The use of illumination during vitreoretinal surgery is a double- edged sword: Although it allows us to see the fundus and other structures of the posterior segment, at the same time we must be concerned about the potential for phototoxicity to these structures. The Stellaris PC (Bausch + Lomb) offers several features that help to improve retinal illumination while at the same time minimizing the potential for damage due to phototoxicity.

In order to appreciate these illumination features, it is helpful to understand the interactions of light with posterior segment tissues. This article discusses those interactions, followed by explanation of the illumination options of the Stellaris PC, their theoretical benefits, and some experimental evidence.


Light's interaction with tissue takes a number of forms. When light passes through tissue without being attenuated, this is called transmission of light, and it is the basis for transparency. Even where tissues are transparent to particular wavelengths of light, the photons can interact with macromolecules in the tissue, causing changes in the direction of the photons. This scattering is most pronounced with short wavelengths of light, at the blue end of the spectrum, and less pronounced for longer wavelengths.

In the eye, scattering is produced by macromolecules in the clear media—the cornea, the lens, and the vitreous. Ophthalmologists exploit this phenomenon when visualizing the cornea at the slit lamp and assessing aqueous flare.

Refraction occurs where 2 tissues that interface have different refractive indices, such as the cornea and the aqueous. This phenomenon does not usually cause difficulty for vitreoretinal surgeons.

Diffuse reflection is back-scattering of incident light by microscopic irregularities in tissue at the cellular, subcellular, and macromolecular levels. This back-scattering is uniform in all directions, so that the tissue appears evenly illuminated from whichever direction it is viewed. Diffuse reflection is distinct from specular reflection, the mirror-like reflection seen from a smooth surface such as a body of water.

Finally, absorption occurs when photons are absorbed by chromophores within tissue. Absorption can lead to a variety of effects in tissue that may be desirable or undesirable, depending on the circumstances.


The 2 main types of undesirable light effects that are relevant to retinal surgeons are the blue-light hazard and retinal hazard. The blue-light hazard is caused by shortwavelength, high-energy light at the blue end of the spectrum. When this is absorbed in the outer retina, it can cause damage to the retina with potential loss of vision that may be clinically unapparent. Retinal hazard is related to the power of the light source used and other factors. In retinal hazard, light absorbed in the retinal pigment epithelium causes pigmentary changes that are clinically detectable.

Van den Biesen and colleagues1 evaluated the safety of several commercially available light sources for vitreoretinal surgery in relation to retinal phototoxicity standards set by the International Commission on Non-Ionizing Radiation Protection (ICNRP). For all the light sources evaluated, the exposure times leading to retinal damage were alarmingly short, at less than 1 minute for the blue-light hazard with the light pipe at 10 mm from the retina, and less than 1 second for retinal hazard with the light pipe in contact with the retina.

Fortunately, as vitreoretinal surgeons know, phototoxic effects are fairly rare in vitreoretinal surgery, so these results appear puzzling. The main reason for the discrepancy may be that the ICNRP standards have a 33-fold safety threshold. Even taking that into account, however, it is possible to exceed that safety threshold within 30 minutes, the duration of many vitreoretinal procedures.


Tissue visualization is primarily a matter of contrast between tissues in the foreground and the background. Contrast arises mainly as a result of differences in color (chrominance contrast), differences in brightness (luminance contrast), and to a lesser extent differences in texture and movement of tissues.

Enhancing tissue visualization is a matter of maximizing chrominance and luminance contrast between tissues. This can be achieved through the selection of illumination sources with emission spectra complementary to the absorption spectra of chromophores and dyes in the eye.

Chromophores are molecules that absorb some part of the visible spectrum and allow us to perceive a tissue as a color. The color we see is the part of the spectrum that is reflected back by the chromophore; that is, the reflection spectrum of the chromophore is essentially the inverse of its absorption spectrum. The principal biologic chromophores in the eye are melanin (brown) and hemoglobin (red).

Dyes can be used to enhance the visualization of tissues, and in vitreoretinal surgery this is referred to as chromovitrectomy.2 Two dyes in common use are brilliant blue G (BBG) and trypan blue, which absorb red light and reflect blue light.

To exploit the luminance characteristics of these chromophores and dyes, the Stellaris PC offers 5 illumination options with different emission spectra. The xenon light source can provide broad-spectrum white light when unfiltered, or it can be modified with yellow, green, or amber filters. A comparison of the emission spectra of these 4 options is seen in Figure 1. A fifth option is a mercury vaporxenon light source, which has an emission spectrum with 2 peaks of emission in the green/yellow part of the spectrum (Figure 2). Because of this narrow spectrum, the mercury vapor-xenon light source does not benefit from the use of filters.

A comparison of the internal limiting membrane (ILM) field in the same eye with different filters can be seen in Figure 3.

Relative to the discussion of contrast above, the vitreous is the foreground and the retina is the background. Under white light, the vitreous scatters blue light more than red, and the fundus reflects red and absorbs blue. To darken the background, thus enhancing contrast, a green filter can be used to filter out red wavelengths (Figure 4). The green filter also helps improve visualization of the vitreous.

The amber filter can be useful when blue dyes are used. The fundus reflects red and absorbs blue, and for the blue dye it is the reverse. When blue light is filtered out by the amber filter, the dye appears almost flat against the background of the fundus (Figure 5).


In a prospective randomized case series, Henrich and colleagues3 assessed the potential of intraoperative light filters on the Stellaris PC to enhance contrast following intraoperative BBG staining of the ILM. They found that the use of filters did not significantly alter the contrast between the ILM and the underlying retina compared with white xenon light, although there was a trend toward improved contrast with the amber filter (P = .13), while green and yellow filters produced slightly inferior contrasts (P =. 37 and .64, respectively). Contrast recognizability was better with the amber compared to the green filter (P = .04).

The study authors concluded that, among the available filters, the best results were achieved using the amber filter, and they recommended use of the amber filter during BBG chromovitrectomy. This study looked only at chrominance contrast, not luminance contrast.4

There is little clinical data on the effect of filter use on outcomes. Coppola et al5 assessed outcomes in 10 patients undergoing 23-gauge epiretinal membrane surgery, 5 with the amber filter and 5 without. They found no difference in clinical outcomes in terms of visual acuity or autofluorescence, but there was a significant difference in reduction of retinal thickness with the amber filter. The clinical significance of these findings is unknown.


What, then, is the ideal illumination for the vitreoretinal surgeon? The ideal illumination exploits the visual properties of chromophores and dyes with the use of complementary emission spectra. The ideal illumination should maximize the contrast in luminance and chrominance between foreground and background. It should minimize phototoxicity risk in terms of emission spectrum, power, and duration of surgery. Ultimately, it should help to improve patient outcomes.

There is a sound theoretical basis for the use of colored filters to help visualize tissues in vitreoretinal surgery. While there is emerging evidence for a clinical benefit, it is far from complete, and more effort is needed in determining the optimum illumination spectra for these light sources, their luminance contrast performance, and their effect on clinical outcomes.

Richard Sheard, FRCOphth, is a Consultant Ophthalmic Surgeon based in Sheffield, South Yorkshire, United Kingdom. He is a consultant for Bausch + Lomb. He may be reached at + 0114 352 0030; fax: 0114 352 0031.

  1. van den Biesen PR, Berenschot T, Verdaasdonk RM, van Weelden H, van Norren D. Endoillumination during vitrectomy and phototoxicity thresholds. Br J Ophthalmol. 2000;84(12):1372-1375.
  2. Costa Ede P, Rodrigues EB, Farah ME, et al. Vital dyes and light sources for chromovitrectomy: comparative assessment of osmolarity, pH, and spectrophotometry. Invest Ophthalmol Vis Sci. 2009;50(1):385-391.
  3. Henrich PB, Valmaggia C, Lang C, et al. Influence of intraoperative light filters on contrast recognizability during brilliant blue G (BBG) assisted chromovitrectomy. a quantitative analysis. Poster presented at: Association for Research in Vision and Ophthalmology annual meeting; May 7, 2012; Fort Lauderdale, FL.
  4. Enaida H, Hachisuka Y, Yoshinaga Y, et al. Development and preclinical evaluation of a new viewing filter system to control reflection and enhance dye staining during vitrectomy. Graefes Arch Clin Exp Ophthalmol. 2012 May 9. [Epub ahead of print]
  5. Coppola M, Marchi S. Morfological and functional outcomes in 10 eyes underwent to 23g vitrectomy: no filter vs amber filter. A compared study. Poster presented at: Association for Research in Vision and Ophthalmology annual meeting; May 7, 2012; Fort Lauderdale, FL.