Early Post-Prk Dry Eye Is Predominantly Phenotype-Driven Rather Than Ablation-Driven Despite Surface Optimization: A Longitudinal Retrospective Analysis

To evaluate clinical and surgical determinants of early postoperative dry eye following photorefractive keratectomy (PRK), with particular emphasis on preoperative ocular surface phenotype and ablation-related parameters.

This single-center study was conducted at the Department of Ophthalmology, Ankara Gazi University Faculty of Medicine Hospital. All procedures were performed by two experienced surgeons using a standardized preoperative surface optimization and surgical protocol.

This retrospective longitudinal study included patients who underwent PRK and had documented tear film break-up time (TBUT), punctate epithelial staining score (Oxford 0–5), and dry eye symptoms in their electronic file. Patient data were analyzed preoperatively and at postoperative months 1, 3, and 6. Clinical dry eye was defined as the presence of at least two of the following criteria: TBUT <10 seconds, Oxford staining score >0, and subjective dry eye symptoms. Patients with preoperative dry eye received standardized 6-week lid hygiene, warm compresses and anti-inflammatory therapy. PRK was performed only after tear film stability and epithelial integrity were confirmed. 

Of 1300 screened patients, 764 (1528 eyes) were included. The overall clinical dry eye rate was 27.5% at month 1 and 23.3% at month 6. Preoperative dry eye requiring treatment was present in 20.9%. In this subgroup, postoperative dry eye occurred in 45.24% at month 1 and 32.81% at month 6. The highest risk was seen in eyes with combined low TBUT and punctate staining (n=73/764), with dry eye in 89.9% at month 1 (p<0.000001) and 55.4% at month 6 (p<0.000001). Among preoperatively normal patients, dry eye developed in 22.86% at month 1 and 20.9% at month 6. Early postoperative dry eye was associated with female sex, age >24 years, preoperative punctate staining, contact lens history, and winter surgery.

Despite successful preoperative surface normalization, patients with baseline tear film instability and epithelial abnormalities remained at significantly increased risk for early post-PRK dry eye. These findings suggest that early postoperative tear film instability is predominantly determined by preoperative ocular surface phenotype rather than by ablation magnitude, as ablation depth and mitomycin-C exposure were not identified as independent predictors.