It is not difficult to imagine that differential thinning of the stroma could lead to differential “bowing” of the corneal layer that produces central bulging.1 Flap creation and tissue removal can also alter the lamellar packing; the peripheral corneal lamellae adjacent to the keratectomized layer are no longer held tense and can relax, potentially causing a pull on the central cornea and subsequent central flattening.2 Also, because the epithelial-thickness profile ultimately determines the majority of the final refractive power of the cornea, alterations in the profile of the epithelial layer can also cause changes in refraction.3
It is widely accepted that the epithelium has an inherent ability to “fill in” postoperative surface irregularities of the stromal surface (Figure 1). This article examines the influence of epithelial and biomechanical changes on customized ablation.HOW INACCURATE IS EXCIMER LASER SURGERY?
The general Munnerlyn equation4 holds that approximately 12µm of central corneal flattening is needed to correct -1.00D with a 6-mm treatment zone. An excimer laser removes tissue in increments of 0.25µm per pulse, a relationship that translates to 48 central corneal pulses for each -1.00D correction. Put another way, each pulse is responsible for a dioptric increment of one–forty-eighth. Per observation, postoperative refractions typically range ±1.00D from what is intended. Why does removing tissue in the increments described produce a result that is 48 times less accurate than the laser itself?REGRESSION, PRIMARY
UNDERCORRECTION, AND ECTASIA
A myopic eye undergoing LASIK for -4.00D is plano on postoperative day 1, -0.25 -0.25 X 180 at 1 month, and
-0.75D sphere at 3 months, 6 months, and 1 year. One would say that the initial correction of -4.00D regressed to
-0.75D where it then stabilized. The question is, why did this eye regress? Did the epithelium thicken centrally? Was the flap too thick, causing forward bowing of the central cornea?
The term regression should be defined as a shift in refraction postoperatively that tends to reverse the intended effect but ultimately stabilizes. It differs from primary undercorrection, in which the postoperative refraction is immediately -1.00D, for example, and remains so. An eye that continues to experience an increase in myopia, however, might be evidence of a cornea that is undergoing plastic deformity or ectasia5 (eg, the corneal lamellae are shearing under IOP, and there may be a concomitant thinning and increase in corneal curvature). Making these distinctions helps point to etiology.
The etiologies of primary undercorrection can be divided into corneal and noncorneal causes. The latter include inaccurate preoperative refraction and the delivery of inadequate laser energy. Among the former are cases where corneal biomechanics change due to the keratectomy, and a stable but unpredicted curvature change is obtained. Such cases can occur when the residual stromal thickness is much less than 250µm but not thin enough to cause long-term destabilization (ectasia).STUDYING FACTORS IN LASIK ACCURACY
My colleagues and I prospectively studied 52 eyes that underwent routine LASIK between 1998 and 1999 with both the Moria LASIK One (Moria, Antony, France) and Hansatome microkeratomes (Bausch & Lomb, Rochester, NY) and either the EC-5000 (Nidek, Inc., Fremont, CA) or the 217C excimer laser (Bausch & Lomb).6 The level of myopia ranged from -1.00 to -10.25D. We scanned all eyes with the 3D Artemis VHF digital ultrasound arcscan (Ultralink LLC, St. Petersburg, FL) in order to obtain the thickness profile and optical power of the epithelium and stroma separately before and after LASIK. We also measured the curvature of the front and back surface of the cornea with the Orbscan II (Bausch & Lomb) before and at least 3 months after LASIK. Three-dimensional maps of epithelial thickness and residual stromal bed thickness were produced from the Artemis data, and Orbscan topography determined anterior and posterior corneal best-fit sphere. We calculated the curvature of Bowman's surface from the anterior best-fit sphere and the epithelial thickness profile. Gradient optics and lens formulae were used to calculate total corneal power from anterior, Bowman's, and posterior corneal interfaces. We defined back-surface curvature change as a bowing factor. Curvature change of the anterior corneal surface was divided into epithelial and bowing components. We calculated the change in corneal power, pre- to postoperatively, with permutations removing “bowing,” epithelial changes, or both. Linear regression and paired T-tests helped us to determine the epithelial and/or bowing contributions to the final refraction.Results
For the cohort of eyes, the minimum residual stromal bed thickness was 262µm. We found that, below 290µm, the change in postoperative back-surface curvature correlated strongly with residual stromal bed thickness (R2=0.5). Clinically, attempted versus achieved manifest refractive change was highly correlated (R2=0.95) with a slope of 0.92. Measuring the change in corneal power by the calculation method was validated by a high association between change in clinical refraction and the calculated corneal power change (R2=0.64, slope=0.91). We then correlated the correction achieved (by refraction) with the calculated change in corneal power, subtracting epithelial and/or bowing factors.
Removing bowing significantly increased the change in corneal power by 15% and produced a higher correlation slope of 0.99 (R2=0.44). Removing epithelial change produced a 5% increase in the change in corneal power (5%) and a slope of 0.94 (R2=0.64). Removing both epithelial and bowing factors increased the change in corneal power by 20% and produced a slope of 1.02 (R2=0.45), (all P<.01). We therefore concluded that significant biomechanical and epithelial effects occurred and that elastic bowing and epithelial changes appeared to account fully for the inaccuracy of LASIK.
In other findings, residual stromal bed thickness was correlated to the mechanical power shifts calculated (R2=0.32), and ablation depth was highly correlated to the mechanical shifts observed (R2=0.89). The thickness of the flap's stromal component bore a significant correlation to the spherical equivalent postoperative error (P<.05). Strong, significant, nonlinear associations existed between the level of myopia treated and the epithelial (P<.001) or biomechanical (P=.011) power-shift measured.Epithelial Profile Changes Characterized
Within the study, we also analyzed the epithelial subset of data apart from the study's epithelial dynamics. To model the epithelial changes, we measured the epithelial thickness profiles obtained with the Artemis before and after surgery in annulus fashion and averaged for the 3-, 4-, 5-, 6-, and 7-mm–diameter zones (Figure 1).
In most cases, before surgery, the mean central epithelial thickness was 51µm (range, 47 to 62µm). By contrast, 3 months after LASIK, the central epithelium had thickened to an average of 61µm ( range, 44 to 75µm). Central thickening can produce a relative increase in corneal curvature and result in a regression of the intended myopic refractive effect of LASIK flattening (Figure 2).
In a raw comparison between the error in the spherical equivalent postoperatively (all patients had an intended postoperative refractive error of zero) and the epithelial power shift as measured by Artemis scanning, we found a statistically significant correlation. In other words, shifts in epithelial power at least partly accounted for the postoperative refractive error. This correlation alone attests to the significance of the epithelium in regression, considering the number of other (biomechanical) factors that were also in play. Nonetheless, a simple linear regression demonstrated a correlation in which the epithelium could account for approximately 25% of every 1.00D of postoperative spherical equivalent error.
Plotting the level of myopia treated against the amount of central epithelial thickening at the 3-, 4-, 5-, 6-, and 7-mm zones (Figure 3) showed a strong correlation between the amount of postoperative epithelial thickening and the level of intended myopia to be treated. Also of interest, there was more thickening in the central cornea when compared to the periphery. In addition, the amount of postoperative epithelial thickening increased linearly as the level of preoperative myopia increased. As the amount of preoperative myopia approached 5.00D, however, this relationship began to stabilize, suggesting that there is a limit to the ability of the epithelium to reverse central corneal flattening.
After dividing the study eyes into three groups (low, moderate, and high myopia), we determined the thickening profile for each. The shift in power due to the epithelium was found to be related to the difference between central and peripheral thickening. If the epithelium were to thicken evenly, there would be no power shift (no change in curvature). In the low myopes, there was considerably more thickening in the center than in the periphery: 8 versus 4µm. Interestingly, a higher level of myopia correlated with a smaller difference between the central and peripheral thickening. The epithelium also appeared to cause more regression in low versus high myopia, despite the fact that the low-myopia group showed less absolute thickening in the postoperative period.
The shift in power due to epithelial profile changes was more significant for lower than higher myopia. Perhaps central epithelial thickening reaches a maximum level beyond which increasing the depth of the myopic ablation will not result in further central epithelial thickening (while the peripheral epithelium can still thicken for higher myopic ablations as the peripheral ablation depth increases).CONCLUSION
This study demonstrates that biomechanical changes appear to be closely correlated to residual stromal thickness, which is known to be a function of preoperative corneal pachymetry and the total depth of intraoperative keratectomy. An accurate knowledge of the intended residual stromal thickness is important when planning retreatments. Specifically, differentiating between an undercorrection mainly due to an elastic yet stable bowing of the cornea and a change in epithelial profile is essential. If the residual stromal thickness is low, it may be responsible for induced mechanical changes, and removing more tissue may produce an inaccurate result and convert a stable, elastic cornea into a potentially unstable, plastic, and perhaps ectatic cornea. This concern is particularly pertinent to wavefront-guided retreatments commonly used to reduce spherical aberration, because they typically consume large amounts of tissue. In the final analysis, true customized ablation may require models to predict both epithelial and biomechanical changes in an effort to achieve best low-aberration vision.
Dan Z. Reinstein, MD, MA(Cantab), FRCSC, DABO, is Medical Director of the London Vision Clinic and Consultant Ophthalmologist at St. Thomas' Hospital–Kings College, London. He states that he holds a financial interest in the Artemis technology but not in the other products and companies mentioned herein. Dr. Reinstein may be reached at +44 20 7224 1005; email@example.com. Seitz B, Torres F, Langenbucher A, et al. Posterior corneal curvature changes after myopic laser in situ keratomileusis. Ophthalmology. 2001;108:666-672; discussion 73.
2. Roberts C. The cornea is not a piece of plastic. J Refract Surg. 2000;16:407-413.
3. Srivannaboon S, Reinstein DZ, Sutton HFS, Silverman RH, D.J. C. [Au: please clarify] Effect of epithelial changes on refractive outcome in LASIK. Invest Ophthalmol Vis Sci. 1999;40(suppl):896.
4. Munnerlyn CR, Koons SJ, Marshall J. Photorefractive keratectomy: a technique for laser refractive surgery. J Cataract Refract Surg. 1988;14:46-52.
5. Barraquer JI. Queratomileusis y queratofakia. Bogota: Instituto Barraquer de America; 1980.
6. Reinstein DZ, Srivannaboon S, Silverman RH, Coleman DJ. The accuracy of routine LASIK; isolation of biomechanical and epithelial factors. Invest Ophthalmol Vis Sci. 2000;41(suppl):318.