Faculty Bibliography

The purpose of this study was to delineate the patient and surgical factors that contribute to the development of corneal neovascularization (CNV) after penetrating keratoplasty (PK). Thirty-six eyes of 36 patients with no antecedent CNV were enrolled in the study. Grafts were sutured to the host with 16 10-0 nylon sutures with the knots buried alternately in either the host or donor corneal stroma. Multiple perioperative factors were recorded for each patient, and at each postoperative visit systematic corneal drawings were used to follow the development of neovascularization. The stroma adjacent to each suture of each graft was given a neovascularization score based on the extent of vessel growth toward the wound interface. Univariate and multivariate analyses were performed, including generalized estimating equations logistic regression where each eye is considered a cluster of observations. Thirty-four patients without preoperative CNV or inflammation were followed prospectively for 6-9 (mean, 7) months after PK. Fourteen eyes (41%) developed some degree of CNV. Indication for keratoplasty, age, gender, phakic status, and size of donor button were not risk factors for CNV development. The most significant risk factor identified for any degree of CNV was placement of the suture knot in the host stroma (p = 0.00007), with the overall relative risk of CNV associated with these knots over 2 (95% confidence interval, 1.1-4.2). Furthermore, the mean recipient size in eyes with postoperative CNV was larger than eyes that did not develop neovascularization (p = 0.015), and active blepharitis was associated with a fivefold increase in the risk of developing CNV to the wound edge (p = 0.008). Embedding suture knots in the host stroma, active blepharitis, and a large recipient bed are significantly associated with postkeratoplasty CNV

A 3-year retrospective study on the risk factors of positive donor rim cultures in penetrating keratoplasty was performed. One thousand and ninety-seven consecutive donor rim cultures were reviewed from the period between June 1990 and October 1993 to determine the rate of culture positivity. The sex, age, diabetes status, use of respirator at time of death, cause of death, harvesting technique, storage time, and corneal storage medium utilized for the donors with positive donor rim culture were compared to those for 100 randomly selected culture negative donor controls. Logistic analysis was performed to eliminate confounding effects. Forty-six of the 1,097 (4.19%) donor rim cultures were positive. We found an association between the in situ technique for donor harvesting and culture negativity (p = 0.03). None of the other donor characteristics was associated with culture positivity. None of the 46 recipients who received the positive culture corneas developed endophthalmitis. In situ cornea harvesting promotes less contamination than enucleation and enriched gentamicin and streptomycin storage medium may further decrease donor rim culture positivity

We examined the effects of contact lens wear on corneal topography following penetrating keratoplasty (PK). We present the long-term follow-up of 23 eyes of 22 patients who underwent PK and rigid gas permeable contact lens fitting. Computerized corneal topography, manual keratometry, and refraction were conducted prior to contact lens fitting and then at 2 weeks, 6 weeks, 3 months, 6 months, and 1 year post-fitting. At least 6 months follow up was obtained for 23 eyes of 22 patients, and at least 12 months follow-up was obtained for 18 eyes of 17 patients. Seventy four percent of the eyes were stable (+/- 1.00 D) by manual keratometry, computerized corneal topography, and refraction (both spherical equivalent and cylinder). Those patients who changed showed a trend toward a small decrease in refractive cylinder and spherical equivalent. Measurements obtained from both computerized corneal topography and manual keratometry showed a small but significant amount of flattening (P < 0.003 and P < 0.04, respectively)

Because suture erosion after keratoplasty is an important risk factor for inflammation, infection, vascularization, and graft rejection, we aimed to delineate patient characteristics associated with these erosions. One hundred eyes of 97 consecutive patients who presented to our service with 10-0 nylon suture erosion after keratoplasty were selected for study. Patient age, preoperative diagnosis, duration from surgery, location and type of eroded suture, vascularity of recipient bed, contact lens or topical steroid use, and presence of infiltrate at the erosion site and subsequent culture results were tabulated. The average duration from keratoplasty to presentation was 33 (range 1-144) months. The locations of the eroded sutures were superior in 53%, nasal in 17%, temporal in 16%, and inferior in 14% of eyes (p < 0.005). Seventy-one percent of the eyes presented with broken sutures (an average 36 months postoperatively) and 29% with intact but loosened eroded sutures (an average 25 months postoperatively, p < 0.05). Sixty-nine percent of the eyes were being treated with topical steroids and presented 11 months earlier (29 months postoperatively) than did those not being treated with topical steroids (40 months postoperatively, p < 0.05). Eyes that had been subjected to keratoplasty for inflammatory conditions presented with suture erosion 10 months earlier than did those that had been subjected to keratoplasty for noninflammatory conditions (p = 0.09). Of the 10 eyes with a suture-related stromal infiltrate, one was culture-positive. Increased elapsed time from surgery, superior position of the suture, topical steroid use, and inflammatory ocular disorders are associated with suture erosions after penetrating keratoplasty

The association of glaucoma and penetrating keratoplasty presents a difficult management problem, particularly when medical therapy and conventional glaucoma surgery have failed. We have found alloplastic tube shunt surgery to be an effective method for intraocular pressure control in glaucomatous eyes subjected to penetrating keratoplasty. We report 46 patients who underwent alloplastic tube shunt implantation before (13 patients, group A), in combination with (17 patients, group B), or after (16 patients, group C) penetrating keratoplasty. All groups had significantly lowered intraocular pressure postoperatively. The most common complication was graft failure, in 31% of patients in group A, 29% in group B, and 44% in group C. Other postoperative complications and results including visual outcome are reviewed. Because the management of advanced glaucoma in patients undergoing keratoplasty is difficult, alloplastic tube shunt implantation should be considered

A prospective study of 84 consecutive patients undergoing penetrating keratoplasty over a 6-month interval was performed to identify risk factors for postoperative epithelial defects. Data collected included donor age, host age, time postmortem until enucleation, general health of donor and host, any history of glaucoma, any surgical procedure, the indication for penetrating keratoplasty, and the presence of any epithelial defect preoperatively in the donor. Patients were evaluated 1 day postoperatively for any epithelial defects greater than 1 mm2 central to the sutures. Single-variable associations and multiple stepwise logistic regression using standard biostatistical techniques were performed. Thirty-one percent (26 of 84) of the patients had epithelial defects on the first postoperative day. Only diabetes in the donor (P = .03) and longer death-to-enucleation times (P = .03) were associated with an increased risk of epithelial defects. We conclude that diabetes in the donor and longer death-to-enucleation times predispose to epithelial defects after penetrating keratoplasty