Keratocyte density in keratoconus. A confocal microscopy study

PURPOSE: To estimate keratocyte density in human corneas with keratoconus by confocal microscopy. DESIGN: Prospective, observational cohort study. METHODS: Twenty-nine unscarred corneas of 19 patients with keratoconus and 29 corneas of 19 controls matched for age (±3 years) and contact lens wear were examined by using confocal microscopy. Images were recorded from the full-thickness central cornea. A masked observer manually counted bright objects (keratocyte nuclei) in images without motion blur. Cell densities in anteroposterior stromal layers of keratoconus corneas were compared with densities in corresponding layers of control corneas. RESULTS: In keratoconus patients, age 40 ± 15 years (mean ± standard deviation), keratocyte density was 19% lower in those who wore contact lenses (16,894 ± 4032 cell/mm³, n = 12) than in those who did not wear contact lenses (20,827 ± 4934 cell/mm³, n = 17, P = .03). In control patients, age 39 ± 16 years, there was no difference in keratocyte density between those who wore contact lenses (n = 12) and those who did not wear contact lenses (n = 17, P = .80). Among contact lens wearers, keratocyte density was 25% lower in keratoconus corneas (16,894 ± 4, 032 cell/mm³, n = 12 [9 = rigid gas-permeable lenses, 3 = soft lenses]) than in control corneas (22,579 ± 2, 387 cell/mm³, n = 12 [3 = rigid gas-permeable lenses, 9 = soft lenses], P = .002), the result of cell density being lower in the most anterior keratocyte layer (P = .001) and the layers between 0% to 10% (P < .001), 67% to 90% (P < .001), and 91% to 100% (P < .001) of stromal thickness. Among noncontact lens wearers, there was no difference in cell density between keratoconus and controls (P = .41). CONCLUSION: Keratocyte density is decreased in the anterior and posterior stroma of keratoconus patients who wear contact lenses.