LASIK was introduced as a more accurate method than PRK, incisional keratotomy, or selective suture removal for surgically correcting regular and irregular astigmatism resulting from penetrating keratoplasty. Recently, it has been shown that performing a hinged lamellar keratotomy (analogous to that which is completed during a LASIK procedure) induces a biomechanical response of the cornea that substantially changes its shape.¹ We must therefore keep in mind that this preparation of the LASIK flap in postpenetrating keratoplasty eyes could induce significant refractive changes prior to excimer laser treatment. It seems reasonable, then, to expect changes in the amount of induced astigmatism upon flap creation, a factor that makes it more appropriate to perform excimer laser ablation as a separate, later procedure.

APPRAISAL OF TWO TECHNIQUES
During our study, we compared a two-step (flap creation with laser ablation as a separate, later procedure) versus a one-step approach (flap creation followed immediately by laser ablation) to the LASIK correction of astigmatism following penetrating keratoplasty. We removed the sutures 12 to 18 months after performing the penetrating keratoplasty and then waited an additional 3 months to permit the refraction to stabilize, as confirmed by two successive, similar topographies, before proceeding with either approach. In the two-step technique, we waited 1 to 3 months between cutting the flap and performing laser ablation (except in one patient who did not require ablation, as described later in this article). The study's objective was to ascertain which technique was more beneficial to the visual and refractive outcome of LASIK in the management of postpenetrating keratoplasty astigmatism.

In our study, we divided 22 cases into two groups. Group one comprised 11 eyes, all of which underwent the two-step LASIK technique. The 11 eyes of group two were treated with the one-step LASIK technique. The penetrating keratoplasty buttons were sized 7.5 to 8.0 mm on average, and all of the LASIK flaps were 9.5 mm in order to avoid interruption of the graft/bed junction. The mean preoperative cylinder was -7.55 D ± 1.68 D. The mean preoperative BCVA was 20/100, and the mean BCVA was 20/40 at the final follow-up visit (P<.001).

We performed a full ophthalmic assessment prior to performing the surgery, and we conducted the postoperative follow-up visits of each case at intervals of 1 week and 1, 3, and 6 months. All of the eyes experienced a decrease in refractive cylinder, but the mean improvement was -4.37 D ± 1.79 D versus -2.38 D ± 1.71 D in group one and group two, respectively. The difference between the two groups was statistically significant in favor of group one (P=.018) (Figures 1 and 2).

We conducted vector analysis of the data from group one. After the lamellar cut, the correction of refractive cylinder achieved was as follows: 1.00 D in one case (10%), 2.00 D in four cases (40%), 3.00 D in three cases (30%), 4.00 D in one case (10%), and 5.00 D in one case (10%). Our postablation vector analysis found the following amounts of correction: 3.00 to 4.00 D in five cases (50%), 5.00 to 7.00 D in four cases (40%), and 8.00 D in one case (10%).

ANALYSIS OF RESULTS
Although PRK and LASIK have been proven to significantly reduce refractive errors resulting from penetrating keratoplasty,² LASIK has become the preferred technique due to the high rate of regression and haze observed with PRK, especially for medium and high myopia.³ The results of LASIK performed on postpenetrating keratoplasty eyes with a high degree of astigmatism are superior to those obtained by incisional surgery.⁴ Our study, however, indicates that the two-step approach is probably the most advantageous solution for eyes with postpenetrating keratoplasty astigmatism and that this approach increases the predictability of LASIK in these cases. As regards refractive stability, this study had a follow-up period of only 6 months. A current study under the direct supervision of Dr. Alió at the Instituto Oftalmologico de Alicante in Spain has a larger number of patients enrolled and will include a longer follow-up period. Its aim is to confirm and evaluate our results over a longer period of time.

Following the lamellar cut, no eye exhibited progression of astigmatism, but all experienced a significant change in astigmatism. We must note that, in group one, a lamellar cut alone corrected the astigmatism of one patient. No laser ablation was required. This finding introduces the possibility of incorporating the lamellar cut as a therapeutic tool in and of itself, rather than simply as a first step in LASIK.

SURGICAL CHOICES
There are three possibilities in an eye with astigmatism resulting from penetrating keratoplasty, and the choice of surgical method depends on the degree of astigmatism. The first option, a lamellar cut only, corrected up to 4.00 D of astigmatism in some cases in our study. The second option is to follow the two-step LASIK approach. The third option is to correct the astigmatism by cutting the flap and prescribing spectacles or contact lenses if the patient can tolerate these modalities. In highly astigmatic patients whose corneas are thinner than average due to previous refractive surgery, correcting a portion of their astigmatism with a lamellar cut may enable them to tolerate contact lenses or spectacles that correct their residual refractive error.

CONCLUSION
Creating a LASIK flap modifies both the cylinder and the spherical equivalent of postpenetrating keratoplasty eyes. We found that the most significant change in cylinder (as recorded topographically) occurs within the first few days of flap creation but observed some further progression and regression during the ensuing 1 to 3 months. Although the exact period of time that should elapse between the lamellar cut and laser ablation is currently unknown and may vary between cases, our study proves that there is a significant refractive difference between the one- and two-step techniques. Moreover, in many cases, the one-step technique induces irregular astigmatism. Finally, our results confirm the safety and efficacy of LASIK in these eyes, and our statistical analysis shows that there is a significant reduction of both the spherical equivalent and the refractive cylinder upon creating the LASIK flap.

Hazem E. Haroun, MD, is Assistant Lecturer of Ophthalmology at Cairo University in Cairo, Egypt. Dr. Haroun may be reached at +20 12 743 04 30; harounhazem@hotmail.com.
Jorge L. Alió, MD, PhD, is Head of the Department of Refractive Surgery, Instituto Oftalmológica de Alicante, and Professor and Chairman of the Ophthalmology Department, Miguel Hernández University Medical School, Alicante, Spain. Dr. Alió may be reached at +34 96 515 00 25; jlalio@oftalio.com.

1. Busin M, Arffa RC, Zambianchi L, et al. Effect of hinged lamellar keratotomy on postkeratoplasty eyes. Ophthalmology. 2001;108:1845-1851; discussion:1851-1852.
2. Lazzaro DR, Haight DH, Belmont SC, et al. Excimer laser keratectomy for astigmatism occurring after penetrating keratoplasty. Ophthalmology. 1996;103:458-464.
3. Alio JL, Artola A, Claramonte PJ, et al. Complications of photorefractive keratectomy for myopia: Two-year follow-up of 3,000 cases. J Cataract Refract Surg. 1998;24:619-626.
4. Webber SK, Lawless MA, Sutton GL, Rogers CM. LASIK for post penetrating keratoplasty astigmatism and myopia. Br J Ophthalmol. 1999;83:1013-1018.