Bitoric laser in situ keratomileusis for astigmatism

Bitoric laser in situ keratomileusis for astigmatism César Albarrán-Diego, OD, Gonzalo Muñoz, MD, PhD, Robert Montés-Micó, OD, MPhil, Jorge L. Alió, MD, PhD Purpose: To evaluate the efficacy, predictability, and safety of bitoric laser in situ keratomileusis (LASIK) for the correction of mixed astigmatism. Setting: Instituto Oftalmológico de Alicante, Alicante, Spain. Methods: This prospective study included 28 eyes of 21 patients with mixed astigmatism who had bitoric LASIK using the Hansatome威 microkeratome (Bausch & Lomb Surgical) and the Chiron Technolas威 217 excimer laser (Bausch & Lomb Surgical). The main outcome measures were uncorrected visual acuity (UCVA), best corrected visual acuity (BCVA), defocus equivalent, blur strength, and refraction. Results: Six months after bitoric LASIK, the mean UCVA was 0.70 ⫾ 0.23 (SD). The percentage of eyes with a UCVA of 20/40 or better was 78.6% and of 20/20, 21.4%. There was a statistically significant increase in the mean BCVA from 0.71 ⫾ 0.19 before surgery to 0.83 ⫾ 0.15 at 6 months (P ⫽ .0004). Three eyes (10.7%) lost 1 line of BCVA; 19 eyes (67.9%) gained 1 or more lines. The mean preoperative astigmatism of ⫺4.04 ⫾ 1.13 diopters (D) was reduced to ⫺0.67 ⫾ 0.79 D after surgery. The defocus equivalent was less than 1.00 D in 75.0% of eyes and less than 0.50 D in 64.3%. Vector analysis showed that the mean achieved correction was 97.4% of the intended correction. Conclusions: Bitoric LASIK was a safe, effective, and predictable procedure in the treatment of mixed astigmatism. It is a means to improving BCVA in a significant percentage of patients. J Cataract Refract Surg 2004; 30:1471–1478  2004 ASCRS and ESCRS V arious techniques have been used to correct mixed astigmatism. One technique, astigmatic keratotomy (AK), uses transverse or arcuate relaxing incisions along the steepest corneal meridian. Although AK is a powerful technique for correcting mixed astigmatism, poor predictability, with undercorrection and overcorrection, currently limits its use.1–3 Another technique, Accepted for publication November 11, 2003. From the Refractive Surgery Departments, Instituto Oftalmológico de Alicante (Muñoz, Montés-Micó, Alió), Alicante, and Hospital Virgen del Consuelo (Albarrán-Diego, Muñoz, Montés-Micó), Valencia, Spain. None of the authors has a proprietary or financial interest in any material described. Reprint requests to Gonzalo Muñoz, Instituto Oftalmológico de Alicante, Avenida Denia 111, 03015 Alicante, Spain. E-mail: gon. [email protected].  2004 ASCRS and ESCRS Published by Elsevier Inc. the excimer laser, uses a negative-cylinder approach to perform a central ablation along the steepest corneal meridian.4 The negative-cylinder approach flattens the steepest meridian but also induces some flattening in the flattest meridian.2,5 In the case of mixed and simple myopic astigmatism, this coupling effect produces a positive sphere that has to be compensated for by a spherical hyperopic ablation, resulting in more tissue removal.6 Positive-cylinder excimer ablation is a different approach in which the laser steepens the flattest meridian with no significant effect on the steepest meridian because the ablation is not performed in the central cornea.2,7 Chayet et al.2 introduced the bitoric or cross-cylinder laser in situ keratomileusis (LASIK) technique for the correction of mixed and simple myopic astigmatism. Bitoric LASIK flattens the steepest meridian with a central cylindrical ablation and steepens the flattest me0886-3350/04/$–see front matter doi:10.1016/j.jcrs.2003.11.054 BITORIC LASIK FOR MIXED ASTIGMATISM Figure 1. (Albarrán-Diego) Corneal changes following bitoric LASIK in an eye with mixed astigmatism. ridian with a paracentral ablation. The bitoric approach has the advantage of less tissue removal for the same refractive defect by balancing negative and positive ablations.6 In this study, we evaluated the efficacy, predictability, and safety of LASIK for mixed astigmatism using the bitoric or cross-cylinder technique. Patients and Methods The inclusion criteria for this study were older than 18 years of age, congenital mixed astigmatism between 2.0 diopters (D) and 6.0 D, and a stable refractive history for more than 1 year. Exclusion criteria were best corrected visual acuity (BCVA) worse than 20/80, pupil diameter larger than 6.0 mm under dim illumination, evidence of developing cataract, history of uveitis or posterior synechias, corneal dystrophy, intraocular pressure higher than 22 mm Hg, and presence of retinal or optic disc pathology. Twenty-eight eyes of 21 patients with mixed astigmatism operated on at the Instituto Oftalmológico de Alicante, Alicante, Spain, were enrolled in this prospective study. All patients were informed about the details and risks of the LASIK procedure, and a written informed consent was obtained in accordance with the Declaration of Helsinki. No institutional review board approval was required for the study. The preoperative evaluation included uncorrected visual acuity (UCVA), BCVA, refraction, slitlamp biomicroscopy, applanation tonometry, fundus examination, ultrasonic pachymetry, keratometry, and corneal topography. Laser in situ keratomileusis was performed in all patients using the Chiron Technolas威 217 excimer laser with the PlanoScan program (Bausch & Lomb Surgical) and a 2 mm flying spot. Antibiotic prophylaxis before surgery consisted of topical ciprofloxacin (Oftacilox威) every 8 hours for 2 days. Antiseptic prophylaxis was performed by applying 1 drop of povidone–iodine 5% solution to the conjunctiva immediately before surgery. Lamellar keratotomy was performed using 1472 the Hansatome威 microkeratome (Bausch & Lomb Surgical). The operating room was maintained within the temperature and humidity ranges stated in the laser manual. Before each procedure, a fluence test was performed, adjusting the energy of the laser to its optimal value. A bitoric or cross-cylinder pattern of ablation without a spherical correction was used. For this, the refraction in each eye was expressed as 2 crosscylinders without sphere and the ablation was performed first in the steep meridian (negative-cylinder ablation) and then in the flat meridian (positive-cylinder ablation), according to the following formulas: CMyo ⫽ S ⫹ C CHyp ⫽ |C| ⫺ CMyo The task was to calculate the refraction in each meridian separately and to perform a proportional treatment to make both focal lines fall on the retina (emmetropia). For example, for a refraction of ⫹2.00 ⫺4.00 ⫻ 90, a negative-cylinder ablation of ⫺2.00 ⫻ 90 was performed, followed by a positive-cylinder ablation of ⫹2.00 ⫻ 180. For a refraction of ⫹1.75 ⫺4.25 ⫻ 70, a negative-cylinder ablation of ⫺2.50 ⫻ 70 and a positive-cylinder ablation of ⫹1.75 ⫻ 160 were performed. Figure 1 shows the corneal changes following bitoric ablation in an eye with mixed astigmatism. After LASIK, topical tobramycin and dexamethasone eyedrops (TobraDex威) were used every 6 hours for 1 week. Follow-up examinations were at 1 day, 1 week, 1 month, and then every 3 months as necessary. Postoperative examinations included UCVA, BCVA, refraction, slitlamp examination, applanation tonometry, and corneal topography. Data for analysis were extracted from chart reviews; if 2 eyes of the same patient were included in the study, they were analyzed separately. When retreatment was needed, it was performed after 6 months and the data were collected between 2 and 3 months afterward, so a final follow-up of 6 months was achieved in all patients before a retreatment was performed. Retreatment criteria were based on patient dissatisfaction, J CATARACT REFRACT SURG—VOL 30, JULY 2004 BITORIC LASIK FOR MIXED ASTIGMATISM which usually occurred when the difference between postoperative UCVA and postoperative BCVA was greater than 1 line. As described by other authors,8 the retreatment data were analyzed separately from the primary outcome; they were not included in the analysis because the objective of the study was to assess the outcome of the primary crosscylinder LASIK technique. There are several ways of analyzing astigmatic data.9–14 In our study, refractions and keratometric data obtained before and 6 months after LASIK were analyzed following the vectorial method proposed by Thibos and Horner.11 One advantage of this mathematical approach is that astigmatism is represented in rectangular vector form. Hence, conventional scalar methods can be applied to each vector component, and this notation allows the application of standard multivariate statistics to compute population means and variances, define confidence intervals, and test hypotheses. Another characteristic of this vectorial methodology is the mathematical independence of each power vector component, which is called orthogonality. Given this orthogonality, statistical analysis can be applied to each component separately. Using the notation of Thibos and Horner, any spherocylindrical refractive error [S (sphere), C (cylinder) ⫻ ␸ (axis)] can be converted into a set of 3 dioptric powers: M, J0, J45, by the following formulas: M ⫽ S ⫹ C/2; J0 ⫽ (⫺C/2) cos (2␸); J45 ⫽ (⫺C/2) sin (2␸); B ⫽ (M2 ⫹ J02 ⫹ J452)1/2. These formulas can be applied whether the conventional notation is made in negative or positive cylinder. Statistical analysis was performed with the SPSS statistics software package SPSS/Pc ⫹11.0 for Windows (SPSS Inc.). Normality of the data in each group was confirmed by the normal probability plots. Monovariate t test and Hotelling T 2 test of multivariate statistics were used for astigmatism analysis. Table 1. Results of bitoric LASIK for mixed astigmatism. Before LASIK After Bitoric LASIK (6 Mo) UCVA 0.46 ⫾ 0.18 0.70 ⫾ 0.23 BCVA 0.71 ⫾ 0.19 0.83 ⫾ 0.15 ⫺4.04 ⫾ 1.13 ⫺0.67 ⫾ 0.79 Measurement Astigmatism (D) ⫾1.00 D of defocus (%) 0 75 ⫾0.50 D of defocus (%) 0 64.3 42.9 78.6 UCVA ⱖ 20/40 (%) Loss ⱖ2 lines BCVA (%) — 0 Gain ⱖ1 lines BCVA (%) — 67.8 Mean ⫾ SD BCVA ⫽ best corrected visual acuity; LASIK ⫽ laser in situ keratomileusis; UCVA ⫽ uncorrected visual acuity. in 6 eyes (21.4%). The mean BCVA was 0.71 ⫾ 0.19 before LASIK and 0.83 ⫾ 0.15 after LASIK. It was 20/40 or better in 28 eyes (100%). Three eyes (10.7%) lost 1 Snellen line of BCVA, 6 eyes (21.4%) gained 1 line, 11 eyes (39.3%) gained 2 lines, and 2 eyes (7.1%) gained 3 lines. The BCVA after bitoric LASIK was statistically significantly better than the BCVA before surgery (P ⫽ .0004). Refraction Preoperatively, the mean astigmatism was ⫺4.04 ⫾ 1.13 D (range ⫺2.00 to ⫺6.00 D). Six months after bitoric LASIK, it was ⫺0.67 ⫾ 0.79 D (range 0.00 to ⫺3.00 D); 16 eyes (57.1%) were within ⫾0.50 D of the intended astigmatic correction and 23 eyes (82.1%), within ⫾1.00 D. Results The mean age of the 9 men and 12 women was 34.2 years ⫾ 6.3 (SD) (range 21 to 47 years). The preoperative refractions ranged from ⫺1.50 to ⫺6.00 D of astigmatism with ⫹2.00 to ⫹4.00 D of sphere. The mean preoperative spherical equivalent (SE) was ⫺0.17 ⫾ 0.80 (range ⫺1.50 to ⫹1.00 D). The post-LASIK followup was at least 6 months in all eyes. A summary of the results is shown in Table 1. Safety, efficacy, and predictability of bitoric LASIK for mixed astigmatism are shown in Figures 2 to 4. Visual Acuity Six months after LASIK, the mean UCVA was 0.70 ⫾ 0.23 (range 0.3 to 1.0). At the last examination, it was 20/40 or better in 22 eyes (78.6%) and 20/20 Figure 2. (Albarrán-Diego) Cumulative decimal visual acuity bar chart comparing BCVA before LASIK and 6 months after LASIK. J CATARACT REFRACT SURG—VOL 30, JULY 2004 1473 BITORIC LASIK FOR MIXED ASTIGMATISM A summary of the distribution of the manifest refractive errors after vector conversion before and 6 months after bitoric LASIK is shown in Table 2. There was a reduction in the blur strength (B) after LASIK, Figure 3. (Albarrán-Diego) Cumulative decimal visual acuity bar chart comparing BCVA before LASIK and UCVA 6 months after LASIK. Figure 4. (Albarrán-Diego) Attempted versus achieved SE (M ) and the astigmatic components (J0 and J45) of the power vector analysis. 1474 showing that a relatively wide range of refractive errors was reduced to a narrow distribution near emmetropia. The power vector magnitude was reduced after surgery, and a compression of the overall refractive error data was observed by the reduction in the standard deviations (from 1.3455 to 0.4554 in J0 and from 1.1994 to 0.2548 in J45). The reduction in the mean magnitude of power vector was 66% for M, 103% for J0, and 98% for J45. Although it might seem that bitoric LASIK had little effect on SE (M), the mean preoperative SE in the present study was near zero because eyes with mixed astigmatism were treated. Given that SE (M) may be a poor indicator of the refractive state in mixed astigmatism,15 Figure 5 has been constructed with the defocus equivalent bar chart. Defocus equivalent is computed as the SE, ignoring the sign of the cylinder. In our study, 64.3% of the eyes had a defocus equivalent refraction of 0.50 D or less and 75% were under 1.00 D. To better view the change in astigmatism caused by LASIK, Figure 6 shows the astigmatic component of the power vector as represented by the 2-dimensional vector (J0, J45). The origin in this graph (0, 0) represents an eye free of astigmatism. The spread in the presurgical data is converted into a concentrated data set around the origin after LASIK. A series of 3 monovariate t tests were performed to test the hypothesis that the 3 components of the power vector analysis had a mean that was not statistically different from zero; that is, the refractive error was adequately corrected by LASIK. The results of the t tests revealed no difference from zero in the 3 cases (t test, P ⫽ .558 for M; P ⫽ .678 for J0; P ⫽ .930 for J45). A multivariate Hotelling T 2 test confirmed that the mean power vector after surgery was not significantly different from a vector of zero length (P ⫽ .8363); ie, bitoric LASIK achieved the objectives of correcting astigmatism. Figure 4 shows the attempted versus achieved plot for SE (M) (top) and for both components of astigmatism: J0 (middle) and J45 (bottom). In the case of SE (M), 100% of cases were within ⫾1.00 D; 71.4% were within ⫾0.50 D, and 50% within ⫾0.25 D. Results for J0 were 92.9% of cases within ⫾1.00 D, 89.3% within ⫾0.50 D, and 64.3% within ⫾0.25 D. For J45, 100% of cases were within ⫾1.00 D, 92.9% within ⫾0.50 D, and 82.1% within ⫾0.25 D. J CATARACT REFRACT SURG—VOL 30, JULY 2004 BITORIC LASIK FOR MIXED ASTIGMATISM Table 2. Refractions after vectorial conversion before and after bitoric LASIK. Before Bitoric LASIK Mean SD After Bitoric LASIK (6 months) M J0 J45 B M J0 J45 B ⫺0.1696 1.0955 0.2103 2.1616 ⫺0.0580 ⫺0.0361 0.0043 0.5193 0.7907 1.3455 1.1994 0.5907 0.5176 0.4554 0.2548 0.5152 Manifest refractions in conventional script notation (S [sphere], C [cylinder] ⫻ ␸ [axis]) were converted to power vector coordinates (M, J0, J45) and overall blurring strength ( B ) by the following formulas: M ⫽ S ⫹ C/2; J0 ⫽ (⫺C/2) cos (2␸); J45 ⫽ (⫺C/2) sin (2␸); B ⫽ ( M2 ⫹ J02 ⫹ J452)1/2. Keratometry Analysis Figure 7, left, shows the scatterplot for J0 and J45 calculated with the presurgical and postsurgical keratometry. Changes in keratometry were highly correlated with refractive changes. The mean manifest astigmatism change after bitoric LASIK was 1.13 ⫾ 1.40 D for J0 and 0.21 ⫾ 1.18 D for J45, whereas the mean change in keratometric astigmatism was 0.92 ⫾ 1.04 for J0 and 0.18 ⫾ 1.06 for J45. There were no statistically Figure 5. (Albarrán-Diego) Comparison of the defocus equivalent before and 6 months after LASIK. significant differences between the mean manifest and the keratometric changes (P ⫽ .5251 for J0, and P ⫽ .9222 for J45). Figure 7, right, shows the predictability of keratometric changes after bitoric LASIK. For J0, 96.4% of eyes were within ⫾1.00 D and 78.5% were within ⫾0.50 D. For J45, 96.5% of eyes were within ⫾1.00 D and 92.8%, within ⫾0.50 D. These results are very close to the manifest refractive results. Complications and Retreatment There were no complications related to the lamellar cut. No complication such as a decentered ablation or epithelial ingrowth was seen in any eye. Seven of 28 eyes (25%) needed retreatment for a significant residual refractive defect after bitoric LASIK. In this group, the mean UCVA after retreatment was 0.70 ⫾ 0.10 and the mean BCVA was 0.79 ⫾ 0.07. The UCVA was better than 20/40 in all the eyes. No eye lost a line of BCVA; 4 of 7 eyes gained 1 line (Table 3). Five of 7 eyes were within ⫾0.50 D of the defocus SE refraction, and all were within ⫾1.00 D. Discussion Figure 6. (Albarrán-Diego) Representation of the astigmatic vector (J0 and J45) before and 6 months after LASIK. Mixed astigmatism can be surgically managed using AK, LASIK, or a combination of the 2 techniques. Astigmatic keratotomy has the advantage that by flattening the steep meridian, a favorable coupling effect is achieved with a consequent steepening of the flattest meridian, theoretically in the same proportion.16 Treatment of mixed astigmatism with AK has been associated with a smaller higher-order aberration increase than treatment with excimer laser ablation.17 This may be because AK does not remove corneal tissue and does not ablate the central cornea since the incisions are located at least 3.0 mm from the optical center of the eye. However, poor predictability, regression, and complications limit the use of AK. J CATARACT REFRACT SURG—VOL 30, JULY 2004 1475 BITORIC LASIK FOR MIXED ASTIGMATISM Figure 7. (Albarrán-Diego) Vector analysis of keratometric data. Left: Scatterplot for J0 and J45 calculated with the presurgical and postsurgical keratometry. Right: Predictability of keratometric changes after bitoric LASIK. Mixed astigmatism can be treated using a cylindrical ablation over the central corneal steep meridian (negative-cylinder ablation) in combination with hyperopic spherical ablation to treat the hyperopic spherical component of the refraction and the induced hyperopia resulting from the cylinder ablation over the steep meridian. This approach requires more stromal tissue removal for the same refractive defect.6 In the positivecylinder ablation technique, the cylinder is corrected by steepening the flat meridian and the remaining myopic sphere is treated with a myopic spherical ablation. For mixed astigmatism, in which refraction can be represented as 2 pure cross-cylinders without sphere, it seems logical to apply an excimer ablation profile without spherical correction. Chayet et al.2 introduced the bitoric or cross-cylinder LASIK technique for the correction of mixed and simple myopic astigmatism. It consists of flattening the steepest meridian with a central cylindrical ablation and steepening the flattest meridian with a paracentral ablation, with no particular treatment of the sphere. As observed by Vinciguerra et al.,18 treating astigmatism by splitting the correction over the flat and steep meridians has the advantage of conserving tissue and improv- ing optics. In the present study, the percentage of eyes with a UCVA of 20/40 or better was 78.6% and of 20/20, 21.4%. Chayet et al.2 report 93% of eyes with a UCVA of 20/40 or better and 57% with 20/20 3 months after bitoric LASIK. It should be pointed out that the preoperative BCVA was 20/20 in 17.9% of our study population versus 57% of the population in the Chayet et al. study.2 Rueda and coauthors19 report 85% of eyes with a UCVA of 20/40 or better and 17.5% with a UCVA of 20/20. Previous reports of bitoric LASIK for mixed astigmatism apply different formulas for calculation of the ablation, and they use hypercorrection of the positivecylinder ablation and undercorrection of the negativecylinder ablation to compensate for hyperopic shift and undercorrection of the cylinder.2,19 There are 2 main methods of bitoric ablation for mixed astigmatism. The first option consists of dividing the total amount of cylinder into 2 equal parts (50% each) for positive and negative cylinder ablation, respectively, with a spherical ablation to compensate for the rest of the refractive defect. Vinciguerra et al.18 suggest that this modality could reduce the amount of root-mean-square for tetrafoil and trifoil after surgery. The second method is Table 3. Summary of bitoric LASIK and posterior retreatment outcomes in the 7 eyes. Before Bitoric LASIK After Bitoric LASIK After Retreatment Eye UCVA Refraction BCVA UCVA Refraction BCVA UCVA Refraction BCVA 1 0.30 ⫹2.00 ⫺4.00 ⫻ 175 0.80 0.40 ⫹0.50 ⫺2.50 ⫻ 90 0.70 0.70 ⫹0.50 ⫺0.25 ⫻ 160 0.80 2 0.60 ⫹0.50 ⫺3.00 ⫻ 150 1.00 0.60 ⫹1.25 ⫺1.25 ⫻ 30 0.90 0.70 ⫺0.50 ⫻ 65 0.90 3 0.20 ⫹1.00 ⫺5.00 ⫻ 140 0.60 0.40 ⫹2.50 ⫺3.00 ⫻ 95 0.70 0.50 ⫺0.50 ⫺0.50 ⫻ 90 0.70 4 0.30 ⫹0.75 ⫺4.25 ⫻ 35 0.60 0.40 ⫺0.75 ⫺0.50 ⫻ 100 0.70 0.80 — 0.80 5 0.30 ⫹1.74 ⫺4.25 ⫻ 180 0.50 0.40 ⫺0.25 ⫺1.00 ⫻ 55 0.60 0.70 — 0.70 6 0.40 ⫹4.00 ⫺5.75 ⫻ 5 0.70 0.60 ⫺1.00 ⫺0.75 ⫻ 170 0.80 0.80 — 0.80 7 0.30 ⫹4.00 ⫺6.00 ⫻ 180 0.70 0.50 ⫹1.25 ⫺1.50 ⫻ 180 0.70 0.70 ⫺0.25 0.80 BCVA ⫽ best corrected visual acuity; LASIK ⫽ laser in situ keratomileusis; UCVA ⫽ uncorrected visual acuity 1476 J CATARACT REFRACT SURG—VOL 30, JULY 2004 BITORIC LASIK FOR MIXED ASTIGMATISM to perform a pure bitoric ablation, dividing the refraction into 2 cross-cylinders without a spherical correction. This modality has the advantage of removing less corneal tissue, at least theoretically for the laser excimer model used in the present study. For example, for a refraction of ⫹3.50 ⫺5.00 ⫻ 90, the half-and-half approach would perform an ablation of ⫹1.00 ⫹2.50 ⫻ 180 and ⫺2.50 ⫻ 90, which theoretically removes a maximum of 79 ␮m of corneal tissue. With the pure bitoric approach, an ablation of ⫺1.50 ⫻ 90 followed by one of ⫹3.50 ⫻ 180 would be made, resulting in a maximum theoretical tissue removal of 64 ␮m. The pure bitoric algorithm has been introduced in the last software version (3.11) of the Technolas 217 laser excimer for performing mixed astigmatism. In the present study, the ablations were performed using the cross-cylinder formula to express the refractive error without any adjustment. Another difference between the present study and the previous ones is that we performed the negative-cylinder ablation first, whereas Chayet et al.2 and Rueda and coauthors19 started over the flat meridian. Whether this has a role in the final outcome of the ablation must be determined. When LASIK is performed for the correction of high astigmatism, it is common to observe an increase in BCVA after surgery. With our approach to bitoric LASIK, more than two thirds of the eyes (67.8%) gained 1 or more lines of BCVA. Bitoric LASIK has been associated with a better than presurgical BCVA, with 25% to 35% of eyes gaining at least 1 line of BCVA.2,19 The increase in BCVA observed after bitoric LASIK might be explained by the fact that the optical compensation for high astigmatism produces a larger distortion of the image than that caused by a lens that compensates for smaller degrees of astigmatism. As described by Edwards and Llewellyn,20 meridional differences in retinal image size can reduce visual acuity in astigmatic subjects and this effect is more appreciable in astigmatism over 3.0 D. Thus, a reduction in the amount of astigmatism such as that obtained in our study (from ⫺4.02 to ⫺0.67 D) would improve visual acuity by the reduction of the image distortion. By measuring the retinal image quality using the doublepass technique, Pujol and coauthors21 demonstrate that higher amounts of astigmatism could reduce visual acuity. One conclusion of this study is that the increase in the amount of astigmatism led to a reduction in the retinal image quality. Bitoric LASIK for mixed astigmatism could be considered not only as a technique for avoiding the need for optical compensation, but also as a method for improving visual acuity in a significant percentage of patients. In the present study, 10.7% of eyes lost 1 line of BCVA and no eye lost more than 1 line. In the study by Chayet et al.,2 no eye lost BCVA postsurgery; in the study by Rueda and coauthors,19 10% of eyes lost 2 lines of BCVA. Although decentered ablation may explain some cases of decreased BCVA, other possible explanations such as increases in higher-order aberrations induced by bitoric LASIK should be considered in future studies. In the present study of bitoric LASIK, 71.4% of eyes had SEs within ⫾0.50 D and 100%, within ⫾1.00 D. Chayet et al.2 report 92% and 100% of eyes with SEs within ⫾0.50 D and ⫾1.00 D, respectively, and Rueda and coauthors,19 57.5% and 82.5%, respectively. Sheludchenko and Fadeykina22 report a reduction in the retreatment rate when bitoric ablation was used to correct astigmatism compared with monotoric ablation. The retreatment rate with bitoric ablation was 25% and with monotoric ablation, 44%. In our study, the retreatment rate was 25%. The follow-up was no longer than 6 months in our study, based on our previous experience. In the study by Chayet et al.,2 bitoric LASIK provided high stability, with a mean change of 0.05 D from the first to the third month and of 0.08 D from the third month to 1 year. As proposed by Thibos and Horner,11 we analyzed both components of astigmatism separately, obtaining a reduction in vectorial astigmatism of 103% and 98% for J0 and J45 components of vectorial analysis. Chayet et al.2 and Rueda and coauthors19 used a methodology for vector analysis slightly different than ours in the formulation. Chayet et al.2 report the mean achieved vector magnitude was 91% of intended, whereas Rueda and coauthors2 report it was 76% of intended. If we compute the vector analysis as done in these studies, a mean achieved vector magnitude of 97.4% of intended correction is obtained. In our study, the mean scalar astigmatism magnitude was ⫺4.04 ⫾ 1.13 D before surgery; it fell to ⫺0.67 ⫾ 0.79 D after surgery, which implies a reduction of 83.4%, similar to that of Chayet et al.2 The mean cylinder in the study by Chayet et al.2 was J CATARACT REFRACT SURG—VOL 30, JULY 2004 1477 BITORIC LASIK FOR MIXED ASTIGMATISM ⫺4.02 ⫾ 1.22 D before surgery and ⫺0.35 ⫾ 0.42 D after surgery, a 91.3% decrease in astigmatism magnitude. Rueda and coauthors19 found a reduction in scalar astigmatism magnitude of 71.9%, from a mean cylinder of ⫺3.17 ⫾ 1.52 D before surgery to ⫺0.89 ⫾ 0.74 D at 6 months. Our results indicate that bitoric LASIK is a safe, effective, and predictable procedure. Seventy-eight percent of the eyes in our study achieved a UCVA of 20/40, and two thirds of the eyes had an increase of at least 1 line of BCVA. Studies that focus on the analysis of corneal aberrometric changes induced by the different LASIK modalities are in progress and may help to ascertain the best ablation profile for LASIK in eyes with mixed astigmatism. References 10. 11. 12. 13. 14. 15. 1. Agapitos PJ, Lindstrom RL. Astigmatic keratotomy. Ophthalmol Clin North Am 1992; 5(4):709–715 2. Chayet AS, Montes M, Gómez L, et al. Bitoric laser in situ keratomileusis for the correction of simple myopic and mixed astigmatism. Ophthalmology 2001; 108: 303–308 3. Güell JL, Vazquez M. Correction of high astigmatism with astigmatic keratotomy combined with laser in situ keratomileusis. J Cataract Refract Surg 2000; 26:960– 966 4. McDonnell PJ, Moreira H, Clapham TN, et al. Photorefractive keratectomy for astigmatism; initial clinical results. Arch Ophthalmol 1991; 109:1370–1373 5. Alió JL, Artola A, Ayala MJ, Claramente P. Correcting simple myopic astigmatism with the excimer laser. J Cataract Refract Surg 1995; 21:512–515 6. Azar DT, Primack JD. Theoretical analysis of ablation depths and profiles in laser in situ keratomileusis for compound hyperopic and mixed astigmatism. J Cataract Refract Surg 2000; 26:1123–1136 7. Argento CJ, Cosentino MJ, Biondini A. Treatment of hyperopic astigmatism. J Cataract Refract Surg 1997; 23:1480–1490 8. Salz JJ, Stevens CA. LASIK correction of spherical hyperopia, hyperopic astigmatism, and mixed astigmatism with the LADARVision excimer laser system; the 1478 9. 16. 17. 18. 19. 20. 21. 22. LADARVision LASIK Hyperopia Study Group. Ophthalmology 2002; 109:1647–1657; discussion by RR Kreuger, 1657–1658 Huang D, Stulting RD, Carr JD, et al. Multiple regression and vector analyses of laser in situ keratomileusis for myopia and astigmatism. J Refract Surg 1999; 15:538–549 Alpins N. Astigmatism analysis by the Alpins method. J Cataract Refract Surg 2001; 27:31–49 Thibos LN, Horner D. Power vector analysis of the optical outcome of refractive surgery. J Cataract Refract Surg 2001; 27:80–85 Harris WF. Analysis of astigmatism in anterior segment surgery. J Cataract Refract Surg 2001; 27:107–128 Naeser K, Hjortdal J. Multivariate analysis of refractive data; mathematics and statistics of spherocylinders. J Cataract Refract Surg 2001; 27:129–142 Kaye SB, Patterson A. Analyzing refractive changes after anterior segment surgery. J Cataract Refract Surg 2001; 27:50–60 Waring GO III. Standard graphs for reporting refractive surgery. J Refract Surg 2000; 16:459–466; errata, 492 and 2001; 17:294 Faktorovich EG, Maloney RK, Price FW Jr. Effect of astigmatic keratotomy on spherical equivalent: results of the Astigmatism Reduction Clinical Trial; the ARC-T Study Group. Am J Ophthalmol 1999; 127:260–269 Montés-Micó R, Muñoz G, Albarrán-Diego C, et al. Corneal aberrations after astigmatic keratotomy combined with laser in situ keratomileusis. J Cataract Refract Surg 2004; 30:1418–1424 Vinciguerra P, Sborgia M, Epstein D, et al. Photorefractive keratectomy to correct myopic or hyperopic astigmatism with a cross-cylinder ablation. J Refract Surg 1999; 15:S183–S185 Rueda L, Pineda-Fernández A, Huang D, Nur J. Laser in situ keratomileusis for mixed and simple myopic astigmatism with the Nidek EC-5000 laser. J Refract Surg 2002; 18:234–238 Edwards KH, Llewellyn RD. Optometry. London, Butterworths, 1988 Pujol J, Arjona M, Arasa J, Badia J. Influnce of amount and changes in axis of astigmatism on retinal image quality. J Opt Soc Am A Opt Image Sci Vis 1998; 15:2514–2521 Sheludchenko VM, Fadeykina T. Comparative results between standard and bitoric nomograms for astigmatism correction. J Refract Surg 2001; 17:S238–S241 J CATARACT REFRACT SURG—VOL 30, JULY 2004