Complications, follow-up & immunosuppression

The three major causes of failure

1. Rejection (mostly endothelial) — see Ch. 06.
2. Late endothelial failure — progressive attrition of cell density, even without rejection, leading to mid/long-term decompensation (particularly in PK and DSAEK).
3. Recurrence of the original disease — stromal dystrophies (deposits on/within the graft), certain surface diseases.

Complications by major technique

Immunosuppression & follow-up

The cornerstone of rejection prevention remains topical corticosteroids, intensive at first then tapered, often maintained at low dose long-term (particularly in endothelial and high-risk cases, where it may be near-indefinite). Follow-up includes monitoring endothelial cell density (specular microscopy), intraocular pressure (corticosteroids raise it), clarity and astigmatism. In high-risk cases, add systemic or topical immunosuppressants (ciclosporin, tacrolimus) and control of neovascularisation.

Special situations

Paediatrics

Congenital opacities (dystrophies, anterior segment anomalies, Peters): grafting is surgically more difficult (soft cornea, positive pressure, brisk inflammation) and the key issue is amblyopia — hence the relative urgency and the importance of visual rehabilitation. Rejection is more frequent and follow-up (examinations under anaesthesia) demanding.

High immunological risk

Neovascularised cornea, repeat grafts, inflammation: favour the least immunogenic technique possible, aggressively control neovascularisation and inflammation, discuss intensified immunosuppression and, in extreme cases, upfront KPro.

Infectious keratitis & herpes

"Hot" grafting (tectonic/therapeutic) treats an uncontrolled perforation or abscess, with a more guarded prognosis and a risk of infectious recurrence. In herpetic keratitis, peri- and postoperative antiviral prophylaxis is decisive in preventing recurrence and rejection: recent data stress the importance of the timing of introduction (prophylaxis started several days before surgery being associated with the best outcomes) and the existence of early ("breakthrough") and late (on withdrawal) recurrences.

Chemical burns & severe dry eye / GVHD

These collapsed surfaces first require reconstruction (amniotic membrane, limbal grafting) and stabilisation; keratoplasty or KPro come only afterwards, on a prepared ground.

Perspectives & synthesis 2026

The trajectory is clear: decline of full-thickness grafting, dominance of endothelial lamellar surgery (DMEK), and rise of regenerative and bioengineered approaches that, for certain indications, replace the graft itself.

The forces at work

• Regenerate or manufacture the endothelium rather than replace it with a donor: DSO ± ROCK, injectable cell therapy (Vyznova in Japan, AURN001 in pivotal development) and the 3D-bioprinted endothelium (Precise Bio PB-001, world-first transplant in 2025) — potential to massively relieve the shortage.
• Manufacture the stroma: acellular substitutes (BPCDX), bioprinting, scaffolds — broadened access, notably in resource-limited countries.
• Prevent recurrence through gene therapy in hereditary dystrophies.
• Artificial intelligence & imaging: surgical planning (DALK/CAIRS nomograms), imaging quantification (anterior-segment OCT, endothelial density), aid to rejection prediction and tissue selection.
• Additive rather than subtractive: CAIRS, rings, upstream cross-linking — preserving the native cornea as long as possible.

* DSO±ROCK: slow and unpredictable recovery, inseparable from ROCK inhibitors; it consumes the peripheral endothelial reserve and may therefore worsen the prognosis of a subsequent DMEK if it fails (see Ch. 10).