The excimer laser has long been used as a tool for altering the refractive properties of the eye by reshaping the anterior curvature of the cornea.1,2 The ultraviolet radiation emitted from the excimer laser breaks up molecular bonds by a photochemical reaction, which causes an ablation of the target tissue. The characteristics of an excimer laser treatment are: 1) precise submicron ablation3, 2) smooth ablated surface, and 3) minimal damage to the adjacent tissue.4 In general, the results of excimer laser surgery for refractive correction have been encouraging, especially in patients with low to moderate myopia.5,6 However, the problem still facing refractive surgeons carrying out photorefractive keratectomy (PRK) are postoperative subepithelial haze formation and, for both PRK and laser in situ keratomileusis (LASIK)7, lack of prediction of refractive outcome. These predicaments are thought to arise from changes in the subepithelial stroma, via deposition of new tissue and/or individual variations in wound healing response.8-12 Regeneration of corneal stromal tissue is an integral part of wound healing, and is predominantly facilitated by the keratocyte, a fibroblast-like cell of neural crest origin located between corneal lamellae. Injury to epithelial cells or keratocytes in one location results in dramatic changes in neighboring keratocytes by virtue of both growth factor interactions13 and the neural-like network of cell-to-cell communications between keratocytes via long cytoplasmic extensions and intercellular gap junctions. Once activated, keratocytes synthesize collagen and glycosaminoglycans, which are thought to give rise to clinically observed loss of transparency.14,15 The extent of collagen and matrix deposition is unpredictable, but in excimer surgery, deeper ablations are reported to produce greater losses of transparency.16
The transparency of the corneal stroma is thought to be due to the regular spacing of the collagen fibrils and their narrow and uniform diameters.17·18 Corneal scar tissue is opaque due to decreased order between collagen fibrils19'21 and after an extended period of healing, transparency improves as ultrastructural order improves.22
Excimer laser surgery induces both transitory haze (which peaks after 1 month) and more persistent or late developing haze, which can remain for many months.11,13 In this study, we aimed to quantify the extent to which disorganized, healing collagen fibrils contribute to persistent corneal haze following PTK.
We treated the right cornea of four young rabbits with PTK using each remaining cornea as a control. Synchrotron x-ray diffraction (XRD) was used to study molecular and interfibrillar spacing within the rabbit stroma. A transmission electron microscope (TEM) was used to provide data from which the percentage transmission of visible light through the newly formed tissue was predicted using the direct summation of fields method.23,24 Corneal haze was objectively determined using a slit lamp-mounted, charged-coupled device (CCD) system.25
MATERIALS AND METHODS
All experimental procedures were carried out in accordance with the ARVO resolution on the use of animals in ophthalmic and vision research. PTK was performed at St. Thomas' Hospital, London, using an Omnimed excimer laser (Summit Technology, Boston, MA) with a wavelength of 193 nm. We treated four young New Zealand White rabbits; typical weight of each rabbit at the beginning of the study was 2.5kg. The pulse energy resulted in a radiant exposure of 180 mJ/cm2 at a pulse frequency of 10 Hz. The beam shape was circular with a fixed diameter of 6.0 mm. Two to three hours before PTK was performed on the right cornea of each rabbit, fentanyl I/M (0.3ml/kg), diazepan I/V (1-2 mg/kg max) and 4 drops of topical 1% proparacaine hydrochloride (right eye) were administered to each animal. During PTK treatment, the animals were placed on a table and a wire Hd speculum was used to hold the eye open. The corneal surface was wiped clean of debris using a swab, and was carefully dried. The beam was aimed and focused on the center of the cornea and the laser activated for 400 pulses (corneal ablation rate approximately 0.25 µm per pulse). Both epithelium and stroma were ablated by the laser. Chloramphenicol antibiotic drops were applied to the wounded eye immediately after surgery and were continued for 3 days.
Rabbits were sacrificed with an intravenous injection of euthatal at 3, 8, 12, and 19 months. Representative samples of corneas with persistent haze following PTK were studied in more detail following dissection at 8 and 12 months. Titled M8 and M 12, the dissected corneas (PTK-treated and control) were frozen in liquid nitrogen-cooled isopentane and stored at -400C until required for x-ray diffraction (XRD). The frozen tissue was allowed to thaw before being placed into an x-ray beam or fixed in glutaraldehyde for transmission electron microscopy (TEM) examination; we have shown previously that freezing and thawing does not alter the collagen spacings in the cornea.26
Corneal haze was quantified in vivo at St. Thomas' Hospital, London by the use of an in-house CCD camera system mounted on a Haag-Streit slit lamp and connected via a frame grabber (Wild Vision, Hawk V 10, London, UK) to an Acorn computer (Archimedes 440, London, UK).25 The slit lamp was positioned with a 40° angle of illumination from the temporal aspect of the cornea, and care was taken to keep corneal light reflection out of the central field. Haze was graded by digitizing the images and analyzing them with in-house software, expressing the haze in relative units (gray levels). Haze was measured directly before PTK to establish a baseline and was then measured at 16 (n=4), 24 (n=4), 36 (n=4), and 570 (n=l) days following PTK.
Synchrotron X-ray Diffraction
Synchrotron x-ray diffraction was carried out at the Central Laboratory of the Research Councils synchrotron facility at Daresbury, UK. The corneas were defrosted and full thickness corneas and anterior lenticules were surgically dissected by an ophthalmic surgeon (A. Brahma) at the synchrotron site and then studied. The anterior lenticules were approximately one-quarter of the total corneal thickness. The corneas (either full or anterior sections) were mounted between Mylar windows in clear plastic cells, which were airtight to avoid tissue dehydration. Plastic cells were placed in the x-ray beam so that the x-rays passed through the center of the tissue, along the optical axis.
Low-angle patterns were collected at Station 2.1, using a fixed camera length of 6 m, an exposure time of 3 minutes, and a 3xl-mm beam size. Rat-tail tendon was used to calibrate data.27 Diffraction patterns were collected on a multi- wire gas proportional detector. The interfibrillar Bragg spacing was calculated from the position of the innermost equatorial reflection after background subtraction.28
Wide-angle patterns were collected at Station 7.2, from full thickness PTK-treated and control corneas, with a camera length of 11 cm, an exposure time of 3 min, and a 200-pm-diameter collimated beam. Wide-angle patterns allow measurements to be taken of the spacing between constituent molecules within the collagen fibrils.29 The 0.305-nm spacing in calcite was used to calibrate data. Diffraction patterns were collected on a MAR image plate detector.
Transmission Electron Microscopy
The corneal tissue was fixed overnight (12 hrs) in 2.5% glutaraldehyde, 0.1M phosphate buffer (pH 7.2) at 4°C, followed by 1.5 hours in 0.1M osmium tetroxide at 20°C. The central 2 mm2 of the anterior stroma was studied using a Jeol 1010 transmission electron microscope. The anterior stroma was defined as an area extending to a depth of approximately 20 µp? below the epithelium; this ensured measurements were taken from newly deposited tissue and not the original tissue. Quantitative image analysis was achieved using a bottom-mounted TEM digital camera (Kodak MegaPlus, Model 1.4i) connected to a personal computer equipped with the image analysis software, AnalySIS 2.1 (Soft Imaging Systems, Münster, Germany).
The diameter and relative positions of the collagen fibrils in cross section were determined by image analysis from at least ten micrographs from the anterior stroma at the center of each cornea. Each micrograph analyzed contained approximately 500 fibrils in cross section. Once collected by digital camera, images were cleaned through a series of algorithms that reduced background noise while retaining the size and position of each fibril. These images were then binarized, allowing fast computation of the relative fibril positions and their radii. This information was then used to calculate a radial distribution function, gir), a mathematical description of the positions of the fibrils with respect to one another, and also to predict the transmission of visible light through the healing anterior stroma using the direct summation of fields method.23,24
Calculation of Radial Distribution Function
The radial distribution, g(r), is a statistical measurement of the average number density of fibril centers at a given distance, r, from any other fibril center, relative to the bulk fibril number density p, resulting in a histogram of g(r) plotted against an increasing r.m30 was calculated from the fibril positions taken from TEM digital images and is an important parameter in calculating the percentage transmission through the cornea.23,24 A distinct peak followed by smaller undulations and eventual stability displayed by a typical radial distribution histogram reflects the short-range order in the structure of the cornea, with neighboring fibrils being relatively uniformly spaced, but without any longrange order. The position of the primary peak ing(r) was used to determine the average interfibrillar spacing of collagen fibrils in cross-section. Comparison of the interfibrillar spacing determined from g(r) with the corresponding value determined from an anterior lenticule by XRD, facilitated a recalibration of the primary peak, g(r), assuming any shrinkage caused by the EM tissue processing occurred equally in all directions.31
Transparency: Direct Summation of Fields Model
The model requires a number of assumptions. Estimates for the refractive indices of the fibrils (nf) and ground substance (ng) are required. The refractive index ng has the value 1.357 in normal rabbit cornea.31 nf is a function of the fibril hydration, if hydration increases, nf will fall. In normal rabbit cornea, nf= 1.416.31 The degree to which refractive indices are altered in ablated tissue will remain unknown until the wounded stroma has been studied in more detail. In the current work, the refractive indices have been assumed to remain normal. All parameters used in the calculations are assumed uniform throughout the tissue.
Details of the calculation are described in original papers23,24 so only an outline of the method is given here. First, the size and position of each fibril is measured from a digital electron microscope image of the fibrils in cross section. The image area is then divided up into a grid and the fibrils are grouped together according to which grid-element they occupy. The field scattered by each fibril is dependent on its radius and therefore varies from one fibril to the next. Taking into account the phase differences introduced by the different fibril locations, the scattered fields from all the fibrils in a grid-element are superimposed. The differential cross section per fibril, σ(Θ^sub s^) is calculated by superimposition, and describes the scattering at one given angle, Θ^sub s^, for all the fibrils in one grid-element. The process is repeated for each grid-element and the ensemble average over the whole grid represents the mean differential scattering cross section per fibril for the entire stroma. Integration from Θ^sub s^ = 0 to 2π radians gives the total scattering cross section per fibril, s, from which the transmittance, F^sub t^ can then be calculated using Equation 1.
Using the objective haze measuring system, the corneas showed an increase in baseline haze, peaking at 24 days after PTK, and remaining elevated for the rest of the experiment (approximately 19 months). This increase was statistically significant (Student t-test, P<.001) when haze values before PTK were compared to haze measurements at either 16, 24, or 35 days (Fig 1).
The corneas of rabbit M8 were prepared for electron microscopy as described in the methods section. The center of the PTK-treated cornea showed a darkly stained, discontinuous basement membrane of variable thickness, a large amount of irregularly orientated collagen fibrils, and no distinct lamellae in an area immediately below the epithelium (Fig 2a). Further into the healing rabbit stroma, fibrils displayed a greater degree of axial alignment (Fig 2b). This differed sharply with the control, which showed a continuous basement membrane of even thickness, a well-ordered spatial arrangement, and axial alignment of fibrils separated into defined lamellae (Figs 3a and b). Keratocytes populated the wounded area but their state of activity could not be judged accurately due to freezing of the tissue before fixation, which distorted the cell morphology. Quantitative image analysis of mean fibril diameter between the wounded area of M8PTK and M8CONT showed no significant change (Table).
Data from low-angle x-ray diffraction gave information on the average interfibrillar Bragg spacing through the anterior stroma of control and PTKtreated corneas. The interference function peaks (obtained from the regular alignment of the collagen fibrils) from the intensity profiles of M8 and M 12 showed an increase in the interfibrillar Bragg spacing for each of the PTK-treated corneas when compared to their respective controls (M8CONT = 57 ± 1.71 nm, M8PTK = 61 ± 1.83 nm, M12CONT = 52 ± 1.56 nm, M12PTK = 59 ± 1.77 nm, where ± indicates the measurement error of the peak position) (Fig 4). The spread of these interference function peaks was also measured by dividing the full height of each peak by its full width at half height. For M8 this ratio equaled 1.53CONT and 1.28PTK, and for M12, 2.39CONT and 1.5PTK.
Figure 1. Haze remains significantly higher for over 1 month following PTK and does not return to normal even after healing for approximately 19 months. Numbers in brackets correspond to the number of days following PTK.
Data from the high-angle camera gave information on the average intermolecular Bragg spacing within the collagen fibrils. The intermolecular spacing of the full corneal thickness PTK-treated and control corneas were compared but no difference was noted (1.36 nm control and 1.40 nm PTKtreated).
Radial distribution functions were calculated from TEM images taken from the subepithelial region of M8. Each radial distribution function was normalized by dividing g(r) by the bulk fibril number density, ρ for the whole area covered. The spread of the primary peaks was then compared by measuring the ratio of full peak height (above 1) to the full width at half height. The ratio equaled 0.04 for M8CONT whereas the spread from the equivalent peak M8PTK was 0.02 (Fig 5).
The effect of 8-month-old PTK wounds on transparency through the cornea was assessed using the direct summation of fields light scattering model. Earlier time points following PTK were not assessed because the wound is unlikely to have attained its original thickness until 6 months after surgery.32 The positions and radii of the collagen fibrils were determined from the TEM images used for the radial distribution function calculations. These images were re-scaled by comparing the interfibrillar spacing from XRD to the position of the g(r) peaks to compensate for any shrinking of the tissue during processing for electron microscopy.
Figure 2. A) Eight months after PTK, electron micrographs show disorganized collagen fibrils in the anterior stroma of rabbit M8 (bar = 500 nm). The tissue also displays a wavy, discontinuous basement membrane of irregular thickness (arrow). Large interfibrillar spaces are also visible (A). B) At a higher magnification the fibrils appear irregularly spaced in the anterior stroma (bar= 100 nm).
The radial symmetry of each TEM image was assessed by plotting the relative displacements of each fibril from every other fibril around a common origin and then dividing the resulting circle's widest diameter by its diameter perpendicular to that. Therefore, any value over 1 has a degree of radial asymmetry. The results shown in Figure 6 suggest that the PTK-treated corneas contain some radial asymmetry (M8PTK = 1.2) whereas the control anterior stromas appear to be radially symmetrical (M8CONT = 1.0).
The direct summation of fields method was used to calculate the percentage transmission through the scar tissue (assuming a tissue thickness of 0. 1 mm, the approximate depth of ablation for the PTK-treated stroma, and therefore the assumed maximum thickness for newly synthesized collagen). To allow a straightforward comparison between the PTK-treated and untreated corneas, the control was assumed to have a corneal thickness of 0.1 mm. The results showed that the percentage transmission through the anterior of PTK-treated rabbit corneas at 500 nm is 96% (Fig 7). This value represents a decrease of percentage transmission in visible light through the PTK-treated corneas of only 2% (control registering 98%). Therefore, there exists an unremarkable difference in percentage transparency, as calculated by the direct summation of fields method between control and PTK-treated corneas after 8 months.
Figure 3. A) At 8 months, the unwounded left cornea (control) of rabbit M8 (bar = 500 nm) displays a smooth, continuous basement membrane of constant thickness (arrow). Unlike the PTK-treated cornea the collagen fibrils have a regular arrangement, ordered into lamellae. B) The fibrils are of a similar diameter to the PTK treated corneas but are more regularly spaced (bar = 1 00 nm) in the anterior stroma.
Difference* Between Mean Fibril Diameters of PTK and Control Rabbit Corneas by Quantitative Image Analysis
It should be noted that the objective haze measurements consisted of both back scattered and reflected light and in patient studies, individuals whose haze was predominately induced by scatter showed more degradation in visual function tests than those whose haze was mostly due to reflection.25 In these patients, an increase in back scattered light was accompanied by a corresponding increase in forward scattered light. Thus, all haze was likely to be accompanied by an effective decrease in tissue transparency.
Figure 4. Low-angle x-ray diffraction profiles from the central anterior stroma of M8 and M12 control and PTK-treated corneas. The peak position of both PTK-treated corneas indicates a lower average interfibrillar Bragg spacing compared to their respective controls.
Figure 5. The lower height of the normalized radial distribution function from the PTK-treated cornea of rabbit M8 Indicates a wider variation in interfibrillar spacing than the control, and the peak also has a slightly higher r-value, which indicates a wider average interfibrillar spacing.
The results from the study of haze in PTK-treated corneas showed a marked response that peaked after 24 days. At similar time points (approximately 1 month after laser treatment) previous studies have shown that the subepithelial layer is highly vacuolated and the influx of cells to the wounded area is at a maximum.3335 However, between approximately 1 and 19 months the haze value of the remaining PTK-treated cornea remained at least 49% higher than before PTK. Haze of this kind, which remains many months after surgery, is termed persistent haze and it is during this period (8 and 12 months) collagen fibril organization and corneal transparency were studied.
Electron micrographs from an 8-month-old wound indicated that directly below the epithelium, the fibrils were neither parallel nor axially aligned, nor enclosed by discrete lamellae but, moving deeper into the stroma, a more ordered arrangement of collagen progressively arose. The interfibrillar spacing was increased compared to normal. Furthermore, the increased spread of the normalized radial distribution function primary peak, plotted from the position of fibrils in cross-section from the anterior stroma, indicated a decrease in fibril order because a smaller proportion of the measured fibrils were at a similar distance to their respective neighbors. However, this increased spread may have also been caused, at least in part, by the radial asymmetry also shown to exist in the PTK-treated corneas. Radial asymmetry of the fibril positions would cause both a widening and lowering of the radial distribution function peaks. Such asymmetry is unlikely to be an artefact caused by processing for TEM since the control stromas appeared radially symmetrical. The measured radial asymmetry was possibly caused by a deformation of the fibril positions during wound healing due to an absence of tightly interwoven anterior lamellae present in normal tissue and thought to play an important role in maintaining a correct corneal shape.36
The low-angle x-ray results also confirmed the presence of a wider average interfibrillar spacing and a less ordered fibril packing in PTK-treated corneas (M8 and M12) since the interference function profiles showed a decrease in the ratio of full height to width at half height compared to the controls - a low value indicating a wider variation of nearest neighbor interfibrillar spacing.37 These data also provide supporting evidence for long-term stromal remodelling previously reported by confocal microscopy10,38 since the variation of interfibrillar spacing lessens over time as wounds heal (Fig 4).
Figure 6. Radial symmetry plots from the anterior stroma of A) control, and B) PTKtreated M8 cornea. The relative positions of anterior fibrils in the PTK-treated cornea show a degree of asymmetry not present in the control, demonstrated by the central ellipse in part B.
Figure 7. The predicted transmission of visible light through the anterior of PTK-treated ?T cornea is less over all visible wavelengths compared to the control, but not to a degree likely to cause visual impairment.
Figure 8. After the depth of the PTK wound was artificially increased to 0.3 mm, the difference in the predicted transmission of visible light between the wounded and the control is considerably larger, indicating that a wound of this depth would probably result in loss of vision.
The refractive indices of the hydrated fibrils were assumed to have remained constant in the PTKtreated and control corneas since no significant change in the intermolecular Bragg spacings was measured. On the assumption that the refractive index of the interfibrillar matrix was also unchanged, we investigated what effect, if any, the altered fibril packing and the presence of varied interfibrillar spaces should have on the transmission of light. A small but insignificant increase in light scattering was predicted by the direct summation of fields method. This small increase in light scatter is unlikely to be responsible for the approximate 50% increase in haze at 8 months, suggesting fibril disorder within the newly formed anterior stroma of PTK-treated corneas is not a major contributor to persistent haze. Interestingly, previous studies using collagenase inhibitors to prevent collagen regrowth in corneas wounded by excimer laser have had a varied and limited effect on the reduction of haze.9·39 Moreover, it seems likely that the small increase in predicted fight scatter was minimal because the layer of disorganized collagen fibrils is not very thick. Thickness of the tissue through which light must pass is an important parameter in calculating light transmission; all other parameters remaining constant, a decrease in thickness will increase light transmission (Equation 1). A thickness of 0.1 mm (maximum wound depth) was used in the transparency calculations but in reality, the disorganized zone is likely to be much thinner at 8 months because of stromal regeneration. To highlight the importance of a thin area of disorganized collagen following PTK, the thickness of the newly formed tissue was artificially raised from 0.1 mm to an extreme value of 0.3 mm, the approximate total thickness of normal rabbit cornea. As expected, this resulted in a decreased transmission of visible light through both the control and PTK-treated cornea (Fig 8). Moreover, the resulting difference in transmission between the control and the PTK-treated cornea was much larger (10% at 500 nm) compared to the difference between the control and the PTKtreated cornea using a 0.1-mm thickness (2% at 500 nm) (Fig 7).
Out results, therefore, imply that collagen deposition is not the cause of persistent haze. This supports recent evidence that stromal rethickening is not associated with haze development over time, suggesting that haze and regression are caused by two independent wound healing mechanisms.40 If fibril disorder within the newly formed anterior stroma of PTK-treated corneas is not a critical component of persistent haze, other causes must be sought. One such cause of haze could be the cut collagen fibers at the PTK wound bed, which persist throughout the healing process as seen by electron microscopy33,41 and whose scattering potential has been evidenced objectively by confocal microscopy.42 Another possible contributor to light scatter and haze might be the wavy basement membrane reported above; the irregularities in this membrane are much larger than half the wavelength of light incident upon them and would in theory cause a diffuse scatter (assuming they are not an artefact of TEM preparation). Yet another potential contributor to corneal haze was proposed recently by Jester and colleagues43, who suggested that keratocytes play a pivotal role in defining the results of refractive surgery. Normal rabbit corneal keratocytes have little reflection from confocal microscopy42 and express high levels of putative enzyme-crystallins, which appear to play a structural function in regulating cellular refractive index and transparency.44 However, there are increased reflections from, and a decreased expression of putative crystalline in, activated and transformed keratocytes in the cornea following PRK. All hazy corneas show increased numbers of anterior stromal wound healing keratocytes with increased reflectivity of both nuclei and cell bodies, suggesting that cellular-based reflections, as opposed to extracellular matrix deposition, are the major origin of increased corneal light scattering after PRK.40,45
These results indicate that following PTK, newly formed anterior corneal collagen fibrils are more widely spaced and misaligned than in the normal cornea (up to 12 months after PTK), particularly those immediately below the epithelium. However, although the causes of persistent haze following PTK have yet to be determined, this study provides strong evidence that the changes in the packing of new collagen fibrils in wounds with a depth of 0.1 mm or less are not the main cause of persistent haze.
1. Krueger RR, Trokel SL, Schubert HD. Interaction of ultraviolet light with, the cornea. Invest Ophthalmol Vis Sci 1985;26:1455-1464.
2. Marshall J, Trokel S, Rothery S, Krueger RR. Photoablative reprofiling of the cornea using eximer laser: Photorefractive keratectomy. Lasers Ophthalmol 1986;1:21-48.
3. Waring GO. Development of a system for excimer laser corneal surgery. Trans Am Ophthalmol Soc 1989;87:854-983.
4. Obata H, Takahashi K, Miyata K, Tsuru T, Ohashi Y, Masuda K. Histological and ultrastructural study of rabbit cornea ablated by scanning excimer laser system. Jpn J Ophthalmol 1994;38:285-291.
5. Stephenson CG, Gartry DS, O'Brart DP, Kerr-Muir MG, Marshall J. Photorefractive keratectomy. A 6-year follow-up study. Ophthalmology 1998;105:273-281.
6. Bower KS, Weichel ED, Kim TJ. Overview of refractive surgery. Am Fam Physician 2001;64:1183-1190.
7. Sugar A, Rapuano CJ, Culbertson WW, Huang D, Varley GA, Agapitos PJ, de Luise VP, Koch DD. Laser in situ keratomileusis for myopia and astigmatism: safety and efficacy. Ophthalmology 2002;109:175-187.
8. Nassaralla BA, Szerenyi K, Wang XW, al Reaves T, McDonnell PJ. Effect of diclofenac on corneal haze after photorefractive keratectomy in rabbits. Ophthalmology 1995;102:469-474.
9. Corbett MC, O'Brart DP, Marshall J. Do topical corticosteroids have a role following excimer laser photorefractive keratectomy? J Refract Surg 1995;11:339-340.
10. Corbett MC, Prydal JI, Verma S, Oliver KM, Pande M, Marshall J. An in vivo investigation of the structures responsible for corneal haze after photorefractive keratectomy and their effect on visual function. Ophthalmology 1996;103:1366-1380.
11. Lipshitz I, Loewenstein A, Varssano D, Lazar M. Late onset corneal haze after photorefractive keratectomy for moderate and high myopia. Ophthalmology 1997;104:369-373.
12. Kim KS, Lee JY, Chung SK, Baek NH. Corneal opacity after repeated photorefractive keratectomy. J Cataract Refract Surg 2001;27:1128-1231.
13. Wilson SE, Mohan RR, Hong JW, Lee JS, Choi R, Mohan RR. The wound healing response after laser in situ keratomileusis and photorefractive keratectomy: elusive control of biological variability and effect on custom laser vision correction. Arch Ophthalmol 2001;119:889-896.
14. Fantes FE, Hanna KD, Waring GO III, PouliquenY, Thompson KP, Saodelli M. Wound healing after excimer laser keratomileusis photorefractive keratectomy. A prospective, randomized, double blind trial. Arch Ophthalmol 1992;110:944-952.
15. Fitzimmons TD, Fagerholm P, Härfstrand A, Schenholm M. Hyaluronic acid in the rabbit cornea after excimer laser superficial keratectomy. Invest Ophthalmol Vis Sci 1992;33:3011-3016.
16. Taylor DM, L'Espérance FA, Del Pedro RA Roberts AD, Gigstad JE, Khntworth G, Martin CA, Warner J. Human excimer laser lamellar keratectomy. A clinical study. Ophthalmology 1989;96:654-663.
17. Maurice DM. The structure and transparency of the corneal stroma. J Physiol (Lond) 1957;136:263-285.
18. Farrell RA. Corneal Transparency. In: Albert DM, Jakobiec FA (eds). Principles and Practice of Ophthalmology. London, UK: WB Saunders; 1994:64-80.
19. Rawe IM, Zabel RW, Tuft SJ, Chen V, Meek KM. A morphological study of rabbit corneas after laser keratectomy. Eye 1992;6:637-642.
20. Malley DS, Steinert RF, Puliafito CA, Dobi ET. Immunofluoresence study of corneal wound healing after excimer laser keratectomy in the monkey eye. Arch Ophthalmol 1990;8:1316-1322.
21. Quantock AJ, Kratz-Owens KL, Leonard DW, Meek KM, Schanzlin DJ. Remodelling of the corneal stroma after lamellar keratoplasty: a synchrotron x-ray study. Cornea 1994;13:20-27.
22. Cintran C, Kublin CL. Regeneration of corneal tissue. Dev Biol 1977;61:346-357.
23. Freund DE, McCaIIy RL, Farrell RA. Direct summation of fields for light scattering by fibrils with applications to normal corneas. Applied Optics 1986;25:2739-2746.
24. Freund DE, McCaIIy RL, Farrell RA, Cristol SM, L'Hernault NL, Edelhauser HF. Ultrastructure in anterior and posterior stroma of perfused human and rabbit corneas: relation to transparency Invest Ophthalmol Visual Sci 1995;36: 1508-1523.
25. Lohmann CP, Fitzke F, O'Brart D, Muir MK, Timberlake G, Marshall J. Corneal light scattering and visual performance in myopic individuals with spectacles, contact lenses, or excimer laser photorefractive keratectomy. Am J Ophthalmol 1993;155:444-453.
26. FuUwood NJ, Meek KM. An ultrastructural time-resolved study of freezing in the corneal stroma. J MoI Biol 1994;236:749-758.
27. Goodfellow JM, Elliot GF, Woolgar AE. X-ray diffraction studies of the corneal stroma. JMoI Biol 1978;119:237-252.
28. Worthington C, Inouye H. X-ray diffraction study of the cornea. Int J Biol Macromol 1985;7:2-8.
29. Meek KM, Fullwood NJ, Cooke PH, Elliott GF, Maurice DM, Quantock AJ, Wall RS, Worthington CR. Synchrotron x-ray diffraction studies of the cornea, with implications for stromal hydration. Biophys J 1991;60:467-74.
30. Farrell RA, Hart RW. On the theory of the spatial organization of macroraolecules in connective tissue, Bull Math Biophys 1969;31:727-760.
31. Leonard DW1 Meek KM. Refractive indices of the collagen fibrils and extrafibrillar material of the corneal stroma. Biophys J 1997;72:1382-1387.
32. Moller-Pederson T, Li H, Petroli W, Cavanagh H, Jester J. Confocal microscopic characterization of wound repair after photorefractive keratectomy. Invest Ophthalmol Vis Sci 1998;39:487-501.
33. Tuft SJ, Zabel RW, Marshall J. Corneal repair following keratectomy: a comparison between conventional surgery and photoablation. Invest Opthalmol Vis Sci 1989;30:1796-1777.
34. Rawe IM, Tuft SJ, Meek KM. Proteoglycan and collagen morphology in superficially scarred rabbit cornea. Histochem J 1992;24:311-318.
35. Corbett MC, Marshall J. Corneal haze after photorefractive keratectomy. Lasers and Light 1996;7:173-196.
36. Muller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol 2001;85:379-381.
37. Fratzl P, Fratzl-Zelman N, Klaushofer K Collagen packing and mineralization. An x-ray scattering investigation of turkey leg tendon. Biophys J 1993;64:260-267.
38. Bohnke M, Schipper I, Thaer, A. Examination of the human cornea with the confocal microscope after excimer-laser photorefractive keratectomy. KHnisceh Monatsblatter fur Augenheilkunde 1997;211:159-167.
39. Corbett MC, O'Brart DP, Patmore AL, Marshall J. Effect of collagenase inhibitors on corneal haze after PRK Exp Eye Res 2001;72:253-259.
40. Moller-Pedersen T, Cavanagh HD, Petroli WM, Jester JV. Stromal wound healing explains refractive instability and haze development after photorefractive keratectomy: a 1year confocal microscopic study. Ophthalmology 2000;107:1235-1245.
41. Cintron C, Hassinger LC, Kublin CL, Cannon DJ. Biochemical and ultrastructural changes in collagen during corneal wound healing. J TJltrastruct Res 1978;65:13-16.
42. M0llerPederson T, Li H, Petroli W, Cavanagh H, Jester J. Confocal microscopic characterization of wound repair after photorefractive keratectomy. Invest Ophthalmol Vis Sci 1998;39:487-501.
43. Jester JV, Moller-Pederson T, Huang J, Sax CM, Kays WT, Cavanagh HD, Petroli WM, Piatigorsky J. The cellular basis of corneal transparency: evidence for 'corneal crystalline'. J Cell Sri 1999;112:613-622.
44. Jester JV, Moller-Pederson T, Huang J, Sax CM, Petroli WM, Cavanagh HD, Piatigorsky J. Rabbit keratocytes express enzyme-crystallins that may modulate cellular corneal transparency following excimer PRK [abstract]. Invest Ophthalmol Vis Sci 1998;39(suppl):S449.
45. Moller-Pederson T. The cellular basis of corneal transparency and haze development. Ophthal Res 2002;34:1-13.
Difference* Between Mean Fibril Diameters of PTK and Control Rabbit Corneas by Quantitative Image Analysis