PURPOSE: Intraocular lens (IOL) formulas based on aphakic refraction alone have been proposed as an alternative to biometry-based formulas. The purpose of this study was to determine the relative accuracy of these simplified formulas. METHODS: Records of patients who received secondary in-the-bag (IOL) implants and for whom aphakic refraction was obtainable were retrospectively studied. The formulas for IOL power calculation by Hug (J Pediatr Ophthalmol Strabismus 2004;41;209-11) and Khan and by AlGaeed (Br J Ophthalmol 2006;90:1458-60; modified to account for greater K values in our population) were compared with the biometry-based Holladay 1 formula by calculating the absolute prediction error for each. RESULTS: Twenty eyes met the inclusion criteria. The average age at IOL implantation was 4.8 ± 3.4 years. The mean of the absolute value of the prediction error when we used the Hug's formula was 2.4 ± 2.1 D; with Khan's formula, 2.4 ± 2.0 D; and with biometry, 1.6 ± 1.4 D. None of these differences was statistically significant (paired t test: Hug vs Khan, P = 0.9; biometry vs Hug, P = 0.3; biometry vs Khan, P = 0.2). The absolute prediction error was more than 2 D in 8 of 20 eyes (40%) when we used Hug's formula, 11 of 20 eyes (55%) with Khan's formula, and 5 of 20 eyes (25%) with biometry. CONCLUSIONS: Although the 0.8 D reduction in accuracy of the refraction-based formulas was not statistically significant, we recommend continued use of biometry-based formulas for IOL power calculation. Aphakic refraction may be used to confirm IOL power prediction when biometry is difficult.
PURPOSE: Intraocular lens (IOL) formulas based on aphakic refraction alone have been proposed as an alternative to biometry-based formulas. The purpose of this study was to determine the relative accuracy of these simplified formulas. METHODS: Records of patients who received secondary in-the-bag (IOL) implants and for whom aphakic refraction was obtainable were retrospectively studied. The formulas for IOL power calculation by Hug (J Pediatr Ophthalmol Strabismus 2004;41;209-11) and Khan and by AlGaeed (Br J Ophthalmol 2006;90:1458-60; modified to account for greater K values in our population) were compared with the biometry-based Holladay 1 formula by calculating the absolute prediction error for each. RESULTS: Twenty eyes met the inclusion criteria. The average age at IOL implantation was 4.8 ± 3.4 years. The mean of the absolute value of the prediction error when we used the Hug's formula was 2.4 ± 2.1 D; with Khan's formula, 2.4 ± 2.0 D; and with biometry, 1.6 ± 1.4 D. None of these differences was statistically significant (paired t test: Hug vs Khan, P = 0.9; biometry vs Hug, P = 0.3; biometry vs Khan, P = 0.2). The absolute prediction error was more than 2 D in 8 of 20 eyes (40%) when we used Hug's formula, 11 of 20 eyes (55%) with Khan's formula, and 5 of 20 eyes (25%) with biometry. CONCLUSIONS: Although the 0.8 D reduction in accuracy of the refraction-based formulas was not statistically significant, we recommend continued use of biometry-based formulas for IOL power calculation. Aphakic refraction may be used to confirm IOL power prediction when biometry is difficult.