Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

Corneal transplantation: DMEK, DSAEK and DALK

Corneal transplantation replaces diseased corneal tissue with donor tissue: full thickness in penetrating keratoplasty (PK), or selectively in lamellar techniques such as deep anterior lamellar keratoplasty (DALK) for stromal disease and endothelial keratoplasty (EK: DSAEK and DMEK) for endothelial disease [S1] [S2]. Lamellar techniques reduce rejection compared with PK [S2] [S10] [S12]. Graft rejection is the most common cause of late graft failure after PK [S4].

Key points - For Fuchs dystrophy and bullous keratopathy, EK appears to give higher 5-year graft survival and lower rejection than PK, on low-level evidence [S10]. - DMEK gives better visual acuity than ultrathin DSAEK but needs more re-bubbling [S9] [S11]. - For keratoconus, DALK and PK give comparable vision; DALK has less rejection [S2] [S12]. - After PK, pseudophakic or aphakic corneal oedema and prior glaucoma raise the risk of graft failure [S3]. - Evidence for immunosuppressants to prevent rejection after PK is largely low quality [S4].

Evidence last checked: 2026-10-17 (PubMed abstracts).

Red flags

• A graft patient with a red eye, reduced vision, light sensitivity or pain: possible rejection or infection; graft rejection is the most common cause of late failure after PK [S4] (see "Refer when").
• Raised eye pressure on post-graft steroids. After EK, ocular hypertension occurred in 20% with standard steroids and 7% with soft steroids [S7].
• Glaucoma history. 10-year graft failure after PK was 58% with prior glaucoma surgery and medication versus 22% without [S3].
• Persistent epithelial defect after keratoplasty [S2].

Recognise it: who needs a graft

• Fuchs endothelial dystrophy: endothelial loss leads to corneal oedema and reduced vision, and replacement with donor tissue is the only successful surgical treatment [S1].
• Pseudophakic or aphakic corneal oedema (bullous keratopathy) [S3] [S5].
• Keratoconus: about 10 to 15% of patients diagnosed with keratoconus require transplantation [S2].

Confirm it: choosing the technique

• Endothelial disease (EK vs PK). A Cochrane review of 3 small RCTs found no strong evidence of a difference in final vision between EK and PK, fewer higher-order aberrations but greater endothelial cell loss after EK [S1]. A systematic review of 50 cohort studies found 5-year graft survival of 0.93 to 0.99 after DMEK versus 0.69 to 0.98 after PK in Fuchs dystrophy, and 5-year rejection of 1.7 to 2.6% after DMEK, 5.0 to 7.9% after DSEK and 11.0 to 28.7% after PK [S10].
• DMEK vs ultrathin DSAEK. A meta-analysis of 7 studies found better 12-month vision after DMEK (SMD 0.50) but more re-bubbling (RR 0.33 favouring UT-DSAEK); other outcomes did not differ significantly [S9]. A review of 4 systematic reviews reached the same conclusion [S11].
• Keratoconus (DALK vs PK). A Cochrane review of 2 RCTs found no difference in vision and less rejection with DALK (OR 0.33, moderate certainty) [S2]. A 2025 meta-analysis of 47 studies (27 018 eyes) found comparable vision, more rejection with PK, but lower graft failure with PK than DALK [S12].

Manage it

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Tier A (primary care / no specialist)

• Refer any grafted eye that becomes red, painful or blurred (see "Refer when"); rejection is the most common cause of late graft failure after PK [S4].
• Post-graft drug regimens are not covered by these abstracts: follow the graft surgeon's plan and local formularies.

Tier B (general eye clinic)

• Check intraocular pressure in patients on post-graft steroids; soft steroids caused less ocular hypertension than standard steroids after EK [S7].
• Monitor endothelial cell density and corneal thickness where possible: after PK, lower density and higher thickness during follow-up were associated with later failure, although most corneas remained clear at 10 years [S3].

Tier C (specialist centre)

• Rejection prophylaxis after EK. In 49 studies (12 893 eyes), DMEK had less rejection than DS(A)EK; standard and soft steroids did not differ in preventing rejection after DMEK [S7].
• Immunosuppression after PK. A Cochrane review of 6 trials found low-quality evidence: systemic mycophenolate may not improve clear graft survival but may reduce rejection, and topical cyclosporine probably makes little difference to rejection after normal-risk grafts [S4].
• Donor and surgical factors. In the Cornea Preservation Time Study (1330 DSAEK eyes), 94% of grafts were clear at 3 years; diabetic donor tissue and operative complications increased early failure, and PACE increased late failure compared with Fuchs dystrophy [S5]. In the Cornea Donor Study, 10-year PK failure was 37% for PACE versus 20% for Fuchs dystrophy [S3].
• Surgeon training in low- and middle-income settings. In an Indian survey, DMEK was performed by 55% of responding surgeons, only 40% of whom had formal training; lack of training was the main barrier and DSEK remained the most common EK [S8]. For DALK, big-bubble technique was most preferred, and 50% of surgeons felt more than 20 cases were needed to pass the learning curve [S6]. These are surveys with low response rates.

Refer when

Pitfalls

• Treating a red grafted eye as simple conjunctivitis [S4].
• Assuming DMEK is always best: it gives better vision but more re-bubbling [S9] [S11].
• Forgetting glaucoma and PACE as risk factors for failure [S3] [S5].
• Missing steroid-induced ocular hypertension [S7].

Evidence notes

• Sources conflict on graft failure after DALK vs PK: no difference in the Cochrane review [S2], lower failure with PK in a larger observational meta-analysis [S12].
• The EK vs PK long-term data are from cohort studies with low level of evidence [S10]; the RCTs were small and lacked masking [S1].
• Most trials come from high-income countries; the two Indian sources are surveys [S6] [S8].
• Not covered by these sources: eye banking and donor supply in low-resource settings, suture management, treatment regimens for acute rejection, and keratoprosthesis.

Related articles

• Keratoconus and corneal cross-linking
• Fungal keratitis
• Herpes simplex keratitis

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.