Predictive Factors For Penetrating Keratoplasty In Patients With Fungal Keratitis
Published 2026 - 44th Congress of the ESCRS
Reference: FP10.09 | Type: Free paper | DOI: 10.82333/a5f5-gg36
Authors: Dr Pedro Mota-Moreira* 1 , Dr Rita Teixeira-Martins 1 , Dr Pedro Marques-Couto 1 , Dr Ana Gama e Castro 1 , Dr Mariana Salsa-Castelo 1 , Dr Rodrigo Vilares-Morgado 2 , Dr Pedro Neves Cardoso 2 , A/Prof. João Pinheiro-Costa 3 , A/Prof. Tiago Monteiro 2 , Dr Luís Torrão 2 , Dr Ana Maria Cunha 2
1Ophthalmology,ULS São João,Porto,Portugal, 2Ophthalmology,ULS São João,Porto,Portugal;Surgery and Physiology,Faculty of Medicine, University of Porto,Porto,Portugal, 3Biomedicine,Faculty of Medicine, University of Porto,Porto,Portugal
Purpose
To identify demographic, clinical, microbiological, and treatment-related factors associated with the need for penetrating keratoplasty (PK) in patients with fungal keratitis.
Setting
Department of Ophthalmology, Unidade Local de Saúde de São João, a tertiary ophthalmic referral centre in Porto, Portugal.
Methods
A retrospective, single-centre, longitudinal study including analysis of 66 eyes from 63 patients (51.5% males) with culture-proven fungal keratitis was conducted between January 2009 until December of 2024. Demographic, clinical, microbiological, therapeutic and disease complications variables were compared between two groups: eyes requiring PK and eyes managed only conservatively (non-PK). Univariate analyses were performed using non-parametric and categorical tests. Variables with p<0.10 were included in a multivariate logistic regression model to identify independent predictors of PK.
Results
Mean age at diagnosis was 65.3±17.5 years, with a mean follow-up of 6.1±3.6 years (no differences between groups). PK was performed in 37 eyes (56.1%). No significant differences were found for initial BCVA, contact lens use, ocular comorbidities, trauma, hypopyon, endophthalmitis, or fungal class. Mean time from swab to PK was 86.3±117.7 days. Final BCVA was borderline better in the PK group (1.3 vs 2.1, p=0.051). Multivariate analysis identified previous PK (aOR 9.09, p=0.021) and anterior chamber wash (aOR 41.13, p<0.001) as only independent predictors for the need of PK. Phthisis rates were low and similar (2.7% vs 3.4%, p>0.99), while evisceration/enucleation was more frequent in non-PK (34.5% vs 13.5%, p=0.044).
Conclusions
In our cohort, previous PK and the need for anterior chamber wash were the only independent predictors for PK in fungal keratitis. Prior PK likely reflects pre-existing ocular comorbidity or advanced structural compromise, whereas anterior chamber wash may represent a marker of greater disease severity requiring aggressive intraocular management. Notably, evisceration/enucleation was more frequent in non-PK eyes, underscoring the importance of early risk stratification and a more proactive, aggressive therapeutic approach. Identification of predictors of PK may support optimized clinical management and timely surgical intervention, potentially improving both anatomical and functional outcomes.