Comparison of Central Corneal Thickness measured by Four

Comparison of Central Corneal Thickness measured by Four methods in normal and Post-Intra. LASIK eyes Hyun Ju Lee, M. D. , Suk Kyue Choi, M. D. , Do Hyung Lee, M. D. , Ph. D. , Jin Hyoung Kim, M. D. Department of Ophthalmology, Ilsan Paik Hospital, Inje University College of Medicine, Gyeonggi, Korea Authors have no financial interest

PURPOSE ü To compare corneal pachymetry assessment using four measurement methods in normal eyes and Post-Intra. LASIK eyes

METHODS ü Corneal pachymetry assessment using 1. Ultrasonic pachymetry (Advent, Mentor O&O, Norwell, MA, USA ) - Default velocity : 1641 (m/s) 2. Scanning slit topography (Orbscan II; Bausch & Lomb, Rochester, NY) - Acoustic equivalent correction factor : 0. 95 3. Rotating Scheimpflug Camera (Pentacam; Oculus Inc, Wetzlar, Germany) 4. Dual Scheimpflug Analyzer (GALILEI™; Ziemer)

METHODS ü Group I - 30 unoperated eyes (Unoperated group) ü Group II - 30 eyes : shorter than 6 months after Intra. LASIK (Early postoperative Intra. LASIK group) ü Group III - 30 eyes : longer than 7 months after Intra. LASIK (Late postoperative Intra. LASIK group)

RESULTS Corneal Thickness Measurement Mean±SD (um) Group Galilei Pentacam Ultrasound Orbscan P Value I 554± 30. 3 544± 30. 7 541± 31. 3 535± 40. 7 0. 202 II 488± 30. 1 479± 28. 6 475± 32. 1 440± 40. 3 . 000 III 483± 32. 9 472± 33. 3 470± 33. 3 432± 51. 7 . 000 One-Way ANOVA ü Group I, the four measurements were similar with a mean difference within 18. 4 μm. (P=. 202) ü Group II & III, Orbscan were thinner than other mesurements. (P=. 000)

RESULTS Corneal Thickness Measurement Group and Measurement Group III U-G U-P U-O Mean differnece (um) 95% Limit of Agreement P Value Pearson Correlation (R) (p < 0. 001) -12. 3 -2. 7 6. 1 -13. 2 -4. 1 34. 4 -12. 8 -2. 4 37. 8 -34. 9, 10. 2 -25. 3, 19. 9 -16. 51, 28. 62 -35. 5, 9. 0 -26. 4, 18. 2 12. 1, 56. 7 -38. 9, 13. 2 -28. 5, 23. 6 11. 7, 63. 8 0. 487 0. 989 0. 897 0. 412 0. 964 0. 001 0. 573 0. 995 0. 001 0. 982 0. 984 0. 932 0. 983 0. 980 0. 903 0. 972 0. 977 0. 918 G = Galilei, P = Pentacam, U = Ultrasound, O = Orbscan II ANOVA, Scheffe multiple comparison test ü Group II & III, Compared to the ultrasound measurement, orbscan significantly underestimated the corneal thickness. (P=. 001)

RESULTS Corneal Thickness Measurement Group and Measurement Group U-O Mean differnece (um) 95% Limit of Agreement P Value I - II -28. 4 -41. 2, -15. 5 . 000 I - III -31. 7 -44. 6, 18. 8 . 000 II - III -3. 3 -16. 2, 9. 5 0. 812 U = Ultrasound, O = Orbscan II ANOVA, Scheffe multiple comparison test

Bland-Altman plots comparing the 4 modalities (Group I) ü Group I, all four methods showed good agreement, with a mean difference being within ± 20 um.

Bland-Altman plots comparing the 4 modalities (Group II) ü Group II, galilei and pentacam showed good agreement with ultrasound, with a mean difference being within ± 10 um. ü Group II, the limits of agreement between orbsan and ultrasound were 0. 5 um and – 69. 3 um with a range of 69. 8 um.

Bland-Altman plots comparing the 4 modalities (Group III) ü Group III, galilei and pentacam showed good agreement with ultrasound, with a mean difference being within ± 10 um. ü Group III, the limits of agreement between orbsan and ultrasound were 11. 1 um and – 86. 5 um with a range of 97. 6 um.

RESULTS ü Group I (CCT) - Similar for all four methods (P=. 202) ü Group II, III (CCT) - No statistically significant difference among Galilei, Pentacam and Ultrasound - Orbscan measurements were thinner than other mesurements ü After Intra. LASIK, Galilei & Pentacam was comparable to Ultrasound(current gold standard for corneal pachymetry) ü Orbscan measurement of CCT after refractive surgery are less than those measured by US

DISCUSSION ü CCT using Galilei and Pentacam before and after Intra. LASIK is also suitable ü CCT using orbscan tends to be underestimated compared with other values after refractive surgery (Group II & III ) ü Further studies are needed to determined which instrument is more accurate in measuring CCT