Abstract
Abstract: :
Purpose: Matching Pursuit (MP) is a time–frequency analysis method to characterize transient activities with high resolution. We used MP to characterize the oscillatory potentials (OPs) in the photopic slow–sequence multifocal (mf) ERG of monkeys, and to compare effects of blocking inner retinal spiking activity and experimental glaucoma on these components. Methods: MfERGs were recorded from anesthetized macaque monkeys using DTL fibers. The stimulus was 103 equal–sized hexagons within 17o of the fovea. Every 200 ms each hexagon had a 50% chance of staying at 20 cd/m2 or flashing briefly (4.7 cd–s/m2). Recordings were made in control eyes (n=18) and after TTX to block Na+–dependent spiking activity (1–2 µM, n=4) or laser–induced monocular experimental glaucoma (n=8). MP was used to extract and quantify OPs in three retinal regions (3 hexagons spanning the horizontal at 7.5–10.5° eccentricity for nasal and temporal, central 7 hexagons for foveal) using three parameters [frequency (peak frequency and bandwidth of ±2SD), position (in time) and energy] of the time–frequency localized weighted Gabor "atoms". Total deviation (TD) from expected perimetric sensitivity was calculated for test locations falling within the three retinal regions. Results: The major macular OPs were adequately represented by two Gabor atoms of different frequencies, designated the Fast OP (mean frequency=153±40 Hz) and the Slow OP (84±32 Hz, n=18). Neither OP showed significant differences in frequency or bandwidth for the three regions (p<0.05). Fast and slow OP frequency bands were distinct in individual eyes, but overlapped slightly across monkeys. TTX significantly reduced Fast OP energy in nasal (by 51%), foveal (45%) and temporal (70%) regions, while frequency and position were stable. In contrast, TTX did not significantly affect Slow OPs in any region. In glaucomatous eyes, Fast OP energy dropped when TD was –5 dB or less; a further gradual decline accompanied increasing sensitivity loss. Slow OPs again were unaffected. In the severest case (TD: nasal=–14 dB, foveal=–17 dB, temporal=–34 dB), Fast OP energy decreased in each region by amount similar to that after TTX. Conclusions:Two major OP components are present in the slow–sequence mfERG, a Fast OP and a Slow OP. TTX and experimental glaucoma reduce the energy of the Fast but not the Slow OP, suggesting that the Fast OP represents spiking activity of ganglion cells and their axons. Because OPs are known to be eliminated by pharmacologic suppression of inner retinal activity, it is likely that the Slow OP is of amacrine cell origin.
Keywords: electroretinography: non-clinical • retina: proximal (bipolar, amacrine, and ganglion cells)