Citation
N. Morishige et al., Abnormal light scattering detected by confocal biomicroscopy at the corneal epithelial basement membrane of subjects with Type II diabetes, DIABETOLOG, 44(3), 2001, pp. 340-345
Categorie Soggetti
Endocrynology, Metabolism & Nutrition","Endocrinology, Nutrition & Metabolism
Abstract
Aims/hypothesis. Abnormalities of the basement membrane are thought to cont
ribute to the complications of diabetes. The suitability of the cornea for
detecting such abnormalities was assessed by determining its light-scatteri
ng index, a quantitative measure of tissue reflectivity in the basement mem
brane zone, with a confocal biomicroscope.
Methods. The light-scattering index was measured in 65 subjects with Type I
I (non-insulin-dependent) diabetes mellitus and 18 control subjects and was
evaluated for its possible relation to the stage of diabetic retinopathy.
Diabetic retinopathy was staged by ophthalmoscopic examination as non-diabe
tic (NDR), simple (SDR), preproliferative (PPDR), or proliferative (PDR).
Results. Examination of the cornea layer-by-layer with a confocal biomicros
cope did not show any marked differences in morphology between diabetic and
control subjects. The LSI (mean +/- SD) was 0.81 +/- 0.13, 0.87 +/- 0.09,
0.90 +/- 0.09, 0.90 +/- 0.13, and 1.02 +/- 0.25 in control subjects and in
diabetic subjects with NDR, SDR, PPDR, or PDR, respectively; the light-scat
tering index of diabetic subjects with PDR was significantly greater than t
hat of the control subjects (p = 0.001). An LSI greater than 1.0 was detect
ed in 5.6, 6.3, 15.0, 15.4, and 50.0 % of control subjects and of patients
with NDR, SDR, PPDR, or PDR, respectively; the percentage of subjects with
an LSI greater than 1.0 was significantly increased in diabetic patients wi
th PDR than for control subjects.
Conclusion/interpretation. These results suggest that the LSI increases wit
h the stage of diabetic retinopathy, and that measurement of corneal light
scattering could provide an index of basement membrane abnormality in peopl
e with diabetes.