To determine whether optical coherence tomography (OCT) epithelial mapping can improve the detection of form fruste keratoconus. French National Eye Hospital, Paris 6 Pierre & Marie Curie University, Paris, France. Retrospective comparative study. Eyes with normal corneas, form fruste keratoconus, moderate keratoconus, or severe keratoconus were assessed using Fourier-domain OCT (RTVue 5.5), scanning-slit corneal topography (Orbscan IIz), and rotating Scheimpflug camera (Pentacam Comprehensive Eye Scanner). Several parameters provided by the software or derived from elevation maps, OCT pachymetric maps, and OCT epithelium parameters were evaluated and compared between the 4 groups. The study involved 145 eyes. There were no significant differences in the keratometry (K) value, inferior-superior value, keratoconus index, central K index, and topographic keratoconus classification indices between the form fruste keratoconus group and the control group (P > .05). Form fruste keratoconic corneas had less epithelial thickness in the thinnest corneal zone than normal corneas, and greater epithelial thickness in the thinnest corneal zone than keratoconic corneas (P < .005). The epithelial thickness in the thinnest corneal zone in form fruste corneas was located inferiorly (P < .005) and corresponded with the zone of minimum epithelial thickness and maximum posterior elevation (P < .005). The receiver operating characteristic curve analysis showed good overall predictive accuracy of the epithelial thickness in the thinnest corneal zone, with a 52 μm threshold value for discriminating form fruste keratoconic corneas from normal corneas. The epithelial thickness in the thinnest corneal zone and its location provided by the OCT epithelial mapping might be useful for the early diagnosis of form fruste keratoconus. No author has a financial or proprietary interest in any material or method mentioned.

Although the existence of the limbal stem cell (LSC) niche is accepted, precise knowledge of its three-dimensional (3D) architecture remains incomplete. The LSC niche was explored on freshly excised and organ-cultured corneoscleral rims from human donors (n = 47), pigs (n = 15) and mice (n = 27) with full-field optical coherence microscopy (FFOCM). Limbal crypt features were detected in 90% of organ-cultured human corneoscleral rims, extending between the palisades of Vogt as radially oriented rectangular (74% of eyes) and/or rounded (23% of eyes) forms, often branching off to, or becoming interconnected by, sub-scleral radially or circumferentially oriented crypts (in 56% of eyes). Mean crypt volume represented 16% of sampled limbal volume on the vertical axis and 8% on the horizontal axis. In pigs, palisades were finer and crypts wider with relatively uniform distribution around the eye, and radial orientation, connecting to numerous narrow criss-crossing invaginations beneath the scleral surface. In mice, only a circumferential limbal trough was detected. Mean crypt volume represented 13% of sampled limbal volume in humans and 9% in pigs. FFOCM combined with fluorescence, and confocal fluorescence microscopy, showed presence of p63-α+ cells and cytokeratin-3+ cells in the limbal crypts. To assess colony forming efficiency (CFE), limbal epithelial cells were cultured at low density with mitomycin-arrested 3T3 feeders. CFE increased with limbal crypt volume and was not significantly decreased in organ-cultured cornea, despite degradation of the epithelial roof, suggesting that stem cells remain protected at the base of crypts during organ culture. CFE in human samples was significantly greater than in pig, and CFE in mouse was zero. Crypt architecture in the three species appears associated with eye exposure to light. LSC density increased with percentage limbal volume occupied by crypts.

Purpose: To compare the efficacy and early morphological changes in the cornea following conventional (C-CXL), transepithelial by iontophoresis (I-CXL), and accelerated collagen cross-linking (A-CXL) in keratoconus. Methods: 45 eyes of 45 patients with progressive keratoconus who underwent corneal collagen crosslinking (CXL) were divided into 3 groups: C-CXL (n=15), A-CXL (n=15) and I-CXL (n=15). Patients were examined before surgery and at 1-month, 3-month, and 6-month intervals following surgery. Density of corneal sub-basal nerves, anterior and posterior keratocytes, corneal endothelium, demarcation line depth and maximal simulated keratometry values (Kmax) were all assessed. Results: Compared with preoperative values, the mean corneal sub-basal nerve and anterior stromal keratocyte densities were significantly lower at 6 months in the C-CXL and the A-CXL groups (p0.05). Kmax, corneal central thickness, and BSCVA remained stable during the whole study period. Conclusion: Iontophoresis was associated with weaker damage of corneal sub-basal nerves and anterior keratocytes compared with conventional procedures, but the demarcation line was present in less than 50% of cases and was more superficial than with the traditional procedure.

To compare the efficacy, safety, and microstructural corneal changes during 2 years after conventional corneal collagen cross-linking (C-CXL) and transepithelial corneal CXL by iontophoresis (I-CXL) for keratoconus. Eighty eyes of 80 patients with progressive keratoconus were treated by C-CXL (n = 40) or I-CXL (n = 40). Patients were investigated before surgery and 1, 3, 6, 12, and 24 months after treatment. We measured central corneal thickness and maximal simulated keratometry values (Kmax) and performed specular microscopy and in vivo confocal microscopy at each time point. The demarcation line was assessed 1 month after treatment. Kmax remained stable after I-CXL during the entire study period (P = 0.56), whereas the average keratometry increased by 0.2 diopter (50.9 ± 5.6-51.1 ± 5.2). Kmax significantly decreased 1 (P = 0.02) to 2 years (P < 0.01) after C-CXL, with an average decrease of 1.1 diopters (49.9 ± 4.5-48.8 ± 4.2). The failure rate of I-CXL was 20% and that of C-CXL 7.5%. The demarcation line was superficially visible in 35% of cases after I-CXL compared with 95% of cases after C-CXL. Endothelial cell density and central corneal thickness remained stable during the entire study period. The change in Kmax 2 years after C-CXL and I-CXL and the preoperative Kmax were negatively correlated (r = 0.14, P = 0.013, and r = 0.17, P = 0.007, respectively). I-CXL halted progression of keratoconus less efficiently than did C-CXL after 2 years of follow-up. Longer prospective studies are still needed to ensure I-CXL efficacy.