Comparison of refractive stability and myopic regression after photorefractive keratectomy with different optical zones using Mel 90 excimer laser

Background: To compare the changes in clinical outcomes and the degree of regression between a optical zone (6.0 OZ) and following (PRK). Methods: The records of 95 eyes that had undergone PRK with a 6.0 OZ (n = 40) and a 6.5 OZ (n = 55) were retrospectively reviewed. We compared data including the spherical equivalent of manifest refraction (SE of MR), simulated K (Sim K), thinnest corneal thickness, change in thinnest corneal thickness (the initial value divided by corrected diopter ΔTCT/CD), Q value, corneal higher order aberrations (HOAs) and spherical aberration (SA) pre-operation, at 3 and 6 months postoperative and at the last follow-up visit (Mean; 20.71 ± 10.52, 17.47 ± 6.57 months in the 6.0 and 6.5 OZ group, respectively). Results: There were no significant differences in the SE of MR, Sim K and UCVA between the 6.0 OZ group and the 6.5 OZ group over 1 year of follow-up after PRK, and the 6.0 OZ group required less ΔTCT/CD than the 6.5 OZ group. The 6.5 OZ group showed better results in terms of post-operative HOAs of RMS, SA and Q value. When comparing that pattern of change in Sim K, there was no significant difference between the 6.0 OZ group and the 6.5 OZ group. Conclusions: The clinical refractive outcomes and regression after PRK using Mel 90 excimer laser with a 6.0 OZ were comparable to those with a 6.5 OZ.


Post-operative management and evaluation
Topical levofloxacin 0.5% solution (Cravit; Santen, Japan) and fluorometholone 0.1% solution (Ocumetholone; Samil, Korea) were started immediately after the surgical procedure. The eye drops were applied four times daily for 1 week, two times daily for 1 month and finally once a day for another 2 months. Routine post-operative follow up visits were scheduled at 1 and 4 days, 1 week, and 1, 3, and 6 months, 1 year and every year thereafter. Uncorrected visual acuity (UCVA) was assessed, and manifest refraction, corneal topography, wavefront analysis and slit lamp biomicroscopy were performed at each visit.

Statistical analysis
Statistical analyses were performed using SPSS software (version 24.0; SPSS Inc.). The mean ± standard deviation (SD) was calculated. Data were first assessed for a normal distribution. In cases of normal distribution, group comparisons were made using an independent two-sample t-test and twoway repeated measures analysis of variance (ANOVA). A minimum sample size of 21 eyes per group was necessary to detect a difference in clinical outcomes between groups to achieve 80% statistical power and 0.05 probability using a two-sided two-sample unequal-variance t-test. The magnitude of the effect chosen was the significant difference previously found when comparing a 6.0 mm optical zone and a 6.5 mm optical zone with the largest required sample size [22]. One-way repeated measures ANOVA with Bonferroni multiple comparisons was performed to determine the difference between data from post-operative visits at 3 and 6 months and from the last follow-up visit for each group. The Mann-Whitney U test and Wilcoxon signed ranks test were used for continuous data that did not approximate a normal distribution. A P-value less than 0.05 was considered significant.

Results
The sample included 95 eyes from 48 patients. The last follow-up visit was done after 12 months postoperatively in both groups, and the longest follow-up period was 38 months in the 6.0 OZ group and 30 months in the 6.5 OZ group. There were no significant differences in mean age of subjects and preoperative manifest refractions, Sim K, thinnest pachymetry, Q value, RMS of HOAs, or spherical aberration between the two groups ( Table 1).
The Sim K values after PRK showed a significant increase in Sim K from the six-month visit to the last visit in the 6.0 OZ group only (p < 0.001). However, there was no significant interaction between optical zone size and the change in pattern by time (p = 0.118, RM ANOVA), and there was not a significant difference in Sim K between the 6.0 OZ group and the 6.5 OZ group at each of the followup visits (3 months, p = 0.594, 6 months, p = 0.907, last follow up, p = 0.667) ( Table 2, Fig. 1).
The thinnest corneal thickness increased from six months post-operatively to the last visit (6.0 OZ, p < 0.001; 6.5 OZ, p < 0.001), but there was no significant change between three and six months postoperatively (6.0 OZ, p = 0.851; 6.5 OZ, p = 0.057). The thinnest corneal thickness after PRK was thinner in the 6.0 OZ group than that in the 6.5 OZ group across all of the follow-up visits (3 months, p = 0.035, 6 months, p = 0.03, last follow up, p = 0.035). However, the thinnest corneal thickness after PRK depended on the amount of correction, thus the changes in thinnest corneal thickness from pre-operative values were divided by the corrected diopter (ΔTCT/CD) to adjust ablation depth by optical zone. Significant differences in ΔTCT/CD were found from the 6-month to the last visit in the 6.0 OZ group (p = 0.001) and the 6.5 OZ group (p < 0.001), but not from the 3-month to 6-month visits (6.0 OZ, p = 0.061; 6.5 OZ, p = 0.058). The 6.5 OZ group had a significantly greater ΔTCT/CD than the 6.0 OZ group at each follow-up visit (3 months, p = 0.018, 6 months, p = 0.043, last follow up, p = 0.024) ( Table 2, Fig. 1). The 6.5 OZ group showed significantly lower RMS of HOAs than the 6.0 OZ group at each follow up visit (3 months, p < 0.001, 6 months, p = 0.001, last follow up, p = 0.005). Significant differences in the RMS of HOAs were found from the six-month to the last visit in the 6.0 OZ group only (p = 0.043). The 6.5 OZ group had a significantly lower SA than the 6.0 OZ group at each follow up visit (3 months, p < 0.001, 6 months, p < 0.001, last follow up, p < 0.001). There were no significant differences from 3 months to 6 months post-operatively (6.0 OZ, p = 0.132; 6.5 OZ, p = 0.667), or from 6 months postoperatively to the last visit (6.0 OZ, p = 0.191; 6.5 OZ, p = 0.137) ( Table 2, Fig. 1).

7
The Q value was significantly lower in the 6.5 OZ group than that in 6.0 OZ group at each follow up visit (3 months, p < 0.001, 6 months, p < 0.001, last follow up, p < 0.001). There were no significant differences from three months to six months post-operatively (6.0 OZ, p = 0.590; 6.5 OZ, p = 0.496), or from 6 months post-operatively to the last visit (6.0 OZ, p = 0.079; 6.5 OZ, p = 0.148) ( Table 2, Fig. 1).

Discussion
Current laser ablation techniques have developed with an increase in frequency and improved tracking system enabling more precise corneal surface remodeling [5]. Furthermore, customized optic design has reduced postoperative mechanical complications and pain, and yielded faster recovery times with better clinical outcomes [6,8,23]. The newer excimer lasers with advanced profile have allowed surgeons to perform surface ablation surgeries with a larger optical zone by enhanced projection error compensation [24].
Previous studies have shown advantages of a larger optical zone for laser refractive surgery. Kim et al. [9] reported more night vision problems and ablation decentration with a 5.0 mm optical zone than with a 6.0 mm optical zone. The study by O'Brart et al. [14] showed less initial overcorrection and regression with a 6.0 mm optical zone than with a 5.0 mm optical zone over 6 months of follow-up.
Rajan et al. [13] reported a larger optical zone had better results in terms of early hyperopic shift, regression, and night haloes for patients who had PRK with 4.0 mm optical zone, 5.0 mm optical zone, and 6.0 mm optical. We compared a 6.0 mm optical zone and the larger optical zone (6.5 mm) using more clinical factors and found no significant differences between the groups. This result may due to the fact that we used the newer excimer laser with a more advanced ablation profile than previous studies.
Seo et al. [22] compared aberration factors after PRK with 6.0 mm optical zone and 6.5 optical zone.
In their study, RMS of HOAs in the larger zone (6.5 mm) group were less than those in the 6.0 mm optical zone group, and both groups showed a significant increase in SA from preoperatively to three months postoperatively. In the present study, we found that the RMS of HOAs and SA in the 6.5 OZ group were lower than in the 6.0 OZ group at 3, 6, and over 12 months after surgery which was 8 longer follow-up period than the previous study of Seo et al [22]. We also focused on intragroup changes after surgery, and found that there was a decrease in the RMS of HOAs from 6 months to beyond 12 months post-operatively in the 6.0 OZ group, but not in the 6.5 OZ group, and no significant changes in SA after 3 months post-operatively in either group.
Based on early clinical studies of PRK, a deeper ablation depth was thought to increase corneal haze [3,17]. However, Rajan et al. [13] reported a larger zone yielded less corneal haze, and recent studies pointed out that a wound edge with a larger ablation zone had a more gradual slope, which enabled smooth epithelial migration and minimized hyperplasia and corneal haze [25,26]. In terms of biomechanical stability, reducing residual bed thickness is related to the weakening of cohesive tensile strength, tangential tensile strength and shear strength [27][28][29], which is more likely to result in mechanical postoperative adverse effects such as keratoectasis [14,18,30]. Therefore, when surgeons set the optical zone for PRK, there should be a limitation on optical zone size based on biomechanical stability and other individual factors such as refractive error, pupil size and age.
Manifest refraction is considered a clinically significant factor influencing patient satisfaction after PRK. In our study, there was no significant regressive change in either the 6.0 OZ group or 6.5 OZ group, and no significant differences between the two groups at each visit. This was consistent with the outcomes for UCVA in the 6.0 OZ group and 6.5 OZ group.
Sim K in the 6.0 OZ group increased significantly from six months to beyond twelve months postoperatively in this study (by paired t-test). However, the difference was 0.3 D, which is not considered clinically significant, and we found that the regression patterns from preoperative visit to any postoperative visit for the 6.0 OZ group and 6.5 OZ group were not significant (p = 0.118, 2 way repeated measured ANOVA). Therefore, there was no significant clinical regression in Sim K in either group.
On the other hand, the RMS of HOAs, SA, and Q value were lower in the 6.5 OZ group than those in the 6.0 OZ group, which meant PRK with 6.5 OZ might yield better visual quality. Corneal haze with epithelial hyperplasia and stromal remodeling can occur after laser refractive surgery [31], and this thickening can be evident up to 12 months after PRK [31][32][33][34], which makes clinical outcomes measured in this period unstable. In this study, the last visits were at least 12 months postoperatively, thus PRK with 6.5 mm optical zone showed lower aberration values than PRK with 6.0 mm zone even in the long term.
The RMS of HOAs showed significant changes between 6 months and beyond 12 months postoperatively likely because it is more reflective of the numerous Zernike's coefficients (SA, COMA, tetrafoil, etc.), but SA is more a clinically interesting factor related to characteristics of visual quality such as glare, halo and monocular diplopia; patients might be clinically more affected by SA than the RMS of HOAs after PRK. Moreover, there is individual variation in neural transfer function, and common clinical measures of visual function are not sensitive at low levels of aberration [35][36][37].
Therefore, better aberration values are not necessarily correlated with improved visual performance.
The thinnest corneal thickness in this study increased significantly until the last follow up visit, but the differences ranged from 3 to 6 μm, which is not clinically significant for patients. The differences may be due to measurement error and/or intra-examination error in thinnest pachymetry, depending on which point of the cornea was selected for measurement. Results would be more reliable if the thinnest corneal thickness was measured at a designated point with ultrasound repeatedly at every visit.
In this study, we performed PRK using the Mel 90 excimer laser with the Triple-A profile which is improved from previous ASA and TSA profiles in that it has better error compensation function and target asphericity control, which minimizes ablation depth, adjusts for spherical aberration and prevents myopic regression [38]. The surgeon can also choose a frequency up to 500 Hz across the surgery in Mel 90 excimer laser, which enables faster ablation time for correction [38,39]. Thus, clinical outcomes after PRK with the same optical zone size can differ according to which excimer laser used, and it therefore seems inappropriate to compare this study with clinical data from studies that used the previous version of the excimer laser for PRK in the same way.
Previous long-term follow up studies after PRK were up to 12 years [13]. However, that study compared different optical zones (4.0 mm, 5.0 mm, 6.0 mm), and in our study (6.0 mm, 6.5 mm) we used a different excimer laser and ablation profile. A longer study seems necessary in that we noted a significant increase in the thinnest corneal thickness from 6 months postoperatively to beyond 12 months, although this study had the longest follow-up period of all clinical studies after PRK with the Mel 90 excimer laser to date.

Conclusions
The 6.5 OZ group had better RMS of HOAs, SA and Q value during the follow-up period in this study.
However, it is questionable whether a difference in wave front index could lead to optical performance improvement. We found that there were no significant differences in SE of MR, Sim K, UCVA, or regression patterns between groups. In other words, clinical outcomes and regression after PRK with a 6.0 mm optical zone was comparable to those after PRK with a 6.5 mm optical zone.
The optical zone selected for refractive surgery depends on individual refraction and keratometer profile, and biomechanical safety, which is based on several factors including total corneal thickness, residual bed thickness and ablation design. Therefore, further studies and effort are necessary for clinical ophthalmologists to compare between 6.0 mm optical zone and 6.5 mm optical zone, and additional studies of the 6.3 mm optical zone are required to better identify optimal optical zones.

Funding
Not applicable.

Ethics approval and consent to participate
This study received ethical approval from the Institutional Review Board of the Kangbuk Samsung Hospital (Seoul, Korea; IRB_2015-02-033). Informed consent was waived because of the retrospective nature of the study.

Consent for publication
Not applicable.