Precision in Intraocular Lens Calculation for Chinese Cataract Patients with Prior History of Radial Keratotomy

Yang Xiang,1,2 Liangpin Li,2 Haiqiang Yu,1,2 Liyun Yuan,2,3 Maoyu Cai,4 Liwei Zhang,5 Xia Hua,6,7,* Xiaoyong Yuan1,2,* 1Clinical College of Ophthalmology, Tianjin Medical University, Tianjin, People’s Republic of China; 2Tianjin Key Laboratory of Ophthalmology and Visual Science, Tianjin Eye Institute, Tianjin Eye Hospital, Tianjin, People’s Republic of China; 3School of Medicine, Nankai University, Tianjin, People’s Republic of China; 4Department of Ophthalmology, The Affiliated Hospital of Guizhou Medical University, Guiyang, People’s Republic of China; 5Affiliated Hospital of Yunnan University, The Second Hospital of Yunnan Province, Yunnan Eye Hospital, Kunming, People’s Republic of China; 6Tianjin Aier Eye Hospital, Tianjin University, Tianjin, People’s Republic of China; 7Changsha Aier Eye Hospital, Changsha, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xiaoyong Yuan, Clinical College of Ophthalmology, Tianjin Medical University, Tianjin Key Laboratory of Ophthalmology and Visual Science, Tianjin Eye Institute, Tianjin Eye Hospital, No. 4 Gansu Road, Heping District, Tianjin, 300020, People’s Republic of China, Email yuanxy_cn@hotmail.com Xia Hua, Tianjin Aier Eye Hospital, 102 Fukang Road, Nankai District, Tianjin, 300191, People’s Republic of China, Email cathayhuaxia@163.comPurpose: To evaluate the predictive accuracy of eight traditional and contemporary intraocular lens (IOL) power calculation formulas in Chinese cataract patients with prior radial keratotomy (RK).Patients and Methods: RK patients who underwent phacoemulsification with IOL implantation were retrospectively recruited. Refractive prediction errors (RPE)were calculated for eight formulas: Haigis-L, Shammas, Barrett True-K, Camellin Calossi Camellin (CCC), Kane, EVO, Pearl-DGS, and Jin-AI. The study compared RPE, mean absolute error (MAE), median absolute error (MedAE), and percentages of eyes achieving prediction errors within ± 0.25, ± 0.50, ± 0.75, ± 1.0, and ± 2.0 diopters.Results: This retrospective study analyzed 34 eyes from 21 patients with prior RK surgery. The superior performing IOL power calculation formulas were Jin-AI (MAE = 0.64 D, MedAE = 0.49 D), Barrett True-K (MAE = 0.58 D, MedAE = 0.535 D), and EVO (MAE=0.59 D, MedAE = 0.538 D). These three formulas demonstrated comparable accuracy, with 47.1%, 44.1%, and 44.1% of eyes achieving prediction errors within ± 0.50 D, respectively. The Haigis-L and Pearl-DGS formulas exhibited significantly higher prediction errors in eyes with more than 12 RK incisions.Conclusion: The Jin-AI, Barrett True-K, and EVO formulas demonstrated superior accuracy for IOL power calculation in Chinese cataract patients with prior RK. The Jin-AI formula, a novel artificial intelligence-based formula derived from Chinese individuals, presents a promising option for post-RK IOL calculation, while external validation studies across diverse ethnic groups are essential.Keywords: radial keratotomy, cataract, IOL power calculation, refractive prediction errors