Abstract
Purpose: :
To compare central corneal thickness (CCT) measurements assessed with contact ultrasound pachimetry (US) and non-contact pachimetry devices: specular microscopy (NCSM) and three spectral-domain anterior segment OCT (AS-OCT) in healthy eyes.
Methods: :
This prospective observational cross-sectional study included 30 healty eyes of 15 subjects ( mean age 28.6±5.84 years, range 24-45). All subjects undergone CCT measurements with NCSM (EM-3000-Tomey) ultrasound pachimeter (Pacline-Optikon) and Spectralis SD-OCT with anterior segment lens (Heidelberg Engineering), Cirrus SD-OCT (Zeiss) and RTVUE OCT with anterior segment lens CAM-L (Optvue). Measurements were repeated three times with each device by two different masked operators during the same session. Differences between CCT values recorded by each non-contact instrument and US pachimetry were calculated using the paired t-test. Agreement between non-contact measurements and US pachimetry was evaluated using the Bland-Altman method and the concordance correlation coefficient (CCC). Inter-observer agreement for all instruments was assessed using the intraclass correlation coefficient (ICC).
Results: :
Mean CCT measurements with NCSM, Spectralis, Cirrus, RTVUE and US pachimetry were respectively 515.68±30.85 µm, 519.78±28.47 µm 515±26.84 µm, 519.08±27.88 µm and 536.78±46.91µm. CCT measured by US pachymetry was statistically significantly higher than by NCSM and AS-OCTs (p<0,001). The mean of the differences between non-contact devices (NCSM, AS-OCTs Spectralis, Cirrus and RTVUE) and contact pachimetry (US) were respectively 21.1±12.25 µm, 17.01±8.6 µm, 21.78±7.46 µm and 17.7±6.3 µm. CCC between non-contact devices (NCSM, AS-OCTs Spectralis, Cirrus, and RTVUE) and US contact pachimetry were respectively 0.72 (range 0.67-0.77), 0.81 (0.76-0.84), 0.73 (0.68-0.77) and 0.81 (0.77-0.84). ICC for inter-observer agreement was 0.98 for the NCSM (range 0.97-0.99), 0.99 for Spectralis AS-OCT (0.98-0.99), 0.97 for Cirrus (0.96-0.98), 0.99 for RTVUE (0.996-0.998) and 0.97 for US pachimetry (0.95-0.98).
Conclusions: :
There was a statistically significant difference between US pachymetry and non-contact devices, with US pachymetry measurements being consistently thicker. There was a good inter-method agreement between contact and non-contact measurements. All the instruments showed an excellent inter-observer agreement.
Keywords: cornea: clinical science • anterior segment • imaging/image analysis: clinical