International Ophthalmology 18:2 l 1-214, 1995. 211 9 1995 KluwerAcademicPublishers. Printedin the Netherlands. Corneal autofluorescence in diabetic and normal eyes Sergio Fantaguzzi, Franco Docchio 1,*, Lauretta Guarisco & Rosario Brancato Department of Ophthalmology, University of Milano, Scientific Institute H-S. Raffaele, Italy; 1 Dipartimento di Elettronica per l'Automazione, University of Brescia, Italy (* Requests for offprints) Accepted3 May 1994 Key words: corneal autofluorescence, diabetes, fluorophotometry, retinopathy grade Abstract Corneal autofluorescence has been lately studied as a predictor of retinopathy severity in diabetic patients. We measured corneal autofluorescence in 138 eyes of 69 diabetic patients and 64 eyes of 32 healthy controls. Diabetic patients were subdivided by the severity of retinopathy according to the Modified Airlie House Classification (stage 1: no or minimal retinopathy; stage 2: minimal background retinopathy; stage 3: background retinopathy; stage 4: (pre-) proliferative retinopathy. The fluorescence peak value and the area underlying the corneal autofluorescence curve were measured with a scanning fluorophotometer (Fluorotron Master, Coherent Radiation Palo Alto CA) Healthy controls' values of corneal autofluorescence (peak value: mean 11.03 + 3.77 ng. equivalent fluorescein/ml; area: mean 163.85 4- 61.40 scan-point x ng. equivalent fluorescein/ml) resulted similar (peak value: p = 0.83; area: p = 0.61) to those of diabetic patients without retinopathy (peak value: mean 11.2 -4- 3.4 ng.eq/ml; area: 170.07 + 57.23 scan-pnt.ng.eq/ml). A statistically significant difference was found between diabetic patients without retinopathy and patients with stage 2, 3, 4 retinopathy. No statistically significant difference was found both for the peak value (p = 0.50) and for the area (p = 0.61) between stage 3 and stage 4 retinopathy. The sensitivity and specificity of corneal autofluorescence as a screening test for diabetic retinopathy were 82% and 62% for the peak value, 87% and 60% for the area; the positive predictive value for the presence of diabetic retinopathy was 65% for the peak and 63% for the area. Introduction An increase of the autofluorescence of ocular tissues has been found in diabetes, aging and chronic exposure to light. Changes in corneal metabolism and thickness can occur as a result of diabetes mellitus [ 1-7]. The val- ues of corneal fluorescence, as measured with a scan- ning computerized fluorophotometer (Fluorotron Mas- ter) have been shown to originate mainly from corneal autofluorescence [8]. It has recently been demonstrat- ed that the filtered peak value of corneal autofluores- cence is correlated to the degree of diabetic retinopathy present [9]. This peak value has been taken into con- sideration as an indicator of the presence and severity of diabetic retinopathy. In this study corneal autofluo- rescence has been evaluated, using both the peak and the area values of the graph. This paper concerns the comparison between the two procedures. Materials and methods 138 eyes of 69 diabetic patients were examined, (of which 58 insulin dependent and 11 non insulin depen- dent) and 64 eyes of normal control subjects. The average age was 36.08 (sd. 14.3) years in diabetic subjects and 44.23 (sd. 15.66) years in healthy sub- jects. The average duration of diabetes was 15.31 (sd. 6.02) years). All the subjects underwent a biomicro- scopic examination of the anterior segment, in order to ensure the absence of corneal disease. No topical medication was given to any of the patients. Contact lens wearers were excluded from the study, as was any subject who had undergone photocoagulation in the three months preceding our evaluation. Diabetic retinopathy was classified in four degrees of severity, using the Modified Arlie House classification [10] on