Visual Outcomes Early after Implantable Collamer Lens V4c Implantation in Different Preoperative Spectacle Correction: Full Correction vs. Under Correction

Abstract Purpose To investigate visual outcomes early after implantable collamer lens (ICL) V4c implantation between patients with fully corrected and under-corrected spectacles preoperatively. Methods Patients who implanted ICL V4c were divided into the full correction (46 eyes/23 patients) and under-correction groups (48 eyes/24 patients) based on preoperative differences between the spherical diopter of the spectacles and the actual spherical diopter. Refractive outcomes, scotopic pupil size, higher-order aberrations, and subjective visual outcomes as assessed using a validated questionnaire were compared between the two groups 3 months postoperatively. Moreover, the relationships between the severity of haloes and postoperative ocular or ICL parameters were analyzed. Results At the 3-month follow-up, the efficacy indices in the full correction and under-correction groups were 0.99 ± 0.12 and 1.00 ± 0.10, respectively; the safety indices were 1.15 ± 0.16 and 1.15 ± 0.15, respectively. Total-eye spherical aberration (p < 0.0001) and internal spherical aberration (p = 0.0005) were significantly different pre- and post-operatively in the under-correction group, while no differences were found in the full correction group. Total-eye spherical aberration (p = 0.002) and the severity of haloes (p = 0.03) were postoperatively different between the two groups. The severity of haloes was associated with postoperative spherical aberration (total-eye spherical aberration: r = −0.32, p = 0.002; internal spherical aberration: r = −0.24, p = 0.02). Conclusion Good efficacy, safety, predictability, and stability were obtained early after surgery regardless of preoperative spectacle correction. Patients in the under-correction group possessed a shift to negative spherical aberration and reported greater severity of haloes at the 3-month follow-up. Haloes were the most common visual symptoms after ICL V4c implantation and the severity of them was correlated with postoperative spherical aberration.


Introduction
The "myopia boom" has attracted widespread attention in the twenty-first century. 1 To date, spectacles remain the most widely used method for myopia correction, 2,3 yet many people with myopia do not wear spectacles on a regular basis. Additionally, long-term use of under-corrected spectacles is relatively common among patients given the traditional belief that under-correction can slow the development of myopia 4,5 , and the purchase of spectacles only based on a cursory exam in an optometry shop rather than a standard hospital-based assessment. Long-term under-correction inevitably modifies the mechanism underlying near triad accommodation to adapt to the new refractive state. 6 In addition, under-correction can blur the retinal image, making it difficult to recognize and easily causing asthenopia. 7 As positive results from longitudinal studies assessing the safety, efficacy, stability, and predictability of implantable collamer lens (ICL, STAAR Surgical, Monrovia, California, USA) V4c implantation are reported, [8][9][10][11] an increasing number of patients have opted for this surgery over wearing spectacles. However, it remains unknown whether visual outcomes in patients with under-corrected spectacles after ICL implantation differ from those of patients with full correction. Previous research has focused on comparing visual outcomes in varying preoperative myopic degrees after ICL implantation [12][13][14] and has partly neglected preoperative spectacle correction. In clinical practice, some patients with under-corrected spectacles preoperatively are not entirely satisfied after surgery due to subjective symptoms, despite postoperative improvements in vision. Thus, evaluating purely objective postoperative measures, such as refractive outcomes is insufficient, highlighting the need for subjective assessments (e.g. questionnaires) as well.
The quality of vision (QoV) questionnaire is a validated and viable tool for assessing subjective visual outcomes after ICL implantation. This questionnaire measures 10 symptoms including glare, haloes, starbursts, hazy vision, blurred vision, distortion, double or multiple images, fluctuation in vision, difficulties in focusing, and difficulties in distance or depth perception. The QoV questionnaire has been used in patients with and without myopia correction by different kinds of spectacles, contact lenses, refractive surgeries, and each symptom is scored along the three subscales: frequency, severity, and bothersomeness. 15 This study aimed to compare early subjective and objective visual outcomes following ICL V4c implantation between patients wearing fully corrected and under-corrected spectacles preoperatively. To our knowledge, no similar study assessing these variables in this study population has been published.

Study design and patients
This prospective study recruited patients who underwent ICL V4c implantation for the correction of myopia at the Department of Ophthalmology, First Affiliated Hospital, College of Medicine, Zhejiang University, in China from March 2022 to June 2022. This research strictly adhered to the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Zhejiang University College of Medicine, First Affiliated Hospital. All the patients voluntarily provided informed consent. The same skilled examiner (TZ) executed each measuring item for three times, and the average value was recorded for analysis.
Patients whose spectacles had a spherical diopter lower than that of the eyes (!0.75 D) were classified into the under-correction group, while those in whom the difference 0.50 D were classified into the full correction group. None of the patients in this study intentionally chose undercorrected spectacles for a more comfortable near vision.

Surgical procedure
Antibiotic eye drops (0.5% levofloxacin, Santen, Japan) were used four times per day starting 3 days before surgery. The same surgeon (YS) performed all surgeries using a routine procedure. After pupil dilation and the application of ocular surface anesthesia, a 3-mm transparent corneal incision was made, following which a viscoelastic agent (sodium hyaluronate, Bausch Lomb, China) was injected. The ICL was implanted into the eye using a customized push injection device (STAAR Surgical, USA). Thereafter, the ICL was adjusted to an appropriate position using a positioning hook, and finally, the viscoelastic agent was manually removed using Ringer's lactate solution. Tobramycin dexamethasone drops (Novartis, Switzerland) and 0.5% levofloxacin were administrated four times daily for a week postoperatively, after which 0.1% fluorometholone (Santen, Japan) was used four times every day for 2 weeks.

Measurement of HOAs
The iTrace aberrometry measured the HOAs under the "Total" pattern in a dark room (under constant luminance 0.1 lux). Patients were asked to blink briskly before capturing measurements to prevent tear film-related wavefront distortion, and a well-focused final image was captured when the total rejection points were 10.
The root means square values of HOAs, including total HOA, coma aberration, trefoil aberration, and spherical aberration (SA) were recorded under the same pupil diameter (4.0 mm).

Subjective visual outcomes
All patients were required to complete the QoV questionnaire (in Chinese version) using Wechat, an online messaging application, to evaluate their subjective visual outcomes three months after ICL implantation. The QoV questionnaire comprises 30 items, each rated along a 4-point scale, and the seven symptoms mentioned firstly are accompanied by a simulated illustration to enhance understanding. 16

Statistical analysis
Statistical analyses were performed by R version 3.6.2 (R Project for Statistical Computing) and GraphPad Prism version 9 (GraphPad Software, USA). Categorical variables are expressed as percentages and frequency, while continuous variables are presented as the mean ± standard deviation (SD). The D'Agostino-Pearson test was used to investigate the normality of continuous data. Student's t-test was performed to inspect age differences between the two groups. The distributions of categorical variables were compared by the chi-square test and Fisher's exact test. Linear mixed model analyses were performed to assess eye-level variables and differences in HOAs under a 4-mm pupil diameter, and correlations were evaluated to determine the relationship between the data from both eyes of the same patients. The Mann-Whitney U test was performed to investigate the differences in subjective outcomes between the QoV questionnaires of the two groups. Spearman's correlation test was applied to analyze the associations between the severity of haloes and postoperative ocular or ICL parameters.

Study population
A total of 47 patients (94 eyes) were included in this study, of these, 23 patients had fully corrected spectacles preoperatively (two men and 21 women) and 24 had under-corrected spectacles (three men and 21 women) for at least 2 years. There were no group distinctions in age, sex, spherical error, cylinder, spherical equivalent, CDVA, IOP, ACD, CCT, WTW, AL, LT, ICL size, ICL power, and optical zone diameter of ICL (all p > 0.05), as shown in Table 1. The mean spherical diopters of spectacles were À8.29 ± 2.11 D in the full correction group (range À13.00 to À4.25 D) and À7.78 ± 2.06 D in the under-correction group (range À11.00 to À2.50 D). The mean insufficient corrected diopters in the under-correction group were À1.36 ± 0.63 D (range À3.00 to À0.75 D).

Efficacy
The mean efficacy indices (ratio of postoperative UDVA/preoperative CDVA) at 3 months postoperatively were 0.99 ± 0.12 in the full correction group and 1.00 ± 0.10 in the under-correction group (p ¼ 0.74). Ninety-eight percent (45/46) of the eyes in the full correction group and 100% (48/48) of eyes in the under-correction group had a visual acuity of 20/25 or better. All eyes had a visual acuity of 20/32 or better (Figure 1(A)). Eighty percent (37/46) of eyes in the full correction group and 83% (40/48) of eyes in the under-correction group had a postoperative UDVA better or equal to the preoperative CDVA.

Safety
Surgical procedures were performed successfully without intra-or postoperative complications observed during the 3-month follow-up. The mean safety indices (ratio of postoperative CDVA/preoperative CDVA) at 3 months postoperatively were 1.15 ± 0.16 and 1.15 ± 0.15 in the full correction and under-correction groups, respectively (p ¼ 0.94). No eyes in either group lost two or more lines of CDVA (Figure 1(B)), and there were no significant differences in ECD before the surgery vs. 3 months after surgery in both groups (full correction group: p ¼ 0.58; under-correction group: p ¼ 0.21). At the 3-month follow-up visit, the ECD and vault values in the full correction vs. the under-correction groups were 2730 ± 198 vs. 2755 ± 220 cells/mm 2 (p ¼ 0.46) and 600 ± 222 vs. 570 ± 224 mm (p ¼ 0.37), respectively. An acceptable vault range between 250 and 750 mm was achieved in 74% (34/46) of eyes in the full correction group and 77% (37/48) of eyes in the under-correction group (p ¼ 0.81). No cataract or anterior subcapsular opacity was observed. Intraocular pressure at 3 months postoperatively was 15.05 ± 2.74 mmHg in the full correction group and 14.87 ± 2.65 mmHg in the under-correction group, respectively (p ¼ 0.39).

Predictability
Ninety-two percent (42/46) of eyes in the full correction group and 90% (43/48) of eyes in the under-correction group had an SE within ± 0.5 D (p > 0.99), and all eyes were within ± 1.0 D (Figure 1(C)). In 65% (30/46) of eyes in the full correction group and 77% (37/48) in the under-correction group, astigmatism was within 0.25D (p ¼ 0. 16), and values for all eyes were within 1.0 D (Figure 1(D)). Seventyeight (36/46) of eyes in the full correction group and 83% (40/48) in the under-correction group achieved an SE within ± 0.5D of the attempted SE (p ¼ 0.61), as shown in Figures 1(E,F).

Stability
One month after surgery, the mean SE values were À0.05 ± 0.42 D and À0.07 ± 0.46 D in the full correction and under-correction group, respectively. At the 3-month followup visit, the mean SE values were À0.04 ± 0.47 D and À0.04 ± 0.52 D in the full correction and under-correction group, respectively (Figure 1(G)). There were no significant differences in stability between the two groups (all p > 0.05).

Scotopic pupil size and higher-order aberration
Measurements of scotopic pupil size and HOAs are summarized in Table 2. No significant differences in scotopic pupil size and HOAs before surgery were found between the two groups. At the 3-month follow-up, only postoperative SA was significantly different between the two groups (totaleye: p ¼ 0.002, internal: p < 0.0001). In the under-correction group, total-eye SA (p < 0.0001) and internal SA (p ¼ 0.0005) were significantly different between the preoperative and postoperative assessments, and the variations in total-eye and internal SA was À0.05 ± 0.06 and À0.08 ± 0.14 mm, respectively. In contrast, there were no significant differences in corneal SA (p ¼ 0.14). SA did not significantly change in the full correction group postoperatively (all p > 0.05). Coma aberration decreased in both groups after surgery (full correction group: p ¼ 0.0005, under-correction group: p ¼ 0.02).

Subjective visual outcomes
The results of subjective QoV outcomes between the two groups are summarized in Figure 2, Tables 3 and S1. Among all symptoms, haloes and glare were the most prevalent in both groups and had the highest ratings on all three subscales. Meanwhile, distortion was the least prevalent symptom, occurring in only two patients in the under-correction group, who ranked the degree of distortion as mild or a little bothersome. A significant difference in the severity of haloes (p ¼ 0.03) was observed between the two groups, and the prevalence of haloes was significantly higher in the under-correction group than in the full correction group (p ¼ 0.002). No other significant differences in frequency, severity, or level of discomfort were observed between the two groups.

Analysis of correlation
Ninety-four eyes from 47 patients were included in the correlation analysis, and the subjective outcomes for the severity of haloes and postoperative ocular/ICL parameters are shown in Table 4. Total-eye SA (r ¼ À0.32, p ¼ 0.002) and internal SA (r ¼ À0.24, p ¼ 0.02) exhibited significant, negative correlations with the severity of haloes. No other significant associations were found.

Discussion
In this investigation, both groups presented good efficacy, safety, predictability, and stability consistent with results from previous studies. [17][18][19][20] As no statistical difference in safety index, efficacy index, ECD, vault, residual SE, and the residual cylinder was found between these two groups 3 months after surgery, this study demonstrated that efficacy, safety, predictability, and stability can be obtained equally early after ICL implantation regardless of preoperative spectacle correction. Regarding subjective visual outcomes, haloes were the most frequently experienced, severely perceived, and bothersome visual symptom, and the severity of haloes in the under-correction group was greater than that in the full correction group 3 months after implantation. Mohr et al. reported that halos were the most commonly perceived long-term visual disturbance after implanted ICL with a central hole; however, the disturbance was minor in most cases. 21 This is thought to be due to refraction, reflection, and beam displacements occurring on the inner wall of a central hole with ICL V4c. 22 Wei et al. reported a higher prevalence of haloes in the TICL-treated group than the ICL-treated group, and the toricity of ICL was found to be correlated with the frequency, severity and bothersome of haloes. 18 In this current study, the severity of haloes was associated with postoperative SA. Macedo-de-Ara ujo et al. postulated that the artificial induction of relatively higher negative spherical aberration values degrades the visual quality of the light source in normal eyes under large pupil dilation conditions, such as night driving or night viewing of bright light sources in a dark environment. 23 Hence, this explains the correlation between postoperative SA and the severity of haloes, since haloes are visual disturbances caused by image distortion under dark conditions. 23 Although there was no significant correlation between the severity of haloes and scotopic pupil size in current results, Lim et al. demonstrated that mesopic pupil size was a risk factor associated with haloes after ICL V4 implantation, 24 more studies on pupil sizes under different conditions are warranted.
A significant shift to internal negative SA of the undercorrection group was observed postoperatively. These findings are similar to those of previous studies related to orthokeratology and soft contact lenses. [25][26][27][28] This is thought to be because compared to an emmetrope, myope in spectacle correction has to accommodate and converge less for a close target due to the prismatic effect. 29 After myopia correction, the demand for near-sightedness accommodation increases, and the imbalance between the demand and the actual function in accommodation also produces a continuous and appropriate amount of accommodative stimulus. 30 Negative SA can enhance the slope of the accommodation stimulus-response curve. 31 Previous studies have reported that the discoverable internal negative SA after using orthokeratology is an ocular adaption caused by the lens as a change in accommodative response. 25,32,33 We speculated that this mechanism might exist in eyes after ICL implantation. Although the implanted ICL can achieve the aspherical effect eliminating part of innate spherical aberrations owing to the hydrophilic nature of the collamer material, considering the shift to negative SA did not present in the full correction group, the postoperative negative SA in the under-correction group was probably induced by the intrinsic lens. Moreover, patients in the under-correction group exhibited lower degrees of accommodation and convergence than those in the full correction group under the same condition, likely due to long-term low-dose mobility of the ciliary muscle in which contraction remains at a low level. 34 A study conducted by Yuan et al. confirmed that the relatively sufficient spectacle correction group possesses more accommodation amplitude than the relatively insufficient group before and 1 month after excimer laser in situ keratomileusis surgery. 35 Therefore, eyes in the under-correction group probably produce more accommodative stimulus as there is a greater imbalance between the demand and the actual function in accommodation compared to the full correction group, and changes in internal SA may lead to an increase in the accommodative response. Future studies should assess postoperative accommodation in patients with long-term insufficient spectacle correction to verify this mechanism. As such, we are currently conducting a prospective study to compare accommodation function between patients with fully corrected and under-corrected spectacles after ICL implantation is proceeding. The decrement of coma aberration and no significant difference in trefoil aberration in both groups are partially consistent with previous studies using iTrace; 36,37 however, other equipment reported contrasting outcomes. 17 The possible explanation is that variability between individualistic HOAs is high, 34 and outcomes might differ based on the apparatus used or under different pupil sizes.
In the current study, we first focused on comparing subjective and objective visual outcomes between the full correction and under-correction groups. Although the findings of the current study are preliminary, we hope this paper will inspire more research related to the rehabilitation of patients with long-term under-condition of spherical errors preoperatively. Furthermore, future studies should investigate different types of binocular visual functions, visual quality, and pupil diameters under different conditions postoperatively between the full correction and under-correction groups.
There were a few limitations in this investigation. Firstly, the follow-up period was short; more follow-up visits would have allowed for longitudinal comparisons to investigate whether the duration time of haloes differed between the two groups. Secondly, the sample size was small, highlighting the need for future studies involving larger populations. Moreover, the range of differences between the spherical diopter of spectacles and the actual spherical diopter was relatively large in the under-correction group; thus, this group should be further divided based on the different degrees of under-corrected conditions in future studies.
In conclusion, ICL implantation obtained good efficacy, safety, predictability, and stability early after surgery  The Mann-Whitney U test was used for intergroup comparison, Ã p < 0.05. regardless of the preoperative conditions of spectacles worn. Notably, there was a shift to negative SA in the under-correction group, and the severity of haloes at the 3-month follow-up was greater in those with under correction than in those with full correction preoperatively. Haloes were the most prevalent visual disturbances in patients after implanting ICL V4c, and the severity of haloes was correlated with postoperative SA. Surgeons should inform patients of the risks of visual disturbances early after surgery in preoperative publicizing and education, especially in patients with long-term under-corrected spectacles. Furthermore, surgeons should pay more attention to postoperative spherical aberration in patients who reported dissatisfaction with subjective visual outcomes at the follow-up visit and explain the commonality of this phenomenon during postoperative rehabilitation to patients.