PURPOSE: To examine the immunohistochemical alterations in the corneal stroma in Peters anomaly (PA) and congenital glaucoma (CG) compared with age-matched normals and acquired adult corneal scarring (AACS). METHODS: The clinical features of PA and CG patients who underwent penetrating keratoplasty were recorded. Immunohistochemistry of cornea and control tissue (normal and acquired corneal scars) was performed with antibodies against collagen types I, III, keratan sulfate, lumican, decorin, and smooth muscle actin followed by semiquantitative analysis of immunolabeling. RESULTS: Clinical features in 2 groups were consistent with PA and CG. Microscopy showed thickened stromal collagen bundles in PA (n = 15), CG (n = 11), and AACS (n = 20) compared with normals (n = 18). PA and CG had distinct immunophenotypes compared with controls. Type I collagen labeling was more intense in CG compared with PA (intensity grading (IG) 2.73 vs. 2.07; P < 0.001). Decorin, lumican, and keratan sulfate labeling was significantly less intense in PA versus AACS (IG; 1.91, 0.38, 1.75 in PA and 2.7, 1.11, 2.61 in AACS. respectively; P = 0.002, P = 0.001 and P = 0.004) and normals (IG 1.92, 1.06, 2.59 respectively; P < 0.001, P < 0.001 and P = 0.005). Collagen I labeling was less intense in CG versus AACS (IG 2.73 vs. 3.09) (P = 0.007). Collagen III labeling was more intense in PA/CG than in normals (IG 0.9, 0.64, 0.62 retrospectively) (P < 0.001 in both). CONCLUSIONS: The immunophenotype of the corneal scar in PA and CG differs from normal corneas and so does PA from AACS. The similarities between CG and AACS suggest that CG scarring has an acquired component.
PURPOSE: To examine the immunohistochemical alterations in the corneal stroma in Peters anomaly (PA) and congenital glaucoma (CG) compared with age-matched normals and acquired adult corneal scarring (AACS). METHODS: The clinical features of PA and CG patients who underwent penetrating keratoplasty were recorded. Immunohistochemistry of cornea and control tissue (normal and acquired corneal scars) was performed with antibodies against collagen types I, III, keratan sulfate, lumican, decorin, and smooth muscle actin followed by semiquantitative analysis of immunolabeling. RESULTS: Clinical features in 2 groups were consistent with PA and CG. Microscopy showed thickened stromal collagen bundles in PA (n = 15), CG (n = 11), and AACS (n = 20) compared with normals (n = 18). PA and CG had distinct immunophenotypes compared with controls. Type I collagen labeling was more intense in CG compared with PA (intensity grading (IG) 2.73 vs. 2.07; P < 0.001). Decorin, lumican, and keratan sulfate labeling was significantly less intense in PA versus AACS (IG; 1.91, 0.38, 1.75 in PA and 2.7, 1.11, 2.61 in AACS. respectively; P = 0.002, P = 0.001 and P = 0.004) and normals (IG 1.92, 1.06, 2.59 respectively; P < 0.001, P < 0.001 and P = 0.005). Collagen I labeling was less intense in CG versus AACS (IG 2.73 vs. 3.09) (P = 0.007). Collagen III labeling was more intense in PA/CG than in normals (IG 0.9, 0.64, 0.62 retrospectively) (P < 0.001 in both). CONCLUSIONS: The immunophenotype of the corneal scar in PA and CG differs from normal corneas and so does PA from AACS. The similarities between CG and AACS suggest that CG scarring has an acquired component.