Comparison of objective and subjective refractive surgery screening parameters between regular and high-resolution Scheimpflug imaging devices

Abstract

Listen

Translate article

To compare objective and subjective metrics from regular and high-resolution Scheimpflug devices (Pentacam) to determine their equivalence and interchangeability for refractive surgery screening. Emory Vision at Emory University, Atlanta, Georgia, USA. Retrospective comparative case series. Eyes of consecutive screened refractive surgery patients were evaluated with high-resolution and regular Scheimpflug devices. Objective parameters evaluated included keratometry (K) values, central corneal thickness (CCT), and device-generated keratoconus screening indices. Masked expert reviewers subjectively graded images as normal, suspicious, or abnormal. One hundred eyes of 50 patients were evaluated. The mean K values were not significantly different (anterior K: high resolution 1.21 diopters [D] ± 1.13 (SD) versus regular 1.15 ± 1.16 D, P = 0.73; posterior K: 0.34 ± 0.23 D versus regular 0.35 ± 0.23 D, P = .67). The mean CCT was significantly thinner in the high-resolution group (514.7 ± 26.6 μm versus 527.6 ± 27.6 μm (P < .0001) with limits of agreement of -12.9 to +17.5 μm. Most keratoconus screening indices were more suspicious with the high-resolution device than with the regular device except the indices of height asymmetry and height deviation. Subjectively, 60% of cases received the same score, high resolution was more suspicious in 28% of cases, and regular was more suspicious in 12% of cases; there was only slight subjective agreement between technologies (κ = 0.26 to 0.31). Regular and high-resolution Scheimpflug imaging devices generated different objective values and significantly different subjective interpretations with poor inter-reviewer agreement. The high-resolution device provided a more conservative overall output. For refractive surgical screening, the 2 devices are not interchangeable. Proprietary or commercial disclosures are listed after the references.