Earlier this year, Glaucoma Today's Chief Medical Editor, Richard A. Lewis MD, wrote an editorial titled, "Not a Bad Gig."1 Like Dr. Lewis, I enjoy my job, and I agree that I can now do much more for my glaucoma patients than I could when I started practicing a quarter of a century ago. In addition, Dr. Lewis commented on an uplifting presentation given by Reay Brown, MD, at the 2008 ASCRS Glaucoma Day concerning "How to Enjoy a Glaucoma Practice."2 Dr. Brown noted that an upbeat attitude is an essential part of every doctor/patient interaction. Physicians who adopt the philosophy that treating glaucoma is a "good gig" and maintain a positive attitude toward their work will not only improve the quality of care they offer their patients, but they will also enhance the efficiency of their practices. These principles can also be applied to perimetry, because they can help us choose the right visual field test for the right patient.
The efficiency of perimetry depends on the size and color of the stimulus, the area of the field we test, and how we interpret the data. For example, a predominantly black grayscale printout impedes our ability to assess the patient's remaining visual field accurately. Inappropriate testing can also make patients feel depressed and anxious. If they cannot see the test objects, they will feel like they are going blind.
Ideally, we want to create a clinical environment in which the information provided by diagnostic tests can help us prevent blindness. This article describes two common errors that reduce the efficiency of perimetry and discusses how choosing properly sized testing stimuli and challenging different areas of the visual field can help us improve the test's utility.
IDENTIFYING PERIMETRIC PITFALLS
Perimetry is a major component of the care we provide to glaucoma patients or suspects. Our job is to obtain the best possible representation of each patient's field of vision, interpret the data, and use the results to formulate a treatment plan.
Patients often dislike undergoing visual field testing, because they worry they will not perform well and become concerned about the repercussions of the test. In addition, perimetric testing increases the amount of time patients spend in your office, especially if you order a suboptimal test. In such situations, they may become apprehensive, and you will find yourself involved in extensive face-to-face discussions trying to explain the problems caused by using the wrong testing protocol.
For example, we could learn more about a patient who has tunnel vision with a 10-2 versus a 30-2 field printout. Nevertheless, I still see doctors ordering the latter test year after year for this type of visual field loss, thereby wasting everyone's time. If the patient has absolute field loss between 15° and 30°, why should we test that area? We must remember that ordering the same visual field test does not work for every patient.
We can improve the efficiency of perimetric testing in our practices by avoiding the following errors3:
- Testing patients with inappropriately sized stimuli
- Failing to examine the portion of the visual field that provides the most data in the shortest feasible time
- Choosing an algorithm that causes excessive fatigue
- Retesting the entire visual field of a patient who has only a central island of vision
- Failing to test specific areas in which patients feel they are losing vision
- Retesting patients who continually produce unreliable data
CHOOSING THE SIZE OF TEST STIMULI
In general, we could do a better job of re-evaluating the visual fields of patients with advanced glaucoma by remembering that one size of stimulus does not fit all. Too often, testing patients who have advanced glaucoma generates zero-decibel printouts characterized by blacked-out grayscales. Even worse, some physicians perpetuate the inefficient use of this test by utilizing the same parameters every year, thus obtaining the same blacked-out results.
Figure 1 shows how we can increase the sensitivity of static automated perimetry by enlarging the size of the test object. Because a size V stimulus is 16 times larger (64 mm2 vs 4 mm2) and subtends more angular degrees at the pupil (1.724° vs 0.431°) than a size III stimulus, it recruits more cells adjacent to the targeted retinal ganglion cells. This process of spatial summation captures more useful information about the patient's remaining vision.4,5
As always, at least two baseline visual fields are optimal to detect change, but it is incumbent on us to identify and map every functioning retinal ganglion cell, especially in patients whose visual acuity has decreased to 20/40 or lower due to glaucoma. If we only test patients with a size III stimulus, we may not obtain an accurate measurement of their remaining retinal sensitivity. As a corollary, we may fail to detect visual loss that is apparent only from the patient's perspective.
Normally, changing the size of a test object from III to V would increase retinal sensitivity from 6 to 10 decibels. A phenomenon called pathologic summation of Dubois-Poulsen, however, allows us to take advantage of an exaggerated improvement in retinal sensitivity that occurs among glaucoma patients. A spot that has zero decibels with a size III stimulus may unexpectedly improve to 20 decibels when it is tested with a size V stimulus. Testing with a larger object therefore identifies areas of functional vision that would be missed with a smaller stimulus. In addition, patients who have advanced glaucomatous visual field loss tend to have better clinical experiences when they interact with a size V stimulus.
Figure 2 shows how we can improve our perimetric efficiency by carefully choosing the testing area. If we know a patient only has a 10° central island of vision, we should not waste time testing his full visual field.6 We will obtain more information about his remaining vision with a 10-2 central field test, because it has a smaller test grid (2° vs 6°) and picks up more points of vision in this test (36 vs four) than the 24-2 protocol. Using a strategy that explores a patient's visual field with the correct test stimulus while targeting areas that retain functional vision not only involves him in the examination and prevents him from falling asleep during perimetric testing, but it also enhances his mental health and increases our chance of collecting data we can use to preserve his remaining vision.
CONCLUSION
The key to efficient perimetric testing is to generate data that help us prevent patients from going blind. I believe that any approach that fails to achieve this goal constitutes suboptimal care. In addition, using the properly sized test object and customizing the testing algorithm to each patient's visual field will help us and our patients maintain the upbeat attitude recommended by Dr. Brown, and consequently, to improve our practices' efficiency.
Ronald L. Fellman, MD, is a glaucoma specialist at Glaucoma Associates of Texas in Dallas. He acknowledged no financial interest in the product or company mentioned herein. Dr. Fellman may be reached at (214) 360-0000; rfellman@aol.com.
