Figure 1. Optical densities of the crystalline lens of the average patient aged 20 years and 70 years together with those of a short-wavelength light–absorbing intraocular lens (SLA IOL) (AcrySof Natural [Alcon]) and a conventional UV-absorbing IOL (AcrySof [Alcon]) plotted against wavelength.1 AU indicates absorbance units.
Figure 2. The so-called aphakic blue-light hazard function. Normalized log efficacy is plotted against wavelength.
Figure 3. Normalized spectral sensitivity curves for the photoreceptors of the retina and melanopsin. Data for the photoreceptors are the estimates of Dartnall and colleagues42 for human photoreceptors. Melanopsin sensitivity is estimated for a photopigment with a peak at 480 nm using the photopigment template of Lamb.43 Normalized log sensitivity is plotted against wavelength.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Simunovic MP. On Seeing Yellow: The Case for, and Against, Short-Wavelength Light–Absorbing Intraocular Lenses. Arch Ophthalmol. 2012;130(7):919–926. doi:10.1001/archophthalmol.2011.1642
Author Affiliations: Sydney Eye Hospital and Save Sight Institute, University of Sydney, Sydney, Australia.
The normal human crystalline lens absorbs UV and short-wavelength visible electromagnetic radiation. Early intraocular lenses (IOLs) permitted the transmission of such radiation to the retina following cataract extraction. Experimental studies of the absorption profile of the crystalline lens and animal studies demonstrating the deleterious effects of short-wavelength radiation on the retina led to the development of UV-absorbing, and later, short-wavelength light–absorbing (SLA) IOLs. Short-wavelength light–absorbing IOLs were designed to mimic the absorption properties of the normal crystalline lens by absorbing some short-wavelength light in addition to UV radiation; however, debate continues regarding the relative merits of such lenses over UV-absorbing IOLs. Advocates of SLA IOLs suggest that they may theoretically offer increased photoprotection and decreased glare sensitivity and draw on in vitro, animal, and limited clinical studies that infer possible benefits. Detractors suggest that there is no direct evidence supporting a role for SLA IOLs in preventing retinal dysfunction in humans and suggest that they may have negative effects on color perception, scotopic vision, and circadian rhythms. This article examines the theoretical and empirical evidence for, and against, such lenses.
The human crystalline lens absorbs UV and short-wavelength visible electromagnetic radiation (Figure 1); this property appears to be derived from cellular components present at birth (such as amino acids) and from so-called lens pigments, which are believed to be derived largely from the amino acid tryptophan and which accumulate with age.2,3 Several roles for UV and short-wavelength visible radiation absorption by the crystalline lens have been mooted, including a reduction in the effects of chromatic aberration by filtering out highly refracted short-wavelength light, glare reduction,4 and photoprotection.2
Early cataract surgery permitted the postoperative transmission of UV and short-wavelength visible radiation to the retina. Studies on the spectral absorption profile of the crystalline lens2,5 and animal experiments demonstrating the deleterious effects of short-wavelength radiation on the retina6-8 led to the development of UV-absorbing intraocular lenses [IOLs] in the early 1980s (conventional UV-absorbing IOLs). This development was followed in the late 1980s by IOLs that also absorbed short-wavelength visible radiation9 (short-wavelength light–absorbing [SLA] IOLs).
Advocates of SLA IOLs suggest that they afford useful photoprotection by absorbing potentially harmful short-wavelength visible light, reduce glare sensitivity, reduce postoperative cyanopsia, and improve contrast sensitivity.10 Opponents of SLA IOLs suggest that they diminish photoreception, interfere with circadian rhythms, and do not provide additional useful photoprotection when compared with conventional UV-absorbing IOLs.11 There is a clear divergence of opinion regarding the benefits and drawbacks of SLA IOLs. This article examines the theoretical advantages and disadvantages of SLA IOLs and also discusses the empirical studies of such lenses.
The so-called sunlight age-related macular degeneration (ARMD) hypothesis suggests that age-related maculopathy (ARM)/ARMD results from sustained or repeated exposure to high-energy short-wavelength electromagnetic radiation.12 As a consequence, some epidemiological studies have examined the relationship between lifetime exposure to sunlight and the development of ARM/ARMD. Results have been mixed12,13; some studies find a positive association, or such an association in patient subgroups,14-19 whereas others find no association20-22 or a negative association.23 However, the crystalline lens may be an important confounding factor in any epidemiological study investigating the role of light in ARM/ARMD. Specifically, variations in the optical density of the lens pigments modulate the dose of short-wavelength electromagnetic radiation incident on the retina; estimates of environmental light exposure are at best estimates at the corneal surface rather than at the retinal surface. The effect of removing the crystalline lens is in itself a controversy. Meta-analysis of the cumulative evidence (from 24 studies involving 113 780 subjects) suggests, however, that cataract surgery may be associated with an increased risk of late ARMD,24 although the contribution of light to this increased risk remains uncertain. In addition, the mechanism of ARMD is generally agreed to be multifactorial; even if we accept that short-wavelength electromagnetic radiation exposure may have a role in the etiology of ARM/ARMD, many variables could modulate the tissue response to such radiation, including previous surgical intervention, diet and other lifestyle factors, systemic disease, macular pigment density, and genetic factors.13,15
Animal studies have investigated the effects of light exposure on the retina, although such studies are inherently limited to studying its acute effects. Two types of light-induced damage are recognized. The first, known as type I, was originally described by Noell and colleagues8 and results from prolonged exposure to light within the short- to medium-wavelength range of the visible spectrum. The peak of the action spectrum roughly coincides with that of the normal human scotopic sensitivity function, which lead to the suggestion that the effect may be mediated by the absorption of light by rhodopsin or one of its intermediates.25 The second type of induced damage, which occurs after exposure to shorter durations of more intense (by approximately 2 log U) light is known as type II phototoxicity and was first described by Ham and colleagues.7 In phakic animals, this so-called blue-light hazard peaks in the short-wavelength range of the visible spectrum and falls with increasing wavelength. Such damage is not only mediated by receptor photopigments, but also by lipofuscin absorption within the retinal pigment epithelium.25 Ham et al26 demonstrated that aphakic rhesus monkeys—in whom the ocular media transmit a significant proportion of UV radiation to the retina—show an increased susceptibility to retinal phototoxicity (Figure 2).
In a thorough theoretical analysis, van de Kraats and van Norren27 measured the transmission spectra and calculated the effects on visual function and photoprotection of a variety of SLA IOLs (in order of increasing short-wavelength visible absorption: YellowFlex [PhysIOL], AcrySof Natural [Alcon], Hoya AF-1 UY [Hoya Surgical Optics GmbH], Optiblue [Abbott Medical Optics Inc], and PC-440 Orange [Ophtec]) and conventional UV-absorbing IOLs (AT45 [Eyeonics], AcrySof [Alcon], and Clariflex [Abbott Medical Optics Inc]). Their data clearly show that all SLA IOLs are not equal, and this finding raises the important issue of standardization. At present, no universally accepted means exists to quickly and conveniently convey the photoreceptive and photoprotective effects of IOLs.27,28 Furthermore, the data accumulated by van de Kraats and van Norren27 show that SLA IOLs that do not have sharp cutoff characteristics (ie, those in which absorption tapers off gradually with increasing wavelength) vary significantly in their transmission with varying lens power. They suggest that sharp cutoff filters may limit this problem,27,28 although changes to the manufacturing process could also reduce the effect. Their calculations also suggest that the addition of an SLA filter does not necessarily render an IOL superior to all conventional UV-absorbing IOLs in terms of photoprotection. Indeed, PhysIOL's YellowFlex SLA IOL offered the worst of both worlds: inferior photoprotection with impaired photoreception when compared with the best-performing conventional UV-absorbing IOL. Such a lens had little to recommend itself in terms of spectral transmission and is no longer commercially available. Calculations by van de Kraats and van Norren27 suggest that all the remaining SLA IOLs should increase the threshold for type II phototoxicity by 0.26 to 0.52 log U when compared with the best-performing conventional UV-absorbing IOL (for northern skylight as the light source and using the blue-light hazard spectrum of Ham and colleagues7). Although they did not directly calculate the theoretical protective effect for type I damage, if we assume its action spectrum resembles that of rhodopsin, the protective effect is more modest, ranging from 0.02 to 0.20 log U. In addition, all conventional IOLs and 1 SLA IOL (YellowFlex) afforded less protection than the average crystalline lens in a 20-year-old subject.
The theoretical protective effects of SLA IOLs are supported by in vitro and animals models. Sparrow and colleagues29 irradiated lipofuscin A2E fluorophore–laden retinal pigment epithelial cells with light filtered by conventional UV-absorbing IOLs (AcrySof, Sensar [Abbott Medical Optics Inc], ClariFlex, and CeeOn Edge [Pharmacia and Upjohn]) or SLA IOLs (AcrySof Natural). They were able to demonstrate that the photoprotective effect of the SLA IOL resulted in a reduction of cell death by about 80%.29 Rezai and colleagues30 performed similar experiments using retinal pigment epithelial cells irradiated with short-wavelength light (430-450 nm) filtered through a conventional UV-absorbing IOL (Alcon AcrySof) or an SLA IOL (AcrySof Natural). They found a 50% reduction in cell death when this light was filtered by the SLA IOL. Marshall and colleagues31 demonstrated that proliferation of human uveal melanoma cell lines stimulated by short-wavelength light (475 nm) was suppressed when the light was first filtered through an SLA IOL (Acrysof Natural); this suppression was not observed with a conventional UV-absorbing IOL (AcrySof). Kurihara and colleagues32 exposed pseudophakic mice implanted with bespoke IOLs created from an SLA IOL (AcrySof Natural) or a conventional UV-absorbing IOL (AcrySof) to white light of an appropriate intensity and duration (5000 lux for 24 hours) to induce retinal phototoxicity. They demonstrated that animals undergoing implantation with an SLA IOL showed significantly less retinal dysfunction—as assessed by electroretinography—and fewer apoptotic retinal cells on histopathological examination than animals implanted with a conventional UV-absorbing IOL.32
Although modeling and in vitro studies using retinal pigment epithelial cells may apply to the blue-light hazard, this itself is an acute condition. Extrapolating such data to model chronic retinal conditions, such as ARM/ARMD, may not be appropriate.11 Similar reservations should be held in regard to in vitro studies using uveal melanoma cell lines33 and to murine models of acute retinal phototoxicity. Nevertheless, short-term photic maculopathy may be underrecognized; for example, as many as one-third of ophthalmologists who view operating microscope lights and/or perform retinal laser procedures without appropriate barrier protection have been shown to have an asymptomatic acquired tritan color vision deficiency.34 The magnitude of this deficiency appears to correlate to the dose of light exposure,34 although the action spectrum of this effect is unknown. Evidence also suggests that individuals with yellower crystalline lenses maintain better short-wavelength cone function in later life,35 suggesting a photoprotective or an adaptive effect of filtering short-wavelength visible light. In a small retrospective study of pseudophakic patients by Miyake and colleagues,36 the incidence of postoperative blood–retinal barrier breakdown and fundus autofluorescence was higher in those with conventional UV-absorbing IOL implants (Hoya MC-5 [Hoya Surgical Optics GmbH]) when compared with those with SLA IOL implants (Hoya UVCY [Hoya Surgical Optics GmbH] or the Yellow-Colored IOL [Menicon Co, Ltd]). More recently, Nolan and colleagues37 used a psychophysical paradigm to investigate the effects of cataract surgery on macular pigment optical density in a group of 42 patients undergoing cataract extraction and implantation of an SLA IOL (AcrySof Natural) or a conventional UV-absorbing IOL (AcrySof). These authors suggest that implantation of an SLA IOL is associated with a progressive increase in macular pigment optical density from 3 months and as long as 1 year after surgery whereas implantation of a conventional UV-absorbing IOL is not associated with a significant change. In a subsequently published nonrandomized study, Obana and colleagues38 used Raman spectroscopy to directly assess macular pigment optical density in 259 subjects with an SLA IOL implant (AcrySof Natural) or a conventional UV-absorbing IOL implant (Acrysof). Although they found that both types of lens were associated with a decline in macular pigment optical density, those receiving SLA IOL implants had significantly higher densities from 1 year postoperatively until their final follow-up point of 2 years. The increased macular pigment optical density in those with SLA IOLs—when compared with conventional UV-absorbing IOLs—may reflect a decreased consumption of macular pigment by retinal biochemical processes triggered by short-wavelength light absorption. The findings of Nolan and colleagues37 would suggest that such a mechanism may not act alone—at least in the year after surgery—because macular pigment optical density was found to be lower in the phakic state (ie, when short-wavelength light absorption would have been even higher than with an SLA IOL). Whatever the mechanism of increased macular pigment optical density, this finding warrants further investigation because it may in turn afford improved photoprotection,39 protection from oxidative stress,39 and optimization of red-green color discrimination via its notch filter absorption profile40 (although possibly at the expense of tritan discrimination40).
Despite the theoretical considerations and empirical results discussed thus far, no studies have been published that directly support the hypothesis that the photoprotection afforded by SLA IOLs results in a decreased incidence of macular dysfunction. In fact, limited evidence to the contrary exists: a recent study by Kara-Junior and colleagues41 of 30 patients with no preoperative retinal pathology undergoing cataract extraction and implantation of discordant IOLs (AcrySof Natural/AcrySof) suggested no interocular difference in retinal structure (as assessed by optical coherence tomography), visual acuity, or contrast sensitivity during a 5-year period after surgery. The limitations of this study include a small sample size and the fact that the results may not be generalizable to patients who could be especially vulnerable to increased short-wavelength light exposure, such as those with preexisting retinal disease or young patients, who would potentially face several decades of increased exposure to such radiation.
All natural and most artificial light sources emit light across a wide range of wavelengths. Similarly, the reflectance spectrum of almost all objects is broad. Furthermore, the visual pigments absorb light across a wide range of wavelengths (Figure 3). For these reasons, the effects of filters such as SLA IOLs under real circumstances may be minimal and restricted to special circumstances, such as visual comparisons in which a target differs from a background in its emission or reflectance within the short-wavelength region of the visible spectrum. The simplest way to demonstrate the inferiority of SLA IOLs is to exploit this prediction by testing sensitivity to near-monochromatic short-wavelength stimuli under conditions in which Weber's law breaks down, to present such stimuli against a background of a longer (dominant) wavelength, or to perform tests that assess relative sensitivity to short- and long-wavelength stimuli. The cynic would suggest that such studies merely recapitulate laboratory absorbance measures, albeit with a noisy and imprecise photodetector (ie, the human visual system). However, such studies are essential in order to place the effects of SLA IOLs into context given the interindividual and intraindividual variances in sensitivity known to affect psychophysical estimates and the possibility of medium- to long-term adaptive mechanisms.35 In addition, the anticipated effects of SLA IOLs are not equal and depend on their absorption profile27; these differences may account for some of the discrepancies in findings between studies.
Because SLA IOLs influence the spectral quality of light incident on the retina, one of the anticipated deleterious effects of such lenses is on color vision. Compared with conventional UV-absorbing IOLs, SLA IOLs would be anticipated to effectively decrease the chromaticity difference between warm and cool colors (ie, they should induce a tritan color vision deficiency).44 In theory, this could be compounded by the recently reported increases in macular pigment optical density found to be associated with SLA IOL insertion.37,38,40 Conversely, one predicted benefit would be a reduction in postoperative cyanopsia.
Experimental studies show that although a theoretical loss of tritan color discrimination should result, the effect is generally too small to be detected by clinical tests of color vision, such as the Farnsworth-Munsell D-15 panel test (and its derivatives),45-48 the Farnsworth-Munsell 100-Hue test,41,47,49-58 and the Moreland equation,46,59 under photopic conditions. Experiments using central color contrast detection tasks in patients with discordant IOL implants in either eye (Hoya AF-1 UV/Hoya AF-1 UY) also suggest that color discrimination is not adversely affected by SLA IOLs.60 One study that used external filters to mimic the effects of SLA IOLs in pseudophakic AMD patients implanted with conventional UV-absorbing IOLs suggests a decreased performance in a practical task of color discrimination: the sorting of black from navy socks.61 However, in comparison with the SLA IOL the authors sought to replicate, the external filter attenuated more short wavelength light.61,62 Several studies suggest that tritan discrimination is slightly impaired in those with SLA IOLs under mesopic conditions,48,49,56,63 and 1 study found that photopic tritan discrimination is slightly impaired in the early (6 months) but not the late (12 months) postoperative period.63 However, the measured impairment of mesopic tritan discrimination is small and certainly far less than what should be produced by the average crystalline lens of age-matched normal eyes (Figure 1).
A few studies have investigated the effects of SLA IOLs on short-wavelength automated perimetric64 thresholds in patients with discordant IOL implants (AcrySof/AcrySof Natural65,66 and Hoya AF-1 UV/Hoya AF-1 UY).46 The reduction in average sensitivity ranged from 1.6 dB66 (AcrySof/AcrySof Natural) to 2.7 dB46 (Hoya AF-1 UV/Hoya AF-1 UY). The difference was found to achieve statistical significance in two46,66 of the 3 studies.
Short-wavelength light–absorbing IOLs appear to reduce the incidence of cyanopsia significantly in the early postoperative period,55,67 although patients generally adapt to such chromatopsias within a few months of cataract surgery regardless.67
The rod photoreceptor's peak wavelength of sensitivity (λmax) lies at about 495 nm,42,68 which is approximately 60 nm below the visual system's peak sensitivity in photopic conditions (see Figure 3).69 Because of this fact and because under extremely low luminance levels sensitivity is independent of background illumination, it has been proposed that scotopic sensitivity should be reduced by SLA IOLs. Calculations suggest that the effect is modest.27,70,71 Data for a variety of SLA IOLs from van de Kraats and van Norren27 suggest a theoretical reduction in sensitivity in rods ranging from 0.04 to 0.22 log U (compared with the best-performing conventional UV-absorbing lens). However, their study used daytime skylight as the assumed light source. Because the natural scotopic light source—the night sky—is often rich in long-wavelength light,72,73 these calculations may represent an overestimation in the negative effects of such lenses. Our visual system is well adapted to extract information about relative differences in brightness/lightness, thus negating the effect of fixed filters under many circumstances; if the spectral quality of the target and background are identical, then one would predict no reduction in scotopic sensitivity under lighting conditions where Weber's law holds. Reductions in sensitivity should only be evident under very low lighting conditions (ie, at absolute threshold), and such conditions are seldom encountered in the modern world.74 Furthermore, these calculations overlook the possibility of long-term compensatory mechanisms.
Despite the fact that absolute threshold should theoretically be minimally increased by SLA IOLs, empirical studies have failed to demonstrate such differences in practice. Greenstein and colleagues51 examined scotopic sensitivity in 9 patients with discordant IOL implants (Acrysof/Acrysof Natural) to narrowband stimuli (440, 500, and 650 nm) as well as to a white stimulus. Although the authors found a trend toward slightly lowered sensitivity in eyes with Acrysof Natural IOL implants, such differences were minimal (<1 dB) and did not achieve statistical significance in their small sample. Muftuoglu and colleagues59 examined the effects of SLA IOLs on scotopic vision using a contrast detection paradigm with and without a glare source in the field of vision. They found no significant difference in scotopic contrast sensitivity between 38 eyes with SLA IOL implants (AcrySof Natural) and 38 eyes with conventional UV-absorbing IOL implants (AcrySof). Kiser and colleagues61 attempted to assess the effects of SLA IOLs on scotopic visual function in 22 elderly pseudophakic patients with AMD. All their subjects had bilateral conventional UV-absorbing IOLs, and their performance at a scotopic threshold task was assessed with and without an external filter supplied by an IOL manufacturer and designed to mimic an SLA IOL (AcrySof Natural). Performance was not found to be significantly affected by the filter.61
Yellow spectacle lenses are anecdotally believed to improve vision under certain conditions, and some empirical studies support the assertion that they may slightly improve contrast sensitivity.75 Part of the potential benefit may be from eliminating those wavelengths that are especially vulnerable to scatter and (in phakic subjects) that induce crystalline lens fluorescence.76 An additional hypothesis is that such lenses may improve the contrast of objects presented against backgrounds that are comparatively richer in short-wavelength light.76
Most studies find no difference in contrast sensitivity between eyes undergoing implantation with conventional UV-absorbing IOLs and SLA IOLs.41,45,47,48,53,54,58,59,63,67 Niwa and colleagues77 assessed contrast sensitivity functions in patients undergoing implantation with conventional UV-absorbing IOLs (Hoya UV) or SLA IOLs (Hoya UVCY). They found a statistically significant superiority for SLA IOLs at medium to low spatial frequencies under photopic and mesopic conditions. Yuan and colleagues55 similarly found that contrast sensitivity at middle to low spatial frequencies was slightly, although significantly, superior in patients undergoing implantation with SLA IOLs (lenses not specified). A recent study by Gray and colleagues78 investigating subjects' performance at a simulated driving task in the presence of a glare source suggests that those implanted with SLA IOLs (AcrySof Natural) perform significantly better than those implanted with conventional IOLs (AcrySof). Hammond and colleagues4 investigated the effects of glare and photobleaching on patients with discordant IOL implants (AcrySof/AcrySof Natural). Their results suggest SLA IOLs provide a significant reduction in susceptibility to glare and improved photostress recovery. This study supported their previous findings in patients with the same type of IOL implants bilaterally.79 Wang and colleagues56 found evidence to suggest that low-contrast acuity was reduced by SLA IOLs under mesopic lighting conditions (Hoya AF-1 vs MC 611 MI [HumanOptics]). Wirtitsch and colleagues46 similarly found a small but significant decrease in contrast acuity associated with SLA IOLs that appears to arise from differences under mesopic conditions (discordant Hoya AF-1 UV/Hoya AF-1 UY IOLs). Pierre and colleagues80 used a minimum-motion paradigm of a red-blue grating to assess changes in contrast perception. This study effectively measured the effect ofSLA IOLs on the spectral sensitivity function of their patients, and perhaps unsurprisingly found a small but significant reduction in relative sensitivity to the blue portion of their stimulus.
One of the most intriguing findings in visual science during the past decade or so is that of the intrinsic photopigment of the retinal ganglion cells, melanopsin, which accounts for the partial maintenance of circadian rhythm in animals deficient of rods and cones.81 Melanopsin's λmax lies at about 480 nm (see Figure 3),82 and thus SLA IOLs would be anticipated to reduce light absorption by this photopigment. The theoretical reduction in light absorption by melanopsin caused by SLA IOLs, when compared with the best-performing conventional UV-absorbing IOLs, is small83-85 and has been estimated to range from 0.08 to 0.27 log U (assuming northern skylight as the source).27 Such a small reduction is unlikely to be of consequence for several reasons. First, such considerations overlook possible compensatory adaptation mechanisms86; second, rods and cones have known input into the circadian mechanism87; and third, complete abolition of light absorption by melanopsin has a surprisingly modest negative effect on circadian cycles.87
The only published empirical evidence comes from a small retrospective study. Landers and colleagues88 administered a standardized questionnaire on sleep patterns to 49 patients with conventional UV-absorbing IOL implants (31 patients with S140NB [Abbott Medical Optics]) or SLA IOL implants (18 patients with AcrySof Natural) during the year preceding their study. They found no significant difference in test scores between the 2 groups, although it should be noted that their study was conducted in a geographical region remarkable for its sunny climate. It has been argued that reduced photoreception by the circadian mechanism may have greater implications for those exposed to chronically low light levels, such as those living at high latitudes and/or those who spend extended periods indoors.9
Although SLA IOLs should theoretically affect color discrimination and absolute thresholds, rigorous analysis suggests that these effects would be minimal27 and reserved for special testing conditions. Furthermore, empirical studies have failed to demonstrate any functional difference in scotopic vision51,59,61 or color discrimination at clinical color vision tests under photopic conditions.45-47,49-54 Sensitivity at certain specialized visual tasks (eg, short-wavelength automated perimetry,46,66 mesopic tritan color discrimination)48,49,56 has been found to be reduced in some studies; however, such differences are small and of questionable functional significance. Although it has been argued that light absorption by melanopsin should be reduced by SLA IOLs, the calculated effect is modest and likely to be of little consequence to circadian rhythms.27,83-85 This conclusion is supported by the currently available empirical data.88 A few studies4,55,77,78 support the assertion that SLA IOLs may improve contrast sensitivity and susceptibility to glare compared with conventional UV-absorbing IOLs, although several studies suggest no significant difference41,45,47,48,53,54,58,59,63,67 and two suggest a slight decrement in contrast sensitivity.46,56
In terms of photoprotection, UV radiation and short-wavelength visible light have clearly been demonstrated to be capable of producing retinal phototoxicity (with the efficacy of such radiation falling with increasing wavelength).26 Consequently, experts generally concur that UV absorption by IOLs is recommended; however, opinions differ on the absorption of short-wavelength visible light. Although exposure to high levels of short-wavelength visible radiation can certainly produce acute retinal phototoxicity in animals and although occupational exposure to intense light sources may produce subclinical photic maculopathy,34 epidemiological studies investigating the role of light in ARM/ARMD have produced mixed results.12,13 Studies have demonstrated that SLA IOLs provide a protective effect in vitro29,30 and in vivo in a murine model.32 Furthermore, they have been shown to be associated with a decreased blood–retinal barrier breakdown in a small retrospective clinical trial.36 Recent clinical trials also suggest that SLA IOLs are associated with increased macular pigment optical densities postoperatively.37,38 No differences in retinal structure were found, however, in a small clinical trial performed on middle-aged patients with no preoperative retinal pathology who underwent implantation with discordant IOLs.41
The normal crystalline lens absorbs UV and short-wavelength visible radiation from an early age. At this time, no convincing theoretical arguments or empirical data suggest that the filtering of short-wavelength visible radiation should be eliminated at the time of cataract surgery. The choice of IOL clearly remains at the discretion of the surgeon, who can weigh the perceived benefits and drawbacks of each type of lens for the individual patient; this choice could be aided in the future by a standardized means of quantifying the photoprotective and photoreceptive effects of IOLs.27,28 In the absence of conclusive evidence, the logical default should be to replace the crystalline lens with an SLA IOL that mimics the absorption properties of the normal adult lens within the short-wavelength range of the visible spectrum until empirical evidence clearly supports a case for doing otherwise.
Correspondence: Matthew P. Simunovic, MB, BChir, PhD, Sydney Eye Hospital, 8 Macquarie St, Sydney, New South Wales, Australia 2000 (email@example.com).
Financial Disclosure: None reported.
Additional Contributions: Jim Schwiegerling, PhD, provided the transmission data for AcrySof IOLs.
This article was corrected for errors on July 11, 2012.