Multicenter biometry study of 1 pair of eyes

Purpose: To assess the accuracy of biometry for intraocular lens (IOL) power calculation. Setting: Six ophthalmic surgery centers in Europe and 1 in the United States. Methods: Biometry was done as if in preparation for cataract surgery in 2 eyes of the same person with 10 combinations of operator and instrument. Data analysis followed a standardized procedure to assess repeatability (within-center variability) and reproducibility (between-center variability) of the test methods. Results: The reproducibility of the corneal radius measurement was 0.06 mm. The use of different keratometric indices made conversion to K-values less reliable. The repeatability of the axial length (AL) measurement of 0.30 mm and the reproducibility of 0.66 mm converted to a calculated IOL power of 0.75 diopter (D) and 1.65 D, respectively. Thus, this potential patient runs the risk of a refractive surprise of up to 1.10 D purely as a result of a measurement error within a center and up to 2.30 D if the patient goes to the worst-case center. Conclusions: The biometry results show that the measured corneal radius can be used with confidence. Reproducibility errors in AL determination require personalization of formula constants or correction at the source by proper calibration. (C) 2001 ASCRS and ESCRS.