COMPARISON OF CONVENTIONAL AUTOMATED PERIMETRY, SHORT-WAVELENGTH AUTOMATED PERIMETRY AND FREQUENCY-DOUBLING TECHNOLOGY IN THE ASSESSMENT OF PATIENTS WITH MULTIPLE SCLEROSIS

GUIDO CORALLO,1 PAOLO CAPRIS,1 STEFANO BARABINO,1 SIMONETTA CICINELLI,1 ANTONIO MASTROMARINO,1 FABIO BANDINI,2 ANTONIO UCCELLI,2 CLAUDIO SOLARO,2 SABRINA ZANARDI3 and

GIOVANNI CALABRIA1

1Eye Clinic and 2Neurological Clinic, Department of Neurological and Visual Sciences; 3Biostatistic Unit, Department of Health Sciences; University of Genoa, Genoa, Italy

Abstract

Purpose: To compare conventional, achromatic, automated perimetry (CAP), short-wavelength automated perimetry (SWAP), and frequency-doubling technology (FDT) perimetry in a group of patients affected by multiple sclerosis (MS) with or without a history of optic neuritis. Material and methods: Thirty eyes of 15 patients (five males, ten females, average age 38 ± 7 years), with a confirmed diagnosis of MS underwent CAP, SWAP (Humphrey 750 – II VFA, program 30-2, full-threshold strategy) and FDT perimetry (program N-30). The patients were selected from the Neurological Clinic of our department. Sixteen eyes had no history of ocular involvement and a completely negative ophthalmological examination. They were matched with a control group of ten healthy volunteers (four males, six females, average age 31 ± 10 years). The mean deviation (MD) and pattern standard deviation (PSD) indices were taken into consideration. The t test was used for comparison between these groups. Fourteen eyes had, on the contrary, a history of optic neuritis. In the analysis of this group, correlations were evaluated (Spearman’s rank test) among the indices MD and PSD of the three techniques. Results: Comparison between the 16 eyes of the MS patients without signs or symptoms of ocular involvement and the group of ten normal controls eyes showed significant differences for both the studied parameters (MD and PSD) with all three techniques: CAP MD (p = 0.0029); CAP PSD (p = 0.0001); SWAP MD (p = 0.0014); SWAP PSD (p = 0.0001); FDT MD (p < 0.0001); and FDT PSD (p = 0.0598). When considering the trend of MD and PSD with the three techniques in the group of MS patients who had a history of optic neuritis, a significant correlation was only found between CAP MD and SWAP MD (p = 0.0057). The other correlations were not significant. Conclusions: SWAP and FDT perimetry are two non-conventional perimetric techniques that were originally developed for the early detection of glaucomatous damage. The results of this study also demonstrate their efficacy for the assessment of patients with neuroophthalmological disorders, such as optic neuritis in patients with MS.

Address for correspondence: Guido Corallo, MD, Dipartimento di Scienze Neurologiche e della Visione dell’Università di Genova, Sezione di Clinica Oculistica, Ospedale San Martino, Pad. 9, Largo R. Benzi 10, I-16132 Genova, Italy. Email: guidocorallo@libero.it

Perimetry Update 2002/2003, pp. 97–101

Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002

edited by David B. Henson and Michael Wall

© 2004 Kugler Publications, The Hague, The Netherlands

Introduction

The earliest symptoms of multiple sclerosis (MS) are often represented by ocular disorders. Optic neuritis is a well-known manifestation, but it is not the only one. Ocular motility disorders with sudden diplopia may also occur. Conventional, achromatic, white-on-white automated perimetry (CAP) is generally performed when suspecting optic neuritis. Threshold tests (central 30° field) allow identification of typical perimetric defects, i.e., central and/or paracentral scotomas, which reflect damage to the optic nerve fibers. In the past few years, various new non-standard perimetric techniques have been introduced, whose original goal was the early detection of glaucomatous damage. Two of them are in widespread use: short-wavelength automated perimetry (SWAP) and, more recently, frequency-doubling technology (FDT) perimetry. Longitudinal studies have shown that SWAP defects in patients with ocular hypertension are predictive of the subsequent glaucomatous visual field (VF) loss observed with CAP.1-3 Moreover, SWAP deficits in patients with glaucoma are larger than areas of loss with CAP, and the rate of progression is about twice as great with SWAP than with CAP. FDT perimetry is a more recent technique, developed by Johnson and Samuels.4 The test is completely different from conventional perimetry. It does not utilize simple spot stimuli, but rather rapidly alternating sinewave gratings. As a glaucoma screening device, the sensitivity of FDT perimetry proved to be similar to that of CAP, with excellent specificity.5-7 There are few published papers describing the use of these techniques in neuro-ophthalmological disorders.8-11 The aim of this study was to evaluate the efficacy of SWAP and FDT in the assessment of patients with MS, and to compare the findings to those obtained with CAP.

Material and methods

Fifteen MS patients (five males, ten females, average age 38 ± 7 years) were recruited from the Neurological Clinic of our department. Both eyes were included in the study and underwent VF testing with three different techniques: CAP, SWAP, and FDT. Sixteen eyes had no history of ocular involvement and a completely negative ophthalmological examination. Fourteen eyes had a history of optic neuritis. Ten healthy, agematched volunteers (four males, six females, average age 31 ± 10 years) were enrolled as a control group, and only the right eyes were considered for statistical analysis (ten eyes). All subjects in both groups gave their consent for participation in the present study and underwent a complete ophthalmological examination at the Eye Clinic of our department. Subjects with myopia or hyperopia > 3 diopters or astigmatism > 2 diopters were excluded, as were subjects affected by any general (other than MS) or ocular disease that might influence the perimetric results. Each technique was repeated twice, on separate days, and only the results of the second examination were taken into consideration. CAP and SWAP were performed with the same perimeter (Humphrey 750 II HFA, Humphrey Instruments, San Leandro, CA) using the full-threshold 30-2 test. FDT perimetry was performed using an FDT instrument (Humphrey-Welch Allyn, Dublin, CA) using the threshold program N-30. This program tests 19 locations (one central location, 16 locations within 25°, and two additional nasal locations within

30°). The MD and PSD indices of the three techniques were evaluated. The t test was utilized to compare the asymptomatic MS group and the normal group. For the symptomatic group, a correlation was evaluated (Spearman’s rank test) among MD and PSD of all three techniques.

Results

There was no statistically significant age difference between the MS patients and normal controls. The comparison between the asymptomatic MS group and the control group showed significant differences for both MD and PSD with all three techniques, except for FDT PSD. The significance levels are reported here: CAP MD (p = 0.0029); CAP PSD (p = 0.0001); SWAP MD (p = 0.0014); SWAP PSD (p = 0.0001); FDT MD (p < 0.0001); and FDT PSD (p = 0.0598). The trend of MD and PSD with the three techniques in the symptomatic MS patients only showed a significant correlation between CAP MD and SWAP MD (p = 0.0057). No other correlation was found.

Discussion

SWAP and FDT are two non-standard perimetric techniques which were originally developed for the early detection of glaucomatous damage. The first was introduced about ten years ago in two independent laboratories, at the University of CaliforniaDavis, Sacramento, and the University of California-San Diego.1,2 The instruments for SWAP are the same as for CAP, but SWAP adopts a high luminance (100 cd/mq) yellow background selectively to adapt the green and red pathways and simultaneously to suppress rod activity. Targets are blue (440 nm) and they preferentially stimulate the short-wavelength sensitive pathway. SWAP isolates the short-wavelength sensitive cones (S-cones) and their connections. The test is processed by the small, bistratified blue-yellow ganglion cells. FDT perimetry uses large low spatial frequency counter phased (25 Hz) gratings as stimuli. The combination of low spatial and high temporal frequency of the bars generates the frequency-doubling illusion when presented at suprathreshold levels. The FDT test isolates a subclass of magnocellular ganglion cells (My cells). All three techniques utilized in this study (CAP, SWAP, and FDT) proved to be sensitive in detecting MS patients, but sensitivity was higher for SWAP and FDT, compared to CAP. A possible explanation for this is that conventional, achromatic perimetry does not isolate the function of a particular subclass of retinal ganglion cells. It simultaneously stimulates most of them, without any selective mechanism. Particularly in cases of mild damage, involving a small proportion of ganglion cells, the impairment of their function might be masked by the normal function of healthy ganglion cells.

SWAP and FDT are selective techniques. The former is processed by the parvocellular component of ganglion cells, in particular by the small, bistratified, blue-yellow ganglion cells, and isolates the short-wavelength sensitive cones (S-cones). The latter isolates a subclass of the magnocellular ganglion cells (My cells). Both these cellular subgroups comprise a small proportion of the total population of gan-

glion cells. The first represents approximately 9%, and the second 3-5%, of all ganglion cells. Their sparse representation may explain their enhanced sensitivity in comparison to CAP.

A statistically significant difference was found, between the asymptomatic MS group and the normal group for both the MD and PSD indices of the three techniques. The level of significance, however, was highest for FDT MD (p < 0.0001), followed by SWAP MD (p = 0.0014) and CAP MD (p = 0.0029). This means that FDT and SWAP are more sensitive in detecting visual field damage in MS patients, compared to CAP. PSD showed the same level of significance (p = 0.0001) for CAP and SWAP, and just below significance level for FDT (p = 0.0598). This may mean that light sensitivity is less homogeneous in asymptomatic MS patients, when compared to normal subjects. In the group of MS patients with a history of optic neuritis, CAP MD and SWAP MD showed similar trends, while no significant correlation was found for PSD. Neither FDT MD nor FDT PSD showed any significant correlations with the corresponding CAP and SWAP indices, confirming the peculiarity of this technique.

Conclusions

SWAP and FDT perimetry are two non-conventional perimetric techniques which were originally developed for early detection of glaucomatous damage. The results of this study also demonstrate their efficacy for the assessment of patients with neuroophthalmological disorders, such as optic neuritis in patients with MS. FDT perimetry, in particular, proved to be an easy, fast, and sensitive technique. The test is well accepted by patients and has a very short duration (about five to six minutes). It may be useful for screening purposes when examining patients affected by MS. SWAP is itself a sensitive technique in these cases, but has some disadvantages: it is timeconsuming (a few minutes for adaptation to the yellow bright screen have to be added to the mean duration of the examination, which is similar to that of CAP); moreover, the fatigue effect is higher and the learning effect longer, compared to CAP.

References

1.Sample PA, Weinreb RN: Color perimetry for assessment of primary open-angle glaucoma. Invest Ophthalmol Vis Sci 31:1869-1875, 1990

2.Johnson CA, Adams AJ, Casson EJ, Brandt JD : Blue-on-yellow perimetry can predict the development of glaucomatous visual field loss. Arch Ophthalmol 111:645-650, 1993

3.Johnson CA, Adams AJ, Casson EJ, Brandt JD: Progression of early glaucomatous visual field loss as detected by blue-on-yellow and standard white-on-white automated perimetry. Arch Ophthalmol 111:651-665, 1993

4.Johnson CA, Samuels S: Screening for glaucomatous visual field loss with frequency-doubling perimetry. Invest Ophthalmol Vis Sci 38:413-425, 1997

5.Cello KE, Nelson Quigg JM, Johnson CA: Frequency doubling technology perimetry for detection of glaucomatous visual field loss. Am J Ophthalmol 129:314-322, 2000

6.Burnstein Y, Ellish NJ, Magbalon M, Higginbotham EJ: Comparison of frequency doubling perimetry with Humphrey visual field analysis in a glaucoma practice. Am J Ophthalmol 129:328-333, 2000

7.Iester M, Mermoud A, Schnyder C: Frequency doubling technique in patients with ocular hypertension and glaucoma: correlation with Octopus perimeter indices. Ophthalmology 107:288-294, 2000

8.Keltner J, Johnson CA: Short-wavelength automated perimetry in neuro-ophthalmologic disorders. Arch Ophthalmol 113:475-481, 1995

9.Fujimoto N, Adachi-Usami E: Use of blue-on-yellow perimetry to demonstrate quadrantanopia in multiple sclerosis. Arch Ophthalmol 116:828-829, 1998

10.Fujimoto N, Adachi-Usami E: Frequency doubling perimetry in resolved optic neuritis. Invest Ophthalmol Vis Sci 41:2558-2560, 2000

11.Wall M, Neahring RK, Woodward KR: Sensitivity and specificity of frequency doubling perimetry in neuro-ophthalmic disorders: a comparison with conventional automated perimetry. Invest Ophthalmol Vis Sci 43:1277-1283, 2002

FAST ‘TOP’ AND NORMAL BRACKETING STRATEGY IN GLAUCOMA

FRITZ DANNHEIM

Department of Ophthalmology, General Hospital Harburg, Hamburg, Germany

Abstract

The reproducibility of visual fields and the conformity of results have been evaluated in 27 glaucomatous eyes (21 subjects) examined twice with normal bracketing and twice with the TOP strategy of the Octopus 1-2-3 perimeter using program G1X. Visual field defects were classified as borderline (n = 4), mild (n = 5), moderate (n = 5) and severe (n = 13). Results: Test duration with the normal strategy was 11.2 ± 0.68 and with the TOP strategy 2.25 ± 0.18 minutes, a gain in time of 80 ± 2%. Reproducibility for each of the two strategies, as represented by the index short-term fluctuation (SF), was 4.04, SD 1.05 and 4.19, SD 1.23 for the normal and TOP strategies, respectively. Three eyes with borderline or mild visual field defects had a considerably larger SF for TOP. Reproducibility, as represented by the correlation coefficient (r) of relative sensitivity values within each strategy, was, in moderate or severe cases, slightly higher and more uniform for the normal strategy than for TOP, with values of around 0.8 (excluding one outlier).

Conformity of results for both strategies, calculated as the SF between the mean sensitivity values for either strategy, was, on average, 3.93, SD 0.94, thus being slightly smaller than the within-strategy SF. Conformity, as represented by the correlation coefficient between mean sensitivity values for either strategy, was, in eyes with moderate and severe defects, in the range of 0.81 to 0.91 (mean, 0.84; SD, 0.09). The independent regression line of this correlation showed a slope of 0.82-0.91, and an intercept of 0.35-2.93. Severe functional loss is underestimated with the TOP strategy, and the best sensitivity values are depressed in comparison with the normal strategy. This attenuation of defect values is reflected by a slightly lower loss variance (LV) for TOP, whereas mean defect is the same for both. Comparison of fields by inspection showed an excellent to reasonably good coincidence of findings for both strategies. Fields with moderate and severe defects had a better coincidence than those with smaller defects. The most disparate results are presented. The TOP strategy proved sufficiently reliable for routine use in glaucoma.

Introduction

The reproducibility of visual fields with the fast ‘tendency oriented perimetry’ (TOP) strategy has been found to be similar to that of the normal bracketing strategy.1 The aim of this study was assessment of the reproducibility of these strategies in eyes with various stages of glaucomatous damage. The second aim was to evaluate the conformity of the results with the two methods.1,2

Address for correspondence: Fritz Dannheim, MD, Dangersener Weg 2, D-21224 Rosengarten, Germany. Email: dannheim@uke.uni-hamburg.de

Perimetry Update 2002/2003, pp. 103–114

Proceedings of the XVth International Perimetric Society Meeting, Stratford-upon-Avon, England, June 26–29, 2002

edited by David B. Henson and Michael Wall

© 2004 Kugler Publications, The Hague, The Netherlands

Material and methods

Twenty-seven eyes of 21 subjects with open-angle or pigmentary glaucoma, aged between 44 and 81 years, who had previous perimetric experience, were examined repeatedly with the Octopus 1-2-3 perimeter using program G1X. A sequence of four tests, including two with the normal strategy (four stages), and two with the TOP strategy, were performed on the same day (examples shown in Figs. 1-4). The order of the tests was randomized and rest periods were taken between each test. All visual fields were included in the analysis and classified by visual inspection and by the average of mean defect (MD) and loss variance (LV) of the two tests with the normal strategy (Fig. 5). The four classes included borderline (n = 4, Fig. 1), mild (n = 5, Fig. 2), moderate (n = 5, Fig. 3), and severe glaucomatous visual field loss (n = 13, Fig. 4). All 27 eyes in Figure 5 are ranked on the x- axis according to their ‘loss grade’ (MD as a percentage of MD 25 dB plus LV as a percentage of LV 100). This grade ranged from 13-156%. A borderline grade represented results of <27%, a mild grade 27-48%, a moderate grade 49-90%, and a severe grade >90%.

Results

Test duration with the normal strategy was 11.2 ± 0.68 minutes, and with the TOP strategy 2.25 ± 0.18 minutes – a time gain of 80 ± 2%.

Reproducibility for both strategies (Fig. 6) was assessed by calculation of the shortterm fluctuation (SF) using the root mean square of differences of n = 59 relative sensitivity values ( ∆S ) within each pair of repeated tests.3

√ n

Σ (∆S)2

i=1

SF = —————

n

All values for SF were twice as high as the conventional within-test index for SF.3 The 27 eyes are sorted into different grades of loss in Figure 5. SF was in the same order with both techniques, although with the normal strategy it was more uniformly distributed (mean, 4.04, SD 1.05; range, 2.59-6.41) than with the TOP strategy (mean, 4.19; SD, 1.23; range, 2.19-6.85). Two of the four eyes with borderline fields (one subject, eyes #3 and 4, see Fig. 1), one eye with mild (#9, Fig. 2), and one eye with severe defects (#21, Fig. 4), had considerably larger SF with the TOP technique. The SF of eye #21 with TOP was in the range of all the other eyes, SF with the normal strategy was, however, exceptionally small. SF was on average 0.16 higher with the TOP strategy in all 27 eyes , 0.65 higher with TOP in the combined borderline and mild grades of visual field loss (n = 9), and 0.09 lower in the moderate and severe grades of visual field loss (n = 18).

Reproducibility of the more densely affected fields was estimated with the correlation coefficient (r) between the point-wise test/retest sensitivity values of each strategy (Fig. 7). For the 18 moderate and severely affected eyes, r was approximately 0.8, except for one outlier with both strategies in the same eye (#14, Fig. 3). With the

glaucoma in strategy bracketing normal and ‘TOP’ Fast

Fig. 1. Visual field results plotted as defect curve, as interpolated gray symbols and numerical values of relative sensitivity. The two upper results are from the normal strategy, while the two lower ones are from the TOP strategy. The fields on the left are the first ones, with the retest values on the right. Results are for eye #4 (‘borderline’). The normal strategy results are borderline with some depression in the lower nasal periphery and good reproducibility (SF, 2.9 dB). The TOP strategy shows more diffuse loss, isolated depression inferior nasally on retest, and less damage (SF for TOP 6.85, conformity 4.37 dB).

105

glaucoma in strategy bracketing normal and ‘TOP’ Fast

Fig. 3. Arrangement as in Figure 1 for eye #14 (most affected field of category ‘moderate’). The normal strategy shows diffuse loss and a dense nerve fiber defect in the lower hemifield, with the upper nasal quadrant affected only in the first test (reproducibility SF, 6.41 dB). The TOP strategy shows more diffuse loss and the local loss is less pronounced inferiorly in the first test, and less extended in the second test (SF, 4.99 dB). The test/retest results for the TOP upper hemifield are slightly different.

107

Fig. 7. Reproducibility expressed as the correlation coefficient (r) between local relative sensitivity values of double examination both with the normal (upper half) and TOP (mirrored lower half) strategies for 27 eyes that have been sorted as shown in Figure 5. Eye #14, marked with an asterisk, is an exception.

normal strategy (Fig. 7, upper half), r is more uniformly distributed (mean, 0.84; SD, 0.08) than with TOP (mean, 0.81; SD, 0.12).

Conformity describes the similarity of findings between the two strategies. It has been evaluated by calculation of SF between the mean sensitivity values of each strategy (Fig. 8, upper half). This inter-strategy SF ranges from 2.02-5.67 (mean, 3.93; SD, 0.94). Eyes with more severe defects presented with a slightly higher inter-strat- egy fluctuation. Conformity of the results was also estimated with the correlation coefficient (r) between point-wise mean sensitivity values of each strategy (Fig. 8, lower half). In the 18 eyes with more severely affected fields, the correlation coefficient was within the range of 0.81-0.91 (mean, 0.84; SD, 0.09) except for one outlier of r = 0.52 (eye #14, Fig. 3), one of the five eyes with the highest SF.

The independent regression line of this correlation (Fig. 9), for the 18 eyes with more severely affected fields, has slopes of 0.82-0.91, excluding the same outlier (eye #14, Fig. 3), with a slope of 0.58 and intercepts between 0.35 and 2.93, with no apparent trend with severity. Eye #14 (Fig. 3) is one of the four eyes with the largest intercept. The characteristics of the independent regression line reveal less depression of sensitivity in deeper defects and more depression in shallower defects with TOP. The associated flattening of the defect curve is apparent in two of the four examples (Figs. 2 and 3).

The correlation of LV (mean of tests 1 and 2) for the two strategies is good (Fig. 10, r = 0.93, slope 0.93, intercept 3.28). TOP has a slightly lower variance (mean of difference, 7.93; SD of difference, 10.75). The smaller loss variance in eye #22 is simply due to a widespread absolute defect, which also accounts for the higher MD

Fig. 10. Mean LV of double examination with normal sensitivity (upper half) and TOP strategy (mirrored lower half) for 27 eyes that have been sorted as shown in Figure 5. Eye #14, marked with an asterisk, is an exception.

(Fig. 5). Correlation of MD (mean of tests 1 and 2) for the two strategies is excellent (Fig. 11, r = 0.97, slope 0.97, intercept 0.27). MD for TOP is, on average, slightly lower than for the normal strategy (mean of difference, 0.59; SD of difference, 1.44).

A comparison of fields by inspection shows an excellent to reasonably good coincidence of findings with both strategies. Fields with dense defects had a better coincidence than fields with borderline or mild defects. The most disparate visual fields are presented in detail in Figures 1-4. Based on the results of the TOP strategy, one of the four borderline fields falls into the mild group, two of the five mild fields into the borderline group, one of the five moderate fields into the mild group, and all the severely affected fields remain in the same group.

Comment

Reproducibility3,4 is an indicator of quality and is relevant to a new and time-saving perimetric strategy.1,2 It is noteworthy that, even during such a short examination time, a reproducible result can be obtained. Good conformity of results is also important, since this will facilitate comparison between strategies. Several authors (for example, Morales et al.1 and Dannheim and Zeyssig2) found a ‘smearing effect’ with the TOP strategy: deep defects turned up as being shallower and with less sharp borders. Such an effect acts like a filter.

A point-wise between-strategy correlation was performed on the 18 clearly defec-

Fig. 11. Mean MD of double examination with normal sensitivity (upper half) and TOP strategy (mirrored lower half) for 27 eyes that have been sorted as shown in Figure 5.

tive fields. We expected the regression line of this correlation to be a good indicator of conformity for high, intermediate, and low sensitivities. We used the ‘independent regression’, the mean between the regression with either method, since both threshold strategies have their own measurement error, which we found to be similar. This regression line would ideally have a slope of 1, an intercept of 0. However, the calculated regression line for each pair of defective visual fields reveals a slope of around 0.85 and an intercept of 0.35-2.93. These values correspond to a slightly narrower span of relative sensitivity values with flattening and a slight upward shift of the defect curve and mildly reduced LV with TOP. This attenuation, which explains the smearing effect,1,2 must be due to the special TOP threshold algorithm which continuously takes information from adjacent locations into account.1 The same discrepancy was found when comparing TOP with SITA Fast strategy.5 The smearing effect may be reduced with the new ‘TOP+’ strategy.6 In follow-up examinations, therefore, strategies should not be changed.

In our opinion, the TOP strategy is sufficiently reliable for the monitoring of glaucomatous eyes. The reproducibility of TOP was, on average, similar to the normal strategy. Even the most pronounced differences reported here were well within the long-term fluctuation.4 The much shorter examination time with TOP is highly valued by all patients and helps to avoid fatigue. After more than four years of experience, we use TOP as our routine perimetric technique.

Acknowledgments

I gratefully acknowledge the helpful advice of Professor Hans Bebié regarding the statistical analysis.

References

1.Morales J, Weitzman ML, Gonzalez de la Rosa M: Comparison between tendency-oriented perimetry (TOP) and Octopus threshold perimetry. Ophthalmology 107:134-142, 2000

2.Dannheim F, Zeyssig S: Octopus 1-2-3 comparison of TOP and normal threshold strategy: examples. Proceedings of the Sixth International Octopus Users’ Visual Field Symposium, Lugano 1998, pp 13-19, Schlieren: Interzeag 1999

3.Bebié H, Fankhauser F, Spahr J: Static perimetry: accuracy and fluctuations. Acta Ophthalmol 54:339346, 1976

4.Flammer J: Fluctuations in the visual field. In: Drance SM, Anderson D (eds) Automatic Perimetry in Glaucoma, A Practical Guide, pp 161-173. Orlando, FL: Grune & Stratton 1985

5.King AJW, Taguri A, Wadood AC, Azuara-Blanco A: Comparison of two fast strategies: SITA Fast and TOP for the assessment of visual fields in glaucoma patients. Graefe’s Arch Clin Exp Ophthalmol 240:481-487, 2002

6.Mesa F, Aguilar J, González-Hernandez M, González de la Rosa M: Second generation of the tendency oriented perimetry algorithm in glaucomatous patients. This Volume, pp 221-226