Biomechanical customization: The next generation of laser refractive surgery

Before the development of wavefront sensors, visual acuity was the standard of care for evaluating vision. However, its limitations became evident in the refractive surgery arena, in which patients with acceptable visual acuity sometimes had debilitating visual complaints. The term quality of vision became more meaningful with the ability to quantify higher-order ocular wavefront errors. In other words, the ability to measure a parameter was the first critical step in making that parameter important in general clinical practice. The ability to measure optical wavefront error also offered the potential to surgically control it, launching a concerted effort toward individually customized wavefront treatments. Yet, wavefront-driven procedures are not optimized for reasons that range from technological to biological. Evidence that the biomechanical properties of the cornea are an important source of individual variation that influences treatment response has existed for some time. However, without the ability to measure biomechanical properties in vivo, their influence cannot be predicted or controlled. As a result, most modern refractive surgical treatments are based in large part on a population-based normative response. As an alternative to knowing individual biomechanical properties, we have attempted to back out the response to treatment from various clinical measures such as corneal topography. The corneal biomechanical landscape is on the verge of great change. This special issue reports 3 new techniques to investigate corneal biomechanical properties (Jaycock et al., pages 175–184; Luce, pages 156–162; Grabner et al., pages 163–174). None is yet commercially available. However, 2 techniques have been used in clinical research studies (Grabner et al. and Luce) and 1 device is approved by the U.S. Food and Drug Administration (Luce), raising hopes that commercial availability will soon follow. All 3 techniques are distinct and give us important biomechanical information not previously available. This represents the critical first step in not only understanding biomechanical response to laser refractive surgery, but also predicting it on an individual basis. With the ability to measure biomechanical properties comes the potential to control biomechanical response. It is likely that the same kind of increased understanding of image quality that accompanied the introduction of wavefront sensors will now develop with respect to corneal biomechanics, accompanying the introduction of devices to measure biomechanical properties. The 3 new devices show a similar trend with respect to biomechanical properties as a function of altered corneal structure from surgical intervention or pathology such as keratoconus. Jaycock et al. demonstrate that the creation of a lamellar flap produces greater corneal distension (expansion) with an increase in intraocular pressure (IOP) than occurs in an intact cornea. Grabner et al. demonstrate greater induced shape change, which implies greater distensibility, in keratoconic corneas than in normal corneas. Luce demonstrates lower hysteresis in keratoconus and after laser in situ keratomileusis (LASIK) than in normal eyes, a manifestation of faster shape recovery, which presumably correlates with a lower modulus of elasticity. Using still another technique, Hjortdal and coauthors (pages 21–29) report a significant long-term decrease in apparent IOP after LASIK and photorefractive keratectomy (PRK) measured by standard pneumotonometry as an indication of corneal “stiffness.” Once again, surgical intervention generated a more easily deformable cornea, which was manifested by an apparent decrease in the measured IOP. Interestingly, LASIK produced a greater change than PRK in corneas of similar overall thickness, highlighting the important concept that biomechanical properties may have a greater impact on IOP measurement artifact than thickness, which is theoretically confirmed by Liu and Roberts (pages 146–155). This highlights the need for a device that is insensitive to biomechanical artifact to measure IOP after LASIK and PRK. The consistency in the results of 4 distinct measurement techniques begins to paint a compelling picture of the biomechanical consequences of laser refractive surgery in lowering the modulus of elasticity of the cornea. Will the apparent increased distensibility or “softening” of the cornea after laser refractive surgery predispose this population to ectasia, as seen in keratoconus with similar biomechanical properties, at least in these preliminary analyses? As a community, we need to be extremely careful not to jump to conclusions based on a single parameter in such a complex structure as the cornea. The work of Vinciguerra et al. (pages 82–87) in ultrathin corneas demonstrates the critically important concept that mechanical stability is not a 1-dimensional function of residual stromal bed thickness. Very large (10.0 mm diameter), ultimately deep phototherapeutic keratectomy ablations were performed with 5 years of stability, providing compelling clinical evidence that a large optical zone may protect against ectasia, even in the presence of a thin cornea. With this shape profile, the stress is distributed more uniformly over the entire cornea. In contrast, a smaller optical zone concentrates the mechanical stress in a smaller region, potentially leading to a progressive process. This may be the rationale for the early effectiveness of newly developing treatments for keratoconus. With the placement of Intacs (Addition Technology) or the use of conductive keratoplasty (Alió and coauthors, pages 190–197), the small diameter, steep zone of the conus is expanded, which redistributes the stress over a larger region and leads to a more regular corneal shape that is ultimately more stable. Is there a relationship between corneal biomechanical properties and the development of ectasia? It can be assumed that the distribution of corneal biomechanical properties in the nonsurgical, nonpathologic population has a normal bell-shaped curve. Based on the new information in this special issue, a hypothesis that corneas with a lower modulus of elasticity preoperatively may be more prone to ectasia postoperatively emerges, given the evidence that the surgical procedure itself further reduces the modulus of elasticity. This might offer a rationale for cases of ectasia that are difficult to explain based on abnormal curvature patterns preoperatively or low residual stromal bed thickness postoperatively. Modulus of elasticity may be an important predictor of ectasia. Patients with a low modulus of elasticity, a small optical zone, and a thin stromal bed are potentially at greatest risk. These are uncharted waters, however, since the ability to measure biomechanical properties in vivo has not previously existed. Much work needs to be done in this area. Current methods of investigating ectasia by examining the posterior corneal surface rely on elevation maps relative to a “best-fit” sphere. This can lead to errors of interpretation due to the relative nature of the measurement. In a large population analysis, Twa and coauthors (pages 61–71) report that not only does the curvature of the posterior cornea increase after a conventional LASIK procedure, but the central elevation above that smaller reference sphere also increases. Does this represent the beginning of an ectatic process? Gryzbowski and coauthors (pages 72–81) present a model for increased posterior corneal curvature after laser refractive surgery resulting from peripheral swelling in the anterior chamber rather than central bulging. This model was validated with a large population analysis, leading to the conclusion that the posterior surface changes that occur after LASIK are the result of stable remodeling. The corneal biomechanical response to laser refractive surgery has 2 main components: structural stability, which has been discussed, and biomechanical remodeling of the anterior surface shape due to the permanent, surgically induced alteration of the corneal structure. A biomechanical model of corneal response, first presented by Dupps et al. in 1995, predicts central flattening accompanied by midperipheral steepening and thickening with any procedure that circumferentially severs tension-bearing lamellae (Dupps WJ, et al. IOVS 1995; 36:ARVO abstract 3257). How is this consistent with the greater distensibility after laser refractive surgery that is reported in this special issue? The answer lies in the complex microstructure of the cornea.1–6 The midperipheral cornea distends and steepens with the sudden reduction of tension in the peripheral lamellar segments that remain after central ablation. The central cornea must flatten to compensate, producing a biomechanical hyperopic shift, all under stress induced by normal IOP and normal hydration. This is true for myopic and hyperopic ablation profiles (Qazi and coauthors, pages 48–60). If the IOP is increased, however, or the hydration state fluctuates significantly, this central flattening response can be modified to produce steepening associated with a myopic shift, as described in 2 case studies (Fam and coauthors, pages 198–201; Toshino et al., pages 202–204). Midperipheral steepening due to the biomechanical response is a likely cause of induced spherical aberration after laser refractive surgery. Another possibility that has been proposed recently is decreased ablation efficiency in the midperiphery due to the nonnormal angle of incidence of the excimer beam with the cornea in this region. The work of Yoon et al. (pages 127–135), however, demonstrates theoretically that even with compensation for this effect in a myopic population, spherical aberration is induced. This has also been demonstrated clinically with an ablation profile that compensates for decreased efficiency in the midperiphery; yet increased steepening is still observed topographically (Figure 1). In addition, the complex corneal microstructure leads to a differential response between the midperipheral nasal and temporal corneal regions, as described by Serrao and coauthors (pages 30–38).Figure 1.: Both eyes of the same patient with the left eye treated using the standard wavefront-optimized profile of a WaveLight system. This profile compensates for the loss of ablation efficiency in the periphery due to the nonnormal angle of incidence of the excimer beam with the cornea. The right eye is treated with the new F-CAT profile, which includes an additional aspheric component that adds more ablation in the midperiphery. Both eyes had similar preoperative refractions with greater than −8.0 diopters of myopia. (Topographies courtesy of Dr. Jerry Tan of Singapore.)Several studies of the response of the anterior corneal surface shape to the creation of a lamellar flap are included in this special issue (Potgeiter et al., pages 106–114; Tran et al., pages 97–105; Güell et al., pages 115–119; Kohnen and Bühren, pages 185–189). The responses from each type of microkeratome are variable. However, a consistent response in most studies is a modest hyperopic shift, measured with multiple techniques. This is consistent with the biomechanical model of central flattening with circumferential severing of tension-bearing lamellae. Tran and coauthors report a hyperopic shift as a reduction in the defocus term of the ocular wavefront, following flaps created with a mechanical microkeratome and a femtosecond laser. Although the femtosecond laser group had significantly fewer higher-order aberrations, the second-order terms were affected despite the very uniform flaps. In addition, Potgeiter and coauthors demonstrate that the magnitude of the change in anterior surface shape due to a flap could be predicted by the residual stromal bed thickness. An important concept highlighted by these 2 articles is that the flap mirrors the response of the bed; thus, identical flaps will not necessarily produce identical responses. How can corneal biomechanics be included in ablation profile design? Multiple approaches are possible and are currently being investigated and implemented. Finite element models are being constructed (Pinsky and coauthors, pages 136–145) and ablation on the back of a flap is being performed (Versace and Watson, pages 88–96) in attempts to understand and optimize biomechanical behavior. In addition, a new aspheric profile, which involves greater ablation in the periphery, is showing great clinical promise. Figure 1 illustrates the difference in corneal shape response between an ablation profile compensated for loss of efficiency in the periphery in the left eye and the same profile with the addition of an aspheric peripheral ablation pattern in the contralateral eye of the same patient. In the right eye, which has greater midperipheral ablation, there is less difference between central and peripheral curvature than in the left eye, providing better vision in the right eye. In other words, the left eye is flatter in the center and steeper in the periphery, with the same central ablation depth as the right eye. This demonstrates that central shape can be modified by changing the ablation pattern predominantly in the midperiphery. The right eye was treated using an ablation profile similar to that shown in Figure 2, which is imposed in addition to the standard profile used in the left eye. Currently, however, all procedures are performed without knowledge of individual corneal biomechanical properties, which would add a new dimension to customization.Figure 2.: Example of a pure aspheric ablation profile with no spherical correction included.We are poised to make important advances in understanding the impact of corneal biomechanics on refractive surgery outcomes. My personal prediction for the future is that the next generation of refractive surgery will include individual biomechanical customization, with full integration of corneal topography, wavefront sensors, and new devices to measure corneal biomechanical properties.