This retrospective single-center series evaluated a predefined, severity-based surgical algorithm for patients with refractory Mooren’s ulcer who failed comprehensive medical therapy. Between February 2016 and January 2024, consecutive patients requiring surgical reconstruction with amniotic membrane transplantation (AMT), lamellar keratoplasty (LK), or penetrating keratoplasty (PK) were reviewed. The principal aims were to assess primary healing, changes in best-corrected visual acuity (BCVA), recurrence, and anatomical success over long-term follow-up.
Patients were included if they had a clinical diagnosis of Mooren’s ulcer characterized by peripheral crescentic corneal ulceration with overhanging edges, had failed comprehensive conservative therapy (maximal topical corticosteroids, adjunctive conjunctival resection, and systemic immunosuppression as indicated), and met the study’s definition of refractory disease: progressive ulceration and stromal thinning despite at least 3 weeks of maximal tolerated medical therapy and conjunctival resection, or rapid melting with impending or actual perforation. Exclusion criteria comprised identified alternative causes of peripheral ulcerative keratitis (systemic autoimmune disease), ocular infection, scleritis, incomplete data, or follow-up under 12 months.
Diagnostic workup included systemic laboratory testing (CBC, ESR, CRP, rheumatoid factor, anti-CCP, ANA, ANCA, anti-SSA/SSB), syphilis serology, chest radiography, and corneal smear and culture to exclude infectious etiologies. For recurrences, systemic autoimmune testing was repeated to rule out newly manifest systemic disease.
The algorithm stratified operations by ulcer depth and perforation size.
AMT: Indicated for ulcers with stromal involvement <2/3 thickness without perforation. The technique used multilayered (6–10 layers) cryopreserved amniotic membrane placed basement-membrane side up, secured with interrupted 10–0 nylon sutures and a bandage contact lens.
LK: Indicated for ulcers ≥2/3 thickness, descemetocele, impending perforation, or perforation ≤4 mm. For perforations 2–4 mm, a double-layer LK technique was performed: a thin inner endothelial-side-down patch sealed the perforation, followed by an oversized outer lamellar graft sutured to the recipient bed. For minimal or no perforation (<2 mm), a conventional single-layer anterior LK was used.
PK: Reserved for perforations >4 mm or when LK was technically infeasible. Full-thickness recipient excision was followed by oversized full-thickness donor button placement and interrupted sutures.
All operations included debridement of necrotic tissue and excision of a 3-mm perilimbal conjunctival and Tenon’s margin adjacent to the ulcer. A single experienced surgeon performed all procedures.
A standardized postoperative regimen was applied to all patients. Topical antibiotics and steroid ointment were used initially. Topical 1% prednisolone acetate was given four times daily for the first month and tapered over 6–12 months; LK/PK recipients were transitioned to 0.1% fluorometholone once daily after 1 year for maintenance when appropriate. Systemic therapy combined oral prednisolone (1 mg/kg/day with tapering over 1–2 months) and cyclosporine A (5 mg/kg/day tapered over 3–6 months).
Recurrences were managed by intensifying medical therapy (hourly topical prednisolone acetate, conjunctival resection, and systemic immunosuppression). Persistent progression despite intensified therapy prompted reassessment with the same severity-based algorithm for possible reoperation.
Primary outcomes included primary healing (complete epithelialization with cessation of stromal thinning), recurrence (marginal ulceration, infiltration, or progressive thinning after initial healing), and anatomical success (globe preservation without evisceration/enucleation or phthisis). Secondary outcomes included BCVA and postoperative complications. Kaplan–Meier analysis estimated recurrence-free survival following LK. Descriptive statistics summarized baseline and outcome data.
After exclusions for insufficient follow-up, 24 eyes from 24 patients were analyzed. Median follow-up was 30 months (range 12–84 months). The cohort had a long median symptom duration preoperatively (7.5 months), extensive peripheral involvement in two-thirds (≥6 clock hours in 66.7%), and a high corneal perforation rate (45.8%). Preoperative vision was poor overall (median BCVA 0.10), with over half of eyes (58.3%) having BCVA ≤0.1.
Baseline severity correlated with surgical tier: AMT recipients had less severe, nonperforated ulcers and better preoperative vision, while LK and PK recipients had deeper, more extensive disease and higher rates of perforation.
Seven eyes underwent AMT. Primary healing was achieved in all AMT eyes. Final anatomical success occurred in 85.7% of AMT eyes. Two recurrences (28.6%) were reported; one was controlled medically and the other required subsequent LK. Functional vision (BCVA ≥ 0.3) was attained in 71.4% (5/7) of AMT eyes. Reported complications included one postoperative cataract in the AMT group.
Fifteen eyes underwent LK (including double-layer techniques for 2–4 mm perforations). Primary healing occurred in 93.3% of LK eyes. Although BCVA improved or stabilized in 86.7% of LK eyes, only 13.3% (2/15) achieved functional vision defined as BCVA ≥ 0.3. Recurrence after initial healing was higher in this group (64.3% when reported as 9/14 in the source), and Kaplan–Meier analysis estimated a median recurrence-free survival of 36 months for LK-treated eyes. Final anatomical success was 93.3% in the LK group.
Two eyes underwent PK for large perforations or when LK was not feasible. Primary healing was reported as 100% in the PK group; however, both PK eyes experienced recurrence (100%) and poor functional outcomes, with final vision ranging from hand motion to no light perception. Final anatomical success was 50.0% for PK-treated eyes within this cohort.
Across all surgical tiers primary healing was 95.8% (23/24). Recurrence rates varied substantially by procedure: 28.6% for AMT, 64.3% for LK, and 100% for PK. Final anatomical success (globe preservation) rates were 85.7% for AMT, 93.3% for LK, and 50.0% for PK, resulting in an overall globe preservation rate of 87.5% (21/24). Visual outcomes favored less-severe cases: AMT recipients more often attained functional vision, whereas keratoplasty-level disease demonstrated worse functional recovery.
The severity-graded surgical approach described preserved globe integrity in the majority of refractory Mooren’s ulcer cases managed in this series. The findings highlight that earlier, tissue-sparing procedures (such as AMT) can yield good structural and functional outcomes for less-severe ulcers, while eyes requiring LK or PK represent more aggressive disease biology with higher recurrence risk and poorer functional prognosis. The authors emphasize the need for vigilant long-term surveillance and sustained immunomodulation after keratoplasty-level interventions.
Strengths include a standardized surgical algorithm, detailed surgical techniques (including a double-layer LK option for 2–4 mm perforations), and long median follow-up. Limitations inherent to the report include its retrospective design, single-center experience, and small sample sizes within each surgical tier. The authors conclude that a severity-based surgical strategy is effective for anatomical preservation in refractory Mooren’s ulcer, but that keratoplasty-requiring cases demand aggressive and prolonged immunosuppression and close follow-up due to higher recurrence risk.