Abstract
Purpose :
Optical coherence tomography (OCT) is the established standard in diagnostics of age-related macular degeneration (AMD). Most of modern OCT devices scan a macular volume of 6 by 6 millimeters (mm) centered over the fovea – however, the necessity of such a big field of view has never been verified.
Methods :
We performed a retrospective analysis of 293 OCT volume scans of 16 patients under long-term treatment of exudative AMD in a pro-re-nata scheme (one eye per patient). All scans were acquired using Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) and consist of 25 individual B-scans with 250µm spacing, covering a 6 by 6 millimeter volume centered over the fovea. In this analysis, we deliberately symmetrically reduced (cropped) OCT volume scans to smaller cubes with an edge length of 3, 2 and 1 mm. We then compared the sensitivity and specificity for detecting the presence of the following relevant OCT biomarkers: subretinal fluid, intraretinal cysts and diffuse retinal thickening. The original and reduced scans were presented to a retina specialist in a random order and were marked as “wet” if any of these biomarkers were present, “uncertain” if presence of biomarkers could not be excluded with certainty, or “dry” otherwise. In addition, the original volume scan was graded again to evaluate re-test probability.
Results :
In the first grading of the original volume, 10 scans were marked as uncertain and excluded from further analysis. Re-test probability for the unreduced volume was 97.8%. In the reduced volumes, for the 3mm volume 6 scans (2.1%) [2mm: 4 (1.4%) / 1mm: 21 (7.4%)] were marked as uncertain. For the remaining scans, the test result correlation between reduced and source volume was 97.8% [95.7% / 88.7%]. The sensitivity was 97.3% [95.1% / 85.3%] with a specificity of 100% [98.2% / 98.1%].
Conclusions :
We could show that relevant AMD biomarkers can be detected reliably when looking only at the very central macular region. The re-test probability for the original scan was not better than the correlation between the original scan and the 3mm volume and only slightly above the 2mm volume. Therefore, small central scans seem sufficient for AMD diagnostics. This may be helpful in the construction of future OCT devices and computer analysis of OCT scans.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.