A comparison of laser-induced retinal damage from infrared wavelengths to that from visible wavelengths

Corneal, lenticular, and retinal damage have been observed following exposures to a laser emitting in the near-infrared wavelength range (Nd:Y AG, 1.318 ~m). Ocular damage thresholds are much higher than for visible wavelengths. However, it was found that infrared (IR) exposures may result in multiple damage sites throughout the ocular medium and retina; that exposure sites which initially appear to be unaffected may reveal slowly developing (days or longer) degeneration; and that late inflammatory responses may ultimately spread to areas of tissue not directly irradiated by the laser. The nature of tissue degeneration following IR laser exposure is examined and compared to that following visible wavelength laser exposures using three approaches: histopathology, scanning laser ophthalmoscopy, and optical coherence tomography. Each approach is shown to reveal unique aspects of the IR laser-tissue interaction when contrasted with effects induced by visible wavelengths. Introduction only begins to drop significantly for IR wavelengths above -1.1 J.lm, where water absorption becomes appreciable. In the far-IR, the composite absorption of the ocular components is such that virtually no incident radiation is transmitted to the retina. In the near-IR, absorption of incident radiation is distributed across the ocular components and, depending upon the precise exposure parameters, damage may be induced in one or more of the cornea, lens, and retina/choroid. The simultaneous induction of laser-induced damage in several tissues is perhaps best understood by schematizing the distribution of energy absorption through the ocular medium. This is done in Figure 2 which compares the absorption of O.SI4-J.lm argon laser radiation (Fig. 2a) to that for the 1.318-J.lm Nd:YAG laser emission used in this study (Fig. 2c). Distance into the eye is measRecent developments in laser technology have yielded a variety of powerful infrared (IR) laser sources, many of which fall in the wavelength range broadly categorized as 'eye-safe', Used in this context, the terminology is frequently divorced from its original intent as a relative descriptor meant only to convey the fact that ocular damage thresholds are significantly higher for 'eye-safe' wavelengths than for elsewhere in the visible and IR spectra. Transmission spectra of the ocular components (cornea, aqueous, lens, and vitreous) of the primate eye are plotted in Figure Ii. From this i1Iustration, it can be seen that transmission through each component (and, therefore, transmission to the retina) is high for visible wavelengths and Addressjor correspondence: Joseph A. Zuclich, PhD, TASC, 4241 Woodcock Drive, Suite B-I00, San Antonio, TX 78228, USA J.A.; Zuc/ich eta/., Infrared /aser-induced retinal damage 29 coherence tomography. Opt Engineer 34:701,710.1995 10. Fujimoto JG. Brezinski ME. Tearney GJ. Boppart SA. Bouma B. Hee MR. Southern JF. Swanson EA: Biomedical imaging and optical biopsy using optical coherence tomography. Nature Med I :970-972. 1995 II. Pan y. Birngruber R, Rosperich J. Engelhardt R: Low-coherence optical tomography in turbid tissue: theoretical analysis. Appl Opt 34:6564-6574. 1995 12. Schuschereba ST. Bowman PO, Vargas JA. Johnson TW. Woo FJ. McKinney L: Myopathic alterations in extraocular muscle of rats subchronically fed pyrodostigmine bromide. Toxicol Pathol 18:103-123, 1990 13. Kamovsky MJ: A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell BioI 27:137A-138A, 1965 14. Frisch GO. Beatrice ES, Holsen RC: Comparative study of argon and ruby retinal damage thresholds. Invest Ophthalmol 10:911-919, 1971 15. Majno G. Joris 1: Apoptosis, oncosis and necrosis: an overview ofcel1 death. Am J Pathol 146:3-15,1995 28 Lasers and Light in Ophtha/m%gy, 8(1} -/997 duced lesions are generally observed shortly following exposure and, within hours, stabi/ize so that no further changes in appearance are noted, save for the gradua/ fading in reflectivity over a period of months. In contrast, the IR retinal lesions were first detected only upon re-examination of subjects on the day following exposure, and thereafter. This delay time in the formation of observable damage is suggestive of a damage mechanism such as thermally-induced programmed cell death or apoptosisls. Funduscopically, the IR lesions reached maximum intensity at -48 hours postexposure. Path%gica/ eva/uation at that time revea/ed the presence of numerous inflammatory cells in the vitreous and at the ILM interface directly above the IR retinal lesion. The suggestion is made that this finding is a prognosticator of the late inflammatory response which, , expressed itself as the large, irregularly shaped opacity found at a 1.318-JJ.m exposure site at two months postexposure (Fig. 9). SLO observation of the two-month inflammatory response indicated strong involvement of the inner retina. Continued monitoring of the subject showed that the inflammatory response gradually cleared during the following month, so that by three months postexposure the underlying circular lesion was again visualized. In summary, the IR wavelength studied ( 1.318 JJ.m) defines the upper limit of wavelength where there is still sufficient transmission through the ocular medium to affect the retina of the eyel.2. Because such a large percentage of the incident IR laser radiation is absorbed by the ocular medium, and because that radiation which reaches the retina is not focused to as small an image size nor absorbed there as strongly as visible wavelengths, a very high dose of IR must be incident at the cornea to cause detectable funduscopic damage to the retina. Thus, the term 'eye-safe' has been applied even to higher power lasers in this segme~~o{;the IR wavelength region. 8Qw;" ever, it should be emphasized that applying the term 'eye-safe' to a given wavelength or wavelength band without regard to the laser power level and other beam characteristics, is inconsistent with published laser safety standards. Further, we report several observations which belie the 'eye-safe' terminology. First, a threshold IR lesion involves a volume of retinal tissue many or-~ ders of magnitude greater than that affected by a threshold-visible laser exposure and. hence, represents a more serious injury. Second. the usual threshold definition of a minimal visible lesion detected by funduscopic observation is not appropriate for the IR wavelength. since the monochromatic SLO imaging can detect retinal effects induced by significantly lower exposure doses. These observations. together with the delayed inflammatory responses observed histologically (Fig. 10) and funduscopically (Fig. 9). suggest that serious visual consequences may develop following a 'threshold. IR laser exposure.