Journal of Refractive Surgery

Review Article 

On the Safety of 193-Nanometer Excimer Laser Refractive Corneal Surgery

Catherine E Van Mellaert, MD; Luc Missotten, MD, PhD



Current literature on the potential hazards of 193-nanometer excimer laser corneal surgery is reviewed. The healing process of the different layers involved in the procedure is examined. The epithelium regenerates without delay or abnormal adhesion. A certain loss of corneal transparency seems to be only a short-term complication. The phototoxic risk for the endothelium and the lens has not yet been fully investigated. This risk is presently thought to be minimal. Factors such as improper corneal centering could lead to degradation of best corrected visual acuity or to loss of contrast sensitivity. Results of the first clinical trials indicate that vision quality does not appear to be adversely affected. The last section reviews studies designed to investigate the mutagenicity and cytotoxicity of 193-nanometer UV radiation. In vivo experiments could find no evidence of cellular transformation. (Refract Corneal Surg 1992;235-239.)



Current literature on the potential hazards of 193-nanometer excimer laser corneal surgery is reviewed. The healing process of the different layers involved in the procedure is examined. The epithelium regenerates without delay or abnormal adhesion. A certain loss of corneal transparency seems to be only a short-term complication. The phototoxic risk for the endothelium and the lens has not yet been fully investigated. This risk is presently thought to be minimal. Factors such as improper corneal centering could lead to degradation of best corrected visual acuity or to loss of contrast sensitivity. Results of the first clinical trials indicate that vision quality does not appear to be adversely affected. The last section reviews studies designed to investigate the mutagenicity and cytotoxicity of 193-nanometer UV radiation. In vivo experiments could find no evidence of cellular transformation. (Refract Corneal Surg 1992;235-239.)

Corneal surgery using ultraviolet radiation from a 193-nanometer excimer laser offers important advantages when compared with conventional techniques. The excimer laser ablates tissue with submicrometer accuracy, leaving minimal damage to adjacent tissue. In 1983, Trokel et al1 suggested that this new ablative tool could be used to remove large areas of corneal tissue by applying a graded intensity from center to edge, to steepen or flatten the cornea, and correct refractive errors of myopia, hyperopia, and astigmatism. Various delivery systems have been designed. Missotten et al2 used a rotating diaphragm perforated by one or several windows to modulate the uniform excimer beam. Other investigators used a constricting diaphragm3,4 or a scanning slit-delivery system.5 Preliminary experiments demonstrated the feasibility of such methods.

For photorefractive keratectomy (PRK) on the human eye to be clinically acceptable, not only the efficacy of this new technology but also the safety of the exposure to far-UV radiation must be established:

1. Wound healing must occur without significant scarring and the ablative process must preserve the optical properties of the central cornea;

2. The endothelium and the lens should not be adversely affected by the procedure;

3. Preservation of best spectacle corrected visual acuity is mandatory and complications such as glare or loss of contrast sensitivity should be avoided;

4. Finally, the most important requirement is to ensure that 193-nanometer excimer laser radiation is not hazardous or mutagenic for the surviving cell populations.


Corneal wound healing has been widely documented. Studies of PRK wounds have not found a significant latency of migration or abnormal epithelial adhesion. Recent reports on long-term healing observed only a transient hyperplasia of the epithelium during the 1st few weeks with a return to the normal number of cell layers at 6 months after surgery.6"8 Several indirect arguments, ie, the recovery of a normal number of hemidesmosomes, a normal innervation, and the lack of recurrent epithelial erosions,3,7,9'10 seem to indicate the presence of strong adhesion between the epithelial cells and their newly synthesized basement membrane. Bowman's layer is absent as it does not regenerate after surgery.

Overall, the healing of the superficial corneal layers appears to be very satisfactory. However, there is some debate in relation to stromal changes. The following questions seem to be appropriate in this context:

1. Is corneal transparency impaired by the healing process?

In many experimental studies with monkeys, a subepithelial haze developed within 5 days to 2 weeks after surgery.3'6'8,10*12 Using a fluorescent dye, dichlorotreiazinyl-amino-fluorescein (DTAF), Tuft et al13 and Goodman et al14 found that this haze could be explained by the deposition of new collagen in the anterior stroma. Malley and colleagues15 studied immunofluorescence staining patterns in corneal wounds after excimer laser corneal surgery and suggested that corneal opacity could be due to the deposition of type III collagen and the absence of keratan sulfate. Collagen III is not present in normal adult corneas. A correlation has been established between the absence of keratan sulfate and the formation of corneal opacities, especially in macular corneal dystrophy. Among monkeys of the same experimental group, individual variations in the amount of opacity were usually observed. Assuming that during these experimental ablations, parameters have been adequately controlled, the observed range of responses must be due to individual differences in wound healing.

2. Does ablation depth play a part in the degree of severity of stromal hazing?

Published information concerning this question is contradictory. Del Pero et al6 found that the haze was more pronounced in the regions of deepest ablation (centrally for a myopic correction and peripherally for a hyperopic correction). Fantes et al7 expressed surprise that shallow ablations led to a greater amount of haze than deeper ablations. Finally, McDonald et al10 reported that no difference in clarity was observed between groups with 1.50 diopters flattening and 3.00 D flattening (12 m deeper).

3. Does corneal clarity depend on the characteristics of the ablated surface?

A common underlying assumption of excimer laser corneal surgery is that a smoother keratectomy bed will cause less scarring. Marshall et al3 used a constricting diaphragm that moved only a few steps in monkeys and reported a mild stromal haze. In a similar experiment, but using a constricting diaphragm with more steps, McDonald et al4 reported opaque corneas in an initial series. Following technical improvements, they reported a milder haze in monkeys and in blind human eyes. Fantes et al16 acknowledged the importance of a smoothly ablated surface and designed their computer-driven rotating-slit apparatus to reduce steps in the stroma in the hope of diminishing the impediment to proper healing. However, corneas treated by this system did not have a perfectly smooth ablation and some did develop a subepithelial haze. More experimental work on this question would be desirable.

4. If healing involves a loss of clarity, will opacities clear with time?

In all studies, progressive clearing did occur except in one6 in which 3 of 12 treated eyes had a moderate haze with a slight progression and increased confluency. The time required for total clearing varies from 17 weeks to 2 years.3,6,10 However, the meaning of the words "clear cornea" might vary from one study to the other. That is, when an author speaks of a clear cornea, does he mean one that is indistinguishable from an untreated one by any method of slit-lamp microscopic examination, or, a functionally clear cornea with perhaps a barely perceptible haze not interfering with vision? Efforts are currently being made to develop a method to objectively and reproducibly grade corneal clarity of excimer treated corneas.17 Preliminary studies have shown that once transparency is acquired, it remains during a 1-year follow up.10 To summarize, the development of a faint haze in the anterior stroma is a complication that tends to resolve with time and that has not been associated with significant visual dysfunction.


Does PRK involve any risk for the endothelium or the lens?

In early studies, Marshall et al18 observed some endothelial cell loss when incisions extended to within a 40-micrometer distance of Descemet's membrane. Acoustic or shock waves were thought to be responsible for this damage. Because clinical procedures using the excimer laser for PRK will likely be confined to the superficial 50 µp? of stroma, concern regarding this question seems unwarranted. This has been confirmed in many studies with primates.3,6,7'11

To our knowledge, no results of experimental studies on lens effects after PRK have been published. Nevertheless, the following considerations offer some degree of reassurance concerning this point. Tuft et al19 characterized the fluorescence spectra emitted by the argon-fluoride laser and estimated the quantum-yield for photons emitted in the 260- to 350-nanometer range as approximately 1 × 10^sup -5^ J. They also calculated that the energy passing through a 4-millimeter diameter pupil to the anterior surface of the Zens- if absorption by the cornea and aqueous, which would cut off radiation below 290 nm is neglected - would be on the order of 5 × 10^sup -6^ J. The phototoxic risk at these low values appears to be very slight when compared to the exposure limits recommended by the International Radiation Protection Association.20 In another study, energy values at the lens level were found to be 1000 X lower than those necessary for acute cataract induction, but the authors argued that the UV exposure from PRK must be added to the accumulated effects of solar exposure.21 The influence of sunlight in the pathogenesis of cataract is still a matter of controversy. To exclude any possibility of damage to the lens caused by excimer laser treatment, further experimental work will be necessary.


Could vision quality suffer from PRK? The aim of refractive surgery techniques is to improve the patient's vision without spectacles. The degree of improvement of the uncorrected visual acuity reflects the efficacy of the technique. Obviously, a procedure may be called safe only if it also maintains best corrected preoperative visual acuity.

The development of stromal hazing as described above could compromise best corrected visual acuity, but it would probably to a greater extent affect contrast sensitivity and induce glare by light scattering in the cornea. Improper centering of the ablation zone relative to the entrance pupil is another factor which could significantly affect quality of vision.22,23 Calculations show the importance of correct centration for optimal retinal image quality. If a 4-millimeter optical zone is decentered by 1 mm relative to a 4-millimeter entrance pupil, 31% of the light rays emanating from a distant object miss the ablated area. These light rays will be refracted to a different extent creating ghost images, glare, or loss of contrast sensitivity.23 Several methods exist for the centering of the optical zone in refractive corneal procedures. Their reliability is still under discussion.23"26

From the first reports of treatment results of sighted human eyes, it would appear that best corrected visual acuity and contrast sensitivity could temporarily suffer from PRK but after 3 to 12 months preoperative values would be attained.27-31


Could surviving cells at the ablation edge be transformed or killed by exposure to 193-nanometer UV radiation? Up to this point, only the possible complications related to the healing process of the ablated surface have been discussed. Questions remain regarding the subcellular effects of high energy UV irradiation surrounding the ablation site.

Although most of the energy is consumed in the process of ablation, some of the 193-nanometer photons may be scattered and subsequently absorbed by the surrounding tissue where they may kill cells or cause genetic damage. Also, the longer wavelengths produced by secondary fluorescence at the site of ablation might pose a danger to adjacent tissue.19

UV radiation at wavelengths less than 300 nm is a known mutagen and carcinogen, and has been associated with basal cell carcinoma, squamous cell carcinoma, and melanoma.32 UV-induced corneal tumors have been reported in rats, mice, and hamsters upon chronic exposure to a broad band source of UV light.33

Primary neoplasms of the human cornea are rare and characterized by very slow growth.34 Corneal tumors almost always represent secondary extensions of lesions that primarily arise from the bulbar conjunctiva and limbus.34,35 Actinic keratosis, pathogenically related to an extended exposure of the conjunctiva to UV light, is precancerous in its severe form.36 In patients with xeroderma pigmentosum, all types of corneal tumors are more common.37 This genetic disease is caused by the absence of the enzyme required for the repair of UV-induced pyrimidine dimerization in DNA.38

UV radiation at wavelengths below 280 nm produces cyclobutyl pyrimidine dimers in DNA and thus has a potential for mutagenesis in mammalian cells.39 Excision repair appears to be the most important mechanism for the removal of the damaged DNA and can be monitored by the amount of unscheduled DNA synthesis (UDS).39

UDS was studied at the edge of an ablation by Green et al in human skin cells in vitro40 and by Nuss et al in corneal rabbit cells in vivo.41 Neither detected UDS in the cells adjacent to the ablated area. The absence of UDS after 193-nanometer excimer laser irradiation could be explained by one or a combination of the following arguments.

First, calculations show that up to 90% of incident radiation between 193 and 200 nm is absorbed by 1 µm of cytoplasm.42,43 The distance between the cell wall and the nucleus in the corneal epithelial cell amounts to about 1.5 to 3 µm. Cytoplasmic components, especially aromatic amino acids and peptides, may shield the nuclear material from incident energy.42

Another explanation may be that other forms of DNA damage may occur that are not measurable by UDS, in particular single-strand breaks (which may be less mutagenic than other lesions) and DNA protein cross links.43-44 Finally, other factors such as thermal effects may temporarily or permanently inhibit the excision repair process.40

Results of experimental systems using criteria other than UDS have been published. Trentacoste et al45 studied the mutagenic potential of the 193nanometer excimer laser on BALB/3T3 mouse fibroblasts. They compared the incidence of malignant foci in controls versus that of excimer laser exposed cells and found no statistically significant increase in mutagenic activity of the excimer-treated cells.

In vitro mutagenesis experiments with different cell lines produced variable results. Photoreactivation experiments with yeast cells showed a significant amount of DNA repair after exposure to 193nanometer radiation.46,47 Yeast cells have an enzyme, photolyase, which repairs UV-light-induced damage in DNA. Chinese hamster ovary (CHO) cells were used to study the induction of resistance to 6-thioguanine or to ouabain.42,48 CHO cells are sensitive to these substances and resistance arises through forward mutations in structural genes. Mutagenesis after 193-nanometer exposure was detected in the ouabain assay48 but not in the 6thioguanine assay.42,48

Two studies compared the cytotoxicity of 193nanometer radiation to that produced by a 254nanometer germicidal lamp by measuring the fluence of radiation at which 37% of the cells survive (D 37X42,43 This D 37 for 193-nanometer radiation in CHO cells was almost eightfold greater than that for 254-nanometer radiation, indicating that 193nanometer radiation is less cytotoxic per incident photon than 254-nanometer radiation.42 Kochevar et al demonstrated that the mechanism for cell killing by 193-nanometer radiation mainly involves damage to proteins in cellular membranes rather than DNA damage.49

Whereas the assays used by most investigators were designed to produce some evidence of DNA damage caused by excimer laser radiation, Gebhardt et al50 preferred to examine the laser's oncogenic potential in the mammalian cornea. For this purpose, they studied the in vivo growth of laser treated mouse corneas and keratocytes implanted in syngeneic recipients. After 8 months, no evidence was found of tumor growth at the subcutaneous injection or implantation site. Many authors agree that the in vivo growth potential in a syngeneic host is one of the most sensitive and relevant measures of cellular transformation.50

From previously mentioned publication on DNA damage and laser-induced cytotoxicity, it would appear that on the whole no adverse effects are to be expected. The few reports about damage to DNA concern in vitro assays. Extrapolation of these results to tissue in vivo should be taken with caution. Sensitivity to 193-nanometer radiation might differ between the cell lines used in vitro. In vivo, cells have a more spherical shape and the shielding role of the cytoplasm may be more important. Corneal tumors are rare and induction of such tumors by UV radiation in animals requires multiple exposures over long periods of time. Thus, carcinogenesis resulting from PRK seems highly improbable.

In summary, current knowledge on the safety aspects of 193-nanometer refractive surgery is limited but suggests that no major problems are to be expected. Further investigations on this matter would be desirable. Uniformity of treatment results is likely to improve with better delivery systems creating smoother ablation surfaces. Individual variation in wound healing capacity could remain an upsetting factor. Modulation of the wound healing response by means of pharmacological intervention with corticosteroids or other drugs may prove to be beneficial. This point remains to be further investigated.

Studies on humans with blind eyes, partially sighted, or normally sighted eyes are underway.27"31,51 Ine acceptance of the ArF excimer laser as a new surgical tool will ultimately depend on the long-term outcome of such clinical trials.


1. Trokel S, Srinivasan K, Braren B. Excimer laser surgery of the cornea. Am J Ophthalmol. 1983;96:710-715.

2. Missotten L, Boving R, François G, Coutteel C. Experimental excimer laser keratomileusis. Bull Soc Beige Ophtalmol. 1987;220:121.

3. Marshall J, Trokel S, Rothery S, Krueger R. healing of the central cornea after photorefractive keratectomy using an excimer laser. Ophthalmology. 1988;95:1411-1421.

4. McDonald M, Frantz J, Santana E, et al. Excimer laser surface shaping of the primate cornea for the correction of myopia. Invest Ophthalmol Vis Sci. 1988;29suppl):310.

5. Hanna K, Chastang C, Asfar L, Samson J, Pouliquen Y, Waring G. Scanning slit delivery system. J Cataract Refract Surg. 1989;15:390-396.

6. Del Pero R, Gigstad J, Roberts A, et al. A refractive and histopathological study of excimer laser keratectomy in primates. Am J Ophthalmol.

7. Fantes F, Hanna K, Waring G, Pouliquen Y, Thompson K, SalvoldeUi S. Wound healing after excimer laser keratomileusis in monkeys. Arch Ophthalmol. 1990;108:665-675.

8. Courant D, Fritisch P, Azema A, et al. Corneal wound after photo-keratomileusis treatment in the primate eye. Lasers and light in Ophthalmology. 1990;3:187-199.

9. Pallikaris G, Patazanaki M, Giorgiadis A, Frenschock O. A comparative study of neural regeneration following wounds induced by an argon fluoride excimer laser and mechanical methods. Lasers and Light in Ophthalmology. 1990;3:89-95.

10. McDonald M, Frantz J, Klyce S, et al. One-year results of central photorefractive keratectomy for myopia in the nonhuman primate cornea. Arch Ophthalmol. 1990;108:40-47.

11. Marschall J, Trokel S, Rothery S, Krueger R. Photoablative reprofiling of the cornea using an excimer laser: photorefractive keratectomy. Lasers in Ophthalmology. 1986;1:21-48.

12. Taylor D, LEsperance F Jr, Warner J, et al. Experimental corneal studies with the excimer laser. J Cataract Refract Surg. 1989;15:384-389.

13. Tuft S, Marshall J, Rothery S. Stromal remodeling following photorefractive keratectomy. Lasers in Ophthalmology. 1987;1:177-183.

14. Goodman J, Trokel S, Stark W, Munnerlyn C, Green W. Corneal healing following laser refractive keratectomy. Arch Ophthalmol. 1989;107:1799-1803.

15. Malley D, Steinert R, Puliafito C, Dobi E. Immunofluorescence study of corneal wound healing after excimer laser anterior keratectomy in the monkey eye. Arch Ophthalmology. 1990;108:1315-1322.

16. Fantes F, Hanna K, Waring GO, et al. Myopic laser keratomileusis on monkeys: clinical, microscopic and ultrastructural observations. Invest Ophthalmol Vis Sci. 1989;30(suppl):217.

17. Andrade H, McDonald M, Liu J, et al. Evaluation of an opacity lensometer for determining corneal clarity following excimer laser photoablation. Refract Corneal Surg. 1990;6:346-351.

18. Marshall J, Trokel S1 Rothery S, Schubert H. An ultrastructural study of corneal incisions induced by an excimer laser at 193 nm. Ophthalmology. 1985;92:749-758.

19. Tuft S, Al-Dhahir R, Dyer P, Zehao Z. Characterization of the fluorescence spectra produced by excimer laser irradiation of the cornea. Invest Ophthalmol Vis Sci. 1990;31:1512-1518.

20. International Non-Ionizing Radiation Committee of the International Radiation Protection Association. Proposed change to IRPA guidelines on limits of exposure to ultraviolet radiation. Health Phys. 1989,56:971.

21. MüUer-Stolzenburg N, Müller G, Buchwald H, et al. UV exposure of the lens during 193 nm excimer laser corneal surgery. Arch Ophthalmol. 1990;108:915-916.

22. Corney L, Kelly C. Ocular aberrations following refractive surgery. Invest Ophthalmol Vis Sci. 1990;4(euppl):481.

23. Maloney R. Corneal topography and optical zone location in photorefractive keratectomy. Refract Corneal Surg. 1990;6:363-371.

24. Walsh P, Guyton D. Comparison of two methods of marking the visual axis on the cornea during radial keratotomy. Am J Ophthalmol. 1984;97:660-661.

25. Steinberg E, Waring GO. Comparison of two methods of marking the visual axis on the cornea during radial keratotomy. Am J Ophthalmol. 1983;96:605-608.

26. Uozato H, Guyton D. Centering corneal surgical procedures. Am J Ophthalmol. 1987;103:264-275.

27. Zabel R, Sher N, Ostrov C, et al. Myopic excimer laser keratectomy: a preliminary report. Refract Corneal Surg. 1990;6:329-334.

28. Epstein D, Fagerholm P, Fitzsimmons T, Tengroth B. Excimer laser photorefractive keratectomy (PRK) for myopia - clinical results in sighted eyes. Invest Ophthalmol Vis Sci. 199l;4(suppl):720.

29. Lohman C, Timberlake G, Fitzke F, et al. The effect of changes in corneal transparency on visual acuity after photorefractive keratectomy using an excimer laser. Invest Ophthalmol Vis Sci. 1991;4(suppl):721.

30. Hogan C, McDonald M, Byrd T, et al. Effect of excimer laser photorefractive keratectomy on contrast sensitivity. Invest Ophthalmol Vis Sci. 1991;4(suppl):721.

31. O'Neill K, Eiferman R, Cook Y. Preliminary results of myopic excimer laser surgery. Invest Ophthalmol Vis Sci. 1991;4suppl):997.

32. Rundel R. Promotional effects of UV radiation on human basal and squamous cell carcinoma. Photochem Photobiol. 1983;38:569-575.

33. Freeman R, Knox J. UV-induced corneal tumors in différent species and strains of animals. J Invest Dermatol. 1964;43:431.

34. Steinhorst U, Von Domaras D. Carcinoma in situ of the cornea. Ophtalmologica. 1990;200:107-110.

35. Waring GO, Roth A, Ekins M. Clinical and pathologic description of seventeen cases of corneal intraepithelial neoplasia. Am J Ophthalmol. 1984;97:547-559.

36. Spencer W. Conjunctiva. In: Spencer W, ed. Ophthalmic Pathology: An Atlas and Textbook. Philadelphia, Pa: WB Saunders; 1985: 177-190.

37. Gaasterland D, Rodrigues M, Mosheil A. Ocular involvement in xeroderma pigmentosum. Ophthalmology. 1982;89:19801986.

38. Newsome D, Kraemer K, Robbins G. Repair of DNA in xeroderma pigmentosum conjunctiva. Arch Ophthalmol. 1975;93:660-662.

39. Sutherland B, Harber L, Kochevar F. Pyrimidine dimer formation and repair in human skin. Cancer Res. 1980;42:3181-3185.

40. Green H, Margolie R, Boll J, et al. Unscheduled DNA synthesis in human skin after in vitro UV-excimer laser ablation. J Invest Dermatol. 1987;89:201-201.

41. Nuss R, Puliafito C, Dehm E. Unscheduled DNA synthesis following excimer laser ablation of the cornea in vivo. Invest Ophthalmol Vis Sci. 1987;28:287-294.

42. Green H, Boll J, Parish G, Kochevar I, Oseroff A. Cytotoxicity and mutagenicity of low intensity, 284 and 193 nm excimer laser radiation in mammalian cells. Cancer Res. 1987;47:410413.

43. Kochevar I, Walsh A, Green H, et al. DNA damage induced by 193 nm radiation in mammalian cells. Cancer Res. 1991;51:288-293.

44. Zavilgelski G, Gurzadyon G, Nikogosyan D. Pyrimidine dimers single-strand breaks, and crosslinks induced in DNA by powerful laser UV irradiation. Chem Photobiophys. 1984^8:175.

45. Trentacoste G, Thompson K, Parrish R, et al. Mutagenic potential of a 193 nm excimer laser on fibroblasts in culture. Ophthalmology. 1987;94:125-129.

46. Seiler T1 Bende T, Winckler K, Wollensak J. Side-effects in excimer corneal surgery, DNA damage as a result of 193 nm excimer laser radiation. Groe fes Arch Clin Exp Ophthalmol. 1988;226:273-276.

47. Winckler K, GoIz B, Laskowski W, Bende T. Production of photo-reactivible lesion in the yeast S. cerevisiae by irradiation with 193 nm excimer laser light. Photochem Photobiol. 1988;47:225-230.

48. Rasmussen R, Hammer M, Berns M. Mutation and sister chromatid exchange induction in Chinese hamster ovary (CHO) cells by pulsed excimer laser radiation at 193 nm and 308 nm and continuous UV radiation at 254 nm. Photochem Photobiol. 1989;49:413-418.

49. Kochevar I, Walsh A, Held K, et al. Mechanism for 193 nm radiation-induced effects on mammalian cells. Radiât Res. 1990;122:142-148.

50. Gebhardt B, Salmerón B, McDonald M. Effect of excimer laser energy on the growth potential of corneal keratocytes. Cornea. 1990;9:205-210.

51. Seiler T1 Kahle G1 Kriegerowski M. Excimer laser (193 nm) myopic keratomileusis in sighted and blind human eyes. Refract Corneal Surg. 1990;6: 165^173.


Sign up to receive

Journal E-contents