SECOND GENERATION OF THE TENDENCY ORIENTED PERIMETRY ALGORITHM IN GLAUCOMA PATIENTS

FÁTIMA MESA, JOSÉ AGUILAR, MARTA GONZÁLEZ-HERNÁNDEZ and MANUEL GONZÁLEZ DE LA ROSA

Hospital Universitario de Canarias, Universidad de La Laguna, Spain

Abstract

Purpose: To compare the results of the second generation of the tendency oriented perimetry algorithm that uses equations for the dependence of close and distant points, glaucoma specific (TOP+GL), with conventional bracketing perimetry (4-2 staircase strategy) in glaucoma patients. Methods: Forty-nine glaucoma patients, at different stages of the disease, were examined with TOP+GL and a bracketing techniques using the Octopus 1-2-3 perimeter. One eye per subject was examined, previous perimetric experience was required, and the order of the examinations was randomized. Results: High correlations between the threshold values and the perimetric indexes (MD and LV) were obtained with both strategies. The square root of LV (sLV) was equivalent for both strategies (average of 5.73 dB (SD = 2.6) for TOP+ and 5.72 dB (SD = 2.51) for bracketing (p = 0.49)). In cases with nasal steps, TOP+GL precisely delimited the border of the defect, without invading the opposite quadrant. Conclusions: TOP+GL produces results that are equivalent to TOP, with better delimitation of the borders of the nasal step. As opposed to TOP, TOP+GL tends to give LV results equivalent to those of a bracketing strategy.

Introduction

Since the introduction of tendency oriented perimetry (TOP) six years ago,1 papers have been published by our own group,2-5 as well as by others.6-13 All these paper described the usefulness of the strategy for study of the visual field. The strategy allows reliable examinations in children,14 a difficult group of patients to examine until this short strategy was introduced. TOP has also been used for the study of other visual functions, such as flicker, contrast,15-16 temporal functions,17 etc.

The smoothing of scotoma edges is a characteristic of the technique which we have described since the introduction of the algorithm.2 This smoothing leads to the loss variance (LV) being 15-20% lower than that produced by conventional bracketing perimetry (4-2 staircase strategy).4,7-8 However, LV values are well correlated, with correlation coefficients (r) between 0.85 and 0.93.2-5

Two years ago, we developed a theoretical procedure for reducing these differ-

Address for correspondence: Manuel González de la Rosa, C/. 25 de Julio, 34, 38004. Santa Cruz de Tenerife, Spain. Email: mgdelarosa@jet.es

Perimetry Update 2002/2003, pp. 221–226

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

222 F. Mesa et al.

ences.18 We attempted to produce a modified TOP strategy in order to obtain results equivalent to the ones given by the conventional bracketing strategy. The linear interpretation applied to the initial program was substituted by an interpretation based on the mathematical relationship between thresholds of different regions of the glaucomatous visual field, obtained with the bracketing strategy. These point-by-point relationships were published recently.19

This paper attempted to evaluate the ability of this modified TOP strategy to simulate the results of the bracketing strategy.

Material and methods

Forty-nine glaucoma patients (mean age, 57.0 ± 14.1 years; 23 males and 26 females) at different stages of the disease were examined with the TOP+GL and standard bracketing techniques, using the Octopus 1-2-3 perimeter and grid 32. While TOP calculates the threshold values from the patient’s responses to the examined and surrounding points, TOP+GL also takes into account responses from other locations with a proven relationship to the test location. In both TOP and TOP+GL, only one stimulus is shown for each point examined.

One eye per subject was examined and the sequence of the examinations was random. Inclusion criteria were prior perimetric experience, minimal visual acuity of 0.5 (20/40), and no other associated pathology that could affect the visual field. Patients were all corrected for distance vision, as required for this perimeter.

The comparison of thresholds and statistical indices obtained using both procedures was carried out using lineal regression, calculating the correlation coefficient and the error of Y in relation to X; that is, the foreseeable error when calculating the index of a strategy from that of the other, using the regression equation. Analysis was carried out globally, and by quadrants.

Results

Correlation coefficient (and error of estimation of Y in relation to X) between both examinations was MD = 0.97 (1.88 dB), sLV = 0.88 (1.21 dB), MD (supero nasal) = 0.97 (2.42 dB), MD (infero nasal) = 0.95 (2.60 dB), MD (supero temporal) = 0.93 (2.89 dB), MD (infero temporal) = 0.96 (1.98 dB), local thresholds = 0.84 (5.51 dB).

Mean MD value was 11.23 (SD = 7.48 dB) for TOP+GL and 11.59 (SD = 7.48 dB) for bracketing (p = 0.41). The mean value of the square root of the loss variance (sLV) was 5.73 (SD = 2.6 dB) for TOP+GL and 5.72 (SD = 2.51 dB) for bracketing (p = 0.49) (Fig. 1).

The local thresholds for each case gave an RMS error for TOP+GL in relation to bracketing that increased from 3 dB for an MD = 0 dB to 8 dB for an MD = 15 dB, and it decreased again for very deep cases (Fig. 2).

The relationship between sLV and MD was similar in both cases. For MD values lower than 15 dB, the correlation coefficient (r) between both indices was 0.88 for bracketing and 0.91 for TOP+GL (Figs. 3 and 4).

The distribution of the frequencies of the local deviations for the 3478 points

Fig. 3. Relationship between MD and sLV using the bracketing strategy.

Fig 4. Relationship between MD and sLV using the TOP+GL strategy.

examined had a similar disposition in both cases, with a slight predominance of the intermediate deviations in the case of TOP+GL (Fig. 5).

In cases with nasal steps, TOP+GL delimited the border of the defect precisely, without invading the opposite quadrant (Fig. 6).

Discussion

This new version of the TOP algorithm gives results that are similar to the ones given by the previous version, with regard to MD values in each quadrant. The local devia-

Fig. 5. Frequencies of the local deviations for the 3478 points examined.

Fig. 6. Example of a visual field using bracketing and TOP+GL.

tions have a distribution that is similar to the standard bracketing strategy. LV values are also similar to the ones given by the bracketing strategy.

Our results should be interpreted carefully. In a previous paper, we observed that the relationship between LV and MD was superior in TOP than in bracketing.20 In the case of the present paper, the relationship was better than in the previous one, and even slightly better than the one initially described by Pearson et al.,21 but there is still a better relationship in the case of TOP+GL.

Finally, in another study presented at this meeting, we found diagnostic advantages with the LV values given by TOP, compared to bracketing.

The LV results in TOP+GL are more equivalent than the ones obtained with the first TOP version to those obtained with the bracketing strategy. However, we still do not know if the diagnostic ability of LV in TOP+GL is as good as that observed for TOP.

References

1.González de la Rosa M, Bron A, Morales J, Sponsel WE: TOP perimetry: a theoretical evaluation. Vision Res 36:88, 1996

2.González de la Rosa M, Martinez A, Sanchez M, Mesa C, Cordovés L, Losada MJ: Accuracy of tendency oriented perimetry (TOP) in the Octopus 1-2-3 Perimeter. In: Wall M, Wild J (eds) Perimetry Update 1996/1997, pp 119-123. The Hague: Kugler Publ 1997

3.González de la Rosa M, Martínez Piñero A, González-Hernández M: Reproducibility of the TOP algorithm results versus the ones obtained with the bracketing procedure. In: Wall M, Wild J (eds) Perimetry Update 1998/1999, pp 51-58. The Hague: Kugler Publ 1999

4.González de la Rosa M, Losada MJ, Serrano M, Morales J: G1-tendency oriented perimetry (TOP): introduction and comparison with G1-standard bracketing. In: Wall M, Wild J (eds) Perimetry Update 1998/1999, pp 43-49. The Hague: Kugler Publ 1999

5.Morales J, Weitzman M, Gonzalez de la Rosa, M: A preliminary comparison between tendency oriented perimetry (TOP) and traditional threshold perimetry. Ophthalmology 107:134-142, 2000

6.Takada S, Matsumoto C, Okuyama S, Iwagaki A, Otori T: Comparative evaluation of four strategies (normal, 2level, dynamic, TOP) using the automated perimeter Octopus1-2-3. In: Wall M, Wild J (eds) Perimetry Update 1998/1999, pp 35-41. The Hague: Kugler Publ 1999

7.Lachkar Y, Barrault O, Lefrancois A, Demailly P: Rapid tendency oriented perimetry (TOP) with the Octopus visual field analyzer. J Fr Ophtalmol 21:180-184, 1998

8.Horikoshi N, Osako M, Goto H, Tamura Y, Okano T: Clinical evaluation of tendency oriented perimetry in Octopus perimeter. Jap J Clin Ophthalmol 53:889-893, 1999

9.Maeda H, Nakamura M: A new rapid perimetry test with dynamic strategy and a tendency oriented program (TOP) to detect glaucoma. Fol Ophthalmol Jpn 50:715-720, 1999

10.Fabre K, Michiels I, Zeyen T: The sensitivity and specificity of TOP, FDP and GDX in screening for early glaucoma. Bull Soc Belge Ophtalmol 275:17-23, 2000

11.Maeda H, Nakaura M, Negi A: New perimetric threshold test algorithm with dynamic strategy and tendency oriented perimetry (TOP) in glaucomatous eyes. Eye 5:747-751, 2000

12.Ruiz Lapuente C, Ruiz Lapuente A, Link B: Influence of topical brimonidine on visual field in glaucoma. Eur J Ophthalmol 11(Suppl 2):S67-71, 2002

13.Wadood AC, Azuara-Blanco A, Aspinall P, Taguri A, King AJ: Sensitivity and specificity of fre- quency-doubling technology, tendency-oriented perimetry, and Humphrey Swedish interactive threshold algorithm-fast perimetry in a glaucoma practice. Am J Ophthalmol 133:327-332, 2002

14.Morales J, Brown SM: The feasibility of short automated static perimetry in children. Ophthalmology 108:157-162, 2001

15.González de la Rosa M, Rodríguez J, Rodríguez M: Flicker-TOP perimetry in normals, patients with ocular hypertension and early glaucoma. In: Wall M, Wild J (eds) Perimetry Update 1998/1999, pp 59-66. The Hague: Kugler Publ 1999

16.Rodríguez J, Cordobés L, Abreu A, González de la Rosa M: TOP-flicker fluctuation in ocular hypertension. In: Wall M, Mills RP (eds) Perimetry Update 2000/2001, pp 149-153. The Hague: Kugler Publ 2001

17.González-Hernández M, Pareja Ríos A, Rodríguez M, González de la Rosa M: Combined spatial resolution and contrast perimetry in normal subjects. In: Wall M, Mills RP (eds) Perimetry Update 2000/2001, pp 109-114. The Hague: Kugler Publ 2001

18.González de la Rosa M, Mesa F, Arteaga V, González-Hernández M: Second generation of the Tendency Oriented Perimetry algorithm: TOP+. In: Wall M, Mills RP (eds) Perimetry Update 2000/ 2001, pp 155-159. The Hague: Kugler Publ 2001

19.González de la Rosa M, González Hernández M, Abraldes M, Azuara-Blanco A: Quantification of inter-point topographic correlations of threshold values in glaucomatous visual fields. J Glaucoma 11:30-34, 2002

20.González Hernandez M, Martinez Piñero A, Fariña FJ, González de la Rosa M: Relation between the mean defect and the loss variance in TOP and bracketing perimetric strategies. (ARVO Abstract No. 361). Invest Ophthalmol Vis Sci 40:S68,1999

21.Pearson PA, Baldwing LB, Smith TJ: The relationship of mean defect to corrected loss variance in glaucoma and ocular hypertension. Ophthalmologica 200:16-21,1990