C constant: New concept for ray tracing-assisted intraocular lens power calculation

PURPOSE: To evaluate the accuracy of the C constant for ray tracing-assisted intraocular lens (IOL) power calculation. DESIGN: Case series. SETTING: Public university hospital and private clinic. METHODS: Preoperatively, all intraocular distances were measured using laser biometry. Various IOL designs were studied; powers ranged from -5.0 diopters (D) to +38.0 D. The IOL power calculation was performed with the Olsen formula using the C constant and compared with the Haigis, Hoffer Q, Holladay 1, and the SRK/T formulas on optimized datasets. Outcome measures were the error of the prediction, expressed as the arithmetic error, and the absolute error between the observed refraction and the predicted refraction. RESULTS: Two thousand forty-three cases from the 2 centers were studied. No significant differences were found between the Haigis, Hoffer Q, Holladay 1, and SRK/T formulas with the exception of the SRK/T formula, which performed better than the other thin-lens formulas in eyes with an axial length (AL) greater than 27.0 mm (P<.01). Compared with the SRK/T formula, the Olsen formula showed an improvement of 15% and 14% in the mean absolute error and a 39% and 85% reduction in the number of large errors (>1.0 D) for the 2 series, respectively (P<.0001). Contrary to the Olsen formula, all thin-lens formulas showed a significant bias in terms of the AL, keratometry reading, and anterior segment length (P<.0001). CONCLUSION: The C constant is a promising concept for ray tracing-assisted IOL power calculation. (C) 2014 ASCRS and ESCRS