ASCRS News
Winter 2025

by Dany Hage, MD,* Johnson Huang, MD,* Jonathan Le, MD,* Karolina Leziak, MD,* Emma Stenz, MD,* Marcus Turner, MD,* Ophthalmology Residents; Parisa Taravati, MD, Residency Program Director, University of Washington
*These authors contributed equally to this work.

Keratoplasty, whether through penetrating keratoplasty (PK) or deep anterior lamellar keratoplasty (DALK), is a cornerstone surgical technique for managing several corneal diseases. Despite its success in restoring corneal clarity, visual acuity is often limited by high degrees of post-keratoplasty astigmatism.¹ Advances in suture management, corneal refraction, contact lenses, and toric lenses have improved outcomes to some extent.² However, further advances are needed to improve vision. Iris-claw phakic intraocular lenses (pIOL) have demonstrated efficacy in correcting astigmatism in both unoperated and post-keratoplasty eyes. In this single-center, retrospective study of patients implanted with iris-claw pIOL, Ribeiro et al. evaluated long-term refractive outcomes, safety profile, and endothelial cell density (ECD) changes in post-keratoplasty eyes. Furthermore, the authors compared the results between post-PK and post-DALK eyes.³

Design and methods

The article described a retrospective analysis of patients who had received iris-claw pIOL for high ametropia and astigmatism after PKP or DALK at the Santo Antonio Local Health Unit, in Porto, Portugal, between January 2013 and December 2023. Eligible patients were 21 years or older, with 12 or more months of refractive stability, best-corrected visual acuity of 20/40 or better, and glasses or contact lens intolerance. Inclusion criteria were anterior chamber depth of 3 mm or greater and ECD of at least 2,000 cells/mm² for DALK and 1,000 cells/mm² for PK. Exclusion criteria were corneal scarring, less than 12 months of follow-up data, past uveitis or ocular hypertension history, and any previous refractive procedures. Surgeries were conducted by three experienced surgeons at a single center.  

Patient data were examined at four visits, immediately prior to implantation, 12–18 months after implantation, and 5- and 10-year subsequent visits. Each visit consisted of corrected and uncorrected visual acuity measurement, intraocular pressure measurement, and complete ophthalmic examination including fundoscopy. Additionally, ECD measurements, corneal tomography, and corneal wavefront analysis were conducted at these visits. Refraction was done 3 months after removal of corneal sutures.  

Implantation of the IOL was performed by three corneal specialists and involved paracenteses, instillation of viscoelastic in the anterior chamber, insertion of the lens using lens spatula or forceps, fixation on the horizontal axis, and surgical iridectomy. Wounds were closed with a single polyamide suture. Patients were given a regimen of postop antibiotic, steroid, and nonsteroidal anti-inflammatory drops. Oral prednisolone was given to those without contraindications. 

Statistical analysis included data normality assessment with the Kolmogorov-Smirnov test, comparisons between continuous variables employed the Student’s t-test and the Mann-Whitney U test, Fisher’s exact test for nominal scaled data, and correlations were analyzed using Spearman’s bivariate correlation test. A p value of less than 0.05 was considered statistically significant, and all values were expressed as mean ± standard deviation. 

Results

This cohort included 39 eyes across 29 patients, 62% of whom were male with a mean age of 35±8.6 years. All patients had undergone keratoplasty for treatment of keratoconus, with 26 eyes (67%) receiving DALK and 13 eyes (33%) receiving PK. Of the 25 eyes (64%) that received the Artiflex lens (Ophtec), 15 eyes (60%) had undergone a prior DALK.

At the intermediate follow-up, postop uncorrected distance visual acuity (UDVA) was ≥20/32 in 49%, ≥20/25 in 28%, and ≥20/20 in 15% of eyes. Full refractive correction without loss of corrected distance visual acuity (CDVA) was identified in 41% of eyes, with 46% gaining ≥1 line and 7.7% losing >1 line. There was an efficacy index of 0.83 and a safety index of 1.18.  

At the 10-year follow-up, postop UDVA was ≥20/32 in 57% of eyes, ≥20/25 in 29%, and ≥20/20 in 14%. Full refractive correction without loss of CDVA was seen in 21% of eyes, whereas 64% gained ≥1 line. There was an efficacy index of 0.91 and a safety index of 1.27. 

Accuracy was assessed by evaluating the final refraction. At the intermediate follow-up, 59% of eyes were within ±0.50 D of their target refraction, and 72% were within ±1.00 D. At the 10-year follow-up, this decreased to 43% within ±0.50 D and 64% within ±1.00 D. There was a residual astigmatism of ≤0.50 D in 49% of eyes and ≤1.0 D in 64%; however, there was no significant difference between DALK and PK.

The predicted annual endothelial cell loss (ECL) was 3.07% in DALK patients and 9.37% in PK patients. Over the 10-year follow-up period, ECD decreased from 2,186 cells/mm² to 1,597 cells/mm² in the DALK group and from 1,243 cells/mm² to 1,013 cells/mm² in the PK group.

No intraoperative complications were reported, and there were no incidences of spontaneous dislocation of the lens, iris atrophy, or cataract development. In the postop period, three patients developed progressive myopia and were treated with PRK. Two patients required IOL explanation, one for corneal ectasia and another for ECL.

Discussion

Postop refractive error currently limits the visual recovery of patients undergoing PK and DALK. In this study, the authors argued that pIOL in this cohort serves the important purpose of restoring vision but also draws attention to the need for patient selection for a safe outcome.

Compared to pIOL in unoperated eyes, pIOL in post-keratoplasty patients is important in restoring functional vision for those having difficulty tolerating spectacle correction due to aniseikonia or contact lens use. PIOL re-establishes functional vision, as nearly 80% of patients had preop UCVA worse than 20/200. At 10 years, nearly half of the cohort still had greater than 20/32 UCVA.  

The authors demonstrated that 2/3 of patients achieved postop UCVA equal to or better than preop CDVA. Compared to unoperated eyes undergoing pIOL, there was more refractive undercorrection. The authors pointed out several factors that may contribute to this finding, including: 1) difficulty refracting keratoplasty eyes, and 2) graft remodeling changes over time secondary to ectasia and scarring (the authors pointed to a shift in keratometry over 10 years). Despite this, based on the authors’ data, astigmatic outcomes were highly predictable. Half the patients were within 0.5 D of refractive cylinder, with 78% within 5 degrees of final refraction. Vector analysis confirmed a significant reduction in astigmatic magnitude and dispersion. Furthermore, wavefront analysis did not show an increase in higher order aberrations postop. 

Short- and long-term safety are important considerations when adopting new surgical techniques. The authors reported no intraoperative complications. In terms of long-term safety, one patient underwent pIOL explantation due to ECL. No patients developed cataract during the follow-up window. However, there was a higher rate of ECL in the study cohort than keratoplasty patients who did not undergo pIOL and normal eyes that underwent pIOL alone. In particular, there was an average annual 9.34% ECL loss in the PK + pIOL group compared to the 5.5% with PK alone. The authors suggested DALK + pIOL as a safer alternative, as a smaller rise in ECL was seen in the study group compared to the expected ECL in DALK (3.07% vs 2.3%, respectively). The measurement of ECL loss as a percentage rather than absolute may also exaggerate the difference between the two groups, as the average ECD in the PK group was much less compared to DALK. The reported 9.34% loss in the PK group vs. 3.07% loss in the DALK group suggests triple the loss; however, the difference in absolute loss of ECD may be similar. Though the authors make comparisons in ECL in the PK vs. DALK groups, these comparisons are limited due to the statistically significant difference in ages of both grafts and patients between groups.

Most importantly, the cumulative effect of endothelial cell injury serves as a caution in patient selection. The authors made a point to select patients who failed spectacles and contact lenses, with extensive discussion on higher chances of corneal endothelial failure postop.

Limitations of this study include a small number of surgeons (3) without a standardized surgical protocol. Different-sized incisions and IOL types were used. Importantly, the Artisan lens is a rigid, non-foldable lens, whereas the Artiflex lens is a rigid, foldable lens, which could lead to differences in surgical handling and therefore affect primary and secondary outcomes of this study. Also, it is unclear whether ECD measurements were standardized between patients. Measuring ECD from a single frame for each patient instead of an average of multiple ECD measurements could affect accuracy.

Future directions include posterior chamber pIOL to preserve endothelial cell count. However, the authors decided against its use due to concern over lens complications, as there is no strong data regarding vault prediction in PK eyes. Further research could determine if posterior chamber pIOL reduces the rate of ECL.

This study is limited by its retrospective nature and small sample size. There were only 11 patients in the PK group, compared to 38 in the DALK group. In addition, there was a difference in follow-up time periods between PK eyes and DALK eyes, limiting the ability to draw conclusions on longer-term safety.

Conclusion

In this retrospective case series, Riberio et al. described the refractive outcomes and efficacy of iris-claw pIOL in eyes with prior PK and DALK. They found that the procedure had acceptable safety indices at all postop time points, with the majority of eyes experiencing stability of or an improvement in CDVA. Additionally, approximately 64% of eyes globally had residual astigmatism of <1.0 D associated with the demonstrated reduction in the average horizontal astigmatic vector. This present series characterizes iris-claw pIOL as a viable surgical option for post-PK and post-DALK astigmatism, with minimal adverse events over a 10-year total follow-up period. 

---

Long term efficacy and safety of iris-claw phakic intraocular lens in the management of post keratoplasty astigmatism

Ribeiro BB, et al. J Cataract Refract Surg. 2025. Online ahead of print. 

• Purpose: To demonstrate the efficacy and safety of iris-claw phakic intraocular lens (pIOL) in the treatment of post keratoplasty astigmatism.
• Setting: Cornea Unit, Ophthalmology Department, Unidade Local de Saúde de Santo António, Porto, Portugal.
• Design: Retrospective longitudinal study.
• Methods: Analysis of patients submitted to iris-claw phakic IOL implantation after previous penetrating keratoplasty (PK) or deep anterior lamellar keratoplasty (DALK) between January 2013 and December 2023. Patients submitted to other refractive procedures and without minimum 1-year follow-up were excluded. Demographic data, spherical equivalent (SE), uncorrected (UDVA), and corrected (CDVA) distance visual acuity was registered preoperatively, at the intermediate visit (12–18 months), 5 years, and 10 years postoperatively. Vector analysis to plot corneal surgically induced astigmatism was performed. Endothelial cell density (ECD) was evaluated at each visit using the same specular microscope.
• Results: We enrolled 39 eyes of 29 patients submitted to keratoplasty due to keratoconus (71.8% [n=28] DALK). At the intermediate visit, 19 eyes (49%) had a postoperative UDVA ≥20/32, and 6 (15%) had ≥20/20. Efficacy index was 0.83. 16 (11%) achieved full refractive correction without loss of CDVA, and 9 (23%) gained ≥1 lines of vision. 28 (72%) eyes were within ±1.00 D of plano refraction. Globally, 25 (64%) eyes showed a residual astigmatism ≤1.00 D. Average annual ECD loss was 3.07% for DALK and 9.34% for PK (p=0.006).
• Conclusion: Iris-claw pIOL is effective for treating post keratoplasty astigmatism. Given the risk of progressive endothelial cell loss, strict follow-up is mandatory, especially after PK.

---

References

1. Deshmukh R, et al. Post-penetrating keratoplasty astigmatism. Surv Ophthalmol. 2022;67:1200–1228. 
2. Fares U, et al. Management of post-keratoplasty astigmatism. J Cataract Refract Surg. 2012;38:2029–2039.
3. Ribeiro BB, et al. Long term efficacy and safety of iris-claw phakic intraocular lens in the management of post keratoplasty astigmatism. J Cataract Refract Surg. 2025. Online ahead of print.

Contact 

Taravati: taravati@uw.edu