As ophthalmologists enter the wavefront age, one area of refractive surgery that has received less attention is the clinical and surgical management of pre- and postoperative astigmatism. Despite the commercial development of modern wavefront treatments, astigmatic lower-order aberrations will continue to play a significant part in patients' and physicians' definitions of successful surgical outcomes. In many current FDA wavefront trials, as well as previous FDA excimer trials, there is a paucity of data for treatment claims regarding astigmatism of greater than 3.0 or 4.0 D. A quick review of the submitted FDA data1 indicates that the sample size for subjects treated in the higher astigmatic range may simply be inadequate for some excimer laser manufacturers' claims of efficacy, safety, and stability.

Proficiency in astigmatic treatments is not only a clinical goal, but a strong financial one, because many astigmatic patients are either unaware of their refractive surgical candidacy or are misinformed that they are not candidates for refractive surgery. The direct word-of-mouth stimulus in building a refractive practice is also closely correlated to patients' perceptions of failure in the form of a decreased quality of mesopic/scotopic vision and increased enhancement rates in the astigmatic patient population. The latter exacerbates the problem by increasing the variable cost per patient and the potential for postoperative complications associated with reoperations.

The advent of modern toric IOLs, phakic IOLs, and excimer keratorefractive techniques has enhanced ophthalmologists' surgical options for astigmatic treatments. Deciding which techniques to use is the true marriage of science and art, with each surgeon selecting a treatment plan based on personal experience and understanding the magnitude, quality, and safety of possible astigmatic correction.

This article provides an overview and also details specific considerations of keratorefractive excimer astigmatic techniques. In order to maximize the final refractive astigmatic outcome, surgeons must address synergistic points at every step of the refractive treatment plan.

THE SOURCE OF UNDESIRABLE OUTCOMES
Undesirable results (including surgically induced astigmatism, astigmatic regression, and undercorrection) can result from pre-, intra-, and postoperative elements of the refractive procedure. The single most common and least glamorous source is a suboptimal refraction during the preoperative evaluation. In terms of identifying the axis and magnitude of cylinder, the most reliable method is the cycloplegic autorefraction, with the noncyloplegic autorefraction a close second.2 Other factors affecting the preoperative assessment of astigmatic axis and magnitude include the simple precaution of discontinuing contact lens wear long enough to allow the manifest refraction to stabilize. Corneal warpage from lens wear can result in soft refractive endpoints and instability as the corneal geometry stabilizes after the patient discontinues lens wear.

There are no set rules to quantify the effect that long-term contact lens wear will have on the fidelity of the manifest refraction, but a few points are worth emphasizing: (1) wearers of soft extended-wear contact lenses may require discontinuation periods of greater than 1 week if the surgeon notes soft refractive endpoint and topographic abnormalities initially; (2) patients wearing soft toric contact lenses may need to discontinue their use for extended periods (2 to 8 weeks) before their eyes reach corneal topographic and refractive stability; and (3) long-term RGP lens wearers may need to discontinue their use for 1 to 6 months or longer before their eyes achieve refractive/topographic stability.

Other preoperative sources of unintended surgically induced astigmatism include transposition errors and the incorrect programming of the excimer laser. A surgeon's reliance on his technical staff to perform refractions and program the desired correction increases the probability of these errors. A surgical system in which there is a double redundancy in checking/confirming the patient's identity, manifest refraction, and laser-programmed data can be highly effective at minimizing this source of error. Finally, if surgeons standardize preoperative refractions to the minus cylinder format transposition, errors will occur less frequently.

MAXIMIZING OUTCOMES
Intraoperatively, the surgical plan based on the microkeratome and excimer laser used may significantly affect the percentage reduction of absolute cylinder, postoperative treatment stability, and subjective optical quality. The following major intraoperative factors maximize astigmatic results.

High-Magnitude Unimeridional Ablations
Depending on the excimer laser, the most common technique for treating myopic astigmatism involves a unimeridional ablation. Unfortunately, the unimeridional ablation provides a short minor ablation axis (5.0 to 6.0 mm) with a relatively abrupt contour change between the treated and untreated portions of the cornea. For lower-magnitude with-the-rule cylinder, this ablation profile is relatively effective. The weakness of the unimeridional profile becomes evident with against-the-rule, myopic astigmatic treatments, which are exquisitely unforgiving of treatment decentration. As the absolute cylinder magnitude increases, even the smallest amount of decentration can create a highly multifocal central cornea with a large dioptric gradient within the entrance pupil. Another limitation of unimeridional, myopic astigmatic treatments is hyperopic coupling, which induces hyperopic sphere for every diopter of astigmatism that is corrected. The coupling is proportional to the magnitude of intended cylinder correction and can result in consecutive hyperopia in myopic astigmatic ablations.

The Bitoric and Cross-Cylinder Alternative
Arturo Chayet, MD, of Chula Vista, California, and Paolo Vinciguerra, MD, of Milan, Italy, pioneered the techniques of bitoric and cross-cylinder ablations.3-5 Both techniques divide the pre-existing astigmatism into hyperopic and myopic components. The key difference is that the cross-cylinder technique divides the preoperative cylinder into equal myopic and hyperopic components and treats the spherical equivalent of the preoperative refraction. Dr. Chayet and his colleagues introduced the bitoric LASIK ablation profile in order to correct mixed and simple myopic astigmatism by dividing the preoperative astigmatism into unequal ratios of myopic and hyperopic cylinder. Theoretically, this technique avoids causing a hyperopic shift in the spherical component of the refraction.

It is not surprising that both techniques were originally developed and patented6,7 by Nidek Co., Ltd., (Gamagori, Japan) for its EC-5000 excimer laser platform, because the circular ablation profile of the standard Nidek myopic cylinder ablation demonstrates greater hyperopic spherical coupling than other modern excimer laser platforms performing unimeridional (oval) ablation profiles. With the possible adoption of both of these ablation algorithms to the EC-5000, Nidek will be able to provide refractive surgeons with a detailed, cognitive, preoperative approach and one of the broadest astigmatic treatment ranges of any excimer laser platform currently available.

The primary goal of the bitoric and cross-cylinder techniques is to create a more stable, predictable optical result with fewer of the side effects commonly associated with unimeridional myopic cylinder treatments (Figure 1). One of the best times to use these alternate approaches to astigmatism treatment is when correcting mixed astigmatism induced by prior refractive procedures. During the preoperative surgical planning, it is important to note that these techniques differ in regard to the location and quantity of corneal tissue ablated. My personal clinical experience is that a cross-cylinder ablation provides the highest quality mesopic/scotopic vision in high astigmatic corrections. The cross-cylinder (50/50 astigmatic split) ablation treats the spherical equivalent of the manifest refraction with the cylinder of the same sign. I generally treat the myopic component first. For example:

Preoperative manifest refraction: -1.0 -5.0 X 180
Treatment Parameters
• First treatment: -3.50 -2.50 X 180
• Second treatment: plano +2.50 X 90
As a general rule, I will make my standard personalized-nomogram adjustments to each component of the treatment.

The cross-cylinder ablation picks up where the unimeridional, compound/myopic astigmatic ablation falters. The other key benefit of the cross-cylinder ablation is manifest in the form of less regression/greater stability of the treatment effect with higher-magnitude cylinder corrections. Cross-cylinder ablations are also tremendously useful when treating mixed astigmatism. For example:

Preoperative manifest refraction: +4.0 -6.0 X 180
Treatment Parameters
• First treatment: plano -3.0 X 180
• Second treatment: +1.0 + 3.0 X 90 (spherical equivalent added to cylinder of same sign)
The main caveat in using the cross-cylinder ablation approach is that it consumes more total tissue than the unimeridional treatment.8

Microkeratome Selection
The microkeratome style, flap thickness, and hinge location/width may have a significant effect on the percentage reduction of absolute cylinder and the surgical induction of astigmatism. The refractive literature supports the concept that the lamellar corneal flap can induce astigmatism.9,10 Corneal biomechanical response models indicate that creating the lamellar flap can have refractive effects.11 As a general rule, nasal hinge flaps appear to induce minus cylinder at 90º, and superior hinge flaps appear to induce minus cylinder at 180º. This effect is more pronounced with large, thick flaps and steep corneal curvatures (wider hinges).

Other lamellar contributions to astigmatism include micro- and macrostriae, as well as epithelial ingrowth. If a surgeon is contemplating a reoperation for surgically induced astigmatism, he needs to rule out striae and epithelial ingrowth as possible etiologies.

Interactions Between the Corneal Stroma and Excimer Laser Many surgeons take great care preoperatively to mark the cornea and align these marks prior to ablation. Much has been written and discussed in regard to proper axis alignment and cyclotorsion, but, in my clinical experience, cyclotorsion is of relatively low clinical significance. No currently available tracker has the ability to intraoperatively compensate for cyclotorsion once the ablation has started. Cyclotorsion is a relatively infrequent occurrence that is most likely in young, nervous patients and occurs dynamically in a fluctuating fashion. If the surgeon notes cyclotorsion during a high-magnitude astigmatic correction, the globe should be manually fixated.

During the excimer ablation, significant amounts of astigmatism can rarely be induced by the excimer laser beam profile12 but are more commonly the result of ablation decentration, drift, and tilt of the corneal bed. Although tracking systems are quite advanced, very little technology is used to ensure adequate xyz orientation of the corneal bed during the laser ablation. Many times the surgeon depends solely on visually aligning the patient's head and eye under the laser. Additionally, pre-medicating patients with oral benzodiazepines (central nervous system depressants) can potentially wreak havoc with the most sophisticated eye tracking system (patient fixation) currently available. Patient fixation during these treatments is absolutely critical to the high-fidelity reproduction of the theoretical ablation profile to the corneal stroma.

Another cause of induced astigmatism commonly occurs during hyperopic treatments. It results from inadequate clearance between the outer zone of the excimer treatment and the flap hinge. If the hinge is ablated, significant surgically induced astigmatism in the axis of the hinge can occur. Although manual and software-related hinge protection systems are available, induced astigmatism still happens as a result of incomplete ablation profile reproduction to the corneal stromal bed. Induced astigmatism associated with hyperopic treatments is multifactorial. Surgically induced biomechanical effects of large-diameter corneal flaps interacting with poor-fidelity reproduction of the theoretical hyperopic ablation profile can produce unpredictable and unstable astigmatic results. The higher-order optical aberrations created by a low-fidelity reproduction of the complex hyperopic ablation profile will also induce irregular astigmatism associated with decreased BSCVA and UCVA.

Although surgeons will employ many different microkeratomes, excimer laser platforms, and surgical techniques, the previously outlined factors are the building blocks of all astigmatic corrections. As technology becomes increasingly sophisticated and the surgical plan is directed by the software of the laser instead of the physician, it is imperative that surgeons use a cognitive approach to the unique preoperative/postoperative assessment and surgical planning of each patient. Understanding the basics of excimer laser ablation profiles/techniques can potentially maximize the outcomes in simple and complex astigmatic cases.

Sam Omar, MD, is from Advanced Vision Institute in Orlando, Florida. Dr. Omar does not hold a financial interest in the product mentioned herein. He may be reached at (407) 389-0800; omar_eye@yahoo.com.
1. LASIK Eye Surgery. Food and Drug Administration Web site. Available at: http://www.fda.gov/cdrh/LASIK/lasers.htm. Accessed 10/7/2002.
2. Walline JJ, Kinney KA, Zadnik K, Mutti DO. Repeatability and validity of astigmatism measurements. J Refract Surg. 1999;15:23-31.
3. Chayet AS, Montes M, Gomez L, et al. Bitoric laser in situ keratomileusis for the correction of simple myopic and mixed astigmatism. Ophthalmology. 2001;108:303-308.
4. Chayet AS, Magallanes R, Montes M, et al. Laser in situ keratomileusis for simple myopic, mixed, and simple hyperopic astigmatism. J Refract Surg. April 1998;14(suppl 2):175-176.
5. Vinciguerra P, Sborgia M, Epstein D, Azzolini M, MacRae S. Photorefractive keratectomy to correct myopic or hyperopic astigmatism with a cross-cylinder ablation. J Refract Surg. Mar-April 1999(suppl 2);15:183-185.
6. Chayet AS, Suzuki Y, inventors; NIDEK Co., LTD, assignee. Apparatus for operation on a cornea. US Patent 6 136 012. October 24, 2000.
7. Vinciguerra P, Sborgia M, Epstein D, Azzolini M, MacRae S, inventors; NIDEK Co., LTD, assignee. Apparatus for corneal surgery. US Patent 6 315 771. November 13, 2001.
8. Doane JF. Unimeridional ablations for compound myopic astigmatism. Cataract & Refractive Surgery Today. 2002;2:4:55-56.
9. Hersh PS. Surgically-induced astigmatism after LASIK for spherical myopia. J Refract Surg. 2001;17:151; discussion:152.
10. Huang D, Sur S, Seffo F, Meisler DM, Krueger RR. Surgically-induced astigmatism after laser in situ keratomileusis for spherical myopia. J Refract Surg. 2000;16:515-518.
11. Roberts C. The cornea is not a piece of plastic. J Refract Surg. 2000;16:407-413.
12. Maloney RK. Surgically-induced astigmatism after LASIK for spherical myopia. J Refract Surg. 2001;17:151-152.