Toric Intraocular Lens versus Peripheral Corneal Relaxing Incisions for Astigmatism between 0.75 and 2.5 Diopters: 5-Years Outcomes

Abstract Purpose To compare 5-year outcomes of toric intraocular lens (tIOL) or peripheral corneal relaxing incision (PCRI) for correction of keratometric astigmatism (KA) between 0.75 and 2.5 diopters (D). Methods Setting: University Hospital. Design: Randomized clinical trial. Eighty eyes (80 participants) received either tIOL or PCRI. Patients were assessed preoperatively, 1-month, 1, and 5 years. Primary outcomes were uncorrected (UDVA) and best-corrected distance logMAR visual acuity (CDVA). Secondary outcomes were a manifest refractive sphere, refractive astigmatism (Ra), spherical equivalent (SEQ), KA & mean keratometry (KM), and Quality-of-Life Impact of Refractive Correction (QIRC) scores. Results There was no difference between the two groups in UDVA, CDVA, refractive sphere, KM, RA, and SEQ. KA reduced after 1 month in the PCRI group and remained stable until 5 years. From 1 to 5 years, the number of eyes with distance emmetropia (within ±0.13D) changed from 59% (20/34 eyes) to 32% (6/19 eyes) for tIOLs and from 43% (15/36 eyes) to 20% (4/21 eyes) for PCRIs with 32% (6/19 eyes) and 20% (4/21 eyes) showing >0.5D change in SEQ at 5-years respectively. Compared to 1-year, Ra significantly increased at 5 years in both groups with no difference between the groups. Mean overall QIRC scores were not different between the groups (tIOL: 49.88 ± 7.47; PCRI: 52.09 ± 7.02; p = .18). Conclusions Although there was no difference between the overall visual and vision-related quality of life outcomes between tIOLs and PCRIs, an increase in refractive astigmatism and reduction in distance emmetropia with time was noted in both groups.


Introduction
Clinically significant astigmatism affects approximately one in five people undergoing cataract surgery, and surgeons can correct it at the time of cataract surgery. 1 Astigmatism correction continues to evolve with sophisticated tools and technology development.][5] A recent Cochrane review concluded that toric IOLs probably provide a higher chance of achieving astigmatism within 0.5D after cataract surgery. 6However, the authors did not find the small mean difference in postoperative astigmatism, favouring toric IOLs, clinically significant.Further, there was no evidence of a difference in postoperative visual acuity or quality of life between the techniques and with uncertain evidence of adverse effects.We previously reported one-year outcomes of keratometric astigmatism correction between 0.75 and 2.5 diopters (D) with tIOL or PCRI during cataract surgery in a randomized clinical control trial.We found no difference in overall visual outcomes between the two groups. 5The purpose of this study was to report long-term (5-years) visual outcomes and vision-related quality of life outcomes in the same cohort of patients.

Materials and methods
This prospective randomized clinical trial was conducted at Sussex Eye Hospital, University Hospitals Sussex University Hospitals NHS Trust, Brighton, England, between March 1, 2014, and September 30, 2021.This study was approved by the National Ethics Committee and followed the tenets of the Declaration of Helsinki.The study was registered with UKCRN (ID:16848) and ClinicalTrials.gov(NCT02067429).The methodology of the study has been previously elaborated in detail. 5he first eye of those with a symptomatic cataract, postoperative visual potential of 0.2 logMAR or better, and corneal astigmatism between �0.75 D and �2.5 D on topography (Pentacam HR; Oculus, Wetzlar, Germany) with post-operative refractive target of emmetropia were included in the study.We excluded those under the age of 18 years, as well as eyes with any ocular comorbidity with cornea, uvea, retina, or optic nerve that may be detrimental to visual outcomes.Diabetic subjects; subjects with glaucoma, abnormal corneal topography, or any other coexisting retinal or cornea conditions; subjects with astigmatism outside the study range; subjects concurrently using ocular medications, including lubricants; and those unable to con-sent and unable to attend follow-up visits were excluded.
Eighty eyes of 80 participants were recruited for this study and were randomly allocated to either intervention arm (tIOL and PCRI) with 40 in each treatment group.Anterior keratometric data from the "4 map refractive" display produced by the Pentacam HR software assessment was exclusively used to calculate the dimensions and power of the tIOLs and PCRIs, using www.raytrace.rayner.comand Donnenfield's nomogram on www.LRIcalculator.comrespectively.All participants underwent optical biometry on IOLMaster 500 (Carl Zeiss Meditec, Jena, Germany), or an A-scan ultrasound biometry (Carl Zeiss Meditec) where optical biometry was not possible.A single experienced right-handed surgeon (M.A.N.) performed all cataract surgeries under topical anesthesia as a day procedure.All eyes were marked at 0 and 180 degrees on the slit lamp under topical anesthesia before the surgery using a Tomark tIOL marker (Geuder, Heidelberg, Germany) fixed to the applanation tonometer.
Patients were followed up at one month, one year, and five years.The primary outcome was uncorrected (UDVA) and best-corrected distance logMAR visual acuity (CDVA).Secondary outcome measures were a manifest refractive sphere, cylinder, spherical equivalent, keratometric astigmatism and mean keratometry, and Quality-of-Life Impact of Refractive Correction.The methods are described in detail in our publication of 12 months' outcome. 5One-month and one-year outcomes have previously been reported. 5Visual acuity was recorded using an ETDRS logMAR chart at 4 m.A single unmasked observer performed all the subjective manifest refractions at five-year follow-up visits.Zernike polynomial data values from the "Zernike's display" Pentacam HR assessment at each visit were trans-ported to a Microsoft Office Excel spreadsheet for analysis.A validated 20-question Quality-of-Life Impact of Refractive Correction (QIRC) score questionnaire 7 was used and administered by an experienced research nurse.
All data were recorded on Microsoft Office Excel 2010 (Microsoft Corporation, Redmond, Washington, USA).The Kolmogorov-Smirnov test tested the normality of all the data.SPSS Statistics version 22.0 (IBM Corporation, Armonk, New York, USA) was used for all statistical analysis.If the participant needed neodymium-yttrium-aluminum-garnet (Nd: YAG) laser capsulotomies at any follow-up, the data were recorded after performing the capsulotomies on separate visits a few days later.ANOVA test was used to compare all the parameters during all visits (preoperatively and postoperatively) within each intervention arm.The t-test was then used to compare all parameters between the PCRI and tIOL groups at each visit.The sample size calculation for this study is already reported in the previous article. 5A p value of <.05 was considered statistically significant.

Results
Eighty participants were enrolled in the study, and followup data were available for 70 patients; of these, 34 had toric IOL implantation, and 36 had PCRIs during their cataract surgery.The mean age of the study population at the time of surgery was 73 ± 3 years (M: F ¼ 30:40); both groups were comparable in terms of preoperative axial length, keratometry, astigmatism, uncorrected logMAR distance visual acuity (UDVA) and best corrected logMAR distance visual acuity (CDVA). 5Toric IOL rotation was assessed at one year follow-up only -all eyes rotated less than 10 degrees between 1 and 12 months, 96% eyes were within 5 degrees of rotation and 89% eyes were within 3 degrees.None of the patients required post-operative repositioning for misaligned tIOL.Fifteen patients from both groups were lost to follow-up at five years.

Primary outcome measure
Mean LogMAR UDVA and CDVA for the tIOL group vs the PCRI group were comparable at 1 year and 5 years (Table 1).At 5 years, 21% (4 out of 19) of eyes in tIOL group and 29% (6 out of 21) in PCRI group had UDVA of 20/20 or better, while 84% (16 out of 19) of eyes in tIOL group and 90% (19 out of 21) in PCRI group had UDVA of 20/40 or better (Figure 1(a)).At 1 year, all eyes in the tIOL group gained one or more lines, compared to 76% (16 out of 21) in the PCRI group. 5These figures were 84.2% (16 out of 19) and 95.2% (20 out of 21) at 5 years, respectively.There was no change in visual acuity in about 10.5% (2 out of 19) of eyes in the tIOL group and 4.8% (1 out of 21) in the PCRI group, while one eye (5%) lost one line in tIOL group at 5 years compared to preoperative CDVA (Figure 1(b)).

Manifest refractive sphere, cylinder, and spherical equivalent
There was no difference in either of these parameters when compared between both groups at five years (Table 1).However, when the parameters were compared at different time points in each group, there was a statistically significant difference in the refractive cylinder.While in the tIOL group, the cylinder progressively increased with time, in the PCRI group, there was an initial decline at 1 year compared to 1 month, which then progressed at 5 years (Table 1).Further, 47% of eyes in the tIOL group were within ±0.5D of the refractive cylinder compared to 57% (12 out of 21) in the PCRI group (Figure 1(c)).Similarly, more eyes (71%; 15 out of 21) were within ±0.5D of spherical equivalent in the PCRI group compared to the tIOL group (68%-13 out of 19) (Figure 1(d)).When comparing the stability of the spherical equivalent, 32% (6 out of 19) of eyes in the tIOL group had a change of >0.5D between 1 and 5 years compared to 20% (4 out of 21) in the PCRI group (Figure 2(a,b)).Double angle plots for comparing preoperative corneal astigmatism to postoperative refractive astigmatism at 1, 6, 12, 60 months are shown in Figures 3.

Keratometric astigmatism (K A ) and mean keratometry (K M )
We did not find any significant difference in anterior K A and K M between the preoperative visits and at preoperative, 1 month, 1 year, and 5 years (Table 2).As in our previous study, the anterior K A differed significantly preoperatively compared to 1 month, 1 year, and 5 years in the PCRI group (Table 2).Anterior and posterior K M and posterior K A were not significantly different between preoperative, 1 month, 1 year, and 5 years in the PCRI (Table 2).

Quality-of-life impact of refractive correction
Mean overall Quality of Life Impact of Refractive Correction (QIRC) scores improved significantly in both groups from a baseline of 46.81 ± 6.90 in tIOL group and 47.62 ± 6.24 in PCRI group to respectively 53.72 ± 7.19 and 54.91 ± 6.27 at 1-month, 50.73 ± 5.88 and 53.67 ± 4.73 at 1-year and finally 49.88 ± 7.47 and 52.09 ± 7.02 at 5 years (p ¼ .18 at 5 years).Overall scores between both groups were comparable at all given time points, and so was the reduction in scores at 5 years relative to 1 year (p ¼ .09).However, there were significant differences for four individual parameters, with the tIOL group performing better for questions 5 and 6 (Category: convenience) and the PCRI group performing better for Questions 1 (category: visual function) and 13 (category: health concerns).PCRI group also had higher scores for question 14 (Category: well-being); however, the difference was just short of reaching statistical significance (Supplementary Table 1

Discussion
In line with other reported studies in the literature, we found PCRIs to be as effective as tIOLs in correcting mild to moderate astigmatism over the long term. 3,4We noted a significant improvement in vision and vision-related quality of life with both treatment modalities, with no statistically significant difference between the two groups.A recent Cochrane review comparing the two modalities of correction reported that though there was a small mean difference in postoperative astigmatism, favoring toric IOLs, this was unlikely to be clinically significant at a shorter follow-up period. 6Further, there was no evidence of any clinically significant difference in postoperative visual acuity or quality of life between the techniques.We found that more patients had uncorrected visual acuity of 20/20 or better in the PCRI group (29%) than in the tIOL group (21%) (Figure 1(a)).However, this was not statistically significant (Table 1), and the trend was not maintained for other visual acuity cutoffs (Figure 1(a)).Visser et al. reviewed the literature on tIOL outcomes and reported pooled estimates of 33% eyes achieving a UDVA of 20/20 or better and 90% achieving a UDVA of 20/40 or better for the Alcon Acrysof tIOL, which was used in a majority of published studies; these parameters were 36% and 77% for the Rayner tIOL. 8However, cumulative pooled estimates were not reported, and the maximum median duration of followup was 13 months.In our study, 84.2% of patients in the tIOL group had one or more-line improvements in CDVA at 5 years compared to 95.2% in the PCRI group (Figure 1(b)).Interestingly, 76.2% of patients who had PCRI had three or more-line improvements as opposed to about 36.8% in the tIOL group (Figure 1(b)).There was no difference in preoperative logMAR best CDVA between the tIOL group (0.46 ± 0.52) and PCRI group (0.72 ± 0.58) (p ¼ .13)for patients who had five-year follow-up data.In a systematic review, Kessel et al. compared toric IOL vs non-toric IOL with or without astigmatic correction (relaxing incisions, opposite clear corneal incisions, and astigmatic keratotomy) for correction of mild to moderate preoperative astigmatism (range: 0.5-3D).with follow-up duration ranging from 3 to 12 months. 2They reported that postoperative UDVA was significantly better in the eyes implanted with a toric IOL, with the mean difference (95% CI) for all eyes being −0.07 logMAR (−0.10 to −0.04).In our study, the number of eyes with distance emmetropia (within ±0.13D) changed from 59% 5 at 1 year to 32% (Figure 1(d)) at 5 years for tIOLs and from 43% 5 to 20% (Figure 1(d)) with PCRIs with 32% and 20% eyes showing >0.5D change in spherical equivalent at 5 years in each group respectively.
Long-term refractive outcomes of astigmatic correction are sparingly reported in the literature.We found that both magnitudes of refractive astigmatism and spherical equivalent increased with time, though comparably, in both tIOL and PCRI groups; however, for each group, the refractive cylinder increased over time and was statistically significant (Table 1 and Figure 3).This can be attributed to long-term against-the-rule (ATR) change in corneal astigmatism with advancing age, which can occur irrespective of cataract surgery. 9,102][13][14][15][16] The reasons for the age-dependent change in the axis remain unclear.Still, they are hypothesized to be due to changes to upper eyelid tension, 17,18 intraocular pressures, 19 and possibly changes to the corneal structure. 10,13ore patients with PCRIs gained more than 3 lines over 5 years compared to patients with tIOLs.However, the overall visual outcomes were not statistically significantly different between the groups (Figure 2(b)).We hypothesize that this could be due to the change in corneal structure with PCRIs compared to tIOLs over 5 years.Further, an increase in refractive astigmatism after small incision cataract surgery  has previously been reported after toric IOLs implantation, more so in eyes with preoperative ATR astigmatism as compared to WTR or oblique astigmatism. 9,20In our study, we did not analyze eyes based on the subtype of preoperative astigmatism; however, a similar change in both groups over time is likely due to age-related corneal changes rather than the intervention.The spherical equivalent also changed similarly in both groups, with the lowest absolute values at 6and 12-month follow-up visits and a gradual but statistically insignificant increase in absolute value until five years (Figure 2).
Further, we compared the quality-of-life outcomes using previously validated QIRC scores, which indicate an initial sharp improvement in QOL in both groups, followed by a decrease in scores at 5 years, though still better than preoperative scores in both groups.Better subjective QOL after toric IOL implantation has been previously reported; 21,22 however, there is limited data on the same after PCRI intervention. 23To the best of our knowledge, ours are the only long-term vision-related QOL outcomes with tIOL and PCRI reported in the literature.Even though there were significant differences in a few parameters at follow-up, we did not find any difference in overall scores between the two groups at 1 month, 1 year or 5 years.(Supplementary Table 1) Similar to our study, Visser et al. did not find any significant difference in refractive error-related quality of life in a multicenter, randomized clinical trial comparing bilateral aspherical toric IOL with bilateral aspherical control IOL implantation. 24he main concerns with relaxing corneal incisions are less predictability and the possibility of increased complications like dry eye or incision site infective keratitis.In our study, we did not have any such complications throughout the follow-up period.Although toric IOLs may be more predictable and have better outcomes, esp.for high preoperative corneal astigmatism, they may still not be available in certain practices or countries.PCRIs are a simple, effective, and low-cost alternative in such situations.There was no out-ofpocket cost for the patient in our study as both interventions were available on the NHS.However, further health economics and cost-effectiveness studies may help evaluate this.The limitation of our study is the loss to follow-up at five years was more than accounted for due to those followups being at the height of the COVID-19 pandemic (2020-21).However, the strengths of our study lie in the randomized controlled design and the length of follow-up.
To conclude, for keratometric astigmatism between 0.75 and 2.5 diopters, cataract surgery combined with PCRIs or toric IOLs offers similar visual outcomes at 1 and 5 years with similar improvement in quality of life.Refractive astigmatism changed in both groups, leading to a change in spherical equivalents.There was a reduction in the number of eyes that have distance emmetropia at 5 years in both groups.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Figure 1 .
Figure 1.(a) Cumulative uncorrected distance visual acuity at 60 months.(b) The number of lines gained with best-corrected visual acuity at 60 months.(c) The refractive cylinder at 60 months postoperatively.(d) Spherical equivalent at 60 months postoperatively

Table 1 .
Comparison between LogMAR uncorrected distance visual acuity and best-corrected distance visual acuity at different postoperative visits (1 month, 1 year and 5 years) and between the groups.