Topical Steroids For Postoperative Haze Prevention After Photorefractive Keratectomy: A Systematic Review And Network Meta Analysis

Despite the well-established methodology of the surgery, haze after photorefractive keratectomy (PRK) remains a major concern for both patients and surgeons. While evidence has shown the value of topical steroids in the postoperative regimen, no consensus is established for a preferred regimen. Our goal was to compare the efficacy and safety of the different topical steroid protocols for prevention of postoperative haze to supply surgeons with an evidence-based selection for their toolbox.

This was a network meta-analysis of randomized control trials (RCTs) derived from a comprehensive search of PubMed, Scopus, and Embase performed through September 24^th, 2025 for RCTs evaluating topical steroidal treatment after PRK in a 3–12-month follow-up. 

Treatments were categorized by distinct active ingredients and concentrations, and pooled into groups by protocol length (1-2 months defined as short, 3-4 months as intermediate, and over that as long). Combination regomens were dealt with separately. After systematic seach and filtration - 14 studies reported outcomes for postoperative haze formation, 12 reported on deviation from target refraction, and 11 reported intraocular pressure (IOP) effects. Data extraction followed PRISMA guidelines, and analyses were conducted using a frequentist approach, random-effects, network meta-analysis model. Standardized mean differences (SMDs) were calculated for comparative analyses and P-scores (P) were used to rank treatment efficacy.

In total, 15 RCTs (1,477 patients; 8 countries) were included. Across the network, a 3-4-month fluorometholone (FML) 0.1% regimen was most studied and ranked best for haze prevention (P=0.837), significantly outperforming the same protocol when started only after bandage contact lens removal (SMD −1.02; 95% CI −2.01 to −0.02). For deviation from target refraction, no between-intervention differences reached significance, however the 3-4-month FML 0.1% regimen ranked second (P=0.683), preceded only by a 5-6-month FML 0.1% protocol (P=0.704). For IOP, FML 0.2% was the only steroid non-inferior to control and ranked best (P=0.975). Dexamethasone 0.1% ranked worst for IOP (P=0.165 for 3-4 months; P=0.256 for 1-2 months).

FML-based regimens administered for intermediate durations offer the most favorable balance of efficacy and safety for haze prevention after PRK and should not be delayed. Corticosteroids with greater intraocular penetration do not provide additional benefit and may increase safety concerns.