Abstract
Purpose :
There is limited evidence about the accuracy of central corneal thickness (CCT) measurements from a new non-contact pachymetry device Tonoref III (NIDEK CO., LTD) compared with standard ultrasound pachymetry. This study aims to examine the accuracy and repeatability of CCT measurements from Tonoref III against two ultrasound pachymetry devices, OcuScan RxP Ophthalmic Ultrasound System (Alcon Laboratories) and SP-100 (Handy; TOMEY, JAPAN).
Methods :
In this cross sectional study, CCT (μm) measurements have been performed on 54 right eyes of 54 patients at the Ophthalmology department of Rethymno General Hospital. Measurements from non-contact Tonoref III (Pach-N) were compared against ultrasound pachymetry CCT measurements of Ocuscan RxP (Pach-O) and SP-100 (Pach-T) during the same visit. Levels of the three CCT measurement methods were compared with repeated measures analysis and with rho-Spearman the relationships with descriptive characteristics of the patients. Differences in pachymetry were classified as follows: Tonoref III vs OcuScan and Tonoref III vs SP-100, and in relation to their subject means were analyzed by paired samples t tests and Bland-Altman plots. The repeatability and reliability of the measurements within groups were assessed by one-way ANOVA and Cronbach alpha respectively.
Results :
29 males (53.7%) and 25 females (46.3%) were included in the study. Intraclass correlation coefficients show an excellent reliability for the three methods (alpha>0.90). Mean CCT (μm) ± stand. dev. (min-max) of Tonoref III, OcuScan RxP and SP-100 were 542.2±28.2 (482-606), 531.2±31.6 (463-604) and 548.2±36.7 (472-665), respectively (p<0.001). No significant relationship with descriptive characteristic of the patients has been found (p>0.05). The mean difference between Tonoref III vs. OcuScan RxP was +10.93μm (95% confidence interval: 8.07, 13.79) (p<0.001) and for Tonoref III vs. SP-100 was -6.06μm (95% confidence interval: -10.63, -1.48) (p=0.010), with repeatability 21 & 23μm, respectively for the two pairs of measurement methods.
Conclusions :
CCT measurements using Tonoref III seems to slightly underestimate CCT compared with SP-100, while overestimating CCT compared with OcuScan RxP measurements. Physicians should consider this difference between devices and its influence in daily clinical practice.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.