A comparison of experienced clinical observers and statistical tests in detection of progressive visual field loss in glaucoma using automated perimetry

[Static automated perimetry in the diagnosis of glaucoma. Assessment of disease progression]

There are several ways to assess glaucoma progression using standard automated perimetry. Most often, ophthalmologists evaluate the stability of visual functions manually when comparing several study protocols. The advantages of clinical assessment are ease of implementation and the ability to interpret data from any device. The main disadvantage of this method is its subjectivity. There are many available automated methods for assessing disease progression involving Humphrey Field Analyzer and Octopus perimeters. Event analysis allows determining glaucoma progression at the time of examination, with consideration of the possible physiological fluctuations in light sensitivity. Trend analysis of perimetric indices makes it possible to assess the rate of glaucoma progression and forecast the trend of changes in visual functions over the next five years. All these methods for assessing progression have certain advantages and disadvantages and cannot be considered ideal. Pointwise and cluster trend analysis are more sensitive in early glaucoma and are being actively researched and developed. These methods have great potential, although they are not yet sufficiently available in clinical practice.

Glaucoma Progression Diagnosis: The Agreement between Clinical Judgment and Statistical Software

Background: To explore the agreement between clinical judgment and Guided Progression Analysis II (GPAII) in the evaluation of visual fields (VF) progression in patients with glaucoma. Methods: Three glaucoma experts and three general ophthalmologists were asked to rate the VF series by classifying them as progressive through the observation of the overview report. The agreement between clinical judgment and GPAII event analysis (EA) and trend analysis (TA) was assessed by Cohen statistic. The sensitivity and specificity of clinical judgment in detecting the presence of progression was evaluated considering the results of GPAII as the reference standard. Results: 66 VF series were included in the study. Glaucoma experts, general ophthalmologists, GPAII EA, and GPAII TA found progression in 39%, 38%, 15%, and 21% of the VF series (p < 0.05). The clinical judgment of glaucoma experts and general ophthalmologists was discordant with GPAII EA in 27.2% and 28.7% (k = 0.35, 95% CI 0.15−0.56 and k = 0.30, 95% CI 0.09−0.52) and with GPAII TA in 21.2% and 25.7% of the VF series examined (k = 0.51, 95% CI 0.31−0.72 and k = 0.41, 95% CI 0.18−0.62). Considering the GPAII EA and TA as reference standard, glaucoma experts showed a sensitivity of 90% and 92.8% and a specificity of 69.6% and 75%, while general ophthalmologists showed a sensitivity of 80% and 78.5% and a specificity of 69.6% and 73%. Conclusions: The agreement between clinical judgment and GPAII ranges from fair to moderate. Glaucoma experts showed better ability than general ophthalmologists in detecting VF progression.

Clinicians' Use of Quantitative Information while Assessing the Rate of Functional Progression in Glaucoma

PURPOSE

Clinicians use both global and point-wise information from visual fields to assess the rate of glaucomatous functional progression. We asked which objective, quantitative measures best correlated with subjective assessment by glaucoma experts. In particular, we aimed to determine how much that judgment was based on localized rates of change vs. on global indices reported by the perimeter.

DESIGN

Prospective cohort study.

PARTICIPANTS

Eleven academic, expert glaucoma specialists independently scored the rate of functional progression, from 1 (improvement) to 7 (very rapid progression), for a series of 5 biannual clinical printouts from 100 glaucoma or glaucoma suspect eyes of 51 participants, 20 of which were scored twice to assess repeatability.

METHODS

Regression models were used to predict the average of the 11 clinicians' scores based on objective rates of change of mean deviation (MD), visual field index (VFI), pattern standard deviation (PSD), the Nth fastest progressing location, and the Nth fastest progressing of 10 anatomically defined clusters of locations after weighting by eccentricity.

MAIN OUTCOME MEASURES

Correlation between the objective rates of change and the average of the 11 clinicians' scores.

RESULTS

The average MD of the study eyes was -2.4 dB (range, -16.8 to +2.8 dB). The mean clinician score was highly repeatable, with an intraclass correlation coefficient of 0.95. It correlated better with the rate of change of VFI (pseudo-R2 = 0.73, 95% confidence interval [CI, 0.60-0.83]) than with MD (pseudo-R2 = 0.63, 95% CI [0.45-0.76]) or PSD (pseudo-R2 = 0.41, 95% CI [0.26-0.55]). Using point-wise information, the highest correlations were found with the fifth-fastest progressing location (pseudo-R2 = 0.71, 95% CI [0.56-0.80]) and the fastest-progressing cluster after eccentricity weighting (pseudo-R2 = 0.61, 95% CI [0.48-0.72]). Among 25 eyes with an average VFI of > 99%, the highest observed pseudo-R2 value was 0.34 (95% CI [0.16-0.61]) for PSD.

CONCLUSIONS

Expert academic glaucoma specialists' assessment of the rate of change correlated best with VFI rates, except in eyes with a VFI near the ceiling of 100%. Sensitivities averaged within clusters of locations have been shown to detect change sooner, but the experts' opinions correlated more closely with global VFI. This could be because it is currently the only index for which the perimeter automatically provides a quantitative estimate of the rate of functional progression.

Applications of deep learning in detection of glaucoma: A systematic review

Glaucoma is the leading cause of irreversible blindness and disability worldwide. Nevertheless, the majority of patients do not know they have the disease and detection of glaucoma progression using standard technology remains a challenge in clinical practice. Artificial intelligence (AI) is an expanding field that offers the potential to improve diagnosis and screening for glaucoma with minimal reliance on human input. Deep learning (DL) algorithms have risen to the forefront of AI by providing nearly human-level performance, at times exceeding the performance of humans for detection of glaucoma on structural and functional tests. A succinct summary of present studies and challenges to be addressed in this field is needed. Following PRISMA guidelines, we conducted a systematic review of studies that applied DL methods for detection of glaucoma using color fundus photographs, optical coherence tomography (OCT), or standard automated perimetry (SAP). In this review article we describe recent advances in DL as applied to the diagnosis of glaucoma and glaucoma progression for application in screening and clinical settings, as well as the challenges that remain when applying this novel technique in glaucoma.

Defining glaucomatous optic neuropathy using objective criteria from structural and functional testing

BACKGROUND/AIMS

To identify objective criteria from optical coherence tomography (OCT) and perimetry that denote a useful, specific definition of glaucomatous optic neuropathy (GON) in eyes with open-angle glaucoma for comparisons among glaucoma research studies.

METHODS

A cross-sectional study of adult patients with glaucoma from nine centres on five continents evaluated de-identified physician diagnosis, OCT and perimetry results for 2580 eyes (1531 patients) in an online database. Each eye was graded by their glaucoma specialist as either definite, probable or not GON. Objective measures from OCT and perimetry, derived from an online consensus panel comprising 176 glaucoma specialists globally, were compared against the three diagnostic levels.

RESULTS

Diagnoses were 54% 'definite', 22% 'probable' and 24% 'not GON'. Using only OCT data or only field data had inadequate specificity (<90%). The best definitional choice for data from either the most recent or the preceding OCT/field pair had 77% sensitivity at 98% specificity and consisted of abnormal OCT superior or inferior nerve fibre layer quadrant with matching, opposite, abnormal Glaucoma Hemifield Test.

CONCLUSIONS

Objective criteria to define GON are practical and may be useful for comparisons among clinical studies to supplement subjective clinical assessment.

An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis

Purpose

To detect visual field (VF) progression by analyzing spatial pattern changes.

Methods

We selected 12,217 eyes from 7360 patients with at least five reliable 24-2 VFs and 5 years of follow-up with an interval of at least 6 months. VFs were decomposed into 16 archetype patterns previously derived by artificial intelligence techniques. Linear regressions were applied to the 16 archetype weights of VF series over time. We defined progression as the decrease rate of the normal archetype or any increase rate of the 15 VF defect archetypes to be outside normal limits. The archetype method was compared with mean deviation (MD) slope, Advanced Glaucoma Intervention Study (AGIS) scoring, Collaborative Initial Glaucoma Treatment Study (CIGTS) scoring, and the permutation of pointwise linear regression (PoPLR), and was validated by a subset of VFs assessed by three glaucoma specialists.

Results

In the method development cohort of 11,817 eyes, the archetype method agreed more with MD slope (kappa: 0.37) and PoPLR (0.33) than AGIS (0.12) and CIGTS (0.22). The most frequently progressed patterns included decreased normal pattern (63.7%), and increased nasal steps (16.4%), altitudinal loss (15.9%), superior-peripheral defect (12.1%), paracentral/central defects (10.5%), and near total loss (10.4%). In the clinical validation cohort of 397 eyes with 27.5% of confirmed progression, the agreement (kappa) and accuracy (mean of hit rate and correct rejection rate) of the archetype method (0.51 and 0.77) significantly (P < 0.001 for all) outperformed AGIS (0.06 and 0.52), CIGTS (0.24 and 0.59), MD slope (0.21 and 0.59), and PoPLR (0.26 and 0.60).

Conclusions

The archetype method can inform clinicians of VF progression patterns.

Non-parametric tests in detecting glaucoma progression

Automated perimetry still represents the gold standard in long term glaucoma monitoring. On a daily practice basis, glaucoma progression analysis could be difficult due to the long time needed to detect, confirm, and quantify the progression rate. Moreover, "trend" and "event" analysis require a good theoretical basis to perform and interpret. Aim of study was to present an alternative method to conventional Glaucoma Progression Analysis (Humphrey Visual Field Analyzer, Carl Zeiss® Inc.) applied for the early detection of glaucoma progression. Such an "event" analysis orients the clinician in a fast manner on the progression profile in glaucoma patients and might adapt the follow up visits accordingly. Method and material: 41 eyes from 41 patients with open angle glaucoma were studied in a longitudinal manner, over a 24 months' time interval from diagnosis. Results: in the GPA analysis, a positive "event" (progression) was detected in 11/ 41 eyes (26.82%). Non-parametric analysis confirmed progression in all GPA cases, and additionally found 8 more eyes with positive progression (46.34% studied eyes). Mc Nemar concordance analysis between tests was good and relevant (kappa index k=0.596, p=0.000), with positive correlation (r=0.652, p=0.008). In conclusion, NPA tends to overestimate the number of progression cases in a cohort, but it can easily orient the clinician on the profile of the followed patients. In the first years, the GPA analysis can be highly inaccurate, but there is a great need to detect which patients are at significant risk for vision loss (fast progressors). Yet, combining the two methods of detection of glaucoma progression, the practitioners might direct their selected interest and attention towards observing a larger than expected number of patients who are at risk for vision loss over time due to glaucoma, but not necessarily in a fast manner.

Expert Evaluation of Visual Field Decay in Glaucoma Correlates With the Fast Component of Visual Field Loss

PURPOSE

To compare the assessment of serial visual fields (VFs) based on subjective expert evaluation with the fast and slow VF component rates determined with pointwise exponential regression (PER) and pointwise linear regression (PLR).

MATERIALS AND METHODS

A total of 5272 VF examinations from 376 eyes diagnosed with open-angle glaucoma were included. Three glaucoma specialists assessed each VF qualitatively to evaluate progression status and the qualitative rate of progression. The rates of VF decay were determined with PER and PLR at each VF location, which were ranked according to the regression coefficient and partitioned into 2 groups (fast and slow). A mean rate for the fast and slow partitions was obtained based on the average of the regression coefficients in each partition. κ-values were used to measure the agreement among the experts and the PER and PLR algorithms.

RESULTS

The average baseline VF mean deviation for the study sample was -6.6 (±5.9) dB. The agreement of the likelihood of progression among the dichotomized experts' score and PER was moderate (κ=0.41, P<0.01) and fair (κ=0.39, P<0.01) for PLR. The agreement of the likelihood of progression among the 3 dichotomized experts' scores was fair (κ=0.22, P<0.01). The agreement of the area of worsening among the dichotomized experts' score and PER and PLR were both moderate (κ=0.48, P<0.01; κ=0.46, P<0.01). The eyes flagged by experts as having "fast" progression rates had a higher average rates of decay for PER and PLR at -2.7 (±4.1) %/year and -0.8 (±1.2) dB/year; eyes flagged as "slow" had lower rates of decay at -0.3 (±1.5) %/year and -0.1 (±0.5) dB/year.

CONCLUSIONS

Expert qualitative evaluation of field series for change and rate of change correlate more closely with the fast component than with the slow component of VF decay.

Enhancement of Visual Field Predictions with Pointwise Exponential Regression (PER) and Pointwise Linear Regression (PLR)

Purpose

The study was conducted to evaluate threshold smoothing algorithms to enhance prediction of the rates of visual field (VF) worsening in glaucoma.

Methods

We studied 798 patients with primary open-angle glaucoma and 6 or more years of follow-up who underwent 8 or more VF examinations. Thresholds at each VF location for the first 4 years or first half of the follow-up time (whichever was greater) were smoothed with clusters defined by the nearest neighbor (NN), Garway-Heath, Glaucoma Hemifield Test (GHT), and weighting by the correlation of rates at all other VF locations. Thresholds were regressed with a pointwise exponential regression (PER) model and a pointwise linear regression (PLR) model. Smaller root mean square error (RMSE) values of the differences between the observed and the predicted thresholds at last two follow-ups indicated better model predictions.

Results

The mean (SD) follow-up times for the smoothing and prediction phase were 5.3 (1.5) and 10.5 (3.9) years. The mean RMSE values for the PER and PLR models were unsmoothed data, 6.09 and 6.55; NN, 3.40 and 3.42; Garway-Heath, 3.47 and 3.48; GHT, 3.57 and 3.74; and correlation of rates, 3.59 and 3.64.

Conclusions

Smoothed VF data predicted better than unsmoothed data. Nearest neighbor provided the best predictions; PER also predicted consistently more accurately than PLR. Smoothing algorithms should be used when forecasting VF results with PER or PLR.

Translational Relevance

The application of smoothing algorithms on VF data can improve forecasting in VF points to assist in treatment decisions.

Comparing glaucoma progression on 24-2 and 10-2 visual field examinations

PURPOSE

To compare the rate of mean deviation (MD) change on 24-2 versus 10-2 VFs in treated glaucomatous eyes with 5 or more examinations.

METHODS

In a retrospective study, 24-2 and 10-2 VFs of 131 glaucoma patients (167 eyes) who had undergone at least 5 VFs examinations during their follow-up were analyzed. All these patients had VF defects both on 24-2 and 10-2 VFs. Rates of MD change were calculated using best linear unbiased predictions (BLUP).

RESULTS

Median age, MD on 24-2 VF at baseline, number of VFs performed during follow-up and follow-up duration were 55 years, -16.9 dB, 9 and 9 years respectively. Median rate of MD change was significantly greater (p<0.001) on 10-2 VF (-0.26 dB/year; interquartile range [IQR]: -0.47, -0.11) compared to 24-2 VFs (-0.19 dB/year; IQR: -0.41, -0.03). Comparing the rates of MD change in eyes with different severities of VF loss (early [MD better than -6 dB], moderate [-6 dB to -12 dB], advanced [-12 to -20 dB] and severe [MD worse than -20 dB]) at baseline (based on the MD on 24-2 VF), median rate of MD change was comparable between 10-2 and 24-2 VFs in mild (-0.45 dB/year vs. -0.40 dB/year, P = 0.42) and moderate (-0.32 dB/year vs. -0.40 dB/year, P = 0.26) VF loss categories, while the same were significantly greater on 10-2 VFs in advanced (-0.28 dB/year vs. -0.21 dB/year, P = 0.04) and severe (-0.18 dB/year vs. -0.06 dB/year, P<0.001) VF loss categories.

CONCLUSIONS

In patients with VF defects both on 24-2 and 10-2 VFs, evaluating the rate of MD change on 10-2 VFs may help in better estimation of glaucoma progression, especially so in eyes with advanced glaucoma at baseline.

Comparison of regression models for serial visual field analysis

PURPOSE

Our aim was to compare fit and predictive performance effectiveness of four pointwise regression models in measuring the visual field (VF) decay rate of progression in patients with open-angle glaucoma.

METHODS

We selected Humphrey VF data of patients with open-angle glaucoma with a minimum follow-up time of 6 years. For each eye (n = 798 from 588 patients), we regressed threshold sensitivity (y) at each VF test location for the entire VF series against follow-up time (x), with four candidate first-order regression models: (1) ordinary least-squares linear regression model (y = β 0 + β 1 x); (2) nondecay exponential regression model (y = β 0 + β 1e (x) ); (3) decay exponential regression model ([Formula: see text]); (4) Tobit-censored, maximum-likelihood linear regression model (y* = [Formula: see text], ε ~ N(0, σ(2))), where x is follow-up time and y is threshold sensitivity.

RESULTS

The average [± standard deviation (SD)] baseline VF mean deviation (MD) was -8.2 (±5.5) dB, the mean follow-up was 8.7 (±1.9) years, and the number of follow-up VFs was 14.7 (±4.4). The decay exponential model was the best-fitting (42.7 % of locations) and best-forecasting (65.5 % of locations) model. The decay exponential model was the best prediction model in all categories of severity.

CONCLUSIONS

It is not clear that the ordinary least-squares linear regression model is always the favored model for fitting and forecasting VF data in patients with glaucoma. The pointwise decay exponential regression (PER) model was the best-fitting and best-predicting model across a wide range of glaucoma severity and can be readily understood by clinicians.

Agreement between event-based and trend-based glaucoma progression analyses

PURPOSE

To evaluate the agreement between event- and trend-based analyses to determine visual field (VF) progression in glaucoma.

METHODS

VFs of 175 glaucoma eyes with ≥5 VFs were analyzed by proprietary software of VF analyzer to determine progression. Agreement (κ) between trend-based analysis of VF index (VFI) and event-based analysis (glaucoma progression analysis, GPA) was evaluated. For eyes progressing by event- and trend-based methods, time to progression by two methods was calculated.

RESULTS

Median number of VFs per eye was 7 and follow-up 7.5 years. GPA classified 101 eyes (57.7%) as stable, 30 eyes (17.1%) as possible and 44 eyes (25.2%) as likely progression. Trend-based analysis classified 122 eyes (69.7%) as stable (slope >-1% per year or any slope magnitude with P>0.05), 53 eyes (30.3%) as progressing with slope <-1% per year, P≤0.05 (sensitive criteria), and 37 eyes (21.1%) as progressing with slope <-1% per year, P≤0.01 (specific criteria). κ between sensitive criteria of GPA (possible combined with likely progression) and trend-based analysis was 0.48, and between specific criteria of GPA (possible clubbed with no progression) and trend-based analysis was 0.50. In eyes progressing by sensitive criteria of both methods (42 eyes), median time to progression by GPA (4.9 years) was similar (P=0.30) to trend-based method (5.0 years). This was also similar in eyes progressing by specific criteria of both methods (25 eyes; 5.6 years versus 5.9 years, P=0.23).

CONCLUSION

Agreement between event- and trend-based progression analysis was moderate. GPA seemed to detect progression earlier than trend-based analysis, but this wasn't statistically significant.

Glaucoma Progression Detection: Agreement, Sensitivity, and Specificity of Expert Visual Field Evaluation, Event Analysis, and Trend Analysis

Purpose.

To assess sensitivity, specificity, and agreement among automated event analysis, automated trend analysis, and expert evaluation to detect glaucoma progression.

Methods.

This was a prospective study that included 37 eyes with a follow-up of 36 months. All had glaucomatous disks and fields and performed reliable visual fields every 6 months. Each series of fields was assessed with 3 different methods: subjective assessment by 2 independent teams of glaucoma experts, glaucoma/guided progression analysis (GPA) event analysis, and GPA (visual field index-based) trend analysis. Kappa agreement coefficient between methods and sensitivity and specificity for each method using expert opinion as gold standard were calculated.

Results.

The incidence of glaucoma progression was 16% to 18% in 3 years but only 3 cases showed progression with all 3 methods. Kappa agreement coefficient was high (k=0.82) between subjective expert assessment and GPA event analysis, and only moderate between these two and GPA trend analysis (k=0.57). Sensitivity and specificity for GPA event and GPA trend analysis were 71% and 96%, and 57% and 93%, respectively.

Conclusions.

The 3 methods detected similar numbers of progressing cases. The GPA event analysis and expert subjective assessment showed high agreement between them and moderate agreement with GPA trend analysis. In a period of 3 years, both methods of GPA analysis offered high specificity, event analysis showed 83% sensitivity, and trend analysis had a 66% sensitivity.

The evidence base to select a method for assessing glaucomatous visual field progression

A large number of methods have been developed for assessing glaucomatous visual field progression, but their properties have not yet been systematically evaluated. In this systematic literature review, we summarize the evidence base for selecting a method by providing answers to ten relevant questions on the variety, validity and reproducibility of methods. In total, we found 301 different methods in 412 articles. The majority of studies (54%) used the Humphrey Field Analyzer. No data have been published about the reproducibility of methods. Although there is no gold standard to assess glaucomatous visual field progression, we found evidence on validity for 48 different methods. Some methods were less capable of distinguishing between progressive and nonprogressive patients. Choosing among twelve methods is supported by some evidence of their validity. These methods still differ in sensitivity, specificity and predictive values of test results within studies comparing several methods. In conclusion, the current evidence base is not perfect. A selection should be made from a limited number of methods, according to the clinical purpose of progression assessment. Methods that quantify the rate of visual field progression seem to be the most appropriate for guiding subsequent medical actions in individual patients. Future studies should investigate whether using one method to monitor patients is superior to another method in preventing loss of quality of life.

Interobserver agreement and intraobserver reproducibility of the subjective determination of glaucomatous visual field progression

PURPOSE

To determine the extent of interobserver agreement and intraobserver reproducibility of the subjective determination of visual field progression with achromatic automated static perimetry in eyes with glaucoma, and to determine the impact of access to Glaucoma Progression Analysis (GPA) data on interobserver agreement.

DESIGN

Retrospective, observational case series.

PARTICIPANTS

Five glaucoma subspecialists from 5 different academic medical centers.

METHODS

Five visual field tests from each of 100 eyes of 83 patients being monitored for glaucoma were retrospectively identified and subjectively and independently evaluated by the 5 glaucoma subspecialists. Each set of visual fields was classified regarding progression as "none," "questionable," "probable," or "definite." More than 1 month later, the same expert observers reevaluated the same sets of visual field tests to allow determination of intraobserver reproducibility. A final subjective evaluation regarding progression was performed 3 months later, at which time the expert observers had access to the GPA printout.

MAIN OUTCOME MEASURES

The level of interobserver agreement and intraobserver reproducibility was estimated using kappa statistics on the raw classification data and also on dichotomized data in which "none" and "questionable" progression were reclassified together as nonprogressed and " probable" and "definite" were reclassified as progressed.

RESULTS

Intraobserver reproducibility was good to excellent (kappa = 0.62-0.78) for the raw data and moderate to good (kappa = 0.58-0.71) for the dichotomized data. Interobserver agreement was moderate (kappa = 0.45; 95% confidence interval [CI], 0.35-0.55) for the raw classification data and also for dichotomized data (kappa = 0.55; 95% CI, 0.46-0.64). Access to the GPA printout did not significantly change the level of interobserver agreement.

CONCLUSIONS

Five glaucoma experts had good to excellent reproducibility of the determination of visual field progression compared with earlier evaluation of the same field sets. Agreement among the experts with each other was only moderate, and did not improve when each had access to GPA results. .

Linear regression analysis of the cumulative defect curve by sectors and other criteria of glaucomatous visual field progression

PURPOSE

To analyze the progression of visual field loss using sector analysis of the cumulative defect curve and other procedures.

METHODS

Visual fields of 260 glaucomatous eyes were analyzed over 2.8 years (SD = 1.2) with at least five examinations (mean 6.9; SD = 2.0) using Octopus 311 perimeter and TOP strategy. The authors applied Threshold Noiseless Trend (TNT) program, which performs local filtering of threshold to reduce fluctuation, and analyzed five criteria: 1) a score based on significant progression of eight sectors of the cumulative defect curve (CD); 2) a score based on the presence of points (PO) with significant progression; 3) global progression (GL) of all local deviations; 4) progression of mean defect (MD); 5) progression of the square root of loss variance (sLV). The authors estimated false diagnoses (FD), randomly reordering examinations of each patient. An index of focality of progression (FI) was obtained.

RESULTS

sLV presented low sensitivity and GL low specificity. CD and PO presented twice the sensitivity of MD, often proving earlier indicators. The authors observed significant progression of some of the three criteria in 17.5% of the cases when MD <6 dB and in 20.7% when MD >6 dB (FD=5.7%). Agreement between two criteria occurred in 6.8% of cases with MD <6 dB and in 11.6% when MD >6 dB (FD=1.9%). Result reproducibility in successive examinations was observed in 9.9% of cases (FD=1.3%). Focality of progression increased with MD.

CONCLUSIONS

PO and CD indicate suspected progression earlier than MD. Reproduction of results in successive examinations or agreement between criteria allows confirmation of progression.

Clinicians agreement in establishing glaucomatous progression using the Heidelberg retina tomograph

PURPOSE

To assess the degree of concordance among clinicians reviewing 3 Heidelberg retina tomograph (HRT) printouts used to detect progression, the Moorfields regression analysis (MRA), the topographic change analysis (TCA), and trend analysis (TA), and to compare with progression identified by stereophotographs.

DESIGN

Observational cohort study.

PARTICIPANTS

We longitudinally followed 237 eyes of 168 patients (50 glaucomatous eyes, 187 glaucoma suspects) from the Diagnostic Innovation in Glaucoma Study (mean follow-up, 46.8+/-14.2 months), with a minimum of 4 HRT images (range, 4-8).

METHODS

Three experienced observers judged the presence of progression using the HRT follow-up printouts available for each HRT method of analysis (MRA, TCA, TA). The overall assessment was based on majority rule, with >or=2 graders agreeing on the classification.

MAIN OUTCOME MEASURES

Observers agreement in assessing HRT progression and agreement for progression or no progression between the HRT methods of analysis and the reference standard represented by masked stereophotograph assessment. The kappa test was used to assess the interobserver agreement.

RESULTS

In general, agreement among clinicians for subjective assessment of progression based on HRT printouts was moderate to good; agreement (kappa) ranged from 0.52 to 0.71 for MRA, 0.61 to 0.63 for TCA, and 0.45 to 0.74 for TA. Of the 237 eyes, 16 (6.8%) were found to progress during follow-up based on masked stereophotograph assessment. Agreement for progression/no progression between the HRT methods and stereophotography was similar among MRA (84.8%, agreement on 5 progressing eyes and 196 nonprogressing eyes; kappa = 0.14), TCA, (82.3%, agreement on 8 progressing eyes and 187 nonprogressing eyes; kappa = 0.2), and TA (84%, agreement on 2 progressing eyes and 197 nonprogressing eyes; kappa = 0.01).

CONCLUSIONS

Clinicians' agreement in identifying suspected glaucomatous progression using different HRT methods of analysis was moderate to good and was similar among all methods, including MRA, which is not designed to detect progression. Agreement between progression identified by HRT and masked stereophotograph assessment was poor. These results suggest that assessment of the HRT and stereophotography may be identifying different aspects of structural change.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosures may be found after the references.

Longitudinal changes in the visual field and optic disc in glaucoma

The nature and mode of functional and structural progression in open-angle glaucoma is a subject of considerable debate in the literature. While there is a traditionally held viewpoint that optic disc and/or nerve fibre layer changes precede visual field changes, there is surprisingly little published evidence from well-controlled prospective studies in this area, specifically with modern perimetric and imaging techniques. In this paper, we report on clinical data from both glaucoma patients and normal controls collected prospectively over several years, to address the relationship between visual field and optic disc changes in glaucoma using standard automated perimetry (SAP), high-pass resolution perimetry (HRP) and confocal scanning laser tomography (CSLT). We use several methods of analysis of longitudinal data and describe a new technique called "evidence of change" analysis which facilitates comparison between different tests. We demonstrate that current clinical indicators of visual function (SAP and HRP) and measures of optic disc structure (CSLT) provide largely independent measures of progression. We discuss the reasons for these findings as well as several methodological issues that pose challenges to elucidating the true structure-function relationship in glaucoma.

Interobserver agreement on visual field progression in glaucoma: a comparison of methods

AIM

To examine the level of agreement between clinicians in assessing progressive deterioration in visual field series using two different methods of analysis.

METHODS

Each visual field series satisfied the following criteria: more than 19 reliable fields, patient age over 40 years, macular threshold at least 30 dB. The first three fields in each series were excluded to minimise learning effects: the following 16 were studied. Five expert clinicians assessed the progression status of each series using both standard Humphrey printouts and pointwise linear regression (PROGRESSOR). The level of agreement between the clinicians was evaluated using a weighted kappa statistic.

RESULTS

A total of 432 tests comprising 27 visual field series of 16 tests each were assessed by the clinicians. The level of agreement on progression status between the clinicians was always higher when they used PROGRESSOR (median kappa = 0.59) than when they used Humphrey printouts (median kappa = 0.32). This was statistically significant (p = 0.006, Wilcoxon matched pairs signed rank sum test).

CONCLUSIONS

Agreement between expert clinicians about visual field progression status is poor when standard Humphrey printouts are used, even when the field series studied are long and consist solely of reliable fields. Under these ideal conditions, clinicians agree more closely about patients' visual field progression status when using PROGRESSOR than when inspecting series of Humphrey printouts.

A method for detecting progression in glaucoma patients with deep, localized perimetric defects

PURPOSE

To define a method for early detection of progressive visual field loss, based on monitoring the "healthy" component of the visual field, in glaucoma patients whose perimetric findings show the co-existence of deep scotomata and normal sensitivity areas.

METHODS

We reviewed all the "central 30-2 threshold tests" stored in the oldest of our Humphrey perimeters (a 640 VFA model, in use at the Glaucoma Service of the University Eye Clinic of Genoa since 1986). Only the perimetric findings of glaucoma patients with pure, deep, localized defects were collected for this study. In accordance with several inclusion criteria, we could select only 12 series of consecutive examinations (12 eyes of 12 patients). Each series included 12 to 20 examinations and the observation period ranged from 6 years 2 months to 9 years 4 months. Some pre-defined criteria made it possible to separate the defective component of the visual field from the "healthy" one. Then two independent "mean deviations" were calculated, one related to the "healthy" area and one to the defective one.

RESULTS

The mean deviation related to the "healthy" component of the visual field showed very little variation (0.6 to -1.3 dB) in the four series that had no increase in defects, even at the end of the observation period. However, in 7 of the 8 series with a tendency to worsen there was a small inter-test increase (-2.2 to -2.6 dB). This finding anticipated the enlargement of the scotomata, confirmed by subsequent examinations. Only in one series did the increase of the mean deviation related to the "healthy" area coincide in time with the real deterioration of the visual field, rather than anticipating it, but the inter-test interval had by chance been much longer than in the other series. The mean deviation related to the defective areas always showed very large changes in all the series, caused by the high variability of thresholds inside scotomata. This was the explanation for the large variations revealed by the "global" mean deviation too.

CONCLUSIONS

Detecting progression is still one of the major problems in evaluating perimetric results. It might be easier to achieve this goal with a method for selectively monitoring light sensitivity inside the "healthy" areas of the visual field.

Infrequent confirmation of visual field progression

OBJECTIVE

To evaluate the effects of the repeatability criteria on the detection of change in visual fields by six progression algorithms used in standard automated perimetry.

DESIGN

Retrospective, observational case series

PARTICIPANTS

Fifty-one glaucoma patients, each with multiple visual fields performed between May 1990 and December 1998, were included.

METHODS

Each patient's set of visual fields were analyzed using the glaucoma change probability, the Early Manifest Glaucoma Trial (EMGT) algorithm, a modified glaucoma change probability score, a modified EMGT score, the Advanced Glaucoma Intervention Study algorithm, and the Collaborative Initial Glaucoma Treatment Study algorithm.

MAIN OUTCOME MEASURES

The effects of repeatability on the detection of field change, the level of agreement among algorithms, as well as the number of eyes identified as changed with each algorithm, were assessed.

RESULTS

Mean follow-up was 34 months (range, 12-87 months). The average percentage of eyes with change based on three consecutive follow-up fields was 8.2% (4.0%-12.5%). However, of those showing change on the initial follow-up, this change from baseline was observed in subsequent examinations on average in 23% (18%-33%), depending on the algorithm. When change was based on just one field, four of the six algorithms noted a significantly greater number of eyes with change. The algorithms, however, did not differ significantly when confirmation of field change required two versus three consecutive follow-up visual fields.

CONCLUSIONS

Although current algorithms may help identify change, there are inconsistencies among them. We found that requiring repeatable change from baseline significantly reduces the number of changed eyes identified with each subsequent follow-up field. Identification and confirmation of change in visual fields plays an important role in helping to identify true glaucoma progression; however, the specific methods to do so have yet to be determined.

Identification of progressive glaucomatous visual field loss

In normal individuals, visual field measures are not perfectly repeatable and individual test locations exhibit both short- and long-term sensitivity variations. This physiologic variability is greatly increased in glaucoma and confounds detection of real progressive loss in visual function. Distinguishing progressive glaucomatous visual field loss from test variability therefore represents a complex task. Procedures used for detection of glaucomatous visual field progression may be broadly grouped into four categories: 1) clinical judgment, which consists of simple subjective observation of sequential visual field test results; 2) defect classification systems, whereby specific criteria are used to stratify field loss by discrete score and define progression as score change over time, such as the Advanced Glaucoma Intervention Study scoring system; 3) trend analyses, which follow test parameters sequentially over time to determine the magnitude and significance of patterns within the data, for example linear regression; and 4) event analyses, which identify single events of significant change relative to a reference examination. All of these methods demonstrate distinct benefits and drawbacks, making each useful in specific circumstances, although no single method appears universally ideal. At the present time the best method of detection of progression may be to rely upon confirmation of change at successive examinations and also by correlation of visual field changes with other clinical observations. Alternative analysis methods may become available in the near future to help identify cases of progressive loss.

The National Survey of Trabeculectomy. II. Variations in operative technique and outcome

PURPOSE

There is a considerable body of literature relating to trabeculectomy; however, there are no data representative of the national experience of trabeculectomy in the United Kingdom (UK). The Department of Health funded a national survey of trabeculectomy to establish current practice patterns and the outcome of trabeculectomy in the National Health Service (NHS). In this paper we report variations in surgical technique and the national success rate of trabeculectomy.

METHODS

A cross-sectional survey was carried out of consultant ophthalmologists performing trabeculectomy in the NHS. Participants recruited their four most recent consecutive first-time trabeculectomy cases according to study eligibility criteria and data were collected by self-administered questionnaire.

FOLLOW-UP

1 year post-trabeculectomy. Main outcome measure of success: final intraocular pressure (IOP) less than two-thirds the pre-operative IOP. Secondary outcome measures of success: final IOP less than 21 mmHg and visual field stability. Success was further defined as unqualified (excluding patients on anti-glaucoma medications at final follow-up) or qualified (including patients on anti-glaucoma medications at final follow-up). The relationship between variables characterising consultants' practice and main outcome measure was examined by chi-square test.

RESULTS

Clinical outcome data were available for 1240 (85.3%) cases. There were wide variations in operative technique. The mean post-operative IOP was 14.4 mmHg (95% CI 14.2-14.7), which is a mean reduction of 11.8 mmHg (95% CI 11.4-12.2). An unqualified success, in terms of the main outcome measure, was achieved in 66.6% of patients and a qualified success in 71.0% of cases. An unqualified success, in terms of a final IOP less than 21 mmHg, was achieved in 84.0% of cases and a qualified success in 92.0%. Visual fields were stable in 84.2%. Outcome was not related to consultants' specialist interest, level of activity, type of hospital or region.

CONCLUSIONS

The success rates reported in this paper represent the national experience of first-time trabeculectomy for open angle glaucoma in the UK. The national success rate at 1 year compares favourably with many studies in the literature. This survey provides valid and clinically relevant measures of success for the production of guidelines and standards for audit at regional, local and individual level and a baseline for the comparison of new therapies.

Mathematical and optimal clustering of test points of the central 30-degree visual field of glaucoma

PURPOSE

To determine a mathematically optimal sector pattern of the central 30 degree visual field for the follow-up of glaucomatous visual field change based on a large number of actual visual field test data of patients with glaucoma.

METHODS

Visual field test data obtained from 1,039 eyes of 1,039 patients with open-angle glaucoma (OAG) using the 30-2 program of the Humphrey Field Analyzer were used for sectorization of the central 30 degree visual field. Of the 1,039 visual field data, 698 (modeling data) were used for determining the sector pattern and 341 (testing data) for checking the sector pattern. The modeling data were further divided into three groups according to the mean deviation (MD) (MD > or = -10 dB, -20 < or = MD < -10 dB, and MD < -20 dB), and the sector pattern was constructed from visual field data of each group using a clustering procedure called VARCLUS. The testing data were used for determining the optimal sector pattern. In a separate set of repeated visual field data of 303 patients with OAG, the fluctuation of MD, sector values of each sector determined, and total deviation of each test point were calculated and compared.

RESULTS

The sector pattern constructed from visual field data of MD > or = -10 dB summarized the visual field performance most effectively. The fluctuation of the sector value of each sector was roughly 1.5 times smaller than the total deviation of each test point.

CONCLUSION

The sector pattern determined may be useful in analyses of the visual field data of patients with glaucoma.

The use of the G1 and Octosmart programs in detecting temporal changes in the visual field

PURPOSE

The Octopus program Octosmart is able to classify visual fields into six classes. In the program a horizontal bar indicates these classes, and an indicator points to the most probable position, related to the measured pathology. The width of this dashed indicator shows the range of possible fluctuations in the measurement and, therefore, its precision. This study sets out to analyse the suitability of this display mode using other visual-field index data.

METHODS

The visual fields of 83 glaucomatous eyes of 61 patients of various etiological groups and glaucoma suspects were studied for periods varying from 1 to 5 years in a retrospective study. All examinations were performed with the G1 Octopus program and analyzed with the Octosmart program. The statistical significance of linear trends of the visual-field indices, mean defect (MD) and corrected loss variance (CLV), and the class shown by the indicator (POI = position of indicator) were determined, and their regression coefficients were analyzed by means of a linear trend test as a function of time.

RESULTS

Of the sample of 83 tested eyes, a total of 18 significant trends were recorded after five examinations. All visual-field indices showed a trend towards amelioration.

CONCLUSIONS

The 18 significant trends observed must be attributed to perturbing long-term fluctuations and, despite their statistical significance, are of little clinical value. It is questionable whether an increased number of examinations per eye would have attenuated the threshold fluctuations sufficiently to make the change in field class more reliable.

Scoring systems for measuring progression of visual field loss in clinical trials of glaucoma treatment

OBJECTIVES

To compare the statistical properties of two visual field scoring systems used by clinical trials of glaucoma treatments with a view to their performance as longitudinal measures of visual field progression.

DESIGN

Cohort study comparing the scoring systems used by the Advanced Glaucoma Intervention Study (AGIS) and the Collaborative Initial Glaucoma Treatment Study (CIGTS) using pairs of visual fields (C-30-2 of the Humphrey Analyzer) spaced 1 year apart.

PARTICIPANTS

Sixty-seven eyes of 56 subjects with glaucomatous visual field loss from a natural history study of glaucoma were examined.

METHODS

Because both scoring systems are scaled so that they range from 0 to 20, the AGIS and CIGTS scores were compared by scatterplot, Pearson correlation, and the mean difference between scores. The difference between pairs of scores 1 year apart was used to compare the temporal variability of each scoring system. The proportion of subjects whose visual field "deteriorated" in 1 year by the amount specified for unconfirmed progression in each clinical trial was calculated.

MAIN OUTCOME MEASURES

Mean difference in scores 1 year apart and the variance of these differences were measured.

RESULTS

The median scores for the first of the two fields was 5.2 (interquartile range [IQR] = 9.8) for CIGTS and 5.0 (IQR = 7.0) for AGIS. The CIGTS scores were slightly larger than AGIS scores by an average of 0.5 (P = 0.06). The mean CIGTS score of the baseline fields was 0.06 (standard deviation = 3.8) higher than the mean of the second fields and 0.12 (standard deviation = 2.8) higher for AGIS. Fifteen percent of eyes had CIGTS scores that deteriorated by three or more (the cutoff for unconfirmed progression) over 1 year, while 7.5% of eyes had AGIS scores that deteriorated by four or more (the AGIS cutoff for unconfirmed progression) over the same period. Twenty-one percent improved by CIGTS criteria, while 12% improved by AGIS criteria.

CONCLUSIONS

CIGTS scores are systematically slightly higher than AGIS scores. The CIGTS scoring system has higher temporal variability than the AGIS system. The CIGTS criterion for unconfirmed progression over a 1-year period is likely to lead to higher estimated rates of progression than the AGIS criteria. This is mostly because of the difference in the definition of progression between the two study criteria.

Inter-observer and intra-observer agreement in the interpretation of visual fields in glaucoma

Visual field changes are one of the main parameters used to monitor progression of glaucoma. This study assesses the degree of intra-observer and inter-observer agreement among nine observers in grading visual fields in glaucoma patients using a visual field system previously described by Jay. The results show a median inter-observer agreement of 61% (median kappa = 0.52) and a median intra-observer agreement of 72% (median kappa = 0.65). This system for grading fields in glaucoma has a high degree of intra-observer agreement, suggesting it is a useful system for longitudinal follow-up of patients by a single observer. The higher degree of disagreement between observers points to the need for careful pretraining of observers in clinical management and research where the results from visual field examinations are to be graded by more than one clinician.

Early detection of visual field progression in glaucoma: a comparison of PROGRESSOR and STATPAC 2

AIM

To compare the performance of PROGRESSOR (pointwise linear regression) and STATPAC 2 (comparison with baseline values) in detecting early deterioration in the visual fields of glaucoma patients.

METHODS

Visual field series from 19 untreated normal tension glaucoma eyes which were deteriorating on clinical grounds were analysed by PROGRESSOR and STATPAC 2. Progression criteria for PROGRESSOR were (1) inner points: slope < -1 dB/year, p < 0.05 and (2) edge points: slope < -2 dB/year, p < 0.05. Criteria for STATPAC 2 were p < 0.05 change probability for any point on three consecutive fields. Detection time was defined as the time interval between the initial field and the first field in which at least one progressing point was identified. Detection times produced by the two techniques were compared.

RESULTS

PROGRESSOR and STATPAC 2 agreed on progression in all 19 eyes. Mean detection time for PROGRESSOR was 1.077 (SD 0.985) years and for STATPAC 2 was 2.161 (1.357) years. PROGRESSOR detected progression sooner than STATPAC 2 in 18 eyes (p < 0.01), Wilcoxon matched pairs signed rank test). PROGRESSOR detected progression earlier by a mean of 1.085 (0.936) years.

CONCLUSIONS

PROGRESSOR consistently detected progression earlier than STATPAC 2. The PROGRESSOR software is a useful tool for the early detection of visual field deterioration in glaucoma.

Comparison of methods to detect visual field progression in glaucoma

OBJECTIVE

The purpose of the study is to develop alternative statistical approaches for evaluating the trend of visual field series over time and to compare the results to human observers.

DESIGN

Retrospective analysis of visual field results.

PARTICIPANTS

Eighty-three eyes of 83 patients (phakic or pseudophakic) with open-angle glaucoma and 5 or more eligible fields were included in the study.

INTERVENTION

Three experienced observers independently reviewed the field series to determine stability or progression.

MAIN OUTCOME MEASURES

The following additional methods to determine progression of visual field loss were used: (1) pointwise univariate regression analysis and a glaucoma change analysis; (2) univariate regression analysis on visual field indices mean deviation, corrected loss variance, and glaucoma pattern index; (3) pointwise multivariate regression analysis with fixed effects on panel data; and (4) clusterwise multivariate regression analysis with fixed effects on panel data. The results of different statistical methods were compared by determining the pairwise agreement (Cohen's weighted kappa) between each technique and three experienced observers.

RESULTS

Patients were observed for a mean (+/-standard deviation) of 5.6 (+/-1.4) years. The visual fields of 27 (33%) and 56 (67%) eyes were considered to have progressed or remained stable, respectively, based on agreement of at least 2 of 3 observers. Univariate regression analysis on visual field indices was not useful for detection of visual field progression. Pointwise and clusterwise regression analyses with fixed effects on panel data performed as well as pointwise univariate regression analysis compared with human observers (kappa = 0.52, 0.53, and 0.55, respectively). Both methods showed better agreement with human observers than with glaucoma change analysis (kappa = 0.41).

CONCLUSIONS

A new statistical model, multivariate regression analyses with fixed effects on panel data, is an appropriate method to evaluate the course of visual field series over time and shows reasonable agreement with experienced observers and pointwise univariate regression analysis.

Estimating progression of visual field loss in glaucoma

PURPOSE

The authors estimated the prevalence and rates of progressive visual field loss in glaucoma patients followed annually for a median of 6.3 years.

METHODS

Linear regression was used to estimate rates of progression of mean deviation, corrected pattern standard deviation (CPSD), clusters of locations based on the Glaucoma Hemifield Test (GHT), and location specific changes in C-30-2 fields of the Humphrey Analyzer.

RESULTS

Sixty-seven eyes of 56 patients whose first two consecutive fields were abnormal on GHT were included. Almost all patients were under treatment or had undergone surgery for glaucoma. Visual field deteriorated in 19 (28%) eyes based on worsening of one or more CPSD, GHT clusters, or individual test locations (regression slopes significantly different from zero). Corrected pattern standard deviation deteriorated in 5 eyes, at least one GHT cluster deteriorated in 17 eyes, and one or more individual test locations deteriorated in 15 eyes. For those whose visual field deteriorated, CPSD increased by 0.9 dB/year. Glaucoma Hemifield Test clusters declined by between 1.4 and 2.4 dB/year. Deterioration at individual locations ranged from 1.0 to 5.0 dB/year. Age, but not baseline visual field severity, was predictive of further visual field loss. The odds ratio for the association between progressive visual field loss and thinning of the nerve fiber layer was 1.81 (95% confidence interval: 0.52, 6.33), and 3.78 (95% confidence interval: 0.80, 18.16) for the association between progressive visual field loss and optic disc changes during follow-up based on masked photograph readings.

CONCLUSIONS

Less than one in three eyes of patients with glaucoma had any progressive field loss. Average changes in threshold sensitivities of less than 1 dB/year could not be detected with seven fields done over 6 years. Larger changes or increased frequency of visual field testing would need to occur before smaller changes could be detected statistically.

Improving the prediction of visual field progression in glaucoma using spatial processing

PURPOSE

The authors show how the predictive performance of a method for determining glaucomatous progression in a series of visual fields can be improved by first subjecting the data to a spatial processing technique.

METHOD

Thirty patients with normal-tension glaucoma, each with at least ten Humphrey fields and 3.5 years of follow-up, were included. A linear regression model of sensitivity against time of follow-up determined rates of change at individual test locations over the first five fields (mean follow-up 1.46 years; standard deviation = 0.08) in each field series. Predictions of sensitivity at each location of the field nearest to 1 and 2 years after the fifth field were generated using these rates of change. Predictive performance was evaluated by the difference between the predicted and measured sensitivity values. The analysis was repeated using the same field data subjected to a spatial filtering technique used in image processing.

RESULTS

Using linear modeling of the unprocessed field series, at 1 year after the fifth field, 72% of all predicted values were within +/- 5 dB of the corresponding measured threshold. This prediction precision improved to 83% using the processed data. At the 2-year follow-up field, the predictive performance improved from 56% to 73% with respect to the +/- 5 dB criterion.

CONCLUSIONS

Predictions of visual field progression using a pointwise linear model can be improved by spatial processing without increased cost or patient time. These methods have clinical potential for accurately detecting and forecasting visual field deterioration in the follow-up of glaucoma.

Visual field progression: comparison of Humphrey Statpac2 and pointwise linear regression analysis

BACKGROUND

Humphrey Statpac2 "glaucoma change probability analysis' is a widely available analysis technique to aid the clinician in the diagnosis of glaucomatous visual field deterioration. A comparison of this technique with the more recently described pointwise linear regression analysis (PROGRESSOR) is given.

METHODS

Series of visual field data from a group of nine eyes of nine patients with normal-tension glaucoma were selected. Each series had 16 fields with mean follow-up of 5.7 years (SD 0.6 years). Statpac2 "glaucoma change probability analysis' was used to define test locations that had unequivocally deteriorated in the last three fields of each series. The accuracy of both Statpac2 and PROGRESSOR in providing early detection of these deteriorated locations was assessed.

RESULTS

The sensitivity and specificity of the two techniques in predicting deteriorated locations were similar when a rate of luminance sensitivity loss of faster than 1 dB/year (2 dB/year for outer locations beyond 15 deg of eccentricity) with a slope significance of P < 0.10 was used as the regression definition of deterioration. The difficulties of comparing two techniques in the early diagnosis of field progression without a true external standard for field loss are illustrated.

CONCLUSIONS

PROGRESSOR closely emulates the performance of Statpac2 in detecting sensitivity deterioration at individual test locations. This new technique, which uses all available data in a field series and gives the rate of sensitivity loss at each location, may provide a clinically useful method for detecting field progression in glaucoma.

Relative dispersion analysis enhances perimetric sensitivity

Objective identification of minor visual field defects is problematic. A possible solution is to examine spatial correlations by means of relative dispersion analysis, a tool of fractal analysis. We studied patients with glaucoma, previous optic neuritis, chiasmal compression and lesions of the brain hemispheres, using high-pass resolution perimetry. One-hundred visual field records were drawn consecutively for each category and ranked according to severity of defects. Records with scores ranking below the 35th percentile, i.e. those with the smallest field defects, were analysed. Relative dispersion analysis recognized 1.3-2.4 times more abnormal subjects than did pattern standard deviation. A previously described form index was intermediate in sensitivity. Specificity was 96%. Relative dispersion analysis appears to capture a novel aspect of visual field abnormality, with good sensitivity and specificity. The analysis is easily performed.

Determining progressive visual field loss in serial Humphrey visual fields

PURPOSE

To evaluate the three commercially available computerized statistical algorithms for determining progression of glaucomatous field loss in serial Humphrey fields.

METHODS

Serial Humphrey fields of 102 ocular hypertensive and glaucomatous eyes were analyzed by linear regression analysis of Statpac, glaucoma change probability and Progressor Programme, and the results compared with a retrospectively determined clinical outcome.

RESULTS

Linear regression analysis of Statpac identified progression in 11 field series, glaucoma change probability identified progression in 23 field series, and Progressor Programme identified progression in 27 field series. Clinical outcome identified much less-progression than any of the three algorithms, determining that only seven eyes showed deterioration. There was a wide variation of results from analysis of the same group of field series by the three different algorithms. In contrast to the other two algorithms, which detected progression in similar percentages of both ocular hypertension and glaucoma field series. Progressor Programme detected progression predominantly in glaucoma field series.

CONCLUSIONS

There was a high degree of variability among the three different algorithms for determining visual field progression, and none of them correlated well with clinical impression.

The Glaucoma Laser Trial (GLT): 6. Treatment group differences in visual field changes. Glaucoma Laser Trial Research Group

PURPOSE

To determine the differences in visual fields during 42 months (3 1/2 years) of follow-up between eyes treated with argon laser trabeculoplasty first and eyes treated with topical medication first in patients with newly diagnosed primary open-angle glaucoma.

METHODS

Visual field examinations were obtained at enrollment, three and six months, and at six-month intervals thereafter during follow-up of 271 patients enrolled in the Glaucoma Laser Trial. Numeric analyses of the examination results, including global indices and patterns of localized changes, as well as masked subjective clinical impression, were used to compare the two treatment groups.

RESULTS

The mean threshold for eyes treated with laser trabeculoplasty first was 0.3 dB better than that for eyes treated with topical medication first averaged over follow-up (95% confidence interval, [-0.1, 0.7]; P = .17). More eyes treated with topical medication first (82 [31%] of 261 eyes) than eyes treated with laser trabeculoplasty first (61 [23%] of 261 eyes) had confirmed localized deterioration at least once during follow-up (P = .02). Improvement was nearly twice as common as deterioration on masked subjective impression in both groups through 30 months (2 1/2 years). Eyes treated with laser trabeculoplasty first were judged to have slightly more improvement and slightly less deterioration than eyes treated with topical medication first.

CONCLUSION

During follow-up, measures of visual field status for eyes treated with laser trabeculoplasty first were slightly better than those for eyes treated with topical medication first. Statistical significance was attained for only some of the differences, and the clinical implications of such small differences are not known.

Technique for determining glaucomatous visual field progression by using animation graphics

The first two visual field examinations of each of 16 glaucoma suspects and 16 glaucoma patients, who were followed up for a median of 7.46 years with seven examinations, were averaged. Three-dimensional color-coded images of the visual field were then generated. After correcting for test-retest variability, the images, as well as additional interpolated images, were aligned and presented in rapid succession to create an animation sequence. Five glaucoma specialists trained themselves to classify the visual fields as either progressing or not progressing. The interobserver and intraobserver agreement rates were then calculated. Perfect agreement (100% concordance) between the observers was obtained for 18 (56.3%) subjects, and at least 80% concordance was obtained for 27 (84.4%) subjects. Of the eight sequences that were repeated to test intraobserver agreement, four of the five observers had a value of at least 87.5%.

The primary treatment trial: changes in the visual field analysis by computer-assisted perimetry

Computer-assisted perimetry was performed 6-monthly on patients entered into the Primary Treatment Trial using the 30-2 program. Those patients with a minimum of five visual fields had pointwise linear regression analysis (Progressor) carried out. Two consecutive slopes significant at the 0.05 level were taken as evidence for change, either improvement or worsening. Forty-eight patients in the surgery group, 40 in the medical group and 20 in the laser group were suitable for analysis. A comparison of the eyes showing 'improvement' or 'worsening' in the surgery and medicine groups showed no difference between them. This lack occurred despite a significant difference in intraocular pressure between the two groups. The reasons for this lack of difference are discussed.

Sectorization of the central 30 degrees visual field in glaucoma

PURPOSE

To determine an optimal sector pattern of the central 30 degrees visual field in glaucoma by mathematically analyzing the visual field data of primary open-angle glaucoma (POAG) without any assumption such as the retinal nerve fiber layer anatomy.

METHODS

One hundred three visual fields of the 30-2 program of the Humphrey Field Analyzer obtained from 103 POAG patients of early to moderately advanced stage were included. Based on the interpoint correlation of deviation of the measured threshold value from the age-corrected normal reference value (total deviation, STATPAC), test points of the 30-2 program were mathematically clustered using the VARCLUS procedure, a new clustering algorithm developed by the SAS Institute. The sector value, which summarizes the visual field performance of the clustered test points (sector), also was calculated.

RESULTS

The 30 degrees central visual field was divided into 15 sectors consisting of at least 3 points. The distribution of sectors was compatible with the projection of nerve fiber layers. There was no sector extending over the horizontal meridian, but the sector pattern was not completely symmetrical around it. Linear regression analysis of the sector values against the mean deviation (STATPAC) suggested that the index is useful in following visual field performance of each sector.

CONCLUSION

The sector pattern and sector values obtained were considered useful in studying the visual field data of glaucoma.

A perimetric nerve fiber bundle map

The so-called 'nerve fiber bundle defects' in the visual field are considered as being the functional equivalent of retinal nerve fiber bundle damage. This study examines the course and the position of scotoma borders in 159 visual fields with complete nerve fiber bundle defects using high resolution test point patterns. The scotoma border lines were averaged geometrically. The result is presented in the form of mean vectors. This functional map of nerve fiber lines shows a close correspondence to the anatomical course of the retinal nerve fibers. To make the result suitable for topographical analysis of visual fields, 21 areas ('Perimetric Nerve Fiber Bundles') are defined. Certain arbitrary assumptions were necessary because data were to scanty in some nasal regions. Also, the lines had to be fitted into a rectangular grid to be transferable to standard perimetric patterns. Topographical analysis on the basis of these areas may be useful for the evaluation of visual fields, particularly those of glaucoma.