Aim: To identify risk factors for the development of dupilumab-induced ocular surface disease (DIOSD) in adult patients with atopic dermatitis (AD) and describe outcomes of treatment.Methods: A retrospective institutional cohort study performed at the Rabin Medical Center, Petach Tikva, Israel. Adult patients with AD who received dupilumab from March 2018 to June 2019 were included. Demographics, AD severity scores, blood IgE levels, previous atopic keratoconjunctivitis (AKC), dermatological response to dupilumab, ophthalmological evaluation and treatment were noted. Univariate and multivariate analyses were used to identify risk factors for DIOSD.Results: Sixteen of 37 patients who were included in the study (43%) had new or exacerbated symptoms of ocular surface disease starting at 2 weeks following the first treatment. Three patients reported transient dry eye sensation which lasted 2 weeks; nine patients reported chronic dry eye sensation, and four patients (25%) had marked blepharoconjunctivitis. The presence of severe AD was the strongest predictor of DIOSD. Not a single patient with moderate AD had DIOSD. In multivariate analysis, prior AKC was a risk factor for DIOSD (R²=15.78, OR=23.28, p=0.005) while a family history of atopy was protective of DIOSD (R²=6.22, OR=0.13, p=0.05). All four patients with blepharoconjunctivitis had resolution of signs, and symptoms within days of starting periocular 0.03%–0.1% tacrolimus ointment.Conclusions: DIOSD is common in patients with AD receiving dupilumab. While most cases are mild, some patients can develop blepharoconjunctivitis which responds well to tacrolimus ointment. AD severity, and previous AKC are risk factors for DIOSD.

Purpose: To identify preoperative and intraoperative factors affecting breakthrough corneal haze incidence after photorefractive keratectomy (PRK) with mitomycin C (MMC). Methods: In this retrospective study of PRK performed at the Care Vision Refractive Laser Center, Tel Aviv, Israel, a total of 7535 eyes (n = 3854 patients; mean age ± SD, 26 ± 6 years; 55% men) underwent PRK with intraoperative MMC application. Patients with histories of corneal pathology or surgery were excluded. Incidence, time of onset, and corneal haze severity were documented on follow-up of 118 ± 110 days. Eyes were grouped by preoperative refraction [low (≤−3D), moderate (−3D to −6D), or high (>−6D) myopia; low (≤3D) or high (>3D) astigmatism; low or high hyperopia]; by intraoperative time (above or below 40 seconds); and by MMC application time (above or below 40 seconds). The main outcome measures were incidence, onset time, and severity of corneal haze. Results: The haze incidence was 2.1% in eyes with high myopia versus 1.1% in those with low to moderate myopia (P = 0.002), and 3.5 times higher in eyes with high than with low astigmatism (P < 0.05). The overall incidence was higher in eyes treated for hyperopia (10.8%) than for myopia (1.3%) (P = 0.0001). In eyes with moderate myopia, the haze incidence was lower in MMC application time ≥40 seconds (0%) than in <40 seconds (1.3%) (P = 0.03). After surgery, a mild early haze incidence peaked at 68.8 ± 6 days and severe late haze at 115 ± 17 days (P = 0.02). Conclusions: Hyperopic and large myopic or astigmatic corrections carry higher risk of haze. Longer MMC application might have beneficial haze prevention.