We studied the epithelial healing of denervated corneas in New Zealand
albino rabbits with their left trigeminal ganglia surgically amputate
d. On the 14th day after amputation, the corneas were keratectomized (
in 8.5 mm diameter) and documentation of the healing process began. We
calculated the epithelial healing rate using simple regression analys
is. We observed a mean healing rate of 0.463 +/- 0.059 mm(2)/hr (mean
+/- SE) in the denervated corneas, compared to 0.609 +/- 0.031 mm(2)/h
r in the control corneas; a statistically significant difference of P
< 0.001. We performed scanning electron microscopic observation (SEM)
at three points; before keratectomy, 48 hrs after keratectomy, and 14
days after keratectomy. SEM observation revealed that, in contrast to
the control corneas, the surface of the epithelial cells in denervated
corneas appeared rough with numerous exfoliating cells observed. This
indicates that the epithelial cells might attach only weakly to the f
loor in denervated corneas. Transmission electron microscopic observat
ion (TEM) performed at 48 hrs and 14 days after keratectomy also suppo
rts this finding. For example, the intercellular space is widened and
fewer desmosomes are observed in denervated corneas. Using immunohisto
chemistry, the surface of the wound bed was covered with fibronectin i
n a similar fashion to the control. In the late stage, the denervated
corneas demonstrated spontaneous epithelial breakdown with 83% of them
having persistent epithelial defects. Epithelial healing in the contr
ol corneas displayed no abnormal signs. On the 14th day after keratect
omy, these eyes were enucleated for immunohistochemistry using bromode
oxyuridine (Brd U) to observe dividing cells. The number of S-phase ce
lls observed in meridional sections averaged 70.3 +/- 10.7 in the dene
rvated corneas, compared to 29.1 +/- 2.30 in controls. We hypothesize
that this increase in denervated corneas occurs in order to compensate
for the excessive desquamation.