Macular Thickness Correlates With Visual Fields in Glaucoma
Seven hundred sixteen glaucoma patients had OCT testing between January 1, 2010, and June 30, 2010. Of the 229 that were performed with the Heidelberg SD-OCT device, 118 of the patients had bilateral scans which were free of vitreo-retinal pathology and segmentation artifact on the OCT. HVFs were available for 109 of the patients; 36 were excluded as unreliable or suspected of artifact on the HVF. One hundred forty-six eyes of 73 study subjects made up the final cohort.
Table 1 summarizes the demographics: 57.5% were women, and 70% were white. The most common glaucoma diagnoses were glaucoma suspect/ocular hypertension (54.8%) and primary open angle glaucoma (26%).
Correlation of Total Macular Thickness With Global Measures of Visual Field
The average total macular thickness (in the 8×8 mm grid) of all subjects was 285 μm (Table 2). Eyes with thinner average total macular thickness (<270 μm) had more negative MD scores (MD −6.76) and higher PSD scores (PSD 5.83), corresponding to significant visual field deficits. In eyes with average macular thickness >300 μm, the visual field parameters were virtually normal (MD 0.39 and PSD 1.71). In eyes with intermediate thickness (between 270 and 300 μm), the visual field parameters were also intermediate (MD −1.27 and PSD 2.30). Bivariate fit analysis (Fig. 2) shows a significant correlation between average macular thickness and both the MD score (Fig. 2A) and PSD score (Fig. 2B). Regarding the correlation between MD and macular thickness, the correlation coefficient is 0.431, R 0.185, linear fit (average macular thickness=287+1.55MD), with P = 0.0001. The correlation between PSD and macular thickness showed a correlation coefficient of −0.508, R 0.258, linear fit (average macular thickness=294–3.61PSD), with P < 0.0001. Results were similar with both the right and left eye (data from the right eye shown).
Macular thickness correlates with mean deviation (MD) and pattern standard deviation (PSD) scores. A, Bivariate fit analysis of average total retinal thickness of the macula compared with the MD score of the Humphrey visual field for each study subject, shown by graphical plot. B, Bivariate fit analysis as in (A), showing the comparison of the macular thickness to the PSD score of the Humphrey visual field.
Correlation of the Asymmetry Between the Right and Left Visual Fields With the Asymmetry in Total Macular Thickness Between the Eyes
Asymmetry is one of the hallmarks of early glaucoma, and is seen in more advanced disease as well. The difference in the average total macular thickness between the 2 eyes was calculated and compared with the difference of the global parameters of the HVF between the 2 eyes. The results show that patients with greater visual field defects in the right eye were found to have thinner macular thickness in that eye, and similarly, greater visual field defects in the left eye was associated with thinner macular thickness in the left eye (Table 3). An absolute difference in MD score of 3 correlated with a difference in average macular thickness of approximately 11 to 13 μm between the eyes.
Correlation of the Asymmetry in the Macular Thickness of the Inferior Macula and Superior Macula With PSD Score
Asymmetry within each eye was observed by noting the difference in the average thickness of the superior macula and inferior macula (Fig. 3). The difference in macular thickness between the superior macula and inferior macula was compared with the PSD score of the visual fields. Greater asymmetry, or difference in thickness between the superior and inferior maculae, correlated strongly with a larger PSD score, both in the event of worse superior thinning and worse inferior thinning. For eyes with superior thinning, the correlation coefficient is 0.575, R 0.331, linear fit (difference in macular thickness=2.03+1.65PSD), with P < 0.0001 (Fig. 3), and for eyes with worse inferior thinning, the correlation coefficient is −0.697, R 0.367, linear fit (difference in macular thickness = −1.54–1.20PSD), with P = 0.0022.
Asymmetry in macular thickness correlates with pattern standard deviation (PSD) scores. Bivariate fit analysis of the difference in average macular thickness (superior-half minus inferior-half) compared with PSD of the Humphrey visual field for each study subject, shown by graphical plot. This plot represents the data from subjects with superior macular thinning; results from subjects with inferior macular thinning are similar (data not shown).
Correlation of the Asymmetry in Macular Thickness With Asymmetry in the RNFL
Given the current understanding of the pathophysiology of glaucoma, correlation between RNFL defects and macular thickness is theorized; data from our study support this (Fig. 4, Table 4). Subjects with a supero-temporal defect in the RNFL were more likely to have superior thinning of the macula (on asymmetry analysis comparing retinal thickness of the superior macula and inferior macula); the correlation coefficient is 0.746, R 0.557, linear fit (difference in sup-temp RNFL = 4.33 + 2.20×difference in macular thickness) with P < 0.0001 (Fig. 4A, Table 4). Similarly, subjects with an infero-temporal RNFL defect were more likely to have inferior thinning of the macula on asymmetry analysis, with a correlation coefficient of 0.845, R 0.715, linear fit (difference in inf-temp RNFL = 4.21 + 2.36×difference in macular thickness), P < 0.0001 (Fig. 4B, Table 4).
Asymmetry of the retinal nerve fiber layer (RNFL) correlates with asymmetry of the macular thickness. A, Bivariate fit analysis of the difference in supero-temporal RNFL thickness between the right and left eyes compared with the difference in macular thickness of the superior-half between the 2 eyes, shown by graphical plot. B, Bivariate fit analysis as in (A), showing the comparison of the infero-temporal RNFL thickness with the thickness of the inferior-half of the macula.