Relationship between fixation stability measured with MP‐1 and reading performance

Abstract

People with visual impairment have reduced reading performance, which is thought to be related to unstable or eccentric fixation. New microperimeters such as the MP-1 offer straightforward analysis of fixation stability. The aim of this study was to investigate the relationship between fixation stability and reading speed in a large cohort of people with diverse causes of visual impairment and to verify the correlation between reading speed and different methods for the quantification of fixation. The better eye of one hundred and twenty subjects was assessed. Fixation values were obtained from the MP-1 microperimeter. Reading speed was evaluated using newspaper text with magnifiers if required. The poorest fixation stability and reading performance was found in people with age-related macular degeneration while the best fixation was in retinitis pigmentosa subjects. A linear relationship was found between reading speed and the proportion of fixations within 2° (r²=0.51, p<0.001) and 4° (r²=0.36, p<0.001). A negative correlation was found between reading speed and all three bivariate contour ellipse areas (BCEA; for log transformation of 1-S.D., 2-S.D. and 3-S.D.: r²=0.39, p<0.001). In a multiple regression model, proportion of points falling within 2° and 4° circle was significantly related to reading speed (r²=0.55, p<0.01; r²=0.43 p<0.01); also BCEAs values were strongly related to reading ability only in patients with central vision loss (r²=0.62, p<0.01 for LogBCEA 68.2%; r²=0.61, p<0.01 for LogBCEA 95.4% and 99.6%) and peripheral defect (r²=0.52, p<0.01 for LogBCEA 68.2%; r²=0.50, p<0.01 for LogBCEA 95.4%; r²=0.49, p<0.01 for LogBCEA 99.6%) but not in combined defect subjects. The study confirms that in people with visual impairment the reduced reading performance is correlated with fixation instability. Moreover, there is a strong relationship between reading speed and both the proportion of fixations falling within 2° and 4° and bivariate contour ellipse area values.