Laser in Situ Keratomileusis to Correct Postkeratoplasty Refractive Errors
Alfred T.S. Leung,Dennis S.C. Lam,Srinivas K. Rao
DOI: https://doi.org/10.1016/s0886-3350(99)00173-x
1999-01-01
Abstract:We congratulate Forseto and coauthors1 on the excellent results they report in eyes having laser in situ keratomileusis (LASIK) to correct refractive errors after penetrating keratoplasty (PKP). This is the largest reported series of LASIK in postkeratoplasty eyes and demonstrates good refractive outcome, minimal endothelial changes, and safety when performed as early as 18 months after PKP. They highlight the importance of performing LASIK after removal of all sutures, when refractive and topographic stability has been attained. A stable refraction indicates a low activity level of wound healing and remodeling and, therefore, adequate wound healing. In addition to improving predictability of the refractive procedure, this also ensures safety of the lamellar cut, which we have reported.2 However, in the current study, the authors do not indicate the following information (for each patient) while presenting the clinical data—timing of LASIK after PKP (mean 5.3 months ± 4.9 (SD); range 1.5 to 19.0 years), the pachymetric readings, the endothelial count (mean 1596 ± 729 cells/mm2), and the keratometric readings (mean 44.64 ± 2.46 diopters [D]). Corneal graft thickness after PKP has been reported to be quite variable.3 A statistically significant increase in corneal thickness, correlating significantly with low endothelial cell densities, has been reported 5 and 10 years after PKP.4,5 A thicker corneal graft, from a stressed endothelium, is likely to have an increased stromal hydration level. This could affect the predictability of LASIK since surgical algorithms assume a state of normal corneal hydration. A thinner than normal corneal graft would limit the amount of myopic correction that can be performed, since at least 250 μm of the posterior stromal bed is required to prevent postoperative keratectasia.6 As the authors point out, Pallikaris and Siganos7 report up to a 4.11% endothelial cell loss after LASIK, although this has not been duplicated in other studies. These authors hypothesized that the endothelial loss was caused by the shock waves of photoablation. Ing et al.5 report a mean endothelial cell density of 958 ± 471 cells/mm2 (coefficient of variation 0.32 ± 0.11; hexagonal cells 56% ± 12%) 10 years after PKP. Thus, possible deleterious effects of LASIK on the corneal endothelium assume greater significance in these eyes with a suboptimal cell count. Theoretically, a scanning-spot laser would be safer, as the shock wave produced by the small beam is much less. Treatment of high amounts of astigmatism, often seen in postkeratoplasty eyes, using this laser would also induce less hyperopic shift. Although average keratometry in the postkeratoplasty patients reported by Forseto and coauthors (44.64 ± 2.46 D) is similar to the normal range, earlier reports have indicated higher values—51.8 D,8 45.8 D, 50.5 D, and 51.75 D.9 Under the standard photorefractive keratectomy (PRK) treatment protocol, the number of laser pulses delivered to the cornea to achieve a desired correction is determined by use of a standard linear nomogram.10 This nomogram is dependent on the patient's intended dioptric correction and optic zone diameter only, without other patient characteristics or preoperative clinical data. The use of a neural network computer program to determine the PRK nomogram has been described. Using preoperative patient data, the program revealed a marked sensitivity to age, sex, keratometry readings, and intraocular pressure.11 The nomogram predicted the need for more pulses in eyes with lower keratometry readings. The standard nomograms work well in normal refractive patients, but a decline in predictability has been noted with an increase in preoperative myopia and attempted dioptric correction.12 Using these nomograms to plan treatment of very steep or flat corneas can affect the predictability of the outcome. These factors could explain earlier reports of uncertain refractive outcome after PRK in postkeratoplasty eyes.13 Reactivation of herpes simplex keratitis and corneal graft rejection have been described after excimer PRK14 and phototherapeutic keratectomy, respectively.15 Although these complications have not been reported after LASIK in postkeratoplasty eyes, they are areas of concern. Thus, while we agree with Forseto and coauthors that studies with larger follow-up are required for a more definitive analysis of the efficacy of LASIK in postkeratoplasty eyes, we feel that graft thickness, endothelial count, and keratometry are important parameters that can help improve the predictability and safety of the procedure in these eyes. Alfred T.S. Leung FRCS, FRCOphth Dennis S.C. Lam FRCS, FRCOphth Srinivas K. Rao MD aShatin, Hong Kong bMadras, India