Advancements in corneal surgery techniques and technology have significantly improved outcomes over time. However, these specialized procedures are not without their challenges. For residents and fellows, mastering the art of corneal surgery involves perfecting surgical techniques and learning how to anticipate, prevent, and manage complications that may arise.
These complications can stem from patient-specific factors (such as anatomical variations), underlying systemic or ocular conditions, intra-operative variables (like surgical technique and instrumentation), and variable healing responses
This article covers three major challenges faced by cornea specialists: corneal graft rejection (CGR), graft detachment (GD) after endothelial keratoplasty (EK), and cystoid macular edema (CME). Each section explores the underlying causes, preventative strategies, and management approaches to these complications in cornea surgery.
Complication 1: Corneal graft rejection
Corneal graft rejection (CGR) is a significant concern following corneal transplantation. Though more common with penetrating keratoplasty (PK), CGR can be seen with lamellar keratoplasty (LK) and EK procedures.
An important aside is necessary here. Primary graft failure (PGF) refers to an immediate, irreversible loss of graft clarity within the first few days to weeks post-keratoplasty, usually due to donor tissue issues or surgical trauma. PGF is relatively easy to diagnose at the slit-lamp because these patients have cloudy corneas post-operatively that never clear.
Secondary graft failure (SGF) is the loss of corneal transparency in a previously clear and functional corneal graft as a result of endothelial failure from immunological (rejection) or non-immunological (late endothelial loss) causes. Therefore, CGR is a subcategory of SGF.
CGR is an immune-mediated response that typically occurs weeks to months (or even years!) post-operatively, presenting with signs like endothelial rejection lines, keratic precipitates, and anterior chamber reaction.
Academically, it is important to appreciate the differences between these two entities, late endothelial failure and CGR; however, clinically, both may present with similar exam findings (such as recalcitrant corneal edema, etc.).
Risk of rejection in CGR
The cornea is considered an “immune privileged” area of the body due to factors such as its absence of vascularity and lymphatics, which prevent access to immune elements.1 This unique immune privileged status translates into relatively high graft survival rates compared to solid organ transplantation in low-risk (non-vascularized/noninflamed) cases.2,3
By contrast, in high-risk PK cases, rejection episodes can be as high as 30 to 60%, and 70% may fail within 10 years.4 Studies have shown that the risk of rejection varies by procedure type, with PK having the highest incidence, followed by Descemet stripping automated endothelial keratoplasty (DSAEK), deep anterior lamellar keratoplasty (DALK), and Descemet membrane endothelial keratoplasty (DMEK).5
CGR can result from various causes; therefore, early recognition and management of rejection episodes are essential to preserving graft function and optimizing long-term visual outcomes.
Common causes of corneal graft rejection
CGR occurs when the host immune system recognizes donor corneal cells as foreign and mounts an inflammatory response.6 The primary mechanisms include epithelial, subepithelial, stromal, and (most commonly) endothelial rejection. There are many potential risk factors for rejection that vary by the type of transplant: pre-operatively, intra-operatively, and post-operatively.
A non-exhaustive list of risk factors/causes is as follows:6,7
- Antigen load of donor
- Duration of tissue storage
- Host corneal vascularization
- Previous failed graft
- Ocular history (uncontrolled glaucoma, inflammatory eye disease, trauma, prior surgery, etc.)
- Poor graft preparation/placement
- Large graft
- Eccentric graft
- Young recipient (< 40 years)
- Loose sutures/exposed knots
Of these, immunological rejection has been cited as the leading cause, followed by late endothelial failure and ocular surface disorders.3 EK procedures, while associated with lower rejection rates than PK, can still fail due to reasons such as donor endothelial cell loss or detachment of the graft, which will be covered later in this article.
Additionally, improper post-operative medication adherence or premature tapering of corticosteroids can precipitate rejection episodes.
Preventative strategies to avoid CGR
Reducing the risk of CGR begins with careful patient selection and pre-operative optimization. Patients with high-risk features, such as significant corneal neovascularization, may benefit from pre-operative treatment in the form of laser photocoagulation, fine needle diathermy, anti-vascular endothelial growth factor (anti-VEGF) injections, topical steroids, or other newer strategies such as mitomycin-C intravascular chemoembolization (MICE).7,8
It is critical to treat and optimize comorbid ocular disease, such as controlling glaucoma or reducing existing ocular surface inflammation. Intra-operatively, careful surgical technique is essential; for example, oversized grafts in PK can increase rejection risk, while precise graft handling in EK can improve adherence and reduce endothelial trauma.3 The use of post-operative prophylactic topical corticosteroids, most commonly prednisolone, remains the treatment of choice for rejection prevention.9
It is important to note that other anti-rejection topical (cyclosporine A, tacrolimus, etc.) and systemic therapies (steroids, immunomodulators, etc.) may be required for high-risk cases.7 While close monitoring in the immediate post-operative period is essential, long-term follow-up is also recommended, as rejection episodes can occur months to years after surgery.6
Management approaches for corneal graft rejection
Early detection and aggressive treatment of CGR are critical to preserving graft clarity and function. Patients should be counselled and given information at each post-operative visit regarding concerning symptoms, such as redness, light sensitivity (photophobia), vision loss, and pain, which should prompt them to seek immediate evaluation; this “RSVP” mnemonic may be a helpful educational tool for patients.6
Clinical signs of CGR vary based on the layer affected but commonly include corneal edema (Figure 1), keratic precipitates on the corneal graft sparing the peripheral recipient cornea, corneal vascularization (Figure 2), stromal infiltrates, endothelial rejection lines (Khodadoust lines, Figure 3), epithelial rejection line, and subepithelial infiltrates.10
Figure 1: Slit lamp image of a corneal graft rejection after DSAEK (notice the "S" stamp still visible) 6 months later. Diffuse 3+ corneal edema present.
Figure 1: Courtesy of Kamran Riaz, MD.
Figure 2: Example of corneal graft rejection in a penetrating keratoplasty graft. Note the corneal neovascularization (arrows) and wedge-shaped corneal edema (asterisk) spreading from the left side with a still relatively clear cornea on the right side of the photo.
Figure 2: Courtesy of Kamran Riaz, MD.
Figure 3: Corneal graft rejection in an eye with penetrating keratoplasty. Note the profound sectoral corneal edema (asterisk) with the presence of a Khodadoust line (arrow).
Figure 3: Courtesy of Kamran Riaz, MD.
If examined early, anterior chamber cell infiltration without flare or graft abnormality may be seen.6 Pachymetry can also be used to detect an increase in edema.6 As a complement to slit lamp examination, imaging modalities such as in-vivo confocal microscopy, specular microscopy, and anterior segment ocular coherence tomography (AS-OCT) have been used to aid the diagnosis of CGR.11,12
CGR outcomes depend on both an early diagnosis and immediate steroid initiation.7 Treatment involves high-dose, frequent topical corticosteroids; adjunctive systemic immunosuppression may be required in severe cases.7,13 Refractory cases may necessitate additional interventions, including subconjunctival, intracameral, or intravitreal steroid injections.7
Other management options include cytotoxic agents (e.g., azathioprine, cyclosporin A, and corticosteroids) and newer immunomodulators (e.g., tacrolimus and rapamycin) with varying degrees of evidence and success.6 Repeated keratoplasty may be necessary in instances of irreversible CGR, though the prognosis diminishes with successive grafts.6
For failed PK grafts, surgeons have traditionally favored a repeat PK; however, there has been growing interest and success in recent years in performing an EK procedure under the existing PK to restore graft clarity and avoid risks associated with an open-sky procedure.
📚
3 Major Complications in Cornea Surgery Quiz
This quiz was designed for ophthalmology residents and fellows and reviews key topics from this article.
Complication 2: Graft detachment (GD)
GD is a complication following EK procedures, such as DMEK and DSAEK.14 Successful outcomes in these procedures depend on proper graft adherence, which can be influenced by donor/recipient risk factors, surgical technique, patient compliance, and underlying ocular conditions.15
Common causes of GD
GD can result from various factors, pre-, intra-, and post-operatively. Recipient-related risk factors include: older age, history of trabeculectomy or glaucoma drainage device (GDD), previous vitrectomy, previous failed PK, presence of an anterior chamber intraocular lens (ACIOL), and partial/complete defects in the iris-lens diaphragm.16,17
Several donor-related characteristics can be significant, such as:16,17
- Donor age
- Donor systemic diseases (e.g., diabetes)
- Tissue endothelial cell count (ECC)
- Death-to-preservation time (DPT)
- Storage time
- Use of preloaded vs. surgeon-loaded tissue,
- Use of thinner lenticules (for DSAEK), graft thickness, and
- Graft size
Intra-operatively, poor graft preparation, such as irregular graft thickness or endothelial cell damage, can compromise adherence.14 Of note, higher rates of GD have been reported post-DSAEK/DMEK during the initial learning curve of the corneal surgeon,17 which does reduce over time as experience is gained—this is very reassuring for the novice surgeon!
Intra-operative challenges, including inadequate air tamponade, improper graft positioning, or insufficient manipulation for centration, can also contribute to detachment.14-17 Post-operatively, patients must maintain supine positioning, and the inability to do so can lead to higher rates of GD.17 Eye rubbing and post-operative hypotony/hypertension have also been implicated as causes of GD.18
Preventative strategies to avoid GD
Preventing GD begins with proper patient selection and surgical technique. Surgeons should handle the graft carefully to minimize endothelial cell loss and ensure proper preparation. Intra-operatively, achieving proper centration and adequate air tamponade is critical.14,15,17,19 In high-risk cases, the use of sulfur hexafluoride (SF6) or perfluoropropane (C3F8) gas can prolong tamponade duration and improve graft adherence.20
Patient education is paramount pre- and post-operatively. To facilitate graft adherence, patients should be instructed to remain supine for at least 24 to 48 hours and avoid any face-down maneuvers. A helpful education tool is to tell patients they have a “2-hour debit card of time” they can use in 10 to 15-minute intervals during this immediate post-operative period to eat meals and perform other necessary activities.
Diagnosing GD can be challenging. In cases of moderate-severe GD, a slit-lamp examination can easily reveal the location and extent of GD (Figure 4). However, in cases of mild or partial GD, especially with thinner DSAEK and DMEK procedures, the diagnosis can be difficult, especially in the presence of post-operative corneal edema. In these cases, AS-OCT can be extremely helpful in determining GD (Figure 5).
Figure 4: Graft detachment in an eye with DMEK. On the left side image, there is prominent edema temporally. On the right side image, thin slit technique can help delineate an area of inferior graft detachment (blue arrow).
Figure 4: Courtesy of Kamran Riaz, MD.
Figure 5: AS-OCT image demonstrating a detached DMEK graft requiring rebubbling.
Figure 5: Courtesy of Kamran Riaz, MD.
Management approaches for graft detachment
The management of GD depends on the extent and location of the detachment. Partial GDs that do not involve the visual axis, especially ones located inferiorly, may be managed conservatively with close observation and have been known to re-attach spontaneously.16
However, significant detachments, superior detachments, or partial detachments where the cornea fails to clear (usually within 1 month) typically require rebubbling, a procedure in which an air or gas bubble is injected into the anterior chamber to reposition the graft.21
Many GDs can be successfully rebubbled at the slit-lamp; while trainee EK surgeons may be primarily focused on learning intra-operative surgical maneuvers, mastery of post-operative rebubbling is a vital skill for all corneal surgeons!
If rebubbling is unsuccessful or extensive graft slippage occurs, surgical repositioning in the operating room may be necessary, including consideration for using expansile gases, lengthy intra-operative tamponade, and/or anchoring sutures (for DSAEK lenticules) at the graft edge. In cases of unsuccessful rebubbling, repeat EK may be required
Complication 3: Cystoid macular edema (CME)
CME is a well-known complication following cataract surgery, particularly in patients with predisposing conditions such as uveitis, diabetes, or underlying retinal disease.22 It is characterized by fluid accumulation within the macula, leading to blurred vision and prolonged visual recovery.
While most trainee surgeons are aware of CME occurring after cataract surgery, awareness of CME after corneal transplant surgeries is less common.
Common causes of CME
CME typically arises from post-operative inflammation, which disrupts the blood-retinal barrier and allows fluid to accumulate in the macular layers.22 Surgical trauma, particularly to the iris, and intra-operative complications can exacerbate this inflammation.22,23
Studies looking at pre-operative risk factors for CME after DMEK have found conflicting results, with some failing to identify any risk factors and others pointing to any indication for endothelial surgery other than Fuchs’ endothelial dystrophy being a potential risk factor.23,24
Patients with pre-existing retinal conditions, such as epiretinal membranes, diabetic retinopathy, or venous occlusive disease, are at higher risk for developing CME, as are eyes that have a history of previous CME.22
Preventative strategies to avoid CME
Pre-operative assessment may be useful for identifying high-risk patients. OCT can be used to screen for pre-existing retinal pathology. Pre-operative laser iridotomy may be associated with a lower risk of post-EK CME and could be considered.25
Intra-operatively, minimizing phacoemulsification energy (in cases of combined cataract and EK surgery) and avoiding excessive iris manipulation can help reduce the risk of CME.22
Post-operative prophylactic initiation of frequent topical corticosteroids and possibly NSAIDS should be considered and balanced with the inherent side effects of these medications.24
Management approaches for cystoid macular edema
Early detection of CME is essential and should be suspected when visual acuity is lower than expected post-operatively, especially in the presence of a clear corneal transplant.24 Thus, it is critical to consider a dilated fundus exam in cases when the cornea and anterior segment are clear, yet the visual acuity is worse than expected.
A manifest refraction yielding a greater-than-expected hyperopic shift can also be helpful. However, the expected hyperopic shift after EK procedures may prevent the clinician from fully appreciating the possibility of EK. Therefore, a macular OCT is critical in diagnosing CME when the clinical exam may be difficult or equivocal. Scans using macular OCT will reveal intra-or sub-retinal fluid in the foveal region in CME.24
Figure 6: Macular OCT showing a case of CME after DSAEK.
Figure 6: Courtesy of Kamran Riaz, MD.
Serial OCT imaging, visual acuity measurements, and manifest refraction assessments are crucial for monitoring resolution and guiding treatment adjustments. While a standardized protocol for managing CME after EK has yet to exist, first-line therapy for CME after cataract surgery includes a combination of topical NSAIDs and corticosteroids to reduce macular inflammation.26
Some clinicians may also use a topical carbonic anhydrase inhibitor, while others eschew these medications out of concern for worsening corneal edema. This concern stems from their potential to impair corneal endothelial pump function, which can result in fluid accumulation within the cornea.27
In cases where CME persists despite topical therapy, periocular or intravitreal corticosteroid injections, such as triamcinolone, may be employed for more potent anti-inflammatory effects.22 Referral to a retina specialist for consideration of intravitreal anti-VEGF therapy may be considered, particularly if underlying vascular pathology contributes to fluid accumulation.22
Future prospective studies on CME management in EK are needed to develop a specific treatment strategy.
Conclusion
Complications, such as CGR, GD, and CME, are inevitable in some cases, but their impact can be minimized through careful pre-operative planning, surgical technique, and timely post-operative management.
By understanding the underlying causes and employing evidence-based strategies, surgeons can navigate these challenges effectively, ensuring better patient outcomes.
