This article is more than 5 years old. Information may no longer be current.
Man experiences painless vision loss after combined DSAEK, cataract surgery
The postoperative course had been uneventful, and the patient had no other complaints.
Click Here to Manage Email Alerts
A 65-year-old Caucasian man was emergently referred to the retina department at the New England Eye Center for less than 24 hours of painless loss of vision in the right eye. He had undergone a combined Descemet’s stripping automated endothelial keratoplasty and cataract extraction in the right eye 6 days before presentation. His immediate postoperative period had been uncomplicated thus far.
The day of presentation, he returned to his surgeon’s office and was immediately referred to our retina department with concern for endophthalmitis. He had no other complaints. He had been seeing well the previous day and had been using his postoperative drops as directed. His ocular history was significant for Fuchs’ dystrophy in both eyes. His medical history was significant only for hyperlipidemia. He was currently using topical ketorolac, ofloxacin, prednisolone and brimonidine in the right eye. His oral medications included pravastatin and aspirin. He did not use tobacco or alcohol, and he did not have known drug allergies.
Examination
On exam, best corrected distance visual acuity was 20/200 in the right eye and 20/25 in the left eye. Both pupils were equally reactive to light with no afferent pupillary defect. Extraocular motility and confrontational visual fields were full in both eyes. IOP was 17 mm Hg in the right eye. The anterior segment exam of the right eye was significant for an attached DSAEK graft centrally but the nasal edge of the graft was curled downward. There were 1+ microcystic edema, 1+ stromal edema and 1+ Descemet’s folds present. There was also a possible fibrin deposit or infiltrate that was visible on the inferior edge of the graft. In the anterior chamber, 4+ cell was present along with a 1-mm hypopyon. The posterior segment exam revealed 2+ vitreous cell, but the fundus was otherwise unremarkable. The exam of the left eye was unremarkable.
What is your diagnosis?
See answer on next page.
Interface opacities
The differential diagnosis for a patient with interface opacities after DSAEK includes infection, graft rejection, epithelial ingrowth and interface deposits. Infection was highest on the differential for this particular patient given the painful presentation, the hypopyon and the time frame.
Graft rejection typically takes place at least 11 months after the initial surgery; this time frame did not fit the patient because he had surgery 6 days before the development of these symptoms. Epithelial ingrowth is also on the differential; however, it is a rare occurrence in DSAEK. If any ingrowth were to develop, it would typically take place within the first 3 months and should not have associated pain. To diagnose epithelial ingrowth, OCT of the cornea or argon laser photocoagulation to the ingrowth can be helpful. Interface deposits were lower on the differential because they are usually observed immediately after the surgery whereas this patient did not have his findings until 6 days after the surgery. The interface deposits may consist of blood, retained Descemet’s membrane, viscoelastic, or surgical debris from blades or lint. The amount of inflammation with interface deposits is variable, with some deposits remaining inert and causing no inflammation.
Diagnosis and management
The day of presentation, the patient underwent an anterior chamber paracentesis followed by an intravitreal injection of vancomycin, ceftazidime and voriconazole. He was continued on his postoperative drops and cyclopentolate was added. The next day, his vision had improved to 20/100, and the size of the fibrin deposits and hypopyon had decreased. IOP of the right eye was slightly elevated to 27 mm Hg. Dorzolamide-timolol was added to further lower the IOP. At this point, anterior chamber paracentesis showed only polymorphonuclear lymphocytes, but the donor corneal rim had stained positive for yeast. Three days after presentation, the anterior chamber tap culture started to grow yeast, and the patient was immediately started on oral voriconazole, topical natamycin and topical cyclosporine. The topical steroids were stopped at this point.
Images: Liu E, Kenyon KR
He continued to be followed closely, but 5 days later he came in with decreased vision and increased pain in the right eye; his vision had decreased to counting fingers and the hypopyon was present again. An anterior chamber paracentesis was repeated, followed by an intravitreal and intracameral injection of voriconazole. He was put on the schedule for return to the OR to remove the graft; however, 2 days later his vision had improved again to 20/100 and the hypopyon had again decreased in size. Given his good response to the voriconazole, he was given another injection at this visit and return to the OR was deferred. He continued to have a rocky course with improvements and setbacks, but the DSAEK graft was finally removed 15 days after his initial presentation, or 21 days after the initial surgery, with plans for a full-thickness penetrating keratoplasty in the future. His second anterior chamber tap did not grow yeast but did grow a gram-positive rod. During the graft explantation, he received another intravitreal injection of vancomycin, voriconazole and ceftazidime and intracameral injection of vancomycin and voriconazole.
Five days after the graft explantation, the patient was doing well (Figure 1). His vision was stable, and his eye felt more comfortable. He was given an intrastromal injection of voriconazole because of persistent posterior corneal infiltrate. Repeat graft was also delayed due to likely persistence of infection. The intrastromal voriconazole was repeated a few days later, and he continued to improve significantly. He was also started on prednisolone and Muro 128 (sodium chloride hypertonicity ophthalmic ointment, Bausch + Lomb) eye drops and continued on voriconazole orally and natamycin eye drops. Three months after the graft removal, BCVA had improved to 20/50 (Figure 2).
At 4 months, he began tapering off natamycin and prednisolone, and BCVA had improved dramatically to 20/30 (Figure 3). Because of this dramatic improvement, he did not undergo a PK.
Discussion
DSAEK was developed in 2005 and now accounts for about half of corneal transplants. There are many different methods used, but in general, the donor graft can be either precut or surgeon cut and typically arrives frozen in Optisol and warmed for 2 hours. The host cornea is marked, and the host endothelium is peeled. The donor tissue is inserted, and an air bubble is injected for approximately 10 minutes to appose the graft to host cornea. A peripheral iridotomy or venting incisions are at times used based on surgeon preference. The rim is cultured in 30% of cases. Sources of infection are suspected to be from harvesting to preservation, during thawing/pre-insertion, venting incisions and postoperatively.
The main advantages of DSAEK over traditional PK include faster surgical and recovery time and lower risk of intraoperative complications. Common complications are graft dislocation and failure. Postoperative fungal infections have been rarely reported, the first appearing in the literature in 2009. In the first case, the rim had grown Candida albicans, and the patient developed small infiltrates by postoperative day 7. The graft was removed on postoperative day 15, and she was continued on topical natamycin and oral voriconazole. She continued to worsen, eventually needing a PK. Her postop course was complicated by persistent epithelial defect, iris neovascularization, pupillary block, pain and eventual enucleation. The two other cases were from the same donor and both rims eventually grew Candida. Both patients developed infiltrates. The first patient underwent a repeat DSAEK, was started on oral fluconazole and topical amphotericin B, and ended up with a BCVA of 20/50. The second patient was given oral fluconazole and topical and intracameral amphotericin B and eventually became Seidel positive in the area of the infiltrate. After gluing and a patch graft, she ended up 20/40.
These cases would argue for increased routine culturing of donor rims and prophylactic treatment as soon as cultures returned, whether or not the patient was showing symptoms. However, in PK, only 14% of fungal culture positive rims resulted in eventual fungal infection. It is possible the delayed onset and refractory nature of DSAEK graft infections may be due to the unique sequestered environment, and as further cases are reported, graft preparation or response to rim cultures will need to differ between PK and DSAEK.
The largest published review of DSAEK postoperative infections with 1,088 eyes found 10 eyes with postoperative infections although cultured donor rims showed no growth of any organisms. This would suggest that negative growth on donor rims does not help predict infection risk. An additional review of 99 cases with literature review also noted correlation with positive rim cultures (seven cases) and eventual infection (two cases). Both patients had repeat DSAEK and eventually required PK despite aggressive therapy.
These studies highlight the need to evaluate further the value of rim cultures, to add antifungal to storage medium, and to evaluate whether the cost of attempted treatment is worth the uncertain result or if signs of infection should lead straight to graft removal or immediate PK.
- References:
- Kitzmann AS, et al. Cornea. 2009;doi:10.1097/ICO.0b013e31819140c4.
- Nahum Y, et al. Cornea. 2014;doi:10.1097/ICO.0000000000000205.
- Ortiz-Gomariz A, et al. Eur J Ophthalmol. 2011;doi:10.5301/EJO.2011.6228.
- Ritterband DC, et al. Cornea. 2007;doi:10.1097/ICO.0b013e31802d82e8.
- Tsui E, et al. Cornea. 2016;doi:10.1097/ICO.0000000000000778.
- For more information:
- Erica Liu, MD, and Kenneth R. Kenyon, MD, can be reached at New England Eye Center, Tufts University School of Medicine, 750 Washington St., Box 450, Boston, MA 02111; website: www.neec.com.
- Edited by Jessica Moon, MD, and Emily C. Wright, MD. They can be reached at the New England Eye Center, Tufts University School of Medicine, 750 Washington St., Box 450, Boston, MA 02111; website: www.neec.com.