To report the accuracy of intraocular lens (IOL) formulas according to axial length, anterior chamber depth, and mean corneal curvature when performing biometry with an immersion type A-scan with mannual keratomery and an IOL Master®.

Retrospective medical chart reviews were carried out for 82 eyes of 65 patients who underwent cataract surgery performed by a single surgeon. Biometry was performed using IOL Master®, mannual keratometry, and immersion type A-scan ultrasound in sequence. Prediction diopter was obtained using Sanders-Retzlaff-Kraff/Theoretical (SRK-T) and Holladay 1 formulas calculated with the biometric value measured by mannual keratomery and A-scan, and using SRK-T and, Holladay 2 formulas with IOL Master®. The final refractive outcome was determined as manifested refraction at least 7 weeks after the surgery, and it was compared with the preoperative prediction dipoter (D) of the IOL formulas.

Mean axial length and mean keratomtric measurements as determined by A-scan with mannual keratomery showed significant statistical differences from those of IOL Master®. However, there was no difference in postoperative mean absolute error between biometric measurements, or among formulas according to axial length, anterior chamber depth, or mean corneal curvature. However, the percentage of actual refraction within ±0.50 D of the intended refraction was dirrerent among the four formalas according to axial length, anterior chamber dept, mean corneal curvature.

Biometry measurement using the immersion-type A-scan with mannual keratomery is as accurate as that using IOL Master® for predicting the postoperative refractive state of cataract surgery. However, it is suggested that the best IOL formula be chosen according to axial length, anterior chamber depth, and mean corneal curvature.