A clinical study of perimetric probability maps

Visual Field Pointwise Analysis of the Idiopathic Intracranial Hypertension Weight Trial (IIH:WT)

Purpose

This study was designed to determine if point analysis of the Humphrey visual field (HVF) is an effective outcome measure for people with idiopathic intracranial hypertension (IIH) compared with mean deviation (MD).

Methods

Using the IIH Weight Trial data, we performed a pointwise analysis of the numerical retinal sensitivity. We then defined a medically treated cohort as having MDs between -2 dB and -7 dB and calculated the number of points that would have the ability to change by 7 dB.

Results

The HVF 24-2 mean ± SD MD in the worse eye was -3.5 ± 1.1 dB (range, -2.0 to -6.4 dB). Total deviation demonstrated a preference for the peripheral and blind spot locations to be affected. Points between 0 dB and -10 dB demonstrated negligible ability to improve, compared with those between -10 dB and -25 dB. For the evaluation of the feasibility for a potential medical intervention trial, only 346 points were available for analysis between -10 dB and -25 dB bilaterally, compared with 4123 points in baseline sensitivities of 0 to -10 dB.

Conclusions

Patients with IIH have mildly affected baseline sensitivities in the visual field based on HVF analyzer findings, and the majority of points do not show substantial change over 24 months in the setting of a randomized clinical trial. Most patients with IIH who are eligible for a medical treatment trial generally have the mildest affected baseline sensitivities. In such patients, pointwise analysis offers no advantage over MD in detection of visual field change.

Visual Field Estimation by Probabilistic Classification

The gold standard clinical tool for evaluating visual dysfunction in cases of glaucoma and other disorders of vision remains the visual field or threshold perimetry exam. Administration of this exam has evolved over the years into a sophisticated, standardized, automated algorithm that relies heavily on specifics of disease processes particular to common retinal disorders. The purpose of this study is to evaluate the utility of a novel general estimator applied to visual field testing. A multidimensional psychometric function estimation tool was applied to visual field estimation. This tool is built on semiparametric probabilistic classification rather than multiple logistic regression. It combines the flexibility of nonparametric estimators and the efficiency of parametric estimators. Simulated visual fields were generated from human patients with a variety of diagnoses, and the errors between simulated ground truth and estimated visual fields were quantified. Error rates of the estimates were low, typically within 2 dB units of ground truth on average. The greatest threshold errors appeared to be confined to the portions of the threshold function with the highest spatial frequencies. This method can accurately estimate a variety of visual field profiles with continuous threshold estimates, potentially using a relatively small number of stimuli.

Using Small Samples to Evaluate Normative Reference Ranges for Retinal Imaging Measures

SIGNIFICANCE

Retinal nerve fiber layer (RNFL) deviation maps often incorrectly score healthy eyes as having wedge defects. This study shows how to identify such problems early in the development of normative databases.

PURPOSE

After reference values are embedded in devices, clinicians and researchers often learn about issues that cause false-positive rates in healthy eyes. Here we show a way to detect and address such issues early on.

METHODS

The thickness of the RNFL was measured for both eyes of 60 healthy younger adults aged 20 to 31 years and one eye each of 30 healthy older adults aged 54 to 82 years. Deviation maps were developed from the left eyes of the first 30 younger adults, and between-subject variability in the shape of the RNFL was assessed. This was repeated in their right eyes, in the second group of younger adults and in the older adults.

RESULTS

For the first group of 30 healthy young adults, between-subject variability in the location of the region of greatest thickness meant that 58% of the pixels below the fifth percentile in the left eyes were from four people whose deviation maps had wedge-shaped patterns, as did the deviation maps for the nine right eyes with 87% of the pixels below the fifth percentile. Wedge patterns were also seen in deviation maps for 8 left eyes and 11 right eyes of the second group of young adults and for 9 eyes of the older adults.

CONCLUSIONS

Evaluation of RNFL thickness maps from 30 young adults was sufficient to determine that between-subject variability in the shape of the RNFL can cause wedge patterns in RNFL deviation maps in many healthy eyes.

Current Practice with Standard Automated Perimetry

This article reviews standard visual field testing. The authors discuss the psychophysics involved in the design of the perimeter and the parameters used to test visual field sensitivity. The authors also explain normal and pathological sensitivity across the visual field, the patient and testing conditions that influence visual field results, and the interpretation of a single visual field, with an emphasis on detection of glaucomatous damage. The new thresholding program for visual fields, Swedish Interactive Thresholding Algorithm, is explained. Finally, the authors give examples of factors that should be considered when setting criteria for abnormality and review how recent studies have used various criteria to identify abnormality.

Characteristics of the normative database for the Humphrey matrix perimeter

PURPOSE

The Humphrey Matrix (Carl Zeiss Meditec, Dublin CA; Welch-Allyn, Skaneateles, NY) is a high-spatial-resolution perimeter that uses frequency-doubling stimuli. It incorporates an efficient test strategy that assumes that age, eccentricity, and test procedure type have only small effects on sensitivity. The results used to create the normative database for the perimeter were examined, to see whether these assumptions were met and to examine the form of the normative data.

METHOD

Visual fields were measured (Matrix 30-2, 24-2, 10-2 and Macula patterns) in >275 subjects judged to be normal by a battery of clinical procedures. The right eye was always tested first.

RESULTS

Sensitivity decreased by approximately 0.7 dB per age decade across all eccentricities; sensitivity decreased with eccentricity, typically by <5 dB at the most peripheral points tested. Although there was no systematic difference in sensitivity between the 30-2 and 24-2 tests, the Macula test sensitivities were typically 1 dB higher than for the 10-2 test. Sensitivity in the left eye was slightly lower than in the right, with the difference being significantly greater in the temporal visual field. In most test locations, the 95% confidence interval of normal sensitivity was approximately 6 dB below the median sensitivity.

CONCLUSIONS

The performance of the test strategy in the Matrix perimeter is appropriately matched to the response characteristics of the normal population. The finding of a spatially nonuniform difference in sensitivity between left and right eyes is attributable to light-adaptation differences between the eyes. This effect is accounted for in the perimeter's normative database.

The sensitivity and specificity of nerve fiber layer measurements in glaucoma as determined with scanning laser polarimetry

PURPOSE

To determine the sensitivity and specificity for detecting glaucoma by scanning laser polarimetry and to assess the relation between nerve fiber layer (NFL) measurements and visual field indices.

METHODS

The peripapillary NFL was divided into four segments: superior, inferior, temporal, and nasal. The mean polarimetric NFL for each segment was calculated out of six selected areas of 256 pixels each. Ratios relative to the nasal segment were determined for the superior and inferior segments. With the use of previously obtained normograms for polarimetric NFL readings, the sensitivity of scanning laser polarimetry was assessed in 200 glaucomatous eyes (155 subjects). The specificity was assessed in a normal population of 150 eyes (150 subjects). The relation between hemifield polarimetric NFL and visual field indices was assessed by linear regression analysis.

RESULTS

The sensitivity of scanning laser polarimetry was 96% and the specificity was 93%. The correlation between NFL parameters and visual field indices ranged from -0.18 to +0.26. The amount of variation by the linear regression model ranged from 3% to 6%.

CONCLUSIONS

Although quantitative measurements of the NFL with scanning laser polarimetry relate poorly to visual field indices, the technique seems to be promising for screening populations for glaucoma. Whether measurements of the NFL with scanning laser polarimetry are also sensitive enough to detect change over time requires further study.

Detection of incident field loss using the glaucoma hemifield test

PURPOSE

To examine different definitions of incident visual field loss among patients with elevated intraocular pressure and varying numbers of abnormal glaucoma hemifield test results over an average of 6 years of follow-up.

METHODS

A cohort of patients with annual C-30-2 Humphrey visual fields were followed for a minimum of 5 years. Three different definitions of field loss were compared: 1, 2, or 3 consecutive annual abnormal glaucoma hemifield test results.

RESULTS

Of 253 subjects, 506 eyes were followed for 5 to 9 years. If incident field loss was defined as one or more normal fields followed by one abnormal glaucoma hemifield test result, the incidence of field loss was 63.6 per 1000 person-years of follow-up. For two or three consecutive abnormal glaucoma hemifield test results, the rates were 27.6 and 19.2 per 1000 person-years of follow-up, respectively. Among patients with field loss in one eye at the start of the study, the incidence of field loss in the fellow eye using 1, 2, or 3 consecutive abnormal fields as the definition of incident field loss was 60.9, 55.5, and 25.5 per 1000 person-years of follow-up, respectively. Three years after incident field loss, 31.9% (1 abnormal test result), 76.5% (2 abnormal test results), and 88.5% (3 abnormal test results) of eyes with incident field loss had an abnormal hemifield test result. For eyes with one, two, and three consecutive abnormal glaucoma hemifield test results at the start of the study, 59.2%, 83.6%, and 89.1%, respectively, had an abnormal field 3 years later.

CONCLUSIONS

One abnormal glaucoma hemifield test result is not a consistent criterion for defining incident field loss. The use of two or three consecutive abnormal fields to define incident field loss makes it more likely that subsequent test results will be abnormal.

Diffuse and localised visual field defects to automated perimetry in primary open angle glaucoma

The occurrence of generalised or diffuse visual field depression in primary open angle glaucoma (POAG), in contrast to the presence of characteristic localised defects, is controversial. The frequency of diffuse visual field loss to automated static perimetry in the early stages of POAG was determined and compared with the frequency of localised defects. Twenty-five eyes of 25 consecutive POAG patients who met the selection criteria were tested on the Humphrey Visual Field Analyser with the Central 30-2 Threshold Test. Patients' mean age was 68 years. All eyes had visual acuity better than 6/9 with correction, refractive error of less than +/- 7 dioptres, no media opacities and normal pupils (3-6 mm). For all eyes, the frequency of abnormality on the STATPAC Total Deviation (TD) and Pattern Deviation (PD) plots were determined for all individual test points. Mean values and standard deviations were compared. The TD plots represent a composite of both diffuse and localised visual field depression, while PD plots are intended to reflect localised field defects. The frequency of involvement of the test points on the TD plots was higher than on the PD plots (mean +/- SD: 41.5 +/- 11.2% vs 27.5 +/- 10.9%) for all presenting defects. When only deep defects (STATPAC p < 1% and p < 0.5%) were evaluated, TD involvement was still more frequent than the PD (20.1 +/- 9.4% vs 13.2 +/- 7.2%). The topographical pointwise incidence of pure generalised sensitivity loss in the visual field was less frequent when only deep defects were taken into consideration (6.9 +/- 5.7%), and the incidence increased progressively with the inclusion of intermediate and shallow field defects (9.6 +/- 6.8% and 14.0 +/- 8.4% respectively). A component of diffuse sensitivity depression is present at all significant levels of visual field loss in glaucoma. This component of generalised loss decreases as the depth of the field defects increases. This finding suggests that early diffuse field loss converts into well-defined pattern defects at later stages.

Effectiveness of unsupervised oculokinetic perimetry for detecting glaucomatous visual field defects

Oculokinetic perimetry (OKP) is a simple, inexpensive technique which was introduced primarily to promote more widespread use of visual field screening for glaucoma. The effectiveness of unsupervised OKP screening was evaluated using 145 non-glaucomatous patients and 64 patients with previously undiagnosed primary open angle glaucoma. Glaucoma cases were validated by the extended 132 point program on the Henson CFS2000 instrument. Both sample populations were representative of patients presenting for primary vision care. The effectiveness of the 26 point OKP test was compared with the 26 point Henson screening program. The results suggest that, in contrast to the Henson screening test (sensitivity = 85.1%, specificity = 93.8%), unsupervised OKP screening (sensitivity = 25.0%, specificity = 93.6%) has limited effectiveness for detecting glaucomatous visual field defects.

Performance of two films for densitometry of retinal photographs

The relative performance was determined of two different photographic films (Kodak Panatomic-X and Kodak Technical Pan) for densitometry readings from retinal photographs of patients who had or were suspected of having glaucoma. The raw data from the two films were significantly different. The high contrast film was particularly sensitive to external variables; however, when normalised in terms of the standardised deviation, the data from the two films were comparable (P < 0.005). We also measured the pattern of nerve fibre layer loss using digital image analysis of red-free photographs of normal, ocular-hypertensive, glaucoma-suspect and glaucomatous eyes. Several mathematical techniques were used to characterise the data from each eye and then to compare these data to the mean photographic density of the normal eyes. Results showed that it was possible to separate normal eyes from glaucomatous eyes. The highest sensitivity achieved was 100% for right and 88% for left eyes, and the highest specificity was 100% for both eyes.

Threshold variation in automated perimetry

A physiologic fluctuation in threshold levels exists in automated perimetry, which is greater in glaucoma and ocular hypertensive patients than in normal subjects. Fluctuation increases with greater eccentricity from fixation and in areas of reduced retinal sensitivity. Other factors related to automated perimetry testing potentially may influence threshold fluctuation including: a learning effect, reliability, pupil size, age, and the mode of stimulus presentation. Statistical software may aid in analyzing both the single field and the changes between successive fields. However, statistical programs cannot replace physician judgment, and all factors that may influence threshold levels should be considered when interpreting automated visual fields.

A comparison of unweighted and fluctuation-weighted indices (within the central 28 degrees of glaucomatous visual fields measured with the Octopus automated perimeter)

The main visual field indices of Flammer and the analogous ones of Heijl have been compared based on a population of 113 glaucomatous visual fields. It is shown that for the population studied, the differences between the two sets of indices are negligible for Octopus G1 program results and that they may thus be used interchangeably.

The effects of weighting the "mean defect" visual field index according to threshold variability in the central and midperipheral visual field

Two visual field indices, the mean defect (MD) of Flammer and the mean deviation (MD) of Heijl, have found wide acceptance among perimetrists. We compared these indices in 169 visual fields from normal- and high-tension glaucomatous eyes. Visual field damage in these eyes varied from slight to severe. In computations of the mean deviation index, the threshold values are weighted by the threshold deviations obtained from normal eyes as a function of eccentricity. However, the present study shows that the differences between the two indices in the population studied are negligible. Thus, subsequent interpretation is not affected by the choice of index, and the two MD indices may be considered to be interchangeable for the types of visual fields used in this study and for program-G1 examinations carried out using Octopus automated perimeters. Since we found smaller increases in local intersubjective fluctuations as a function of eccentricity in 274 normal visual fields as compared with results published by others, caution is indicated for interpretation of the visual field using probability weighting.