Electroretinographic electrodes: general introduction

Today, many designs of electroretinographic (ERG) electrodes are available, including contact lens, gold foil, gold wire, corneal wick, wire loop, microfiber, and skin electrodes. The clinical ERG is obtained with an electrode placed at some distance from the neural elements that are producing the signals of interest. The electrical current induces an electric field within and around the eyeball and orbit, with both spatial and temporal variations.³⁴ The ERG signals are conducted from their retinal generator sites through various tissues to the surface electrode. Each electrode type has its own characteristic impedance, recording characteristics, and inherent artifacts. Clinical electroretinographers should be aware of electrode characteristics and should carefully choose the electrode type that is most practical for their recording situation. Typically, an ionic bridge is used to establish an electrically conductive medium between the metallic electrode and the surface of the eye or skin.

Artifacts during ERG recording can almost always be traced to the electrode system.^25,71 Characteristically, these electrode-related recording artifacts fall into one of three categories: those related to electrode polarization, those related to electrode slippage or movement, and the photovoltaic artifact. Care must be taken so that the active ERG electrode and its reference electrode are of the same metallic type. When two electrodes are made from the same metal, the potential difference between them is usually around zero, but slight impurities in the metal and possible surface contamination can cause differences in electrical potential. Potential differences between metals can be very large in comparison with the magnitude of the ERG activity measured at the cornea or surface of the skin.

Fortunately, when a.c.-coupled amplifiers are used, these steady potentials are blocked by the coupling capacitors in the input circuit, thus reducing the effect of electrode potentials. However, motion artifacts are more troublesome. But when d.c.-coupled amplification is used, the electrode potential artifacts are very important; they may cause amplifier blocking and can give rise to baseline drift that may be reduced only by using more stable electrodes. Electrode potentials can be minimized by careful preparation, by ensuring that all electrodes are of the same metal and by avoiding contamination of the electrode surfaces. Eye movement and electrode movement are two sources of potential artifact that can greatly affect the quality of ERG recordings. Electrode movement or uneven fitting of contact lens electrodes on the cornea can produce artifacts in the ERG recording. Unfortunately, ERGs recorded from any electrode can be contaminated by eye movement artifact. Techniques for the digital subtraction of eye movement artifact from flash ERGs¹⁹ as well as pattern ERGs²⁰ have been described and found useful for obtaining pure ERG tracings. An additional source of artifact occurs when photic stimulation strikes the electrode surface and generates a photovoltaic signal that appears as a spike early in the ERG recordings. Fortunately, photovoltaic artifacts can be dealt with quite easily by shielding the electrode surface from the light source.

Contact Lens Electrodes

The use of contact lens electrodes for clinical ERG recording was first described by Lorrin Riggs, who developed a clear, nonirritating lens that could be fitted to the subject's eye for prolonged ERG recording sessions.⁵⁹ Several variants of contact lens electrodes have been developed over the years.^{13,37,38,63,65,66} Because of their ability to give reliable and reproducible recordings,⁴⁹ contact lens electrodes are the recommended standard for clinical ERG recordings to flash stimulation. However, they are not recommended for pattern ERG recording, as they can degrade image quality on the retina.⁷

The most frequently used contact lens electrodes are the Burian-Allen (Hansen Ophthalmic Laboratories, Iowa City) and Henkes Lovac (Medical Workshop, Groningen, Holland) assemblies as illustrated in figures 17.1A and 17.1B.^13,29 Recently, the Goldlens (Diagnosys LLC, Littleton, MA) has enjoyed popularity as a well-designed low-noise contact lens electrode (see figure 17.1C). The Burian-Allen assembly makes use of a large speculum that holds the eyelids apart and contacts the scleral surface. A smaller clear corneal contact lens is held against the cornea with a spring assembly. The force exerted on the cornea by the spring mechanism of the Burian-Allen speculum contact lens has been measured at 10g.⁶⁹

Figure 17.1.  

Contact lens electrode assemblies for clinical electroretinography. Electrodes illustrated are the Burian-Allen (A), Henkes Lovac (B), Goldlens (C), and ERG Jet (D).

A circular silver wire around the circumference of the contact lens makes the actual contact with the cornea and provides an active electrode. A reference electrode is formed by a coating composed of silver granules in polymerized plastic within the surface of the scleral speculum for bipolar ERG recordings. For monopolar ERG recordings, a forehead or indifferent ear electrode reference is used.

Although uncomfortable, the Burian-Allen electrode assembly can be used for a period of several hours, though a session of no more than 30 minutes is recommended. Disadvantages of this lens include the possibilities of corneal abrasion, conjunctival abrasion, and irritation produced by movement of the lens assembly. Although frank abrasions are uncommon, they do occur, and minor trauma can be seen in some cases if the patient's cornea is stained by fluorescein. In cooperative patients, corneal staining is often directly proportional to the skill of the examiner in inserting and removing the lenses. A 1% methylcellulose solution drop on the contact lens prior to insertion helps to protect the cornea. Naturally, contact lens electrodes are poorly tolerated by young children, who often require sedation for ERG recording.

Glass contact lens electrodes employing a dome or cup containing isotonic saline and methylcellulose have been reported.^{29,33,64–66} In 1951, Henkes described an electrode that enjoys popularity today among clinical electroretinographers.²⁹ This corneal contact lens electrode maintains electrical contact with the surface of the cornea through a dome containing isotonic saline and methylcellulose. A low vacuum is maintained by suction, and this ensures a good electrical contact with the surface of the eye.

While the Burian-Allen and Henkes corneal contact lens electrodes have similar impedance and recording characteristics, a significant difference in the degree of susceptibility to corneal injury has been demonstrated in diabetic patients. Vey et al. investigated a series of 57 diabetic patients who underwent standard ERG recordings with subsequent slit-lamp examination utilizing fluorescein strips and cobalt blue light.⁶⁹ Twenty-eight patients were examined with the Burian-Allen speculum-type corneal electrode, and 29 patients had ERG recordings performed by using the Henkes bipolar low-vacuum corneal electrode. Subsequent biomicroscopy with fluorescein demonstrated that over 30% of the patients who were examined by using the Burian-Allen electrode demonstrated disruption of corneal epithelium, whereas corneal changes were observed in only 7% of the patients who were tested with the Henkes electrode assembly. The authors concluded that owing to the abnormal susceptibility to corneal injury displayed by diabetics, the Henkes low-vacuum electrode was their recommendation for standard ERG testing in these patients. However, it should be noted that many investigators have not found significant corneal changes from the use of Burian-Allen lenses when they are placed carefully and not left in for excessive amounts of time. Recently, there has been a reported study of the use of contact lens electrodes for recording the ERG in extremely small immature preterm infants within the nursery setting.⁴⁷ By using infant monkey size 4 bipolar contact lens electrodes (Hansen Ophthalmic Laboratories), flash ERGs were recorded in seven infants with corresponding conceptional ages of 32–34 weeks. The authors concluded that ERG recording using contact lens electrodes in preterm infants is a safe and practicable procedure.

The use of a soft contact lens under a hard lens to cushion and disperse the direct pressure on the cornea has been advocated, and comparable recording characteristics have been demonstrated.^10,23 Schoessler and Jones have described a soft hydrophilic contact lens electrode with excellent recording characteristics.⁶² This electrode was formed by a fine gold or platinum wire sandwiched between two soft contact lenses. These investigators claimed that their recording electrode was more comfortable and stable than hard contact lens electrodes and provided minimum obstruction to vision. A comparison of the recording characteristics of the standard Burian-Allen speculum-type electrode assembly and a soft contact lens electrode revealed that while the amplitude of the b-wave recorded with the soft contact lens assembly was comparable, the signals were less stable over recording sessions lasting several hours.¹⁷ This lack of recording stability was evidenced by increased high-frequency noise superimposed over the ERG waveform. Rehydration of the soft contact lens system with 0.9% saline only temporarily reduced the high-frequency noise artifact. While the soft lenses were more comfortable to wear, they required frequent rehydration, since the outer soft lens was not in contact with the moistened conjunctiva and would rapidly dehydrate. As the outer lens desiccated, it began to deform and allow air to enter the interlenticular space, thereby producing an increasingly noisy ERG recording.¹⁷ The soft lens electrode assembly was less stable and did not remain centered on the cornea. In addition, the hydrogel lens sandwich was too delicate to be useful in clinical ERG recording situations. Other attempts at hydrogel lens construction have been more successful.⁹ The ERG-Jet (Universo S.A. La Chaux-De-Fonds, Switzerland) gold foil corneal contact lens electrodes have been advocated, as they are light and relatively inexpensive and are disposable following testing. The ERG-Jet electrode is illustrated in figure 17.1D. The recording characteristics compare favorably with the Burian-Allen system.²⁶ However, a photoelectric artifact with longer-duration light flashes has been described.²⁵ As with all corneal lenses, topical anesthesia was required. While all corneal electrodes generally give excellent-quality ERG waveforms in cooperative adults, with a few exceptions,⁴⁸ they may be poorly tolerated by younger children, who may require restraint or sedation. A simple modification of the standard ERG-Jet contact lens recording electrode for use in infants and small children was developed by Brodie and Breton.¹² A small Plexiglas cylinder was fixed to the front surface of the electrode, thereby preventing lid closure and facilitating insertion and removal. The authors reported the use of this modified ERG-Jet electrode in over 400 patients and claimed that with its use, there was seldom need to use anesthesia or oral sedation in their pediatric population.

Lid-Hook Electrodes

In addition to corneal contact lens electrodes for recording ERGs, there are circumstances in which it is preferable to use other ERG recording electrodes. An electrode that employed a polyethylene film (Mylar) strip coated on one side with aluminum and bent into a J shape was first reported to produce high-quality ERG recordings without the use of topical anesthetic.¹⁵ Unfortunately, the aluminum coating was less than ideal because it tended to unbond from the Mylar surface at low levels of alternating current.^4,11 A gold-coated Mylar electrode was first described in which the gold surface did not unplate and with which excellent ERGs were obtained.¹¹ The curved tail of the J shape rode on the cornea and produced some changes in the corneal epithelium of 35% of the patients who were tested.

A low-mass, inexpensive gold foil electrode (C. H. Electrodes, Bromley, Kent) described by Arden and associates⁴ was constructed from gold foil applied to Mylar (see figure 17.2A) and provided ERGs that were very similar to those obtained by more conventional contact lens electrodes. The gold foil electrode (GFE) was very flexible and, when inserted into the lower fornix and bent to lie on the cheek, barely touched the corneal margin. A junction wire was connected to the gold foil electrode and led to an insulated standard electrode wire that was also taped against the cheek. The uninsulated junction wire was not allowed to touch the skin. Although initial corneal anesthetic was recommended in the past, current practice is to use no anesthetic. In any case, once the anesthetic wears off, no additional topical anesthetic is usually required. Arden reported that the GFE produced slightly smaller responses than the Karpe lens did and had some higher-frequency components superimposed on the ERG waveform.⁴ One problem with gold foil electrodes is their flexibility, in that they may shift or fall out during testing, especially in elderly patients with lid laxity, and if they are uncomfortable to the patient, tearing may effectively short the electrode and give no signal. The GFE is better tolerated by patients than standard contact lens electrode assemblies are, and this makes it possible to more easily record ERGs using the GFE on young children with congenitally malformed or narrow palpebral fissures as well as patients immediately following cataract or corneal surgery. While gold foil electrodes would appear to be less injurious to the cornea than contact lens electrodes, a small proportion of patients still suffer transient symptoms, including blurred vision, ocular discomfort, and tearing, which occasionally persists. Aylward et al. described the corneal changes in a consecutive series of 50 normal subjects undergoing electroretinography with gold foil electrodes.⁶ Transient corneal changes were observed in 31 subjects (62%), which included punctuate epithelial keratitis, corneal erosions, and stromal thinning. There was a significant association between the severity of corneal changes and the subject's age and the use of topical anesthetic. Because topical anesthetics typically reduce the frequency of blinking and reduce the amount of lacrimation, these may be significant factors in relation to the corneal changes.

Figure 17.2.  

Lid-hook, microconductive fiber, and skin electrodes used in clinical electroretinography.

The fact that the GFE does not interfere with the optics of the eye makes it ideal for pattern ERG recording.⁷ There have been numerous reported studies in which gold foil electrodes have been compared to contact lens systems.^24,26,50,53 Gjotterberg compared several contact lens electrodes with the gold foil lid-hook assembly in 11 healthy eyes recorded on two occasions.²⁶ It was reported that 9 of the 11 patients preferred the GFE to the Burian-Allen or low-vacuum contact lens electrodes, although these latter electrodes produced larger-amplitude responses to scotopic flash with smaller intertest amplitude variability and therefore were deemed better for research protocols. Other disadvantages of the GFE included the fact that it was easily blinked out, and the junction was fragile, so after several usages, the electrode became noisy. If it was placed more than 15mm from the medial fornix, the voltage recorded declined.

Since the pattern ERG is a small evoked response, the question of an electrode's intrasessional variability and its test-retest reliability is important. Odom et al. reported a two-center study of the intrasessional variability of the pattern ERG using gold foil recording electrodes.⁵⁶ The coefficient of variation and the coefficient of repeatability were used as measures of intrasession variability or precision. They observed that for pattern ERGs recorded using gold foil electrodes, the intrasession pattern ERG reliability was 0.95 or higher, and intrasession coefficients of variability could be 0.05 or less.

Several investigators have looked at the test-retest reliability of the GFE in recording the pattern ERG.^5,57,58 Prager in two studies reported that while gold foil electrodes produced larger pattern ERG amplitudes than microconductive threads, their test-retest reliability was low.^57,58 These initial results,⁵⁷ which were repeated in a three-center study, showed that the reduction in pattern ERG amplitude and an increase in variability occurred when recordings were made with used GFEs.⁵⁸ For new gold foil electrodes, pattern ERGs were recorded with higher amplitudes than electrodes used three times in 90% of the trials. Electrodes that were used three times demonstrated an average change in the coefficient of variation of 14%. At two of these study sites, test-retest pattern ERGs on a total of 18 patients demonstrated a test-retest coefficient of variation of 30% and 47% for new and used electrodes, respectively. It was deemed that these sources of variation could have resulted from the presence of small cracks, the number and configurations varying in each electrode. Light microscopy demonstrated these small cracks, and it was proposed that constant flexion and variation would result in significantly different impedances over time. Thus, the more an electrode is used, the lower is the pattern ERG amplitude, and the resulting poorer test-retest reliability would occur. The authors recommended that gold foil electrodes be used only a single time.⁵⁸ These findings were in stark contrast to those reported in a two-center study of GFE test-retest reliability.⁵ When similar gold foil electrodes were inserted in the same fashion but without using topical anesthesia, pattern ERGs were recorded after either the first or the third or fourth use in four normal subjects. The test-retest interval on the normal subjects was five to six working days, and the same electrodes were used on routine clinical patients in the interim. In these trials, the GFEs were extensively examined before and after the tenth use. The authors reported that in the results from three separate experiments, the P50 and N95 components of the transient pattern ERG do not alter significantly with electrode use. Interestingly, in some cases, the pattern ERG amplitude increased, although not significantly, with electrode use. The investigators were unable to replicate the results described by Prager et al.^57,58 In addition, they were unable to observe any cracks in the gold surface of the electrode similar to those reported by Prager et al.^57,58 Arden et al. suggested that the main source of variation was likely to be either subject related or the technical expertise of the tester rather than GFE aging.⁵ Wong and Graham examined qualitative defects in gold foil electrodes and resistances in 94 used gold foil electrodes.⁷¹ They divided the electrodes into four groups of varying resistances and gold coating defects. In addition, they measured photopic flash ERGs in a single subject with GFEs randomly selected from each group. They reported no significant difference among electrode groups for ERG peak implicit times or amplitudes, although a slightly greater amplitude variability was observed for electrode groups with more defects. Their data suggested that whether or not to use a GFE depends not on the degree of visible defects on the electrode, but rather on the overall resistance of the electrode.

Gold foil electrodes are also suitable for multifocal electroretinography. Mohidin et al. recorded multifocal electroretinograms (mERGs) from 12 subjects on three separate days to investigate the repeatability and variability of the mERG using four different electrodes.⁵³ They reported a coefficient of variation of 0.19 for the GFE, which was similar to the coefficient of variation of 0.21 reported by Prager et al. for the pattern ERG.⁵⁷ Prager's findings were in contrast to a two-center study reported by Odom et al.,⁵⁶ who reported a coefficient of variation as low as 0.05 for the gold foil electrode. They claimed that their low coefficient of variation was attributable to stringent technical control and use by experienced experimenters.

Other materials have been used to develop lid-hook electrodes, including carbon fiber and polyvinyl (PV) gel. Honda et al. developed a disposable electrode for ERG recording that was made from anomalous polyvinyl alcohol (PVA) gel.³² This new PVA hydrogel electrode was made of low-cost material that could be discarded after use. The electrode was cut in the shape of a lid hook and could be inserted in the lower fornix like the gold foil lid-hook electrode. The PVA electrode was considered to be very stable without electrical polarization because it contains no metal. It was claimed that patients felt no discomfort during recordings and that the electrodes produced no corneal injury.^18,32

The C-glide (Unimed Electrode Supplies, Farnham, Surrey, UK) is a carbon fiber lid-hook electrode that has been popular among some for ERG recording for some time (see figure 17.2B).⁸ Unlike the extremely flexible gold foil lid-hook electrodes, the C-glide electrode is more rigid in construction, and it has been reported to require longer insertion time.⁵⁰ This greater rigidity may be an advantage in that the C-glide electrode may be more difficult to blink out of the eye than the highly flexible gold foil lid-hook electrode. Esakowitz et al. compared the C-glide carbon fiber electrode with other standard contact lenses, GFE, DTL, and skin electrodes for recording scotopic and photopic electroretinograms.²⁴ They reported that the coarse recording tip of the C-glide electrode and its preshaped plastic hook inevitably touched the eyelashes and made it the most unpopular of the eye contact electrodes tested, whereas the GFE was well tolerated and relatively simple to insert. When compared to the Burian-Allen electrode, the b-wave amplitudes produced by the C-glide were approximately 77%. The gold foil electrode produced b-wave amplitudes that were only 56% as great as those produced by the Burian-Allen contact lens electrode.²⁴ McCulloch et al. compared a number of electrodes including the Burian-Allen contact lens, the C-glide, and GFE in recording pattern ERGs.⁵⁰ They reported that the Burian-Allen recorded the largest amplitude pattern ERGs, whereas the C-glide and GFE gave significantly better within-session quality of recordings. From their studies, they found no compelling reason to recommend a particular type of foil or fiber electrode for pattern ERG recording, as each electrode type had advantages and disadvantages.⁵⁰ The C-glide electrode gave the largest amplitude but had the longest time for insertion and a significantly lower comfort rating than the GFE. Mohidin et al. examined the repeatability and variability of the C-glide electrode compared to several other electrode types and found that the C-glide electrode produced a significantly larger coefficient of variation of 0.31 compared to the contact lens (0.15) and GFE (0.19) when recording multifocal ERGs.⁵³

A novel disposable ERG electrode was developed by Hiroi et al.³¹ consisting of a thin gold filament 0.12mm in diameter. The gold wire could be bent and placed in the lower fornix or touching the cornea. They reported electrode performance equal to that of the ERG-jet for flash and pattern ERG recording.³¹ Another novel and popular noncorneal electrode for clinical ERG was developed that consists of a thin wire forming a loop modeled to fit in the lower conjunctival sac (see figure 17.2C).²⁷ The HK-loop electrode (HK Med, Ljubljana, Slovenia) is formed from a loop of thin stranded or monofilament silver, gold, or platinum wire that is Teflon coated over its entire length except for three small windows, which allow exposed portions of the metal. The loop is bent so that it appears U-shaped in side view, similar to a lid-hook electrode, and this portion is placed in the lower conjunctival sac, making electrical contact with the sclera, while the Teflon insulation provided shielding from unwanted extraneous potentials. The HK-loop is adaptable to unusual anatomic configurations of the patient's eye or eyelids. The ERGs that are obtained with the HK-loop electrode are similar in amplitude to those recorded with gold foil electrodes.^27,50

DTL Fiber Electrodes

Dawson, Trick, and Litzkow (DTL) described a very low-mass, silver-impregnated microfiber corneal electrode for clinical electroretinography.²¹ This DTL electrode was based on an extremely low-mass conductive thread that makes contact with the tear film of the eye and is electrically coupled to an insulated wire. The individual fibers of the nylon thread are approximately 50µm in diameter and are impregnated with metallic silver. The thread is usually draped in the lower fornix, although alternative methods using a holder to position the electrode across the eye have been described.^52,68 Simultaneous flash ERGs recorded from DTL fiber and Burian-Allen electrodes have been compared, and the DTL b-wave amplitudes are generally lower than the corresponding responses recorded with the Burian-Allen corneal contact lenses.^22,23,25,51

The advantages of the DTL electrode lie in the area of subject comfort, optical quality, and reduced electrode impedance.²³ The DTL system is well tolerated by children and by adults with keratitis. Like the gold foil lid-hook electrode, the DTL does not obscure the optics of the eye and therefore is superior for recording the pattern ERG. It is not blinked out of the eye like the gold foil electrode, but this may not be an advantage, for if the fiber is displaced toward the lower fornix, the signal size is reduced.⁵² Hebert et al.²⁸ examined the reproducibility of flash ERGs recorded with DTL electrodes. Subjects were tested on two occasions separated by an interval of 7–14 days, and Naka-Rushton parameters were derived from DTL-recorded ERG responses. The Naka-Rushton parameters Vmax, the maximum b-wave amplitude observed at the saturation point of the luminance curve, and log K, the intensity necessary to produce the semisaturation of Vmax amplitude, were examined. It was found that there was a high test-retest intraclass coefficient of r = 0.97 for Vmax and r = 0.91 for log K in their subjects. The authors concluded that the DTL electrode yields stable and reproducible ERG recordings, provided that care is taken to ascertain proper electrode positioning.

While DTL electrodes have been found to yield highly reproducible signals for flash and pattern ERG, they were thought to represent an interesting challenge to record reproducible oscillatory potential signals.⁴⁶ Suprathreshold photopic oscillatory potentials were recorded with either a DTL electrode or a Henkes corneal contact lens electrode. While the summed oscillatory potential amplitude index recorded with the DTL electrode was one half of that obtained with the Henkes electrode, the timing and amplitude ratios were consistent for all the oscillatory potential peaks.⁴⁶ Other investigators have also found that DTL electrodes provide robust and reproducible ERG recordings compared to corneal contact lenses and gold foil electrodes for flash ERG recording,^8,24,30,45 pattern electroretinograms,^50,57,58 oscillatory potentials,⁴⁶ and multifocal electroretinograms.^51,53

Recently, the DTL electrode has been modified, and it is now commercially available (DTL-Plus, Diagnosys LLC, Littleton, Massachusetts) and is illustrated in figure 17.2D. The DTL-Plus electrode is composed of 7-cm-long, low-mass spun nylon fibers impregnated with metallic silver. Small sponge pads at each end with adhesive backing are secured to the nasal and temporal canthi, securing the electrode and positioning the microconductive thread at the limbus.

Skin Electrodes

The possibility of recording ERGs in response to flash and pattern stimulation without placing electrodes in the eye has specific advantages, especially in the testing of children and infants. Although skin electrodes are in general not recommended as active recording electrodes, there are situations in which there is no other method of obtaining a reliable ERG from a patient. ERGs recorded through the periorbital skin surface have been described.^{1,18,35,36,61,63,67} Tepas and Armington used skin electrodes placed on the nasal and temporal canthi and reported that averaged ERGs could be reliably recorded in a wide range of stimulus conditions.⁶⁷ Larger-amplitude ERGs have been reported with subsequent head rotation to displace the cornea toward the active electrode. The ERG signals that were produced were smaller in amplitude, noisier, less reliable, and more variable than corneal ERG recording. Signal averaging can significantly improve the signal-to-noise ratio of ERGs recorded with skin electrodes. Standardized placement of the skin electrodes is important to an individual laboratory's ability to give reliable interpretation when using this technique. The dermal electrodes (see figure 17.2E) are better tolerated, and the recording quality may be acceptable for many clinical recording situations in infants and young children, particularly if the patient will not allow testing with a corneal contact lens.⁵¹

While skin electrodes have been used for ERG recording for some time, there are few reports that directly compare their performance against standard contact lens assemblies.^24,44,50 The advocates of the use of skin electrodes for ERG recording have primarily been those experienced who are in pediatric electroretinography. While most, though not all, adults will accept insertion and prolonged wearing of contact lens electrodes, the majority of infants, toddlers, and young preschool children are by no means impassive and will not willingly allow insertion and wearing of contact lenses. Typically, restraint with forcible electrode placement is necessary, or electrode placement is performed while the child is sedated or anesthetized. It is in the infant population with apparently poor vision that clinical electroretinography probably plays its most crucial diagnostic role, and many pediatric electrophysiologists have turned to the skin electrode, which has allowed them to record reliable ERG recordings.⁴⁴ Kriss reported the great Ormond Street experience of over 4000 recordings using skin ERGs in a wide variety of ophthalmological and neuro-ophthalmological conditions.⁴⁴ In this study, four healthy young adults had ERG recordings from one eye using six electrode types, including skin electrodes. The researchers reported that for flash ERGs, the b-wave amplitude of skin ERGs was about 14% that recorded to a Burian-Allen electrode.^24,44 As has been previously described, several ocular and technical factors can affect skin ERGs, but the authors concluded that, provided that one recognizes these confounding factors and deals with them accordingly, skin ERGs can provide reliable and diagnostically useful information in young children.⁴⁴

Coupland and Janaky had previously compared the performance of dermal electrodes against the popular DTL fiber and the PVA hydrogel electrode in 32 eyes,¹⁸ where ERGs were simultaneously recorded with skin electrodes along with either DTL or PVA gel electrodes under four standardized flash conditions. The use of simultaneous skin ERG recordings ensured that any differences observed were not due to adaptation, endogenous, or environmental factors. Under all four stimulus conditions, the skin ERGs had consistently shorter a- and b-wave implicit times. This was attributed to the positioning of the skin electrode on the infraorbital ridge, the active electrode being closer to the posterior pole of the eye. Generally, the skin ERGs were about half as large in amplitude as the averaged DTL or PVA ERG recordings.

Because of the advantages of signal-averaging techniques in increasing the signal-to-noise ratio, skin electrodes have been used in detecting retinal oscillatory potentials⁶¹ as well as in recording the pattern ERG.^1,35,36 Wali and Leguire compared skin and ERG-Jet electrodes over a wider range of luminances and compared the b-wave amplitudes for each electrode fitted by the Naka-Rushton equation.⁷⁰ As expected, Vmax for the skin electrode was smaller than that for the corneal electrode. The value of log K differed by 0.3, with the skin electrode giving a lower value of log K. There was no significant difference between the values of n for the two electrodes. A comparison of electrodes in terms of general variability as measured by the coefficients of variation showed that while the skin electrode yielded less b-wave implicit time variability, the corneal contact lens yielded less b-wave amplitude variability.⁷⁰ While skin electrodes have worked well in experienced hands, the problem of increased variability and a lower signal-to-noise ratio in ERG recording has led the ERG standardization committee of ISCEV to recommend that they should be used only in exceptional circumstances.