Accuracy Of Intraocular Lens Power Calculation Formulas In Sequential Versus Simultaneous Penetrating Keratoplasty And Cataract Surgery
Published 2026 - 44th Congress of the ESCRS
Reference: POCAT614 | Type: Free paper | DOI: 10.82333/2t0y-w242
Authors: Dr Kazuhiro Nakamoto* 1 , Dr So Goto 1 , Dr Takeshi Soma 1 , Dr Yoshinori Oie 1 , Dr Akihiko Shiraki 1 , Dr Masaki Suzue 1 , Dr Yu Yoshinaga 2 , Dr Naoyuki Maeda 3 , Prof. Dr Kohji Nishida 4
1Ophthalmology,The University of Osaka Graduate School of Medicine ,Osaka,Japan, 2Ophthalmology,Sumitomo Hospital,Osaka,Japan;Ophthalmology,The University of Osaka Graduate School of Medicine ,Osaka,Japan, 3Kozaki Eye Clinic,Osaka,Japan, 4Ophthalmology,The University of Osaka Graduate School of Medicine ,Osaka,Japan;Integrated Frontier Research for Medical Science Division, Institute for Open and Transdisciplinary Research Initiatives (OTRI),The University of Osaka,Osaka,Japan
Purpose
Intraocular lens (IOL) power calculation in eyes undergoing penetrating keratoplasty (PKP) remains inherently challenging. Refractive predictability may be further compromised when cataract extraction is performed simultaneously with PKP, as corneal power must be assumed rather than measured. This study quantitatively compared the refractive predictive accuracy of three intraocular lens (IOL) power calculation formulas in sequential and simultaneous penetrating keratoplasty and cataract surgery.
Setting
This retrospective, single-center clinical study was conducted at The University of Osaka Hospital, Osaka, Japan. Medical records of consecutive patients who underwent cataract surgery between January 2017 and November 2025 were reviewed.
Methods
Eyes undergoing simultaneous penetrating keratoplasty and cataract surgery or sequential penetrating keratoplasty followed by cataract surgery were analyzed. IOL power was calculated using SRK/T, Haigis, and Barrett Universal II formulas. In the simultaneous surgery group, keratometry was assumed to be 41–43 D. Axial length was measured by optical biometry. Prediction error was defined as postoperative spherical equivalent minus predicted refraction. Refraction at 3 months was used in the sequential surgery group, and refraction at visual stabilization within 1 year was used in the simultaneous surgery group. Median absolute error (MedAE) and the proportion within ±1.00 D were evaluated.
Results
Nineteen eyes were included in the sequential surgery group and thirteen in the simultaneous surgery group. In the sequential surgery group, mean prediction error (mean ± SD) was −0.68 ± 2.04 D (SRK/T), 0.013 ± 2.37 D (Haigis), and 0.072 ± 2.13 D (Barrett Universal II). In the simultaneous surgery group, it was 0.16 ± 5.49 D, −0.45 ± 5.51 D, and 0.16 ± 5.55 D, respectively. MedAE was 1.20 D (SRK/T), 1.18 D (Haigis), and 1.00 D (Barrett) in the sequential group, compared with 2.90 D, 2.77 D, and 2.60 D in the simultaneous group. Within ±1.00 D using SRK/T, 47% (9/19) of sequential eyes achieved target refraction compared with 0% (0/13) in the simultaneous group (p = 0.004). Similar trends were observed with the other formulas.
Conclusions
The sequential surgery group demonstrated superior refractive predictability compared with the simultaneous surgery group. Despite the use of modern IOL formulas, refractive outcomes in the simultaneous surgery group remained highly variable, likely reflecting assumed keratometric values and postoperative corneal curvature changes. Sequential surgery may provide more reliable refractive outcomes when refractive precision is prioritized. However, compared with normal eyes, refractive accuracy remains limited, and further refinement of IOL power calculation strategies in keratoplasty cases is required.