Investigating the impact of residual refraction within ±1.0 dioptre on uncorrected distance visual acuity in pseudophakic eyes: a cross-sectional study

Abstract

Objectives

To investigate the impact of residual refraction within ±1.0 dioptre (D) on uncorrected distance visual acuity (UDVA) in pseudophakic eyes.

Design

Cross-sectional study.

Setting

This study was based on retrospectively collected electronic refraction records from a tertiary care academic ophthalmology centre in southern China between May 2022 and July 2025.

Participants

Patients aged ≥40 years who underwent uneventful phacoemulsification cataract surgery with in-the-bag monofocal intraocular lens implantation and achieved a postoperative corrected distance visual acuity (CDVA) of ≤0.1 logarithm of the minimum angle of resolution were enrolled. They were stratified by astigmatism subtypes: minimal astigmatism (<0.50 D), with-the-rule (WTR) astigmatism, against-the-rule (ATR) astigmatism and oblique astigmatism.

Outcomes measures

Postoperative evaluation (≥1 month) included spherical equivalent (SE) refraction, UDVA and CDVA. UDVA was compared across eyes with SE intervals of 0.50 D within ±1.0 D. ORs were calculated to assess the relative risk of failing to achieve a UDVA of 0.1 or better for postoperative SE within ±1.0 D, using 0.00 D as the reference.

Results

The study included 1333 eyes from 1333 patients (mean (SD) age, 66.1 (8.96) years; 532 male (39.9%)). Overall, and particularly in the minimal astigmatism (<0.50 D), ATR astigmatism and oblique astigmatism subgroups, hyperopic eyes exhibited significantly better UDVA than their myopic counterparts. Slight myopia [−0.50 D, 0 D) significantly worsened UDVA versus 0 D in both the overall population and the minimal astigmatism subgroup. Slight hyperopia (0 D, +0.50 D] minimally affected UDVA, whereas an equivalent degree of myopia increased the odds of not achieving UDVA ≤0.1 by 1.55-fold (95% CI 1.08 to 2.21) overall and by 3.14-fold (95% CI 1.49 to 6.58) in the minimal astigmatism subgroup. Additionally, UDVA was optimal with minimal astigmatism and decreased progressively with each 0.50 D increment in residual astigmatism magnitude, a dose-dependent trend consistent across astigmatism subtypes.

Conclusions

The impact of refractive errors (≤1.0 D) on UDVA was associated with the magnitude and type of astigmatism. Residual astigmatism of ≥0.50 D exerted a significant negative effect on UDVA. A plano SE (0 D) was optimal for minimum and WTR astigmatism, whereas slight hyperopia yielded superior UDVA in ATR and oblique astigmatism.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Comprehensive data collection and analysis provides quantitative evidence for the synergistic effect of spherical error together with astigmatic magnitude and meridian orientation on uncorrected distance visual acuity (UDVA).

• Real-world observational design reflected routine clinical practice in postoperative refractive outcomes following monofocal intraocular lens implantation, especially with regard to optimising refractive targets to enhance UDVA for patients with different astigmatism subtypes.

• The study was conducted at a single centre and did not include assessments of near vision and patient satisfaction.

Introduction

Monofocal intraocular lens (IOL) remains the most commonly implanted type, owing to their cost-effectiveness, low risk of photic phenomena (eg, halo and glare), and superior contrast sensitivity compared with multifocal designs.^{1 2} Emmetropia is traditionally considered the optimal target for patients prioritising distance vision after monofocal IOL implantation. However, the manufacturing standard for most IOLs in 0.50 dioptre (D) increments makes targeting exact zero refraction commonly unattainable. Therefore, surgeons are often confronted with the clinical choice of targeting either slight myopia or slight hyperopia.

Advances in cataract surgical techniques, IOL technology, preoperative diagnostics, IOL power calculations and postcataract astigmatism prediction technology have narrowed the gap between postoperative refractive outcomes and the intended targets.³,⁶ Nevertheless, studies show that 50%–70% of patients achieve postoperative refractions within 0.50 D of the target, and 79%–94% fall within 1.0 D.⁷,⁹ While cataract surgery outcomes are typically deemed successful when postoperative refraction falls within ±0.50 D or ±1.0 D,¹⁰ even small refractive errors can significantly impact patient satisfaction as expectations for visual quality continue to rise.^{11 12} However, the impact of these small refractive errors on uncorrected distance visual acuity (UDVA) has not been fully elucidated yet.

Residual astigmatism is a common postoperative issue, with preoperative corneal astigmatism being a primary contributor.^{6 13 14} While toric IOLs and limbal relaxing incisions can partially address corneal astigmatism, factors such as cost, the risk of IOL rotation and residual astigmatism limit their universal applicability.^{15 16} IOL power calculations are typically based on spherical equivalent (SE), but both the magnitude and orientation of residual astigmatism significantly influence postoperative UDVA and patient satisfaction.^{1112 17},¹⁹ Previous studies have explored the isolated effects of residual astigmatism and spherical error on UDVA.^{11 12} However, the synergistic impact of spherical error together with astigmatic magnitude and meridian orientation has yet to be comprehensively investigated.

This study aims to investigate the impact of small postoperative refractive errors (≤1.0 D) on UDVA in a large sample of patients with monofocal IOLs. By analysing the interplay between postoperative SE and astigmatism types, we seek to provide evidence-based recommendations for optimising refractive targets to enhance UDVA.

Methods

This was a hospital-based cross-sectional study conducted at the Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China. We retrospectively reviewed all electronic refraction records from May 2022 to July 2025. Patients meeting the predefined inclusion and exclusion criteria were subsequently included as study participants.

Eligibility criteria

Inclusion criteria were as follows: (1) Patients aged 40 years or older; (2) Uneventful phacoemulsification with in-the-bag monofocal IOL implantation; (3) Postoperative subjective refraction performed at least 30 days after surgery, with a corrected distance visual acuity (CDVA) of 0.1 logarithm of the minimum angle of resolution (logMAR) or better;²⁰ and (4) Postoperative SE between −1.0 D and +1.0 D. Patients who underwent combined surgical procedures during phacoemulsification (eg, capsular tension ring implantation, iridotomy) or had any other vision-impairing ocular pathology (eg, corneal leucoma, retinal detachment, macular degeneration, optic neuritis or abnormal pupil size) were excluded. If both eyes of one patient met the inclusion criteria, one eye was randomly selected for analysis to ensure data independence.²⁰ This selection was performed retrospectively using a computer-generated simple randomisation procedure (Microsoft Excel, function). Specifically, a random number between 0 and 1 was generated for each patient; the right eye was selected if the number was ≤0.5, and the left eye was selected otherwise. This process was conducted solely for analytical data set creation and did not influence clinical procedures.

Measurements and surgical technique

All patients underwent a standardised preoperative ocular examination, including slit-lamp examination, intraocular pressure measurement and ocular biometry. All surgeries were performed under topical anaesthesia (0.5% proparacaine hydrochloride, Alcon laboratories) using a standardised phacoemulsification technique (without femtosecond laser assistance). A 2.2–2.8 mm clear corneal main incision was created temporally, followed by a 1.0 mm auxiliary side-port incision at an appropriate position to facilitate instrument manipulation and intraocular pressure stabilisation. Postoperative evaluations included subjective refraction (sphere, cylinder and axis), UDVA and CDVA, all measured by experienced optometrists and recorded in the electronic medical record system. Monocular subjective refraction was performed under non-cycloplegic conditions using the System Chart SC-1600 Pola (SC, Nidek, Gamagori, Japan) with E optotypes. Extracted parameters included demographic data (age and sex), ocular history (previous eye diseases or surgeries), surgical details (type and power of implanted IOL) and postoperative refractive outcomes at least 1 month after surgery. SE was calculated as: SE = sphere + (cylinder/2), where cylinder was the cylindrical power in negative notation. The SE=0 D classification was based on an exact plano subjective refraction, not a rounded value. Visual acuity measurements were converted from decimal notation to logMAR units for analysis using the formula: logMAR = −log₁₀ (decimal visual acuity).

Statistical analysis

Patients were stratified by astigmatism subtype: (1) Minimal astigmatism (< 0.50 D), (2) With-the-rule (WTR) astigmatism (steeper meridian within ±30° of the vertical axis), (3) Against-the-rule (ATR) astigmatism (steeper meridian within ±30° of the horizontal axis) and (4) Oblique astigmatism (all other orientations). In the overall data set and astigmatism subgroups, patients’ eyes were stratified by SE into five categories: [−1.0 D, –0.50 D), [−0.50 D, 0 D), 0 D, (0 D, +0.50 D] and (+0.50 D, +1.0 D].

Statistical analysis was performed using SPSS software (V.26, IBM Corp). One-way analysis of variance (ANOVA) was applied to compare postoperative UDVA across different ranges of SE refraction for both the total data set and the four astigmatism subtypes, and to assess differences among the astigmatism subtypes. Post hoc multiple comparisons were conducted using Bonferroni’s correction to adjust for type I error. A quadratic polynomial trend analysis was conducted within the one-way ANOVA framework using polynomial contrasts. The required sample size was calculated using the F test in one-way ANOVA, based on the following assumptions according to the historical data: a medium effect size of 0.35 and a significance level of 0.0083 (adjusted to 0.05/6 using the Bonferroni correction for multiple post hoc comparisons). To achieve 80% power, a minimum of 140 participants per group was determined. To robustly conduct subgroup analyses for the four residual astigmatism types, this initial estimate was multiplied by approximately 10. G*Power V.3.1.9.2 was used for the calculation.

Univariate analyses were performed to identify potential covariates, and variables with a significance level of p<0.05 were then entered into a multivariable logistic regression model to evaluate the effect of residual SE on postoperative monocular UDVA. Adjusted ORs with 95% CIs were calculated using 0.00 D SE as the reference, assessing the impact of each additional 0.50 D of hyperopic or myopic SE within the range of ±1.00 D on UDVA. A value of p<0.05 was considered statistically significant for all analyses.

Patient and public involvement

None.

Results

A total of 1333 pseudophakic eyes of 1333 patients (39.9% male) were included in the study, with a mean age of 66.1±8.96 years (range 40–90 years) (table 1). Among these eyes, 81.5% (1087/1333) achieved a postoperative SE within ±0.50 D. The mean UDVA in logMAR units was 0.25±0.12 for eyes with myopia [−1.0 D, –0.50 D), 0.18±0.12 for eyes with slight myopia [−0.50 D, 0 D), 0.10±0.11 for eyes with zero residual refraction (0 D), 0.14±0.12 for eyes with slight hyperopia (0 D, +0.50 D] and 0.17±0.12 for eyes with hyperopia (+0.50 D, +1.0 D] (figure 1A, p<0.001). Post hoc analysis revealed that the eyes with zero residual refraction had significantly better UDVA compared with both slightly myopic eyes (p<0.001) and slightly hyperopic eyes (p<0.01). The quadratic trend was statistically significant (p<0.001), indicating a U-shaped pattern across ordered groups. In addition, both slightly hyperopic eyes (0 D, +0.50 D] and moderately hyperopic eyes (+0.50 D, +1.0 D] demonstrated significantly better UDVA than their myopic counterparts (−0.50 D to 0 D and −1.0 D to −0.50 D, respectively; both p<0.001).

Table 1. Baseline characteristics and postoperative refraction outcomes at ≥1 month follow-up.

Parameter	Total	Type of residual refractive astigmatism				P value*
		Minimum (<0.50 D)	WTR	ATR	Oblique
Number of subjects, n (%)	1333	297 (22.3)	157 (11.8)	724 (54.3)	155 (11.6)
Age (years)	66.1±8.96	64.0±8.90	59.9±10.2	68.7±7.63	64.4±9.00	<0.001
Male, n (%)	532 (39.9)	125 (42.1)	56 (35.7)	302 (41.7)	49 (31.6)	0.126
Post-op refractive sphere (D)	0.35±0.52	0.00±0.37	0.36±0.50	0.50±0.51	0.28±0.53	<0.001
Post-op refractive astigmatism (D)	0.82±0.63	0.07±0.11	0.91±0.44	1.12±0.58	0.72±0.31	<0.001
Post-op SE (D)	−0.06±0.42	−0.03±0.37	−0.09±0.47	−0.06±0.40	−0.07±0.49	0.651
Post-op UDVA (logMAR)	0.16±0.12	0.09±0.10	0.16±0.14	0.19±0.12	0.19±0.13	<0.001
Post-op CDVA (logMAR)	0.02±0.04	0.01±0.05	0.02±0.04	0.02±0.04	0.03±0.04	<0.001

Data are expressed as mean±SD or n (%).

Minimum, postoperative astigmatism <0.50 D.

^*P values obtained using one-way analysis of variance (ANOVA) or χ2 test.

ATR, against-the-rule astigmatism; CDVA, corrected distance visual acuity; D, dioptre; logMAR, logarithm of the minimum angle of resolution; Post-op, postoperative; SE, spherical equivalent; UDVA, uncorrected distance visual acuity; WTR, with-the-rule astigmatism.

Figure 1. Comparison of mean postoperative UDVA (logMAR) across refractive groups and astigmatism subtypes. Mean postoperative UDVA (logMAR) for patients stratified by postoperative spherical equivalent (SE): [-1.0 D, -0.50 D), [-0.50 D, 0 D), 0 D, (0 D, +0.50 D], (+0.50 D, +1.0 D] in the overall data set and astigmatism subgroups. Four astigmatism subgroups include minimal astigmatism (<0.50 D), with-the-rule (WTR) astigmatism, against-the-rule (ATR) astigmatism and oblique astigmatism. Statistical significance was assessed using one-way analysis of variance (ANOVA), with post hoc pairwise comparisons adjusted using Bonferroni’s correction. D, dioptre; logMAR, logarithm of the minimum angle of resolution; Minimum, postoperative astigmatism <0.50 D; Post-op SE, postoperative spherical equivalent; UDVA, uncorrected distance visual acuity. Statistical significance: ***p<0.001, **p<0.01, *p<0.05, ns: not significant.

This phenomenon was likewise evident in most astigmatism subtypes. The minimal astigmatism, ATR astigmatism and oblique astigmatism subgroups showed similar postoperative UDVA outcomes: slight hyperopia (0 D, +0.50 D] demonstrated better UDVA than slight myopia [−0.50 D, 0 D) (all p<0.05); moderate hyperopia (+0.50 D, +1.0 D] outperformed moderate myopia [−1.0 D, –0.50 D) (minimum: p<0.001; ATR astigmatism and oblique astigmatism: p<0.05). However, no significant differences were observed in the WTR astigmatism subgroup, or between any hyperopic groups and 0 residual refraction across all subgroups (figure 1B–E).

Compared to 0 D, myopia exerted a more pronounced adverse effect on UDVA in the minimal astigmatism group. This was the only subgroup in which even slight myopia [−0.50 D, 0 D) led to significant visual degradation relative to zero refractive error. Specifically, the mean UDVA values in this group were as follows: 0.23±0.12 for myopia [−1.0 D, –0.50 D), 0.12±0.08 for slight myopia [−0.50 D, 0 D), 0.05±0.09 for eyes with zero residual refraction (0 D), 0.07±0.07 for slight hyperopia (0 D, +0.50 D] and 0.09±0.08 for hyperopia (+0.50 D, +1.0 D] (figure 1B, p<0.001).

Additionally, a trend was observed that zero residual refraction achieved the best UDVA in the minimum astigmatism and WTR astigmatism groups, whereas in the ATR astigmatism and oblique astigmatism groups, slight hyperopia (0 D, +0.50 D] appeared to yield better UDVA than 0 D, though the differences were not statistically significant. The mean magnitude of residual astigmatism (D) did not differ significantly across refractive groups in the astigmatism subgroups (all p>0.05, online supplemental figure 1).

Figure 2 illustrates the impact of magnitude of residual astigmatism on UDVA across astigmatism subtypes in patients with a refractive error of SE within ±0.50 D. Residual astigmatism exerted a clear dose-dependent influence on postoperative UDVA. Eyes with minimal cylinder (<0.50 D) achieved the best mean UDVA (0.08±0.09 logMAR) both in the overall data set and astigmatism subtypes. UDVA declined progressively as residual astigmatism increased: 0.13±0.12 logMAR (0.50 D to <1.0 D), 0.18±0.11 logMAR (1.0 D to <1.5 D) and 0.25±0.10 logMAR (≥1.5 D). Post hoc Bonferroni tests confirmed that every pairwise comparison between the strata was statistically significant (p<0.001). Notably, this stepwise decline in UDVA was consistent when stratified by astigmatic orientation (WTR, ATR and oblique subtypes). Online supplemental figure 2 further illustrates the impact of residual astigmatism type on UDVA in patients with SE within ±0.50 D and residual astigmatism <1.0 D. Mean astigmatism magnitude was matched to eliminate differences among the WTR astigmatism, ATR astigmatism and oblique astigmatism groups. A significant overall difference in mean postoperative UDVA was observed (p<0.001): the minimal astigmatism group (<0.5 D) achieved optimal UDVA (0.08±0.09), outperforming the WTR astigmatism (0.13±0.16), ATR astigmatism (0.13±0.10) and oblique astigmatism (0.16±0.12) groups. However, no statistically significant differences in UDVA were observed between any pairwise comparisons of the WTR astigmatism, ATR astigmatism and oblique astigmatism subtypes (all p>0.05).

Figure 2. Comparison of mean postoperative UDVA (logMAR) across residual astigmatism. The mean postoperative UDVA (logMAR) for patients categorised by the magnitude of residual astigmatism: minimal astigmatism (<0.50 D), [0.50 D, 1.0 D), [1.0 D, 1.5 D), ≥1.5 D in the overall data set and astigmatism subtypes. Astigmatism subtypes include with-the-rule (WTR) astigmatism, against-the-rule (ATR) astigmatism and oblique astigmatism. Statistical significance was assessed using one-way analysis of variance (ANOVA), with post hoc pairwise comparisons adjusted by Bonferroni’s correction. D, dioptre; logMAR, logarithm of the minimum angle of resolution; Minimum, postoperative astigmatism <0.50 D; UDVA, uncorrected distance visual acuity. Statistical significance: ***p<0.001, **p<0.01.

To identify the impact of postoperative SE over UDVA, a multivariate regression model was performed (table 2). An SE of [−0.50 D, 0 D) significantly increased the odds of not achieving a UDVA of 0.1 logMAR or better, with an OR of 1.55 (95% CI 1.08 to 2.21; p<0.05) overall and 3.14 (95% CI 1.49 to 6.58; p<0.01) in the minimal astigmatism subgroup. The impact of myopic SE on UDVA became more pronounced at the −1.00 D to −0.50 D level. For eyes with an SE of [−1.00 D, –0.50 D), the OR increased by a factor of 7.11 (95% CI 4.17 to 12.13; p<0.001) after adjusting for age, sex and residual astigmatism type. In the minimal astigmatism subgroup, the OR increased by a factor of 22.15 (95% CI 7.48 to 65.64; p<0.001) after adjusting for age and sex. Similarly, the OR increased by 4.38 (95% CI 1.18 to 16.24; p<0.05) in the WTR astigmatism group, 4.01 (95% CI 1.86 to 8.64; p<0.001) in the ATR astigmatism group and 14.88 (95% CI 1.65 to 134.23; p<0.05) in the oblique astigmatism group. In contrast, hyperopic SE within 1.00 D had a minimal effect on postoperative UDVA overall and in all astigmatism subtypes. The number of eyes and mean UDVA for each subgroup are detailed in online supplemental table 1. These findings indicated that increasing hyperopic SE had a less pronounced effect on UDVA compared with myopic SE. The summary of the IOL models included in the data set is provided in online supplemental table 2.

Table 2. Multivariate logistic regression for failure to achieve UDVA (logMAR) ≤0.1, stratified by postoperative spherical equivalent (D).

Post-op SE (D)	Total	Type of residual refractive astigmatism
		Minimum (<0.50 D)	WTR	ATR	Oblique
[−1.00, –0.50)	7.11 (4.17 to 12.13)***	22.15 (7.48 to 65.64)***	4.38 (1.18 to 16.24)*	4.01 (1.86 to 8.64)***	14.88 (1.65 to 134.23)*
[−0.50, 0.00)	1.55 (1.08 to 2.21)*	3.14 (1.49 to 6.58)**	1.50 (0.53 to 4.26)	1.20 (0.72 to 2.01)	1.09 (0.39 to 3.03)
0.00	Reference	Reference	Reference	Reference	Reference
(0.00, +0.50]	0.95 (0.65 to 1.38)	1.06 (0.42 to 2.71)	0.92 (0.29 to 2.96)	0.92 (0.54 to 1.54)	0.30 (0.09 to 1.02)
(+0.50, +1.00]	1.43 (0.85 to 2.41)	2.34 (0.66 to 8.32)	0.98 (0.25 to 3.80)	1.20 (0.55 to 2.65)	1.08 (0.31 to 3.78)

Data are expressed as OR (95% CI).

Adjusted for age, sex and type of residual refractive astigmatism in the overall data set; adjusted for age, sex in the residual astigmatism subgroups.

Statistical significance: bold indicates significance at p<0.05; *p<0.05; **p<0.01; ***p<0.001.

Minimum, postoperative astigmatism <0.50 D.

ATR, against-the-rule astigmatism; D, dioptre; logMAR, logarithm of the minimum angle of resolution; post-op, postoperative; SE, spherical equivalent; UDVA, uncorrected distance visual acuity; WTR, with-the-rule astigmatism.

Discussion

In this study, we investigated the impact of postoperative SE on UDVA in patients with or without astigmatism following monofocal IOL implantation. Our findings demonstrated that minimised residual refractions (closer to 0 D) provided optimal UDVA, especially slight hyperopia outperforming slight myopia following cataract surgery without astigmatic correction. Notably, any astigmatism ≥0.50 D should be actively corrected.

Eyes with slight hyperopia demonstrated superior UDVA compared with those with an equivalent degree of myopia. Previous research has indicated that the residual myopic sphere ≤−0.25 D and hyperopic sphere ≥+0.50 D can have a clinically meaningful impact on UDVA.¹² Although a direct comparison between hyperopic and myopic eyes was not performed in their analysis, the trend indicating greater decline of UDVA with myopia paralleled our current observations. This phenomenon is likely attributable to pseudoaccommodative mechanisms. In phakic eyes, the accommodative amplitude declines significantly after age 40 years.²¹ In contrast, for pseudophakic eyes implanted with monofocal IOLs, the reported accommodative amplitude ranges from 0.50 D to 2.53 D, based on different measurement methods and optotypes.^{22 23} The contraction of the ciliary body induces forward axial IOL movement.²⁴ Studies have reported that approximately 28% of patients exhibit anterior chamber depth changes of at least 0.3 mm due to this movement.²² The accommodative effect generated by this axial IOL movement varies significantly with axial length. For every 1.0 mm forward IOL movement, the amount of accommodation achieved varies from 0.8 D in a long eye to 2.3 D in a short eye.²⁵ In addition to axial IOL movement, eyes with senile miosis, higher corneal higher-order aberrations and greater corneal multifocality are more predisposed to exhibit greater apparent accommodation, which primarily affects intermediate and near vision.^{22 26} Hyperopic defocus may be partially compensated through apparent accommodative ability. In contrast, no analogous compensatory mechanism exists for myopic defocus.

We confirmed that even small myopic refractive errors can significantly impair UDVA in patients with monofocal IOLs. A previous study of 493 eyes recommended targeting a myopic SE of −1.00 D to −1.50 D in unilateral cataract surgery with monofocal IOL to enhance visual function.²⁷ However, the reported UDVA in those eyes ranged from 0.26 logMAR to 0.32 logMAR, which may fall short of contemporary patient expectations for high-quality distance vision. Our findings demonstrated that slight myopia [−0.50 D, 0 D) significantly impaired UDVA overall and in the minimal astigmatism subgroup. This effect was further amplified in the −1.00 D to −0.50 D range across all astigmatism subgroups. In contrast, hyperopic SE within 1.00 D did not significantly increase the odds of not attaining UDVA ≤0.1, either overall or in any astigmatism subgroup. For patients opting for monofocal IOL, the primary goal is typically to achieve excellent UDVA, while accepting the need for spectacles for near vision. Therefore, targeting emmetropia or slight hyperopia is more likely to meet their visual expectations.

Our findings underscored the necessity of adjusting the target refraction based on the specific classification of astigmatism. A slight hyperopic bias was beneficial for uncorrected distance vision in eyes with ATR astigmatism and oblique astigmatism, whereas emmetropia (0 D) remained optimal for eyes with WTR astigmatism and minimal astigmatism (<0.50 D). Previous studies have indicated that both the magnitude of residual astigmatism and spherical error independently affected UDVA, irrespective of astigmatic orientation.^{11 12} They found that residual astigmatism ≥0.25 D, myopic sphere ≤−0.25 D and hyperopic sphere ≥+0.50 D significantly impaired UDVA.^{11 12} However, these earlier investigations did not evaluate the potential interaction between residual sphere and astigmatism, nor did they exclude participants with suboptimal CDVA. In contrast, our study identified the optimal refractive targets across different types of astigmatism. In the WTR group, small refractive errors were found to have a comparatively lower impact on UDVA. Unlike other astigmatism subtypes, eyes with WTR astigmatism exhibited greater tolerance to myopic refractive errors, whereas the minimal astigmatism group showed the poorest tolerance. This difference may be attributed to eyelid-mediated optical modulation. As proposed by Huber,²⁸ in WTR astigmatism, eyelid squeezing can induce a horizontal slit pupil effect, enhancing depth of focus through a pinhole effect and consequently improving UDVA by selectively sharpening the vertically blurred meridian. Supporting this, Rhim et al²⁹ further demonstrated that eyelid squinting improved distance vision in pseudophakic eyes with WTR astigmatism by shifting the circle of least confusion from the middle of the two focal lines to a more distant point (closer to the retina). In contrast, other astigmatism subtypes lacked this compensatory mechanism.²⁹

Notably, although UDVA was maximised at zero or slight hyperopic refractions, concomitant astigmatism ≥0.50 D remained a vision-limiting factor necessitating targeted correction. Our study demonstrated that residual astigmatism below 0.50 D had a negligible effect on UDVA. These findings supported prioritising surgical correction for predicted postoperative astigmatism exceeding this threshold, which aligned with previous research.¹⁸ Specifically, predicted postoperative astigmatism ≥0.50 D was suggested to be corrected as much as possible, either through toric IOL or corneal incisions.

Advancements in medical technology have made it possible to achieve precise control over postoperative refractive outcomes, including the management of both astigmatism and SE.³,⁶ However, most commercially available types of IOL are manufactured in 0.50 D increments, leading to non-zero postoperative refractive errors for the majority of patients. Consequently, the introduction of IOLs with finer dioptric gradients is suggested to minimise residual refractive error and optimise visual outcomes. Our study provided clinically relevant guidance for IOL power selection. In order to achieve better UDVA, primary consideration should be given to correcting astigmatism, ensuring that postoperative astigmatism is controlled at a minimal level. Simultaneously, a strategic approach to IOL power selection should be adopted. By using IOL with smaller power increments, the target postoperative SE should be emmetropia (0 D) or deliberately biased towards a slightly hyperopic outcome.

Our study found that hyperopia and myopia of the equivalent magnitude exerted asymmetrical effects on UDVA. This clinical observation was consistent with prior research using defocus curves in which the simulated hyperopia resulted in a more gradual decline in distance visual acuity compared with simulated myopia.³⁰ Defocus curves simulate refractive error across a range of distances, thereby reflecting visual acuity at various dioptric levels. However, the interpretation of defocus curves is limited by a lack of standardisation, as multiple factors can influence their results, including pupil size, contrast level, sphere versus cylinder defocus, viewing distance, monocular versus binocular assessment, chart type, etc.³¹ An earlier investigation of defocus curves in pseudophakic eyes suggested that hyperopic defocus may be less detrimental to visual acuity than myopic defocus.³⁰ Through the analysis of real-world clinical data, our study provided empirical validation of this optical phenomenon, thereby offering evidence-based guidance for refractive target selection in routine practice.

This study had several limitations. First, this single-centre study was single ethnic. As such, these findings provided valuable reference for Asian populations. Second, the study did not include assessments of subjective visual function, near vision or patient satisfaction. Third, our regression models adjusted only for age and sex. However, it is important to note that our strict inclusion criterion (CDVA ≤0.1 logMAR) inherently excluded participants with significant vision-impairing conditions, such as severe posterior capsular opacification or serious ocular surface disease, which were therefore not included as covariates. Finally, although all visual acuity tests were conducted under standardised photopic conditions to induce physiological miosis and patients with pupillary abnormalities were excluded, individual physiological variations in pupil size may remain a potential confounding factor affecting postoperative UDVA.

Conclusion

In this study, the impact of refractive errors (≤1.0 D) on UDVA was associated with the magnitude and type of astigmatism. Residual astigmatism of ≥0.50 D exerted a significant negative effect on UDVA. A plano SE (0 D) was optimal for minimum and WTR astigmatism, whereas slight hyperopia yielded superior UDVA in ATR and oblique astigmatism.

Data availability statement

Data are available upon reasonable request.