Implantation of the Visian Implantable Collamer lens (ICL, STAAR Surgical) provides significant benefits across a broad spectrum of daily activities and serves as a life-changing intervention for numerous high myopia patients (1–5). ICL implantation is now considered a feasible refractive surgery option for moderate and low myopia, given its rapid recovery, excellent refractive outcomes, and reversibility, while avoiding corneal ablation (6). Therefore, heightened standards are essential for the visual and refractive outcomes following ICL implantation.

Surgically induced astigmatism (SIA) resulting from a clear corneal incision (CCI) during surgery is inevitable and represents a crucial factor influencing postoperative visual and refractive outcomes in ICL implantation. The refractive power is consistently diminished along the meridian by corneal incisions, resulting in changes to corneal astigmatism (7, 8). When preparing for the placement of an ICL, different methods are used to adapt to the limited size of the ICL and various eye shapes, with the aim of achieving an ideal vault for long-term safety assurance (9). Combining particular CCI locations with distinct ICL placement orientations serves to minimize rotational angles in surgery, ultimately lowering operation time and mitigating intraoperative complications. Temporal and superior CCIs are widely employed as corneal incision locations in clinical practice. A study by Kamiya et al. (10) revealed a marked increase in corneal astigmatism post-ICL surgery among individuals in the temporal CCI group, while a reduction was noted in the vertical CCI group. Further investigation is required to understand the differences in corneal SIA induced by various CCI locations and the influencing factors to achieve optimal visual and refractive outcomes after ICL placement.

Apart from the effects of CCI sites, corneal SIA following surgery is influenced by parameters such as incision width, incision length, and the distance from the incision to the corneal apex. Numerous studies have established that wider and longer CCIs located in proximity to the corneal apex yield greater corneal SIA after cataract surgery (11–13). ICL implantation differs from cataract surgery in that it is generally undertaken on young adults with superior recovery capabilities and differing corneal astigmatism (14, 15). Additionally, ICL implantation omits the need for phacoemulsification procedures and has relatively less effect on the cornea, potentially offering a more precise assessment of the effect of CCI on corneal astigmatism.

Besides corneal SIA, the existence of increased higher-order aberrations (HOAs) post-refractive surgery significantly influences visual acuity and leads to complications such as glare, ghosting, and compromised night vision (16, 17). Although the introduction of HOAs following ICL implantation is notably lower than that of corneal refractive surgery, there are still observed increases in total HOAs, trefoil, and spherical aberrations (SA) (18–20). Several factors, including corneal incisions, the ICL lens, and the eccentricity and tilt of the ICL lens, are likely responsible for the heightened presence of HOAs post-ICL surgery. However, to the best of our knowledge, still, no studies have reported the correlation between corneal morphological features and corneal HOAs/SIA following temporal or superior corneal incisions during ICL implantation.

In this study, patients were categorized into two groups based on the location of CCIs: the temporal CCI group and the superior CCI group. The visual and refractive outcomes, CCI features, corneal SIA, and corneal HOAs were analyzed for the two groups. Furthermore, associations between CCI characteristics and the levels of corneal SIA and HOAs were investigated in both groups. These findings provide valuable insights into optimizing ICL surgical techniques regarding the corneal incision creation, appropriate ICL refraction selection, and ICL implantation procedure design.

This retrospective investigation involved the analysis of data from patients who received same-day bilateral phakic intraocular lens implantation at the Eye Center, Second Affiliated Hospital, Zhejiang University, Hangzhou, China between April 2021 and March 2023. Informed consent was obtained from all participants. The study received approval from the hospital’s ethics committee and adhered to the principles outlined in the Declaration of Helsinki.

To prevent any confusion from aberration symmetry and Zernike polynomial reconstruction in the fellow eyes, only the right eyes were considered in this study (21, 22). The inclusion criteria were as follows: (i) age between 18 and 45 years, (ii) myopia ranging from −0.50 to −18.00 D with or without astigmatism (−0.50 to −6.00 D), and myopic spherical equivalent refraction (SER) > −18.00 D, (iii) stable refraction for at least 2 years with no increase greater than 0.50 D per year, (iv) anterior chamber depth (ACD) ≥ 2.80 mm, (v) corneal endothelial cell density (ECD) ≥ 2000 cell/mm², (vi) anterior corneal astigmatism<3.00 D. Patients with a history of ocular trauma or surgery, coexisting ocular diseases such as severe dry eyes, corneal scar, keratoconus or topographic suspicions of keratoconus, cataract, glaucoma or suspected glaucoma, uveitis and retinopathy; pregnancy or breastfeeding; systematic diseases that may interfere with corneal healing such as diabetes mellitus and connective tissue disorders were excluded from the study.

Prior to surgery, all patients underwent a comprehensive ophthalmic examination that included slit-lamp biomicroscopy, intraocular pressure (IOP), manifest and cycloplegic refraction, uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), ECD, tear break-up time, corneal tomography (Pentacam HR, Oculus Optikgerate GmbH), optical biometry (IOLMaster 700, Carl Zeiss Meditec AG), ultrasound B scan, fundus photography, and ultrasound biomicroscopy (UBM, Model SW 3200 L, Tianjin Suowei Electronic Technology Co., Ltd.). Prior to the preoperative evaluation, patients were directed to cease wearing soft contact lenses for at least 1 week; rigid gas-permeable contact lenses for at least 1 month; and orthokeratology lenses for at least 3 months.

Through preoperative assessments and the STAAR surgical calculator, the surgeon determined the size and placement orientation of the ICL. Toric implantable collamer lenses (TICL) would be chosen if subjective astigmatism was over 0.75 D or if correcting astigmatism could enhance CDVA by more than 2 lines.

Following the application of the specified criteria, 75 eyes met the requirements for inclusion in the temporal CCI category, with 47 eyes meeting the criteria for the superior CCI category (Figure 2). The distribution of corneal CCI locations (central point) in both groups is presented in Figure 1B.

The demographic and baseline data of the temporal and superior groups are presented in Table 1. A significantly higher proportion of TICL was observed among individuals in the temporal CCI group (p < 0.001). Both groups had comparable baseline clinical characteristics, except for cylindrical diopter (p < 0.001) and ACD (p = 0.003). Furthermore, the postoperative vault in the superior CCI group was significantly lower than that observed in the temporal CCI group (p = 0.009) (Table 1).

Temporal and superior CCIs are commonly used CCI locations during ICL implantation in clinical practice. Consistent with the previous study (10), our data validated the satisfactory safety and efficacy of ICL implantation in both temporal and superior CCI groups (Supplementary Figure S1). Given that a majority of patients underwent ICL implantation for high myopia, postoperative CDVA consistently surpassed preoperative CDVA, owing to the elimination of image reduction caused by glasses wear.

Corneal SIA caused by CCI during ICL surgery is inevitable, which may affect postoperative visual and refractive outcomes. SIA resulting from CCI can be influenced by multiple factors, including surgeon’s experience, incision locations, incision width, IL, Dis-Ex, etc. (13, 28–31). It is widely accepted that the magnitude of corneal SIA is positively correlated with CCI width and IL, while negatively correlated with Dis-Ex (11–13). Our data indicated that longer Dis-Ex and Dis-En were observed in the temporal CCI group compared to the superior CCI group due to an oblate shape of the anterior corneal surface (Table 2). Since both groups used keratome with identical widths, shorter Dis-Ex and Dis-En resulted in a larger Angle-W in the superior CCI group (Table 2). The negative correlation between corneal SIA and Dis-Ex has been reported by previous studies (12, 13). The results of our investigation revealed analogous corneal SIA outcomes in both the temporal and superior CCI groups (as shown in Table 2), and a negative correlation between anterior corneal SIA and Dis-Ex/En was observed solely in the superior CCI group. An explanation could be that the temporal corneal CCI was located at a considerable distance from the corneal apex, causing only a minor alteration in corneal SIA. Moreover, our prior research determined that increased CCI length combined with reduced angles of incision entry and exit could lead to higher SIA values on the anterior and posterior corneal surfaces (12). However, no such typical correlation was observed in this study. This can be ascribed to the discrepancy of surgical maneuver and invasion between ICL and cataract surgery, as no phacoemulsification and I/A procedures are required in ICL surgery, leading to the better uniformity of CCI morphology and hence minor correlations to corneal SIA.

In line with the previous report (10), our data also showed a significant increase in corneal astigmatism in the temporal CCI group and decreased astigmatism in the superior CCI group following ICL implantation (Supplementary Table S1). Most ICL candidates are younger patients, who predominantly have with-the-rule astigmatism. Thus, the superior CCI certainly can ameliorate the corneal postoperative with-the-rule astigmatism, compared with the temporal CCI. Nevertheless, it should be noted that corneal astigmatism can be influenced by other factors such as palpebral fissure size and eyelid compression (32). Squinting caused by refractive errors or tension can lead to a substantial increase in vertical corneal refractive power (32). Correcting ocular refractive error, followed by the natural widening of the palpebral fissure, may cause a decrease in the vertical meridian corneal diopter. Caution should be taken when planning the ICL astigmatism and CCI locations for patients with relatively smaller palpebral fissures or corneal diameters in order to achieve optimal visual and refractive outcomes.

Some patients still experience halo symptoms after ICL surgery (33). In addition to the ICL central hole, the increase of HOAs is the significant reason for postoperative halo (34, 35). Previous studies have reported a significant increase in total HOAs, trefoil, and coma aberration as well as a decrease in SA (negative increase) after ICL implantation (34, 36–39). The substantial reduction in SA was mainly due to the negative SA of the ICL itself that increased with its diopter (40), while the changes in coma and trefoil may be associated with the corneal incision (41).

In our patients, both temporal and superior CCI groups showed stable corneal tHOAs, coma, trefoil, tetrafoil, and 2nd astigmatism over the 4-mm optical zone, except for posterior corneal tHOAs and trefoil (Figure 3). However, for the 6-mm optical zone, significant changes of corneal tHOAs, trefoil, and tetrafoil in both groups, along with 2nd astigmatism in the temporal CCI group, were noted (Figure 3). These findings should partially contribute to the night vision symptoms experienced by most patients postoperatively.

For the 6-mm optical zone Z (3, 3) Zernike term, the two groups exhibited opposing changing patterns, while for 4-mm optical zone Z (3, 3), significant changes were only noted in the superior CCI group (Figure 4). No significant correlation between corneal Z (3, 3) and CCI features was found in both groups (Figure 5). These results suggested that the changes of corneal Z (3, 3) seemed to be mainly influenced by CCI locations, not CCI morphological features. The superior CCI potentially introduces more corneal Z (3, 3) and more postoperative corneal aberrations, as Z (3, 3) is an important HOAs term influencing optical quality.

Another important Zernike third-order term [Z (3, 1)] showed a significant increase in the temporal CCI group (Figure 4). It is understandable that the temporal CCI can affect the horizontal corneal coma, evidenced by the remarkable decrease of anterior corneal K_f in the temporal CCI group (Supplementary Table S1). On the other hand, the superior CCI group showed no significant coma changes, consistent with the insignificant changes of anterior corneal K_s or K_f in the superior CCI group (Supplementary Table S1). The underlying mechanism ought to be complex, including the additional effects caused by upper eyelid cover or compression as described above, besides the incision relaxing effect itself. Therefore, compared with the temporal CCI, the superior CCI location might be helpful to minimize the introduction of postoperative coma.

Similarly, only the superior CCI group displayed a connection between corneal HOAs and CCI features. The shorter Dis-Ex/En result was closely related to the changes of Z (4, −4) and Z (4, 4). An increase in the Angle-W resulted in higher values for corneal Z (3, −3) and Z (4, −4) (Figure 5F). The findings suggested a potential link between post-surgery corneal HOAs and specific characteristics of CCI such as Angle-W, Dis-En, and Dis-Ex, particularly when the incision was located superiorly. Incorporating an extended Dis-Ex/En and a diminished Angle-W in the CCI design could be beneficial in decreasing corneal HOAs following ICL implantation.

When determining the optimal placement of CCIs for ICL surgery, both superior and temporal incisions offer unique benefits and challenges. The selection of incision site should be guided by a comprehensive assessment of factors that impact postoperative success and complication risk. Superior incisions are advantageous for correcting pre-existing with-the-rule astigmatism, particularly when non-toric ICLs are used. By strategically placing the incision to flatten the steeper corneal meridian, superior incisions can effectively address astigmatism (42). They may also be less influenced by eyelid movements, which can result in reduced postoperative discomfort, enhanced wound stability, and accelerated healing. Furthermore, the support provided by the patient’s forehead can be beneficial for less experienced surgeons (43). However, superior incisions carry a higher risk of postoperative ocular surface fluid entering the anterior chamber, potentially increasing the likelihood of infection and inflammation (44). Conversely, temporal incisions provide better access in patients with prominent brows or deep-set eyes and are less likely to induce astigmatism, making them an attractive option for patients with minimal pre-existing astigmatism (45). However, temporal incisions may be less effective in astigmatism correction compared to superior incisions and are more susceptible to exposure-related complications. The choice between superior and temporal CCIs should be tailored to the individual patient’s ocular anatomy, the presence of pre-existing astigmatism, and the surgeon’s expertise and preference.

There are several limitations in this study. Firstly, compared with the ICL, more repetitive checking and adjusting maneuvers were performed to ensure accurate alignment of the TICL axis with the intended orientation during TICL implantation, which might have a subtle influence on the corneal contour. Secondly, there were discrepancies in the angles of the central point of CCI (β) within the superior and temporal groups (Figure 1B), potentially introducing a bias in the analysis. Therefore, conducting additional research with a larger number of participants could lead to more conclusive findings. Thirdly, the single-center observational retrospective design employed in our study is a significant limitation. To strengthen our conclusions, it is imperative to conduct prospective multi-center studies with larger sample sizes in the future. Lastly, our study was limited in scope as it primarily focused on objective assessments of corneal SIA and HOAs, with visual acuity being the sole subjective measure. Given that patients may encounter a range of visual symptoms beyond mere changes in visual acuity, such as glare, halos, night vision difficulties, and fluctuating vision, our analysis did not encompass the full spectrum of postoperative visual experiences. Further research is required to investigate the correlation between the location of CCIs and ocular HOAs, Also, the effect of CCI location on visual quality, as quantified through questionnaires, warrants further investigation to provide a more holistic understanding of the implications of incision site selection on overall visual performance and patient satisfaction.

Collectively, the corneal SIA after ICL implantation was negatively correlated with Dis-En and Dis-Ex and was comparable in magnitude between the 3.0-mm temporal and superior CCIs. The Zernike 3rd and 4th order terms were significantly associated with the CCI location along with specific CCI characteristics, especially in the superior CCI group. Our findings indicated that the CCI location design can cause subtle differences in postoperative corneal SIA and HOAs. For better corneal optical quality and optimal visual outcomes post-ICL surgery, it is advantageous to have a CCI that is distanced further from the corneal apex, irrespective of whether it is located superiorly or temporally.