spotlight on technology & technique
Beyond the Amsler Grid
An ingenious new instrument may detect CNV far earlier than the traditional test.
By Christopher Kent, Senior Associate Editor

An estimated 8 million people in the United States have intermediate age-related macular degeneration (AMD) and are at risk of developing "wet" AMD. Unfortunately, most of those suffering from wet AMD are diagnosed too late for useful treatment.

Recently, however, Carl Zeiss Meditec Inc. obtained distribution rights for a novel new perimetry device -- called the Preview PHP (Preferential Hyperacuity Perimeter) -- for the early detection and management of AMD. The device, de-veloped by Notal Vision Ltd., in Tel Aviv, Israel, has been shown to be twice as likely to detect the presence of AMD as the Amsler grid.

	The Preview PHP from Carl Zeiss Meditec, Inc.

How the Test Works

Here's what happens during the test: The patient looks at a computer screen, blank except for a fixation dot in the center. A single straight line, consisting of a series of closely spaced dots with one dot out of alignment, is flashed somewhere on the screen for 160 milliseconds. (The human visual system is very good at detecting misalignment, a talent referred to as "hyperacuity." In fact, this ability is 10 times more sensitive than standard visual acuity.) The patient uses a stylus to touch the screen where he experienced a distortion in the line.

A new dot appears on the screen where the patient touched it with the stylus; the new dot then moves across the screen toward the fixation dot in the center, drawing the patient's attention. When the dots merge, another line is flashed at a different location on the screen, and the process repeats.

About 100 lines are flashed, at enough locations to cover the entire macular area.

Bypassing the Grid's Shortcomings

To develop a better method of detecting early-stage AMD, the researchers at Notal Vision pinpointed six specific limitations of the Amsler grid and designed the PHP to bypass those limitations:

Ensuring correct use of the test. Because the PHP test is performed in the doctor's office, there are no concerns that patients would misuse the test at home -- or not use it at all.

Completion. The brain automatically compensates for a certain amount of distortion or missing in-formation by "correcting" or filling in (as it does with the physiological blind spot in our field of vision.) The Amsler grid is easy to remember, so the brain can reflexively compensate for small distortions.

The "crowding" phenomenon: When staring at the cross, multiple sources of information make it far more difficult to identify the letter "R" away from the center of vision.

To avoid this problem, the PHP test flashes lines for a very short time; the brain doesn't have time to "correct" distortions.

Crowding. The human brain perceives details located away from the center of vision more clearly when they're isolated from competing information. (See example, below.) Thus, it's harder to perceive a defect in a line (away from the center of our visual field) if the line is surrounded by other lines.

Because the PHP flashes one line at a time, there's no lateral inhibition of perception; the patient can detect distortions more accurately.

Fixation. Most of us perceive our entire field of vision as sharp and clear. In reality, only the foveal region is really sharp and clear. The eyes compensate for this by scanning the entire area we're looking at; the brain uses the resulting detailed information to present us with a seemingly sharp complete picture.

Fixation is hard to maintain because we instinctively want to scan to fill in detail -- especially when we perceive a distortion. With the Amsler grid, this can result in a mental image mapped out by the fovea that doesn't reveal distortions in other parts of the visual field.

Because the PHP test lines are flashed so briefly, the eye doesn't have a chance to fixate on them to scan them with the fovea. Also, when the new dot appears (after the patient touches the screen) and moves back to the center, the eye instinctively follows it. This ensures that the eye is fixated at the center point when the next line is flashed.

Impact of visual deficits. Hyperacuity is nearly unaffected by color, contrast sensitivity, dark adaptation or patient age; even media opacities such as cataracts don't undercut this ability, unless vision is seriously curtailed.

Differentiating between lesions of different size. Because CNV generally causes greater retinal pigment epithelium (RPE) elevation than drusen (the debris found in dry AMD), it's important to be able to determine the relative amount of RPE elevation -- something the Amsler grid can't do.

The PHP test can manage this because our attention is always drawn to the greatest distortion we're experiencing. Once a patient's response indicates a visual distortion caused by RPE elevation, the instrument flashes other lines that pass through that point. As long as the bend in the line created by the patient's visual flaw is greater than the one presented by the instrument, the patient will select the physiological flaw. Once the artificial distortion is perceived as greater than the physiological one, the patient will select the artificial distortion instead.

By flashing lines with different degrees of artificial distortion that pass through the patient's area of physiological distortion, the instrument can determine how much elevation in the RPE has occurred. That, in turn, indicates whether the patient has developed CNV.

Gauging the Results

In addition to presenting a map showing field defects, the PHP provides a confidence level for the results, using a normative database. Rather than an accuracy rating, the confidence level indicates the likelihood that the results show a progression to wet AMD. For example, if a test indicates no progression to wet AMD, but confidence level is low, the patient is borderline.

The instrument designers elected not to provide the doctor with numerical data -- at least for now. "The PHP is currently designed to detect evidence of progression as a diagnostic test," explains Yair Alster of Notal Vision. "Some early evidence suggests the instrument may be useful for monitoring progression over time. Once that use is clinically established, we'll adapt the software to provide comparative, numerical data."

Comments from the Field

Richard Rosen, M.D., is a retinal specialist at the New York Eye and Ear Infirmary; he's used the PHP to test close to 40 patients. "So far, results seem to correlate well with the presence of the pathology we're looking for," he says. "We've checked patients with other macular complaints but no evidence of neovascular membrane, and they come up negative.

"It's an easy test for the patient; it takes between 5 and 10 minutes, depending on the patient's facility. Unlike a conventional visual field, many patients see this as a kind of video game; they have to touch the screen where they see changes in the line. Although it's a serious test, there's an element of play, which improves patient cooperation.

"Eventually, this could become a home test, perhaps using a Palm Pilot-type handheld screen. In the meantime, it could screen people in environments where there's not always a physician present, perhaps when buying glasses. It's an exciting technology with a lot of potential."

Appearing Soon

Carl Zeiss Meditec hopes to have the Preview PHP on the market by early 2004. For more information, contact the Carl Zeiss Meditec Customer Direct Center at (800) 342-9821.