Clinical comparison of biometry using the non-contact optical low coherence reflectometer (Lenstar LS 900) and contact ultrasound biometer (Tomey AL-3000) in cataract eyes.

摘要

A-scan ultrasound biometers enable measurements of axial length (AL), anterior chamber depth (ACD), lens thickness (LT) and usually also central corneal thickness (CCT). The new Haag-Streit Lenstar LS 900 measures all of the above parameters and additionally retinal thickness in the optical line utilising optical low coherence reflec-tometry (OLCR) technology in a non-contact fashion. Furthermore, the corneal radius in the steep and flat meridian can be determined. In a prospective, non-randomised study, we evaluated the reproduc-ibility and correlation of biometry using the Lenstar LS 900 compared with contact A-scan ultrasound using the AL-3000 in 83 cataract eyes of 54 subjects (mean age 73.9 years, SD 11.2). Of these 83 eyes, 58 eyes were classified as having nuclear (NUC) cataract grade 1, 17 presented with grade 2, and eight cataracts were classified as grade 3. AL measurements were possible in 68 (82%) eyes with the Lenstar versus 82 (99%) eyes with the AL-3000. The automatic analysis of the Lenstar was not possible to provide reliable AL values in three eyes with NUC 1, in five eyes diagnosed with NUC 2 and in seven eyes with NUC 3, whereas the same was found in one eye with NUC 3 for the AL-3000. Except for one patient (NUC 2) in all other unmeasurable eyes, supcapsular opaci-fication was present to some extent. ACD values were measured with the Lenstar in 81 eyes (vs 82 eyes with AL-3000); LT in 65 eyes (vs 77 eyes) and CCT in 82 eyes (vs 82 eyes). The mean AL for all eyes that could be measured by both devices was 23.84 +-1.9 lmm using the Lenstar and 23.55+-1.85 mm with the AL-3000 (p<0.0001). No significant differences between both devices could be shown for the mean ACD and LT (p>0.05 each) whereas it was shown to be significant for CCT (p=0.0048) (table 1, figure 1). Correlation analysis revealed a highly significant correlation for all parameters between the two devices (table 1). Evaluating the means of standard deviations (SD) for the measurements of each eye and their SDs to assess reproducibility of measurements, a significant higher reproducibility was found for the Lenstar compared with AL-3000 regarding AL, ACD and CCT (p<=0.001, each), whereas no significant difference in repeat accuracy was found for LT (p=0.14).