Electrophysiological characterization of active versus inactive birdshot retinochoroidopathy

Abstract

Purpose : Birdshot retinochoroidopathy (BSRC) is a chronic disease that typically requires treatment, and monitoring of disease course is supported by imaging, perimetry, and full field electroretinography (ffERG), which is an objective measure for quantifying generalized retinal function. The aim of this study is to characterize the ffERG features of active versus inactive BSRC.

Methods : A retrospective case-control study was performed at Casey Eye Institute. All patients with BSRC and ffERG testing between the years 1999–2019 were included. Active or inactive disease status was determined by chart review. Healthy controls were used for comparison. Amplitude (A) and implicit time (T) for the standard components of the waveforms were analyzed using a linear nested mixed-effects model, controlling for inter-eye correlation, age, and multiple comparisons (Bonferroni adjusted α=0.017).

Results : 70 patients were included: 47 control, 17 active, and 6 inactive. The following were statistically significant (see Table): Active BSCR had higher a-wave A and lower b:a wave A ratio than controls for the photopic responses (p=0.0007, 0.0002). Active BSCR had slower T than either inactive BSCR or controls for bright scotopic a-wave (p=0.0003, 0.0004), photopic b-wave (p=0.015, <0.0001), and 30 Hz flicker (p=0.002, 0.0003). Inactive BSCR also had slower T than controls for 30 Hz flicker (p<0.0001).

Conclusions : This study confirms that active BSRC frequently slows responses arising from dark- and light-adapted photoreceptors and bipolar cells. Decreased photopic b:a wave ratios were associated with supernormal amplitudes of the a-wave without a reduction in the b-wave, suggesting that inflammation in BSRC may cause irritability of light-adapted cone photoreceptor-driven responses. Indeed, supernormal photopic a-wave amplitudes has been reported in early stage BSRC and other causes of intraocular inflammation. The 30 Hz flicker peak time could differentiate between active BSRC, inactive BSRC, and control, which further supports its usefulness as a ffERG parameter in BSRC. While the 30 Hz flicker response is often regarded as a sensitive measure of cone-driven responses, the cellular mechanism behind its sensitivity as a biomarker of clinical disease status in BSRC is unknown.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.