Severity Assessment of Acute Hydrops Due to Recurrent Keratoconus after Penetrating Keratoplasty Using Anterior Segment Optical Coherence Tomography.

Purpose: To compare the clinical characteristics of acute hydrops between eyes with recurrent keratoconus and eyes with treatment-naïve keratoconus. Methods: Six eyes with acute hydrops caused by recurrent keratoconus (recurrent group) and 29 eyes with acute hydrops due to treatment-naïve keratoconus (naïve group) were included. We collected data from anterior-segment optical coherence tomography (AS-OCT), corneal topography and medical records. With AS-OCT, we investigated the location of Descemet's membrane break, the length and frequency of Descemet's membrane detachment, and the speed of corneal thinning before acute hydrops. Results: Descemet's membrane breaks were detectable in 6 cases in the recurrent group (100%) and 7 cases in the naïve group (70%). Distance from the break to the central line was 2-3 mm in the recurrent group, and less than 1 mm in the naïve group (p < .001). Detachment length was significantly greater in the recurrent group than in the naïve group (p = .003). Steep corneal power values tended to increase before acute hydrops in the recurrent group, but the difference between groups was not significant (observation period; 6.03 ± 2.67, 7.36 ± 5.53 years, respectively). Penetrating keratoplasty was required in 5 cases (83%) in the recurrent group, and 8 cases (28%) in the naïve group (p = .02). Conclusions: Acute hydrops due to recurrent keratoconus occurred around graft-host junctions and was more severe than that in eyes with treatment-naïve keratoconus. Severity assessment of Descemet's membrane detachment using AS-OCT provided clues not only for diagnosis, but also for predicting prognosis.