Light microscopic evaluation and scanning electron microscopic analysis of horse eyes following deep anterior lamellar keratectomy.

OBJECTIVE  To describe the technique of deep anterior lamellar keratoplasty (DALK) with Descemet's membrane (DM) exposure in horse eyes. Also, to compare the efficacy and safety of viscodissection and big-bubble techniques for DALK. ANIMALS STUDIED  Thirty-four ex vivo horse eyes. PROCEDURE  Deep anterior lamellar keratoplasty was performed in 34 ex vivo horse eyes. Two groups (Group V--viscodissection--2% sodium hyaluronate; Group A--air--big-bubble) of 17 eyes were studied. Other than the substance used, the surgical technique was similar for both groups. Nonperforated eyes were submitted for light microscopic histologic evaluation and scanning electron microscopic (SEM) analysis. RESULTS  Group V--Perforations occurred in 18% of the eyes during surgery. Light microscopy revealed exposure of DM in 28% of the eyes with mean thickness of the remaining stroma being 70.4 μm. Group A--Perforations occurred in 42% of the eyes. Light microscopy revealed exposure of DM in 60% of the eyes with mean thickness of the remaining stroma being 23.3 μm. No significant differences in safety, efficacy and thickness of the remaining stroma (including all eyes or excluding those with DM exposure) were observed. SEM of the surgical site revealed a more even surface in those eyes with DM exposure compared to eyes with thicker remaining stroma in both groups. CONCLUSIONs  We describe two DALK techniques (viscodissection and big-bubble) for use in horses. No significant differences in safety, efficacy and thickness of the remaining stroma were observed. However, a nonsignificant trend toward the big-bubble technique being more efficacious but less safe was observed.