Ocular surface changes in type II diabetic patients with proliferative diabetic retinopathy

Abstract

Aim: To detect and analyze the changes on ocular surface and tear function in type II diabetic patients with proliferative diabetic retinopathy (PDR), an advanced stage of diabetic retinopathy (DR), using conventional ophthalmic tests and the high-resolution laser scanning confocal microscopy.

Methods: Fifty-eight patients with type II diabetes were selected. Based on the diagnostic criteria and stage classification of DR, the patients were divided into the non-DR (NDR) group and the PDR group. Thirty-six patients with cataract but no other ocular and systemic disease were included as non-diabetic controls. All the patients were subjected to the conventional clinical tests of corneal sensitivity, Schirmer I Test, and corneal fluorescein staining. The non-invasive tear film break-up time (NIBUT) and tear interferometry were conducted by a Tearscope Plus. The morphology of corneal epithelia and nerve fibers was examined using the high-resolution confocal microscopy.

Results: The NDR group exhibited significantly declined corneal sensitivity and Schirmer I test value, as compared to the non-diabetic controls (P< 0.001). The PDR group showed significantly reduced corneal sensitivity, Schirmer I test value, and NIBUT in comparison to the non-diabetic controls (P < 0.001). Corneal fluorescein staining revealed the progressively injured corneal epithelia in the PDR patients. Moreover, significant decrease in the corneal epithelial density and morphological abnormalities in the corneal epithelia and nerve fibers were also observed in the PDR patients.

Conclusion: Ocular surface changes, including blunted corneal sensitivity, reduced tear secretion, tear film dysfunction, progressive loss of corneal epithelia and degeneration of nerve fibers, are common in type II diabetic patients, particularly in the diabetic patients with PDR. The corneal sensitivity, fluorescein staining scores, and the density of corneal epithelial cells and nerve fibers in the diabetic patients correlate with the duration of diabetes. Therefore, ocular surface of the patients with PDR should be examined regularly by conventional approaches and confocal microscopy to facilitate early diagnosis and treatment of keratopathy.

Keywords: confocal microscopy; corneal sensitivity; ocular surface; proliferative diabetic retinopathy; tear film break-up time; type II diabetes.

Figures

Figure 1. Corneal sensitivity was measured by a Cochet-Bonnet esthesiometer

Corneal sensitivity was significantly reduced in both NDR and PDR group in comparison to the non-diabetic controls; corneal sensitivity in the PDR group was also significantly lower than that in the NDR group. bP< 0.001.

Figure 2. Basic tear secretion was measured by Schirmer I Test

Basic tear secretion in the NDR and PDR group was both significantly lower than that in the non-diabetic controls. The tear section in the PDR group was also significantly diminished as compared to that in the NDR group. dP< 0.001.

Figure 3. NIBUT was measured by a Tearscope Plus

The NIBUT in the PDR group was significantly shorter than that in the NDR group and the control group. The NIBUT in the NDR group was not significantly different from that in the controls. fP< 0.001.

Figure 4. Corneal epithelial cells were examined by a high-resolution confocal microscope

A: Corneal epithelia from the controls were in a regular shape and with a discerned contour; B: The irregular and swelled corneal epithelia were observed sporadically in the NDR group; C: Corneal epithelial exfoliation and edema were seen more frequently, and the epithelial cell edges were irregular and blurred in the patients with PDR; D: Scale bar = 25 µm. Density of corneal epithelial cells was measured by a HRT-II confocal microscope. The epithelial cell density in the PDR group was significantly lower than the controls and the NDR group. hP< 0.001.

Figure 5. Corneal nerve fibers were examined by confocal microscopy

A: In the control group, nerve fibers were straight and thick. The axonal fibers branched at acute angles; B: In the NDR group, nerve fibers were more tortuous and branched less than those in the control group; C: The nerve fibers in the PDR patients were sparse, thin and discontinued. The fibers had even fewer branches and exhibited more tortuous axonal trajectories than the NDR group. Scale bar = 25 µm; D: The density of corneal nerve fibers was examined by a HRT-II confocal microscope. The nerve fiber density in the PDR patients was significantly decreased as compared to that in the NDR and the control group. jP< 0.001.