ASCRS News: EyeWorld Journal Club
Winter 2024

by Eli Pratte, MD,* John Wilson, MD,* Maggie Xing, MD,* Washington University Department of Ophthalmology and Visual Sciences residents, Arsham Sheybani, MD, Residency Program Director
*All authors contributed equally to this work.

Toric intraocular lens (TIOL) implantation aims for correction of pre-existing corneal astigmatism during cataract surgery and has been shown to improve uncorrected distance visual acuity (UDVA) and increase spectacle independence.¹ While traditional teaching cautions against flipping the axis of astigmatism (i.e., overcorrection of the astigmatism) during TIOL placement, recent publications reported no intolerance in patients with astigmatic overcorrection as well as possible benefit from axis-flip to with-the-rule astigmatism.^2,3,4 In this study, Naderi et al. performed a post hoc analysis to investigate the effects of axis-flip after TIOL placement on visual performance and patient-reported outcomes. The authors also recalculated TIOL power with inclusion of posterior corneal astigmatism (PCA), which was not utilized in the initial study, to further identify any effects on axis-flip in this cohort of patients.

Methods

This was a post-hoc analysis of data from patients who received a TIOL for intraoperative astigmatism correction from a prospective randomized case-control study conducted at Guy’s and St. Thomas’ Hospital, NHS Foundation Trust between October 2019 and March 2022. Participants had symptomatic cataracts and regular corneal astigmatism of greater than or equal to 1.50 D of cylinder and less than 5.00 D of cylinder, as measured on the IOLMaster 700 biometer’s (Carl Zeiss Meditec) anterior keratometry function and confirmed by Scheimpflug tomography (Oculus). Participants were free of other vision-impairing comorbidities. All participants received aspheric RayOne single-piece, hydrophilic acrylic TIOL RAO610T (Rayner) initially sized per the Raytrace online premium IOL calculator (Rayner) without inclusion of PCA data. In the initial study, participants were randomly assigned into a control group that received a TIOL calculated to achieve full correction of astigmatism or an intervention group that utilized a 2.00 D cylinder (DC) or 4.00 DC TIOL with residual corneal astigmatism corrected by an opposite clear cornea incision. In the post-hoc analysis, 6-month postop subjective refractive positive cylinder was compared with baseline biometric steep axis to analyze the number of cases of axis-flip (defined as a change in cylinder axis of 90±22.5 degrees) from the baseline corneal astigmatism. Quality of life scores CATPROM-5 and EuroQoL EQ-5D-3L were used as proxies of patient tolerance of any residual postoperative cylinder. Additionally, PCA inclusion in TIOL calculation was retrospectively analyzed to identify cases in which axis-flip may not have occurred with inclusion of this coefficient in original lens calculations.

Results

The original study enrolled 102 patients, 92 of whom underwent cataract surgery and received a TIOL. Of those, 91 followed up at 4 weeks and 84 at 6 months. Demographic data was similar between the flipped cases (FC) and the unflipped cases (UF) in terms of age, sex, preoperative measures including UDVA, CDVA, biometric K1-K2, Pentacam K1-K2, axial length, CATPROM-5, EQ-5D-3L value and raw scores, and intervention group-to-control group ratio. There were fewer eyes with against-the-rule (ATR) and oblique (OB) astigmatism in the flipped group.

At 6 months postoperatively, about 1/3 (29/84, 34.5%) of patients had axis-flip. The mean UDVA (logMAR) was similar (0.13 for both groups), as was CDVA (0.01 [flipped] vs. 0.00 [unflipped]) and refractive cylinder (0.74 [flipped] vs. 0.93 [unflipped]). Quality of life measures CATPROM-5 and EQ-5D-3L were similar as well.

Among the flipped group, 28 eyes flipped from with-the-rule (WTR) to ATR, none flipped from ATR to WTR or OB, and 1 flipped from OB to OB. The mean change in axis was greater for the flipped group. Regarding vector analysis, the target induced astigmatism was similar, but the surgically induced astigmatism (SIA) was greater for the flipped group, 2.90 vs. 2.14. The difference vector and degrees of angle of error were not significantly different. The magnitude of error was greater in the flipped group, as was the correction index.

The authors recalculated TIOL powers with adjustment for PCA and found that the recommended cylindrical power decreased for FC with preoperative WTR, while it increased for UC with ATR. Among FC, 6 eyes (21%) may have avoided axis-flip had PCA been adjusted for. In contrast, among UF, 2 eyes (4%) would have had axis-flip with PCA adjustment.

Discussion

This study certainly provides important evidence that axis-flip, even WTR to ATR, may not impact final VA or affect patient satisfaction.

In this post-hoc analysis of 84 patients from a prospective randomized study of toric intraocular lens outcomes, Naderi et al. set out to determine the effects of cylinder axis-flip after TIOL insertion on visual acuity and patient-reported outcome measures (PROMs). This study sets out to challenge the common consideration of erring on the side of undercorrection of cylindrical axis during TIOL insertion to avoid possible negative outcomes with axis-flip.⁵ The initial prospective study from which they performed their post hoc analysis included patients whose astigmatism was completely addressed with TIOL (in addition to the SIA from the main wound), while others were a combination of TIOL with an opposite clear corneal incision. Ultimately, their study did not reveal any difference in UDVA, CDVA, or any of the 3 utilized PROMs between flipped and unflipped cases. Interestingly, despite the accepted notion that WTR is better tolerated than ATR, there were no differences in vision or PROMs despite 28/29 (96.6%) FCs that were originally WTR that flipped to ATR.

The authors also discussed further sub-analysis of their patient data retroactively calculating TIOL power given consideration of PCA, which was not measured or considered in preoperative planning during the initial prospective trial. They determined that for both UF with preoperative ATR astigmatism and FC with preoperative WTR astigmatism, there was a significant reduction in calculator-recommended TIOL cylindrical powers when PCA was considered. This reaffirmed the understanding that PCA is typically greater vertically, inducing a larger minus lens effect in the vertical axis and therefore shifting any residual postoperative cylinder ATR. This leads to interesting discussion on age-related drift to ATR astigmatism and how future studies may provide more basis for patient-specific considerations for under- or overcorrection of WTR or ATR, respectively, to optimize visual outcome.

This study challenges the suggestion that flipping the patient’s preoperative baseline cylindrical axis might result in reduced postoperative visual acuity and patient satisfaction. Their conclusion that axis-flipping did not impact post-TIOL visual acuity or PROMs may support the surgical decision to completely correct all preoperative astigmatism regardless of axis.

While this study does oppose common notions of axis-flip that can lead to more optimized visual outcomes for post-TIOL patients, there are certainly limitations as well. Although they do indicate that the initial prospective study was sufficiently powered to determine significant differences in outcomes between their control group and investigative group, their follow-up attendance of 84 patients at the 6-month time point does make extrapolating clinical significance from their study more difficult. Additionally, they do comment that there was no significant difference in the mean postoperative axis change between the control group and investigative group from the original prospective study, however, combining these two cohorts together is less ideal than if there were a single group of patients being studied where preoperative cylinder had been addressed in a single fashion. Lastly, their proxies for tolerance of axis-flip were visual acuity and PROMs and do not include a defocus curve. VA and PROMs are important to assess and do clearly suggest tolerance of axis-flip given their results. However, any true effect of these factors to suggest astigmatic intolerance may be missed due to the smaller cohort of patients, which might have been revealed if defocus curves could have been completed.

Nevertheless, this study certainly provides important evidence that axis-flip, even WTR to ATR, may not impact final VA or affect patient satisfaction. Factoring this into preoperative selection of TIOLs may lead to more accurate and effective reduction of a patient’s cylinder power. This mindset in conjunction with consideration of PCA and the drift to ATR with time provides an exciting future opportunity to optimize visual outcomes for patients. 

---

The effects of axis-flip of the refractive cylinder on vision and patient-reported outcome measures following toric intraocular lens implantation

Naderi K, et al. J Cataract Refract Surg. 2024. Online ahead of print.

• Purpose: To investigate the effects of cylinder axis-flip after toric intraocular lens (TIOL) implantation on vision and patient-reported outcome measures (PROMs).
• Setting: Teaching hospital in the United Kingdom.
• Design: Post hoc analysis of data from patients who participated in a prospective randomized study of TIOLs.
• Methods: Axis-flip was defined as a change in postoperative refractive cylinder (RC) axis of 90±22.5 degrees from the preoperative biometric axis. Uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), residual RC, and CATPROM-5 and EQ-5D-3L quality-of-life scores were analyzed.
• Results: At 6 months, axis-flip occurred in 29 (34.5%) of 84 eyes of which 28 had with-the-rule (WTR) astigmatism preoperatively. Mean (±SD) UDVA (logMAR) was 0.13 (0.16) in flipped cases (FCs) and 0.10 (0.14) in unflipped cases (UFs) (P=.88). CDVA was 0.01 (0.11) in FC and was 0.00 (0.09) in UF (P=.68). Mean RC was 0.74 diopters (D) (0.41) in FC and 0.93 D (0.47) in UF (P=.08). Mean CATPROM-5 score was 6.22 (2.56) in FC and 5.52 (3.03) in UF (P=.29). Mean EQ-5D-3L calibrated score was 0.89 (0.19) in FC and 0.85 (0.19) in UF (P=.35). Retrospectively applying coefficients of adjustment to account for posterior corneal astigmatism (PCA) suggested that 6 eyes (21%) of FC with WTR might have avoided axis-flip.
• Conclusions: Axis flipping after TIOL implantation did not adversely influence visual acuity or PROMs scores. Most FC had WTR preoperatively. Adjusting for PCA might have reduced axis-flip in some of these eyes.

---

References

1. Kessel L, et al. Toric intraocular lenses in the correction of astigmatism during cataract surgery: a systematic review and meta-analysis. Ophthalmology. 2016;123:275–286.
2. Kent C. 37 ways to get great outcomes with torics. Review of Ophthalmology. 2012. Accessed October 19, 2024. www.reviewofophthalmology.com/article/37-ways-to-get-great-outcomes-with-torics
3. Hoffmann PC, et al. Results of higher power toric intraocular lens implantation. J Cataract Refract Surg. 2011;37:1411–1418.
4. Beheregaray S, et al. Astigmatic overcorrection and axis flip for targeting minimal remaining refractive astigmatism with toric intraocular lenses. J Cataract Refract Surg. 2018;44:109–110.
5. Hasegawa Y, et al. Type of residual astigmatism and uncorrected visual acuity in pseudophakic eyes. Sci Rep. 2022;12:1225.

Contact 

Sheybani: sheybaniar@wustl.edu