Redistribution of the neurosensory retina in inferior limited macular translocation: an evaluation using image registration

• Published in: Graefe's archive for clinical and experimental ophthalmology 2007-03, Vol.245 (3), p.437-442
• Main Authors: Denion, Eric, Ordonez, John-Richard, Klein, Jean-Claude, Glacet-Bernard, Agnès, Walter, Thomas, Caputo, Georges
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Algorithms; Coloring Agents; Female; Fluorescein Angiography - methods; Humans; Indocyanine Green; Macular Degeneration - surgery; Male; Middle Aged; Ophthalmology; Retina - physiology; Retina - transplantation; Retrospective Studies
• ISSN: 0721-832X; 1435-702X
• DOI: 10.1007/s00417-006-0408-1

The present work aims to demonstrate the role of redistribution of the neurosensory retina (RNR)-that is, retinal tissue stretching and compression-in inferior limited macular translocation (LMT), to perform displacement mapping all over the area centralis, and to try and improve displacement evaluation precision. Using piecewise linear image registration (through Matlab) on indocyanine green angiographic frames, we have evaluated the deformation (on both the retinal and choroido-scleral layer) of a pre-operative grid superimposed on the area centralis, plotted displacement vectors all over the area centralis, and calculated a scale ratio between pre- and post-operative frames. The scale ratio tests whether or not the same physical distance is represented by the same distance on these frames. The choroido-scleral layer is not deformed following surgery. The retinal layer shows areas of stretching and compression, demonstrating the RNR. Displacement vectors feature a rotating pattern around the optic disk (except if a macular fold exists) and a positive naso-temporal gradient. The median scale ratio is 3.8% (2.5% to 5.8%). LMT involves RNR accounting for inferior retinal displacement, despite the supposedly limited internal scleral shortening. Our study confirms that the optic disk temporal edge is the displacement axis (except if a macular fold exists). Our method, unaffected by a non-zero scale ratio, provides more precision than previously published.