Confocal microscopy measures corneal nerve density

October 1, 2005

Fort Lauderdale, FL—Confocal microscopy allows corneal nerve density to be measured in vivo, which has been especially helpful in identifying dramatic decreases in the nerve densities after PRK and LASIK. Using this technology, investigators can measure the increases in the densities over time following the surgeries. There are, however, artifacts that may interfere with microscopy results that must be controlled, according to William Bourne, MD.

He described the advances in confocal microscopy at the annual meeting of the Association for Research in Vision and Ophthalmology.

Dr. Bourne, professor of ophthalmology, Mayo Clinic College of Medicine, Rochester, MN, warned that corneal morphology is not the same as function. In most patients, he noted, the sensitivity or the function has not been correlated with the corneal morphology, although a correlation may be assumed to exist.

The clinical confocal microscope allows visualization of the nerve fiber bundles; in most cases, the individual nerve fibers cannot be visualized. The nerve fiber bundles tend to lie in one plane and can be seen easily with the clinical confocal microscope (Tandem Scanning microscope), he said.

Another device (Confoscan 3, Nidek), which is commercially available, is advantageous in that it allows visualization of nerves and allows more contrast than the Tandem Scanning microscope. The depth of the section is about 12 μm deeper with the Confoscan 3 microscope. Dr. Bourne demonstrated that the instrument allows views of the nuclei of anterior keratocytes in the same section with the nerves. He explained that there is 16 μm of Bowman's membrane between the keratocytes and the nerves, indicating that the section is thicker than with the Tandem Scanning microscope, which provides a section of about 12 μm.

The Confoscan 3 cannot be used to obtain thickness measurements because it is positioned farther from the cornea and there is a great deal of anterior and posterior movement. The Confoscan 4, the newest-generation device, may solve some problems of thickness measurements encountered with the previous generations.

Using the confocal scanning microscope, Dr. Bourne and colleagues measured the length of the subbasal nerves per area of field. Between three and six scans of the central cornea, covering slightly different areas of the cornea, will yield distinct samples and allow estimations of the nerve density.

Dr. Bourne described two clinical research studies using these technologies. In the first, a 5-year longitudinal study, he and his colleagues studied the effect of PRK and LASIK on corneal morphology by comparing the corneal nerves preoperatively and postoperatively.

"With PRK," he explained, "the epithelium is removed and Bowman's layer and the anterior stroma are ablated. After the procedure, the epithelium and the nerves heal over the stromal surface. In LASIK, the stroma is ablated behind the flap, leaving Bowman's layer intact. Most of the nerves are cut when the flap is created, and they must eventually return anterior to Bowman's layer," Dr. Bourne explained.

Dr. Bourne and his colleagues found that after PRK, the nerve densities decrease and gradually increase; by 2 years after PRK there is no significant difference in nerve density from preoperatively. After LASIK, the nerve density also decreases and returns much more slowly; 3 years after surgery, there is still a significant difference from the preoperative density, but by 5 years after surgery there is no significant difference from preoperatively.

The second study, which was part of the Ocular Hypertension Treatment Study, also was conducted at the Mayo Clinic and evaluated the effect of glaucoma medications on corneal nerves. Patients with ocular hypertension either received medications or were observed for 6 years. Dr. Bourne and coworkers examined the 38 of 54 patients who were studied at the Mayo Clinic using confocal microscopy 6 to 9 years after the start of the study. Of the 38, 19 patients had received various medications for a mean of 8 years. Nine patients were observed and switched to treatment at the end of 6 years when they were offered treatment or they received treatment if glaucoma progressed; 10 patients were only observed.

"The nerve density was about 40% less in the treated patients compared with those who were observed, a difference that was statistically significant," Dr. Bourne said.