Abstract
Purpose :
Early detection of ganglion cell (GCL) and retinal nerve fibre layer (RNFL) thinning, using automated optical coherence tomography (OCT) segmentation, may facilitate the prevention of glaucomatous vision loss. However, long term treatment may be unwarranted if there are false positive artefacts in the analysis. The purpose of this study is to determine the nature of artefacts in a consecutive series of patients with vitreomacular traction (VMT).
Methods :
A retrospective series of patients with VMT was compiled from a retinal clinic for the period between October 2012 and September 2014. VMT was defined as vitreous adhesion at the foveal center with deformation of the foveolar contour on OCT (Cirrus HD, Carl Zeiss Meditec). Scans were obtained after spontaneous resolution of VMT or pars plana vitrectomy and were compared. Diabetic retinopathy, macular degeneration, glaucoma or other macular pathology were exclusion criteria.
All OCT scans were conducted using a standardized protocol without pupil dilation. Poor quality scans (motion artifact, poor fixation, blinking or signal strength <6) were excluded from this study. The proprietary Ganglion Cell Analysis (GCA) calculated the combined thickness of the ganglion cell complex (GCC), comprised of the RNFL, GCL and inner plexiform layer (IPL). All images were analysed and classified into pattern groups .
In addition, abnormal VMT OCT and normal OCT images were manipulated, shifting the automated foveal localisation determined by the GCA in 6 directions; the resulting CGA images were analysed.
Results :
Thirty-seven eyes of 26 patients (11 bilateral, 6 left eyes and 9 right eyes) were eligible for this study; 27 eyes had follow-up OCT scans available. Five distinct patterns of GCA were identified: normal torus, single spoke defect, wedge defect, horizontal step and complex defect.
On manually shifting the foveal centre in selected images away from the automated foveal localisation (AFL), all five patterns could be elicited in normal eyes and VMT eyes.
Conclusions :
In the presence of VMT, interpretation of abnormal GCA is of limited benefit as artefacts are introduced by the associated difficulty in automated foveal localisation and software assumptions about the pattern of variation in thickness of the GCC at the centre of the macula. In the absence of VMT, it is important to check the AFL before interpreting abnormal findings.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.