DSEK for Corneal Dystrophy in Dogs

• Descemet's stripping endothelial keratoplasty (DSEK) is an effective surgical treatment for corneal endothelial dystrophy in dogs
• The study aims to describe and assess the clinical outcome and complications of using DSEK to treat canine corneal endothelial dystrophy.
• Six dogs with progressive corneal edema from abnormal dystrophic endothelial cells underwent DSEK.
• Patients were examined pre- and postoperatively over three months, assessing corneal edema and thickness with ultrasonic pachymetry.
• DSEK graft positions were evaluated via optical coherence tomography at three months post-surgery, and intra- and postoperative complications were noted.
• Corneal edema and thickness improved postoperatively in all six patients.
• Fibrin was observed intraoperatively in 1/6 eyes and postoperatively in 2/6 eyes
• A DSEK graft was partially scrolled in 1/6 eyes.
• Secondary ocular hypertension was seen in 1/6 eyes.
• Corneal vascularization was found in 4/6 patients.
• The early postoperative results suggested DSEK as an effective treatment for corneal endothelial dystrophy in dogs by resolving corneal edema and reducing corneal thickness.
• Further studies are needed to document long-term complications and graft longevity.

Introduction

• Corneal endothelial dysfunction in dogs results from canine endothelial dystrophy, intraocular surgery, glaucoma, diabetes, and uveitis.
• Canine endothelial dystrophy is similar to Fuchs' dystrophy in humans.
• Endothelial cell population decreases with age, natural endothelial cells are thought to have limited regenerative capacity, which can lead to fluid accumulation
• Untreated endothelial decompensation can result in transparency loss and vision loss.
• Secondary bullous keratopathy and ulcerative keratitis can ensue, causing ocular discomfort.
• Palliative therapy is often used for canine endothelial dystrophy, including hypertonic saline and bandage contact lenses.
• Surgical management includes collagen cross-linking, thermokeratopasty, and conjunctival flaps.
• Additional physician-based techniques are anterior stromal puncture, mid-infrared laser coagulation, amniotic membrane transplantation and phototherapeutic keratectomy.
• These are often palliative to manage bullous keratopathy and recurrent ulcers with limited visual potential.
• Visual patients with Fuch's corneal endothelial dystrophy (FCED) commonly undergo corneal transplantation procedures like DSEK (Descemet's stripping endothelial keratoplasty), DMEK (Descemet's membrane endothelial keratoplasty), and PK (penetrating keratoplasty).
• Conventional penetrating keratoplasty is being replaced by selective lamellar keratoplasty like DMEK and DSEK, because these procedures improve visual outcomes and have fewer complications.
• DMEK involves selective transplantation of DM and endothelium.
• DSEK involves selective transplantation of DM, endothelium, and posterior stromal lamellae.
• Of these three surgical techniques, only PK has been reported in canines with a rejection rate of 56%.
• The study evaluates DSEK's efficacy in treating canine corneal endothelial dysfunction and reports any intra- and postoperative complications.

Materials and Methods

• A prospective, nonrandomized clinical study included cases of progressive corneal edema.
• Progressive corneal edema was the inclusion criterion.
• Exclusion criteria: active ulceration and/or corneal vascularization; concurrent ocular disease; and patients ≥12 years of age.
• The New South Wales Government, Department of Primary Industries approved the study.
• Owner consent was obtained and procedures followed Association for Research in Vision and Ophthalmology guidelines for animal research.

Ophthalmic examination

• Eyes underwent ophthalmic examination with slit lamp biomicroscope, indirect ophthalmoscope, Schirmer tear test-1, fluorescein staining, and applanation tonometry before surgery and over three months after.
• Corneal edema was graded from 0 to 4.

Measurement of corneal thickness

• Corneal thickness was measured using ultrasonic pachymetry pre- and post-surgery over three months.
• Central corneal thickness was measured using ultrasonic biomicroscopy if ultrasonic pachymetry could not be obtained.

Retrieval of donor corneas

• Donor corneas were retrieved from animals without systemic disease that could compromise the corneal health.
• Eyes were examined postmortem using slit lamp biomicroscope.
• Only healthy eyes, as outlined by the Eye Bank Association of America, were used
• Globes were retrieved within 4 hours of death using aseptic technique.
• Globes were placed in 5% polyvinyl pyrrolidone iodine solution for 5 minutes, followed by sterile 0.5% sodium thiosulphate, then rinsed off with sterile saline.
• Corneoscleral tissue was retrieved by incising the sclera approximately 3mm from the limbus.
• The incision was extended along the entire limbus, a stab incision made into the ciliary body, and a final excision made with Westcott scissors.
• Under sterile conditions, the corneoscleral tissue was placed in a unit with the endothelium uppermost, containing Optisol GS, and stored at 4°C.
• Transplantation of the cornea took place within 24 hours following globe retrieval.

Endothelial cell count

• The corneoscleral tissue was grasped and rinsed in sterile petri dish of balanced saline solution (BSS) in laminar flow cabinet.
• It was then placed on a clean microscope slide, endothelial side up, and two drops of 1.8% sterile sucrose solution were applied to the endothelial surface.
• cell counts were performed on alight microscope equipped with a camera.
• Image capture and analysis was performed with corneal endothelial cell analyzer.
• Two drops of Trypan blue 0.06% were applied to the endothelial surface for 45 seconds, then rinsed off with BSS solution.
• Cell viability was estimated by visual inspection of the endothelial cell layer as functional endothelial cells that do not retain stain.
• Corneal thickness data was recorded in the center and peripheral cornea.
• Only donor corneas with endothelial cell densities >2000 cells/mm² were qualified for transplantation.

Preparation of the DSEK graft

• The corneal structure was maintained using a custom built artificial anterior chamber loaded with a bubble of Optisol GS.
• Corneal thickness was measured with an Accutome Pachpen.
• Accurate depth knives were used to make a partial thickness stab incision through the stroma, aiming to leave a posterior stromal depth of 150 - 200µm.
• The lamellar keratectomy red spatula was then used to dissect the corneal lamella at the desired depth made by the initial stab incision followed by dissection of the proximal two-thirds of the cornea with the lamellar keratectomy white spatula.
• Then, the lamellar keratectomy blue spatula was used to complete the dissection across the cornea to the limbus.
• Left to right corneal section scissors were then used to separate the anterior stromal flap from the underlying posterior stroma, leaving a short 5mm circumferential segment still attached to the limbus.
• A sponge was used to remove excess fluid from the anterior stroma so that a small "F" mark could be made with a sterile gentian violet marking pen.
• The dissected anterior stroma was then placed back on top of donor graft and stored in Optisol GS until the tissue was used in surgery.
• The prepared DSEK graft tissue was then returned to cold storage at 4°Celsius until the time of surgery.
• Immediately prior to surgery, the precut cornea was loaded onto a silicone punch block with the endothelial side up.
• A 13 mm trephine blade was then applied to the endothelial surface.
• Firm pressure was applied onto the blunt end of the trephine to cut through the posterior lamellar graft and anterior stromal cap.
• The endothelial surface was covered in Optisol solution to maintain hydration.

Preparation of recipient cornea

• A circular impression was made using a gentian violet stained 13 mm trephine blade on the recipient corneal surface to mark the outline of where Descemet's membrane was to be stripped.
• A 7 mm partial-thickness corneal groove was made at 12 o'clock.
• A 15°, 3 mm depth Beaver Micro-Sharp blade was used to perform a side port incision at the 6 and 10 o'clock position at the limbus.
• Trypan blue 0.06% was then injected intracamerally for 45 seconds to stain the endothelium and its borders and thus facilitate stripping of DM.
• Then, 1.6% sodium hyaluronate was injected intracamerally to maintain the anterior chamber.
• A 2.75mm slit knife was used to enter the anterior chamber through a 2-step incision at the limbus at 12 o'clock.
• Using the marked 13mm corneal circle as a guide, the DM was stripped along the marked circle with a Terry Reverse known as descemetorhexis.
• The stripped DM-endothelium complex was removed through the 12 o'clock incision.
• A 9/0 ethilon suture was looped through the 12 o'clock limbal incision and across the anterior chamber to exit at the 6 o'clock side port incision.
• Viscoelastic was aspirated through the 12 o'clock limbal incision using irrigation/ aspiration (I/A) on an Infiniti Vision System and BSS solution mixed with 1IU/ml heparin sodium.
• Fibrin was encountered intraoperatively in patient 1, which prevented the complete unfolding of the DSEK graft because heparin was not added into the BSS solution.
• Patients 2-6 had heparin added to the BSS solution prophylactically.
• The DSEK graft (endothelial side up) and anterior stromal cap were transferred onto the surface of the recipient cornea, and the 9/0 ethilon suture needle was threaded through the DSEK graft.
• The anterior stromal cap was removed, and 1.6% sodium hyaluronate was injected onto the upper conjunctival fornix to prepare a protective bed.
• Graft exposure to air should be kept to a minimum at ≤15 minutes.
• The 12 o'clock incision was then extended to full-thickness with corneal section scissors.
• The graft was then pulled through the 12 o'clock incision into the anterior chamber via traction on the preplaced 9/0 ethilon suture through the 6 o'clock incision.
• The graft was then unfolded with surface compression and stroking using the shaft of the viscoelastic cannula.
• Any residual interface fluid was massaged away from the center to the periphery.
• The anterior chamber was filled completely with air(hyperinflated) to a pressure of ≥30 mm Hg and left undisturbed for 10 minutes.
• A small amount of air was removed from the anterior chamber, leaving behind a 7-9 mm air bubble.

Postoperative medical therapy

• Postoperative treatment: topical 1% prednisolone acetate 6 times a day for 1 month then tapering down, chloramphenicol 10 mg/g, polymyxin B sulfate 5000 µg/g twice daily for 2 weeks, and topical tacrolimus 0.02% ointment twice daily for 3 months.
• Oral medications - amoxicillin-clavulanic acid 12.5 mg/kg BID for 10 days, prednisolone 1 mg/kg BID for 7 days, then 1 mg/kg once daily for 7 days, then 0.5 mg/kg once daily until corneal vascularization was no longer observed, and tramadol hydrochloride 4 mg/kg BID for 10 days.
• Cyclosporine 5 mg/kg once daily for 30 days was started at Day 7 postoperative in patients 1 and 2 and immediately postoperative in patients 3-6.
• An Elizabethan collar was used for 2 weeks postoperatively to prevent self-trauma.

Evaluation of graft position

• DSEK graft position was evaluated with optical coherence tomography (OCT) device at 3 months postoperatively.
• Interface between the DSEK graft and recipient's cornea and respective thicknesses were also evaluated.

Results

• Six dogs were included, mean age of 11.2 ± 1.4 years
• Mean Schirmer tear test-1 values and mean intraocular pressures of the operated eyes were 17.8 ± 2.8 mm/min and 9.5 ± 3.9 mm Hg, respectively.
• Median duration of corneal edema before DSEK as assessed by the owner or diagnosed by a veterinary ophthalmologist was 14.5 ± 3.4 months and 10.5 ± 6.5 months, respectively.
• Two dogs had one episode of ulcerative keratitis in the year preceding surgery, but no patients had ulcerative keratitis at the time of surgery.
• Patient 5 had diabetic cataracts and underwent bilateral phacoemulsification with placement of intraocular lenses two years prior to DSEK surgery.
• He also had an acute episode of marked aqueous flare and superficial ulcerative keratitis three weeks postphacoemulsification that ultimately resolved following aggressive anti-inflammatory management.
• Patient 5 also had axial stromal bullae in the operated eye at the time of DSEK surgery
• Patient 6 had an acute flare of atopic dermatitis on the day of DSEK surgery.

Postoperative examination findings and complications

• Corneal clarity and visualization of intraocular structures were notable in all six eyes postoperative.
• No corneal ulcers were identified in any of the six patients during the study period.
• In four out of the six patients, corneal edema score was worse at day 1 postoperative versus preoperative, due to direct contact of the recipient stroma post-descemetorhexis with irrigation fluid intraoperatively, leading to corneal hydration and ineffective pumping by the donor corneal endothelium.
• Gradual reduction in corneal pigmentation was observed in patient 1.
• In patient 2, partial detachment and scrolling of the graft at the lateral edge, leading to temporary increase in corneal edema score on Day 21 to Day 28.
• Graft scrolling did not advance further after Day 28 and corneal fibrosis occurred.
• Anterior stromal vascularization was observed in patients 1, 2, 5, and 6 on Days 7, 7, 6, and 3, respectively.
• Deep posterior stromal vascularization was seen in patient 6 on Day 39.
• Patient 5 experienced markedly elevated intraocular pressure, 70 mm Hg OS, on Day 6.
• Intraocular pressure reduced to 34 mm Hg on medical therapy with topical brinzolamide, topical latanoprost, and oral glycerol.
• Fibrin was observed on Day 12 when corneal clarity had improved and subsequently lysed with an intracameral injection of tPA 20 µg.
• IOPS were subsequently controlled with topical brinzolamide BID.
• On Day 26, patient 5 was started on atorvastatin 40 mg SID to treat hyperlipidemia.
• Glucose levels were uncontrolled in spite of increasing his insulin dosage 12 days after being on atorvastatin,
• Atorvastatin was stopped after 22 days of treatment, and glycemic control was re-established.
• Patient 6 had persistent uveitis with ocular hypotension and miosis up to Day 25 postoperatively.
• Fibrin was also noted in patient 6 on Days 11 and 25, which required lysis with intracameral tPA 20µg injections.
• Patient 2 had diarrhea 17 days postoperatively that required a temporary 3-day cessation of oral cyclosporine; subsequent resumption of cyclosporine was uneventful.

Optical coherence tomography

• Corneal cross-sectional images demonstrated that all the transplanted grafts were well positioned against the posterior stroma of the recipients.
• Mild central posterior stromal fibrosis, seen as host-donor interface reflectivity, was noted in one patient.
• Lateral edge of the DSEK graft was partially detached and scrolled in one of the six patients
• Thick hyper-reflective areas on the underlying endothelium were noted in patients 5 and 6

Statistical analysis

• Statistical analyses were conducted in GenStat v17 (VSNi), and a P-value of <0.05 was considered significant.
• Correlation between corneal thickness and corneal edema score was assessed.
• Reduced Maximum Likelihood (REML) models were used to determine the effect of time post-surgery on corneal thickness and edema score, with a random effect of dog included.
• These models were conducted on data obtained from patients 1-4 as they had no existing systemic problems prior to surgery.
• Post hoc testing using least significant differences (LSD) was conducted to determine pairwise differences.

Corneal edema grading

• Reduced Maximum Likelihood modeling showed a significant change in edema score over time.
• The score reduced most significantly from Day 0 to Day 28 in patients 1-4 Post hoc testing indicated that there was a significant difference between Day 0 and Day 28; Day 28 and Day 56; and Day 56 and Day 84.
• Stable corneal edema scores were obtained at a mean time of 39 ± 14 days postoperative.
• In four out of the six patients, corneal edema scores were greater 1 day postoperatively compared to preoperative scores.
• Patient 5 had a decrease in corneal edema score on Day 14-60; however, there was a marked increase in corneal edema score on Day 74.
• Postoperative corneal edema scores for patient 6 did not differ from preoperative scores.

Central corneal thickness

• Reduced Maximum Likelihood modeling showed a significant change in corneal thickness over time.
• the thickness was most significantly reduced from Day 0 to Day 28 after surgery in patient 1-4.
• Post hoc testing indicated that there was a significant difference between Day 0 and Day 28; Day 28 and Day 56; and Day 56 and Day 84.
• In patients with no postoperative intraocular complications (patients 1, 3, and 4), mean time to the first appearance of stable central corneal thickness was 26 +4 days.
• A correlation of 0.88 between central corneal thickness and corneal edema score in all six dogs was obtained.

Discussion

• This study showed DSEK is a viable surgical treatment of progressive corneal edema owing to canine corneal endothelial dysfunction and is the first report of this surgery in this species.
• Decreased corneal thickness and improvement of corneal edema were observed in four dogs at approximately one month postoperatively.
• Two dogs experienced substantial postoperative complications and thus did not improve following surgery.
• The study demonstrated an excellent correlation between central corneal thickness and the corneal edema score suggesting that that corneal edema score can be used as a surrogate if corneal thickness cannot be measured in dogs whose corneal thickness cannot be measured due to advanced corneal opacity
• Partial graft detachment was observed in one patient.
• Factors that could have caused the focal graft detachment include: the absence of recipient stromal fibrils to initially bind to the donor stromal fibrils, viscoelastic in the graft/ recipient interface, mismatch of the donor and recipient curvatures causing a portion of the donor to arc away from the recipient, trapped fluid in the graft/ recipient interface, and the patient rubbing the eye.
• Stromal vascularization was observed in four out of six patients during the study, of which three had a 7mm limbal incision
• In the Patient that had already presented pre-existing corneal pigmentation, had vascularization in the pigmented area.
• Corneal grafts that have not cleared after 2 months are classified as primary graft failures.
• Graft failures are usually due to lack of endothelial function from unhealthy tissue, unhealthy recipient circumstances, or surgical technique.
• The corneas of patients 5 and 6 failed to clear as expected: Fibrin was observed in the anterior chamber post-operatively in patients 5 and 6, which necessitated intra-cameral injection of tissue plasminogen activator in both cases, it had been made necessary since the patient eyes where producing an overabundance of inflammatory cells.

Corneal Endothelial Cell Loss

• A review of the patients charts indicated it was a result of a partial folding of the donor graft, which then required for an extended period of time that the cell culture be suspended in air.
• Patient 6 had persistent post operative uveitis , which had its roots both surgically and from a graft rejection, since normally graft rejection only occurs 2 weeks after with evidence of segment inflammation.
• The report suggested that the cause was surgical since that was what distinguished him from other patients.

Study Complications

• The only difference that made them distinguishable from the other patient in study, was the use of 6 sheets Glide to help with limbal incision.

DSEK Conclusion

• DSEK can therefore carefully selected to assist with canine patients Complications can include: fibrous uveitis, and graft dislocation With the goal to increase successful transplantation in Dogs DSEK requires: careful attention to patient history, surgical selection, and an in depth analysis donor tissue analysis.