Abstract
Abstract: :
Purpose: To evaluate diffuse loss and test-retest variability in patients with advanced glaucoma using two different macular perimetric algorithms. Methods: We tested one eye each of 11 patients with stable, advanced glaucoma and a control group of 10 age-similar normal volunteers. All subjects were experienced and reliable visual field testers with good visual acuity, clear ocular media and no concomitant conditions affecting visual function. Differential light sensitivities were assessed using the Full Threshold (FT) and SITA Standard algorithms with the Humphrey Field Analyzer 10-2 pattern. All tests were repeated twice (t1, t2) within 5 +/- 6 days. To assess diffuse loss, we compared average differential light sensitivities at the 10 most sensitive points for patients and controls. To assess test-retest variability for each group, we used the standard deviation of (t1-t2) at all seeing points. To evaluate the effects of sensitivity on test-retest variability in the patient group, we performed linear regression on |t1-t2| vs. mean sensitivity. Finally, we compared mean test-retest variability for two subsets of points: ?normal sensitivity? (26 to 35 dB) and ?reduced sensitivity? (5?25 dB). Results: Mean diffuse loss for patients was -4.7 dB (FT) and -5.2 dB (SITA) (t greater than 3.8, p less than 0.001). The number of patients with significant diffuse loss was 7 for FT and 8 for SITA (chi squared greater than 8.4, p less than 0.004). For both FT and SITA, test-retest variability was higher for the patient group than the control group (F greater than 4.2, p less than 0.0005) and linear regression on the patient data showed increased variability with depth of defect (r greater than 0.23, p less than 0.00005). Direct comparison of ?normal sensitivity? and ?reduced sensitivity? points confirmed that variability was greater in abnormal areas (t greater than 5, p less than 0.0005). Conclusions: Results were quite similar for both FT and SITA algorithms: patients showed diffuse loss on the order of 5 dB and test-retest variability was greater in damaged vs. normal areas.
Keywords: perimetry • visual fields