S. Parks

Peripheral retinal dysfunction in patients taking vigabatrin

Objective: To assess the wide-field multifocal electroretinogram (WF-mfERG) for assessment of retinal function in vigabatrin-treated patients. Methods: Thirty-two adults who had taken vigabatrin for at least 3 years for localization-related epilepsy underwent WF-mfERG, ERG, logMar visual acuity and color vision evaluation, Humphrey visual field analysis (static perimetry), and funduscopy. The group was matched with a cohort of patients who had never received vigabatrin. Results were compared with a normative data set (120 drug-free controls) with respect to potential bilateral abnormalities in response timing. Results: There were no significant differences between groups in visual acuity or color vision testing. Of the vigabatrin patients, 19 (59%) had bilateral visual field defects compared to none of the controls. Using WF-mfERG, all patients on vigabatrin with visual field defects showed abnormalities (100% sensitivity) and only 2 of the 13 patients without a field defect showed retinal abnormalities (86% specificity). Conclusions: WF-mfERG may be useful for detecting retinal pathology in patients taking vigabatrin. The majority of previous reports based on subjective testing may have underestimated the prevalence of peripheral retinal toxicity related to the drug.

Technical aspects of multifocal ERG recording

There are a wide range of variables which can influence the quality of the multifocal response. It is possible to place these variables into one of four categories. First, the method of stimulus delivery will determine the field of view, interference levels and the duration of on-state stimulation. Second, data acquisition variables such as electrode type and placement, amplifier specifications and filter bandwidth settings will have a direct impact on waveform shape and on the topographic distribution of signal amplitudes. Third, patient variables such as fixation, pupil dilation and refractive error will also contribute to the multifocal response. Fourth, there are many measurements that can be taken from multifocal recordings. In addition to standard amplitude and implicit time measures (the implicit time measure in the multifocal response is becoming increasingly important particularly in early stages of disease processes), the scalar product measure provides information on waveform shape. The conventional impulse and higher order responses will be different for different modes of stimulation such as Cathode Ray Tube (CRT) and Liquid Crystal Display (LCD) systems and latency shifts will be introduced if not corrected in software. Procedures which could lead to misleading interpretation include artefact rejection, averaging with neighbours and summing of responses. These procedures should be handled with caution.

The multifocal ERG: unmasked by selective cross-correlation

The purpose of this paper is to provide the reader with a better insight into the mechanisms of multifocal ERG (mfERG) recording. The construction of the first and second order mfERG responses were examined by recovering the response to specific pulse trains embedded in the m-sequence.A custom built pc based multifocal system driving a LED stimulator was used to record a 61 element mfERG and a global ERG. The global ERG recording was used to enable the recovery of different pulse trains embedded in the m-sequence. Summation of these individual pulse trains was performed and the results compared with the standard full cross-correlation. An isolated pulse response is defined as a flash of light that has no other flashes within two m-sequence base periods before or after the flash. This isolated pulse response was recovered from the raw data and this response input into a simple superposition model to predict the waveform shape for specific pulse trains. The superposition model was compared with the actual selective cross-correlation for a particular pulse train. The summations of the selective cross-correlation components give identical responses to the full cross-correlation. The superposition model also predicts the waveform shapes recovered by the selective cross-correlation procedure. The mfERG response is a complex composite response from a number of different pulse trains. Examination of the individual waveform shapes provides some insight into the origin of the mfERG waveform. The main contributions to the P1 component are the same as for an isolated response and as with the standard ERG this component is likely to be dominated by the mid retina. The N1 component is also likely to have similar origins to that of the isolated response but the amplitude is dominated by contributions from pulse trains where there is no change of state and therefore includes a component from the interaction between two consecutive stimuli. The N2 component is a composite response dominated by the interaction between two successive stimuli two base periods apart and the P1 component of a second stimulus delayed one frame from the first stimulus.

A comparison of CRT and Digital stimulus delivery methods in the multifocal ERG

The purpose of this paper is to compare and evaluate the multifocal ERG response from raster based CRT and Digital Projection (LCD) stimulus delivery systems. A custom built p.c. based multifocal system was used to generate a 61 hexagonal element stimulus array. The stimulus was presented on a high luminance CRT display and on a back projected screen using a Digital polysilicon projection system. A fast response photodiode was used to analyse the stimulus pulse characteristics of both systems. A number of recordings were performed to assess the effect of stimulus delivery on a standard m-sequence response, inserted full-field filler response and on separation of onset and offset components. The pulse width for a CRT system is dependant on the type of phosphor and is typically 2 msec whereas the Digital Projection system produces a 13.3 msec pulse equivalent to the frame rate for the system. Slowing down the m-sequence by a factor of eight results in a pulse width of 106 msec which should enable the recovery of true offset responses. The CRT stimulus consists of a series of eight pulses of 2 msec duration each separated by 11.3 msec. First order responses are larger from the CRT system and second order responses larger from the Digital system. In conclusion, there are fundamental differences in the two delivery systems. The CRT system may have more potential in examining non-linear aspects of the multifocal response. Although both systems may be able to record offset responses, the Digital system will generate true offset responses whereas the CRT system may not allow true separation of these components.

The impact of fixation on the multifocal electroretinogram

There are a number of variables which can influence the quality of multifocal ERG waveforms. In common with visual field measurements, fixation quality may be an important parameter on the integrity of the acquired data. A low cost, fixation-monitoring device was used to assess fixation quality on a group of normal volunteers. Data was successfully acquired while five subjects viewed a fixation target for a period of time equal to that of a single multifocal recording segment. The target was presented on a stationary grey background and as the central fixation mark on a 61-element multifocal flicker stimulus. The results show no significant difference with or without the flickering pattern. The percentage of samples falling within 1.2° of the point of fixation was 51%. This suggests that fixation quality is adequate for scaled stimuli where the central element subtends 2.4°. High resolution stimuli of less than 2.4° may be more susceptible to fixation fluctuations during the recording process.

Binasal visual field defects are not specific to vigabatrin

This study investigated the visual defects associated with the antiepileptic drug vigabatrin (VGB). Two hundred four people with epilepsy were grouped on the basis of antiepileptic drug therapy (current, previous, or no exposure to VGB). Groups were matched with respect to age, gender, and seizure frequency. All patients underwent objective assessment of electrophysiological function (wide-field multifocal electroretinography) and conventional visual field testing (static perimetry). Bilateral visual field constriction was observed in 59% of patients currently taking VGB, 43% of patients who previously took VGB, and 24% of patients with no exposure to VGB. Assessment of retinal function revealed abnormal responses in 48% of current VGB users and 22% of prior VGB users, but in none of the patients without previous exposure to VGB. Bilateral visual field abnormalities are common in the treated epilepsy population, irrespective of drug history. Assessment by conventional static perimetry may neither be sufficiently sensitive nor specific to reliably identify retinal toxicity associated with VGB.

Wide field multifocal and standard full field electroretinographic features of hemi retinal vein occlusion.

The purpose of this study was to document the standard full field electroretinographic (ERG) and wide field multifocal electroretinographic (WF-mfERG) findings in eyes with recent onset hemi-retinal vein occlusion (HRVO) and to compare the electro-diagnostic findings in the affected and fellow eyes with reference to normative data. Eight patients with HRVO were assessed using ERG and WF-mfERG. WF-mfERG first order responses from the affected hemi-retinae and the unaffected hemi-retinae in each affected eye were compared. WF-mfERG responses from each affected hemi-retina and from the symmetrical hemi-retina of each fellow eye were compared. ERG responses between affected and unaffected eyes were also compared. All electrodiagnostic tests were compared to normative data (5–95% confidence limits derived from age-related controls). WF-mfERG P1 and N1 implicit times were greater for the affected hemi-retinae than for the unaffected hemi-retinae (p <0.05). WF-mfERG N1 and P1 implicit times were prolonged (p < 0.05) and WF-mfERG P1/N1 amplitude ratios were significantly reduced (p < 0.05) for the affected eyes when compared with the fellow eyes. Maximal b-wave, cone b-wave and flicker implicit times were prolonged (p < 0.05) when comparing affected and fellow eyes. These results indicate that retinal injury due to HRVO culminates in significant delay of both ERG and WF-mfERG implicit times. These results suggest that WF-mfERG in combination with ERG may have a role in the management of HRVO.

An instrument to investigate temporal processing mechanisms with the multifocal ERG

The multifocal ERG technique is a powerful method of studying the function of different areas of the retina. Display systems such as the CRT, which are commonly used for stimulation, are subject to limitations such as those imposed by the raster method of scanning. This work describes a novel stimulating display using LEDs that retains the established hexagonal areas but overcomes some of the limitations of the CRT display systems. The design and construction of the instrument is described together with some preliminary results.

Three-dimensional electromagnetic model of the human eye: advances towards the optimisation of electroretinographic signal detection.

Classical electromagnetic theory is used to examine the topographical variation in electrical potentials over the corneal surface resulting from specific retinal stimuli. Results from a three-dimensional mathematical model show that over 97% of calculated electromagnetic field potentials lie within 3% of previous analytical model data for an axially symmetric case. Maps of corneal potentials are produced that are shown to be characteristic of specific retinal stimuli and location. The maximum variation in corneal potential for a full field global stimulus is found to be approximately 1%. This is considered encouraging, as current electrophysiology techniques measure ocular potentials from a single corneal or scleral site, the position of which is often difficult to localise and reproduce. The model is used to simulate both central and peripheral stimuli and scotoma conditions. A 20° central scotoma simulation shows an overall reduction in central corneal potential of only 3%, whereas peripheral stimuli are found to cause up to 10% variations in this potential. There is therefore a possibility that a single recording site for multifocal retinal stimulation is not ideal. These data may be used to suggest more appropriate electrode recording positions for maximum signal recovery, not least in optimising signal detection for multi-focal electroretinography stimulation.

Identification of appropriate primitive polynomials to avoid cross-contamination in multifocal electroretinogram responses

The basis of the multifocal/electroretinogram is the use of a decimated m-sequence for simultaneous and independent stimulation of many areas of the visual pathway. The purpose of this study was to investigate the effects of cross-contamination from higher orders of the response. A series of primitive polynomials were found by construction of finite fields. The first-order ERG response was formed by cross-correlation of m-sequence with the physiological response. A second-order response was formed by investigation of particular flash sequences of the stimulation sequence and cross-correlation of a second-order m-sequence with the physiological response. Zech logarithms were used to identify cross-contamination between the various first and second-order sequences. Tables of good and bad primitive polynomials were constructed for degrees 12–16, and the effects of window length and decimation length were examined. When the sequence was decimated into 128 areas, and a window of length 16 was examined, cross-contamination occurred in all sequences generated from primitive polynomials of degree less than or equal to 12, but in only 26% of degree 14, and 5.6% of degree 16. A photodiode (artificial eye) was used in an experiment to construct trace arrays showing responses from 61 individual areas. Additional waveforms were present on the trace array when the experiment was carried out with a bad primitive polynomial. The use of finite field theory to generate primitive polynomials and Zech logarithm analysis allowed prediction of which primitive polynomials were suitable for m-sequence generation for multifocal electroretinography. Practical investigations supported the theoretical analysis. This has important implications for developers of multifocal electrophysiology systems.