Comparison of optical low-coherence reflectometry and applanation ultrasound biometry on intraocular lens power calculation

Background The aim of the study was to determine whether the innovative non-contact optical low-coherence reflectometry method utilized by the Lenstar LS 900® agrees sufficiently with applanation ultrasound A-scan technique in routine biometric measurement and intraocular lens power calculation to replace it. Methods Twenty-two patients hospitalized at our eye clinic undergoing cataract surgery were assigned to have five consecutive measurements of axial length by two examiners in a single session using applanation ultrasound and the Lenstar. The applanation ultrasound intraocular lens power calculation was based on automated keratometry and applanation ultrasound axial length measurements. The Lenstar intraocular lens power calculation was based on its measurement of keratometry and axial length. Bland–Altman analysis was used to assess interobserver repeatability of applanation ultrasound and the Lenstar as well as agreement between the Lenstar and applanation ultrasound for axial length measurement and intraocular lens power calculation. Results Thirty-two eyes of 22 patients were analyzed. In 95% of the observations, predicted refractive error corresponded to –0.26 ± 0.62 D and 0.01 ± 0.20 D obtained with applanation ultrasound and the Lenstar, respectively. Conclusions Based on excellent repeatability of the Lenstar and acceptable repeatability of applanation ultrasound, two techniques may be used interchangeably. The predicted refractive error of ±0.20 D in 95% of the observations has never been achieved. Optical low-coherence reflectometry might become a new standard method for biometric measurement needed for intraocular lens-power calculation in patients with cataract.