In vivo confocal microscopy for the evaluation of corneal microstructure in keratoconus.

PURPOSE: To evaluate the corneal microstructure in keratoconus with in vivo confocal microscopy (IVCM).

METHODS

Unscarred corneas of 68 patients with keratoconus and 22 controls were evaluated with slit lamp examination, corneal topography, and IVCM (Confoscan 3.0, Vigonza, Italy). One eye was randomly chosen for analysis. Keratocyte, endothelial cell and basal epithelial densities, sub-basal and stromal nerve structure, and severity of stromal haze were evaluated.

RESULTS

Compared with corneas of control subjects, patients with keratoconus had a significantly lower anterior stromal keratocyte density (1279+/-197 cells/mm(2) vs. 1132+/-178 cells/mm(2), p=0.002), lower midstromal keratocyte density (904+/-213 cells/mm(2) vs. 770+/-120 cells/mm(2), p<0.001), lower posterior stromal keratocyte density (935+/-113 cells/mm(2) vs. 725+/-113 cells/mm(2), p<0.001), lower endothelial cell density (2924+/-300 cells/mm(2) vs. 2719+/-279 cells/mm(2), p=0.004), lower basal epithelial cell density (5987+/-699 cells/mm(2) vs. 4365+/-537 cells/mm(2), p<0.001), lower sub-basal long nerve density (32.0+/-6.5 nerves/mm(2) vs. 19.6+/-6.5 nerves/mm(2), p<0.001), thicker sub-basal (3.2+/-0.4 microm vs. 3.7+/-1.1 microm, p=0.01) and stromal nerves (5.0+/-1.2 microm vs. 8.0+/-2.9 microm, p<0.001), and higher proportion of corneas with haze (40.9% vs. 92.6%, p<0.001). A history of contact lens use (n=12) was not associated with lower keratocyte, endothelial cell, or basal epithelial cell counts.

CONCLUSIONS

Corneal microstructure is abnormal in patients with keratoconus. Keratocyte and endothelial cell loss appears to be present in keratoconic corneas.