Comparison of Glaucomatous Visual Field Defects Using Matrix Perimetry and Swedish Interactive Thresholding Algorithm (SITA) Perimetry

Abstract

Abstract: : Purpose: To compare glaucomatous visual field defects found by Swedish Interactive Thresholding Algorithm (SITA) perimetry and Matrix perimetry (Carl Zeiss Meditec, Dublin, CA), a new iteration of frequency doubling technology (FDT) using a 24–2 test pattern. Methods: SITA and Matrix visual field (VF) testing was performed on 50 subjects from UPMC Eye Center. One eye of each subject was included, and the diagnosis of glaucoma was based on clinical findings and SITA visual field defects. Similar analyses were conducted on a group of 41 subjects who had glaucoma diagnosed by Matrix VF defects. Test duration, mean threshold, mean deviation (MD), pattern standard deviation (PSD), glaucoma hemifield test (GHT), and number of abnormal points on the pattern deviation (PD) plot were evaluated for each device. Results: The following results were obtained from the SITA based analyses. Test duration was significantly shorter for Matrix (p=0.002). The mean threshold value of SITA tests (25.91 ± 5.26 dB) was significantly higher than that of Matrix tests (21.10 ± 5.66 dB) (p<0.0001). SITA MD (–3.54 ± 5.10dB) was significantly greater than that found by Matrix (–5.10 ± 5.13dB) (p=0.002). There was no significant difference in PSD between the two devices; however, SITA detected significantly more abnormal points on the pattern deviation plot as compared to Matrix (p=0.005). In 18 eyes (36%) where SITA showed glaucomatous visual field defects, Matrix visual fields were normal. All findings were similar in the group with glaucoma defined by abnormal Matrix visual fields, except that the difference in the number of abnormal points was not significant between technologies. Conclusions: There was little difference in the detection of glaucomatous visual field defects comparing Matrix and SITA, regardless of whether the definition was based on Matrix or SITA VF abnormalities. Taking into account that the dB units reported by Matrix and SITA represent different physical properties the comparison of the threshold values should be assessed with caution. The number of abnormal points identified by the pattern deviation analysis was significantly higher in SITA. This might reflect a limitation of the normative dataset used for Matrix analysis. PSD may be useful for comparison between the perimetric modalities.

Keywords: visual fields