema. Invest. Ophthalmol. Vis. Sci. 2014;55(13):1750.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Choroidal thickness(CT) measurement using Enhanced Depth Imaging Optical Coherence Tomography(EDI-OCT) is often performed with 100 frames per line to decrease the signal to noise ratio. However, a high scan density can make it difficult to image large areas due to the increased time required. This issue may be compounded by diabetic macular edema(DME), where reduced acuity may cause poor fixation and edema may obscure the underlying choroid. Measurement of CT with a smaller number of frames would reduce patient discomfort, test time and allow for larger areas to be imaged. However, the accuracy of reduced frame scans is unknown. We investigated the accuracy of CT measurements in patients with DME with a reduced number of frames as compared to the standard 100 frames.
Ten patients with DME(any severity), aged 54 to 74, were recruited from laser and injection clinics in Victoria, Australia. Consecutive images of the study eye were taken with 10, 25, 50 and 100 frames. The average value of sub-foveal choroidal thickness for each scan density, manually measured three times per image, was compared to the standard 100-frame image. Within subject differences were assessed by repeated measures analysis of variance (ANOVA) and Dunnett-adjusted pairwise comparisons, while Pearson’s correlation coefficient was utilized to assess the correlations between these measurements.
Median (interquartile range [IQR]) best-corrected visual acuity using the logarithm of the minimum angle of resolution (logMAR) chart was 0.44 (0.3-0.6), and median duration of diabetes was 13 years(IQR 5-18). Mean CT measurement was 249.9 ± 59um, 254.0 ± 56.0μm, 264.4 ± 69.6μm and 257.8 ± 66.6um for 10, 25, 50 and 100 averaged scans respectively. There was no significant difference between the means (P>0.36 using ANOVA) and pairwise comparisons showed no differences in thickness between 10, 25 and 50 scans vs. 100 scans (all P>0.05). Pearson’s correlation coefficient showed high correlations between the standard 100 frames with 10 frames (correlation coefficient: 0.98), 25 frames (correlation coefficient: 0.84) and 50 frames (correlation coefficient: 0.95).
Measurement of CT with 10 A-scans per line in patients with DME is comparable to the standard 100 scans per line. Our finding allows for shorter imaging times, which is helpful in those with impaired fixation, and is likely to increase patient comfort.
This PDF is available to Subscribers Only