AIM: To study the reliability of Pentacam, OCT, Tomey OA-2000, IOL Master700, and A ultrasound pachymetry in the measurement of central corneal thickness(CCT)in myopic eyes.

METHODS: CCTs of 56 myopic patients(112 eyes)were measured using Pentacam, RTVue OCT, Tomey OA-2000, IOL Master 700, and A ultrasound pachymetry(NIDEK US-500). The difference, correlation, and consistency were statistically analyzed using average values taken from the five devices and the measured values produced by each device.

RESULTS: CCTs of 112 eyes were determined using Pentacam(530.17±25.08)μm, OCT(519.79±26.90)μm, Tomey OA-2000(521.75±26.51)μm, IOL Master 700(519.53±28.15)μm, and A ultrasound pachymetry(542.23±26.88)μm. The average value across all five devices was 526.69±26.08μm. The results of root mean square error(RMSE)and theil inequality coefficient(TIC)showed that the degree of deviation from the average of the values measured using Pentacam was the smallest. The other four devices were, from least deviation to most, Tomey OA-2000, OCT, IOL Master 700, and A ultrasound pachymetry. Typical correlation analysis showed that the correlation coefficients between the measured values of Pentacam, OCT, Tomey OA-2000, IOL Master 700, and A ultrasound pachymetry and the average values were ρ=0.957, P=0.950, P=0.953, P=0.930, and ρ=0.949, respectively. The results of correlation coefficient showed that Pentacam and Tomey OA-2000 had the closest correlation with the average value. The other three were, in order of closest to furthest correlation, OCT, A ultrasound pachymetry and IOL master 700. The results also showed that the measurements of Pentacam and Tomey OA-2000 were more reliable, which was consistent with the clinical situation. The results of Bland-Altman analysis showed that the five groups of measurements were closely consistent with the average values, and the CCT measurements of Pentacam and Tomey OA-2000 were the most consistent with the average values.

CONCLUSION: Among the five instruments, Pentacam and Tomey OA-2000 produced results that were closest to the average value in CCT measurement, with the best correlation and consistency. Their measurement methods are more objective and safer than A-scan, so they may provide better reference data for clinical practice.