DWEK (Descemetorhexis Without Endothelial Keratoplasty) is a surgical procedure for Fuchs endothelial corneal dystrophy (FECD) that involves only surgically stripping the central Descemet membrane (DM) without transplanting donor corneal endothelium. It is also called Descemet stripping only (DSO) 1).
FECD is the most common indication for corneal transplantation worldwide, accounting for about 39% of all corneal transplants 2). In recent years, selective endothelial transplantation (DSAEK/DMEK) has become mainstream, replacing penetrating keratoplasty (PKP), but issues such as graft rejection and donor tissue availability remain.
The concept of DWEK originated from case reports of spontaneous resolution of corneal edema after graft detachment following endothelial transplantation or after Descemet membrane removal 1). Such spontaneous resolution is unique to FECD and is not observed in bullous keratopathy, a disease of endothelial cell depletion 1). Based on this observation, several research groups began investigating intentional Descemet membrane stripping alone 1). Subsequent studies reported promising results, but also revealed unpredictable variability in outcomes 2).
They are different names for the same procedure. In addition to DWEK (Descemetorhexis Without Endothelial Keratoplasty), it is also called DSO (Descemet stripping only), Descemet stripping without endothelial keratoplasty, among other names. All involve stripping the central Descemet membrane without transplanting a donor cornea.
DWEK requires careful patient selection. The optimal candidates are FECD patients with centrally localized lesions, clear peripheral cornea, and abundant endothelial cells.
Indications
Confirmed FECD: Cases confirmed as FECD by clinical examination and confocal microscopy
Central guttae: Determined to be the main cause of visual symptoms (decreased visual acuity, contrast sensitivity, and glare)
Peripheral ECD: ≥1000 cells/mm² by specular microscopy or confocal microscopy
Lens status: Both phakic and pseudophakic eyes are acceptable
Contraindications
Severe corneal stromal edema: Advanced cases with opacity, bullae, and Descemet’s folds
Low peripheral ECD: <1000 cells/mm²
Secondary corneal pathology: When other corneal diseases coexist
History of viral keratitis: History of herpes simplex virus or cytomegalovirus keratitis
A peripheral endothelial cell density of ≥1000 cells/mm² is an important candidate criterion. Central corneal thickness (pachymetry) has been reported to have a weak correlation with postoperative reaction rate and final visual acuity2). The effects of age, sex, smoking history, and genetic factors on CEC migration are not fully understood2).
DWEK is performed under local or general anesthesia. The standard surgical time is approximately 6 minutes.
It has been reported that surgical technique significantly affects outcomes 2).
In a retrospective study, some cases treated with the 360-degree scoring method did not achieve corneal clarity. In contrast, the “two-flap method,” which involves complete Descemet membrane stripping after scoring two clock hours, achieved clarity in all cases 2).
The size of Descemet membrane stripping is also an important factor. Large-diameter stripping of 6.0–9.0 mm used in early studies frequently resulted in persistent corneal edema 2). A small-diameter stripping of 4.0 mm is recommended; expanding the stripping diameter by 2.0 mm more than doubles the area that remaining CEC must regenerate, so stripping larger than 4.0 mm is considered excessive 2).
Additionally, irregularity of the stromal surface impedes CEC migration and causes prolonged localized edema 2).
It can be performed simultaneously with cataract surgery (phacoemulsification + intraocular lens implantation), known as “triple-DWEK.” Simultaneous surgery is not considered to adversely affect DWEK outcomes.
The outcomes of DWEK in major reports are shown below.
| Report | Clearance rate | Mean recovery period |
|---|---|---|
| Report 12) | 14/17 eyes (82%) | Approximately 3 months |
| Report 21) | 9/12 eyes (75%) | Not reported |
| Report 31) | 10/13 eyes (77%) | Not described |
Studies using 4mm small-diameter Descemet membrane stripping have reported a clearance rate of approximately 75-82%. In Report 2, two eyes that did not achieve clearance were additionally treated with topical ripasudil, and eventually clearance was achieved1).
Recovery time is longer compared to standard endothelial transplantation (DSAEK/DMEK)1). There is wide individual variation, ranging from cases that rapidly clear within 1 month postoperatively to cases that require 6-8 months2).
Rapid Response Group
Timing: Within 1 month postoperatively
Findings: Appearance of endothelial cell mosaic pattern in the center, resolution of corneal edema
Response Group
Timing: Within 3 months postoperatively
Findings: Gradual reduction of corneal edema and completion of clearance
Slow Response Group
Timing: After 3 months postoperatively
Findings: Gradual progression of clearance over 6-8 months
Non-responder group
In 4 cases where DWEK was performed on both eyes, similar clearing times were observed in the right and left eyes, suggesting that genetic factors and the growth factor environment in the anterior chamber may be involved in the recovery rate2). High preoperative pachymetry has been reported to be weakly correlated with a lower response rate and final visual acuity2), but no significant differences in age, baseline ECD, or pachymetry were found between the responder and non-responder groups2).
Corneal clearance after DWEK is thought to occur due to migration of healthy peripheral corneal endothelial cells (CECs) to the central denuded area 2). The observation of relatively low central CEC density after DWEK suggests that migration, rather than proliferation, is the primary mechanism 1).
In FECD, central CECs are preferentially damaged, while the periphery remains relatively normal 2). Rosettes (structures of degenerated CECs arranged around guttae) are also more common centrally and less common peripherally 2).
Normally, peripheral CECs do not spontaneously migrate to the center because contact inhibition is maintained 2). Endothelial progenitor cells derived from the neural crest (NCC) in the periphery are inactivated by contact inhibition 2). When contact inhibition is released by Descemet membrane stripping, proliferation and migration of progenitor cells and mature CECs are stimulated 2).
The cytokine TGFβ may also be involved in maintaining contact inhibition of the CE monolayer 2). In an ex vivo human corneal culture model, the presence of intact Descemet membrane, younger donor age, and supplementation with the Rho kinase inhibitor Y-27632 have been shown to promote CEC migration 2).
It has been reported that the size of guttae affects CEC behavior 2). Cultured CECs (HCEnC-21T) seeded on Descemet membrane from FECD patients showed delayed cell adhesion compared to normal Descemet membrane 2). CECs require a normal Descemet membrane, and healthy CECs are essential for normal ECM production. This bidirectional interaction between CECs and ECM is called “dynamic reciprocity” 2).
Descemet membrane stripping larger than 4.0 mm results in an area too large for remaining CECs to regenerate 2). An increase in diameter to 2.0 mm more than doubles the surface area, potentially exceeding the migration and regeneration capacity of CECs 2). This is thought to be the reason for poor outcomes with the initial large-diameter stripping of 6.0–9.0 mm 2).
In some reports, for cases where transparency was not achieved with DWEK alone, the addition of topical ripasudil ultimately achieved transparency. This suggests that the indications for DWEK may be expanded 1).
Rho kinase inhibitors have been confirmed in basic research to promote CEC migration 2), and are expected to play a role as an adjunctive therapy after DWEK. The use of netarsudil is also being attempted.
The main reason DWEK has not become mainstream treatment is the difficulty in predicting whether a patient will benefit from the surgery 1). Research is ongoing to clarify the relationship between preoperative, intraoperative, and postoperative factors and corneal transparency 2). Genetic factors, anterior chamber environment, and standardization of surgical techniques are future challenges.
Since DWEK does not require donor corneas, it is expected to be applied in regions with severe donor shortages. If patient selection criteria and surgical techniques are optimized, it may become established as a treatment option for early-stage FECD.
