Perimetry software measures functional progression of glaucoma using trend, event analyses

Key Points

"And it builds in [the] concept of needing to confirm progression not only on just a single field but on two fields," he added.

Dr. Budenz, professor of ophthalmology, Bascom Palmer Eye Institute, Miller School of Medicine, University of Miami, advised clinical correlation, however.

Use of the software requires a particular, newer perimeter (Humphrey Field Analyzer II-i, Carl Zeiss Meditec), Dr. Budenz said. The program "adjusts for the reduced hill of vision in many of our elderly patients who develop cataract, and it works with the old full threshold fields from the early and mid 1990s and the newer [Swedish interactive threshold algorithm] fields from the late 1990s to the present," he said.

EMGT criteria

The new software was developed based on parameters for change used in the Early Manifest Glaucoma Treatment (EMGT) study, Dr. Budenz said.

"In the EMGT trial, a patient would have reached a visual field endpoint if he or she had deterioration of three or more test points at a statistically significant level at the exact same location on three consecutive visual fields," he said.

The new software represents as dots the individual visual field test points that have not changed from the previous test. The first time an individual point shows deterioration from two baseline fields, the software will display an open white triangle at that point.

"Certainly, most of our patients, we hope, are not [having progression] on visual field, so we may see an occasional open white triangle that goes away on a repeat test, and we have to use our heads here and not overinterpret occasional white triangles," Dr. Budenz said. "If there aren't three of them in an expected area of the visual field, we can probably wait until the next routine visit to get visual field tests."

If the next visual field test performed confirms the deterioration of that individual point, then the software will display a partially darkened triangle, and if another visual field test is performed and confirms that the point has deterioration, the software will display a fully darkened triangle.

"But one actually needs three or more of these points to have changed in the exact same location on three consecutive tests to have a message of 'likely progression' in the software," he said. "Intermediately, one sees 'possible progression,' which should prompt you to get another confirmatory field."

Initial values too low to be analyzed are indicated with an X on the printout, he said.

Trend analysis

The latest version of this software includes the ability to perform trend analysis in addition to performing the event analysis of which its predecessor was capable, Dr. Budenz said.

"What the new trend analysis does is allow us to judge the rate of visual field progression as well as the endpoint criteria," he added.

Instead of using mean deviation, trend analysis plots a visual field index (VFI) against a patient's age, indicated on the X axis instead of time, and shows the rate of change thus far. It also predicts the visual field loss that would be expected to occur in the future if the patient's disease course is not changed. The VFI, which weighs central points more than peripheral points, ranges from 0% to 100%, with 100% indicating a normal visual field.

The new software determines change at specific points by comparing information from the pattern deviation plot instead of information from the total deviation plot as used in glaucoma change probability software. In theory, Dr. Budenz said, this approach eliminates the confounding factor of change caused by diffuse depression of the visual field, which, for instance, can occur in eyes with cataract.

The resulting graph also can help explain actual and possible progression to patients, Dr. Budenz said.