Thermal Load from a CO₂ Laser Radiant Energy Source Induces Changes in Corneal Surface Asphericity, Roughness, and Transverse Contraction

Abstract

Purpose.: We examined corneal surface response to an isolated thermal load.

Methods.: Cadaveric porcine eyes were pressurized and stabilized for processing and imaging. A carbon dioxide (CO₂) laser (1.75 W) delivered a uniform disk of continuous wave thermal radiant energy to the exposed corneal stromal surface without ablation. Thermal load was determined by measuring corneal surface temperature during CO₂ laser irradiation. Corneal profilometry was measured with broad-band optical interferometry before and after CO₂ laser irradiation. Photomicrographs of the stromal surface were taken before and after irradiation, and the images were superimposed to examine changes in positional marks, examining mechanical alterations in the stromal surface.

Results.: Thermal load from uniform laser irradiation without ablation produces central corneal steepening and paracentral flattening in the central 3-mm diameter. Q values, measuring asphericity in the central 2 mm of the cornea increased significantly and it was correlated with the temperature rise (R ² = 0.767). Surface roughness increased significantly and also was correlated with temperature rise (R ² = 0.851). The central stromal surface contracted and underwent characteristic morphologic changes with the applied thermal load, which correlated well with the temperature rise (R ² = 0.818).

Conclusions.: The thermal load created by CO₂ laser irradiation creates a characteristic spectrum of morphologic changes on the porcine corneal stromal surface that correlates to the temperature rise and is not seen with inorganic, isotropic material. The surface changes demonstrated with the CO₂ laser likely are indicative of temperature-induced transverse collagen fibril contraction and stress redistribution. Refractive procedures that produce significant thermal load should be cognizant of these morphologic changes.

Introduction

Excimer laser ablation, using argon-fluoride (ArF) excimer wavelength (193 nm) laser radiation to change surface shape of the human cornea by ablative tissue removal, has been used successfully for more than 20 years. Thermal load associated with excimer ablation historically has been considered to be negligible. Many refractive procedures use thermal load directly to change the shape of the cornea with laser and radiofrequency radiation. Thermal load during the excimer ablation has become a greater concern with high fluence, and high repetition pulse rates. ¹ Elevated temperatures of the corneal surface have been demonstrated with excimer ablation. ¹ Furthermore, corneal collagen denaturation and alteration of corneal stroma have been demonstrated at those temperatures. ² Thermal load is an important parameter to be considered in any procedure on the cornea, and quantifying its effects is essential to understanding processes in which very subtle alterations can affect the quality of vision dramatically. 

Thermal energy denatures the triple helix structure of the corneal collagen, creating a contractile force along its lengthwise dimension. ² Thermally denatured collagen can shrink between 8% and >30% of its original length as seen with CO₂ laser studies. ³ Collagen contraction and denaturation without thermal necrosis begins at 40°C and achieves a maximum at approximately 62°C. ^2,4,5 The human cornea has been shown to be heated briefly to 60–70°C during ablation with the ArF excimer laser. ⁶  

A thermal response was noted with the excimer laser at its inception, but it was believed to be negligible. ⁷ Studies also show a thermal response to a brief temperature rise associated with the excimer laser, which was thought previously to be strictly ablative in nature. ^6,8 This temperature rise is caused by ablation fluencies used in broad-beam and flying spot lasers. ⁸ Other studies have shown this thermal response to be magnified with increasingly desirable pulse rates and laser fluencies. ⁹ Some lasers running in excess of 1000 Hz and 500 mJ/cm² are capable of reaching maximal sustained temperature increases of almost 20°C. ¹  

High frequency pulse rates in excess of 500 Hz are being used commercially more frequently. A thermal effect and potential tissue damage reducing vision quality may be associated with high frequency pulse rates. ^6,8,10 Baseline corneal surface temperature has been measured between 29°C and 32°C before ablation. ^11,12 Researchers have seen temperature elevations with scanning lasers of 6–8°C with a fluence of 118 mJ/cm² at 30 Hz and from 6–12°C with 81 mJ/cm² at 30 Hz. ^11,13 These researchers looked at temperatures averaged over long time constants. However, Ishihara et al. examined the temperature elevation in the microsecond domain and noted temperature elevations of 112°C. ⁸ The stromal architecture is modified at several temperatures beginning at 45°C with high thermal damage noted at 62 degrees. Some have indicated that temperatures during excimer ablation exceeding 40°C may cause damage. ^6,12 Linear corneal shrinkage was noted to be up to 2% percent at temperatures from 40–60°C and from 2%–25% at 60–70°C. ¹⁴  

The response of enucleated porcine corneal tissue is similar enough to live rabbit or live human eyes to justify their use and is supported in many studies. ^{6,8,9,15–18} There is evidence of increased amplitudes in the irregularities seen using enucleated eyes, possibly from increased hydration of the tissue. ¹⁶ Q values have been used clinically and mathematically to evaluate asphericity changes in living human corneas and mathematical models. ^19–25 Studies also have looked at surface roughness following excimer ablation as a cause of subepithelial haze, and suggested some interaction between the laser and residual corneal tissue. ²⁶ The residual unablated corneal tissue is noted to have numerous histologic changes suggestive of an interaction between the excimer laser and the residual corneal stromal surface. ^9,16,27–31 Histologic changes include a thin “coagulation layer” on the residual surface and irregular stromal “undulations” with elevated temperature of the stroma. ^4,32,33  

We examined and quantified the effects of thermal load applied to the surface of the porcine corneal stromal tissue. Our study is designed specifically to look at changes in central surface shape, structure, and roughness, while correlating them with thermal load. The ultimate purpose was a clearer understanding of this interaction to facilitate predictive models that are able to design better future excimer lasers using altered parameters, such as increased pulse rates and fluencies. Also, these findings will help facilitate predictive models for other modalities used to alter the shape of the corneal surface, such as laser thermo-keratoplasty and ultraviolet exposed vitamin A, which rely directly upon collagen contraction. 

Methods

We are in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Whole fresh cadaveric porcine eyes were used less than 24 hours postmortem, stored in ice with preservative solution. All eyes were corneal disease- and injury-free, and the corneas were visually clear. The eyes were heated, via water bath, to 29 degrees Celsius just before use. A corneal pachymeter (Medtronic, Minneapolis, MN) was used to measure the central thickness, which was 1002 ± 71 μ. The porcine eyes were stabilized and maintained in a constructed chamber leaving the cornea and surrounding sclera exposed (Fig. 1). The eyes were pressurized in the chamber with an intravenous gravity feed infusion set-up of lactated Ringer's solution into the ocular vitreous. Intraocular pressure was maintained at 16 mm Hg and verified by pneumotonometry, correcting for corneal thickness (Model 30; Medtronic). 

Figure 1. 

 

Globe stabilization chamber with infusion.

A model Excellus Hansatome microkeratome (Bausch & Lomb, Rochester, NY) with an 8.5 mm suction ring was used to create a corneal (LASIK type) flap of 180 μ thickness. This hinged flap was laid back to expose the bare stroma for ablation and measurement. At other times the flap was replaced to prevent surface desiccation and stromal edema. Additionally, hydration with basic salt solution in a sealed bag around the stabilizing chamber was used to prevent desiccation. 

A carbon dioxide (CO₂) laser (10.8 μm) delivery system (Fig. 2) was designed and constructed using a continuous Synrad 20 W CO₂ laser model 48-0 (Synrad, Inc., Mulkilteo, WA). The CO₂ laser was created to provide uniform irradiance of thermal load to the anterior corneal stroma, without ablation. The theoretical Fresnel number of the optical system was F = 9.8, negating most of the effects of diffraction. It was designed to deliver a 4 mm diameter nearly constant fluence profile to the surface of the cornea, and thereby produce a disk of stromal temperature rise and related surface collagen contraction. The designed temperature rise was such that it would model the thermal load from the excimer laser without ablation. Temperature elevation was designed to achieve a temperature change of 0–14°C in the surface contraction series of 21 samples, and of 5–19°C in the asphericity and surface morphology series of 8 samples. A radially symmetric 3.5 mm diameter Gaussian beam was transmitted through a 3× CO₂ laser beam expander and then through a 4 mm aperture to achieve a nearly constant fluence profile over the 4 mm diameter. Finally, the beam was redirected vertically to the target corneal stroma. A diode laser (<0.1 mW) was used to center the target cornea. Laser output at the target was measured at 1.75 W over a 4 mm diameter area. A trigger switch and timer maintained the set irradiation time, thereby controlling temperature rise. Irradiation times were varied between 0.5 and 10 seconds. As there are significant variations in a given sample's absorption of the radiant energy and the timing of the peak emission thermal measurement, the temperature rise may be ±1°C for any two identical exposure times. 

Figure 2. 

 

CO₂ laser (10.8 μm) delivery system (1.75 W).

Stromal surface topography was measured on eight prepared porcine eyes before and after CO₂ laser irradiation, while maintaining the orientation of the eye. A broad-band optical interferometric profilometer was used to produce very high resolution (sub-micrometer) images of the central corneal stromal surface (model NT9800; Veeco, Tucson, AZ). This testing was completed to determine changes in profile and closest spherical fit as well as surface roughness. The profilometer software was used to provide raw profile data as well as noise reduced data with a Gaussian Fourier filter and tilt adjustment. 

The conic constant Q value was used as an indicator of asphericity of the central corneal stromal surface. The interferometer measured profile was used to calculate the spherical fit of the central 1 mm diameter of the stromal surface and the sag of the surface was determined at a 2 mm diameter. The Q value was calculated from a modification of Baker's equation.  

 

where the shape factor p = 1 + Q, and R is the central 1 mm radius of curvature. 

The surface roughness was determined on eight irradiated eyes from the raw measured profile data and a summed average of 10 randomly chosen places in the central 2 mm of the stroma. The following equation was used to calculate the roughness in micrometers, by determining the absolute value of the raw surface deviation x_k from that of the smoothed surface at n random locations.  

 

 

Transverse contraction (or expansion) of the corneal stromal surface was determined by using a diamond blade to make two parallel opposed positional marks in the stromal surface of all 21 prepared eyes (20 irradiated eyes and 1 unradiated, re-measured eye). The marks were made to a depth of 200 μ, approximately 2 mm apart and centered in the 6 mm ablation crater. A CCD telocentric microscope with measurement software (Amscope, Irvine, CA) was used to produce standardized photomicrographs of the stromal surface maintaining a constant magnification and focal distance. Standardized photomicrographs of the stromal surface were taken before and after uniform CO₂ laser irradiation, and the images were superimposed to examine transverse positional changes in the marks quantifying a transverse mechanical alteration in the stromal surface (Fig. 3). The apex medial and lateral edges of the V-shaped incisions were visible in the photomicrograph. Measurements were made from the medial edge of the incisions (Fig. 4). Measurements of the distance between the two marks before and after irradiation were made at 4 to 5 separate locations by a blinded observer and averaged. 

Figure 3. 

 

Sample of the superimposed images of pared stromal incisions before (blue) and after (red) treatment with uniform CO₂ laser irradiation. Inner edge of incision is marked for clarity. Odd numbered measurements are before treatment and even numbered measurements are after treatment.

Figure 4. 

 

Photomicrograph of stromal corneal incision with diamond blade set at a 200 μ depth. The apex, as well as medial and lateral edges of the V-shaped incision are visible.

Transmission electron microscopy was completed on the post-irradiated corneal stroma to examine morphologic changes in the residual stroma and to examine for possible histologic alterations consistent with temperature elevation or surface contraction. 

The temperature of the corneal surface during CO₂ laser irradiation was determined by evaluating the irradiance recorded from a Radiance 1 (Amber Engineering, Goleta, CA) indium antimonide infrared camera (3–5 μ) running at a 60 Hz frame scan time. Blackbody standardized Matlab software (Mathworks, Natick, MA) was used to convert the irradiance pattern to a thermal map. Infrared camera measurements were completed maintaining a constant image view angle, and focal distance of 45° and 9 cm, respectively (Figs. 5, 6). 

Figure 5. 

 

Infrared camera sample of maximum irradiance image during CO₂ laser irradiation on the stromal surface and its associated thermal map (120 × 170 pixels) after processing. The temperature profile of the surface (Celsius) is provided.

Figure 6. 

 

(A) Sample of maximum temperature profile during 3-second CO₂ laser irradiation. (B) Temperature progression with time during 3-second CO₂ laser irradiation.

To insure the validity of the findings four, 8 mm radius, optical quality, PMMA plastic test spheres were irradiated in an identical fashion, and measured for profile changes, surface roughness, and transverse contraction of surface. Separately, four prepared eyes examined the corneal porcine stroma looking for changes in profile, and transverse contraction of the surface with desiccation over 25 minutes at 25°C and 30% relative humidity. 

Results

Uniform irradiation with the CO₂ laser, producing a thermal load on the porcine cornea without ablation, produces central corneal steepening and paracentral flattening of the corneal “optic zone” (central 3 mm diameter). The curvature of the central 1 mm of cornea increased in all samples, and the curvature in the central 2 mm of the cornea increased to a lesser degree in all samples (Fig. 7). 

Figure 7. 

 

Two corneal stromal profiles showing results before and after CO₂ laser irradiation. (A) Maximum temperature rise 5°C. (B) Maximum temperature rise 10°C. Profiles show a central steepening and peripheral flattening. The graphs are superimposed by setting the average of the each profile to zero.

All samples showed an increase in asphericity when closest spherical fit is applied to the pre- and post-ablated profiles as shown below by example in Figure 8. 

Figure 8. 

 

Corneal stromal profile showing samples of deviation from closest spherical fit before and after CO₂ laser irradiation. (A) Maximum temperature rise 5°C. (B) Maximum temperature rise 10°C. Aspheric profile is magnified greatly following CO₂ laser irradiation.

A total of 20 irradiated eyes all showed contraction (1 unradiated, re-measured eye showed negligible contraction) of the stromal surface following CO₂ laser irradiation to varying degrees, correlated to temperature rise. The central 2 mm of the stromal surface contracted with the applied thermal load in all samples. The contraction correlated well with the central maximum temperature rise using linear regression (R ² = 0.818, Fig. 9). Data showed considerable scatter as would be expected with biologic tissue and complex procedures/measurement techniques. In addition, 18 eyes were measured previously without irradiation to examine variation in measurement technique. The unradiated, re-measured eyes indicated a change of −0.045% ± 0.235% (95% confidence interval [CI]). ³²  

Figure 9. 

 

Percent corneal surface contraction with maximum observed temperature rise on the stromal surface.

Q values, measuring asphericity in the central 2 mm of the cornea, increased significantly following all CO₂ laser irradiated samples. The Q value increase was correlated logarithmically with the temperature rise (R ² = 0.767, Fig. 10). Figure 10 correlation was determined with logarithmic closest fit regression analysis using Microsoft Excel 2007 (Microsoft, Redmond, WA). 

Figure 10. 

 

Change in asphericity over the center 2 mm of the cornea measured by Q value with increasing maximal temperature rise.

Surface roughness increased significantly following CO₂ laser irradiation and was correlated parabolically with temperature rise (R ² = 0.851, Fig. 11). Figure 11 correlation was determined with the closest fit power regression analysis through the origin using Microsoft Excel 2007. 

Figure 11. 

 

Change in stromal surface roughness with increasing maximal temperature rise.

The corneal surface subject to increased thermal load undergoes characteristic morphologic changes beginning with increased asphericity and roughness, then a central reticulated island, and finally large central reticulation with radial folding (Fig. 12). 

Figure 12. 

 

Stromal corneal topography samples imaged with optical interferometry, showing morphologic progression of the corneal surface appearance with increasing thermal load. (A) Pre-irradiation. (B) 5°C maximum temperature rise. (C) 13°C maximum temperature rise. (D) 19°C maximum temperature rise.

Transmission electron microscopy was completed on four post-CO₂ laser irradiated samples to examine morphologic changes in the tissue. Photomicrographs of the ablated stromal surface indicate a 1–5 μ region of abnormal “undulating” collagen fibrils with underlying normal straight layered fibril appearance. Additionally, remnants of an electro-dense “coagulation” layer also are seen on the surface (Fig. 13). 

Figure 13. 

 

Transmission electron microscopy of the irradiated stromal surface after 12°C maximum temperature rise. Note a 5 μ depth region of abnormal “undulating” stromal fibrils and electro-dense “coagulation.” Arrows: layer of morphologic change.

CO₂ laser irradiation of all four optical quality, 8 mm radius, plastic PMMA spheres produced no significant changes (or even measurable) changes in asphericity, surface roughness, or transverse contraction. The spheres were irradiated in an identical fashion to the porcine corneas. Corneal thickness was re-measured after irradiation with the 1.75 W CO₂ laser delivery and no change was noted, confirming that the irradiation was not ablative. The corneal thickness was measured as 1002 ± 71 μ (95% CI) before irradiation and 1041 ± 108 μ after irradiation. 

The stromal surface was examined previously for changes in asphericity and transverse contraction under conditions of moderate desiccation. The surface was not affected significantly by moderate desiccation (25 minutes, 29°C, 30% relative humidity). Shown are desiccation time graphs examining surface contraction. No significant changes were noted in surface contraction of the four samples and changes in profile (Figs. 14, 15). Q values changed insignificantly on four samples from 0–90 minutes of desiccation, from −11.0 ± 9.8 (95% CI) initially to −7.6 ± 8.8 after 90 minutes. ³²  

Figure 14. 

 

Percent contraction with moderate desiccation over 25 minutes at 25°C and 30% relative humidity. Note variations in measurement without significant change in the surface.

Figure 15. 

 

Sample of corneal stromal surface changes in profile with desiccation up to 90 minutes at 25°C and 30% relative humidity. Note there are negligible changes in profile with the possible suggestion of central flattening.

Discussion

Uniform, continuous wave, CO₂ laser irradiation without ablation creates an isolated thermal load and measurable temperature rise in proportion to the exposure time. The thermal load from the CO₂ laser interacts with the stromal surface in a predictable, characteristic and progressive fashion with increased temperature. The thermal load creates a characteristic spectrum of morphologic changes on the porcine corneal stromal surface that correlates to the temperature rise and is not seen with inorganic, isotropic material. Those changes demonstrated include an increase in asphericity, increased surface roughness, surface contraction, and histologic alterations in the superficial stromal collagen fibrils. The surface changes demonstrated with the CO₂ laser likely are indicative of temperature-induced transverse collagen fibril contraction and stress re-distribution. 

Our previous research has shown that uniform excimer laser ablation from a pulsed flying spot source also interacts with the unablated residual stromal surface in a predictable characteristic fashion. ³² The results from pulsed excimer laser ablation are very similar to those seen here with the isolated thermal load from the CO₂ laser. Uniform scanning excimer laser ablation on porcine corneal stroma produces a significant central steepening and a peripheral flattening in the “optic zone” of the cornea. ³² Surface roughness of the bare stroma increased significantly following excimer ablation. The maximum central corneal temperature elevated 5.0°C after 20 seconds of ablation. All excimer ablated corneal samples showed contraction of the stromal surface following ablation. Transmission electron microscopy of the ablated stromal surface indicate a 1–2 μ region of abnormal “undulating” collagen fibrils with underlying normal straight-layered fibril appearance. ³²  

The correlations showed a general trend, which was the purpose of our study. We fitted them with closest fit basic linear or non-linear curves, which may or may not be the true model for the parameter's behavior. To our knowledge, there is no prior model to base the particular linear or non-linear fit, but they are presented as a basis for consideration for future modeling. 

It was assumed originally that a uniformly distributed pulsed excimer laser profile would ablate a uniform thickness of the cornea as it does with isotropic plastic materials, and the result of ablation would be the same as the original surface profile translated to some depth that was a function of total fluence applied. ³⁴ However, it was noted early that the central cornea appeared to be under-ablated significantly, creating a severe asphericity termed “central islands.” ³⁵ These central irregularities were correlated with visual disturbances, such as decreased best corrected visual acuity, glare, and halos. ^36,37 Excimer-induced asphericity has been the accepted reason that the original Munnerlyn formula works well on a plastic sphere, but not as well on a human cornea. Increased asphericity has been shown following commercial excimer myopic correction. Before the use of central overablation algorithms, the incidence of clinically evident central asphericity measured by Placido ring topography in human eyes was between 50% and 88%. ³⁸ Empiric adjustments to the ablation algorithm were made that mostly solved the problem at the time, without a clear understanding of the cause. ^20,39–41 Even with these interventions, research has noted that the largest contributor to post-ablation ocular aberrations, by a factor of 4, is spherical aberration. ⁴² Despite the empiric changes to the excimer ablation algorithms, irregularities persist, albeit less frequently and with less severity. The major causes in symptomatic patients with poor visual outcomes are linked to persistent central corneal irregularities with associated aberrations, mostly spherical aberration. ^42–44 Also, excimer laser ablations still lead to a decrease in the quality of vision, particularly at night with decreased contrast sensitivity. ^43,45,46 Even recent studies highlight the postablative change in asphericity and the need to understand better its etiology. ^23,47–50  

Over the past 15 years, three major theories emerged as to the cause of the apparent “under-ablation”: (1) redeposition of material ejected from the ablation back on the central cornea, ^16,51,52 (2) ejected plume material blocking the subsequent pulse, ³⁶ and (3) corneal edema preferential to the center of the cornea. ^36,51,52 Additional theories have included that the ablation is a function of distance from the ablation edge. ^18,20 Also, transverse energy along the ablated surface interacts with the tissue, decreasing the ablation threshold preferentially in the central cornea. ³⁴ Several other theories have been considered, but have not gained support. ^7,53–55 None of the theories explains adequately the apparent “non-uniform” ablation and some investigators have questioned the validity of these theories. ^20,34,52  

Many researchers examining the excimer laser interaction with the cornea have noted the changes in asphericity clinically, even with aspheric and central island correcting algorithms. ^23,47–50 At least one researcher suggested a thermal response for the central irregularities. ⁵⁶ Q values used to quantify asphericity are useful in our study, but cannot be compared directly to the clinical studies, as the ablation was completed on enucleated porcine eyes in a uniform manner without aspheric correction algorithms and measured the central 2 mm of the optic zone. Noack et al. also quantified the large changes in central apparent “under-ablation” with a uniform excimer ablation, ¹⁶ and our results parallel those findings. The changes in asphericity are not seen with isotropic, homogeneous, plastic materials. This increase in asphericity seen only with corneal tissue supports the theories of preferential central corneal edema and transverse energy on the corneal surface, and that the asphericity is a function of distance from the ablation edge. ^{18,20,34,36,51,52} Also, increases in surface roughness and surface contraction are supportive of this interaction, but likely exclude corneal edema as an etiology. The uniformly ablated cornea increases its central curvature with a relative paracentral flattening out to a 3 mm diameter in the “optic zone,” and has been demonstrated clinically with increased asphericity and central island formation. These clinical alterations have been predicted and treated empirically with all commercially available lasers by central overablation. 

One plausible theory is that corneal asphericity is a consequence of the stress re-distribution in the stromal layer as a result of the thermal load produced by the excimer laser. The thermal load induces disruption of the integrity of the corneal collagen fibrils at the stromal surface, and a subsequent deformation of the surface consistent with the new stress/strain equilibrium represented by a central peak and paracentral flattening. The cornea can be considered a “thick membrane” with little resistance to shear, and as a “thick-walled pressure vessel” where internal swelling (hydrostatic) stresses maintain equilibrium with the collagen fibril network. Modeled in this manner, it conceivably would require little contraction of the surface fibrils to create these deformations. 

Prior theories relate the aspheric effect with excimer ablation to the distance from the edge of the ablation, and it is likely that the same observations would be seen with our CO₂ laser irradiated samples had we used varied diameters. ^18,20  

A theory based upon stress redistribution has been evaluated here by examining the isolated thermal response without ablation or a pulsed laser. The CO₂ laser radiant energy source was used to measure the cornea's response to an isolated thermal source. Additionally, a mathematical model would be helpful. A corneal finite element model is being developed that may confirm a contraction mechanism producing the aspheric changes. This phenomenon and its associated etiology may be of increased importance using excimer lasers with increased thermal load, which has been demonstrated in newer high pulse frequency, high fluence lasers. 

Previous research has noted the increased surface roughness as a possible reason for subepithelial haze. ²⁶ Increased surface roughness following excimer ablation is observed clinically, immediately following ablation, and appears to be uniform across the stromal surface. Our previous raw profile data confirm these observations, which are suggestive of a uniform interaction between the excimer laser and the residual stroma. ³² The thermal load from the non-ablative CO₂ laser in our study also produced increased surface roughness with increased temperature to a similar degree as the excimer laser for an equivalent temperature rise. 

Histologically, “undulations” in the collagen sheets increase progressively from the normal straight horizontal pattern between 50 and 60°C. ⁴ Our transmission micrographs of the excimer ablated surface have shown these undulations to 2 μ below the ablated surface with a normal appearance to the underlying stroma. ³² The post-ablated micrographs also showed remnants of the thin, electrodense “coagulation” layer reported previously. ³³ The “coagulation” layer is consistent with gelatinization reported at temperatures above 70°C. ⁴ Transmission photomicrographs completed on the post-CO₂ laser irradiated corneal stroma also indicated a 1–2 μ region of abnormal “undulating” collagen fibrils with underlying normal straight layered fibril appearance. Additionally, possible remnants of an electro-dense “coagulation” layer were seen on the surface. This is further evidence of temperatures that are able to denature and possibly contract the surface collagen fibrils as seen with excimer laser and the thermal load supplied by the CO₂ laser. 

Contraction of the corneal stromal surface following excimer ablation has been shown in our prior study. ³² Elevated temperatures of the corneal surface have been demonstrated with excimer ablation and contraction of corneal stroma has been demonstrated at those temperatures. Researchers have seen temperature elevations with scanning lasers of 6–8°C with a fluence of 118 mJ/cm² at 30 Hz and 6–12°C with 81 mJ/cm² at 30 Hz. ^11,13 These researchers looked at temperatures averaged over long time constants. However, Ishihara et al. examined the temperature elevation immediately following commonly used fluence excimer pulses in the microsecond domain and measured temperature elevations of 112°C. ⁸ The stromal architecture is modified at several temperatures beginning at 45°C with high thermal damage noted at 62°C. Linear corneal shrinkage was noted up to 2% at temperatures from 40–60°C and from 2%–25% at 60–70°C. ¹⁴ In addition, this study shows temperature induced histologic changes with associated contraction in collagen fibrils, ¹⁴ consistent with the changes seen in the top 2 μ of our corneal samples. The linear corneal collagen shrinkage seen in our prior study was approximately 2%, which correlates well with that seen at temperatures between 40 and 60°C. ³² However, our maximum steady-state temperature was 35°C. With our infrared camera, it was not possible to measure the brief temperature rise following the excimer pulse due to the relatively long sensor integration time. This transient temperature rise likely exceeds 100°C. ⁸ Our histologic findings with excimer samples are consistent with higher temperatures. It is possible that the very rapid thermal-relaxing high temperatures may add effectively several degrees to the surface and is able to conduct this heat 1 to 2 μ into the stromal surface, creating the surface shrinkage seen. 

Desiccation was examined as a possible factor altering the surface profile or creating contraction. It has been shown that desiccation will change the amount of ablation per pulse. As we noted here, it had no significant effect upon asphericity or transverse contraction following excimer ablation. 

The porcine cornea is quite robust compared to the human cornea with a significantly increased corneal thickness, 1000 versus 550 μ, as well as increased tensile strength and modulus of elasticity. ⁵⁷ Central corneal thickness in live pigs has been measured to be considerably less than in cadaveric samples. ⁵⁸ The amplitudes of deformations following excimer ablation to the stromal surface do appear magnified 2-fold in cadaveric eyes when compared to live human eyes. ¹⁸ It could be postulated that our results possibly are attenuated in live human corneas. 

Optical interferometric profilometry is a direct measurement technique that does not rely on the rate of change of surface curvature, and also can image discontinuities of the surface. Interferometric profilometry can provide high resolution (sub-micrometer) surface elevation and lateral imaging of the exposed central corneal stroma within 0.5 mm of the corneal apex, which is not possible with most other topographic measurement techniques. The method currently is not useful clinically due to the time required for measurement. At least two other researchers have used this technique indirectly in the lab. ^16,59  

Acknowledgments

Charles Munnerlyn, Veeco Corporation; Eustace Dereniak, and Thomas Milster, in associations with the OSC Remote Sensing lab, and the Maskless Lithography Lab provided expertise and use of equipment. Tony Day, AHSC imaging facility assisted with the study, and Hodges Eye Care allowed laser use. 

Footnotes

 Disclosure: S.J. McCafferty, None; J.T. Schwiegerling, None; E.T. Enikov, None