Review evaluates corneal-sparing techniques for ectatic diseases, suggesting a new functional restoration framework.
Corneal ectatic and degenerative disorders have traditionally been managed with tissue-replacement strategies such as penetrating keratoplasty (PK) and deep anterior lamellar keratoplasty (DALK). However, an improved understanding of corneal biomechanics, stromal wound healing, and immune privilege has driven a paradigm shift toward tissue-sparing and layer-selective interventions. In particular, biologically integrated stromal procedures, including Bowman layer transplantation (BLT), corneal allogenic intrastromal ring segments (CAIRS), and stromal lenticule addition keratoplasty (SLAK), aim to restore structural integrity, improve corneal shape, and preserve native tissue while minimising the risks associated with full-thickness transplantation. In this review, we summarise the structural basis of corneal biomechanics, the pathological mechanisms of stromal failure in keratoconus and related disorders, and the biological responses that follow stromal interventions. We then evaluate current evidence on corneal cross-linking (CXL), stromal reinforcement, and stromal addition procedures, with emphasis on indications, mechanisms, clinical outcomes, and limitations. Reflecting this evolution, we frame these procedures as the restoration of specific biomechanical and optical functions rather than mere tissue replacement, aligning corneal surgery with the broader trend toward selective tissue restoration and minimally invasive biological augmentation. To support clinical decision-making, we propose a three-tier Stabilise-Reshape-Replace (SRR) framework. Each tier addresses a dominant clinical problem: SRR is intended as a pragmatic expert-derived scaffold rather than a validated treatment algorithm.
No takes yet. Share an insight, caveat, or question.
Hayashi et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: