ISCEV guide to visual electrodiagnostic procedures

The ISCEV standard full-field ERGs (Fig. 1a) are global responses of the retina to brief flashes of light and provide an assessment of generalized retinal function under light- and dark-adapted conditions. A ganzfeld (German for “whole field”) stimulator, which provides a uniformly illuminated field, is used to deliver a range of flash stimuli that evenly illuminate the maximal area of retina. The ERGs are recorded with electrodes in contact with the cornea or conjunctiva or with skin electrodes attached to the lower eyelids. Several types of corneal electrode may be used including contact lens, fiber, jet and gold foil electrodes. The pupils are dilated to maximize retinal illumination and to minimize inter-subject and inter-visit variability. Reliable interpretation of recordings requires comparison with electrode-specific and age-matched normative data. The normal test–retest variability of ERG parameters is also an important consideration if used to monitor disease progression or the safety or efficacy of treatments.

The ISCEV standard protocol includes dark-adapted (DA) recordings after 20-min dark adaptation to flash strengths of 0.01, 3.0 and 10.0 cd s m⁻² (DA 0.01; DA 3.0; DA 10.0). The weak flash (DA 0.01) ERG arises in the inner retinal rod bipolar cells and is the only standard test that selectively monitors rod system function. Abnormality of the DA 0.01 ERG can be caused by either rod photoreceptor dysfunction or selective dysfunction occurring post-phototransduction or at the level of the inner retinal rod bipolar cells. The DA 3.0 (standard flash) and DA 10.0 (strong flash) ERGs have input from both rod and cone systems, but the DA rod system contribution dominates in a normal retina. Approximately the first 8 ms of the cornea-negative a-wave reflects rod hyperpolarizations, and as the a-wave in the DA 10.0 ERG is of shorter peak time and larger than in the DA 3.0 ERG, it provides a better measure of rod photoreceptor function. The subsequent cornea-positive b-wave arises largely in the rod On-bipolar cells and reflects function that is post-phototransduction. Thus, the DA strong flash ERG enables localization of dysfunction to the rod photoreceptors (a-wave reduction and concomitant b-wave reduction) or to a level that is post-phototransduction or inner retinal (sparing of the a-wave with b-wave reduction). The DA oscillatory potentials (OPs) are small high-frequency components normally visible on the rising limb of the DA 3.0 and DA 10.0 ERG b-waves and are thought to reflect amacrine cell signaling. Reduction in the OPs is often associated with other ERG abnormalities but may occur selectively in some disorders. The cone system contribution to both DA ERG a- and b-waves is minor in a normal retina but can be of greater significance in patients with disease primarily or exclusively affecting the rod system.

Standard light-adapted (LA) ERGs provide two measures of generalized cone system function; both are obtained to a flash strength of 3.0 cd s m⁻², after a standard period of 10-min light adaptation in the Ganzfeld with a constant background luminance of 30 cd m⁻². A 30 Hz flash stimulus, superimposed on the background, is used to elicit the LA 30 Hz flicker ERG, generated largely by post-receptoral retinal structures. The single flash cone (LA 3.0) ERG consists mainly of a- and b-waves. The LA 3.0 ERG a-wave arises in the cone photoreceptors and Off-bipolar cells; the b-wave is dominated by a combination of cone On- and Off-bipolar cell activity, and a reduced b/a ratio suggests cone system dysfunction that is post-phototransduction or post-receptoral.

The full-field ERG enables the distinction between generalized outer and inner retinal dysfunction and predominant rod or cone system dysfunction. Symptoms and/or clinical signs may suggest a retinopathy, but the presence, severity and nature of retinal dysfunction cannot always be inferred from the clinical findings and ERGs can help differentiate between a wide range of disorders when appropriately placed in clinical context (see below and Table 1). It is stressed that the full-field ERG is largely generated by the retinal periphery and there is minimal contribution from the macula. Electrophysiological assessment of macular function requires the use of different techniques such as the pattern ERG or multifocal ERG.

The ISCEV standard PERG is derived largely from the macular retinal ganglion cells and complements the full-field ERG, in differentiating between maculopathy and generalized retinopathy. PERG also enables a more meaningful evaluation of a VEP, to exclude a macular cause of VEP abnormality and to provide an additional assessment of retinal ganglion cell involvement (see below). The PERG is recorded to an alternating high-contrast checkerboard using a corneal electrode. PERGs are attenuated by poor refraction and ocular media opacity, and care must be taken to optimize the optical quality of the checkerboard stimulus; for this reason, contact lens electrodes are not suitable.

The transient PERG has two major components of diagnostic value: a positive polarity P50 and a negative polarity N95 (Figs. 1a and 2). Both components reflect macular retinal ganglion cell function, but there is an additional more distal retinal contribution to the P50 component. Both P50 and N95 depend on the function of the macular cones, and P50 reduction and/or delay can characterize macular dysfunction. Selective reduction in N95 with preservation of P50 suggests dysfunction at the level of the retinal ganglion cells. In severe or chronic retinal ganglion cell dysfunction, there may be P50 reduction, but in such circumstances P50 usually shortens in peak time, reflecting loss of the retinal ganglion cell contribution to P50. Preservation of P50 helps to establish the effective stimulus quality and contrast of the checkerboard in patients who may have poor visual acuity for reasons other than maculopathy. Comparison of responses to a standard and additional large-field stimulus may help characterize the area of macular dysfunction, although spatial resolution is lower than for the mfERG.

The ISCEV standard mfERG (Fig. 3a) provides a measure of cone system function over 61 or 103 discrete hexagonal retinal areas, within the central 40°–50° of the posterior pole centered on the macula. The hexagons of the ISCEV standard stimulus array are scaled to elicit comparable response amplitudes from each stimulus region, resulting in larger hexagons with increasing eccentricity. Reliable recording requires good patient fixation, and corneal electrodes are required as signals are small.

Each hexagonal stimulus element is modulated rapidly to display white or black frames according to an irregular but predetermined binary sequence known as a “pseudorandom” or “m-sequence.” The signal associated with a particular hexagon is extracted from a single continuous recording from each eye, using automated cross-correlation analysis. The responses can be mathematically stratified into components associated with single illumination events (the first-order kernel), used for ISCEV standard testing. The optical quality of the stimulus is important, and patients should be optimally refracted and must fixate accurately on a central target or cross-hairs throughout the recording period. There is a compromise between spatial resolution (smaller, more numerous hexagons) and the recording duration necessary to obtain responses with a satisfactory signal-to-noise ratio. There are two major response components; an early negative polarity N1 component is derived from cone bipolar cells with a cone photoreceptor contribution and a later positive polarity P1 component that arises in cone bipolar cells.

The spatial resolution of the mfERG is better than for the PERG and full-field ERGs, and this enables improved characterization of focal central, annular, hemifield or discrete paracentral areas of posterior pole dysfunction, but reliable recording requires good patient fixation. If the area of dysfunction shows reasonably good radial symmetry, interpretation may be facilitated by averaging waveforms from all the hexagons in each concentric ring in the stimulus pattern (ring-averaging). Illustrative examples of mfERG recordings are shown in a case of retinitis pigmentosa (RP) with central macular sparing (Fig. 3b), in macular dystrophy (Fig. 3c) and in a patient with an enlarged blind spot (Fig. 3d). The mfERG is also a useful adjunct to the VEP and is less affected by optical factors than the PERG; there is no retinal ganglion cell contribution to the mfERG, and a normal response excludes primary macular photoreceptor dysfunction as cause of VEP abnormality or central visual loss. However, in some conditions such as cystoid macular edema (CME), the mfERG may be preserved or less affected than the PERG.

The ISCEV standard EOG is used to assess generalized retinal pigment epithelium (RPE) function. There is a potential difference between the apical and basal surfaces of the RPE that results in a dipole across the eye, with the cornea being positive with respect to the back of the eye. This potential difference, the standing potential of the eye, is recorded using skin electrodes placed at the medial and lateral canthus of each eye during uniform 30-degree horizontal saccades, made periodically during dark and then light adaptation. During the standard 15-min period of dark adaptation, there is a fall in the recorded standing potential, typically reaching a minimum at 10–15 min, referred to as the dark trough (DT). The dark phase is followed by a 15-min period of continuous light adaptation to a standard white background (100 cd m⁻²), provided by a Ganzfeld stimulator. Following light onset, there is an increase from the standing potential resulting in the EOG light peak (LP). The LP/DT ratio (Arden ratio) provides a measure of the generalized function of the RPE/photoreceptor complex. The development of a normal EOG light peak requires not only a normally functioning RPE, but also normally functioning rod photoreceptors, with the degree of EOG abnormality broadly corresponding to the degree of rod-derived ERG abnormality. An EOG assessment of generalized RPE/photoreceptor function is most useful when interpreted in the context of normal or only mildly subnormal rod-mediated ERG findings. In the presence of severe rod dysfunction from any cause, the EOG will be abnormal, and not additionally informative about the function of the RPE. Common causes of generalized RPE dysfunction are outlined below.

The ISCEV standard VEPs provide an important objective test in the investigation of suspected optic nerve disease or post-retinal visual pathway dysfunction. The VEPs are electrical potentials recorded from the scalp derived from electrical currents generated in the visual cortex in response to visual stimulation. The VEP indicates the function of the entire visual pathway from the retina to area V1 of the visual cortex and primarily reflects the central retinal projection to the occipital poles. Recording electrodes are positioned on the scalp according to anatomical landmarks using a standardized “International 10/20 system” measurement method. The recording montage includes at least one occipital electrode (Oz) referred to a mid-frontal reference (Fz). Computerized signal averaging is used to extract the time-locked VEP from spontaneous brain activity (the electroencephalogram or EEG).

The ISCEV standard for VEP testing describes three stimulus modalities: pattern-reversal, pattern onset–offset and diffuse flash stimulation. A reversing checkerboard is used to record the pattern-reversal VEP, generally most useful for the assessment of optic nerve function, but requiring an adequate level of fixation and compliance. The normal pattern-reversal VEP has a prominent positive component at approximately 100 ms (P100; Fig. 2), although normal ranges differ and are age and laboratory dependent. Pattern onset–offset (pattern appearance) stimulation is less commonly used in the diagnosis of optic nerve disease than pattern reversal, but has the advantage of being less affected by nystagmus. Flash VEPs are generally less sensitive to dysfunction than pattern VEPs, but may be used in young children or when patients cannot fixate or comply with testing. They are also useful in the presence of media opacity when the use of stronger non-standard flashes may be helpful to establish the integrity of the visual pathway. There is wider variability in normal ranges than for pattern VEPs, and an inter-ocular comparison is often most useful. Flash VEPs may occasionally reveal abnormalities in the presence of normal pattern VEPs, and this can occur in rare cases of optic neuritis, in some cases of optic nerve sheath pathology or due to unsuspected retinopathy.

Multichannel VEPs, in excess of the current ISCEV standard, are needed to detect optic nerve misrouting or to detect and characterize chiasmal or retrochiasmal dysfunction. Multichannel flash VEPs can also reveal the visual pathway misrouting associated with albinism in children, but flash VEPs are usually normal in adults with albinism.

The timing, amplitude and waveform shape of the P100 component are used to evaluate pattern-reversal VEPs, which provide an important objective test in the investigation of suspected optic nerve disease or post-retinal visual pathway dysfunction. However, abnormalities are not specific and can reflect, for example, optic nerve or macular dysfunction and can also be caused by poor compliance or sub-optimal refraction. Reliable interpretation of pattern VEP abnormality usually requires complementary assessment to exclude a macular cause. Similarly, a flash ERG may exclude a retinal cause of flash VEP abnormality. There are numerous causes of optic nerve disease, and VEPs may suggest or support a suspected diagnosis when interpreted in clinical context. Common causes of optic neuropathy are outlined below.

Visual acuity loss may be caused by inherited and acquired causes of maculopathy (with or without retinopathy), optic nerve and visual pathway disease, but this may not be obvious on clinical grounds alone and the distinction is enabled by electrophysiological testing.

Night blindness (nyctalopia) can result from generalized rod system dysfunction, and this may be confirmed or excluded using a full-field ERG. The DA 3.0 and DA 10.0 ERGs enable localization of dysfunction to the rod photoreceptors (a-wave reduction and concomitant b-wave reduction) or to a level that is post-phototransduction or inner retinal (sparing of the a-wave; b-wave reduction).

Photophobia is commonly associated with generalized cone system dysfunction and can be an early symptom in cone and cone-rod dystrophies. In cone dystrophies, the LA 30 Hz and LA 3.0 ERGs show delay and/or amplitude reduction, and in cone-rod dystrophy, there is additional abnormality of the DA ERGs (Fig. 1c). In both conditions, there is usually severe macular dysfunction evident on PERG (Fig. 1c) or mfERG testing. Rod monochromacy (achromatopsia) is characterized by severe cone system dysfunction from early infancy; the LA ERGs are typically undetectable, but the DA 0.01 ERG, selective for the rod system, is normal, and the DA 10.0 ERG is normal or shows mild reduction in the a- and b-waves, due to a loss of the normal dark-adapted cone system contribution. The ERG findings in S-cone monochromacy (a form of “X-linked incomplete achromatopsia”) are similar, but DA ERGs may be additionally attenuated due to high myopia; there may be a markedly abnormal (but detectable) LA 3.0 ERG and the short-wavelength (“blue”) flash ERG is relatively preserved. Congenital photophobia may also be a feature of albinism. Photophobia is rarely caused by dysfunction confined to the macula. Acquired causes of photophobia include retinal inflammatory disease such as uveitis and birdshot retinochoroidopathy (BRC), both associated with a high incidence of generalized cone system dysfunction, AIR and CAR. Photophobia is a rare feature of optic nerve disease but can also occur in neurological disorders such as migraine, meningitis and in carotid artery or vertebral artery disease.

Peripheral visual field constriction is a common feature of rod-cone dystrophy (RP), and this can occur without classical intraretinal pigment deposition, particularly in children. Cone and cone-rod dystrophies may present with visual field defects including central scotomata, generalized depression of sensitivity, ring scotomata and peripheral field loss if there is relative sparing of central macular function. Peripheral visual field loss may also occur in inflammatory retinal disorders such as BRC, associated with variable retinal dysfunction but often characterized by generalized cone system dysfunction, manifest as delay in the LA 30 Hz ERG, and sometimes associated with additional inner retinal rod system involvement (reduction in DA 10.0 ERG b:a ratio) which may be reversible following treatment (Fig. 1g, h). In acute zonal occult outer retinopathy (AZOOR), there is usually field loss disproportionate to visible fundus changes and persistent photopsia within the scotoma. Full-field ERG abnormalities are common, and some may show a reduction in the EOG light peak-to-dark trough ratio, not explained by abnormalities in rod function. Autoimmune disorders, such as CAR and AIR, may also present with rapid visual field constriction and marked ERG abnormality (see above). Homonymous hemianopic visual field defects usually reflect chiasmal or retrochiasmal brain lesions, and these may be detected by multichannel VEP recordings and require prompt further investigation. Field loss may also be seen in shallow retinal detachments and retinoschisis with concomitant full-field ERG changes, and clinical or ultrasound eye examination is essential.

Disk pallor may be a feature of optic neuropathy or retinopathy, including cone and cone-rod dystrophies. In central retinal artery occlusion (CRAO), there may be unilateral retinal edema and a “cherry red” spot at the fovea in the acute phase, but after a few weeks, this resolves as disk pallor develops. The subacute and chronic phases may be mistaken for ischemic optic neuropathy, and the electrophysiology enables the distinction. The ERG in CRAO has an electronegative DA 3.0 or DA 10.0 ERG, and there is usually marked involvement of the LA ERGs, in keeping with generalized inner retinal dysfunction. There are several other potential masquerades of optic neuropathy including occult maculopathy (inherited or acquired) and central serous chorioretinopathy (CSR); both may manifest PERG P50 or central mfERG abnormalities. In acute idiopathic blind spot syndrome (AIBSS), the mfERG may characterize the nasal area of reduced function (Fig. 3d). Occult retinopathy (including AZOOR), autoimmune and paraneoplastic retinopathies typically show marked ERG abnormalities, and in posterior scleritis, which like optic neuritis may be accompanied by pain on eye movement, there may be inflammatory changes affecting the retina that cause ERG abnormality.

Glaucoma is a progressive optic neuropathy associated with injury to retinal ganglion cell axons, frequently due to elevated intraocular pressure. Common signs include a characteristic pattern of optic atrophy (enlargement of the optic nerve cup), sectoral nerve fiber layer defects, often best visible with red-free light and evident on optical coherence tomography (OCT). There are often characteristic visual field defects, including arcuate “nerve fiber bundle defects” which reflect the distribution of optic nerve fibers emanating from the optic nerve, and “nasal steps” at the horizontal raphe.

The pattern ERG is sensitive to macular ganglion cell dysfunction and nerve fiber layer loss in glaucoma and can be of value in the evaluation of “glaucoma suspects” with glaucomatous risk factors such as elevated intraocular pressure, or optic nerve head changes, prior to the measureable loss of visual field. There may be reduction in the N95 (and also the P50) component in transient recordings, but steady-state PERG recordings are more affected. Traditional full-field ERG parameters, such as a-wave and b-wave amplitudes, are insensitive to ganglion cell injury, but there is increasing interest in the photopic negative response (PhNR). This is a late, cornea-negative deflection in the full-field ERG which is often recorded to red flashes presented on a blue background. The PhNR reflects global retinal ganglion cell function and offers the possibility of detecting and monitoring glaucomatous progression. ISCEV standard multifocal ERGs (first-order kernels) are driven primarily by photoreceptor and bipolar cells and are thus relatively insensitive to ganglion cell damage, although subtle effects of glaucoma have been described in the second-order kernels or with special stimulation paradigms. Multifocal recording technology has also been adapted to produce low-resolution visual field-like maps of VEP responses to spatial stimuli for eccentricities out to approximately 20° (e.g., dartboards), although standardization and clinical utility have yet to be established.

Congenital nystagmus is a feature of several ocular and neurological disorders. Isolated idiopathic congenital motor nystagmus (CMN) is not associated with other ocular or neurological abnormalities, and although pattern-reversal VEP and PERG may be difficult or impossible to record due to eye movements, flash VEPs and full-field ERGs are normal. Common retinal causes of nystagmus include Leber congenital amaurosis (LCA), associated with severe generalized photoreceptor dysfunction (DA and LA ERGs are severely reduced or undetectable), cone and cone-rod dystrophy, rod and S-cone monochromacy and complete and incomplete CSNB, characterized by different ERG phenotypes (see above). Nystagmus is also associated with ocular and oculo-cutaneous albinism (see above), and diagnosis in the former may be difficult in the absence of obvious skin depigmentation.

Acquired nystagmus may result from drug toxicity or medication that impairs the function of the labyrinth, thiamine or vitamin B12 deficiency, head injury, stroke, multiple sclerosis or any disease or injury of the brain that affects neural centers that control eye movements. Exclusion of afferent visual pathway dysfunction with electrophysiology may provide an important contribution to the management of such cases.

The full-field ERG is sensitive to retinal ischemic disorders affecting the inner retina. There may be reductions in the DA 3.0 and DA 10.0 ERG b:a ratios, the DA oscillatory potentials are usually abnormal or extinguished, and LA 30 Hz ERGs show prolonged peak times and waveform distortions. The ERG may be invaluable in detecting ischemic central retinal vein occlusion (CRVO), progression of non-ischemic to ischemic CRVO and in the diagnosis of ocular ischemic syndrome especially when the carotid Doppler scans are normal or equivocal. The ERG has advantages over commonly used fluorescein angiography in being safe and noninvasive, providing information on deeper layers and peripheral areas of retinal blood supply and may be informative in patients with systemic co-morbidities or pregnancy, in patients allergic to fluorescein dye or in cases of vitreous hemorrhage obscuring the fundus view. Prolonged LA 30 Hz ERG peak times are frequently seen in diabetic retinopathy and are associated with increased risk of disease progression. Peak time delays can be useful for screening, and loss of oscillatory potentials can occur in some diabetic patients without diabetic retinopathy and may identify patients at increased risk.

Full-field ERGs and flash VEPs can provide valuable information in patients with suspected retinal or visual pathway disease when the fundus is obscured or when the use of retinal imaging techniques is precluded by an opaque ocular media. Integrity of retinal and visual pathway may be important considerations prior to treating patients with corneal lesions, cataracts or vitreous hemorrhage, particularly if there is a history of retinal detachment, retinal or neurological involvement. A normal or relatively preserved ERG or flash VEP may suggest a better prognosis for improved vision. An abnormal full-field ERG may suggest generalized retinal dysfunction but may also occur in vitreous hemorrhage. An abnormal ERG does not exclude central visual recovery because it does not assess macular function. It is noted that the ERG is usually abnormal in the presence of intraocular silicone oil tamponade (for retinal detachment), but interpretation is confounded because the oil impedes conduction of the electrical signals from the retina to the corneal surface.

Visually asymptomatic patients with a family history of retinal or optic nerve disease or suspected cases of syndromic retinal dystrophy may require electrophysiological testing for evidence of subclinical disease. For example, visually asymptomatic obligate carriers of X-linked RP usually manifest abnormal and asymmetrical ERG abnormalities, irrespective of whether there is fundus abnormality, whereas the ERGs in carriers of X-linked choroideremia are usually normal until late in life. Carriers of X-linked ocular albinism and patients with rubella retinopathy may also have abnormal fundus pigmentation; the ERGs are normal in the former and normal or near-normal in the latter. There is variable expressivity in (autosomal dominant) Best disease such that some heterozygotes have a normal fundus and an EOG may be needed to confirm the diagnosis. Similarly, VEP and PERG N95 abnormalities may indicate optic nerve and retinal ganglion cell dysfunction in cases of suspected dominant optic atrophy.

Serial testing may assist the distinction between stationary and progressive conditions, important for diagnosis and patient counseling. Pattern and flash VEPs have diverse applications and may be used to monitor visual pathway maturation in infants with poor vision or amblyopia or to monitor optic nerve function in patients with known neurological disease. In inflammatory retinal diseases such as BRC, the ERGs can be used to monitor efficacy of treatment objectively (Fig. 1g, h), thus informing clinical management and titration of potentially toxic medication. Worsening VEPs may prompt the need for surgical intervention in dysthyroid eye disease or in neurological disorders, irrespective of stable neuroradiology. Several medications commonly administered systemically for non-ocular conditions are potentially toxic to the macula, retina or optic nerves, and pre-treatment assessment and monitoring may be considered. The multifocal ERG, for example, may reveal annular or parafoveal macular dysfunction that can manifest as an early stage of hydroxychloroquine toxicity, before the development of a visible “bulls-eye” lesions and before structural changes are evident or obvious on retinal imaging. Intraocular drugs, intraoperative dyes and bright lights of ophthalmic surgical equipment have become another source of toxic/phototoxic maculopathy that may need retinal and macular electrophysiology testing for monitoring, for clinical evaluation or for diagnosis. ERG evaluations are also becoming an integral part of various clinical trials comparing outcome efficacies of various surgical or medical procedures involving the macula such as macular holes, epiretinal membranes, anti-VEGF treatments, macular detachments and central serous chorioretinopathy. Similarly, ERGs may be used to monitor retinal safety of new treatments and as objective outcome measures in clinical trials that aim to restore visual function or arrest disease progression.

Absent or impaired visually mediated behavior may indicate a disorder affecting any level of the visual system. Babies who do not fix and follow and presumed amblyopic patients that fail to respond to treatment may require testing to confirm or exclude pathology. Early diagnosis of retinal dystrophy may be essential to identify young candidates who are potentially amenable to future experimental treatments. A normal ERG may also prompt the need for further investigations such as VEPs or neuroradiology. Non-organic visual loss is relatively common in older children, and in such circumstances, the electrophysiological data are usually normal even though there may be reported profound visual loss.

The differential diagnosis includes several retinal disorders such as Leber congenital amaurosis, congenital stationary night blindness, and rod and S-cone monochromacy. The ERG will help differentiate these conditions. Young children with albinism show multichannel flash VEP evidence visual pathway misrouting, although with increasing age (above about 5 years) this may be best demonstrated with pattern onset–offset VEPs. Flash VEPs and ERGs are normal in idiopathic CMN. Clinical examination is also needed to investigate or exclude TORCH infections like viral retinitis that result in nystagmus and variable ERG abnormalities.

The ERG may be helpful in advising families with patients at risk of hereditary retinal disorders. The extent to which the various retinal dystrophies are detectable in early infancy is frequently not known, but a normal ERG at age 7 or 8 years of age largely excludes X-linked RP. Night blindness may be associated with RP or CSNB and ERGs help differentiate between progressive and stationary disorders. In young cases of suspected Best disease, children may be unable to comply with EOG testing, but testing of the parents will almost invariably identify the parent carrying the mutation, irrespective of whether the fundi are normal.

Perinatal infections, particularly the “TORCH” agents, may attack ocular tissues, with possible profound associated dysfunction. The most common perinatal infection is probably rubella retinopathy, which frequently results in mottled RPE pigmentation ad a “salt and pepper” appearance, but in such cases the ERG is usually normal or near-normal.

Perinatal brain damage may lead to severe visual impairment. VEPs enable objective determination of the nature of the deficit and may help grade the severity of cortical dysfunction. However, it is important to recognize that VEPs do not reflect higher processing required for normal vision.

In children who have suffered head/orbital trauma or suspected visual pathway injury, complementary retinal and VEP testing may localize dysfunction and help to confirm, exclude or distinguish between retinopathy and optic nerve or post-retinal dysfunction, particularly in those unable or too young to communicate verbally.

Infants often present with “visual indifference,” showing little or no reaction to visual stimuli for several months. If the eye examination, ERG, and VEP are normal or near-normal, this provides reassurance, and the prognosis for development of normal or near-normal vision is reasonably good.

The most common indication in this category is vigabatrin, which is used for the treatment of infantile spasms (West syndrome). The drug causes peripheral visual field constriction in approximately 30% of adults. The ERG is helpful in monitoring patients who are too young or lack the ability to perform visual field testing.

Children suspected of having amblyopia are often referred for electrophysiology to exclude other causes of poor vision, for example when visual acuity has not improved with patching and the fundi are normal or when visual acuity is reduced bilaterally. In amblyopic eyes, pattern-reversal VEPs may show amplitude reduction; delays in the major positive (P100) component can occur, but this tends to be more prominent in strabismic rather than anisometropic amblyopia. Pattern VEPs may also be used to monitor the efficacy of occlusion therapy in amblyopic and fellow eyes, but subjective assessment of vision (if possible) should generally take priority.

Fundus photography has been available as a clinical tool since 1926, and fluorescein angiography was introduced in 1959. More recently, advances in fundus imaging have appeared with increasing frequency, not only documenting ophthalmoscopic findings, but extending the range of clinical perception in depth (ICG angiography) and resolution; spectral domain OCT now approaches the resolution of low-power microscopy, without the need to remove tissue from the eye for histologic processing. However, the enhanced capability of fundus imaging has not displaced electrophysiological methods of testing function. The need to complement anatomical methods with studies of visual function is as keen as ever and perhaps more so as increasing detail in fundus imaging allows ever finer diagnostic distinctions to be made, for which the functional consequences must be determined.

Electrophysiology has a pivotal role to characterize disorders and the phenotypic variability associated with a known genotype or to guide the screening of genes associated with a known electrophysiological phenotype. Advances in molecular biology have enabled genotyping of many inherited retinal and macular dystrophies, but the functional consequences remain difficult to predict due to allelic heterogeneity, genetic modifiers and other factors. In rare retinal dystrophies, ERGs can be used to identify the gene responsible, e.g., in enhanced S-cone syndrome (NR2E3), “cone dystrophy with supernormal ERG” (KCNV2) and RGS9/R9AP-retinopathy, as outlined in Table 1. It is more usual for the ERGs to suggest a range of disorders or possible genotypes, e.g., in complete CSNB, the ERG phenotype is common to X-linked and autosomal recessive forms with mutations in 1 of several different genes and ERGs are additionally identical to those in melanoma-associated retinopathy, highlighting the importance of interpretation in clinical context. The emergence of unbiased whole exome and whole genome sequencing may reveal novel or unexpected genetic alterations and electrophysiological interrogation likely to prove increasingly important to establish the functional consequences and genotype–phenotype correlations.