Fig. 8.1
Optical coherence tomography images of donor tissue obtained (a) before debulking cut, (b) after debulking cut, and (c) after refinement cut
8.4.2 Surgical Technique
Patients are sedated with intravenous droperidol, 3 mL (7.5 mg), immediately before the peribulbar injection of local anesthetic, which consists of a mixture of 10 mL of L-bupivacaine, 0.75 %, and 100 IU of hyaluronidase. The entire procedure is performed with the surgeon sitting at the 12 o’clock position.
A marker 9 mm in diameter can be used at the beginning of the procedure to outline the limits of the internal surface from which the endothelium is peeled off.
Then the anterior chamber is entered at the 12 o’clock position with a 25-gauge needle mounted on a 2.5-mL empty syringe.
Aqueous (approximately 0.2–0.4 mL) is aspirated and air is injected, filling the anterior chamber.
A 25-gauge blunt cannula mounted on a syringe is introduced and used to cut through the endothelium and Descemet membrane, following the contour of the superficial mark. The cannula is therefore used to sweep away the endothelium and Descemet membrane, usually in a single piece. Whenever air is lost, the anterior chamber is reformed by injecting additional air with the syringe. Performing the entire maneuver under air enables perfect visualization of the Descemet membrane (no dye is ever used) and eliminates the need for any viscoelastic substance in the anterior chamber.
A clear cornea tunnel, 1 mm long and 3.2 mm wide, is prepared nasally.
At this point, an inferior peripheral iridectomy is performed, if not already present.
An anterior chamber maintainer is placed at the 12 o’clock position to enable continuous irrigation while performing the next surgical steps.
After punching the donor tissue to the desired size, the hollow of the punching block is filled with balanced salt solution (BSS) and the tissue left to float on the liquid cushion. The donor tissue is not folded and therefore inserted into the anterior chamber with the so-called taco technique. A corneal forceps is used to grasp the edge of the graft and drag it onto the plate of the side platform of a modified Busin glide (mini-glide or Mini Busin spatula) used to scoop the tissue, keeping the endothelial side up. As the funnel of the modified glide is smaller than the conventional glide, the tip can be inserted into the wound during delivery, thus preventing squeezing of the tissue while entering the anterior chamber through the corneal incision.
The tissue is pulled into the funnel-shaped part of the Busin glide using a microincision forceps until it engages the mini-glide opening.
A side entry is created temporally.
The mini-glide is then inverted and positioned at the entrance of the nasal clear cornea tunnel. The same microincision forceps (held in the left hand when operating on the left eye and in the right hand when operating on the right eye) is inserted through the side entry and passed across the anterior chamber, exiting throughout the clear cornea tunnel to grab the graft from the mini-glide and drag it into the anterior chamber. No additional viscoelastic substance is required during insertion. The donor lamella is allowed to unfold spontaneously under continuous irrigation from the anterior chamber maintainer. When necessary, centration of the graft is achieved by gentle tapping on the corneal surface.
Figure 8.2 shows the different steps of the surgical procedure.
Fig. 8.2
Photographs showing the steps of the surgical technique from Descemet stripping, debulking cut, refining cut, tissue marking, graft scooping, and implantation using the pull-through technique
Both the clear cornea tunnel and the side entry are sutured watertight with interrupted 10-0 nylon sutures. The graft is attached to the posterior corneal surface by filling the anterior chamber with air injected through the temporal side entry.
Triamcinolone acetonide and gentamicin sulfate, 0.3 %, are injected subconjunctivally at the end of the procedure.
8.5 Results
In patients undergoing a triple procedure (UT-DSAEK and phacoemulsification with IOL implantation), phacoemulsification and IOL implantation are performed prior to UT-DSAEK surgery. No viscoelastics have to be used throughout the procedure; capsulorrhexis is performed using a bent needle mounted on a syringe filled with saline to maintain a closed system, while the IOL is implanted under continuous irrigation from an anterior chamber maintainer placed at the 12 o’clock position. The anterior chamber maintainer has to be inserted through a side port constructed with a 15°-disposable blade introduced in the anterior chamber under continuous irrigation delivered by the irrigation-aspiration system throughout the main port. Intracameral acetylcholine chloride is used to constrict the pupil prior to UT-DSAEK surgery.
8.6 Postoperative Care
After surgery, patients are instructed to lie supine for at least 2 h. Once this period is passed, the patients are examined about 3 h after surgery at the slit lamp to check if the graft is fully attached and in position. Some air is removed from one of the side ports using a blunt cannula if no aqueous has entered the anterior chamber from behind the iris through the inferior peripheral iridectomy.
When the graft detaches from the recipient stroma and a false chamber is seen, the graft has to be repositioned, preferably in the operating room, injecting air into the anterior chamber similarly to what is done during the primary surgery. The supine position is requested again for about 2 h.
In dubious cases, particularly when the space between the cornea and donor tissue is minimal, anterior segment optical coherence tomography (ASOCT) can be used to help diagnosing the presence of a double chamber.
Postoperatively, all patients are given topical tobramycin, 0.3 %, and dexamethasone, 0.1 %, suspension (TobraDex; Alcon, Fort Worth, Texas) combination therapy every 2 h for 2 weeks, then every 3 h for 2 additional weeks. The treatment is switched to pure steroidal eyedrops (dexamethasone 0.1 %) qid for 1 month, tid for 1 month, bid for 1 month, then finally qd to be continued indefinitely unless the patient is aphakic or steroid responder.
All sutures are removed in all cases between 4 and 6 weeks from surgery.
8.7 Complications
Primary complications associated with UT-DSAEK usually occur during the preparation of the donor tissue. Graft dislocation and primary graft failure follow in the list as more frequent complications.
Microkeratome-related complications were recorded in 7.2 % of all the dissections performed in our original work [8]. All complications occurred during the second refinement pass. Buttonholing with incomplete central dissection of the donor tissue can be handled by means of hand refinement and the tissue still used. Perforation occurred, with tissue being discarded or used anyway depending on the location of the perforation; punching the tissue eccentrically was compatible with rather peripheral sites of perforation. Improper dissection outside of the 9-mm zone, i.e., at the very end of the cut, was compatible with the normal use of the tissue. Complications during the refinement cut were more common with the use of a 50-μm microkeratome head and to a lesser extent with a 90-μm microkeratome head.
Postoperatively, graft detachment and dislocation occurred in 4.6 % of all cases. Total detachment was seen in most cases and generally successfully managed by re-bubbling (with a single or double injection). When the detachment is limited to the periphery of the graft, there is usually no interference with the vision and therefore it is not treated. Postoperative graft dislocation requiring air reinjection occurred much less often than reported after DMEK by Price et al. [6] and Guerra et al. [9] (60 %) and Laaser et al. (92 %) [10].
Primary endothelial graft failure was recorded in 2 % of the cases at 12 months follow-up and was generally successfully treated with systemic steroid treatment (1 g per kg of body weight), sub-Tenon steroid injections (Depo-Medrol every 2 weeks), and topical steroid drops. The success of the therapy depends on the delay of patient presentation to the ophthalmic care, being higher in patients presenting with initial and mild symptoms like mild loss of vision, glare, haloes, and redness.
Additional complications of the procedure include pupillary block, which is immediately treated by dilation and/or air removal at the slit lamp, persistent epithelial defect, cystoid macular edema, persistent interface haze, interface infections, and cataract formation.
Once established, Urrets-Zavalia syndrome requires surgical treatment (pupilloplasty).
Persistent epithelial defects are treated with bandage soft contact lenses and antibiotic ointment up to reepithelialization.
Cystoid macular edema is treated with systemic dorzolamide and nonsteroid anti- inflammatory medications. Intravitreal injections are seldom required.
Interface bacterial or fungal infections, which are resistant to conservative multispectrum antibiotic treatment, are surgically faced by “en bloc” removal of the recipient cornea and DSAEK graft and replaced by full-thickness keratoplasty, thus eliminating the interface infections “tout court.”
The incidence of early cataract formation in phakic eyes could be significantly reduced by modifying the incision site from 3 and 9 o’clock position to 2 and 10 o’clock position delivering the graft through a clear cornea tunnel with Busin glide and pulling it with proper forceps more superiorly than the level of the pupil aperture.
8.8 UT-DSAEK Results
The prospective data from a large clinical series of 285 ultrathin DSAEK eyes by M. Busin has documented the benefits of the procedure at 2 years after surgery. This procedure offers the potential to achieve the visual results of DMEK with the ease of handling and tissue preparation of DSAEK.
The basic benefit of ultrathin DSAEK is a faster visual recovery than conventional DSAEK and an equivalent visual recovery than DMEK. The proportion of eyes achieving final acuity of 20/20 is higher than either conventional DSAEK or DMEK.
In addition, comparing UT-DSAEK to DSAEK and DMEK, all three offer an acceptable rate of cell loss.
8.8.1 Visual Outcomes
As early as 1 month postoperatively with all sutures still in place, BSCVA ≥20/20 was recorded in 11.7 % and BSCVA ≥20/40 in 63.8 %, of all eyes. BSCVA kept on improving over time, with a percentage of patients reaching 20/20 or better of 26.4 % at 6 months, 39.5 % at 1 year, and 48.8 % at 2 years after DSAEK. In particular, phakic patients younger than age 50 performed best, with 76.5 % of them reaching a BSCVA ≥20/20 as early as 6 months after UT-DSAEK.
The steep curve obtained for the UT-DSAEK values shows a continuous improvement in average logMAR BSCVA from 0.76 ± 0.49 preoperatively, to 0.35 ± 0.40 at 1 month, 0.16 ± 0.13 at 3 months, 0.12 ± 0.12 at 6 months, 0.08 ± 0.12 at 1 year, and 0.03 ± 0.09 at 2 years after UT-DSAEK, respectively.
Quite strikingly, the mean postoperative logMAR BSCVA curves of UT-DSAEK and DMEK almost overlap throughout the entire follow-up period considered, whereas the one of conventional DSAEK remains at a lower level [9, 11]. The speed of visual recovery after UT-DSAEK is somewhat slower than that recorded after DMEK, but as early as 1 year postoperatively, the percentage of eyes with BSCVA of 20/20 or better are substantially identical [9]. Based on this report, in terms of 20/20 visual recovery, UT-DSAEK performs much better than conventional DSAEK as early as 1 year in the postoperative course, with about three times as many patients achieving 20/20 or better vision.
The visual outcome recorded after UT-DSAEK indicates that the use of thin grafts allows a higher number of eyes to achieve a better visual performance in a shorter period of time than after DSAEK. However, not differently than after DMEK, the percentage of eyes without comorbidities recovering 20/20 vision after UT-DSAEK is lower than expected [6, 9, 12, 13]. This indicates that probably other factors besides the presence of a stromal interface of any type can determine the final visual result of EK. Patel et al. [14] demonstrated a small peak in corneal backscatter originating from the posterior stroma of the recipient cornea, which could be related to long-standing edema and consequent anatomical and functional changes in this part of the cornea. Recovery of normal stromal architecture might very often require a prolonged period of time, during which vision would be negatively affected. If this were true, the excellent results obtained in the eyes of younger patients undergoing phakic UT-DSAEK would be a simple consequence of a “better” recipient cornea.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
