Toric versus Standard Implantable Collamer Lens Implantation for Low Myopic Astigmatism: A 12-Month Retrospective Study Authors: Haiyan He, Zhanchi Hu, Runhua Peng, Liangqi Feng, Chi Xiao Published: 2026 ABSTRACT AbstractBackgroundImplantable collamer lens (ICL) and toric ICL (TICL) have shown effectiveness for myopic and astigmatic patients. This study aims to compare the visual and refractive outcomes of TICL and ICL implantation for low myopic astigmatism.MethodsThis single-center, retrospective, non-consecutive, and non-randomized study included 124 eyes of 124 patients with low myopic astigmatism who underwent TICL or ICL implantation. Visual acuity, refractive outcomes, visual quality, and ocular parameters were assessed at baseline, 3 months, and 12 months postoperatively. The surgical results were reported using standard graphs and double-angle astigmatism vector analysis.ResultsBoth groups demonstrated excellent safety (TICL: 1.28±0.18; ICL: 1.19±0.16) and efficacy (TICL: 1.21±0.19; ICL: 1.09±0.16), with high predictability for spherical equivalent refraction (R2 of 0.980 and 0.984, respectively) and stable refractive outcomes at 12 months. UDVA was marginally better in the TICL group (−0.08±0.07 vs −0.03±0.06, p<0.001). Crucially, TICL yielded significantly lower residual astigmatism (−0.16±0.28 D vs −0.33±0.40 D, p < 0.05) and a smaller prediction error on vector analysis. Despite these refractive differences, objective visual quality metrics improved significantly in both groups, with no statistically significant intergroup differences observed.ConclusionsBoth TICL and ICL are safe and effective for low myopic astigmatism (0.75–1.0 D). Although TICL yielded significantly lower residual astigmatism and higher uncorrected visual acuity than ICL, the nonrandomized design and selection bias necessitate caution in claiming comparative superiority. Toric correction for such low astigmatism should be individualized, integrating clinical benefits with economic and safety factors.Keywords: implantable collamer lens, low astigmatism, vector analysis, visual outcome, visual quality Introduction Myopia and astigmatism are common refractive errors in clinical practice that can adversely impact visual acuity and quality of life.1 Large-scale epidemiological studies indicate that astigmatism ≤1.0 D predominates among adults.2,3 Astigmatism exceeding 0.5 D may result in clinically significant visual impairment, thereby underscoring the need for precise correction.4 With advances in refractive surgery, corneal‑based procedures such as laser in‑situ keratomileusis (LASIK) and small incision lenticule extraction (SMILE) have demonstrated high efficacy and safety profile.5,6 More recently, implantable collamer lens (ICL) implantation has emerged as an effective alternative for correcting low to moderate myopia.7,8 Specifically, the toric implantable collamer lens (TICL) has shown surgical outcomes comparable to those of LASIK and SMILE in patients with astigmatism.9 For astigmatism ≤2.0 D, TICL has demonstrated superior predictability and refractive accuracy compared to ICL combined with corneal relaxing incisions.10,11 Moreover, Wei et al confirmed the superiority of TICL over standard non-toric ICL in astigmatic correction, albeit across a broad spectrum of preoperative astigmatism (0–3.25 D).12 However, managing astigmatism within the 0.75–1.0 D range poses a persistent clinical challenge. Although TICL provides effective optical correction for astigmatism,13 its widespread adoption is constrained by practical considerations. First, concerns regarding postoperative rotational stability may incline surgeons toward standard ICL for low astigmatism,14,15 given that a 1° misalignment alone diminishes corrective efficacy by approximately 3.3%.16,17 On the other hand, patient-reported symptoms exhibit considerable variability within this range—ranging from significant visual disturbances to an absence of symptoms—complicating uniform treatment decisions. Furthermore, the premium associated with the toric design prompts questions regarding its cost-effectiveness among patients. Nevertheless, direct comparative evidence specific to the ≤1.0 D subgroup remains scarce. The clinical threshold at which the benefits of TICL clearly outweigh its potential drawbacks and costs is inadequately defined. Accordingly, this study aimed to delineate the efficacy, safety, and astigmatic correction outcomes of TICL vs standard ICL in patients with low myopic astigmatism (0.75–1.0 D). We sought to provide robust evidence to guide surgical decision-making in this prevalent yet less well-characterized population. Materials and Methods Participants This single-center, retrospective, non-consecutive, and non-randomized study was performed in line with the principles of the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of Dongguan Guangming Ophthalmic Hospital (approval number: 202510-1). Patients who underwent bilateral EVO Visian TICL or ICL (V4c, STAAR Surgical, Monrovia, CA, USA) implantation at Dongguan Guangming Ophthalmic Hospital from April 2021 to December 2024 were reviewed. The inclusion criteria of patients were: (1) bilateral TICL or ICL implantation; (2) age from 21 to 45 years old; (3) stable refractive status [≤0.5 D progression of spherical equivalent refraction (SEQ) per year]. The inclusion criteria of research eyes were: (1) 1.0 D or 0.75 D myopic astigmatism preoperatively; (2) preoperative corrected distance visual acuity (CDVA) of 20/25 or better; (3) dominant eye if both eyes meet the previous criteria (the dominant eye was determined using the Dolman method).18 The exclusion criteria were: (1) corneal endothelial cell density (ECD) <2000 cells/mm2, or anterior chamber depth (ACD) <2.8 mm; (2) presence of any ocular inflammation, corneal diseases, cataract, glaucoma, fundus diseases, or any other ophthalmic diseases; (3) prior ocular surgery; (4) systemic diseases that may involve the eyes; (5) pregnant females. Patients initially assessed their CDVA with and without astigmatism correction. TICL was preferentially selected for eyes demonstrating superior CDVA with astigmatic correction. In cases where both lens types yielded comparable visual outcomes, the final decision was deferred to patient preference regarding cost and risk tolerance. Sample Size This study was designed as a retrospective comparative study to evaluate surgical outcomes for low astigmatism. The primary outcome measure was the magnitude of prediction error (PE) for residual astigmatism, defined as the vector difference between the actual and predicted residual astigmatism, as described by Koch et al.19 Based on preliminary data from the initial 20 eyes in each group, the mean difference in PE magnitude was estimated at 0.193 D, with a pooled standard deviation of 0.34 D (Cohen’s d= 0.57). Sample size estimation was performed using G*Power (version 3.1.9.7). Considering the non-normal distribution typical of refractive error data, the Mann–Whitney U-test was designated as the primary statistical test. Setting α = 0.05 and power (1 − β) = 0.80, the calculation indicated that 106 eyes (53 per group) were required to detect the specified difference. Consequently, we retrospectively identified and enrolled eligible participants from our electronic medical record system. Examinations All participants underwent comprehensive baseline evaluations and were followed up at 3 and 12 months postoperatively. Uncorrected (UDVA) and corrected (CDVA) distance visual acuities were measured using standard Snellen charts. Manifest refraction was determined using an autorefractor (AOS-1500, Nidek). Biometric measurements were obtained using multiple devices: axial length was measured via optical biometry (IOLMaster 700, Carl Zeiss Meditec AG); anterior chamber depth (ACD), anterior chamber angle (ACA), and lens vault were assessed using the Pentacam Scheimpflug system (Oculus Optikgeräte GmbH); and endothelial cell density (ECD) was analyzed using a specular microscope (CEM-530, Nidek). Additionally, corneal parameters—including keratometry, white-to-white distance (WTW), and central corneal thickness—were acquired from the Pentacam database. Optical quality metrics, including higher-order aberrations (HOAs), modulation transfer function (MTF), and Strehl ratio (SR), were evaluated using an iTrace Prime aberrometer (Tracey Technologies) under a 3-mm pupil diameter. Finally, intraocular pressure (IOP) was measured using a non-contact tonometer (NT-530P, Nidek). All preoperative assessments included slit-lamp biomicroscopy and fundus examination. Surgical Procedure The dioptric power and planar size of both TICL and ICL were determined using the STAAR Surgical Online Calculator (https://evo-ocos.staarag.cn/), with lens sizing based on WTW and ACD measurements. All procedures were uneventfully performed by a single experienced surgeon (C.X.). Prophylactic topical antibiotics (0.5% levofloxacin) were instilled four times daily for three days preoperatively. Preoperative peripheral laser iridotomy was omitted, given the presence of a central perforation in the EVO V4c platform. On the day of surgery, toric axis orientation was marked at the limbus under a slit lamp. Following topical anesthesia (0.5% proparacaine hydrochloride), ICL implantation was performed via a 3.0-mm temporal clear corneal incision, adopting the one-step technique described by Wei et al.20 A cohesive viscoelastic agent (1.7% sodium hyaluronate) was injected to coat the corneal endothelium. The ICL was then advanced into the ciliary sulcus; for TICL, the lens was subsequently rotated to align the toric axis markers with the limbal references. Residual viscoelastic material was thoroughly irrigated with balanced salt solution (BSS). Postoperatively, the same antibiotic regimen was continued four times daily for two weeks. A tapering anti-inflammatory protocol was adopted: 1% prednisolone acetate four times daily for three days, followed by 0.1% bromfenac sodium twice daily for three weeks. Additionally, preservative-free artificial tears (0.3% sodium hyaluronate) were prescribed four times daily, and 0.3% gatifloxacin ointment was applied nocturnally for one month. Astigmatism Analysis In accordance with the guidelines of the Journal of Refractive Surgery and the Journal of Cataract and Refractive Surgery, vector analysis was performed using double-angle plotting techniques.19,21–24 All refractions were converted to the corneal plane to ensure consistency. The back vertex distance (BVD) for manifest refraction was set at 12 mm, while for toric lens power transformation, the BVD was adjusted to the preoperative ACD to enhance precision, particularly given the low magnitude of astigmatism. The predicted refractions were generated using the EVO Toric Calculator (version 5.01). All refractive cylinders were transposed to the positive cylinder format, specifying both the magnitude and the axis of the steep meridian. Target-induced astigmatism (TIA) was defined as the vector representing the intended astigmatic correction. For eyes receiving non-toric ICLs, TIA was set to zero. For TICL eyes, TIA corresponded to the toric power and axis specified in the surgical plan. Vector calculations for PE were executed following double-angle transformation. Double-angle polar plots with convex polygons were generated using the wrap-up functions of the Wilcox-Holladay-Wang-Koch (WHWK) Statistical Software implemented in R (version 4.4.1).25 Alpins vector analyses were conducted using the ASSORT Group Analysis Calculator (version 1.1.4, American Academy of Ophthalmology).23,26 Statistical Analyses To account for inter-eye correlation and satisfy the assumption of independence, only data from the dominant eye were included in the analysis, consistent with published recommendations for ophthalmic outcomes.27,28 Snellen visual acuity was converted to the logMAR scale for statistical testing. Data distribution was assessed using histograms and the Shapiro–Wilk test. Depending on the distribution, continuous variables were compared using Student’s t-test (between groups) or the Mann–Whitney U-test for non-normally distributed data; within-group comparisons were performed using paired t-tests, or the Friedman test, as appropriate. Post Hoc analysis using Wilcoxon Signed-Rank Test with Bonferroni correction was adopted when applicable. Categorical variables are presented as frequencies (percentages) and compared using the Chi-square test or Fisher’s exact test when expected cell counts were low. Correlation analysis was conducted via simple linear regression. To mitigate confounding from non-randomized allocation and baseline disparities in preoperative cylinder power, multiple linear regression and analysis of covariance (ANCOVA) models were constructed using the general linear model (GLM) framework. Key outcomes—including postoperative UDVA, PE, and residual astigmatism—were adjusted for baseline covariates. Adjusted marginal means and pairwise comparisons with Bonferroni correction are reported. Statistical significance was defined as a two-tailed p< 0.05, with p< 0.0167 retained for post hoc analyses to account for multiple comparisons. Data were managed using Microsoft Excel (Version 2411, Microsoft Office 2019), and all statistical analyses were performed using IBM SPSS Statistics (Version 26.0, IBM Corp.). Graphical representations of visual outcomes were generated using GraphPad Prism (Version 9.0, GraphPad Software). Results A total of 124 eyes from 124 patients were enrolled in this retrospective study, comprising 68 eyes in the TICL group and 56 eyes in the ICL group. All participants completed the scheduled 12-month follow-up. Baseline demographic and ocular characteristics are summarized in Table 1. Postoperative complications were generally mild and self-limiting. These included dry eye (8 eyes), night glare/halo (5 eyes), ocular pruritus (3 eyes), transient asthenopia (3 eyes), conjunctival hyperemia (2 eyes), photopsia (1 eye), and ICL surface pigmentation (1 eye). Importantly, no intraoperative complications or severe postoperative adverse events—such as infectious keratitis, cataract progression, or glaucoma—occurred in either group throughout the study period. Table 1Baseline Demographics and Preoperative Ocular CharacteristicsCharacteristicsTICL (n = 68 Eyes)ICL (n = 56 Eyes)p-valueMean ± SDRangeMean ± SDRangeAge (years)29.1 ± 5.821, 4028.6 ± 4.122, 351.000Gender (male/femal)7/26 (21/79)6/28 (18/82)0.765Eyes (right/left)16/17 (48/52)16/18 (47/53)1.000WTR/ATR/OBL53/7/8 (78/10/12)49/5/2 (87/9/4)0.238aRefraction sphere (D)−5.94 ± 2.24−11.00, −2.75−6.73 ± 1.77−10.50, −3.500.119Refraction cylinder (D)−0.95 ± 0.11−0.75, −1.00−0.84 ± 0.12−0.75, −1.000.001Spherical equivalent (D)−6.41 ± 2.24−11.5, −3.25−7.16 ± 1.79−10.88, −3.880.147CDVA (LogMAR)0.00 ± 0.03−0.08, 0.100.00 ± 0.03−0.08, 0.101.000IOP (mmHg)13.8 ± 2.210.0, 18.014.0 ± 2.89.7, 24.00.968ECD (cell/mm2)2811 ±2392232, 33142808 ± 1992366, 31670.959ACD (mm)3.11 ± 0.212.82, 3.523.17 ± 0.262.82, 3.760.523ACA (°)41.4 ± 4.635.5, 51.740.1 ± 5.630.8, 57.60.304Axial length (mm)25.96 ± 1.1424.34, 28.6126.13 ± 1.2824.31, 29.030.697Lens size (mm)12.7 ± 0.412.1, 13.212.7 ± 0.312.1, 13.20.674Notes: aFisher’s exact test.Abbreviations: ACA, anterior chamber angle; ACD, anterior chamber depth; ATR, against-the-rule astigmatism; CDVA, corrected distance visual acuity; ECD, endothelial cell density; ICL, implantable collamer lens; IOP, intraocular pressure; OBL, oblique astigmatism; TICL, toric implantable collamer lens; WTR, with-the-rule astigmatism. Baseline Demographics and Preoperative Ocular Characteristics Notes: aFisher’s exact test. Abbreviations: ACA, anterior chamber angle; ACD, anterior chamber depth; ATR, against-the-rule astigmatism; CDVA, corrected distance visual acuity; ECD, endothelial cell density; ICL, implantable collamer lens; IOP, intraocular pressure; OBL, oblique astigmatism; TICL, toric implantable collamer lens; WTR, with-the-rule astigmatism. Safety and Efficacy At 3 and 12 months postoperatively, the TICL group exhibited higher safety indices (1.24 ± 0.16 and 1.28 ± 0.18, respectively) compared to the ICL group (1.21 ± 0.16 and 1.19 ± 0.16). Similarly, efficacy indices favored the TICL group at both time points (1.20 ± 0.16 and 1.21 ± 0.19 vs 1.15 ± 0.16 and 1.09 ± 0.16, respectively). Notably, both safety and efficacy outcomes were significantly superior in the TICL group at the 12-month mark (p < 0.01). Correspondingly, mean UDVA at final follow-up was significantly better in the TICL group (−0.08 ± 0.07 logMAR) than in the ICL group (−0.03 ± 0.06 logMAR; p < 0.001). Furthermore, a greater proportion of eyes in the TICL group achieved UDVA of 20/20 or better and exceeded their preoperative CDVA thresholds (Figure 1A and B). Importantly, CDVA remained stable in both groups throughout the 12-month observation period, with no instances of vision loss detected (Figure 1C). The standard graphs for reporting refractive outcomes of TICL and ICL implantation for low myopic astigmatism. (A) Cumulative percentage of eyes at different Snellen lines of preoperative CDVA and postoperative UDVA for both groups. (B) Difference in Snellen lines between postoperative UDVA and preoperative CDVA for both groups. (C) Changes in Snellen lines of postoperative CDVA and preoperative CDVA for both groups. (D) Scatterplot of attempted seq refraction and achieved seq refraction for both groups. (E) Accuracy of seq to the intended target for both groups. (F) Stability of seq at the 12-month interval for both groups. (G) Refractive astigmatism before and after surgery for both groups. (H) Scatterplot of surgically induced astigmatism and target induced astigmatism for the TICL group. (I) Refractive astigmatism angle of error for the TICL group. Abbreviations: CDVA, corrected distance visual acuity; ICL, implantable collamer lens; seq, spherical equivalent; TICL, toric implantable collamer lens; UDVA, uncorrected distance visual acuity. Predictability Both groups exhibited robust linear associations between attempted and achieved spherical equivalent (SEQ) at the 3- and 12-month follow-ups (both p < 0.001). At 12 months, the predictability of SEQ correction was exceptionally high in both cohorts, with coefficients of determination (R2) of 0.9799 and 0.9839 in the TICL and ICL groups, respectively (Figure 1D). Regarding absolute accuracy, 96% (65/68) of TICL eyes and 95% (53/56) of ICL eyes achieved a postoperative SEQ within ±0.50 D of the intended target (Figure 1E). Stability Both groups demonstrated excellent postoperative refractive stability throughout the follow-up period. At 3 months, the mean SEQ was 0.10 ± 0.19 D in the TICL group and 0.02 ± 0.21 D in the ICL group (p = 0.031). By 12 months, these values had stabilized at 0.03 ± 0.32 D and −0.09 ± 0.27 D, respectively, maintaining a statistically significant intergroup difference (p = 0.029; Figure 1F). Postoperative Astigmatism Analysis At both 3 and 12 months postoperatively, the TICL group exhibited significantly lower mean residual astigmatism compared to the ICL group (−0.03 ± 0.12D vs −0.17 ± 0.26D at 3 months; −0.16 ± 0.28D vs −0.33 ± 0.40D at 12 months; both p < 0.05). At the final follow-up, a greater proportion of eyes in the TICL group achieved residual astigmatism within 0.25 D (76% [52/68] vs 57% [32/56]; Figure 1G). Vector analysis further corroborated the superior astigmatic correction accuracy of TICL. The mean PE magnitude was 0.083 D in the TICL group, substantially lower than the 0.276 D observed in the ICL group. The distribution of PE, represented by mean and median convex polygons, is depicted in Figure 2. Notably, a significant correlation between TIA and SIA was observed exclusively in the TICL group (R2 = 0.1528, p = 0.001; Figure 1H). At 12 months, 74% (50/68) of TICL eyes exhibited an angle of error (AE) within 5°, whereas only 4% (3/68) demonstrated an AE exceeding 35° (Figure 1I). Furthermore, the Alpins vector analysis confirmed a favorable alignment between TIA and SIA, with arithmetic mean values of 0.84 D and 0.70 D, respectively (Figure 3). The mean difference vector was 0.19 D × 178°, with a corresponding mean correction index of 0.83. Double-angle plots of PE with mean (top) and median (bottom) convex polygons (each ring = 1.0 D) for the TICL and ICL groups at 12 months postoperatively. Abbreviations: PE, prediction error; ICL, implantable collamer lens; TICL, toric implantable collamer lens. Astigmatism vector analysis of TICL eyes based on Alpins method. The double-angle vector diagrams demonstrate target induced astigmatism, surgically induced astigmatism, difference vector, and correction index at 12 months postoperatively. The vector means are plotted as a red diamond and the 95% confidence ellipses of the dataset are drawn as a red circle. Abbreviation: TICL, toric implantable collamer lens. Ocular Biological Parameters As detailed in Table 2, postoperative values for IOP, ECD, and ACA remained stable relative to baseline in both groups, with no statistically significant intergroup differences observed throughout the follow-up period (all p > 0.05). The mean percentage change in ECD at 12 months was −0.88% ± 6.85% in the TICL group and −0.03% ± 6.29% in the ICL group (p > 0.05). In contrast, while vault height decreased significantly over time within both groups, it was consistently and significantly lower in the ICL group compared to the TICL group at both 3 and 12 months postoperatively (p < 0.05). Table 2Postoperative Ocular Parameters at 12 monthsTICLICLp-valuePostop 3m IOP (mmHg)13.7 ± 2.614.0 ± 2.40.479 ECD (cell/mm2)2845 ± 2902800 ± 2360.354 ECD loss (%)−0.11 ± 7.050.24 ± 5.890.853 ACA (°)23.2 ± 4.1a24.1 ± 4.2a0.160 Vault height (μm)497 ± 204417 ± 1790.023Postop 12m IOP (mmHg)14.0 ± 2.414.8 ± 2.6b0.075 ECD (cell/mm2)2868 ± 3012807 ± 2360.105 ECD loss (%)−0.88 ± 6.85−0.03 ± 6.290.363 ACA (°)23.8 ± 3.9a24.3 ± 4.1a0.510 Vault height (μm)433 ± 184c366 ± 161c0.021Notes: aP < 0.0167, vs preoperative values (Friedman test and Post Hoc analysis, with Bonferroni correction). bP < 0.0167, vs 3-month values (ANOVA test and Post Hoc analysis, with Bonferroni correction). cP < 0.05, vs 3-month values (paired samples t-test).Abbreviations: ACA, anterior chamber angle; ECD, endothelial cell density; ICL, implantable collamer lens; IOP, intraocular pressure; TICL, toric implantable collamer lens. Postoperative Ocular Parameters at 12 months Notes: aP < 0.0167, vs preoperative values (Friedman test and Post Hoc analysis, with Bonferroni correction). bP < 0.0167, vs 3-month values (ANOVA test and Post Hoc analysis, with Bonferroni correction). cP < 0.05, vs 3-month values (paired samples t-test). Abbreviations: ACA, anterior chamber angle; ECD, endothelial cell density; ICL, implantable collamer lens; IOP, intraocular pressure; TICL, toric implantable collamer lens. Objective Visual Quality Both groups demonstrated significant postoperative improvements in HOAs, coma, MTF, and SR compared to baseline (all p < 0.0167; Table 3). Trefoil was also significantly reduced within the ICL group (p < 0.0167). Crucially, all visual quality metrics remained stable across the 12-month follow-up period in both cohorts, with no statistically significant intergroup differences observed for any parameter (p > 0.05). Table 3Comparison of Visual Quality ParametersTICLICLp-valuePostop 3m HOAs0.108 ± 0.050a0.099 ± 0.038a0.407 Spherical0.004 ± 0.0270.004 ± 0.0240.717 Coma0.059 ± 0.044a0.053 ± 0.031a0.835 Trifoil0.058 ± 0.0380.054 ± 0.029a0.910 MTF0.231 ± 0.102a0.226 ± 0.104a0.857 SR0.06289 ± 0.05954a0.06298 ± 0.06028a0.940Postop 12m HOAs0.115 ± 0.055a0.111 ± 0.108a0.109 Spherical0.014 ± 0.0250.003 ± 0.0340.174 Coma0.070 ± 0.0470.064 ± 0.095a0.084 Trifoil0.068 ± 0.0680.060 ± 0.041a0.694 MTF0.227 ± 0.101a0.197 ± 0.096a0.068 SR0.06327 ± 0.07585a0.06055 ± 0.11655a0.099Notes: aP < 0.0167, vs preoperative values (Friedman test and Post Hoc analysis, with Bonferroni correction).Abbreviations: ICL, implantable collamer lens; MTF, modulation transfer functions; TICL, toric implantable collamer lens; HOAs, high-order aberrations; SR, Strehl ratio. Comparison of Visual Quality Parameters Notes: aP < 0.0167, vs preoperative values (Friedman test and Post Hoc analysis, with Bonferroni correction). Abbreviations: ICL, implantable collamer lens; MTF, modulation transfer functions; TICL, toric implantable collamer lens; HOAs, high-order aberrations; SR, Strehl ratio. Discussion While the efficacy of TICL for correcting low astigmatism is well established, prior evidence predominantly pertains to astigmatism exceeding 1.0 D, demonstrating its superiority over non-toric ICL.12,13 Direct comparative evidence for managing astigmatism ≤1.0 D remains scarce. Addressing this gap, the present study specifically evaluated TICL and ICL outcomes in eyes with 0.75 D or 1.0 D of astigmatism. Consistent with existing literature, both TICL and ICL implantation yielded satisfactory safety and efficacy profiles in this cohort.29,30 Postoperative SEQ predictability and stability were comparable between groups. Notably, however, the TICL group achieved significantly superior UDVA and CDVA (p < 0.05). To account for the non-randomized design and the resultant baseline imbalance in preoperative cylinder, we employed ANCOVA with preoperative cylinder as a covariate. This analysis confirmed that preoperative cylinder was not a significant predictor of postoperative UDVA (p = 0.136), PE (p = 0.880), or residual astigmatism (p = 0.606). After adjusting for this baseline confounder, the estimated mean difference in postoperative UDVA remained robust at −0.050 logMAR (95% CI: −0.074 to −0.026, p< 0.001; Table 4), mirroring the unadjusted analysis. Table 4Estimated Marginal Means and Group Comparison Regarding Postoperative OutcomesParametersEMMsEMM Difference95% CIp-valueTICLICLPostoperative UDVA (logMAR)−0.079−0.029−0.050−0.074 to −0.026<0.001PEa (D)0.2230.324−0.101−0.209 to 0.0060.065Residual astigmatismb (D)−0.153−0.3320.1790.050 to 0.3070.007Notes: arefers the magnitude of PE. bnegative cylinder format is used for preoperative and residual astigmatism.Abbreviations: CI, confidence interval; EMM, estimated marginal mean; ICL, implantable collamer lens; PE, predicted error; TICL, toric implantable collamer lens; UDVA, uncorrected distance visual acuity. Estimated Marginal Means and Group Comparison Regarding Postoperative Outcomes Notes: arefers the magnitude of PE. bnegative cylinder format is used for preoperative and residual astigmatism. Abbreviations: CI, confidence interval; EMM, estimated marginal mean; ICL, implantable collamer lens; PE, predicted error; TICL, toric implantable collamer lens; UDVA, uncorrected distance visual acuity. Consequently, we attribute the superior visual acuity in the TICL cohort primarily to its effective astigmatic correction rather than baseline disparities. First, the mean postoperative cylinder was approximately 0.17 D lower in the TICL group than in the ICL group (p = 0.028). After adjusting for baseline cylinder, this mean difference widened slightly to 0.18 D (95% CI: 0.05 to 0.31, p = 0.007; Table 4). Vector analysis further corroborated the enhanced accuracy of TICL. Although the overall magnitude of PE did not differ significantly between groups before or after adjustment (both p > 0.05; Table 4), the 95% CI for the adjusted mean difference (−0.209 D to 0.006 D) approached significance, suggesting a trend toward superior predictive accuracy with TICL. Particularly, decomposition of the PE vector revealed that the mean horizontal component was significantly lower in the TICL group (p < 0.001), whereas the vertical component showed no intergroup difference (p = 0.392). This anisotropy likely reflects the axis alignment strategy for young adults with with-the-rule (WTR) astigmatism, wherein TICL axes are predominantly placed horizontally. This interpretation is supported by the double-angle plots, which localized the vector means of both TIA and SIA within the WTR sector (Figure 3). Given the uniform temporal incision site across all participants, the visual superiority of TICL cannot be ascribed to surgical technique; rather, it substantiates the optical benefit of mitigating lens-induced astigmatism through toric correction. From an optical standpoint, the TICL incorporates a cylindrical correction aligned to the corneal principal meridians. This design effectively neutralizes astigmatic focal lines, minimizing the blur circle diameter on the retina and thereby enhancing image sharpness. In contrast, SEQ compensation merely superimposes the circle of least confusion of the conoid of Sturm onto the retinal plane. While this may yield marginal improvements in visual acuity, it fails to resolve the fundamental astigmatic blur. Supporting this principle, several studies have identified a significant correlation between uncorrected astigmatism and elevated higher-order aberrations (HOAs), particularly under mesopic conditions and in astigmatism exceeding 2.0 D.31–33 Notably, Mohammadpour et al found no correlation between residual astigmatism and HOAs after refractive surgery.33 In the present study, no significant intergroup differences in HOAs or other objective visual quality metrics were observed between TICL and ICL cohorts. These findings align with prior reports demonstrating comparable postoperative improvements in visual quality following both procedures.9,12,34 Furthermore, we observed significant reductions in HOAs, coma, and trefoil at 12 months postoperatively compared to baseline. This contrasts with Wei et al, who reported increases in total HOAs and trefoil following ICL implantation.12 This discrepancy may be attributed to methodological variations, specifically the scotopic measurement pupil diameter: our assessments utilized a 3-mm aperture, whereas Wei et al employed a 5-mm aperture, where optical aberrations are inherently more pronounced. Guo and Atchison demonstrated that an induced cylinder of merely 0.33 D can degrade visual acuity by approximately 0.0 logMAR lines.35 In our cohort, the TICL group achieved a mean UDVA slightly better than the 0.0 logMAR threshold, accompanied by significantly lower residual astigmatism (−0.16 D) compared to the ICL group (−0.33 D). This intergroup difference closely approximates the 0.33 D threshold proposed by Guo and Atchison, providing a potential optical explanation for the visual superiority observed with TICL. Furthermore, they posited that with-the-rule (WTR) astigmatism may be more tolerable neurologically than optically. Neural adaptation, characterized by a positive coupling between higher-order aberrations and astigmatism, potentially mitigates the subjective impact of low astigmatism (0.3–0.5 D) on image quality.4 Given that astigmatism <0.5 D exerts only a marginal effect on visual performance, a threshold of ≥0.75 D is widely adopted in modern cataract surgery to justify toric IOL implantation.4,36 Collectively, our findings substantiate that even for low astigmatism as mild as 0.75 D, precise toric correction yields superior visual and refractive outcomes compared to spherical equivalent compensation with standard ICLs. The AE serves as a sensitive proxy for postoperative rotational stability. In this cohort, the TICL demonstrated exceptional axial stability, with 74% (50/68) of eyes exhibiting an AE within 5° at 12 months, while merely 4% (3/68) presented with an AE exceeding 35° (Figure 1I). This narrow distribution underscores the accurate alignment between the TIA and SIA. Our findings are consistent with the established long-term stability of the EVO TICL platform.14,15,37 For instance, Lee et al reported a mean absolute rotation of 3.87 ± 3.07°at 6 months,37 and Chen et al documented a comparable 6.00 ± 6.70° at 3 years.15 Importantly, Chen et al emphasized that such minor rotation in TICL with low cylinder power may not result in clinically significant astigmatism.15 This phenomenon is particularly relevant to our study population (0.75–1.0 D astigmatism), suggesting that even subtle misalignments are unlikely to compromise the clinical outcome. Both groups maintained mean vault heights within the clinically safe range. Consistent with prior reports,38–40 the TICL group exhibited a significantly higher vault compared to the ICL group. For instance, Zhao et al documented mean vaults of 554.11±219.36µm and 449.70±172.47µm in TICL and ICL eyes, respectively.39 In a paired-eye study, the ICL size and spherical power of ICL were mostly associated with vault.38 However, these parameters were compared in this study. The vault difference is likely attributable to the increased lens thickness inherent to the toric design, wherein the incorporation of the cylindrical component necessitates a thicker optic zone.39,40 Consequently, this dimensional increase may confer a mechanical advantage by providing stronger resistance to iris apposition or dynamic touch.38 Other potential confounders such as age, ACD, and ACA were comparable between groups, reinforcing that the vault disparity stems primarily from the lens design rather than baseline anatomic variation.39 This study has several limitations that must be acknowledged. First, the sample size was modest, and the non-randomized design resulted in an inherent baseline imbalance in preoperative astigmatism. Future prospective, randomized controlled trials with larger cohorts are warranted to validate these findings and minimize selection bias. Second, although the distribution of astigmatism axes did not differ significantly between groups, the uniform temporal incision may restrict the interpretability of our results for high-volume surgeons employing varied techniques, such as superior or steep-axis incisions. Subsequent studies should consider stratifying patients by axis type or conducting subgroup analyses to isolate the incisional effect. Third, while 12-month data suggest excellent initial stability, the long-term rotational behavior of TICL in the low-astigmatism spectrum requires further surveillance. Finally, the absence of patient-reported outcome measures, such as standardized visual disturbance questionnaires, represents a gap in our assessment of subjective visual quality. Future comparative studies incorporating these metrics are essential to fully elucidate the patient experience with TICL vs ICL for low astigmatism. In conclusion, both TICL and standard ICL implantation demonstrated favorable safety, predictability, and 12-month refractive stability in patients with low myopic astigmatism (0.75–1.0 D), with no severe complications observed. TICL achieved significantly lower residual astigmatism and better uncorrected visual acuity compared with ICL, as supported by vector analysis showing smaller prediction error and close alignment between target and surgically induced astigmatism for TICL. Objective visual quality metrics improved similarly in both groups. Toric correction for low astigmatism offers measurable refractive and visual acuity benefits. However, limited by the nonrandomized design and the selection bias introduced by treatment allocation based on visual simulation and patient preference, these advantages must be interpreted with caution. Although our adjusted analyses reduced confounding from baseline cylinder differences, they cannot fully eliminate this bias. Clinicians should be cautious when considering TICL for patients with low astigmatism, even when preoperative CDVA improves noticeably with cylinder correction. In clinical practice, the decision-making process must integrate these potential benefits with economic and safety considerations. The higher cost of toric implants and the potential risk of postoperative rotation may negate the benefits of astigmatic correction. Reserving TICL for cases where the patient’s visual demands are high and preoperative assessment predicts a suboptimal outcome with a standard ICL may be a more prudent strategy. Future cost-benefit analyses incorporating patient satisfaction scores are warranted to refine these thresholds. Funding Statement The authors did not receive support from any organization for the submitted work. Abbreviations ACA, anterior chamber angle; ACD, anterior chamber depth; AE, angle of error; CDVA, corrected distance visual acuity; CI, confidence interval; ECD, endothelial cell density; HOA, higher-order aberrations; ICL, implantable collamer lens; IOP, intraocular pressure; LASIK, laser in-situ keratomileusis; MTF, modulation transfer functions; PE, prediction error; SEQ, spherical equivalent; SIA, surgically induced astigmatism; SMILE, small incision lenticule extraction; SR, Strehl ratio; TIA, target induced astigmatism; TICL, toric implantable collamer lens; UDVA, uncorrected distance visual acuity; WTR, with-the-rule astigmatism; WTW, white-to-white. Data Sharing Statement All data generated or analyzed during this study are included in this published article. Ethics Approval and Consent to Participate This study was performed in line with the principles of the Declaration of Helsinki. Review and approval were granted by the Ethics Committee of Dongguan Guangming Ophthalmic Hospital (approval number: 202510-1). Surgery informed consent was provided to all subjects. All the data were anonymous, and research informed consent was therefore waived. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors have no competing interests to declare that are relevant to the content of this article. A previous relevant preprint of this manuscript is available on ResearchSquare https://www.researchsquare.com/article/rs-6897555/v1. References Associated Data Data Availability Statement All data generated or analyzed during this study are included in this published article.