Graft central thickness measurement by rotating Scheimpflug camera and ultrasound pachymetry after penetrating keratoplasty

PURPOSE: To assess agreement between rotating Scheimpflug camera and ultrasound pachymetry in measuring graft central thickness, and compare reproducibility/repeatability of these methods in corneal grafts and normal corneas. DESIGN: Experimental study. PARTICIPANTS: Sixty-five patients with corneal grafts after penetrating keratoplasty and 20 controls with normal corneas (1 eye per patient). METHODS: In 45 eyes with clear grafts after penetrating keratoplasty, graft central thickness measurements were compared between the 2 methods (examiner 1). In another 20 eyes with clear grafts after penetrating keratoplasty and in 20 normal corneas, 2 independent examiners (1 and 2) each employed both methods in a first session to assess interexaminer reproducibility; measurements were then repeated by examiner 1 alone in a second session, and differences with his first session measurements used to assess intraexaminer repeatability. Paired t test, intraclass correlation coefficient (ICC) and 95% limits of agreement (95% LoA) were calculated to assess differences, correlation, and variability of methods, examiners, and first-second measurements. MAIN OUTCOME MEASURES: Graft central thickness measurements by 2 methods. Difference of measurements by 2 examiners; and difference of first-second measurements by 1 examiner, in corneal grafts and normal corneas with both methods. RESULTS: Mean graft central thickness measurement was 556.9+/-41.8 microm with the rotating Scheimpflug camera and 561.8+/-40.8 with ultrasound pachymetry (P = 0.012). There was a significant linear correlation in graft central thickness measurement between the 2 methods (r = 0.93; P<0.001) and 95% LoAs were -34 to +23.4 microm. Interexaminer and intraexaminer correlations were high with both methods: ICCs were > or = 0.94 in corneal grafts and > or = 0.98 in normal corneas. Interexaminer and intraexaminer variability was slightly higher with the rotating Scheimpflug camera than with ultrasound pachymetry, and in corneal grafts than in normal corneas. CONCLUSIONS: Measurements of graft central thickness with the rotating Scheimpflug camera, although slightly lower, were comparable to those with ultrasound pachymetry. The reproducibility and repeatability of these methods in corneal grafts are only slightly lower than in normal corneas.