From the Department of Ophthalmology, King’s College London (Knox Cartwright, Jaycock, Marshall), London; and Wolfson School of Mechanical and Manufacturing Engineering, Loughborough University (Tyrer), Loughborough, United Kingdom.
This research was supported by grants from the Royal College of Surgeons and TFC Frost Charitable Trust.
Dr Marshall was formerly a consultant for IntraLase. The remaining authors have no proprietary interest in the materials presented herein.
Study concept and design (N.E.K.C., J.R.T., P.D.J., J.M.); data collection (N.E.K.C.); analysis and interpretation of data (N.E.K.C., P.D.J., J.M.); drafting of the manuscript (N.E.K.C.); critical revision of the manuscript (J.R.T., P.D.J., J.M.); administrative, technical, or material support (J.M.); supervision (J.R.T., J.M.)
Correspondence: Nathaniel E. Knox Cartwright, MA, MRCOphth, Bristol Eye Hospital, Lower Maudlin St, Bristol BS1 2LX, United Kingdom. E-mail: firstname.lastname@example.org
Studies with up to 14 years of follow-up have shown that refractive outcomes of surface ablation are maintained and imply long-term postoperative biomechanical stability of the cornea.1,2 However, photorefractive keratectomy (PRK) is associated with pain and delayed visual recovery due to the resultant epithelial defect3 and risk of postoperative haze due to wound healing interactions between the epithelial cells and keratocytes.4
By contrast, although visual recovery is rapid and pain and haze negligible following LASIK, concerns exist regarding possible corneal biomechanical instability,5,6 as creation of the microkeratome incision results in a significantly greater number of collagen lamellae and fibrils being severed, compounded by more fibers being injured during ablation. It has been implied that the limited number of reports of postoperative ectasia indicate that the risk of this complication is low, especially if patients with recognized risk factors are excluded.7 This implication has been used to argue that LASIK itself is not a cause of biomechanical instability. The counterargument is that virtually all long-term studies of LASIK have been complicated by high retreatment rates8–12 or a trend towards refractive regression,13–16 implying that progressive postoperative change in corneal shape occurs.
The evolution of mechanical microkeratomes17 and the subsequent advent of femtosecond laser technology has resulted in the potential to create reliably thinner intrastromal flaps with low risk of complication.18 Laboratory studies have provided evidence suggesting that the major structural components of the cornea are situated in the anterior third of the stroma adjacent to the epithelium.19,20 Given certain technological innovations, it has been suggested that making thinner flaps adjacent to Bowman layer will result in a procedure that simultaneously approximates the biomechanical advantages of PRK while preserving the superior wound-healing response of LASIK.
The purpose of this study was to compare the change in biomechanical properties elicited by femtosecond laser incisions at depths comparable to those used in thick- and thin-flap LASIK. The secondary aim was to evaluate separately the biomechanical effects of the delamination incision that creates the bed of a flap by separating lamellae in a defined plane and the side cut that divides multiple lamellae. An investigation was undertaken to determine whether varying side cut angulation could improve postoperative biomechanical performance. All biomechanical measurements were performed using radial shearing speckle pattern interferometry (RSSPI), a non-contact optical interferometric measurement technique well-established in mechanical engineering and sensitive to changes in radial strain caused by submicron variation in displacement.21,22
Materials and Methods
The tenets of the Declaration of Helsinki were followed and ethical approval was obtained from St Thomas’ Hospital Research Ethics Committee. Forty-two experimental specimens (comprising 24 paired and 18 unpaired human corneoscleral buttons), with measurements that had previously been included in a published analysis of age-related variation in corneal elasticity,21 were obtained from the United Kingdom Corneal Transplant Service and immersed in Eagle minimal essential medium containing 5% dextran (molecular weight 500 000), 2% fetal bovine serum, penicillin 100 units/mL, streptomycin 0.1 mg/mL, and Amphotericin B 0.25 μg/mL. On arrival in the laboratory, the specimens were placed in fresh medium without fetal bovine serum and placed in a humidified incubator with 5% carbon dioxide at 37°C.
After 24 hours, the corneoscleral buttons were rinsed in culture medium without dextran and central corneal thickness was determined ultrasonically (DGH-550 Pachette 2; DGH Technology, Exton, Penn-sylvania). Specimens were mounted in a modified Barron artificial anterior chamber (Katena Products, Denville, New Jersey). Central corneal thickness measurements were repeated at 1 week.
To provide the corneal endothelium with its physiological environment, the internal reservoir of the chamber was filled with culture medium and maintained at a hydrostatic pressure of 15.0 mmHg (2000 Pa) using a digital manometer with a resolution better than 1 Pa.23 A Petri dish with a hole in its base was glued to the locking ring of each artificial anterior chamber and filled with culture medium to the level of the limbus, then covered with a lid ensuring a moist air environment for the epithelium (Fig 1). Additional drops of medium were applied to the epithelial surface every 2 hours and the culture media in both the Petri dishes and internal reservoirs were replaced every 48 hours. Control and experimental specimens were retained in their respective chambers throughout the duration of the experimental period.
Figure 1. Photograph of the modified Barron artificial anterior chambers used to hold the specimens throughout the experimental period and enable the external corneal surface to be bathed in culture medium to the level of the limbus (left – top view; right – side view). Between measurements, and while in organ culture, each Petri dish was covered by its lid to maintain surface humidity.
Twelve of the 42 corneoscleral buttons studied were used as controls and did not undergo surgery. Eighteen of the remaining specimens were divided into 6 equal subgroups in which 8.0-mm diameter lamellar flap, side cut, and subsurface delamination incisions were created at 90- and 160-μm depths. In the remaining 12 specimens, 8.0-mm diameter 90-μm depth side cuts angled between 30° and 150° were created. All incisions were made using a 60-kHz femtosecond laser (FS60; IntraLase Corp, Irvine, California) with the settings listed in Table 1. The IntraLase software available at the time only permitted flap side cut angles <90° so the keratoplasty software was used to create side cuts with greater angles.
Table: Femtosecond Laser Parameters Used
Measurements of corneal biomechanical integrity were obtained at three time points: preoperatively, immediately postoperatively, and 1 week later following a period of wound healing. All measurements were recorded using a RSSPI device following a transient artificial anterior chamber hydrostatic pressure increase from 15.0 to 15.5 mmHg. The pressure transient was applied for approximately 2 seconds and its magnitude was an attempt to emulate the cardiac cycle.
Details of the experimental apparatus used and the method of data analysis have been described in detail elsewhere.21 In brief, the corneoscleral buttons were covered with a stretched layer of polytetrafluoroethylene tape and illuminated normal to their apex by the expanded beam of a 500-mW, 532-μm laser (GLC-050-S; CrystaLaser LC, Reno, Nevada). The resultant granular speckle pattern generated was captured as a reference image on a digital camera with 12-bit resolution (MDC 1004; Imperx, Boca Raton, Florida) and linked to a PXI0-1002 computer (National Instruments, Newbury, Berkshire, United Kingdom). A second image was captured following the pressure increase and subtracted from the reference image to produce the compound interferogram used for data analysis.
The strain maps obtained were integrated to determine change in central displacement and thin shell theory was used to calculate Young’s modulus (E).24 Specifically,
where the assumed constants ν
(Poisson’s ratio) = 0.49,25R
(anterior corneal radius of curvature) = 7.5 mm, Ri
(transverse radius of the artificial anterior chamber) = 6 mm, and η = sin−1
/R); and the measured variables d
= change in central displacement, p
= pressure change, t
= corneal thickness and .
Following the final strain measurements, the corneas were fixed in glutaraldehyde 2.5% buffered in 0.1 M sodium cacodylate containing 10 mg/mL calcium chloride at a final pH of 7.4. The corneas were washed in sucrose buffer and post-fixed for 1 hour in osmium tetroxide 2% before being dehydrated in alcohol and embedded in araldite (CY212) via epoxypropane. Glass knives were used to cut semi-thin sections on a Cambridge-Huxley ultramicrotome (Cambridge Instrument Company Ltd, London, United Kingdom) and stained with toluidine blue 1% for light microscopy.
R-2.8.1 statistical software (R Foundation for Statistical Computing, Vienna, Austria) was used. The changes in apical displacement caused by the differing surgical incisions were compared using the Kruskal–Wallis one-way analysis of variance test. Linear regression analysis was used to characterize the effect of variation in side cut angle. P<.05 was considered statistically significant.
Age of the corneal donors ranged from 24 to 102 years (mean: 77.8 years) and all were free from known ophthalmic disease. Postmortem times ranged between 5 and 24 days, but did not significantly affect the results of subsequent experiments. Biomechanical measurements were obtained from all corneas and typical measurement time was approximately 3 minutes.
Mean initial central corneal thickness was 532±17 μm, which increased to 559±22 μm after 1 week in organ culture. Lamellar flap depths with intended depths of 90 and 160 μm were measured to be 93±8 and 154±7 μm deep, respectively. Delamination incisions with intended depths of 90 and 160 μm were measured to be 82±4 and 167±6 μm deep, respectively. It was not possible to measure side cut incision depth in the same manner, but depths of side cuts were comparable to those of flaps and delaminations in histological specimens.
In control corneas, no significant change in apical displacement occurred throughout the study period. Following incisions at 90-μm depth, mean apical displacement increased by 9% from 4.3±0.2 to 4.7±0.2 μm after flap formation, by 9% from 4.3±0.4 to 4.7±0.2 μm after 90° side cut creation, and by 5% from 4.1±0.1 to 4.3±0.2 μm after delamination. No significant change occurred during wound healing. The equivalent responses to incisions at 160 μm were increases in apical displacement of 32% from 4.4±0.2 to 5.8±0.0 μm after flap formation, 33% from 4.3±0.4 to 5.7±0.1 μm after 90° side cut creation, and 5% from 4.1±0.1 to 4.3±0.1 μm after delamination. No significant changes occurred during wound healing. Only the displacement changes following creation of 160-μm depth flaps and 160-μm depth side cuts were statistically significant. These results are summarized in Figures 2 and 3.
Figure 2. Bar plots showing the changes in mean corneal apical displacement following the increase in apical displacement for the control specimens and each major experimental group following the increase in artificial anterior chamber pressure from 15.0 to 15.5 mmHg. Mean values that are statistically significantly greater than before surgery (paired t test, P<.05) are indicated with an asterisk. No changes that occurred during the first postoperative week were statistically significant.
Figure 3. Plots showing mean radial change in displacement following the increase in artificial anterior chamber hydrostatic pressure from 15.0 to 15.5 mmHg.
In the 15 corneas in which 90-μm flaps with varying side cut angle were created, a negative correlation was noted between side cut angle and strain increase. Compared to the 5% displacement increase already described for 90° side cuts at 1 week, the displacement increase was 12% when the side cut angle was 30° and the flap epithelial diameter was greater than the stromal diameter but was only 2% when the side cut angle was 150° and the flap geometry reversed (Fig 4). This trend did not reach statistical significance.
Figure 4. Effect on increase in corneal apical displacement of variation of 90-μm depth side cut angulation following the increase in artificial anterior chamber hydrostatic pressure from 15.0 to 15.5 mmHg.
Light microscopy of the fixed corneas confirmed the expected precise nature of femtosecond laser incisions with side cuts being linear, delamination incisions running parallel to the corneal surface, and the angle between the two incision types being distinct. No evidence of collagenous healing across wound interfaces was noted, and epithelial ingrowth was not observed in any cornea. As side cut angle increased from 30° to 150° and flap configuration became increasingly undercut, wound gape decreased, and fewer activated keratocytes were present (Fig 5).
Figure 5. Light micrographs of the anterior stroma obtained 1 week following creation of a 90-μm flap with a conventional 90° side cut (top) and a 90-μm flap with an undercut 135° side cut (bottom) obtained following fixation and staining with toluidine blue. The insets illustrate these flap configurations graphically. Wound gape was less and fewer keratocytes (the darker intrastromal opacities) were visible in the specimen with the undercut (135°) side cut.
The biomechanical properties of the cornea are determined by the arrangement of fibrillar collagen in Bowman layer and the stroma.26 Stromal groups of parallel collagen fibrils, each comprised of many collagen molecules, are grouped into lamellae that run from limbus to limbus parallel to the corneal surface. In the anterior third of the stroma, fiber arrangement was less ordered than in its posterior two thirds with lamellae being both smaller and more interwoven.25,27 This difference results in the anterior third of the stroma having approximately twice the cohesive tensile strength of the posterior two thirds.20 Although a rudimentary scar containing fibrin and tenascin forms after injury, the ends of severed corneal collagen fibers do not reconnect and corneas remain significantly and permanently weaker following surgery.28
The highly ordered and layered structure of the cornea indicates a direct relationship between the depth of incisions and the number of load-bearing collagen fibers severed. Consequently, the finding that equivalent depth flap and side cut creation resulted in similar displacement increases is to be expected, because both incision types sever the same number of collagen fibers. The finding that delamination incisions did not cause any significant change in displacement is consistent with the concept that these cuts divide few collagen fibers, as they run in the same plane as the collagen fibers.
The 32% increase in displacement observed after 160-μm depth flap creation was more than 3.5 times greater than the 9% increase measured when 90-μm flaps were created. This difference is greater than would be predicted by the less than two-fold difference in flap thickness. This can be attributed to the fact that the epithelium does not contribute significantly to the tensile strength of the cornea, ie, the weakening effect of surgery depends on how far cuts extend into the stroma rather than absolute incision depth. On this basis, if it assumed that the corneal epithelium is 50-μm thick and the stroma 500-μm deep, then 160-μm incisions extend almost three times further (110 μm/22%) into the stroma than 90-μm cuts (40 μm/8%).
The effect of this difference, together with the stromal mechanical anisotropy described earlier, is modeled theoretically in Figure 6 where change in apical displacement is plotted against incision depth. This assumes that 1) the epithelium has insignificant mechanical function, 2) the anterior third of the stroma has twice the mechanical strength of its posterior two thirds, and 3) the stiffness of the stroma relates directly to minimum stromal residual bed thickness and is contrasted with a plot where it is assumed that displacement is determined solely by total residual corneal bed thickness. Although an oversimplification of the actual situation, this illustrates how factors such as these cause incisions extending into the mid-stroma to have a far greater biomechanical impact than is predicted by depth alone.
Figure 6. Theoretical plots of change in corneal apical displacement against maximum incision depth where it is assumed that displacement is determined solely by residual overall corneal bed thickness (dashed line) and where it is assumed that 1) the epithelium is 50-μm thick and has insignificant mechanical function, 2) the stroma is 500-μm thick and its anterior third is twice as stiff as its posterior two thirds, and 3) the stiffness of the stroma relates directly to its minimum residual bed thickness (solid line).
The undercut flap configuration found to be mechanically optimal was equivalent to the scarf joints used by mechanical engineers to repair composite materials on account of this connection’s tensile high strength.29 This strength results from the fact that scarf joints not only incorporate large areas of contact between opposing surfaces, but also do not contain angulations where strain can concentrate and crack propagation initiate. An additional advantage of undercut corneal flaps is that wound apposition will increase as intraocular pressure rises, reflected in this study as decreased wound gape and presence of fewer activated keratocytes, factors that may reduce the possibility of flap slippage and epithelial ingrowth.
Comparison of the results of the current study with those of other studies is not straightforward because this is the first investigation to use RSSPI to study the effect of surgical intervention. However, a laboratory study that used electronic speckle pattern interferometry to study sheep eyes found a 20.7% displacement increase following 130-μm depth microkeratome flap creation.23 The majority of clinical studies of the biomechanical changes caused by refractive surgery have depended on dynamic pneumotonometry for their measurements and interpretation of such results is complicated by the use of non-standard metrics. Although no studies have directly compared thin-flap LASIK and thick-flap LASIK, thin-flap LASIK causes similar changes to PRK, whereas the response to LASIK differs.3,30
The fact that these experiments were performed in organ culture rather than in vivo is both a strength and weakness of the study. It has been demonstrated that the model used permits physiological wound healing, and equivalent experiments are presently impossible to perform in vivo.31 Caution must be applied when extrapolating these results to the clinical situation. It is possible that the use of an artificial anterior chamber may have altered the boundary conditions for sustaining intraocular pressure and affected the results obtained. Only a limited period of wound healing was permitted in this study, but it is unlikely that the results would have been different if the specimens had been maintained for longer in organ culture. It has been shown that the cohesive tensile strength of healed LASIK wounds averaged just 2.4% of virgin corneas centrally and paracentrally and 28.1% at the peripheral flap–wound interface.28
In the future, it may be possible to prevent haze pharmacologically thereby eliminating the most significant complication of surface ablation (the keratorefractive procedure with the least biomechanical impact). Until then, with this finding that thin-flap LASIK causes only insignificant mechanical changes and with other investigations demonstrating that patient recovery is equivalent after thin-flap LASIK to that following conventional LASIK and much more rapid than after PRK, these results support the clinical trend towards creation of thinner flaps.3
- O’Brart DP, Patsoura E, Jaycock P, Rajan M, Marshall J. Excimer laser photorefractive keratectomy for hyperopia: 7.5-year follow-up. J Cataract Refract Surg. 2005;31(6):1104–1113. doi:10.1016/j.jcrs.2004.10.051 [CrossRef]
- Rajan MS, Jaycock P, O’Brart D, Nystrom HH, Marshall J. A long-term study of photorefractive keratectomy; 12-year follow-up. Ophthalmology. 2004;111(10):1813–1824.
- Slade SG, Durrie DS, Binder PS. A prospective, contralateral eye study comparing thin-flap LASIK (sub-Bowman keratomileusis) with photorefractive keratectomy. Ophthalmology. 2009;116(6):1075–1082. doi:10.1016/j.ophtha.2009.01.001 [CrossRef]
- Baldwin HC, Marshall J. Growth factors in corneal wound healing following refractive surgery: a review. Acta Ophthalmol Scand. 2002;80(3):238–247. doi:10.1034/j.1600-0420.2002.800303.x [CrossRef]
- Seiler T, Quurke AW. Iatrogenic keratectasia after LASIK in a case of forme fruste keratoconus. J Cataract Refract Surg. 1998;24(7):1007–1009.
- Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia after laser in situ keratomileusis in patients without apparent preoperative risk factors. Cornea. 2006;25(4):388–403. doi:10.1097/01.ico.0000222479.68242.77 [CrossRef]
- Randleman JB, Russell B, Ward MA, Thompson KP, Stulting RD. Risk factors and prognosis for corneal ectasia after LASIK. Ophthalmology. 2003;110(2):267–275. doi:10.1016/S0161-6420(02)01727-X [CrossRef]
- Randleman JB, White AJ Jr, Lynn MJ, Hu MH, Stulting RD. Incidence, outcomes, and risk factors for retreatment after wavefront-optimized ablations with PRK and LASIK. J Refract Surg. 2009;25(3):273–276.
- Alio JL, Muftuoglu O, Ortiz D, et al. Ten-year follow-up of photorefractive keratectomy for myopia of more than −6 diopters. Am J Ophthalmol. 2008;145(1):37–45. doi:10.1016/j.ajo.2007.09.009 [CrossRef]
- Alio JL, Muftuoglu O, Ortiz D, et al. Ten-year follow-up of photorefractive keratectomy for myopia of less than −6 diopters. Am J Ophthalmol. 2008;145(1):29–36. doi:10.1016/j.ajo.2007.09.007 [CrossRef]
- Alio JL, Muftuoglu O, Ortiz D, et al. Ten-year follow-up of laser in situ keratomileusis for myopia of up to −10 diopters. Am J Ophthalmol. 2008;145(1):46–54. doi:10.1016/j.ajo.2007.09.010 [CrossRef]
- Alio JL, Muftuoglu O, Ortiz D, et al. Ten-year follow-up of laser in situ keratomileusis for high myopia. Am J Ophthalmol. 2008;145(1):55–64. doi:10.1016/j.ajo.2007.08.035 [CrossRef]
- Jaycock PD, O’Brart DP, Rajan MS, Marshall J. Five year follow-up of LASIK for hyperopia. Ophthalmology. 2005;112(2):191–199. doi:10.1016/j.ophtha.2004.09.017 [CrossRef]
- Kymionis GD, Tsiklis NS, Astyrakakis N, Pallikaris AI, Panagopoulou SI, Pallikaris IG. Eleven-year follow-up of laser in situ keratomileusis. J Cataract Refract Surg. 2007;33(2):191–196. doi:10.1016/j.jcrs.2006.11.002 [CrossRef]
- Esquenazi S. Five-year follow-up of laser in situ keratomileusis for hyperopia using the Technolas Keracor 117C excimer laser. J Refract Surg. 2004;20(4):356–363.
- O’Doherty M, O’Keeffe M, Kelleher C. Five year follow up of laser in situ keratomileusis for all levels of myopia. Br J Ophthalmol. 2006;90(1):20–23. doi:10.1136/bjo.2005.075127 [CrossRef]
- de Ortueta D. Planar flaps with the Carriazo-Pendular microkeratome. J Refract Surg. 2008;24(4):322.
- Chang JS. Complications of sub-Bowman’s keratomileusis with a femtosecond laser in 3009 eyes. J Refract Surg. 2008;24(1):S97–S101.
- Muller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol. 2001;85(4):437–443. doi:10.1136/bjo.85.4.437 [CrossRef]
- Randleman JB, Dawson DG, Grossniklaus HE, McCarey BE, Edelhauser HF. Depth-dependent cohesive tensile strength in human donor corneas: implications for refractive surgery. J Refract Surg. 2008;24(1):S85–S89.
- Knox Cartwright NE, Tyrer JR, Marshall J. Age-related differences in the elasticity of the human cornea. Invest Ophthalmol Vis Sci. 2011;52(7):4324–4329. doi:10.1167/iovs.09-4798 [CrossRef]
- Hung YY. A speckle-shearing interferometer: a tool for measuring derivatives of surface displacements. Optics Communications. 1974;11:132–135. doi:10.1016/0030-4018(74)90200-4 [CrossRef]
- Jaycock PD, Lobo L, Ibrahim J, Tyrer JR, Marshall J. Interferometric technique to measure biomechanical changes in the cornea induced by refractive surgery. J Cataract Refract Surg. 2005;31(1):175–184. doi:10.1016/j.jcrs.2004.10.038 [CrossRef]
- Young W, Budynas R, Roark R. Roark’s Formulas for Stress and Strain. Beijing, China: Tsinghua University Press; 2003.
- McPhee TJ, Bourne WM, Brubaker RF. Location of the stress-bearing layers of the cornea. Invest Ophthalmol Vis Sci. 1985;26(6):869–872.
- Komai Y, Ushiki T. The three-dimensional organization of collagen fibrils in the human cornea and sclera. Invest Ophthalmol Vis Sci. 1991;32(8):2244–2258.
- Abahussin M, Hayes S, Knox Cartwright NE, et al. 3D collagen orientation study of the human cornea using X-ray diffraction and femtosecond laser technology. Invest Ophthalmol Vis Sci. 2009;50(11):5159–5164. doi:10.1167/iovs.09-3669 [CrossRef]
- Schmack I, Dawson DG, McCarey BE, Waring GO III, Grossniklaus HE, Edelhauser HF. Cohesive tensile strength of human LASIK wounds with histologic, ultrastructural, and clinical correlations. J Refract Surg. 2005;21(5):433–445.
- Harman AB, Wang CH. Improved design methods for scarf repairs to highly strained composite aircraft structure. Composite Structures. 2006;75:132–144. doi:10.1016/j.compstruct.2006.04.091 [CrossRef]
- Wallau AD, Campos M. One-year outcomes of a bilateral randomised prospective clinical trial comparing PRK with mitomycin C and LASIK. Br J Ophthalmol. 2009;93(12):1634–1638. doi:10.1136/bjo.2008.152579 [CrossRef]
- Rajan MS, Watters W, Patmore A, Marshall J. In vitro human corneal model to investigate stromal epithelial interactions following refractive surgery. J Cataract Refract Surg. 2005;31(9):1789–1801. doi:10.1016/j.jcrs.2005.02.047 [CrossRef]
Femtosecond Laser Parameters Used
|Depth||90 or 160 μm|
|Spot & line separation||7 μm|
|Bed energy||0.8 μJ|
|Depth||90 or 160 μm|
|Cut angle||30° to 150°|