The a-wave latency in control subjects and patients with retinal diseases.

Purpose: To determine the a-wave latency of the electroretinograms (ERGs) recorded from control subjects and patients with retinal diseases.

Methods: The a-wave latency and implicit time (IT) were measured retrospectively from the ERGs of 40 control subjects and 99 patients. The patients included 9 with complete congenital stationary night blindness (cCSNB), 13 with achromatopsia or cone dystrophy, 5 with supernormal and delayed rod ERG syndrome, and 72 with retinitis pigmentosa (RP). To assess whether latency measurements can be obtained reliably by different observers from patients with smaller a-wave amplitudes and noisier baselines, the a-wave latency and IT of the ERG of the right eye of 10 control subjects and 10 patients with RP were measured by three observers.

Results: The mean a-wave latency measured for the same 10 control ERGs by three observers differed by less than 1 millisecond while the mean IT differed by 1.7 milliseconds. For 10 ERGs from RP patients, the mean for the a-wave latency measured by the three observers differed by less than 2.0 milliseconds and by 1.1 millisecond for the IT. The coefficient of variation varied from 24.8% to 36.7% for the latency and from 11.5% to 16.0% for the IT. The a-wave latencies elicited by the 0-dB stimulus under scotopic and photopic conditions from the 40 control subjects were not statistically different. The a-wave latency in patients with cCSNB did not differ significantly from that in control subjects. The longer a-wave latency in patients with achromatopsia suggested that the rods have a longer latency than cones. The scotopic and photopic a-wave latencies were significantly longer in RP patients. The longer latency in RP patients was not due to smaller a- or b-wave amplitudes.

Conclusions: The a-wave latency can be measured as reliably as the IT in control subjects but the reliability is not as good for the latency as for the IT in RP patients. The larger coefficients of variation in RP patients were most likely due to the measurements being made from RP patients at different stages of their disease. Our results suggest that the a-wave latency in control subjects is determined by cones under both scotopic and photopic conditions. The longer a-wave latency in RP patients suggests that the rods and cones are altered over a significant area of the retina.