Purpose
To evaluate the clinical results of Descemet membrane endothelial keratoplasty (DMEK) without any intraoperative or postoperative posturing.
Design
Retrospective, noncomparative, and interventional case series.
Methods
One hundred thirty-four (134) consecutive eyes of 101 patients requiring DMEK for any cause of endothelial failure were included. Descemet membrane endothelial keratoplasty was performed with an intraoperative inferior PI. After unscrolling and centration of the graft, the anterior chamber was near completely filled with 20% SF6. After filling the eye with gas there was no intraoperative tamponade time, and patients did not posture postoperatively. Main outcome measures were: graft detachment rate, rebubbling rate, primary and secondary failure rates, rejection rate, intraoperative and postoperative complications, and visual acuity 3 and 12 months after surgery.
Results
Five of 134 (3.7%) developed graft detachment greater than one-third of the graft area, and 19 of 134 (14.2%) developed graft detachment less than one-third of the graft area. A rebubble procedure was performed in 19 of 134 (14.2%); of those, the average number of rebubbling procedures performed was 1.13 (range, 1-2). Primary failure occurred in 6 of 134 (4.5%).Twelve-month graft survival occurred in 122 of 134 eyes (91.0%), with secondary failure within the first 12 months accounting for failure of 6 of 134 (4.5%). Nine of 134 (6.7%) cases had a recorded episode of endothelial rejection. The average time for rejection episode was 9.5 months (range, 1-20 months) after surgery.
Conclusions
A total of 134 ‘posture-less’ DMEKs did not suffer very high rates of graft detachment, rebubbling, or other postoperative complications, even with immediate erect posturing after surgery.
INTRODUCTION
D escemet’s membrane endothelial keratoplasty (DMEK) has become a standard treatment option, with excellent visual outcomes, for patients requiring a corneal endothelial graft. Despite the successful outcomes of this procedure, many variations in the postoperative management of DMEK have been described, including: the choice of gas (room air versus SF6), thresholds for rebubbling, intraoperative tamponade time, and postoperative posturing regimen. During the immediate postoperative phase, the presence of a gas bubble and its surface tension prevent significant graft detachment; hence, graft detachment occurs only after resorption of the bubble, with the average time to rebubbling reported between 7-92 days (median 19 days) postoperatively. There is no consensus on the optimum intraoperative or postoperative posturing regimes. Santander-Garcia and associates found no difference in graft detachment rates between various intraoperative tamponade times, and the variation in postoperative posturing times has been described from 90 minutes up to 7 days for SF6, with no evidence to support different detachment rates with differing posturing durations.
It is believed that there are no reports of DMEK performed without the need for any intraoperative tamponade time or postoperative supine posturing. This paper aimed to present the results of DMEK surgery in which there was no intraoperative or postoperative posturing required.
METHODS
STUDY DESIGN
This retrospective case series was approved as a Clinical Audit report by the Clinical Audit Committee at Moorfields Eye Hospital (London, UK) and was performed in accordance with the tenets of the Declaration of Helsinki. The medical records of all DMEKs for all indications performed between October 2016 and September 2021, with a minimum follow-up of 3 months, were evaluated. All operations were performed at Moorfields Eye Hospital by a senior surgeon (MW) or by a fellow supervised by MW. All data were prospectively collected and entered into the patient’s electronic medical records from the Moorfields Electronic Healthcare Record System (OpenEyes, Apperta Foundation CIC). Patients were excluded who were postured for any amount of time on the day of the surgery or in the first 7 days of the postoperative period. If a patient did not posture after the DMEK and was subsequently rebubbled and asked to posture after the rebubble, he/she was included, as the initial DMEK was posture-less.
SURGICAL TECHNIQUE
Donor preparation is performed using the SCUBA technique within the receptacle of a coronet donor punch (Network Medical Products Ltd). After staining the endothelium with 0.06% trypan blue (Vision-blue, DORC) the Descemet membrane (DM) is scored with a reverse Sinskey hook. The edge of the DM is grasped with plain microforceps and peeled from one edge until the leading edge of the graft peel is 2-3 mm beyond the midline of the cornea, leaving a 2-3 mm hinge of attachment in one quadrant. The stromal bed is then punched with a 3-mm skin trephine (Stiefel Laboratories UK Ltd) in the center and the 3 mm plug left in situ. The DMEK is then left in a well of balanced saline solution until the recipient eye is ready for DMEK insertion. The DM aspect of the graft is then marked with an inked S-marker through the 3-mm trephine. The donor cornea is replaced in the Coronet receptacle and the DM is punched with an 8-mm punch. The graft is then stained with trypan blue 0.15%, BBG 0.025%, and 4% PEG (Membrane Blue-Dual) (DORC) for 1 minute. The graft is then loaded into a 2.4-mm Geuder injector (Geuder AG).
In combined cases, descemetorrhexis is performed after intraocular lens (IOL) implantation under cohesive viscoelastic (Provisc, Alcon), whereas descemetorrhexis is performed under filtered room air in solitary DMEK.
Prior to DMEK insertion the pupil is miosed with acetylcholine 1% (Miochol, Novartis). An inferior peripheral iridotomy (PI) is created with a vitrector.
Following graft insertion with a 2.4-mm Geuder injector (Geuder AG), the graft is positioned and orientation confirmed before the anterior chamber (AC) is filled to 100% with a 20% concentration of SF6, aiming for an IOP within the physiological range as assessed using the tip of the cannula. Corneal suturing is not routinely performed unless the anterior chamber is unstable at the end of surgery.
No intraoperative posture time occurs. Immediately following injection of the gas bubble, cefuroxime 1 mg/0.1 mL (Aprokam, Thea Pharmaceuticals) is injected into the anterior chamber by one of the existing paracenteses and a slight meniscus is visualized just inside the limbus (estimated AC fill >90%). This concludes the operation, and after placing a patch and shield on the eye the patient is sat up and taken to the ward where he/she remains erect until postoperative review 1-2 hours later. At this review the IOP is measured with Goldmann tonometry and the inferior meniscus is confirmed to be above the inferior PI, otherwise some gas is released from the AC via an existing paracentesis until the PI is uncovered. Following discharge no instructions are given to the patient to posture. The patient is reviewed at days 1 and 8 following the surgery and then at the discretion of the clinical team, but usually at months 1, 3, 6 and 12.
POSTOPERATIVE MANAGEMENT
Postoperatively, all patients were given topical preservative free (PF) moxifloxacin 0.5% or chloramphenicol 0.5% four times a day for 7 days. The default postoperative steroid tapering regimen for uncomplicated cases was dexamethasone sodium phosphate 0.1% four times a day for 1 month, three times a day for 1 month, twice a day for 1 month, then once a day until the 1 year review.
DMEK DETACHMENT DIAGNOSIS AND MANAGEMENT
The protocol for postoperative examinations at days 1 and 8 includes visual acuity testing with a Snellen chart at 6 m, IOP measurement with Icare (Icare tonometer TA01i; Icare Finland Oy), anterior segment optical coherence tomography (AS-OCT) with MS-39 (CSO), and slit lamp biomicroscopy examination.
Rebubbling is always performed if there is a graft detachment greater than one-third of the graft area. Partial detachments in cases with less than one-third of the detached graft could be rebubbled at the discretion of the treating surgeon or observed. If observed, AS-OCT is performed on a weekly basis and observation continues as long as the pachymetry map, visual acuity, and area of detachment are consistently improving ( Figure 1 ).
RESULTS
A total of 134 eyes from 101 patients who underwent DMEK during the study period were included in the analysis. Mean age at the time of surgery was 71.7±11.1 years (range, 32-95 years). Sixty-nine of 134 (51.5%) surgeries were performed by a consultant surgeon (MW), with the remainder (65 of 134, 48.5%) performed by a senior fellow under the supervision of MW. Indications for surgery were: pseudophakic bullous keratopathy (n=22, 16.4%), aphakic bullous keratopathy (n = 1, 0.75%), Fuchs endothelial dystrophy (n = 89, 66.4%), herpetic endothelial failure (n = 1, 0.75%), failed penetrating keratoplasty (n =4, 3.0%), failed Descemet stripping automated endothelial keratoplasty (DSAEK) (n = 8, 6.0%), and failed DMEK (n = 9, 6.7%). DMEK surgery was combined with cataract surgery in 61 of 134 cases (45.5%), revision of glaucoma tube in 2 of 134 (1.5%), rotation of implantable collamer lens in 1 of 134 (0.75%), synechiolysis in 1 of 134 (0.75%), surgical capsulotomy in 1 of 134 (0.75%), implantation of secondary IOL in 1 of 134 (0.75%), and iridoplasty in 1 of 134 (0.75%). Anatomical comorbidities included corneal scarring in 12 of 134, a glaucoma drainage device (GDD) in 8 of 134, angle closure in 6 of 134, keratoconus in 3 of 134, and uveitis in 5 of 134. Visual comorbidities included corneal scarring in 12 of 134, glaucoma in 20 of 134, age-related macular degeneration in 4 of 134, amblyopia in 7 of 134, keratoconus in 3 of 134, uveitis in 5 of 134, cystoid macular edema in 5 of 134, retinal vascular disease in 5 of 134, epiretinal membrane/macular hole in 10 of 134, and other macular pathology in 2 of 134 cases. Average follow-up duration was 19.6±12.8 months (range, 3-49)
GRAFT DETACHMENT
Five of 134 (3.7%) eyes developed graft detachment greater than one-third of the graft area, 19 of 134 (14.2%) developed graft detachment less than one-third of the graft area, and 110 of 134 (82.1%) were fully attached at all postoperative reviews. A rebubble procedure was performed in 19 of 134 (14.2%) and in 11 of 89 (12.4%) of those with FED. One of 4 (25%) of those with previous PK and 2 of 8 (25%) of those with a GDD required rebubbling. Details of these cases are found in Table 1 . For those requiring rebubbling, the quadrant that best described the detachment was inferior in 42.1%, superior in 15.8%, nasal in 15.8%, temporal in 15.8%, and mixed in 10.5%. Of those requiring rebubbling the average number of rebubbling procedures performed was 1.13 (range, 1-2). Sixteen of 19 rebubbling procedures were performed on the slit lamp with filtered room air, 3 of 19 were performed in the theatre and rebubbled with SF6; one of these required further unscrolling and thereafter had an uneventful course, and another required repositioning of an inverted graft, this case suffered primary failure and subsequently underwent an uneventful repeat DMEK. The most common time to perform rebubbling was at the day 8 review (range, day 1-3 months).
| Indication for surgery Demographic PBK FED Failed PK Failed DMEK | n = 5/22 (22.7%) n = 11/89 (12.4%) n = 1/4 (25%) n = 2/9 (22%) |
| Age at time of surgery, years | 70±14 (range, 32-88) |
| Gender Male Female | 9/19 10/19 |
| Laterality Right Left | 13/19 6/19 |
| Associated additional surgical procedures Cataract surgery Revision of glaucoma tube Rotation of ICL | n = 6 (32%) n = 1 (5%) n = 1 (5%) |
| Grade of surgeon Consultant Fellow | n = 10 (53%) n = 9 (47%) |
| Anatomical comorbidities None Corneal scarring Glaucoma drainage device Angle closure Keratoconus Anterior uveitis | n = 13 n = 1 n = 2 n = 1 n = 1 n = 1 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
