Purpose: Marfan syndrome (MFS) is diagnosed based on Ghent nosology, including major and minor criteria such as increased axial length (AXL) and flattened corneal curvature (higher K-values) or myopia of more than -3 diopters (D) in its latest revision. Because corneal flattening may, in part, be caused by AXL increase, it may be helpful to consider K-values separately. We present statistical evaluation of using corneal K-values for identifying MFS.
Methods: A retrospective study of K-values of 74 right eyes of 74 patients with MFS, who were compared with an age- and AXL-matched group of 74 right eyes of 74 patients without MFS. After multivariate analysis, receiver operating characteristic curves were calculated.
Results: Mean age was 44.1 years versus 44.9 years (P = 0.834). Mean AXL was 25.22 mm versus 25.47 mm (P = 0.661). K-values showed significant differences: mean Kmax was 8.25 mm (40.91 D) versus 7.9 mm (42.72 D) (P < 0.001) and mean Kmin was 8.22 mm (41.06 D) versus 7.69 mm (43.89 D) (P < 0.001). The area under the curve in receiver operating characteristic analysis was 0.82 for Kmax and 0.78 for Kmin; the best cutoff was seen at a Kmax of at least 8.16 mm (41.36 D; sensitivity 73% and specificity 81%).
Conclusions: Although both K-values differ significantly, Kmax is the better marker to identify MFS. A Kmax of more than 8.16 mm (41.36 D) seems to be a reasonable cutoff (specificity of 81% and sensitivity of 73%). Because this biometric value is easily obtained and standardized, we see it as a good supporting screening tool for MFS suspects.