Clinical Considerations

The ERG provides objective retinal functional data and is therefore important in the diagnosis, management, and follow-up of retinal disease. Symptomatic indications include night blindness, in which the potentially blinding rod–cone dystrophies must be distinguished from the relatively benign congenital stationary night blindness (CSNB). The dystrophies are associated with markedly abnormal a-waves in the dark-adapted bright-flash ERGs; CSNB is usually associated with a normal a-wave and a “negative” ERG waveform (see Fig 3-2). Other symptomatic indications include photophobia, which indicates generalized cone dysfunction (as in cone dystrophy), and photopsia or shimmering, which can sometimes signal the development of autoimmune retinopathy, possibly paraneoplastic. The ERG is increasingly used in the assessment and monitoring of inflammatory disorders such as birdshot chorioretinopathy; objective functional data allow clinicians to make management decisions with more confidence. The ERG facilitates an objective assessment of disease severity, aiding in decisions on when and how to treat; following treatment, it provides a valuable measure of treatment efficacy that is more sensitive than conventional clinical parameters.

Figure 3-2 Opposite: ERG findings in various disorders, with typical Normal waveforms shown at the top. The timing of the 30 Hz flicker ERG is shown by the vertical dashed line. For all responses, the flash occurs at time 0 ms. Macular dystrophy, Full-field ERG responses are all normal, but the pattern ERG (PERG) is undetectable. Cone dystrophy, DA 0.10 (rod-specific) and DA 10.0 ERG responses are normal; photopic flicker and single-flash ERGs are reduced and delayed; the PERG response is subnormal, indicating macular involvement. Rod– cone dystrophy, Retinitis pigmentosa (RP) with macular involvement; all ERG responses are markedly subnormal, with rod ERGs more affected than cone ERGs, severe reduction of the DA 10.0 a-wave, indicating photoreceptor disease, and delayed 30-Hz and single-flash cone ERGs, indicating generalized cone system dysfunction. Abnormal PERG response shows macular involvement. Rod–cone dystrophy (RP); macula spared, Full-field ERGs show an abnormal rod–cone response pattern similar to that shown directly above, but a normal PERG response shows macular sparing. cCSNB, “Complete” congenital stationary night blindness. Findings show loss of on-pathway function at a postreceptoral level. DA 0.01 response is undetectable; DA 10.0 response is profoundly electronegative, with the normal a-wave reflecting normal photoreceptor function and the marked relative b-wave reduction showing inner retinal disease; 30-Hz flicker ERG shows only minor changes in waveform and mild delay; LA 3.0 ERG shows changes diagnostic of loss of cone on-pathway function but sparing of the off-pathway. The a-wave commences normally but then shows a broadened trough. The b-wave rises sharply, with loss of the photopic oscillatory potentials, and marked reduction in the b:a ratio. The PERG response is markedly subnormal. iCSNB, “Incomplete” CSNB. DA 0.01 response is subnormal but detectable; DA 10.0 response is markedly electronegative; 30-Hz flicker ERG response is markedly subnormal, showing delay and a characteristic triphasic wave-form; LA 3.0 single-flash photopic ERG shows a subnormal a-wave and a markedly subnormal b-wave, reflecting involvement of both on- and off-cone pathways; the PERG response is sub-normal. XLRS, X-linked retinoschisis. DA 0.01 response is severely reduced; DA 10.0 response is profoundly electronegative; 30-Hz flicker ERG shows delay; LA 3.0 ERG shows delay and marked reduction in the b:a ratio; the PERG response is markedly subnormal. BCM, Blue-cone (S-cone) monochromatism. DA 0.01 and DA 10.0 ERG responses are normal; 30-Hz flicker ERG is virtually undetectable; LA 3.0 response shows only a small b-wave at approximately 50 ms consistent with an S-cone origin; PERG is undetectable.

(Courtesy of Graham E. Holder, PhD.)

Figure 3-3 Multifocal (mfERG), full-field, and pattern ERG (PERG) recordings from a patient with ABCA4 retinopathy (Stargardt disease; fundus flavimaculatus) demonstrate the importance of fixation in mfERG recording and interpretation. A, Fundus photography, B, fundus autofluorescence imaging, and C, near-infrared imaging and SD-OCT show macular atrophy centralized on the fovea. D, ERG responses are normal (see Fig 3-1 for terminology; x-axis = μV; y-axis = ms); PERG response to a 15° field is undetectable, but PERG response to a 30° field is present but subnormal. E, mfERG shows an area of dysfunction that is localized, but apparently not around the fovea, which simply reflects the eccentric fixation often present in a patient with a central scotoma.

(Courtesy of Graham E. Holder, PhD.)

Figure 3-4 mfERG, full-field, and PERG recordings in a patient with hydroxychloroquine toxicity. A, Fundus autofluorescence imaging. B, Near-infrared imaging and spectral domain–optical coherence tomography (SD-OCT). The changes shown on OCT are less marked, particularly temporal to the fovea, than may have been predicted by the degree of functional loss. C, ERG responses are normal (see Fig 3-1 for terminology); PERG response to a 15° field is barely detectable. D, mfERG response shows marked abnormality with some sparing of the response to the central foveal hexagon but loss of parafoveal responses.

(Courtesy of Graham E. Holder, PhD.)

ERGs must always be taken in a clinical context, and to enable accurate ERG diagnosis, a careful clinical history should include previous drug and/or surgical treatment as well as a family history. Results are diagnostic (pathognomonic) only for 3 relatively rare inherited disorders: bradyopsia (mutation in RGS9 or R9AP), enhanced S-cone syndrome (NR2E3), and “cone dystrophy with supernormal rod ERG” (KCNV2).

The ERG can be useful in assessing patients with vascular disease. In patients with central retinal artery occlusion (CRAO), the ERG is characteristically negative, reflecting the dual blood supply to the retina; the photoreceptors are supplied via choroidal circulation, but the central retinal artery supplies the inner nuclear layer. Thus, the b-wave amplitude is reduced but the a-wave is relatively preserved. In eyes with central retinal vein occlusion (CRVO), a negative ERG or delay in the 30-Hz flicker response suggests significant ischemia.

The ERG can also be helpful in determining the carrier state of individuals with X-linked disease. For example, carriers of X-linked RP usually have abnormal ERG findings that reflect lyonization, even with a healthy-appearing fundus. However, in choroideremia, carriers usually exhibit a normal ERG response despite an abnormal fundus appearance (also resulting from lyonization).

Electroretinography is suitable for use with children of all ages, providing objective functional data for patients who may not be able to describe their symptoms. In addition, the ERG may reveal retinal abnormalities prior to the development of fundus abnormalities. For young subjects, sedation, general anesthesia, or eyelid electrodes may be used. The latter are usually well tolerated and require neither sedation nor anesthesia. Interpretation of pediatric ERGs involves special considerations. Adult ERG values are not reached until 6–9 months of age, and, if general anesthesia is used, the effect of the anesthetic on the ERG must be considered.

• Johnson MA, Marcus S, Elman MJ, McPhee TJ. Neovascularization in central retinal vein occlusion: electroretinographic findings. Arch Ophthalmol. 1988;106(3):348–352.

• Vincent A, Robson AG, Holder GE. Pathognomonic (diagnostic) ERGs. A review and update. Retina. 2013;33(1):5–12.

Excerpted from BCSC 2020-2021 series: Section 10 - Glaucoma. For more information and to purchase the entire series, please visit https://www.aao.org/bcsc.