Ocular surface inflammation due to allergy and blepharitis can lead to corneal complications and visual impairment. The aim of this study is to evaluate the efficacy of a cyclosporin 0.1% topical treatment achieving steroid-sparing. Eighty pediatric patients with moderate and severe vernal and blepharitis-related keratoconjunctivitis were included. Symptoms (photosensitivity, itching, discharge, tearing), signs (corneal fluorescein staining, papillary hypertrophy) and patients’ subjective assessment were evaluated during a 6-month follow-up. At the follow-up, all patients treated with topical cyclosporin showed a significant improvement in all subjective symptoms and objective signs (p < 0.001). The total number of courses of rescue steroids courses decreased from 3.71 ± 1.72 to 0.25 ± 0.49 at month 3 and to 0.13 ± 0.38 dropping bottle at month 6 (p < 0.001 at both time points). The 96.1% of the allergic cohort and 96.4% of the blepharitis cohort experienced a satisfactory good or rapid and good effect during the 6-month follow-up. The probability of needing rescue corticosteroids increased with an odds ratio of 1.98, (95% CI: 1.19–3.28, p = 0.008) for each unit increase in Oxford score when analysing the whole cohort. Topical cyclosporin seems to be very effective reducing the number of recurrences of corneal involvement and the need for steroid treatment.

Objectives: Our objective was to evaluate the diagnostic accuracy of clinical diagnoses in patients with chronic conjunctival lesions (CCLs) in relation to prevalence data and to identify clinical features that improve diagnostic performance. Methods: Retrospective data were collected for all patients with CCLs at a tertiary eye clinic between 2006 and 2024. A total of 1304 patients were reviewed; 391 had available histopathological data. Results: Misclassification was the highest for clinically benign, non-melanocytic, non-degenerative CCLs (referral rates: 0.446; positive predictive value [PPV]: 0.289). The referral rates for histology assessment were comparable for malignant melanocytic and non-melanocytic CCLs (0.923 and 0.989, respectively) and for benign melanocytic and non-melanocytic CCLs (0.319 and 0.446, respectively). In contrast, the diagnostic agreement was double for melanocytic malignant CCLs (0.800 vs. 0.391) and triple for melanocytic benign CCLs (0.934 vs. 0.289) compared with non-melanocytic CCLs. However, the limbal involvement increased the odds ratio (OR) for proper diagnostic performance to 7.0 (95% confidence interval (CI): 2.7–19.1), whereas the limbal overlapping increased the OR to 11.2 (95% CI: 4.0–35.9) among malignant CCLs. Conclusions: The clinical diagnosis of non-melanocytic CCLs is less frequently and less accurately confirmed histologically. In this CCL subgroup, where the lesion involves or overlaps the limbus, the likelihood of malignancy is increased sevenfold to elevenfold.