Excimer Laser Ablation Rate and Corneal Hydration

Excimer Laser Ablation Rate and Corneal Hydration

Excimer Laser Ablation Rate and Corneal Hydration Paul J. D o u g h e r t y , M.D., Kent L. W e l l i s h , M.D., and Robert K. M a l o n e y , M.D. ...

914KB Sizes 0 Downloads 76 Views

Excimer Laser Ablation Rate and Corneal Hydration Paul J. D o u g h e r t y , M.D., Kent L. W e l l i s h , M.D., and Robert K. M a l o n e y , M.D.

During excimer laser photorefractive keratectomy, dehydration of the cornea begins as soon as the epithelium is removed. Corneal hydration might affect the excimer laser abla­ tion rate, which could affect the accuracy of correction. We studied the effect of corneal hydration on the excimer laser ablation rate in bovine eyes. To control hydration, bovine corneoscleral rims were equilibrated in dextran solutions of varying concentrations. One but­ ton trephined from each rim underwent laser ablation. Hydrated tissue ablation rates (amount of collagen, ground substance, and wa­ ter removed per pulse) and dry component ablation rates (amount of collagen and ground substance removed per pulse) were calculated from mass removed. The hydrated tissue abla­ tion rate at physiologic hydration was 0.40 μιη/pulse. As corneal hydration increased, the hydrated tissue ablation rate increased by 5.6 μg/cm 2 /pulse per increase in unit corneal hy­ dration (simple linear regression analysis, P = .0001). The dry component ablation rate de­ creased linearly by 0.82 μg/cm 2 /pulse per unit increase in corneal hydration (simple linear regression analysis, P = .0001). Both clinical data and theoretical arguments imply that dry component ablation rate determines refrac­ tive outcome after photorefractive keratectomy. Since the dry component ablation rate increases as the cornea dries, significant dehy­ dration of the cornea before ablation might lead to relative overcorrections of myopia. Surgeons should use a technique that mini­ mizes changes in hydration to maximize the

Accepted for publication March 1 1 , 1994. From the Jules Stein Eye Institute a n d t h e UCLA School of Medicine, Los Angeles, California. Dr. Malo­ ney is supported by a Research to Prevent Blindness Career Development Award, New York, New York. Dr. Wellish is an institutional Klara Spinks Fleming Fellow in Cornea and External Ocular Disease. Reprint requests to Robert K. Maloney, M.D., Jules Stein Eye Institute, 100 Stein Plaza, UCLA, Los Angeles, CA 90024-7003.

©AMERICAN JOURNAL OF OPHTHALMOLOGY

predictability of excimer laser photorefractive keratectomy. 1 HOTOREFRACTIVE KERATECTOMY with the argon-fluoride excimer laser can correct myopia by excising tissue from the anterior cornea. 16 The results are somewhat variable, however, with imperfect predictability in refractive out­ come.2"7 The refractive correction achieved de­ pends on the amount of tissue removed, modi­ fied by wound healing. This lack of precision in predicting refractive outcome is probably caused, in part, by variability in wound healing, but it may also be caused by variability in the original ablation. One source of variability in tissue removal might be variability in the abla­ tion rate, which is the amount of tissue re­ moved by each pulse of the laser. 8 In photorefractive keratectomy, the epitheli­ um is removed before ablation. Evaporative dehydration of the anterior cornea begins im­ mediately and depends in part on the length of time between scraping and ablation, the pres­ ence or absence of nitrogen flow, and the use of irrigation after epithelial scraping. If changes in corneal hydration affect the ablation rate, they could lead to a change in the amount of dioptric correction after photorefractive keratectomy. 810 Hydrated tissue ablation rate is the mass (or depth) of hydrated corneal tissue (collagen, ground substance, and water) removed per la­ ser pulse (Fig. 1). This is the term that is commonly used in published reports. We define dry component ablation rate as the mass of dry corneal tissue (collagen and ground substance only) removed per laser pulse (Fig. 1). Dry component ablation rate might differ signifi­ cantly from hydrated tissue ablation rate and may be affected differently by the level of cor­ neal hydration. In this experiment, we determined the effect of corneal hydration on both the hydrated tis­ sue excimer laser ablation rate and dry compo­ nent laser ablation rate in bovine corneas. We argue that dry component ablation rate is a more important determinant of refractive out­ come than hydrated tissue ablation rate. Final-

118:169-176, AUGUST, 1994

169

170

AMERICAN JOURNAL OF OPHTHALMOLOGY

o o o o o o o o o o o o_m oo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo

m

en 00 00

oo oo oo oo oo ablation oo oooooooooooooo oooooooooooooo Laser

Fig. 1 (Dougherty, Wellish, and Maloney). Diagram of a portion of the anterior corneal stroma after epithelial scraping before and after laser ablation. The dry component ablation rate equals the mass of collagen and ground substance (~) removed per pulse. The hydrated tissue ablation rate equals the mass of corneal water (o) plus the mass of collagen and ground substance (~) removed per pulse. In this case, assuming that a single laser pulse is responsible for the depicted ablation and that each unit of corneal water (o) has a mass of 2 mg and each unit of collagen and ground substance (~) has a mass of 3 mg, the dry component ablation rate equals 30 mg/pulse and the hydrated tissue ablation rate equals 110 mg/pulse. ly, we calculated the effect of changes in corne­ al hydration on refractive outcome.

Material and Methods Eyes were harvested from freshly slaughtered steers. Each cornea used in the experiment was free from epithelial defects or scars. Within six hours of death, 20 corneoscleral rims were dissected and rinsed with normal saline and placed in individual beakers containing 50 ml of a solution of 298,000-dalton average molecu­ lar-weight dextran (Sigma Chemical Co., St. Louis, Missouri) at concentrations of 2.5%, 5.0%, 7.5%, or 10.0%. Five rims were soaked at each concentration for 24 hours to establish an equilibrium hydration. Detailed preliminary experiments with bovine corneoscleral rims in dextran indicated that equilibrium hydration was reached at three hours and remained stable for 40 hours (unpublished data). Each rim was removed from solution, blotted dry, and rinsed with 5 ml of normal saline to remove residual dextran. The epithelium was removed by blunt scraping after loosening with an isopropyl alcohol-soaked pad. Experimental and control buttons, each 7.75 mm in diameter, were trephined from the paracentral portion of each of the 20 rims. Buttons were immediately weighed to 10 μg accuracy on an electronic scale. The buttons were then placed into an antidesiccation chamber and transported to the laser room. The antidesiccation chamber was a

August, 1994

low-volume Lucite chamber with an airtight lid that was used to minimize changes in corneal hydration from evaporation during transport to and from the laser room. In preliminary stud­ ies, buttons were weighed at successive time intervals inside the antidesiccation chamber. These weights were found to be stable with time, verifying that water did not escape from the chamber during experimentation (unpub­ lished data). An argon-fluoride excimer laser (ExciMed UV200, Summit Technology, Waltham, Massa­ chusetts) with a beam diameter of 4.0 mm, pulse rate of 10 per second, and fluence of 180 mj/cm 2 was used. Before experimentation, the fluence and beam homogeneity of the device were confirmed by ablation of a gelatin filter and a polymethylmethacrylate disk. Each ex­ perimental button received 1,200 laser pulses. Each control button was not ablated but other­ wise handled identically to the ablated button by transporting it in the same antidesiccation chamber and exposing it to the atmosphere adjacent to the laser during ablation of the experimental button. After laser ablation of each experimental button, both the experimen­ tal and the control button were weighed imme­ diately. After weighing, all buttons were desic­ cated for 36 hours with anhydrous calcium sulfate crystals in a desiccating oven set at 60 C and - 6 0 kilopascals. To calculate the mass of tissue removed by the laser, an estimate of the amount of mass lost to evaporation from the experimental button was needed. We assumed that this was equal to the mass lost to evaporation from each control button adjusted proportionally for the relative masses of the lased and control buttons. The estimated evaporative loss was subtracted from the experimental button's loss of mass to arrive at the hydrated mass removed by the laser. The average percentage of mass that evaporated from the control buttons from trephination un­ til desiccation was 6.0% (S.D. = 1.5%) of their initial mass. One matched pair of buttons was excluded from the analysis because a presumed recording error led to a calculated increase in mass of the experimental button after laser ablation. Corneal hydration was defined as milligrams of water per milligrams of dry corneal tissue. Hydration was calculated by subtracting dry corneal mass from hydrated corneal mass and dividing by dry corneal mass. 11 The hydrated tissue ablation rate in micrograms per square centimeter per pulse was

Exdmer Laser and Corneal Hydration

Vol. 118, No. 2

calculated by dividing the hydrated mass re­ moved from each experimental button by the number of pulses and the area of ablation. The hydrated tissue ablation rate in micrometers per pulse was calculated by dividing the hydrat­ ed tissue ablation rate in micrograms per square centimeter per pulse by the density of bovine cornea at the given hydration. The corneal den­ sity at each hydration was calculated using the density of dry cornea and the density of water, assuming that volume is conserved when water is added to dry cornea. 12 Hedbys and Mishima 13 calculated a density of 141 g/ml for dry bovine cornea. Using this figure, the calculated densi­ ties ranged from 1.07 g/ml for corneas with a hydration of 5.32 mg of water per milligram of dry tissue to 1.13 g/ml for corneas with a hydration of 2.24 mg of water per milligram of dry tissue (Table). Before the dry mass removed by laser abla­ tion from each experimental button could be calculated, the hydration of the tissue removed needed to be estimated. The hydration of the control button was used as an estimate of this

171

hydration. The dry mass removed was calculat­ ed using the control button hydration and the hydrated mass removed from the experimental button. Since hydration is equal to wet mass minus dry mass divided by dry mass, 11 by alge­ braic rearrangement, the dry mass removed is equal to the wet mass removed divided by one plus tissue hydration. The dry component abla­ tion rate in micrograms per square centimeter per pulse was calculated by dividing the dry mass removed from each experimental button by the number of pulses and the area of abla­ tion. The relationship between ablation rates in micrograms per square centimeter per pulse and hydration was analyzed statistically using simple linear regression analysis and Pearson correlation analysis under the null assumption that slope and correlation were zero. The dioptric error in refractive outcome when photorefractive keratectomy is performed in corneas at nonphysiologic hydration was calcu­ lated. We assumed that the amount of collagen and ground substance removed (dry component

TABLE SUMMARY DATA* CONTROL HYDRATION

PRELASER

HYDRATED

CORRECTED

TISSUE

NO.

MASS

MASS REMOVED

MASS REMOVED

DENSITY

DRY MASS REMOVED

1 2 3 4 5 6 7 8 9 irjt 11 12 13 14 15 16 17 18 19 20

2.24 2.33 2.42 2.47 2.53 2.92 2.99 3.01 3.02 3.11 3.56 3.73 3.76 3.90 4.13 4.53 5.11 5.20 5.32 5.32

57.25 38.74 36.33 34.15 33.38 42.56 38.98 37.11 40.31 45.62 46.72 52.02 47.44 47.43 52.99 53.72 60.29 62.05 60.55 61.48

8.18 8.11 8.32 7.74 8.55 8.69 8.48 8.58 9.42 4.90 9.22 9.55 9.54 10.78 10.43 11.92 10.52 10.23 10.51 12.70

4.92 5.25 5.81 5.69 5.65 7.09 6.38 6.47 6.46 -3.31 6.71 6.23 6.64 8.31 7.57 6.85 7.50 8.28 8.72 7.64

1.13 1.12 1.12 1.12 1.11 1.11 1.10 1.10 1.10 1.10 1.09 1.09 1.09 1.08 1.08 1.07 1.07 1.07 1.07 1.07

1.52 1.58 1.70 1.64 1.60 1.81 1.60 1.61 1.61 -0.81 1.47 1.32 1.40 1.70 1.47 1.24 1.23 1.34 1.38 1.23

RIM

*AII values are given in milligrams except for hydration, which is in milligrams of water per milligram of dry tissue, and density, which is in milligrams per cubic centimeter. 'Eliminated from data set because of a presumed data recording error that led to the calculated addition of mass to the experimental button after laser ablation.

172

AMERICAN JOURNAL OF OPHTHALMOLOGY

ablation rate) determines the refractive out­ come, because, when epithelial healing is com­ plete, the cornea returns to physiologic hydra­ tion regardless of the amount of water that was removed at the time of ablation. Dioptric cor­ rection is proportional to the depth of abla­ tion,14 which is proportional to the ablation rate. Then the ratio of the achieved dioptric correction to the intended dioptric correction equals the ratio of the dry component ablation rate at the nonphysiologic hydration to that at physiologic hydration.

Results The hydrations of the control buttons achieved with the various concentrations of dextran ranged from 2.24 to 5.32 mg of water per milligram of dry mass (Table). As expected, the buttons in the 2.5% dextran group were the most hydrated, and those in the 10.0% dextran group were the least hydrated. Normal bovine corneal hydration is 3.45 mg of water per milli­ gram of dry mass, 16 so hydrations above and below a physiologic level were achieved with this protocol. The hydrated tissue ablation rate ranged from 33 to 58 μg/cm 2 /pulse (Fig. 2). Figure 2 demon­ strates that the hydrated tissue ablation rate increased linearly with increasing hydration, with a slope of 5.6 μg/cm 2 /pulse per unit

August, 1994

hydration (95% confidence interval, 4.0 to 7.2). This slope and the resultant Pearson correlation coefficient of 0.859 were significantly different from zero (P = .0001). The hydrated tissue ablation rate at a physiologic hydration of 3.45 mg of water per milligram of dry tissue was 44 μg/cm 2 /pulse (95% confidence interval, 38 to 49). This rate yields a hydrated tissue ablation rate of 0.40 μιη/pulse (95% confidence inter­ val, 0.35 to 0.45) at physiologic hydration. The dry component ablation rate ranged from 8.1 to 12.0 μg/cm 2 /pulse (Fig. 3). The dry component ablation rate at physiologic hydra­ tion was 10.0 μg/cm 2 /pulse (95% confidence interval, 8.9 to 11.1). Figure 3 demonstrates that the dry component ablation rate decreased linearly with increasing hydration, with a slope of —0.82 μg/cm 2 /pulse per unit hydration (95% confidence interval, - 0 . 5 2 to -1.14). This slope and the resultant Pearson correlation coefficient of —0.773 were significantly differ­ ent from zero (P = .0001). The calculated dioptric error in photorefractive keratectomy using dry component ablation rate for corneas at nonphysiologic hydration for attempted 3-, 6-, and 12-diopter corrections are shown in Figure 4. Corneas above physiologic hydration will be undercorrected, and those

s

SI

I

!i

11

■at

I

I Corneal Hydration (mass of water/mass of dry tissue) Fig. 2 (Dougherty, Wellish, and Maloney). The effect of corneal hydration on the hydrated tissue ablation rate. The hydrated tissue ablation rate in­ creases as hydration increases (P = .0001).

Corneal Hydration (mass of water/mass of dry tissue) Fig. 3 (Dougherty, Wellish, and Maloney). The effect of corneal hydration on the dry component ablation rate. The dry component ablation rate de­ creases as hydration increases (P = .0001). The dry component ablation rate should be more important than the hydrated tissue ablation rate in determining refractive outcome after photorefractive keratecto­ my, since the corneal hydration returns to a physio­ logic level after ablation.

Vol. 118, No. 2

Excimer Laser and Corneal Hydration

i Physiologic hydration

Corneal Hydration (mass of water/mass of dry tissue) Fig. 4 (Dougherty, Wellish, and Maloney). The calculated effect of hydration on refractive outcome after photorefractive keratectomy for attempted cor­ rections of 3, 6, and 12 diopters. If the ablation is performed at physiologic hydration (dashed vertical line), the attempted and achieved corrections are equal. If the cornea is dehydrated, then overcorrection results; conversely, superhydration leads to undercorrection.

below physiologic hydration will be overcorrected. Greater disparities of hydration from physiologic and greater attempted dioptric cor­ rections led to greater overcorrections or undercorrections.

Discussion Seiler and associates 9 and Waring10 found that increasing corneal hydration increases the hy­ drated tissue ablation rate, measured in mi­ crometers removed per pulse, at hydrations below physiologic levels. The data of Seiler and associates showed that the amount of water, collagen and ground substance removed in­ creases as the corneal water content increases. Our data also indicate that the hydrated tissue ablation rate increases with hydration. This might represent increased removal of water, increased removal of solid material, or both. We hypothesize that as corneal hydration decreas­ es, less water is carried off passively with ablat­ ed collagen and ground substance, decreasing the hydrated tissue ablation rate (Fig. 5). Our data indicate that the dry component ablation rate increases with decreasing hydra­

173

tion. This finding suggests that water in the corneal stroma masks the effectiveness of the laser energy in removing the dry component. This is curious, since water does not absorb the laser energy at 193 nm. One possibility is that dissolved ions might act as absorption sites for laser energy. 16 Nevertheless, the wet stromal component somehow attenuates the laser ener­ gy because increased amounts of water de­ crease the amount of dry mass removed. De­ spite the higher dry component ablation rate, corneas with lower hydration would have a lower hydrated tissue ablation rate because of the decreased water available to be carried off passively (Fig. 5). Variation in corneal hydration will affect the hydrated tissue and dry component ablation rates in an opposite manner. The question aris­ es of which ablation rate is important in deter­ mining refractive outcome. We believe that the amount of collagen and ground substance re­ moved (dry component ablation rate) is more important than hydrated tissue ablation rate in determining refractive outcome, on the basis of both theoretical grounds and clinical data. The cornea returns to physiologic hydration when the epithelium heals, because the factors controlling hydration, such as stromal swelling pressure, membrane integrity, evaporation, and intraocular pressure, 11 are unchanged by the laser ablation. Since corneal water content returns to a physiologic level (3.45 mg of water per milligram of dry tissue) after epithelial healing, the amount of corneal solids removed rather than the amount of water plus corneal solids removed will determine the initial refrac­ tive result before wound healing. Therefore, dry component ablation rate should be more important than hydrated tissue ablation rate in determining the refractive outcome. Clinical data also support the possibility that dry component ablation rate is most important in determining refractive outcome. In a study by Maguen and associates, 17 when nitrogen purge gas was used to remove the laser ablation plume, a significant number of overcorrections resulted. When the purge gas was discontinued, the number of overcorrections decreased signif­ icantly. This phenomenon may be explained by assuming that the nitrogen purge gas dehy­ drates the cornea after epithelial scraping and increases the ablation rate that determined the refractive outcome (dry component ablation rate), resulting in an increased incidence of overcorrections (Fig. 5). If hydrated tissue abla­ tion rate determined refractive outcome, the

174

August, 1994

AMERICAN JOURNAL OF OPHTHALMOLOGY

Physiologic Hydration

Dehydration 1 1 1 M 1 1 M l oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo

1 I I I I 1 I II

oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo

m

m

m

en

m

oo oo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo

LTPt oo oo oo oo oo oo oo oo c IIM II N oooooooooooooo oooooooooooooo

B CO. _m oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo oooooooooooooo

oo oo oo oo oo oo oo oo oooooooooooooo oooooooooooooo

D=

Epithelial

~=

Collagen/ground

oooooooooooooo oooooooooooooo oooooooooooooo

oo oo oo x oo oooooooooooooo

cells substance

Water

i-

Laser

energy

Fig. 5 (Dougherty, Wellish, and Maloney). This diagram shows the effect of the laser energy on a cornea at physiologic hydration at the time of laser ablation (column left) compared to a cornea that is dehydrated from nitrogen purge gas or excess time from scraping to ablation (column right). Row A represents a portion of the anterior aspect of each cornea before epithelial scraping. Row B represents each cornea just before laser ablation. Row C represents each cornea just after laser ablation with the resultant loss of collagen and ground substance (~) and corneal water (o). Using the assumptions from Figure 1, the normally hydrated cornea would have a dry component ablation rate of 30 mg/pulse and a hydrated tissue ablation rate of 110 mg/pulse. The dehydrated cornea would have a larger dry component ablation rate of 60 mg/pulse because less water is present to attenuate the laser energy. The hydrated tissue ablation rate of 100 mg/pulse is smaller than that of normally hydrated cornea because less water is available to be carried off passively with each laser pulse. Row D shows each cornea after healing. Because more collagen/ground substance is removed from the dehydrated cornea than the normally hydrated cornea, the dehydrated cornea will have a greater myopic correction than the normally hydrated cornea when the epithelium heals and corneal hydration returns to a physiologic level.

Vol. 118, No. 2

Excimer Laser and Corneal Hydration

data of Seiler and associates 9 and Waring10 would predict a decreased ablation rate and a smaller refractive change as the cornea dehy­ drates with the use of nitrogen purge gas. This prediction contradicts the clinical findings of Maguen and associates 17 of increased overcorrections with decreasing hydration, implying that dry component ablation rate determines refractive outcome in excimer photorefractive keratectomy. In previous studies, three methods were de­ scribed for calculating the excimer laser abla­ tion rate. One involves counting the number of pulses needed to perforate a cornea of pachymetrically determined thickness. 1819 The pachymetric method averages the ablation rate of the anterior and the posterior stroma. The posterior stroma is more hydrated than the anterior stro­ ma and should therefore have a different abla­ tion rate. 20 The ablation rate of anterior stroma should determine the refractive change. Another method for determining the ablation rate involves measuring the depth of ablation with a stage micrometer on a light microscope and plotting the result against the pulse count. The slope of the plot yields the ablation rate. 16 This focusing method is complicated by subjec­ tivity in determining the depth of ablation, since any inhomogeneity in laser beam would cause an irregular ablation trough. The third method involves measuring exci­ sion depth histologically and dividing by the number of pulses. 8,21 The histologie method is confounded both by the imprecision of measur­ ing excision depth because of peaks and valleys in the ablation trough and by tissue shrinkage or swelling from fixation. None of the three previously described meth­ ods can be used to calculate a dry component ablation rate. The current method of determin­ ing ablation rate based on mass removed elimi­ nates subjectivity in the measurement of tissue ablated and allows the determination of dry tissue removed by the laser. Our calculated hydrated tissue ablation rate of 0.40 μπι/pulse in bovine stroma at physio­ logic hydration is significantly lower than the hydrated stromal tissue ablation rate of 0.55 ± 0.1 μπι/pulse in human cornea found by Seiler and Wollensak. 3 Our lower hydrated stromal tissue ablation rate is, in part, caused by our lower radiant exposure (180 mj/cm 2 vs 205 mj/cm 2 ), but it may also reflect interspecies differences in corneal collagen or ground sub­ stance. Clearly, one must be cautious in directly

175

translating findings in bovine cornea to human cornea. However, bovine cornea offers a useful model to study the effect of excimer laser-tissue interaction. 161819 Advantages of bovine cornea include presence of a histologically distinct Bowman's layer,22 easy availability of fresh tis­ sue, and large size. The size of the bovine corneoscleral rim allowed us to trephine both donor and control buttons from the same eye. Our calculations demonstrating that dioptric outcome varies with hydration have significant clinical implications in photorefractive keratec­ tomy. Our data imply that any delay in tissue ablation after de-epithelialization might cause dioptric overcorrection, because the resulting evaporation in the anterior stroma would lead to a higher-than-expected dry component stro­ mal ablation rate. The use of nitrogen purge gas would increase the amount of overcorrection. The in vivo effect of evaporation on corneal hydration and refractive outcome in photore­ fractive keratectomy requires further study. Our current findings imply that each surgeon should develop a reproducible technique for performing photorefractive keratectomy that allows for equal stromal dehydration from pa­ tient to patient. The laser corneal surgeon may then need to examine the refractive outcome to obtain the mean refractive deviation from that anticipated. This information could be used to modify the attempted correction to compensate for errors in outcome induced by the changes in hydration specific to the particular technique. ACKNOWLEDGMENTS

Statistical consultation was provided by Lida Hadjiaghia, M.S., Department of Biomathematics, UCLA School of Medicine. Bartly J. Mondino, M.D., Michael O. Hall, Ph.D., and Suraj P. Bhat, Ph.D., Jules Stein Eye Institute, provid­ ed assistance with laboratory techniques.

References 1. Trokel, S. L., Srinivasan, R., and Braren, B.: Excimer laser surgery of the cornea. Am. J. Ophthalmol. 96:710, 1983. 2. McDonald, M. B., Frantz, J. M., Klyce, S. D., Beuerman, R. W., Varnell, R., Munnerlyn, C. R., Clapham, T. N., Salmeron, B., and Kaufman, H. E.: Central photorefractive keratectomy for myopia. The blind eye study. Arch. Ophthalmol. 108:799, 1991. 3. Seiler, T., and Wollensak, J.: Myopic photore­ fractive keratectomy with the excimer laser. Oneyear follow-up. Ophthalmology 98:1156, 1991.

176

AMERICAN JOURNAL OF OPHTHALMOLOGY

4. Sher, N. A., Chen, V., Bowers, R. A., Frantz, J. M., Brown, D. C , Eiferman, R., Lane, S. S., Parker, P., Ostrov, C , Doughman, D., Carpel, E., Zabel, R., Gothard, T., and Lindstrom, R. L.: The use of the 193-nm excimer laser for myopic photorefractive keratectomy in sighted eyes. A multicenter study. Arch. Ophthalmol. 109:1525, 1991. 5. Eiferman, R. A., O'Neill, K. P., Forgey, D. R., and Cook, Y. D.: Excimer laser photorefractive kera­ tectomy for myopia. Six-month results. Refract. CornealSurg. 7:344, 1991. 6. Gartry, D. S., Kerr-Muir, M. G., and Marshall, J.: Excimer laser photorefractive keratectomy. 18month follow-up. Ophthalmology 99:1209, 1992. 7. Gartry, D. S., Kerr-Muir, M. G., Lohman, C. P., and Marshall, J.: The effect of topical corticosteroids on refractive outcome and corneal haze after photo­ refractive keratectomy. A prospective, randomized, double-blind trial. Arch. Ophthalmol. 110:944, 1992. 8. Seiler, T., Kriegerowski, M., Schnoy, N., and Bende, T.: Ablation rate of human corneal epithelium and Bowman's layer with the excimer laser (193 nm). Refract. Corneal Surg. 6:99, 1990. 9. Seiler, T., Fantes, F. E., Waring, G. O., and Han­ na, K. D.: Laser corneal surgery. In Waring, G. O. (ed.): Refractive Keratectomy for Myopia and Astig­ matism. St. Louis, C. V. Mosby, 1992, p. 690. 10. Waring, G. O.: Development of a system for excimer laser corneal surgery. Trans. Am. Ophthal­ mol. Soc. 87:854, 1989. 11. Dohlman, C. H.: Physiology of the cornea. Corneal edema. In Smolin, G., and Thoft, R. A. (eds.): The Cornea. Scientific Foundations and Clini­ cal Practice. Boston, Little, Brown and Co., 1987, pp. 3-16. 12. Ehlens, N.: The fibrillary texture and the hydration of the cornea. Acta Ophthalmol. 44:620, 1966.

August, 1994

13. Hedbys, B. O., and Mishima, S.: The thickness-hydration relationship of the cornea. Exp. Eye Res. 5:221, 1966. 14. Munnerlyn, C. R., Koons, S. J., and Marshall, J.: Photorefractive keratectomy. A technique for laser refractive surgery. J. Cataract Refract. Surg. 14:46, 1988. 15. Duane, T. D.: The steady state of corneal hydration. Am. J. Ophthalmol. 32:203, 1949. 16. Van Saarloos, P. P., and Constable, I. J.: Bo­ vine corneal stroma ablation rate with 193-nm ex­ cimer laser radiation. Quantitative measurement. Re­ fract. Corneal Surg. 6:424, 1990. 17. Maguen, E., Nesburn, A. B., Papaioannou, T., Salz, J. J., Macy, J. I., and Warren, C : 193-nm ex­ cimer laser photorefractive keratectomy. Short-term visual rehabilitation with and without nitrogen flow. Refract. Corneal Surg. In press. 18. Puliafito, C. A., Wong, K., and Steinert, R. F.: Quantitative and ultrastructural studies of excimer laser ablation of the cornea at 193 and 248 nanome­ ters. Lasers Surg. Med. 7:155, 1987. 19. Krueger, R. R., and Trokel, S. L.: Quantitation of corneal ablation by ultraviolet laser light. Arch. Ophthalmol. 103:1741, 1985. 20. Turss, R., Friend, J., Reim, M., and Dohlman, C : Glucose concentration and hydration of the cor­ neal stroma. Ophthalmic Res. 2:253, 1971. 21. Aron-Rosa, D. S., Boulnoy, J. L., Carre, F., Delacour, J., Gross, M., Lacour, M., Olivo, J. C , and Timsit, J. C : Excimer laser surgery of the cornea. Qualitative and quantitative aspects of photoablation according to energy density. J. Cataract Refract. Surg. 12:27, 1986. 22. Gelatt, K. N.: Textbook of Veterinary Ophthal­ mology. London, Bailliere Tindall, 1981, p. 26.