Interpretation of a Series of Visual Fields and Detection of Visual Field Progression
Interpretation of serial visual fields should meet 2 goals:
-
separating real change from ordinary variation
-
using the information obtained from visual field testing to determine the likelihood that a change is related to glaucomatous progression
Visual field testing is a subjective examination, and different responses may be obtained each time the test is performed or even during the same test. This fluctuation can greatly confound the detection of disease progression. In order to detect true visual field progression, the clinician needs to evaluate whether the observed change exceeds the expected variability for a particular point or area.
There are, in general, 2 main approaches to analyzing visual field progression. The first approach is to compare the results of the current test (or a few recent tests for confirmation) with those obtained at baseline (often a pair of baseline visual field tests). If the results of the follow-up examination(s) are significantly worse than baseline, progression is said to have occurred. This approach, which is called event-based analysis, defines progression based on a predefined incremental deterioration compared to baseline.
In the second approach, called trend-based analysis, instead of comparing the current test with a baseline test, the clinician looks for progressive change by analyzing changes in visual field points or summary statistics for all tests available in a specific period. Change is observed as a trend in the values plotted over time, and significant deterioration can be assessed by observing the slope of the regression line. In addition to evaluating whether progression has occurred, trend-based analysis allows estimation of the rate of progression. It is well known that some patients decline faster than others; estimating each individual’s rate of progression is helpful for predicting the risk of functional impairment and determining how aggressive treatment should be.
There are a variety of tools available to assist clinicians in identifying visual field progression, and there is no consensus about the best method for detecting change. The simplest and most general method uses the MD index plotted against time. A statistically significant decline would indicate progressive deterioration. Deterioration of the MD index may represent glaucomatous progression or progression of cataract or other media opacities. Conversely, in a glaucoma patient who has undergone cataract surgery, true underlying glaucoma progression may be masked by this method.
The Humphrey perimeter provides Guided Progression Analysis (GPA) software to assist in detection of visual field progression (Fig 12-6). This software presents an event-based method that is based on the pattern-deviation plot and, therefore, adjusts for the potential confounding effects of diffuse loss of sensitivity from media opacities. New or worsening visual field defects are identified by comparison to a pair of baseline tests; therefore, it is critical to have reliable baseline examinations. Often, the patient experiences a learning effect, and the second visual field may show substantial improvement over the first. To address this phenomenon, at least 2 visual field tests should be performed as early as possible in the course of a patient’s disease. If the results are quite different, a third test should be performed. The software automatically selects the first 2 available examinations as the baseline tests. However, this selection can be easily overridden to a more suitable time point (eg, change in therapy after progression) or to avoid initial learning effects (which could reduce the sensitivity to detect progression). The software then compares each follow-up test to the average of the baseline tests (Fig 13-6). It identifies points that show change greater than the expected variability (at the 95% significance level), as determined by previous studies with stable glaucoma patients. If significant change is detected in at least 3 points and repeated for the same points in 2 consecutive follow-up tests, the software will flag the last examination as Possible Progression. If significant change is detected and repeated for the same 3 or more points in 3 consecutive follow-up tests, the GPA software will flag the last examination as Likely Progression.
GPA software also provides the Visual Field Index (VFI) and VFI progression plot (Fig 14-6). The VFI is calculated as the percentage of normal visual field, after adjustment for age. Therefore, a VFI of 100% represents a completely normal visual field, while a VFI of 0% represents a perimetrically blind visual field. The VFI is shown on the GPA printout as a percentage value for each examination. A trend-based analysis of VFI as a function of the patient’s age is presented with a future projection that predicts the VFI, assuming the same regression slope, 5 years in the future. Whereas the MD is based only on the total-deviation map, and is thus affected by cataract, the VFI is based both on the pattern-deviation probability map, for the identification of possibly progressing points, and on the total-deviation map, used for the actual calculation of change of the total-deviation value. In addition, the algorithm applies different weights to different locations, giving more weight to more central points, which have greater impact on the patient’s quality of vision. The final VFI score is the mean of all weighted scores.
The Octopus perimeter also provides a comprehensive statistical package (EyeSuite) for evaluation of visual field progression. The software calculates rates of progression in terms of mean defect change per year (in dB/year), similar to the MD index from Humphrey perimetry. In addition, the software provides an analysis of progression by individual test points (pointwise linear regression) and by clusters, where test locations are combined according to nerve fiber bundle patterns.
Frequency of Testing
Accurate and timely detection of progressive changes can be difficult because of the inherent variability of visual field testing. The time required to detect new visual field loss depends on the frequency of testing and follow-up scheme used. The recommended frequency of testing varies depending on patient characteristics such as age, disease severity, presence of risk factors for progression, concomitant clinical findings, and risk for functional impairment. In order to precisely estimate a rate of progression, many tests are needed over time. Studies have shown that the strategy of acquiring only 1 visual field test per year may be insufficient in a large number of cases, resulting in delayed detection of disease progression and imprecise estimation of rates of change. Although the optimal testing strategy will vary, obtaining at least 2 or 3 tests per year during the first 2 years of follow-up is recommended in order to exclude the possibility of fast progression. Thereafter, a testing strategy consisting of 2 tests per year may be sufficient in most cases.
-
Wu Z, Saunders LJ, Daga FB, Diniz-Filho A, Medeiros FA. Frequency of testing to detect visual field progression derived using a longitudinal cohort of glaucoma patients. Ophthalmology. 2017;124(6):786–792.
Excerpted from BCSC 2020-2021 series: Section 10 - Glaucoma. For more information and to purchase the entire series, please visit https://www.aao.org/bcsc.