Wavefront Guided LASIK with a High-Resolution Aberrometer Induces Less Higher-Order Aberration than Early Custom LASIK

Background: The use of 'high-resolution' aberrometer wavefront-guided LASIK system offers the theretical advantages. To date there have been a few papers that proved the clinical superiority of the 'high-resolution' wavefront-guided LASIK. The purpose of this present study was to To evaluate the surgical outcomes of wavefront-guided laser in-situ keratomileusis (LASIK) via a high-resolution aberrometer to treat myopic refractive error. Methods: This study involved 51 eyes of 26 consecutive patients (high-resolution group) who underwent wavefront-guided LASIK from 2012 to 2015 via high-resolution aberrometer and excimer laser surgical unit at the Baptist Eye Institute, Kyoto, Japan and who were followed for 6-months postoperative. Postoperative outcomes were compared with our historical control data (early custom group) from 2002 to 2007 (51 eyes of 26 patients) treated with the same correction amount. Measured parameters included Visual acuity (VA), spherical equivalent refractive error (SER), higher-order aberration (HOA), and contrast sensitivity (CS). Results: In the high-resolution group and the early custom group, the mean SER was -5.86±2.57 diopters (D) and -5.99±2.52D, respectively, and the mean patient age was 33.8±9.2 and 34.3±6.9 years, respectively. LogCS of 18 cycles-per-degree (cpd) change was more satisfactory in the high-resolution group than in the early custom group. Postoperative spherical aberration change in the high-resolution group was signicantly lower than that in the control group (P<0.05). Decrease of induced spherical aberration produced a positive effect on the post LogCS of 18 cpd. Discussion: The latest wavefront-guided LASIK induced less HOAs than early custom LASIK, and although similar VA and refractive precision ndings were observed, the high-resolution group tended to acquire better visual function post surgery.


Introduction
Since the early 2000s, laser-assisted in situ keratomileusis (LASIK) has been recognized as a standard technique for refractive surgery [1,2]. When performing LASIK, a microkeratome or femtosecond laser rst creates a thin circular " ap" in the cornea, with the surgeon then ablating some corneal tissue using an excimer laser under the ap [3,4]. A 'custom LASIK' measures an eye's optical system and corrects higherorder aberrations (HOA) or suppresses the induction of HOA. Custom ablation includes wavefront-guided ablation [5,6], wavefront-optimized ablation [7,8]. (designed to minimize the induction of spherical aberrations), and topography-guided ablation [9].
The accuracy of wavefront-guided LASIK is ensured via a wavefront sensor that measures the aberrations of the optical wavefront. Initially, wavefront sensors for wavefront-guided ablation had only 75 measurement points within a 7-mm diameter [10,11]. However, the iDESIGN REFRACTIVE STUDIO (Johnson & Johnson, New Brunswick, NJ) makes measurements of 1257 points in that same 7-mmdiameter area [12,13]. This type of wavefront sensor is known as a 'high-resolution' aberrometer. The use of this wavefront-guided LASIK system offers the following advantages. First is the increased number of measurement points. Second is the creation of an ablation pro le using Fourier analysis [14]. The third point is the accuracy of ablation by iris recognition registration [15]. To date there have been a few papers that proved the superiority of latest wavefront-guided LASIK.
The purpose of this present study was to clarify whether or not wavefront-guided LASIK performed with a high-resolution aberrometer improves postoperative visual function more than initial custom LASIK.

Subjects
This study involved the consecutive patients (high-resolution group) who underwent wavefront guided LASIK using the iDESIGN REFRACTIVE STUDIO [12] at the Baptist Eye Institute, Kyoto, Japan between 2012 and 2015, and who were followed up for 6-months postoperative. The control group (early custom group) underwent custom LASIK at Baptist Eye Institute between 2002 and 2007, and in whom the correction amount and patient age were matched to the patients in the high-resolution group. Approval for this study was obtained from the Institutional Review Boards of Kyoto Prefectural University of Medicine and the Baptist Eye Institute, Kyoto, Japan, and prior to enrollment, written informed consent was obtained from all patients in accordance with the tenets set forth in the Declaration of Helsinki.

Indication
Indication criteria for LASIK were as follows, the preoperative corneal topography deemed normal, corrected distance VA (CDVA) better than decimal 0.5, and the residual corneal bed more than 250 μm in depth. As it was required for the preoperative examination, all patients were strictly instructed to discontinue hard contact lens use for 3 weeks prior to surgery, and soft contact lens use for 2 weeks prior to surgery. Wavefront-sensor examination was performed 3 times for natural pupils and 3 times for dilated pupils. Among the wavefront data of the natural pupils, the one with the least excessive accommodation was selected and used as a data source for the corneal ablation pro le. If there was excessive accommodation, the measurement was once-again performed on the day of the operation, and the measurement with the smallest difference from the subjective refraction was used for the surgery.
CS and Visual Acuity (VA) were subjected to a logarithmic transformation and analyzed as a continuous variate with log CS and logMAR. The predictability was calculated from the difference between the target and the refractive error measured 6 months post LASIK. The obtained wavefront data was tted to a sixorder Zernike polynomial, total HOA, coma-like aberration (third order + fth order), and spherical-like aberration (fourth order + sixth order), with the spherical aberration then being calculated. Changes in the CS and HOA values were calculated by subtracting the preoperative values from the postoperative.

Statistical Analyses
The mixed-effect model was used to analyze study variables and changes (i.e., CS at 3 cpd, 6 cpd, 12 cpd, 18 cpd, and in 4mm and 6mm total HOA, Coma-like aberration, spherical-like aberration, spherical aberration) prior to surgery and at 6 months postoperatively with the preoperative and postoperative changes in both eyes as the objective variables, an identi cation (ID) number as the random effect, and each group as a xed effect. The proportion was tested by the Fisher's direct test. The con dence level of the con dence interval was 95%, and a P-value of <0.05 was considered statistically signi cant. Linear regression analysis was used to clarify the relationship between the HOA change and the CS changes. Statistical analysis was performed with JMP PRO VERSION 14 STATISTIC SOFTWARE FOR WINDOWS (SAS Institute Inc., Cary, NC).

Results
This study involved 51 eyes of 26 consecutive patients (high-resolution group). The control group (early custom group) involved 51 eyes of 39 patients. In the high-resolution group and the early custom group, the average SER was -5.87±2.58D and -5.99±2.52D, respectively, and the average patient age was 33.8±9.2 and 34.3±6.9 years, respectively. No signi cant difference was found between the two groups. The background of the patients in both groups is summarized in Table 1.
At 6-months postoperative, the proportion of predictability within ±0.5D and ±1.0D in the high-resolution group was 88.2% (45/51) and 98.0% (50/51), respectively, while that in the early custom group was 90.2% (46/51) and 96.1% (49/51), respectively. In the high-resolution group and the early custom group, the mean SER was -0.25±0.41D and -0.20±0.58D, respectively. The proportion of CDVA decreased by two or more lines in comparison with preoperative results, 2% for the high-resolution group (1/51) and 0% for the early custom group. It should be noted that the CDVA in that 1 eye in the high-resolution group recovered to the preoperative level during the follow-up observation period. No complications resulting in visual impairment were observed in both groups. The postoperative clinical variables of the patients in this study are shown in Table 2.
Comparison of the changes in CS between the high-resolution group and the early custom group revealed no statistically signi cant differences in all cycles. Post LASIK. the mean CS in the high-resolution group at all frequencies was found to have improved compared with that prior to surgery, however, it decreased in the early custom group. The maximum difference between the two groups was 18 cpd, and the average difference was 0.14 log CS ( Table 3).
The HOA change with a pupil diameter of 4 mm between both groups prior to and at 6-months after surgery was compared. A lesser amount of spherical aberration change was observed in the highresolution group compared to the early custom group. Moreover, in the high-resolution group, the changes in total HOA, coma-like aberration, and spherical-like aberration were smaller. The comparison in 6 mm pupil diameter group prior to and 6-months postoperatively revealed statistically signi cant decrease in the total HOA, spherical-like aberrations, and spherical aberrations changes between the high-resolution group and the early custom group. The average value of the coma-like aberration change was also lower in the high-resolution group without reaching statistically signi cance (Table 4). A statistically signi cant correlation was found between the change in the CS of 18 cpd and spherical aberration change in pupil diameter of 4 mm (P=0.007) (Figure 1).

Discussion
In this study, we compared two techniques within the same refractive error and precision outcomes. Our results demonstrate that wavefront-guided LASIK using a high-resolution aberrometer produced less induced HOA, and that the average change in CS was improved compared with early custom LASIK. In addition, our ndings revealed a correlation between the decrease of HOA and the improvement of CS.
It has been reported that wavefront-guided LASIK proved the superiority of postoperative CS in comparison to other types of custom ablation [19,20]., and standard LASIK [21,22]. A previous study reported that the CS after wavefront-guided LASIK using the iDESIGN® Refractive Studio was sharper than that produced by small incision lenticule extraction (SMILE) [23,24]. In this present study, wavefrontguided LASIK using a high-resolution aberrometer induced smaller spherical aberration in the 4mm and 6mm pupils. Moreover, the CS change correlated signi cantly with the increase of spherical aberration change in 4mm pupils.
Our ndings appear to clarify the superiority of performing surgery with a high-resolution aberrometer. In the early custom group, we theorize that the spherical aberration tended to increase due to the fact that wavefront-optimized ablation does not correct the spherical aberration [25,26] and the early wavefrontguided ablation does not consider induced spherical aberration [27]. Another reason for the decrease of spherical aberrations was the femtosecond laser ap [28], as it has been reported that the total HOA and spherical aberration decreased when femtosecond laser was performed in comparison with a microkeratome.
Between 2002 and 2007, three different lasers were assigned randomly to the patinets in our facility. No obvious clinical differences were found in the treated eyes (data not published). However, possible obscuring of the interpretations cannot be excluded.
In conclusion, our ndings revealed that LASIK using a high-resolution wavefront sensor and femtosecond laser produced less induction of HOA compared with the custom LASIK systems that were available for use 10 years ago, and that via this method, CS tend to improve. Therefore, the current LASIK procedure seems to have almost resolved the in uence of induced HOA on visual function, and our ndings indicate that LASIK with a high-resolution aberrometer results in a better visual function post surgery.

Declarations
Acknowledgements Funding/Support: The authors indicate no funding support.