RESUMO
PurposeTo evaluate and compare the accuracy of modern intraocular lens (IOL) power calculation formulae in pediatric eyes and compare prediction error (PE) obtained with manufacturer's vs personalized lens constant.Patients and methodsAn observational case study was conducted in 117 eyes (117 patients) undergoing pediatric cataract surgery with IOL implantation. PE was calculated as predicted refraction minus actual postoperative refraction, and absolute PE as absolute difference independent of the sign, (APE)=predicted refraction minus actual postoperative refraction. This was done for each formula using manufacturer's and personalized lens constant. Further, PE and APE were evaluated according to axial length (AL).ResultsMean age of children was 2.97 years. About 66/117 eyes (56.4%) were below 2 years of age. Using Holladay 2, Holladay 1, Hoffer Q, and SRK/T formulae with manufacturer's lens constant, mean PE was 0.36, 0.41, 0.69, and 0.28 diopter (D), respectively. With personalized lens constant, it was 0.16, 0.15, 0.50, and -0.12 D, respectively. Difference in mean PE between the formulae was statistically significant (P<0.0001). SRK/T and Holladay 2 formulae had the least PE, both with manufacturer's and personalized constant. For eyes with AL<20 mm, SRK/T and Holladay 2 formulae gave the least PE. Personalizing the lens constant led to a decrease in mean PE in all formulae, except the Hoffer Q formula. However, personalizing the lens constant did not significantly improve the APE. At least 21% eyes had an APE of >2 D with all formulae, even with personalized lens constants.ConclusionIn pediatric eyes, SRK/T and the Holladay 2 formulae had the least PE. Personalizing the lens formula constant did reduce the PE significantly for all formulae except Hoffer Q. In extremely short eyes (AL<20 mm), SRK/T and Holladay 2 formulae gave the best PE.
