Five-stage glaucoma damage classification using FDT indices.

Abstract

Frequency doubling technology perimetry (FDT) is a new unconventional method of visual field testing, which analyses selectively the My ganglionic cells (Johnson & Samuels 1997). The test is quick and easy for the patients and results are well comparable with Standard Automated Perimetry (SAP) test results (Trible et al. 2000). However, FDT does not include any methods to calculated or stage the severity of damage, which may be of crucial interest in chronic glaucoma, both for research purposes and for day-to-day clinical practice. The purpose of our investigation was to investigate the possibility of using FDT, DM and DSM indices together to stage the severity of functional damage in a population of patients with chronic glaucoma. Some 121 eyes of 77 patients affected by chronic open-angle glaucoma (mean age 61.3 ± 11.4 years; range 22–83 years) were assessed. All the patients had previously been tested with the Humphrey 30–2 (full threshold strategy) test. The stage of damage was classified on the glaucoma staging system (GSS; Brusini 1996). The distribution of the cases into GSS stages was as shown in Table 1. These patients were then tested using FDT perimetry (Welch-Allyn, Skaneateles Falls, NY & Zeiss-Humphrey Systems, Dublin, CA, USA; N-30 threshold test, which examines 17 areas in the 20° central visual field + 2 areas in the nasal periphery). A new 2-axis diagram was created, inserting the DM and DSM indices, analogous to MD and PSD SAP indices, on the x and y axes, respectively. (Warning: DM and DSM indices normal limits are not the same as the MD and PSD limits used in the first FDT version.) To obtain this diagram, a further population of 200 glaucomatous patients was analysed previously. The correlation between the two classification systems was studied using the gamma test, considering a two-way contingency table with ordinal data. The correspondence between the GSS stages and FDT stages is shown in Table 1. The correlation was statistically significant (gamma test value = 0.95; ASE = 0.016; P < 0.001). A standardized staging of glaucomatous functional damage severity would be very useful both in the research field and in day-to-day clinical practice; the same would be true for a reliable and simple classificatioin of visual field defects. A number of methods have been proposed in the past to obtain this goal, starting from SAP data. At present, several techniques of unconventional visual field examination have been developed for the detection of glaucomatous functional damage earlier than SAP. Among them, FDT is one of the most interesting and popular methods. Apart from a score, recently proposed for detecting visual filed damage defects using FDT in screening mode (Patel et al. 2000), no system exists for staging the defects in a standardized and reliable manner. In 1996, we introduced the glaucoma staging system (GSS), a new method for classifying automated visual field tests in glaucoma. The GSS uses the SAP MD and CPSD (or CLV) indices on a Cartesian coordinate diagram. The diagram is divided into 6 different stages of increasing severity by curvilinear lines: from stage 0 (completely normal visual fields) to stage 5 (very low threshold readings, with only small remnants of sensitivity). Moreover, every stage, apart from stage 0, is subdivided into 3 groups by two oblique straight lines: generalized visual field in the upper area, mixed defects in the central area and localized defects in the lower left area. This system proved to be very useful both to stage the damage severity and to separate the different components of visual field loss (Koçak et al. 1997; Koçak et al. 1998). We then tried the same approach with FDT. Our results seem to demonstrate that this is possible. The correspondence between this new classification and the results obtained with the GSS was very good. A slight overlap was found in the intermediate stages, that is to say between stages 2 and 3, where the mid defects are to be found. The subdivision of visual field defects into three types (generalized, localized and mixed) seems to be less precise than with the GSS. This may be due either to the different number of locations tested with these two methods of perimetry (76 points with SAP, 17 areas with FDT) or, more likely, to the fact that distinct components of the visual streams are analysed. However, also in the FDT diagram, the generalized defects tend to group in the upper area, while the localized defects are usually located in the lower left sector. The combined use of FDT, DM and DSM indices allows the user to stage accurately the severity of functional damage in chronic glaucoma and can help to separate localized and diffuse components of visual field loss.