To review 12 cases of postoperative detachment and spontaneous reattachment of Descemet stripping automated endothelial keratoplasty (DSAEK) lenticles.
Retrospective, observational case series.
This was a review of patients undergoing DSAEK at 7 institutions. Patients who had a significant detachment of their DSAEK lenticle during the postoperative period were identified and divided into 2 groups. Significant detachment was defined as either complete central interface fluid with bare peripheral attachment (group 1) or a free-floating lenticle in the anterior chamber (group 2). Patients who subsequently had a spontaneous reattachment of the lenticle were identified, with data regarding surgical technique and intraoperative and postoperative complications collected for analysis.
Our cohort consisted of 12 eyes of 12 patients who met the definition of significant postoperative detachment with subsequent spontaneous reattachment. Four patients had complete central detachment with peripheral attachment (group 1), whereas 8 patients had a free-floating lenticle (group 2). Ten of the 12 patients had a successful outcome as defined as an attached and clear DSAEK lenticle. In our study, reattachment was seen as early as 5 days and as late as 7 months after surgery, with reattachment in 9 of 12 patients by day 25.
Spontaneous reattachment of detached DSAEK lenticles may occur during the postoperative period. The decision of when to bring the patient back for a rebubble ultimately must be made on a case-by-case basis.
Descemet stripping automated endothelial keratoplasty (DSAEK) is an alternative to penetrating keratoplasty (PK) for the treatment of endothelial dysfunction in cases of Fuchs dystrophy, pseudophakic and aphakic bullous keratopathy, and failed PKs. In DSAEK, the posterior stroma and endothelium are transplanted selectively, thereby decreasing the risk of potential complications of PK, including wound dehiscence, intraoperative suprachoroidal hemorrhage, postoperative ametropia, and high astigmatism. In addition, DSAEK is a relatively refractive-neutral procedure and provides faster visual rehabilitation than PK.
A number of studies have shown graft detachment to be the most common complication of DSAEK, ranging from 1% to 82%, depending on surgeon experience and surgical technique used. Graft detachment can be handled with repositioning of the graft and injection of an air bubble (rebubbling) or, alternatively, with observation. Koenig and Covert reported a single case of spontaneous reattachment of a partially dislocated graft among their 9 patients with a dislocated graft, and Suh and associates reported a singe case of spontaneous reattachment among their 27 patients with a dislocated graft. Price and Price also reported a single case of spontaneous reattachment.
We report the largest series of patients, to our knowledge, with spontaneous reattachment of a DSAEK graft after complete postoperative detachment, with 12 patients from 7 institutions. Eleven of these cases are previously unreported, and we provide additional information regarding the patient previously reported by Price and Price.
This was a retrospective review of all patients undergoing DSAEK at 7 institutions. Patients who had a significant detachment of their DSAEK lenticle during the postoperative period followed by spontaneous reattachment were identified. Significant detachment was defined as either complete central interface fluid with bare peripheral attachment or a free-floating lenticle in the anterior chamber. These patients then were subdivided into group 1 (central detachment) or group 2 (free-floating lenticle) for data analysis.
Our cohort consisted of 12 eyes of 12 patients who met the definition of significant postoperative detachment with subsequent spontaneous reattachment (see Tables 1 through 4 ). Patients 1 through 4 had complete central detachment with peripheral attachment (group 1), whereas Patients 5 through 12 had a free-floating lenticle (group 2). Indications for surgery for patients in group 1 included Fuchs endothelial dystrophy (n = 1), failed PK (n = 2), and pseudophakic bullous keratopathy (PBK; n = 1). Indications in group 2 included Fuchs endothelial dystrophy (n = 3), failed PK (n = 2), PBK (n = 1), Fuchs and PBK (n = 1), and corneal edema (n = 1). The overall rate of detachment at our institutions ranged from 1.8% to 7.6%.
|Patient 1||Patient 2||Patient 3||Patient 4|
|Indication for surgery||Fuchs||PBK||Failed PK||Failed PK|
|No. of medications||N/A||4||4||1|
|Prior glaucoma surgery?||N/A||No||No||No|
|Size of lenticle (mm)||8.0||8.0||8.0||8.0|
|Wound size (mm)||5.0||5.0||5.0||5.0|
|Intraoperative complications||None||Hypotony POD 1||IOP 43 POD 1||None|
|Type of dislocation||Central only||Central only||Central only||Central only|
|Day reattached||25||13||5||3 mos|
|Techniques used to reattach||Face-up extra 24 h (48 h total)||None||None||None|
|Air bubble left on day off?||Yes||Yes||Yes||Yes|
|Outcome: lenticle attached and cleared?||Yes||Yes||Yes||No|
|VA at 6 mos||20/25||20/100||20/200|
|If fewer than 6 mos, VA (time)||20/200 (2 mos)|
|Patient 5||Patient 6||Patient 7||Patient 8||Patient 9||Patient 10||Patient 11||Patient 12|
|Indication for surgery||PBK||Fuchs||Fuchs||Fuchs, PBK||Failed PK||Corneal edema||Failed PK||Fuchs|
|No. of Medications||3||N/A||N/A||N/A||N/A||3||Not available||N/A|
|Prior glaucoma surgery?||No||N/A||N/A||N/A||N/A||No||Yes||N/A|
|Lens status||Pseudophakic||Phakic||Pseudophakic||Pseudophakic||Pseudophakic||Aphakic||Sutured PCIOL||Pseudophakic|
|Vent incisions? (no.)||Yes (4)||Yes (4)||No||Yes (2)||No||No||Yes (3)||No|
|Size of lenticle (mm)||8.75||9.0||8.0||8.25||8.75||8.0||9.0||9.0|
|Insertion technique||Lens glide over ACIOL||Forceps||Forceps||Forceps||Forceps||Forceps with suture pull||Forceps||Forceps|
|Wound size (mm)||5.5 (scleral tunnel)||5.0||5.0||4.5||5.2||5.2||4.5||5.0|
|Intraoperative complications||None||Donor edematous||None||None||None||Iris prolapse after surgery||None||None|
|Type of dislocation||Complete||Complete||Complete||Complete||Complete||Complete||Complete||Complete|
|Day reattached||5||7||5||7||20||8||7 mos||6 wks|
|Postoperative complications||None||None||Could not position||None||None||None||None||None|
|Techniques used to reattach||None||Face-down position||Face-down position||None||Significant time face down at home||None||Rebubble x2 in office||Rebubble x2 in office|
|Air bubble left on day off?||Yes||No||Yes||Yes||No||No||Yes—but in vitreal cavity||Yes|
|Outcome: lenticle attached and cleared?||Yes||Yes||Yes||Yes||Yes||Yes||No||Yes|
|VA at 6 mos||20/40||20/30 + 2||20/60||20/400||20/40||20/30||20/400 (best visual potential)||20/40|
|Group 1 (n = 4)||Group 2 (n = 8)|
|Age (yrs)||76 ± 5.1||70.25 ± 13.73|
|No. male||2 (50%)||3 (37.5%)|
|No. Female||2 (50%)||5 (62.5%)|
|Fuchs||1 (25%)||3 (37.5%)|
|PBK||1 (25%)||1 (12.5%)|
|Fuchs/PBK||0 (0%)||1 (12.5%)|
|Failed PK||2 (50%)||2 (25%)|
|Corneal edema||0 (0%)||1 (12.5%)|
|Prior PK||2 (50%)||2 (25%)|
|Glaucoma||3 (75%)||3 (37.5%)|
|POAG||2 (50%)||2 (25%)|
|PXF||1 (25%)||0 (0%)|
|Mixed mechanism||0 (0%)||1 (12.5%)|
|Prior glaucoma surgery||0 (0%)||1 (12.5%)|
|Pseudophakic||4 (100%)||6 (75%)|
|Phakic||0 (0%)||1 (12.5%)|
|Aphakic||0 (0%)||1 (12.5%)|
|Group 1 (n = 4)||Group 2 (n = 8)|
|Descemet stripped||2 (50%)||6 (75%)|
|Peripheral scraping||4 (100%)||4 (50%)|
|Vent incisions||0 (0%)||4 (50%)|
|Forceps||4 (100%)||6 (75%)|
|Forceps with suture pull||0 (0%)||1 (12.5%)|
|Lens glide over ACIOL||0 (0%)||1 (12.5%)|
|Hypotony POD 1||1 (25%)||0 (0%)|
|Elevated IOP POD 1||1 (25%)||0 (0%)|
|Donor cornea edematous||0 (0%)||1 (12.5%)|
|None||2 (50%)||7 (87.5%)|
Six patients had glaucoma. Patients 2, 3, 4, 5, and 10 were medically managed with intraocular pressure (IOP)-lowering drops, and Patient 11 had had a previous glaucoma tube shunt. Ten patients were pseudophakic, including all 4 patients in group 1. Patient 10 was aphakic, and Patient 6 was phakic. Patient 6 underwent a triple procedure with DSAEK combined with phacoemulsification and implantation of a posterior chamber intraocular lens.
In terms of surgical procedure, 8 patients underwent stripping of the Descemet membrane. Those who did not undergo stripping of Descemet membrane were undergoing DSAEK for a failed PK. Vent incisions were performed in Patients 5, 6, 8, and 11, all of whom were in group 2. The size of the lenticle ranged from 8 to 9 mm, and the wound size ranged from 4.5 to 5.5 mm, using both clear corneal and scleral tunnel approaches. Forceps insertion of the DSAEK lenticle was performed in 10 patients, including all 4 patients in group 1. Patient 10 had the lenticle inserted with forceps with suture pull through, and Patient 5 had the lenticle placed with a lens glide over an anterior chamber intraocular lens.
There were no intraoperative complications, but Patient 6 was noted during surgery to have an edematous donor cornea. Patient 10 had postoperative iris prolapse, which was managed surgically after the lenticle spontaneously reattached.
Patient 2 was hypotonus on day 1 (see Figure 1 ). This patient was an 83-year-old man with exfoliative glaucoma managed medically who underwent DSAEK for PBK. He had uneventful surgery, but sought treatment after surgery with a centrally detached lenticle and hypotony. He was Seidel negative on examination. On postoperative day 2, the eye remained soft and he was advised to stop all glaucoma medications. On day 12, his IOP was 19 mm Hg. The lenticle reattached on day 13.