Evaluation of phakic intraocular lens power calculation using the new Linz-Homburg-Castrop formula and comparison with four conventional methods

Purpose: Evaluating the accuracy of phakic intraocular lens power calculation in a middle European patient cohort. Setting: Eyelaser Clinic, Linz, Austria. Design: Single center, single surgeon retrospective consecutive case series. Method: ology: Inclusion of patients after uneventful phakic IOL surgery implanting 91 non-toric and toric Visian ICL model V4c. OCOS Software, JPhakic Software, Olsen-Feingold formula, Holladay Formula and Linz-Homburg-Castrop (LHC) formula were compared. When possible, lens constants were optimized for the patient cohort. Data of one single eye per patient was included. Outcome measures were: mean absolute prediction error (MAE), median absolute prediction error, mean prediction error with standard deviation and median prediction error, as well as percentage of eyes with an absolute prediction error within limits of 0.25 Dpt, 0.5 Dpt, 0.75 Dpt and 1.0 Dpt. Results: A total of 91 eyes of 91 patients were assessed. After application of the Cochrane Q test, the Olsen-Feingold formula achieved a significantly lower percentage of eyes within an absolute prediction error of 1.0 Dpt than all other methods. Conclusion: In our patient cohort OCOS Software, JPhakic Software, Holladay and LHC formulae showed equal results and can be cross-checked. The LHC formula was not published before. A ready-to-use Excel sheet is available as an addendum.