Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
PT" Investigative Ophthalmology & Visual Science, Vol. 30, No. 6, June 1989 Copyright © Association for Research in Vision and Ophthalmology Retinal Damage in Macaque after White Light Exposures Lasting Ten Minutes to Twelve Hours Jan J. M. Kremers and Dirk van Norren We induced photochemical damage in small parts of the retinas of anesthetized macaques after light exposures of varying intensity, lasting between 10 min and 12 hr. Damage was assessed both with funduscopy and densitometry at several periods after exposure. Damage was most extensive 2 days post-exposure, with similar thresholds for both methods. Reciprocity between exposure time and irradiance was found for all exposures at a threshold irradiant dose of 230 J/cm2. This is in good agreement with part of the literature data on monkeys, yet contradicts another report (Sykes et al) in which a much lower threshold dose was found. The latter data probably concern a different class of damage. It remains unclear what critical factors distinguish the two classes. Observations more than 70 days post-exposure show a divergence between funduscopic and densitometric thresholds. Although the appearance of funduscopic lesions had changed, the threshold dose remained 230 J/cm2. Densitometry showed full recovery of the amount of visual pigment for doses below 600 J/cm2. Invest Ophthalmol Vis Sci 30:1032-1040,1989 old damage should increase, rather than decrease. Griess and Blankenstein12 found the time constant of the repair processes to be almost exactly 4 days. From this study alone reciprocity is expected to hold up to about 2 days. But, as Griess and Blankenstein determined the time constant in an indirect way, this is only true when no other, unexpected, processes interfere at long exposures. Sykes'" experiments actually suggest such an interference. The aim of the present study was to investigate whether or not there is a deviation from reciprocity after long exposures, as described above. Therefore, we explored the exposure range from a few hundred seconds to 12 hr. The conditions were, apart from irradiance and time, similar to those used by Ham et al,9 that is, anesthetized macaque monkeys and exposures restricted to small patches of retina. In addition to funduscopy, we introduced retinal densitometry as a new technique for assessing damage. This technique has the advantage of being both quantitative and noninvasive. Also, direct information is obtained on the kinetics of visual pigments after photochemical damage. Finally, damage was also assessed between 2 and 7 months after exposure, to study the effect of repair mechanisms. When the primate retina is exposed to intense light for periods varying between several seconds and 4 hr, extensive damage to the pigment epithelium and to the neural retina may be found.1"4 In patients such damage is also reported after ocular surgery in which use was made of bright light sources.5"8 For periods between 10 and 1000 sec, Ham et al 2 9 1 0 verified that reciprocity holds, that is, irradiance is exchangeable for exposure time. The threshold dose for white light in Ham's experiments varied between 200 to 400 J/cm2. With exposures shorter than 10 to 100 sec, depending on the size of the retinal image, reciprocity fails because the rise in retinal temperature is such that the domain of thermal -damage is entered. It is unclear how far reciprocity holds for exposures beyond 1000 sec. One report is available in which macaque monkeys were exposed to 12 hr of continuous fluorescent light.'' We calculated that the threshold dose in the latter experiment was about 16 J/cm2, an order of magnitude lower than Ham et al's 200-400 J/cm2. The direction of the deviation from reciprocity is unexpected. Naively, it is expected that with increasing exposure time (and decreasing irradiances) repair processes gradually overtake damage induction. As a result the total dose to reach threshFrom the TNO Institute for Perception, Soesterberg, The Netherlands. Supported by the Netherlands Organization for Scientific Research, through the Foundation for Biophysics. Submitted for publication: May 23, 1988; accepted December 5, 1988. Reprint requests: D. van Norren, PhD, TNO Institute for Perception, P.O. Box 23, 3769 ZG Soesterberg, The Netherlands. Materials and Methods Animal Preparation Monkeys were sedated with 25-40 mg ketamine hydrochloride (Ketalar®, Park-Davis, Barcelona, Spain) and anesthetized with an initial intravenous 1032 Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 RETINAL DAMAGE AFTER WHITE LIGHT EXPOSURES / Kremers and van Norren No. 6 dose of 20 mg/kg of pentobarbital (Nembutal®, CEVA, Paris, France), followed by a continuous infusion of 2-3 mg/kg/hr. After intubation, artificial respiration was started with a mixture of 70% nitrous oxide and 30% oxygen. To prevent eye movements an intravenous infusion of a muscle relaxant pancuronium bromide (Pavulon®, Organon, Oss, The Netherlands) was given. The initial dose was 30 Mg/kg followed by a continuous infusion of 30 jtg/kg/hr. Pupils were dilated with a drop of Phenylephrin® (Bournonville-Pharma, Almere, The Netherlands), combined with a drop of Cyclogyl® (Schieffelin and Co., New York, NY). The head was fixed in place with ear bars and a jaw rest. A hard contact lens was ,used to prevent drying of the cornea. All preparation •was done in yellow light to minimize the possibility of retinal damage before exposure. This investigation adhered to the ARVO Resolution on the Use of Animals in Research. Irradiation Whenever we use the term irradiation, we refer to exposures of white light to establish damage thresholds. We used a dual beam ophthalmoscopic stimulator for irradiation under Maxwellian view conditions. The light source was a 450 W xenon arc with light stabilized power supply. Infrared radiation was eliminated by a Schott KG 3 filter. UV radiation was reduced by the glass lenses. Details of this apparatus can be found in Valeton and van Norren. 13 Four patches of retina were defined using a diaphragm with four holes imaged on the retina. The holes measured 4° diameter and were arranged in a two by two array, about 1 ° separated. The center of the array was positioned about 8° above the fovea. To vary the irradiance level of individual patches, neutral density filters were placed on the holes. The irradiances of the four patches were regularly calibrated with a Laser Precision RK-5100 Pyroelectric Radiometer. The spectral output of the white light at the corneal level was measured with a scanning spectroradiometer (Photo Research Spectrascan S.N. 2186). The results of the latter measurement are presented in Figure 1. The average color temperature was 5400°K. Retinal irradiance (Eret) was calculated from the measured irradiance (Eair) with4: J-ret '-'air ^' in which d is the distance between focus and the irradiated plane of the measuring device of the radiometer, f the focal length of the monkey eye (about 1.35 cm), n the refractive index (about 1.33), and t the ocular transmission (about 0.9). The maximal available retinal irradiance was 400 mW/cm2. 1030 Fig. 1. Spectral distribution of the damage-inducing light at the corneal level. Funduscopy The irradiated patch (seen through the ophthalmic stimulator) was marked on a fundus photograph. Two days after exposure, changes were visible as white-yellowish spots. Funduscopic observations were performed with an indirect ophthalmoscope (28 diopter lens), because the large field of view made lesions easier to recognize due to the comparison with the undamaged surroundings. We classified the observations into three categories: (1) no change, indicating subthreshold damage or absence of damage; (2) a just visible change (a patch somewhat smaller than the irradiated spot, and with fuzzy edges), as an indication for threshold damage; and (3) a distinct funduscopic change (a patch with a size comparable to the irradiated spot or somewhat larger, and with sharp edges), indicating suprathreshold damage. Funduscopic observations were made at four different time intervals after exposure: (1) 0.5-2 hr; (2) 4-12 hr; (3) about 2 days; and (4) minimally 70 days. The most complete observations were made 2 days post-exposure because at that time funduscopic changes were most clearly present (in accordance with observations of, eg, Ham et al u 0 ). Densitometry In retinal densitometry a beam of light is shone into the eye, reflected at the fundus and collected by a light-sensitive device, usually a photomultiplier (PM). Essential in the technique is: first, that measuring light bleaches only a small fraction of the photopigment. Second, measurements are done in two conditions, one with high adapting illuminance which virtually bleaches all visual pigment, resulting in a relatively high number of reflected photons, and Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 1034 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1989 one with lower adapting illuminances in which the visual pigment is partially regenerated, resulting in a lower number of reflected photons. Comparison of the two conditions yields a quantity bearing a relation to the density of the visual pigment. In our experiments, the density of the rod visual pigment was assessed with a laboratory-built device in which several features of the Utrecht densitometer14 were incorporated, but which had the advantage of measuring at seven wavelengths in rapid succession. It was the prototype of a densitometer described by van Norren and van de Kraats.15 Briefly, in a 100 msec cycle the eye is repetitively exposed to seven periods of light with different wavelengths, a dark period, and four periods of high adapting illuminance, each period lasting 5 msec. The mean retinal level of bleaching light in the monkey eye was equivalent to 3.2 X 105 td (2.2 X 105 scot td) for humans; the retinal illuminance due to the sequential exposure to the various wavelengths of the measuring light amounted an average of 3000 equivalent td (104 scot td). The measuring beam illuminated a field of 4.3°; the bleaching beam was 10°. The photomultiplier in photon counting mode collected light from a 2.6° area concentric with the illuminated field. The bleaching light was yellow (cut off at 530 nm) in order to avoid interference between bleaching and damaging light. A microprocessor sorted the counts of the photomultiplier for the proper wavelengths into seven channels, and these were all subsequently corrected for the counts collected in the dark period. The running average (with a time constant of four sec) of the counts at a chosen wavelength could be displayed as a function of time on a monitor, and all results could be stored on disk. Density was defined as the logarithm of the ratio of the number of counts at 511 nm and at that at 730 nm,14 the latter being a wavelength not appreciably absorbed by the visual pigment. Densitometric damage was defined by the damage index (D): D = (A - B)/A (2) where A and B are the density differences between high and low light level conditions, before and after exposure, respectively. Procedure At the beginning of an experimental session the anesthetized animal was placed behind the irradiating device and the block of four patches of retina was selected and marked on a fundus photograph. Next, the animal was placed behind the densitometer and the density of rhodopsin was measured at minimally two of the four selected patches of retina. When these Vol. 30 densities were similar, the other two patches were not measured. Then, the animal was placed back behind the irradiating device and irradiation was started. At predetermined times subsequent holes of the diaphragm were covered to allow for different exposure times. To assess short-term damage, density was measured at a number of patches between 0.5 hr and 2 hr after irradiation was completed. On other occasions the observations and measurements were done between 4 and 12 hr post-exposure. Therefore, they were grouped into a separate category. Density was nearly always measured 2 days after irradiation. Depending on the length of the irradiation this was between 33-40 hr after completion of irradiation. Additional density measurements were obtained between 70-210 days after irradiation. Whenever density was measured, notes were made on the appearance of the fundus. Some eyes were used more than once. Eye movements during exposure excluded some patches from evaluation. Results Funduscopy Two days post-exposure, the retinal patches that underwent a funduscopical change showed a brighter appearance than undamaged retina. The size of the damaged patches varied slightly, depending on the irradiant dose. Figure 2A through 2D summarizes the results of funduscopic observations for different irradiances and exposure times. The data are based on observations on 41 irradiated patches in ten eyes of seven animals. Up to 12 hr post-exposure not much damage was observed; only in a limited number of cases were threshold funduscopic lesions visible (half-open circles in Fig. 2A, B). In Figure 2C the results of the most extensive observations, made 2 days post-exposure, are given. In any single experiment, we noted that the transition between no damage and evident damage always was within a factor of two of irradiant dose. The spread in threshold dose between individuals was not large either: it was easy to draw a line by eye with slope —1 (constant dose) which separated sub- from suprathreshold data. This line represents a well defined and remarkably constant threshold dose of 230 J/cm2 over two decades of exposure times. To enable comparisons, the threshold dose line was drawn in all plots of Figure 2. It shows that it represents a conservative estimate of a threshold course for the post-exposure observations at other times. With the longest exposure time used, 12 hr, threshold damage is found at about 5 mW/cm2. For a human eye, this would correspond to about 3 X 106 td white light. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 No. 6 1035 RETINAL DAMAGE AFTER WHITE LIGHT EXPOSURES / Kremers ond von Norren 1 o00 10A 1 A:0.5-2hrs <^ © « o \ \ 1 ' B:4-12hrs o \ 3 O o^. \ o \ Fig. 2. Data on funduscopic observations 0.5-2 hr (A),4-12hr(B), 2 days (C), and more than 70 days (D) post-exposure. Filled circles represent a clear funduscopic lesion (suprathreshold damage), half-filled circles a just visible lesion (threshold damage), and open circles no visible lesion (subthreshold damage). The dashed line represents an irradiant dose of 230 J/cm2, which is the bestfitby eye for threshold damage 2 days post-exposure. 1 \ \ O \ °_o o en o \ \ o \ 103 \ °\ a.10' C: 2doys D:>70 days o \* \ •• • \ • » o V^»« 6 10 • • o \ \ \ o* ""* * j® ""* * s °N .* 10 3 -- 10' i 1 10u After 70 or more days, damaged patches which two days post-exposure were brighter than undamaged retina, now appeared darker and had a granular texture. They were surrounded by an annulus of about half a degree, with a somewhat brighter appearance than the rest of the retina. Recovery, in the sense that patches returned to normal, was minimal, as can be seen from Figure 2D. Densitometry An example of densitometric measurements at 511 nm is given in the left part of Figure 3. In the upper panel a measurement before exposure is given. The record starts with the bleaching light on; after 2 min the light was switched off and pigment regeneration was followed for 15 min in the dark. Thereafter, the bleaching light was switched on again and the trace is seen to return to its original fully bleached level. The density difference between bleached and regenerated was 0.09, fairly typical for the records obtained in these experiments. The accuracy with which the density differences could be assessed was between 10 and 20%. In the right part of Figure 3, density difference is given as function of wavelength. The difference is 102 10310° retinal irradiance (mW.cnrr2) i 103 maximal at 511 nm, indicative for rod involvement. The middle panel of Figure 3 presents the results of measurements taken 2 days after 4800-sec exposure with an irradiance of 220 mW/cm2, which was 4.6 times the funduscopic threshold dose. No changes in density are observed following an identical bleaching and regeneration sequence, which is indicative for a serious disturbance in the regeneration cycle of rhodopsin. In the lower panel of Figure 3, a measurement in the same patch 85 days after exposure is presented. Partial recovery of the density of rhodopsin was found. In Figure 4A densitometric damage indices obtained for 2 days post-exposure are given as a function of irradiant dose. The threshold dose obtained from funduscopy, the vertical line, provides a satisfactory separation between high and low damage. In Figure 4B the same mode of presentation is chosen for densitometric changes after prolonged recovery. The threshold dose for long-term densitometric damage, though less accurately determined, lies at about 600 J/cm2, which is a factor of 2.6 higher than the dose for funduscopic and densitometric threshold damage 2 days post-exposure. So, again, the fundu- Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 1006 INVESTIGATIVE OPHTHALMOLOGY & VI5UAL SCIENCE / June 1989 Vol. 30 0.12 pre 0.08 0.1 0 a> I.-**"* \ — ^ iJ) c 0.1 2 days post Ui 0 0.1 0 Jl 0 2 days post CD 85 days post kMIM 0 "°0.04 85 days post •VMUMMJ 8 10 12 time (min) U 16 18 20 400 500 600 700 wavelength (nm) Fig. 3. Example of densitometry before exposure (top panels), 2 days after a 4800-sec exposure to 220 mW/cm 2 (middle panels), and 85 days after the same exposure (bottom panels). The left panels show the density traces for the measuring light of 511 nm wavelength. A trace starts with the pigment bleached. After 2 min the bleaching light is turned off, and the visual pigment is allowed to regenerate for 15 min. Then, the bleaching light is turned on again, and the density trace is seen to return to its previous level. The right panels show the difference spectrum, which resembles the rhodopsin density spectrum. Note the absence of visual pigment 2 days post-exposure and the partial recovery of visual pigment after 85 days. scopic criterion 2 days post-exposure proves to be the most conservative damage criterion. In Figure 5, funduscopic and densitometric results are compared for 2 days, and more than 70 days post-exposure, respectively. Correspondence is high for observations 2 days post-exposure, but much poorer after prolonged recovery. In distinct funduscopic lesions partial or even complete densitometric recovery was measured in many instances. The data obtained between 0.5 and 2 hr post-exposure are not presented in a figure, because none of the ten measurements showed a densitometric change, despite the fact that the irradiant total dose was about 550 J/cm2, and despite the fact that in one of the measured patches a vague funduscopic change was observed (after a 12 hr exposure to 11.5 mW/cm2). Only two measurements were done between 4 and 12 hr post-exposure. The irradiant doses in the two patches were 351 and 1200 J/cm2, respectively, and both displayed a vague funduscopic change. Only the high dose exposure resulted in a densitometric damage (D = 1). Discussion Relation between Irradiance and Exposure Time at Threshold The funduscopic threshold damages, obtained 2 days post-exposure, (Fig. 2C) occurred at a constant 230 J/cm2 irradiant dose for exposure conditions of up to 12 hr. This result is interesting in two ways. In the first place it extends the domain of time-irradiance reciprocity. From Ham et al's9 data, reciprocity was clear up to at least 1000 sec. By combining data in the literature for various animal models, we could already tentatively extend that period to about 4 hr.16 This study now provides experimental evidence for reciprocity up to 12 hr in one animal model, the macaque. In the second place the value of the threshold dose of 230 J/cm2 merits attention. It is only slightly lower than the 410 J/cm2 value derived from the data of Ham et al.9 This difference may have a technical explanation. Ham et al9 used the integrated value from 400-800 nm, whereas ours ranged from 400-700 nm. Since all evidence suggests that wavelengths over 700 nm do not contribute to photochemical damage (provided they do not increase retinal temperature), our threshold data are comparable to Ham's data. However, our results are clearly at variance with those of Sykes et al," who found threshold damage already at 16 J/cm 2 in an animal model closely related to ours. The contradiction with Ham's data was not obvious as Sykes' exposure was outside the irradiance/time domain covered by Ham. The controversy has now become much more manifest by our extension of the reciprocity domain to 12 hr. Despite similarities in exposure time and animal model, there are differences between our and Sykes' experiments, which might explain the different threshold irradiant doses. These differences are the use of anesthetics, the size of the exposed retinal area, Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 No. 6 1037 RETINAL DAMAGE AFTER WHITE LIGHT EXPOSURES / Kremers and van Norren the technique of damage assessment and the light source. 1. Anesthetics: We exposed the monkey's eye under Nembutal anesthesia, whereas the animals in Sykes' experiments were unanesthetized. There is evidence indeed that Nembutal provides protection.17 On the other hand, rough estimations of the damage dose in (unanesthetized) human sungazers are certainly not lower than our 230 J/cm2 value, because only minor damages were reported even after 1 hr of sungazing18 (retinal irradiance estimated to be about 104 mW/cm 2 , no corrections for eye movements). 2. Size of exposed retinal area: Sykes et al used large-field exposure, whereas we used 4° field sizes. 2 days 1.0 0.8 0.6 • x 0.4 I 0.2 - • o 1.0 - • A: 2 days o 0.0 - r1 >70days . o 1.0 l_ 0.8 • * n=1 - n=2 n=10 Q) 0 8 0.6 | 0.4 I 06 0.2 0.4 - 0.0 - • 0.2 - -0.2 B:>70days 1.0 • 0.0 0.8 no just clear funduscopic damage 0.6 Fig. 5. Relation between funduscopic and densitometric damage indices two days and more than 70 days after exposure. The mean densitometric damage and standard deviation are presented at three funduscopic damage indices, together with the number of observations. Correspondence between both indices is most clearly present 2 days after exposure. 0.2 0.0 -0.2 500 1000 irradiant dose (J.cnr 2 ) Fig. 4. Densitometric damage index as a function of the irradiant dose, for measurements 2 days (A) and more than 70 days (B) post-exposure. The dashed lines represent the best fitting doses for threshold densitometric damage: 230 J/cm2 after 2 days, and 600 J/cm2 after more than 70 days. As a matter of fact, Ham et al19 have suggested, but never substantiated, such an explanation. 3. Technique of damage assessment: Sykes used histologic criteria, whereas our data were based on funduscopic and densitometric observations. Both this study and our earlier literature survey have shown, however, that different techniques do not markedly differ in results. Fuller et al20 found that the irradiant dose for a funduscopic threshold was only Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 1038 INVESTIGATIVE OPHTHALMOLOGY & VI5UAL 5CIENCE / June 1989 107 •• B 10 1 week-* \ a \ 3 105CD e \ a a A a A • o Ooo A class I •i90-580 V. 0 1 day-* 12hrs-* V o o ^ o class n • 1hr-* O Q. 3 a 10 10' 3 10" 10A 10,-2 2 irradiance (mW.crrr ) about a factor of two higher than the dose for a histologic threshold. 4. Light source: We used a xenon arc lamp, whereas fluorescent bulbs were used in Sykes' experiments. However, taking account for the action spectra of photochemical damage, it is very unlikely that the difference in spectral distribution of the radiation might cause such large differences in threshold doses. Another difference might be the flicker in the fluorescent lights due to the AC power supply. We do not find any mechanistic justification for this possibility. Therefore, it seems very improbable that the differences in threshold doses would have been caused by the use of different light sources. The importance of these possible differences may become clear when things are put in a different perspective by noting that the results of the Sykes experiments do not form just an odd point in a further consistent picture of threshold doses (Fig. 6). They seem to fit into a series of threshold data mainly found in rats (eg, 21-23) but also in pigeons24 and in other monkey experiments. 2526 These are the socalled Class I damages as defined by Mainster27 and Kremers and van Norren.16 Class I damage is found Vol. 30 Fig. 6. Funduscopic data from the present experiments (circles), brought together with literature data. Open symbols denote subthreshold damage, halffilled symbols threshold damage, and closed symbols represent suprathreshold damage. The drawn curve is a theoretical threshold curve for Class II damage.1216 The following Class I data were used: V: rats21 (denoted with 490-580, indicating the wavelength contents of the irradiant light); D: rats23; A: pigeons24; >: monkeys". The literature data on Class II damage shown here (0) originate from Ham et al.9 Except for the experiments of Noell et al21 all exposures involved white light sources. One of our observations (12 hr exposure to 0.7 mW/cm2) was not shown in Figure 2, as it did not provide additional information on threshold data, but it has been included in this figure to emphasize the controversy with the data of Sykes etal." in experiments that have the following characteristics in common: unanesthetized animals, mainly rats, are exposed to large-field, long-term (12 hr or longer) irradiation, with retinal irradiance seldom exceeding 1 mW/cm2 (data on white light sources16). On histological examination of the retina, damage is mainly restricted to the photoreceptor level (for a review see Lanum28). The action spectrum, found in rats, is similar to the absorption spectrum of visual pigment.21-22 Harwerth and Sperling's25 experiments, in which photoreceptor systems could be selectively damaged by colored light, suggest an action spectrum resembling visual pigment absorption spectrum also exists in monkeys. The damages produced in our experiments are typically of the Class II type. Conditions in which photochemical damage of Class II is found can be characterized as follows. Small patches of retina in usually anesthetized animals were exposed, for limited periods of time, to irradiances typically exceeding 10 mW/cm2 using white light sources. Damage is assessed at least 24 hr after exposure; in general the first signs of damage appear in the pigment epithelium.19 The action spectrum, in macaque monkeys, showed Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 No. 6 RETINAL DAMAGE AFTER WHITE LIGHT EXPOSURES / Kremers and von Norren maximum sensitivity in the UV. Animal models in these experiments involved primates and rabbits.29"32 Human subjects, about to undergo eye enucleation, were also studied.1833 The amount of unbleached visual pigment has been suggested as crucial factor for class determination, resulting in lower threshold doses than expected for Class II damage for exposures of 12 hr or longer.16 Although this hypothesis seemed attractive for explaining literature data, our experiments do not support it. Possibly, a combination of some of the aforementioned factors is required for class determination. The unresolved controversy between Sykes' and our results should, therefore, not be viewed as just differences between two experiments, but should be placed in the more interesting, unresolved distinction between Class I and Class II damage. Recovery Two days post-exposure, damage seems to be most distinct. Thereafter, the appearance of the funduscopic lesions changes and the regeneration cycle of the visual pigments may recover. Two days after exposure, the damaged part of the retina appeared lighter than the rest and occasionally it seemed edematous. This is in line with reported depigmentation and mild edema.1'34'35 After more than 70 days, the lesion was less obvious (also observed by Ham et al1), darker than the rest of the retina, and it had a granular appearance. The granularity after prolonged recovery could be attributed to the irregular pigmentation of the RPE cells as described by Tso.35 We found substantial recovery from damage with densitometry. Thus, the disturbance of the visual cycle may vanish completely with the funduscopic lesion still present. The extent of recovery was found to depend on the irradiant dose: when the dose was higher than about 600 J/cm 2 , recovery was never complete. These results are in qualitative agreement with the effects reported by Moon et al.36 After exposing the macula of a rhesus monkey to 441 nm light, inducing a funduscopic threshold lesion, they found visual performance (visual acuity measured with Landolt rings) to be temporarily impaired, and returning to normal in about 10 days. After a suprathreshold exposure (1.5 times threshold dose) visual performance only showed partial recovery. We found incomplete recovery only after an irradiant dose of 2.6 times threshold. This might indicate a lingering damage to the neural retina despite full recovery of the visual pigment regeneration cycle. Key words: retina, photochemical damage, white light, irradiant dose, densitometry 1039 Acknowledgments The authors thank Drs. J. J. Vos and B. de Graaf for useful discussions, Dr. G. Szanto for correcting the manuscript's grammar, and Mr. F. Fiirrer and Mrs. N. Blokland for assistance during the experiments. References 1. Ham WT Jr, Ruffolo JJ Jr, Mueller HA, Clarke AM, and Moon ME: Histologic analysis of photochemical lesions produced in rhesus retina by short-wavelength light. Invest Ophthalmol Vis Sci 17:1029, 1978. 2. Ham WT Jr, Mueller HA, RufFolo JJ Jr, Guerry D III, and Guerry RK: Action spectrum for retinal injury from near-ultraviolet radiation in the aphakic monkey. Am J Ophthalmol 93:299, 1982. 3. Lawwill T, Crockett S, and Currier G: Retinal damage secondary to chronic light exposure: Thresholds and mechanisms. Doc Ophthalmol 44:379, 1977. 4. Calkins JL, Hochheimer BF, and D'anna SA: Potential hazards from specific ophthalmic devices. Vision Res 20:1039, 1980. 5. de Laey JL, de Wachter A, van Oye R, and Verbraeken H: Retinal phototrauma during intra-ocular lens-implantation. Int Ophthalmol 7:109, 1984. 6. Headon M, Jacobs N, and Rosen ES: Solar hazard: Eclipse viewing in Manchester. In Hazards of Light; Myths and Realities; Eye and Skin, Cronly-Dillon J, Rosen ES, and Marshall J, editors. Oxford, Pergamon Press, 1986, pp. 251-255. 7. Pigeaud-KJessens M and van der Heijde G: Operating microscope induced maculopathy: A case history. In Hazards of Light; Myths and Realities; Eye and Skin, Cronly-Dillon J, Rosen ES, and Marshall J, editors. Oxford, Pergamon Press, 1986, pp. 257-261. 8. Khwarg SG, Linstone FA, Daniels SA, Isenberg SJ, Hanscom TA, Geoghegan M, and Straatsma BR: Incidence, risk factors, and morphology in operating microscope light retinopathy. Am J Ophthalmol 103:255, 1987. 9. Ham WT Jr, Mueller HA, Ruffolo JJ Jr, and Guerry D III: Solar retinopathy as a function of wavelength: Its significance for protective eyewear. In The Effects of Constant Light on Visual Processes, Williams TP and Baker BN, editors. New York, Plenum Press, 1980, pp. 319-346. 10. Ham WT Jr, Mueller HA, and Sliney DH: Retinal sensitivity to damage from short wavelength light. Nature 260:153, 1976. 11. Sykes SM, Robison WG Jr, Waxier M, and Kuwabara T: Damage to the monkey retina by broad-spectrum fluorescent light. Invest Ophthalmol Vis Sci 20:425, 1981. 12. Griess GA and Blankenstein MF: Additivity and repair of actinic retinal lesions. Invest Ophthalmol Vis Sci 20:803, 1981. 13. Valeton JM and van Norren D: Fractional recording and component analysis of primate LERG: Separation of photoreceptor and other retinal potentials. Vision Res 22:381, 1982. 14. van Norren D and van de Kraats J: A continuously recording retinal densitometer. Vision Res 21:897, 1981. 15. van Norren D and van de Kraats J: A densitometer with the size of a funduscamera. Vision Res 29:369, 1989. 16. Kremers JJM and van Norren D: Two classes of photochemical damage of the retina. Lasers and Light in Ophthalmology 2:45, 1988. 17. Sperling HG: Spectral sensitivity, intense spectral light studies, and the color receptor mosaic of primates. Vision Res 26:1557, 1986. 18. Tso MOM and LaPiana FG: The human fovea after sungazing. Trans Am Acad Ophthalmol Otolaryngol 79:788, 1975. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017 1040 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1989 19. Ham WT, Allen RG, Feeney-Burns L, Marmor MF, Parver LM, Proctor PH, Sliney DH, and Wolbarsht ML: The involvement of the retinal pigment epithelium. In Optical Radiation and Visual Health, Waxier M and Hitchins VM, editors. Boca Raton, CRC Press, 1986, pp. 43-67. 20. Fuller D, Machemer R, and Knighton RW: Retinal damage produced by intraocular fiber optics light. Vision Res 20:1055, 1980. 21. Noell WK, Walker VS, Kang BS, and Berman S: Retinal damage by light in rats. Invest Ophthalmol 5:450, 1966. 22. Williams TP and Howell WL: Action spectrum of retinal light-damage in albino-rats. Invest Ophthalmol Vis Sci 24:285, 1983. 23. Rapp LM and Williams TP: A parametric study of retinal light damage in albino and pigmented rats. In The Effects of Constant Light on Visual Processes, Williams TP and Baker BN, editors. New York, Plenum Press, 1980, pp. 135-159. 24. Marshall J, Mellerio J, and Palmer DA: Damage to pigeon retinae by moderate illumination from fluorescent lamps. Exp Eye Res 14:164, 1972. 25. Harwerth RS and Sperling HG: Effects of intense visible radiation on the increment-threshold spectral sensitivity of the rhesus monkey eye. Vision Res 15:1193, 1975. 26. Zwick H and Beatrice ES: Long-term changes in spectral sensitivity after low-level (514 nm) exposure. Mod Probl Ophthalmol 19:319, 1978. Vol. 30 27. Mainster MA: Light and macular degeneration: A biophysical and clinical perspective. Eye 1:304, 1987. 28. Lanum J: The damaging effects of light on the retina: Empirical findings, theoretical and practical implications. Surv Ophthalmol 22:221, 1978. 29. Lawwill T: Effects of prolonged exposure of rabbit retina to low-intensity light. Invest Ophthalmol 12:45, 1973. 30. McKechnie NM and Johnson NF: Light damage to the retina. Graefes Arch Klin Exp Ophthalmol 203:283, 1977. 31. Skoog KO and Jarkman S: Photic damage to the eye: Selective extinction of the c-wave of the electroretinogram. Doc Ophthalmol 61:49, 1985. 32. Rinkoff J, Machemer R, Hida T, and Chandler D: Temperature-dependent light damage to the retina. Am J Ophthalmol 102:452, 1986. 33. Robertson DM and Feldman RB: Photic retinopathy from the operating room microscope. Am J Ophthalmol 101:561, 1986. 34. Tso MOM, Fine BS, and Zimmerman LE: Photic maculopathy produced by the indirect ophthalmoscope: 1. Clinical and histopathologic study. Am J Ophthalmol 73:686, 1972. 35. Tso MOM: Photic maculopathy in rhesus monkey: A light and electron microscopic study. Invest Ophthalmol 12:17, 1973. 36. Moon ME, Clarke AM, Ruffolo JJ Jr, Mueller HA, and Ham WT Jr: Visual performance in the rhesus monkey after exposure to blue light. Vision Res 18:1573, 1978. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933376/ on 06/12/2017