P. Heinz et al., IN-VIVO OBSERVATION OF CORNEAL NERVE REGENERATION AFTER PHOTOREFRACTIVE KERATECTOMY WITH A CONFOCAL VIDEOMICROSCOPE, German journal of ophthalmology, 5(6), 1997, pp. 373-377
The aim of the present study was to study corneal nerve regeneration a
fter photorefractive keratectomy (PRK) in vivo by means of a confocal
videomicroscope. A total of 25 myopic eyes of 16 patients were examine
d biomicroscopically and with the confocal microscope (Microphthal, Hu
nd Company, Wetzlar, Germany) before and after PRK. The mean age of th
e patients was 30.4 (range 21-44) years, the mean follow-up period was
13.7 (11.0-15.1) months, the mean preoperative refraction was -5.5 (-
1.75 to -9.0) D, the mean amount of attempted correction was -4.75 (-1
.75 to -7.5) D, the mean ablation depth was 50 (30-75) mu m, and the s
ingle central ablation zone measured 6.0 mm in diameter. The postopera
tive refractive and visual outcome wag favorable in most eyes The mean
postoperative refraction was -0.5 (-2.75 to +0.5) D. Among the 25 eye
s, 22 (88%) had +/- 1.0 D of attempted correction at 1 year and only 2
eyes showed a loss of 2 Snellen lines. Despite increased light scatte
ring and reflection due to haze, sufficient postoperative imaging of t
he corneal nerve architecture was possible in all cases. Recovery of c
orneal innervation started from the margin of the ablation, being dire
cted toward the center of the cornea and the epithelium. At 5-8 months
following PRK, stromal nerves and the subepithelial nerve plexus took
on their definite appearance. The original nerve structure was not re
established, abnormal branching and accessory thin nerve fibers being
present without exception. Confocal slit-scanning videomicroscopy is a
very useful technique for in vivo investigation of corneal nerve rege
neration after PRK and, presumably, other keratorefractive procedures.
In comparison with histology studies in animals, reinnervation of the
human cornea after PRK seems to be completed earlier but follows the
same morphologic principles.