To evaluate the visual refractive and aberrometric outcomes of laser-assisted in situ keratomileusis (LASIK) surgery for the correction of high mixed astigmatism using a new-generation excimer laser and optimized aspherical profiles.
Retrospective interventional case series.
Fifty-two eyes of 36 patients (21-53 years) with primary mixed astigmatism over 3.0 diopters (D) were included. All cases underwent LASIK surgery using the sixth-generation excimer laser Amaris with cyclotorsion control and a femtosecond platform for flap creation. Visual, refractive, corneal topographic, and aberrometric outcomes were evaluated during a 3-month follow-up. Refractive astigmatic changes were analyzed by Alpins method.
A significant reduction of refractive sphere and cylinder was observed 3 months postoperatively ( P = .001), with an associated improvement of uncorrected distance visual acuity ( P = .001). Best-corrected distance visual acuity (CDVA) remained unchanged in 31 eyes (59.6%), while 3 eyes (5.76%) lost 2 lines of CDVA. Fourteen eyes (26.9%) had spherical equivalent (SE) within ±0.5 D of emmetropia and 34 (65.3%) had SE within ±1.0 D of emmetropia. No significant difference was observed when comparing surgically induced and target astigmatism. A significant induction of higher-order aberration attributable to increase of spherical aberration was found ( P = .003). Seven eyes (13.4%) required retreatment.
LASIK for primary high mixed astigmatism using optimized aspherical profiles and a fast-repetition-rate excimer laser with cyclotorsion control is a safe, effective, and predictable procedure. Induction of higher-order aberrations is still present in the correction of the refraction error of the magnitude included in this study.
The treatment of mixed astigmatism, one of the most peculiar astigmatic errors, has been a challenge for refractive surgeons over the last 3 decades. Mixed astigmatism exists when one meridian is in focus in front of the retina and the other meridian is in focus behind the retina. To correct this refractive error with the excimer laser, the cornea must be relatively steepened in one meridian and flattened in the other.
Nowadays, new excimer laser platforms are designed to carry out a better ablation profile with a larger optical zone, relatively steepening the flat meridian for the hyperopic part of mixed astigmatism and flattening the steep meridian of the myopic part. Centration has been optimized at the corneal vertex, which is the nearest point to the visual axis. The cyclotorsion of the eye is also considered, leading to a lesser induction of aberrations and, hence, better visual acuity of the patients postoperatively.
Flap creation has also been recently improved by the introduction of femtosecond laser. A better control of the parallel flap faces and the creation of larger and more accurate femtosecond laser flaps might be beneficial to perform larger-diameter mixed-astigmatism ablation with laser in situ keratomileusis (LASIK) and thus achieve more positive visual outcomes. The use of femtosecond LASIK surgery has demonstrated better refractive results for the correction of low to moderate hyperopia and high myopia.
In spite of technical innovations and recent surgical developments, no reports have been published over the past 10 years, to the best of our knowledge, on the use of LASIK for mixed astigmatism higher than 3.0 diopters (D).
The aim of the current study was to measure the efficacy, predictability, safety, and aberrometric outcomes of femtosecond LASIK surgery in primary astigmatism over 3.0 D using a high-repetition-rate excimer laser platform with cyclotorsion control and an optimized aberration-free profile, and to analyze the astigmatic changes using the Alpins vector analysis method.
Patients and Methods
This was a retrospective, consecutive, interventional, nonrandomized, noncomparative case series study.
This study evaluated 52 eyes of 36 patients with ages ranging from 21 to 53 years (mean age of 36 ± 8.7 years) that underwent uneventful primary LASIK surgery between March 2008 and January 2010. The postoperative follow-up period was 3 months.
Inclusion criteria were patients with a minimum age of 20 years and motivated for the refractive surgical correction of their refractive error, eyes with primary mixed astigmatism over 3.0 D, stable refractive error for 12 months before surgery, normal peripheral retina or treated with photocoagulation when necessary, and no previous ocular surgery or other ocular comorbidities such as corneal disease, glaucoma, or history of ocular trauma. If patients wore contact lenses, they were discontinued for at least 4 weeks before surgery. The exclusion criteria were patients younger than 20 years, lens opacities, irregular cornea at corneal topography, and eyes with amblyopia and a potential for far vision of less than 20/40 (0.3 logMAR). According to the pachymetry, calculated postoperative corneal stromal bed thicknesses of more than 250 μm at the thinnest corneal area were left following surgery. The refractive outcomes of the postoperative visits were separated into 2 groups: treatment outcomes and retreatment outcomes. Before the surgical intervention each patient was adequately informed about the surgery, its risks, and its benefits, and patients signed an informed consent in accordance with the Helsinki Declaration. Approval from the Ethical Board Committee of our institution was obtained for this investigation to perform the retrospective analysis of this series of cases.
Preoperative Examination and Preparation
The preoperative examination included the following tests: uncorrected distance visual acuity (UDVA); corrected distance visual acuity (CDVA); refraction (manifest and cycloplegic); slit-lamp biomicroscopy; applanation tonometry; ultrasonic pachymetry (OcuScan RxP; Alcon Laboratories Inc, Fort Worth, Texas, USA); scotopic, low, and high mesopic pupillometry (Procyon Pupillometer P2000SA; Procyon Instruments Ltd, London, UK); and corneal topography and corneal aberrometry (CSO, Firenze, Italy) with 6-mm pupils in all cases and fundus evaluation.
One day before surgery, the patients were given tobramycin and dexamethasone (Tobradex; Alcon Cusi SA, Barcelona, Spain) eye drops. After providing informed consent, the patients were prepared for surgery by applying 1 drop of topical anesthesia of oxybuprocaine hydrochloride and tetracaine hydrochloride (colircusi Anestesico Doble; Alcon, Barcelona, Spain).
Excimer Laser and Ablation Profile
LASIK procedures were performed using the sixth-generation Amaris excimer laser (Schwind Eye-Tech-Solutions, Kleinostheim, Germany). This excimer laser belongs to the newest generations of excimer laser technology that integrates diagnostic data (corneal topography and aberrometry, total eye aberrometry), intraoperative pach-ymetry information, optimized centration and tracking control of the ablation, optimized or customized ablation profiles, and an improved control of the energy and thermal energy delivered on the cornea. The Amaris Schwind system has a very fast repetition rate of 500 Hz and incorporates 2 levels of fluence. A high fluence level is used in the first 80% of the treatment to speed up the procedure, and for the remaining 20% of the treatment a low fluence level is used to ensure the smoothness of the ablated surface. A small beam size of 0.54 mm with a super-Gaussian ablative spot profile is delivered in a randomized flying spot pattern to reduce the successive overlapping of the laser spot and minimize the thermal load to the cornea.
This laser platform has incorporated a 5-dimension high-speed eye tracker with an acquisition speed of 1050 Hz that tracks both the pupil and the limbus simultaneously with a reaction time of less than 3 ms. This platform also includes a high-resolution online pachymetry that provides information about the thickness of the cornea in a real-time basis throughout the entire duration of the treatment.
All treatments performed in the current series were based on optimized aspherical ablation profiles and calculated using the commercially available software ORK-CAM from Schwind (Schwind Eye-Tech-Solutions). These optimized aspherical aberration-free profiles consider a focus-shift balance because of tissue removal and a compensation factor for the loss of efficiency when the laser hits the cornea in a nonnormal incidence in order to avoid the induction of aberrations and to balance the aberrations that are present in the treated eye. With the same purpose, a multi-dynamic aspherical transition zone was always created.
All surgical procedures were performed at Vissum Corporacion, Alicante, Spain by the same experienced surgeon (J.L.A.). First, the designed treatment with the ORK-CAM software was loaded into the excimer laser computer and reviewed by the surgeon to confirm the data. As previously noted, the Amaris excimer laser from Schwind (Schwind Eye-Tech-Solutions) was used to perform all the LASIK treatments. The corneal flap was created with the IntraLase femtosecond laser (60-kHz IntraLase femtosecond system; IntraLase Corp, Irvine, California, USA), using the following parameters: temporal hinge, 9.5 mm diameter, pocket function in status “ON,” and flap thickness of 100 μm.
Ablations were centered on the corneal vertex using the pupillary offset; that is, the distance between the pupil center and the normal corneal vertex measured by videokeratoscopy (CSO). The pupillary offset measurement was translated into the treatment plan as cartesian coordinates and then manually entered in the excimer laser computer. The optical zone of the treatment was selected according to the preoperative scotopic pupil size. Depending on the pachymetry, optical zones with at least the same diameter as the scotopic pupil were targeted in order to avoid uncomfortable optical effects. In the current series, the mean optical zone was 6.50 ± 0.16 mm (ranging from 6.5 to 7 mm) and the mean ablation zone (automatically calculated by the system when the optical zone is selected) was 8.04 ± 0.32 mm (ranging from 7.61 to 8.5 mm). The targeted postoperative refraction was emmetropia in all cases.
Static Cyclotorsion Between Upright and Supine Positions
Cyclotorsion control and axis cognition mode were used in all cases. The AMARIS system includes an eye registration module for the eye tracker subsystem in which the diagnostic image is taken as a reference and compared with an eye tracker image under the AMARIS system obtained before the ablation is started to determine the static cyclotorsion component (SCC).
Dynamic Cyclotorsion During Ablation
The AMARIS system includes an eye registration module for the eye tracker subsystem in which the first eye tracker image under the AMARIS system obtained when starting the ablation is taken as a reference and compared with any further eye tracker images to determine the dynamic cyclotorsion component (DCC). The mean, minimum, and maximum DCC values were recorded for this study.
After surgery patients were treated with a standard combination of tobramycin and dexamethasone (Tobradex; Alcon) eye drops 4 times daily for 1 week. Postoperative examinations were at 1 day and 1, 3, and 6 months. On the first postoperative day, a detailed slit-lamp examination was performed to evaluate the flap position and the integrity of the cornea. At the other postoperative visits, UDVA, manifest and cycloplegic refractions, CDVA, slit-lamp biomicroscopy, and total and corneal wavefront aberrations were recorded. Independent observers performed the examinations and collected the data. The astigmatic vector analysis was performed at 3 months after surgery.
The criteria for retreatment included at least 2 of the following 3 parameters: (1) manifest spherical equivalent (SE) of −1.0 D or greater; (2) UDVA of 0.3 logMAR or worse; and (3) patient dissatisfaction with the visual result. Undercorrection was defined as a SE of −1.0 D or greater at the first postoperative visit. LASIK retreatments were performed by lifting the flap and re-ablating the stromal bed with the Amaris excimer laser (Schwind Eye-Tech-Solutions). All retreatments were customized according to the corneal wavefront inferometry provided by the corneal topography. For the purpose of correction, wavefront refraction and correction of spherical and coma aberrations were performed on 7th order of Zernike polynomials. After ablation, the flap was placed to its original position and the interface was irrigated copiously. All enhancements were performed at least 3 months after the initial surgery. Postoperative visits after retreatment are included but are separated from the primary treatment outcome data in the present report.
Vector Analysis of Astigmatic Changes
All astigmatism vector analyses were carried out according to the Alpins method with ASSORT software (ASSORT Pty Ltd, Cheltenham, Australia). The following para-meters were determined and evaluated: target astigmatism, which is the vector of the intended change in cylinder for each treatment; surgically induced astigmatism, which is the vector of the real change achieved; and the difference vector, which is the additional astigmatic change that would enable the initial surgery to achieve its intended target. From the following parameters the following vectors were calculated and analyzed at each postoperative visit: magnitude of error, which is the arithmetic difference between the magnitude of the surgically induced astigmatism and target astigmatism; angle of error, which is the angle described by the vectors of the surgically induced astigmatism and target astigmatism; and correction index, which is the ratio of surgically induced astigmatism and target astigmatism (ideal value is 1, with overcorrection for values larger than 1 and under-correction for values lower than 1). This analysis was performed for refractive and corneal astigmatism.
Statistical analysis was performed with the SPSS statistics software package version 10.1 for Windows (SPSS Inc, Chicago, Illinois, USA). Normality of the data analyzed was confirmed by the Kolmogorov-Smirnov test. When parametric analysis could be applied, the Student t test for paired data was used for the comparison between preoperative and postoperative data, and also between postoperative consecutive visits. However, when nonparametric tests were needed, the Wilcoxon rank sum test was applied. Differences were considered statistically significant when the associated P value was less than .05. Correlation coefficients (Pearson or Spearman, depending on whether normality condition of the data could be assumed) were used to assess the correlation between different variables. Visual acuity was converted to logarithm of the minimal angle of resolution (logMAR) from the decimal notation for statistical analysis, using a visual acuity conversion chart.
The main outcome measures of this study were efficacy (percentage of eyes that showed equal or better UDVA compared to preoperative CDVA), safety (percentage of eyes that lost ≥2 lines [Snellen] of CDVA after the primary procedure compared to preoperative CDVA), predictability (percentage of eyes within ± 0.5 D and ± 1.0 D of the intended correction), and anterior corneal surface high-order aberrations and percentage of eyes losing or gaining lines of CDVA.
A total of 52 eyes were followed up for 3 months. The mean (± standard deviation) age of 38 men and 14 women was 35.92 ± 8.7 years (range 21-53 years). The preoperative refraction ranged from −3 to −5.25 D of astigmatism with 0.25 to 5.00 D of sphere. The mean SE was 0.47 ± 1.09 preoperatively and −0.09 ± 0.69 D at 3 months ( P = .001). The mean defocus equivalent was 4.36 ± 1.51 D preoperatively and 1.10 ± 0.80 D at 3 months ( P = .001).
As SE may be a poor indicator of the refractive state in mixed astigmatism, the defocus equivalent was also calculated as the SE, but ignoring the sign of the cylinder.
Postoperative UDVA improved significantly at 3 months after the surgery ( P = .001). No significant change in CDVA was observed preoperatively and 3 months postoperatively ( P = .63). Table 1 summarizes the preoperative and postoperative visual and refractive outcomes obtained in this study.
|Preoperative||3 Months After Surgery||P b|
|Sphere (D)||2.41 ± 1.26 (0.25 to 5)||0.46 ± 0.61 (−0.75 to 2.50)||.001|
|Cylinder (D)||−3.89 ± 0.70 (−3 to −5.25)||−1.11 ± 0.67 (−3.50 to 0)||.001|
|Spherical equivalent||0.47 ± 1.09 (−1.63 to 2.5)||−0.09 ± 0.41 (−1 to 0.88 )||.001|
|Defocus equivalent||4.36 ± 1.51 (1.88 to 7.63)||1.10 ± 0.80 (0 to 4.25)||.001|
|UCVA Snellen||0.31 ± 0.20 (0.05 to 0.85)||0.80 ± 0.19 (0.30 to 1)||.001|
|CDVA Snellen||0.94 ± 0.14 (0.6 to 1.2)||0.93 ± 0.11 (0.70 to 1.2)||.63|