Combined corneal ulcer debridement, conjunctival flap grafting, and amniotic membrane transplantation for perforated or impending perforated necrotizing stromal herpes simplex keratitis

To evaluate short-term anatomical, inflammatory, and visual outcomes of combined corneal ulcer debridement, conjunctival flap grafting, and amniotic membrane transplantation (AMT) in patients with severe necrotizing stromal herpes simplex keratitis (HSVNSK) with corneal perforation or impending perforation. This retrospective observational case series included 14 eyes of 14 patients with clinically diagnosed severe HSVNSK treated at Qiandongnan Miao and Dong Autonomous Prefecture People’s Hospital between January 2020 and December 2025. All eyes had a minimum stromal thickness of 210 μm or less at the ulcer bed and either established perforation or impending perforation. In addition to standardized topical and systemic anti-infective and anti-inflammatory therapy, all eyes underwent corneal ulcer debridement, conjunctival flap grafting, and AMT. Follow-up was scheduled on postoperative day 1, week 1, week 2, month 1, month 3, month 6, and month 12. Main outcomes included best-corrected visual acuity (BCVA), corneal opacity score, corneal neovascularization grade, conjunctival and limbal injection scores, anterior chamber depth, thickness of the composite repair layer at the ulcer bed, time to epithelialization, resolution of hypopyon, and postoperative complications. Paired t-tests and Wilcoxon signed-rank tests were used to compare all parameters between baseline and 1 month postoperatively. Follow-up at 3, 6, and 12 months relied on qualitative telephone interviews. The cohort comprised 6 men and 8 women with a mean age of 55.3 ± 7.6 years. Six eyes (42.9%) had overt corneal perforation and eight eyes had hypopyon before surgery. Time to full epithelialization averaged 5.6 ± 1.0 days, and all six eyes with positive preoperative Seidel testing exhibited no detectable aqueous leakage after postoperative week 1. In hypopyon-positive eyes, complete hypopyon resolution occurred at a median of 7.0 days (range, 0–14 days); stromal edema resolved within 9.6 ± 3.2 days. All conjunctival flaps remained viable without necrosis, retraction, or displacement. At postoperative month 1, BCVA (Logarithm of the Minimum Angle of Resolution, LogMAR) improved from 2.07 ± 0.66 to 1.52 ± 0.69 ( P < 0.001); median corneal opacity score declined from 4 (3, 4) to 2 (2, 3) ( P = 0.003). Median conjunctival and limbal injection grades were markedly reduced from 2 (1, 3) to 0 (0, 1) (both P = 0.001), and anterior chamber depth increased from 1.84 ± 0.92 mm to 2.49 ± 0.24 mm ( P = 0.048). The thickness of the composite repair layer at the ulcer bed increased from 99.53 ± 63.38 μm to 516.49 ± 84.14 μm ( P < 0.001). Corneal neovascularization grade increased from 2 (2, 3) to 3 (3, 3) ( P = 0.008). No secondary glaucoma, endophthalmitis, persistent epithelial defects, flap necrosis, or HSV recurrence were documented across variable follow-up durations. Note that these observed clinical changes represent associations derived from combined surgical and pharmacologic management and cannot be solely attributed to the surgical intervention alone. Combined corneal ulcer debridement, conjunctival flap grafting, and AMT were associated with short-term preservation of corneal structure and suppression of intraocular inflammation in patients with severe HSVNSK complicated by corneal perforation or impending perforation. This multimodal globe-salvaging tectonic procedure yielded limited visual rehabilitation despite statistically improved short-term best-corrected visual acuity and is not intended for optical reconstruction. This retrospective observational case series was not registered as a clinical trial. Not applicable.