Chapter 8 Visual electrophysiology
how it can help you and your patient
Pediatric visual electrophysiology can be a challenge, but provides information about the working of retina and visual pathways that we cannot achieve by other means. This functional assessment helps us with early diagnosis, prognosis, and an objective means of monitoring neurologic and ocular sequelae.
There are international guidelines and standards for performing visual electrophysiologic tests (e.g. ISCEV, the International Society for Clinical Electrophysiology of Vision, available at http://www.iscev.org, or International Federation of Clinical Neurophysiology at http://www.ifcn.info). We apply, and extend, ISCEV adult protocols in able children, i.e. children who can sit still and follow instructions for 30 minutes or more. With younger, or less compliant, children we use adapted protocols that are robust enough to provide comparable information without restraint, sedation, or anesthesia.
As children have short attention spans we may need to use distraction to encourage reproducible results. We need to be flexible and responsive during the test to adapt the protocol, and the order of tests within a protocol. This may be prompted by ongoing analysis and interpretation, or a change in compliance. This enables us to meet the needs of a child in a way they enjoy, yet answer the clinical question in a time efficient way! The overarching aim is to minimize stress and anxiety to child, carer, and staff. This optimizes results and enhances our chance of reliable future monitoring.
Clinical visual electrophysiologic tests include the electro-oculogram (EOG) and the electroretinogram (ERG), which assess the function of the retinal pigment epithelium (RPE) and retina, and the visual evoked potential (VEP), which assesses the integrity of the postretinal pathways to the striate visual cortex particularly the macular pathways. The retinotopic representation of the macula on the gyri of the occipital lobes is most accessible to surface VEP electrodes.
Behavioral compliance is the limiting factor in pediatric visual electrophysiology testing. To record an EOG the child will have to sit still and make saccades every minute for 15 minutes in the dark then 15 minutes in the light; for the pattern ERG (PERG) or multifocal ERG (mfERG) they need steady fixation with good focus and for an ISCEV ERG they will need to dark adapt for 20 minutes and light adapt for 10 minutes. This may be possible for exceptional youngsters, but it is more likely from ages 5 years upwards. Our adaptive protocol may be applied from birth onwards and aims for a total chair time of 30 minutes during which time pattern VEPs (PVEPs) and flash ERGs are carried out contemporaneously with flash VEPs.
To illustrate this we have applied our GOSH (Great Ormond Street Hospital) protocol to two common questions arising at different ages. The flow chart outlines the diagnostic algorithm and hierarchy of testing (Fig. 8.1). We have added short notes on the technical aspects of the methodology at the end of the chapter. Artefacts can mimic physiologic responses and must be excluded before findings from complementary tests may be interpreted as consistent (see Fig. 8.1).
Fig. 8.1 All testing can start with pattern reversal 50′ checks presented to both eyes. Depending upon the response we may proceed to smaller reversing checks and monocular testing, or divert to pattern onset stimulation. After pattern stimulation is completed flash stimulation is used to contemporaneously record ERGs and VEPs. Transoccipital asymmetries are noted throughout and explored in all three stimulus modalities and when possible with half field stimulation. With suspected voluntary defocus the PERG is recorded simultaneously with the pattern reversal VEP. This robust combination strategy can be used to investigate diverse clinical questions as the two examples outline:
The EOG is used to investigate if a maculopathy is due to retinal pigment epitheliopathy. The standing potential of the eye due to voltage across apical and basal surfaces of the RPE is around 6 millivolts, with maximal positivity detected at the corneal apex. Electrodes placed on the medial and lateral canthi measure a large potential change during a saccade: the electrode closest to the cornea becomes positive relative to the electrode furthest from the cornea. This is the EOG and it is displayed as a voltage/time plot. Eye movements, including nystagmus, can be characterized graphically. The EOG potential increases in light and decreases in the dark.1 With reproducible eye movements, e.g. saccades of known size, this variation can be measured and expressed as the Arden ratio (light rise/dark trough) and values below 1.6–1.8 are abnormal. The voltage change with light is a consequence of the phagocytosis of outer segment discs and transport of retinal binding proteins at the apical end of the RPE. If the photoreceptors are sick, both the ERG and the EOG are affected. The EOG is diagnostically potent when the Arden ratio is abnormal and the rod ERG is normal as this discriminates a primary retinal condition from an epitheliopathy, e.g. bestrophin mutations (see Chapter 45).2
The ERG is used to distinguish cone and rod dysfunction and photoreceptor from inner retinal dysfunction. The ERG is measured in microvolts and its size and shape depend upon the relative proportion and extent of rods and cones that are excited and the size of the retinal area stimulated.3 Rods and cones can be preferentially stimulated by flashes of different colors, strength, and duration presented under different states of dark and light adaptation. The gradual evolution of the ERG waveform with increasing flash strength is shown in Fig. 8.2A scotopically and Fig. 8.2B photopically.
Fig. 8.2 Scotopic (A) and photopic (B) luminance response series. ERGs to ISCEV standard flash strengths are shown in red. (A) from bottom to top shows first the development of the rod driven b-wave and then the a-wave as flash strength increases. (B) shows the photopic hill phenomenon with smaller b-wave amplitude to higher flash strengths.
To very dim lights, a small scotopic threshold response (STR) has been described but clinically this is difficult to achieve and is used rarely. As flash strength increases a late (60 ms), round, positive b-wave emerges. This is the rod driven b-wave, which reflects inner retinal activity associated mainly with depolarizing on-bipolar cells. The change of b-wave amplitude with flash strength can be described by a Naka-Rushton function, derived from the Michaelis-Menton equation, but the derived parameters will vary according to the method of curve fitting. In clinical circumstances these need to be interpreted with care.4
As the flash strength further increases an early negative a-wave precedes the b-wave. The a-wave becomes larger and faster with increasing flash strength reflecting photoreceptor hyperpolarization. To the brightest flashes the leading edge of the scotopic a-wave models rod phototransduction.5
4. The d-wave – an off pathway response. Rods use the on pathway through the inner retina; cones use both on and off pathways. The d-wave is associated with decreases in light under photopic conditions, and is best seen in response to prolonged on–off flashes (on > 90 ms). This is an important extra stimulus for investigating “negative,” no b-wave ERGs, e.g. subtyping CSNB (see Chapter 44).8 Usually b- and d-waves are superimposed in the ERGs to short duration (< 10 ms) flashes.