Abstract
Abstract: :
Purpose: To compare the pattern and extent of abnormalities detected by short-wavelength automated perimetry (SWAP) and the multifocal electroretinogram (mfERG) in eyes of diabetics that have no retinopathy (by stereo and color fundus photography) and in diabetics with early non-proliferative retinopathy. Methods: One eye from each of 30 diabetic subjects, 15 with early retinopathy and 15 without retinopathy, was tested on SWAP and mfERG on the same day. SWAP was tested using standard procedures (Humphrey 24-2, full threshold procedure), as was the mfERG (dilated pupil, 103 elements, 200 cd/m^2 white, 2 cd/m^2 dark; 10-100 Hz filtering). MfERG implicit times (IT), measured using Hood and Li’s (1997) method, were compared for each of the 103 stimulus locations to normative values obtained from 20 age-similar controls. SWAP data were analyzed using age-based control data and analysis methods provided by Johnson et al. (Discoveries in Sight, Portland, OR). Total threshold deviation results were used for mfERG comparisons. Abnormalities in both measures were defined as Z scores > 2. Results: Local functional abnormalities were detected in 93% of the subjects (13/15) with early retinopathy by either mfERG IT or SWAP or both. The sensitivity of the two measures was similar; 73% of the subjects (11 out of 15) had mfERG delays, and 60% (9 out of 15) had SWAP defects. Although there was overlap between the 2 measures (7/15 eyes had abnormalities on both), it was nearly as common (6/15) to see defects of only one type in a given eye. When abnormalities on both measures were present in an eye, they tended to be extensive, making quantification of spatial overlap difficult. Of those eyes without diabetic retinopathy, about half (7/15) showed either mfERG or SWAP abnormalities or both; this is less than the retinopathy group. Only 2 of 15 eyes were abnormal on both measures; in both of these eyes, the SWAP and mfERG abnormalities were seen in the same quadrants. MfERG and SWAP identified functional losses in similar numbers of subjects in the group without retinopathy (4 subjects for mfERG and 5 subjects for SWAP). Conclusions: Almost all diabetics with early retinopathy have either mfERG or SWAP abnormalities (or both). For context, patients with this level of diabetes generally have few, if any, conventionally measured ‘white on white’ visual field defects in clinical testing. The different defect locations shown by mfERG or SWAP suggest that these two sensitive functional measures may tap overlapping but distinctly different retinal anomalies associated with diabetes.
Keywords: diabetic retinopathy • electroretinography: clinical • visual fields