Comparison of a new optical biometer using swept-source optical coherence tomography and a biometer using optical low-coherence reflectometry

PURPOSE: To investigate agreement between the IOLMaster 700 based on swept-source optical coherence tomography (SS-OCT) and the Lenstar LS 900 based on optical low coherence reflectometry (OLCR). SETTING: Private practice, Castrop-Rouxelle, Germany. DESIGN: Prospective evaluation of diagnostic test. METHODS: Axial length (AL), keratometry (K), anterior chamber depth (corneal epithelium to lens) (ACD), lens thickness, and central corneal thickness (CCT) were measured in 183 eyes of 183 patients. Analyses used a paired t test, Pearson correlation coefficient (r), and Bland-Altman plots. Intraocular lens (100 power calculations were compared using the Hoffer Q, Holladay 1, and SRKIT formulas. RESULTS: The difference in the mean AL between SS-OCT and OLCR was statistically significant but clinically insignificant (23.61 mm +/- 1.27 [SD] and 23.60 +/- 1.27 mm, respectively; P<.0001); the agreement and correlation were excellent. The mean K with OLCR was flatter by 0.02 diopter (D) than the mean K with SS-OCT, 43.82 +/- 1.43 diopters (D) and 43.84 +/- 1.43 D, respectively; the difference was not statistically significant. The mean ACD with SS-OCT was deeper by 0.03 mm than the mean ACD with OLCR, 3.22 +/- 0.44 mm and 3.19 +/- 0.44 mm, respectively; the difference was statistically significant (P <.001) but clinically insignificant. The differences in the mean lens thickness (OLCR 4.63 +/- 0.44 mm and SS-OCT 4.59 +/- 0.43 mm) and the mean CCT (OLCR 559 +/- 37 mu m and SS-OCT 554 +/- 36 mu m) were not statistically significant. There was no statistically significant difference between the median absolute errors of the 3 formulas. CONCLUSIONS: Agreement between SS-OCT and OLCR was very good. The clinically insignificant but statistically significant differences in AL, ACD, and lens thickness did not reflect a statistically significant difference in IOL power calculation using 3 third-generation formulas. (C) 2016 ASCRS and ESCRS