Purpose
To report the technique and outcome of Descemet membrane endothelial keratoplasty (DMEK) in pediatric patients older than 6 years of age.
Design
Institutional interventional retrospective case series.
Methods
This study included 5 eyes of patients less than 15 years of age with endothelial dysfunction who underwent DMEK. Three eyes had Descemet stripping done of the same size as the donor graft. Two eyes underwent non–Descemet stripping endothelial keratoplasty. Attachment of DMEK scroll and improvement in corneal clarity, vision, pachymetry, and intraoperative or postoperative complication was noted. We defined primary graft failure as nonclearing corneal edema despite a well-attached lenticule on anterior segment optical coherence tomography.
Results
A total of 5 eyes of 5 children (all male) with a mean (± standard deviation) age of 9.2 ± 3.42 years underwent DMEK. The mean preoperative visual acuity of 1.93 ± 0.25 logMAR units improved postoperatively to 0.98 ± 0.29 (95% confidence interval, P = .03). Anatomic success (well-attached scroll with a more transparent cornea with a decrease in pachymetry) was seen in 4 of 5 eyes (80%). One eye had a primary donor failure. Two out of the 5 eyes (40%) required rebubbling. The mean pachymetry in eyes with successful procedure reduced from 1094.5 ± 101.5 µm to 619.25 ± 150.3 µm ( P = .03).
Conclusions
Although DMEK in pediatric eyes is challenging, the early results are encouraging, and it is a viable option in pediatric patients with endothelial failure.
T illett’s first attempt at posterior lamellar keratoplasty in the early 1950s met with moderate success. Later in the early 2000s, there was renewed interest in endothelial keratoplasty (EK). Older techniques of posterior lamellar keratoplasty (PLK) such as large-incision PLK, microkeratome-assisted PLK, small-incision PLK, and deep lamellar EK have now given way to the present EK techniques of Descemet stripping endothelial keratoplasty / Descemet stripping automated endothelial keratoplasty (DSAEK), ultra-thin/nano-thin DSAEK, pre-Descemet endothelial keratoplasty, and Descemet membrane endothelial keratoplasty (DMEK). Melles reported the first case of DMEK in 2006. DMEK is now widely considered the procedure of choice for endothelial dysfunction. Better anatomic integration, faster visual rehabilitation, smaller incision size, and lower risk of rejection make DMEK a preferred choice over PK or even DSAEK. , These are of great significance, especially in a pediatric age group. DMEK is widely performed in adults with Fuchs endothelial corneal dystrophy and pseudophakic corneal decompensation. There are, however, several challenges in performing endothelial transplant in pediatric eyes. These include the relatively small anterior segment, decreased scleral rigidity with an increased propensity for anterior chamber collapse, and a clear crystalline lens exposed to the risk of surgically induced cataract. In cases like congenital hereditary endothelial dysfunction (CHED), the diffuse corneal haze hampers intraoperative visualization, and the dark iris of Asian eyes often makes the surgery technically more challenging. The likelihood of an increased inflammatory response in pediatric eyes, the difficulty in enforcing a supine posture, and preventing eye rubbing are additional challenges in the postoperative period. We herein report our initial experience of the surgical procedure and outcomes of DMEK in pediatric eyes.
Methods
This study was deemed exempt by the Vision Research Foundation, and all data collection was in conformity with federal or state laws and was in adherence to the tenets of the Declaration of Helsinki after ethical committee clearance. Detailed file review of all pediatric patients who underwent DMEK by a single surgeon (B.S.) between March 2017 and January 2020 at a tertiary care center in the south of India was done. This retrospective interventional case series included DMEK performed in 5 eyes of 5 patients (4 with CHED and 1 with a failed penetrating keratoplasty). Preoperatively all patients underwent a detailed ophthalmic examination including visual acuity testing, slit-lamp evaluation, intraocular pressure measurement, pachymetry / optical coherence tomography (OCT), and retinal evaluation. Visual acuity was measured using a Snellen chart, Lea symbols, or ability to fixate and follow the light. The measured visual acuity was converted to logMAR units for statistical analysis. Posterior segment evaluation by ultrasound was normal in all patients.
Surgical Technique
Surgeries were performed under general anesthesia. Corneal epithelial debridement helped improve the intraoperative visibility. Descemet membrane was stained with trypan blue (0.06%) and Descemet stripping under cohesive viscoelastic was completed in 3 out of the 5 eyes. An external illumination using a light pipe placed at the limbus helped improve visibility for the descemetorrhexis. Non–Descemet stripping EK was performed in 2 eyes, 1 with a failed graft and 1 where the Descemet stripping could not be initiated. Inferior limbal paracentesis was performed, and iris tissue was grasped with a 23 gauge forceps and externalized to perform an inferior peripheral iridectomy in all cases. Pupillary constriction was achieved using intracameral pilocarpine (0.2%). A 2.8 mm clear corneal incision was made to accommodate the Geuder injector. Donor cornea was prepared as per standard protocol, and an “F” stamp was marked on the Descemet side to facilitate correct orientation of the scroll. The scroll orientation in the Geuder injector was reconfirmed before injecting the scroll into the anterior chamber. The surgical wound was secured with a single 10-0 nylon suture. The orientation of the scroll was confirmed using the tangential illumination of the light pipe. Using a combination of tapping movements on the corneal surface and fluid jets, the tissue was unfolded and centered. A total of 0.3 cc of 20% SF6 gas and air was used to float the tissue up against the recipient cornea and achieve a complete fill. Subconjunctival antibiotics and steroids were injected at completion and the eye patched. The parents were asked to maintain the child in a supine position as much as possible. Postoperatively patients were started on 1% Prednisolone acetate (Pred Forte® ; Allergan, Inc., Irvine, CA, USA) 6 times a day, tapered every 2 weeks to reach a maintenance dose of 1 time a day that was continued indefinitely. Topical antibiotic drops (0.5% moxifloxacin, Vigamox®; Novartis, Alcon Laboratories, Inc., Fort Worth, TX, USA) were used 4 times a day for a week.
Results
All 5 patients in our series were male, with a mean (± standard deviation) age of 9.2 ± 3.42 years (range 6-15 years) and a mean follow-up of 13.6 ± 6.7 months (range 6.7-23.3 months). Anatomic success, defined as well-attached DMEK lenticule with a clearing of the cornea, was seen in 4 of 5 eyes (80%) ( Figure 1 ). Two eyes had a detachment of the Descemet membrane scroll in the early postoperative period that settled with air injection. However, 1 eye did not clear at 2 months postoperative despite a well-attached Descemet membrane scroll on OCT, suggestive of primary donor failure.
The mean preoperative logMAR visual acuity was 1.93 ± 0.25, and it improved to a mean postoperative visual acuity of 0.98 ± 0.29 logMAR units ( P = .03). Refraction was possible in 4 eyes, and the mean spherical equivalent was 0.4 diopters (D) ± 8.7 D (range, -7 D to +10 D) and mean astigmatism -2.75 D ± 0.4 D (range, -3.25 D to -2.5 D) ( Table ).
| Age (y) | Sex (M/F) | Laterality (OD/OS) | Primary Indication | Preoperative Pachymetry | Preoperative Vision | Surgery | Specular Count of Donor | Age of Donor Tissue | Outcome | Postoperative Vision | Postoperative Pachymetry | Postoperative Refraction |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 6 | M | OD | CHED | 1,111 | 6/60 | nDMEK and rebubble | 2,832 | 55 | Primary graft failure | Not recorded | 1,187 | Not recorded |
| 8 | M | OS | CHED | 1,027 | Counting fingers | DMEK | 3,322 | 50 | Clear graft | 6/60 | 652 | -1.75/-2.5 × 150 |
| 8 | M | OD | Failed PK graft | 1,240 | Hand movements | nDMEK and rebubble | 2,777 | 52 | Clear graft | 3/60 | 490 | -7/-3.25 × 140 |
| 9 | M | OD | CHED | 1,088 | Counting fingers | DMEK | 2,631 | 61 | Clear graft | 6/24 | 517 | +10/-2.5 × 140 |
| 15 | M | OD | CHED | 1,023 | 2/60 | DMEK | 2611 | 60 | Clear graft | 6/60 | 818 | -2.00/ -1.50 × 10 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
