Clinical display of mfERG data.

PURPOSE: Many mfERG displays show normal responses that are larger at the center than peripherally, and the typical linear display of signals is inaccurate with respect to the retinal location of the signals. Printouts do not always indicate retinal or field view, they sometimes emphasize 3-D topographic plots which are not always representative of physiologic signals, and they show ring response densities which are different in every ring and hard to interpret without norms. These problems limit the clinical usefulness of the mfERG and limit communication in the literature. We share our Stanford Display to illustrate possible solutions to these problems. METHODS: We have changed the scaling factor for our mfERG unit to produce a trace array with near equal signals everywhere. We display responses is a spatially scaled array, in a retinal view, so that signals appear in their correct anatomic locations relative to a fundus image. The 3-D display is minimized on the page of signal analysis, and we emphasize ring response averages rather than ring response densities. RESULTS: The new scaling and trace array display greatly facilitate the analysis of retinal disease. Regions of loss are easily recognized in their fundus location. Ring ratios based upon response amplitudes all have a normal value of 1.0 which simplifies analysis. A case of early hydroxychloroquine retinopathy demonstrates the use of this Stanford display. CONCLUSIONS: Recognition of these recording and display options may help mfERG users to maximize the value of the test. Proper scaling of the mfERG stimulus array facilitates localization of retinal disease and simplifies ring response analysis. Different laboratories will have different priorities for signal analysis, but mfERG displays should always indicate the eccentricity of responses, and the use of a retina or field view.

Bacterial contamination of ocular surface and needles in patients undergoing intravitreal injections.

PURPOSE: To evaluate potential sources of bacterial contamination during intravitreal (IVT) injection procedures. METHODS: Patients scheduled for IVT injection were asked to enroll in the study at the California Vitreoretinal Center (Menlo Park, CA) and the Vantage Eye Center (Salinas, CA) between October 2004 and April 2005. A total of 104 patients participated in the study, with a total of 118 IVT injection procedures performed on 107 eyes. Standard microbiological techniques were used to culture, identify, and quantify bacterial contamination of injection needles and the bulbar conjunctiva at the injection site in patients undergoing IVT injections. The main outcomes measured were type and quantity of bacterial isolates. RESULTS: Two (2%) of 114 needles collected were contaminated with bacteria. The prevalence of bacterial contamination of the injection site on the bulbar conjunctiva was 43% before prophylaxis on the day of the injection with topical antibiotics and povidone-iodine, with a statistically significant reduction to 13% after prophylaxis (P < 0.0001). Coagulase-negative Staphylococcus, the most common bacterium isolated from the ocular surface, was isolated from both culture-positive needles. CONCLUSIONS: IVT injection needles became contaminated with bacteria during the injection procedure. Although the contamination rate was low, this supports a mechanism of postinjection endophthalmitis in which there is direct inoculation of ocular surface flora into the vitreous cavity by the injection needle.

Utility in clinical practice of standard vs. high-intensity ERG a-waves.

PURPOSE: Standard ERG a-waves represent contributions from both photoreceptor and inner retinal cells, while the leading edge of the high-intensity a-wave is produced only by photoreceptors. This has raised questions about the value of the a-wave as an indicator of photoreceptor disease, and has led to suggestions for standardizing higher-intensity stimuli. Our objective was to compare the behavior of standard and high-intensity a-waves in clinical practice. METHODS: Standard ISCEV (International Society for Clinical Electrophysiology of Vision) a-waves and high-intensity a-wave responses were recorded under scotopic and photopic conditions from normal subjects and from patients with photoreceptor dystrophies and other diseases. RESULTS: The standard scotopic a-wave amplitude followed the high-intensity a-wave closely among patients with different diagnoses, and the results did not change significantly when cone a-waves were subtracted to isolate rod signals. The only exception was one patient with the enhanced S cone syndrome (ESCS) whose dark-adapted responses were cone-driven. Initial peak times clustered in a small range for both standard and high-intensity responses, and were not very sensitive to disease. CONCLUSION: High-intensity a-waves can show photoreceptor characteristics directly, and may help analyze some rare disorders. However, in our study the amplitude of conventional scotopic a-waves mirrored that of the high-intensity responses quite closely over a wide range of patients. This suggests that for practical purposes even if it is not perfect, the standard ERG is an excellent indicator of photoreceptor disease.

Are circadian variations in the electroretinogram evident on routine testing?

We sought to determine whether routine ERGs using ISCEV standard stimuli, would show a pattern of circadian variation. We examined ERGs from 40 successive normal subjects who were tested at different times during regular laboratory operating hours of 9 am to 4 pm, and also reviewed high intensity a-waves from a subgroup. There were no obvious associations of either ERG amplitude or implicit time with time of day. No statistically significant difference was found between average ISCEV ERG parameters or high-intensity a-wave parameters obtained in the morning (9 am to 1 pm) and afternoon (1 pm to 4 pm). We conclude that time of day is not critical for routine ERG recordings, although small, variable, circadian changes may well be present. We suggest that the time of day be noted on clinical recordings, in case this information becomes relevant for a particular patient.

Clinical S-cone ERG recording with a commercial hand-held full-field stimulator.

Our purpose was to explore S-cone ERG protocols for a commercial full-field hand-held stimulator that contains colored LEDs, and to see whether the test would be useful as a part of routine ERG testing. S-cone responses were elicited by blue flashes over a longer-wavelength background. With the standard stimulator containing blue (461 nm), green (513 nm) and red (652 nm) LEDs, we were unable to obtain satisfactory responses. Reproducible S-cone ERGs were obtained with a stimulator that had been custom-fitted with shorter-wavelength blue (440 nm) LEDs for stimulation, and orange (590 nm) LEDs for background adaptation. S-cone responses took only a few minutes to record, and the typical waveform showed a slow peak at 45-50 ms with amplitude 3-9 microV, but ranging from 0 microV to more than 10 microV. Larger waves appeared in a patient with enhanced S-cone syndrome. S-cone responses could also be obtained with an alternating blue-orange flicker protocol. We added the S-cone response to our regular ERG protocol for a number of months. Although most normal subjects and patients showed recognizable S-cone responses with this stimulator, the amplitudes were small and there was too much variability to make the technique effective for routine clinical testing. In general, the S-cone responses followed the standard cone ERG responses in disease.