O R I G I N A L A R T I C L E Nonselective Less of Contrast Sensitivity in Visual System Testing in Early Type I Diabetes MAURO A.S. DI LEO, MD SALVATORE CAPUTO, MD BENEDETTO FALSINI, MD VlTTORlO PORCIATTI, PHD ANGELO MlNNELLA, MD ALDO V. GRECO, MD GIOVANNI GHIRLANDA, MD OBJECTIVE— Psychophysical methods in patients with diabetes mellitus reveal deficits of central or foveal vision. Our aim was to evaluate the contrast-sensitivity thresholds in 24 insulin-dependent (type I) diabetic patients with a short disease duration and without retinopathy, taking into account metabolic control. RESEARCH DESIGN AND METHODS— The control group consisted of age- matched nondiabetic subjects. None had visual or systemic symptoms. Contrast sensitivity measured at eight different spatial frequencies to sinusoidal bar patterns of 0.6-12.2 cycles/deg can detect functional defects in the spatially sensitive retinal ganglion cells or in higher visual pathways. We performed two different temporal types of contrast-sensitivity testing, dynamic (8 Hz) and static (0 Hz). RESULTS — Significant losses with dynamic contrast-sensitivity test at all but the highest spatial frequencies (i.e., 12.2 cycles/deg) were shown, whereas there was significant attenuation of contrast sensitivity at five spatial frequencies (1.0, 1.4, 2.2, 7.1, and 9.6 cycles/deg) in the static mode. Grating losses (<2SD of control means) of contrast sensitivity were found in 33.3% (dynamic) and in 72.9% (static) of eyes of diabetic patients. HbA lc values were positively correlated at variable spatial frequencies (1.0, 1.4, and 2.2 cycles/deg for dynamic test and 0.6, 1.0, 1.4, 2.2, 4.8, and 7.1 cycles/deg for static test). CONCLUSIONS — Our results suggest an early, generally nonselective neuronal damage of visual pathways that occurs before the onset of clinically detectable retinopathy. The visual deficit may be related directly to the effects of diabetes; repetitive minor hypoglycemic insults may contribute more than a marked hyper- glycemic condition to the mechanisms underlying physiological changes along the optic nerve. FROM THE DEPARTMENTS OF INTERNAL MEDICINE AND OPHTHALMOLOGY, CATHOLIC UNIVERSITY, ROME; AND THE INSTITUTE OF NEUROPHYSIOLOGY, NATIONAL COUNCIL OF RESEARCH, PISA, ITALY. ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO MAURO DI LEO, MD, ISTITUTO DI CLINICA MEDICA, UNIVERSITA CATTOLICA DEL SACRO CUORE, LARGO A. GEMELLI, 8, 00168 ROMA, ITALY. RECEIVED FOR PUBLICATION 25 MARCH 1991 AND ACCEPTED IN REVISED FORM 25 SEPTEMBER 1991. C ontrast sensitivity is a measure of the amount of contrast required to detect or recognize a visual target. The usefulness of contrast sensitivity measurements as an indication of early- stage disorders of diabetic retinopathy, has been shown in many studies (1-8). To clinically understand the physiologi- cal basis of these defects, techniques based on stimuli that use grating patterns consisting of alternating vertical light- dark bars presented on a television mon- itor yield the most information about vi- sual pathways. Gratings that can vary in spatial frequency (i.e., the number of light-dark cycles/degree of visual angle) are used to consider the quality of the optic system. Psychophysical (contrast sensitivity) and electrophysiological (vi- sual-evoked potentials and pattern elec- tro retinogram [P-ERG]) studies have benefited from the application of spatial- frequency methods (9). With contrast- sensitivity testing, the lowest detectable contrast across a range of spatial frequen- cies can be measured. In this way, a contrast-sensitivity function is derived for a subject. The spatial response of the visual system can be discriminated as a series of multiple channels, each channel sensitive to a narrow band of spatial frequency (10,11). Channels may be the expression of single-neuron function. Activities of ganglion cell subpopulations of the vi- sual pathway that respond well to grat- ings of different spatial frequency can be psychophysically isolated (12,13). The number of distinct neural channels in the primary visual pathway that respond to gratings justifies the use of spatial- frequency techniques in clinical psycho- physics to assess visual function deficien- cies. In our previous studies (14,15), the early effect of insulin-dependent (type I) diabetes on the central retina was shown with P-ERG gratings, but no in- formation about the overall function of the visual system in diabetic patients was given. In this study, our aim was to in- vestigate possible spatial frequency- 620 DIABETES CARE, VOLUME 15, NUMBER 5, MAY 1992