The Risk of Sympathetic Ophthalmia Associated with Open-Globe Injury Management Strategies: A Meta-analysis.

TOPIC: Sympathetic ophthalmia (SO) is a sight-threatening granulomatous panuveitis caused by a sensitizing event. Primary enucleation or primary evisceration, versus primary repair, as a risk management strategy after open-globe injury (OGI) remains controversial. CLINICAL RELEVANCE: This systematic review was conducted to report the incidence of SO after primary repair compared with that of after primary enucleation or primary evisceration. This enabled the reporting of an estimated number needed to treat. METHODS: Five journal databases were searched. This review was registered with International Prospective Register of Systematic Reviews (identifier, CRD42021262616). Searches were carried out on June 29, 2021, and were updated on December 10, 2022. Prospective or retrospective studies that reported outcomes (including SO or lack of SO) in a patient population who underwent either primary repair and primary enucleation or primary evisceration were included. A systematic review and meta-analysis were carried out in accordance with Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Random effects modelling was used to estimate pooled SO rates and absolute risk reduction (ARR). RESULTS: Eight studies reporting SO as an outcome were included in total. The included studies contained 7500 patients and 7635 OGIs. In total, 7620 OGIs met the criteria for inclusion in this analysis; SO developed in 21 patients with OGI. When all included studies were pooled, the estimated SO rate was 0.12% (95% confidence interval [CI], 0.00%-0.25%) after OGI. Of 779 patients who underwent primary enucleation or primary evisceration, no SO cases were reported, resulting in a pooled SO estimate of 0.05% (95% CI, 0.00%-0.21%). For primary repair, the pooled estimate of SO rate was 0.15% (95% CI, 0.00%-0.33%). The ARR using a random effects model was -0.0010 (in favour of eye removal; 95% CI, -0.0031 [in favor of eye removal] to 0.0011 [in favor of primary repair]). Grading of Recommendations, Assessment, Development, and Evaluations analysis highlighted a low certainty of evidence because the included studies were observational, and a risk of bias resulted from missing data. DISCUSSION: Based on the available data, no evidence exists that primary enucleation or primary evisceration reduce the risk of secondary SO. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

The Use of Preoperative Prophylactic Systemic Antibiotics for the Prevention of Endopthalmitis in Open Globe Injuries: A Meta-Analysis.

TOPIC: This study reports the effect of systemic prophylactic antibiotics (and their route) on the risk of endophthalmitis after open globe injury (OGI). CLINICAL RELEVANCE: Endophthalmitis is a major complication of OGI; it can lead to rapid sight loss in the affected eye. The administration of systemic antibiotic prophylaxis is common practice in some health care systems, although there is no consensus on their use. METHODS: PubMed, CENTRAL, Web of Science, CINAHL, and Embase were searched. This was completed July 6, 2021 and updated December 10, 2022. We included randomized and nonrandomized prospective studies which reported the rate of post-OGI endophthalmitis when systemic preoperative antibiotic prophylaxis (via the oral or IV route) was given. The Cochrane Risk of Bias tool and ROBINS-I tool were used for assessing the risk of bias. Where meta-analysis was performed, results were reported as an odds ratio. PROSPERO registration: CRD42021271271. RESULTS: Three studies were included. One prospective observational study compared outcomes of patients who had received systemic or no systemic preoperative antibiotics. The endophthalmitis rates reported were 3.75% and 4.91% in the systemic and no systemic preoperative antibiotics groups, a nonsignificant difference (P = 0.68). Two randomized controlled trials were included (1555 patients). The rates of endophthalmitis were 17 events in 751 patients (2.26%) and 17 events in 804 patients (2.11%) in the oral antibiotics and IV (± oral) antibiotics groups, respectively. Meta-analysis demonstrated no significant differences between groups (odds ratio, 1.07; 95% confidence interval, 0.54-2.12). CONCLUSIONS: The incidences of endophthalmitis after OGI were low with and without systemic antibiotic prophylaxis, although high-risk cases were excluded in the included studies. When antibiotic prophylaxis is considered, there is moderate evidence that oral antibiotic administration is noninferior to IV. FINANCIAL DISCLOSURE(S): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Pain Control Following Photorefractive Keratectomy: A Prospective Clinical Trial Comparing Codeine Versus Oxycodone for the Management of Postoperative Pain.

PURPOSE: To compare the efficacy of oral codeine plus acetaminophen versus oxycodone plus acetaminophen for severe pain control following photorefractive keratectomy (PRK). METHODS: This single-center trial randomized 200 patients to receive codeine 30 mg/acetaminophen 325 mg (codeine group) or oxycodone 5 mg/acetaminophen 325 mg (oxycodone group)every 4 hours as needed for severe pain for 4 days following PRK. Patients recorded postoperative pain, tablet consumption, and tetracaine use. Patients were monitored at postoperative 1 day, 1 week, and 1, 3, and 6 months for visual acuity and follow-up. Study outcomes were mean postoperative pain, treatment and tetracaine use, and visual acuity. RESULTS: Analysis of 197 patients who completed the trial (97 codeine group and 100 oxycodone group) showed mean pain scores were lower in the codeine group throughout the intervention period. Mean pain scores were higher in the oxycodone group than the codeine group on postoperative days 2 and 4 (P = .017 and P = .034, respectively). The oxycodone group consumed more tablets than the codeine group, with a difference on postoperative day 2 (P = .019), and used a greater number of tetracaine drops (P = .015). Repeated measures analysis of variance showed significant improvement in visual acuity in both groups with no difference in visual outcomes (P = .81). CONCLUSIONS: Codeine/acetaminophen is as effective and safe as oxycodone/acetaminophen for pain control following PRK, with no clinical difference in overall pain control and long-term visual outcomes. This implies that treating postoperative pain after PRK with a Schedule III opioid (codeine) is effective and potentially decreases the risk of misuse by a higher regulated Schedule II opioid (oxycodone), lowering the potential for abuse and dependence. [J Refract Surg. 2021;37(9):582-589.].

Dexamethasone-Eluting Contact Lens for the Prevention of Postphotorefractive Keratectomy Scar in a New Zealand White Rabbit Model.

PURPOSE: To evaluate the safety and efficacy of an experimental dexamethasone-eluting contact lens (DCL) for the prevention of postphotorefractive keratectomy (PRK) corneal haze in a New Zealand White (NZW) rabbit model. METHODS: Both eyes of 29 NZW rabbits underwent PRK. The rabbits were randomized to one of the 5 study arms for 4 weeks: tarsorrhaphy only, tarsorrhaphy and bandage contact lens (BCL) replaced weekly, tarsorrhaphy and BCL for 1 week plus topical 0.1% dexamethasone ophthalmic solution (drops) for 4 weeks, tarsorrhaphy and BCL replaced weekly plus topical dexamethasone for 4 weeks, and tarsorrhaphy and DCL changed weekly for 4 weeks. Each week for 4 consecutive weeks postoperatively, the tarsorrhaphies were opened, the eyes underwent evaluation and imaging, and the tarsorrhaphies were replaced. Contact lenses were cultured on removal. Central corneal haze was assessed weekly with corneal densitometry. After 4 weeks, the animals were killed, and the eyes were enucleated for histopathologic analysis. RESULTS: The tarsorrhaphy only group displayed more haze with a greater change in optical densitometry from pre-op compared with the other treatment groups. There was no difference between the DCL group and the groups receiving a BCL and dexamethasone drops in densitometry or histopathology. No NZW rabbits developed clinical signs of infection, and cultures from DCLs and BCLs grew similar organisms. CONCLUSIONS: In the post-PRK rabbit model, DCLs worn weekly for 4 weeks were safe and as effective at preventing corneal haze as 0.1% dexamethasone drops applied 4 times a day for 4 weeks.

Assessment of Factors Associated With Postoperative Pain After Photorefractive Keratectomy.

PURPOSE: To define the factors that affect patient's self-assessed postoperative pain after photorefractive keratectomy (PRK). METHODS: Patients who underwent PRK in 2016 were evaluated. Anonymized data collected included patient gender, age, and season at the time of surgery, ablation depth, surgeon status (attending vs. resident), topical tetracaine use, and subjective pain scores at postoperative days (PODs) 1 and 7. Average pain scores and amount of pain medication taken were analyzed for each of the previously mentioned variables. RESULTS: Overall, 231 patients who underwent PRK were analyzed. The mean pain score and SD were 0.78 ± 1.87 on POD 1 and 0.03 ± 0.37 by POD 7. Patients who used topical tetracaine reported significantly higher pain on POD 1 and 7 compared with patients who did not use tetracaine (P < 0.001 and P = 0.038, respectively). No significant differences in pain scores were seen based on surgeon status, ablation depth, gender, and season. Patients who used topical tetracaine took a higher amount of oral pain medication (9.44 ± 6.01) compared with those who did not (7.02 ± 4.71) (P = 0.022). CONCLUSIONS: Postoperative pain was significantly elevated in patients who used tetracaine on POD 1 and POD 7. These patients were also more likely to take oral pain medication than those who did not use topical tetracaine. Surgeon status, season, gender, and ablation depth showed no significant differences in subjective pain scores. Oral pain medication should be evaluated to assess efficacy and safety in inhibiting ocular pain after PRK.

Dose Uniformity of Topical Corticosteroids: A Simulated Trial of Fluorometholone Acetate 0.1% and Loteprednol Etabonate Gel 0.5.

PURPOSE: The purpose of the study was to determine the concentrations of Flarex® and Lotemax® when shaken and not shaken. Many patients fail to shake or inappropriately shake suspensions of corticosteroids before instillation as directed. This study was designed to help determine what concentration of corticosteroid these patients are receiving. In addition, independent confirmation of loteprednol etabonate ophthalmic gel dose uniformity was determined and compared as a possible alternative. METHODS: Drug concentrations of shaken versus unshaken Flarex and Lotemax were determined over a 20-day simulated tapered course in our institutional laboratory. Collected samples were analyzed by reversed-phase high-performance liquid chromatography with photodiode array detection at 240 nm. RESULTS: Flarex had a mean concentration of 93.7% of the declared concentration when shaken and 7.25% when not shaken. The difference between these groups was statistically significant (P = 0.0001). Lotemax had a mean concentration of 96.74% of the declared concentration when shaken and a mean concentration of 98.97% when not shaken. The difference between these groups was not statistically significant (P = 0.194). CONCLUSIONS: Flarex maintains dose uniformity when shaken. When not shaken, it has poor dose uniformity. Lotemax was consistent whether shaken or not in our study and can be considered to eliminate the variability of poor patient compliance with shaking. The manufacturers of both drugs recommend shaking before application.

Immunomodulatory therapy with anti-interleukin-1 after photorefractive keratectomy in the New Zealand white rabbit.

PURPOSE: To examine the safety and efficacy of topical anti-interleukin-1 (anti-IL-1) following photorefractive keratectomy (PRK) in rabbit eyes. SETTING: Joint Warfighter Refractive Surgery Center, Lackland Air Force Base, Texas, USA. DESIGN: Experimental study. METHODS: After standard PRK, 48 eyes of 24 New Zealand white rabbits were divided into 4 treatment arms and 1 control arm. Eyes in the treatment arms were randomized to receive fluorometholone 0.1% or an anti-IL-1 suspension (2.50 mg, 1.25 mg, or 0.25 mg doses) plus standard moxifloxacin, balanced salt solution (BSS), and an ocular lubricant (Systane) 4 times a day. Control eyes received only moxifloxacin, balanced salt solution, and ocular lubricant. RESULTS: No adverse events were observed with anti-IL-1. The safety of anti-IL-1 was affirmed because there was no statistically significant difference in time to epithelial closure, foam-layer histology and thickness, or final stromal thickness measurements between the anti-IL-1 and the steroid or control groups. No increase in haze was observed with anti-IL-1. There was a trend toward decreased haze with anti-IL-1 at several data points compared with the control and steroid groups. Finally, there was a trend toward less haze in all metrics at almost every timepoint for the 2.50 mg anti-IL-1 group compared with lesser concentrations. CONCLUSIONS: Anti-IL-1 therapy might be a safe, effective alternative to steroids for haze prevention after PRK. Of the doses studied, 2.50 mg of anti-IL-1 4 times a day appeared to be most effective. Further studies in human eyes are needed. FINANCIAL DISCLOSURE: Dr. Reilly has been a consultant to Alcon Laboratories, Inc., and Abbott Medical Optics, Inc. None of the authors has a financial or proprietary interest in any material or method mentioned.

Visual outcomes after femtosecond laser in situ keratomileusis flap complications.

PURPOSE: To compare visual outcomes in normal and complicated laser in situ keratomileusis (LASIK) flaps constructed with a femtosecond laser. SETTING: Wilford Hall Ambulatory Surgical Center, San Antonio, Texas. DESIGN: Retrospective chart review. METHODS: The main outcome measures were uncorrected distance visual acuity (UDVA) and best-corrected distance visual acuity (CDVA) at 1 month post-surgery, which were compared in patients with and without complications during flap creation. RESULTS: This retrospective chart review identified 586 eyes of 293 consecutive patients who had bilateral simultaneous femtosecond LASIK for myopia. A normal flap was attained in 539 (91.8%) of patients. At 1 month postoperatively, the mean UDVA was 20/19.1, and the CDVA was 20/16.8. 3 (0.51%) of patients experienced an incomplete side cut or intraoperative flap tear (the mean 1-month uncorrected distance visual acuity [UDVA] was 20/17.5 and the corrected distance visual acuity [CDVA] was 20/16.9 in both groups). Microstriae were seen in 28 (4.8%) of patients (mean 1-month UDVA, 20/21.7; CDVA, 20/17.4). Diffuse lamellar keratitis appeared in 4 (0.68%) of patients (mean 1-month UDVA, 20/21.7; CDVA, 20/15.6). Epithelial ingrowth developed in 9 (1.5%) of patients (mean 1-month UDVA, 20/18.7; CDVA, 20/16.9). There was no statistically significant difference in UDVA or CDVA at 1 month postoperatively between complicated and uncomplicated flaps (P > .05), and complicated flaps did not lose a statistically significant amount of CDVA or low-contrast acuity (CDVA 5%) compared to preoperative values. CONCLUSIONS: Intraoperative and postoperative complications of LASIK flaps created with the femtosecond laser were rare, and their visual outcomes were equivalent to those of normal flaps. FINANCIAL DISCLOSURE: Charles D. Reilly is a consultant to Alcon Labs and Abbott Medical Optics at present, but was not at the time of the study. For the remaining authors none are declared.

Retrospective Analysis Comparing the Preoperative and Postoperative "Q" Values for 2 Different Lasers in Refractive Surgery.

PURPOSE: To compare the presurgical and postsurgical corneal asphericity, or "Q value," between VISX Wavefront-guided treatment and Allegretto Wave Wavefront-optimized treatment in photorefractive keratectomy for the correction of myopia. METHODS: This is a retrospective database analysis performed at the Joint Warfighter Refractive Surgery Center, Lackland Air Force Base, TX. All data were accessed through the Institutional Review Board-approved Joint Warfighter Refractive Surgery Center database. Fifty-one patient records (102 eyes) were reviewed to determine the change in the Q value after photorefractive keratectomy for the correction of myopia. RESULTS: The average change in the Q value per diopter of treatment sphere over the entire study population was 0.12 (±0.04) for Allegretto and 0.14 (±0.04) for VISX (P = 0.004). There was no statistical difference among the preoperative and postoperative best-corrected visual acuities (BCVAs), low-contrast BCVA (BCVA 5%), or quality of vision complaints between both platforms. CONCLUSIONS: Low to moderate myopic correction with the Allegretto laser was associated with a smaller change in the Q value per diopter of treatment and therefore more closely maintained the original prolate shape of the cornea when compared with VISX. No difference in the postoperative visual outcome between both groups was detected.

Comparison of pachymetry measurements between the Alcon Wavelight EX500 and Sonogage Corneo-Gage plus platforms.

PURPOSE: To compare the agreement of intraoperative central corneal thickness (CCT) measurements of the Wavelight EX500 (Alcon Laboratories, Fort Worth, TX) that uses optical low coherence reflectometry to the Corneo-Gage Plus ultrasound pachymetry device (Sonogage, Cleveland, OH). METHODS: In this retrospective study, 50 eyes of 26 patients undergoing LASIK were evaluated. Following LASIK flap creation, each eye was measured by both optical low coherence reflectometry and ultrasound pachymetry immediately prior to flap lifting and then again after flap lifting. RESULTS: The mean CCT value before lifting the flap was 556.9 and 557.78 µm as measured by ultrasound pachymetry and optical low coherence reflectometry, respectively. After lifting the flap, the mean ultrasound pachymetry value was 440.96 µm and the mean optical low coherence reflectometry value was 441.7 µm. A two-sample Kolmogorov-Smirnov test demonstrated that the ultrasound pachymetry and the optical low coherence reflectometry distribution of measurements were the same. A Shapiro-Wilk test of normality could not be rejected. Bland-Altman plots showed strong agreement. The correlation between the two tests was 0.98 before flap lifting and 0.97 after flap lifting, both with a 95% confidence interval. CONCLUSIONS: The pachymetry measurements by the optical low coherence reflectometry correlated with those of the ultrasound pachymetry device. The Wavelight EX500 optical low coherence reflectometry may be used in place of the ultrasound pachymetry device for measuring CCT.

The effect of silicone hydrogel bandage soft contact lens base curvature on comfort and outcomes after photorefractive keratectomy.

OBJECTIVE: To evaluate the relative pain and the relative amount of contact lens loss experienced using two different base curvatures (BCs) of the Acuvue Oasys bandage soft contact lens (BSCL) after photorefractive keratectomy (PRK). METHODS: One hundred forty patients undergoing PRK on either the Allegretto or the VISX laser at the Joint Warfighter Refractive Surgery Center in Lackland AFB, TX, were randomized to one of the two different BCs of the Acuvue Oasys BSCL: 8.4 or 8.8 mm. Patients were evaluated on postoperative days 1 and 4 during which they completed a survey rating absolute pain in each eye on a visual analog pain scale. Lens loss was recorded throughout the study. RESULTS: Patients treated on the Allegretto laser preferred the 8.4-mm BC lens, whereas comfort after treatment on the VISX depended on corneal shape. For VISX, patients with very flat corneas (steep K ≤42 preoperative or ≤38 postoperatively) preferred an 8.8-mm BC lens while patients with very steep corneas (steep K >45 preoperative or >42 postoperative) preferred an 8.4-mm BC lens, though these results were largely not statistically significant. Patients who lost their lenses prematurely tended to be those whose corneal curvature did not match their contact lens BC. CONCLUSIONS: Individuals treated with the Allegretto laser or individuals with more prolate corneas should likely be fit with an 8.4-mm BC Acuvue Oasys BSCL while individuals with more oblate corneas should likely be fit with an 8.8-mm BC lens to minimize postoperative pain and premature BSCL loss.

Comparison of 3 silicone hydrogel bandage soft contact lenses for pain control after photorefractive keratectomy.

PURPOSE: To evaluate the relative pain with 3 U.S. Food and Drug Administration-approved bandage soft contact lenses (SCLs) applied after photorefractive keratectomy (PRK). SETTING: Joint Warfighter Refractive Surgery Center, Lackland Air Force Base, Texas, USA. DESIGN: Prospective comparative case series. METHODS: Patients having PRK were randomized to a senofilcon A (Acuvue Oasys), balafilcon A (Purevision), or lotrafilcon A (Air Optix) bandage SCL in each eye postoperatively. Patients were evaluated 1 and 4 days postoperatively and completed a survey rating absolute pain in each eye on a visual analog scale. RESULTS: The study enrolled 54 patients. At 1 and 4 days, eyes with the senofilcon A lens had the lowest pain scores followed by eyes with the lotrafilcon A lens and then eyes with the balafilcon A lens. Averaging qualitative results from 1 and 4 days showed that eyes with the senofilcon A lens were reported as having more pain by 4% of patients, eyes with the lotrafilcon A lens by 27%, and eyes with the balafilcon A lens by 53%; 16% reported no difference (P<.001). Quantitatively, the senofilcon A lens was 40% more comfortable than the lotrafilcon A lens and 65% more comfortable than the balafilcon A lens on average. The lotrafilcon A lens was 38% more comfortable than the balafilcon A lens on average (P<.01). CONCLUSIONS: There was a statistically and clinically significant difference in post-PRK pain between the 3 bandage SCLs. The senofilcon A lens caused the least pain. FINANCIAL DISCLOSURE: Dr. Reilly is a consultant to Alcon Laboratories, Inc. and Abbott Medical Optics, Inc. but was not at the time of the study. No other author has a financial or proprietary interest in any material or method mentioned.

Novel approach for the treatment of corneal ectasia in a graft.

PURPOSE: The aim of this study was to report the use of novel masking agents during an anterior lamellar keratoplasty performed using a femtosecond laser in a patient with corneal ectasia that was consistent with recurrent keratoconus. METHODS: This is a case report. RESULTS: A 55-year-old man, with a 23-year status after penetrating keratoplasty for keratoconus, presented with a chief complaint of ocular discomfort in the right eye. On slit-lamp examination, the physician estimated 70% to 80% inferior thinning at the graft-host interface with inferior corneal neovascularization. Because of the high risk of developing corneal perforation and the patient's desire to minimize visual recovery time, anterior lamellar keratoplasty was chosen. To minimize the risk of perforation during femtosecond dissection of the anterior lamellar bed, gentian violet and cyanoacrylate glue were used in the area of thinning as masking agents. CONCLUSIONS: This represents the first documented use of gentian violet and cyanoacrylate glue as double masking agents to defocus the femtosecond laser raster pass during keratoplasty.

Air bubble-associated endothelial trauma in descemet stripping automated endothelial keratoplasty.

PURPOSE: To evaluate endothelial cell trauma by anterior chamber (AC) air bubbles in Descemet stripping automated endothelial keratoplasty (DSAEK). DESIGN: Laboratory investigation. METHODS: Twelve human donor corneas (6 pairs) were sectioned using an automated microkeratome system (Moria ALTK System, Antony, France). One cornea of each pair was mounted on a Moria artificial AC, and an air bubble was injected to fill 40% of the AC. The apparatus was rotated 180 degrees for a total of 50 times to simulate air bubble trauma. The fellow corneas were used as controls. Each endothelial graft was stained with 0.25% Trypan blue for 90 seconds followed by 0.2% alizarin red for 2 minutes, and digital photomicrographs were obtained. Abnormally staining areas indicative of graft injury were removed digitally from the total graft area. The proportion of uninjured corneal endothelium was calculated, and differences were analyzed. RESULTS: In this ex vivo model of air bubble trauma, the proportion of viable graft endothelium after air bubble injury was 79.8 +/- 0.04% (n = 6). The proportion of viable endothelium in the control group was 89.9 +/- 0.02% (n = 6). The statistically significant mean difference of 10.1% (P = .03) is indicative of greater endothelial injury after air bubble trauma. CONCLUSIONS: Using this model, a moderate but significant amount of endothelial cell damage was associated with air bubble trauma compared with the control group. Air bubble trauma may account partially for the loss of endothelial cell density after DSAEK surgery and may impact graft survival.

The histology of graft adhesion in descemet stripping with endothelial keratoplasty.

PURPOSE: To describe the histologic findings of full-thickness corneas from penetrating keratoplasty (PK) in 10 patients with a prior history of Descemet stripping with endothelial keratoplasty. DESIGN: Retrospective review of the histopathology of 10 PK specimens. METHODS: We reviewed histologic sections stained with hematoxylin and eosin, periodic acid-Schiff, and colloidal iron stains from 10 PK specimens, which had been trisected and submitted in their entirety. Clinical data were abstracted from the medical record. RESULTS: The interface between host and graft in most cases was barely perceptible with minimal changes in the degree of tissue eosinophilia, a subtle increase in keratocyte cellularity focally, or the presence of melanin granules within keratocytes. Residual host Descemet membrane (DM) was found in 8 of the 10 cases, most often at the edges of the graft. In no case did the presence of DM appear to hinder graft adhesion. CONCLUSION: The adhesion of Descemet stripping with endothelial keratoplasty grafts is not associated with significant scarring or keratocyte proliferation. Contrary to previous assumptions, retained DM did not appear to hinder graft adhesion, raising the possibility that removal of DM may be unnecessary for endothelial transplantation.

Candida glabrata endophthalmitis following penetrating keratoplasty.

We report a case of Candida glabrata endophthalmitis following penetrating keratoplasty in a 57-year-old man. The infection was thought to be treated successfully with intravitreal amphotericin B but flared 7 months later following cataract extraction and eventually required explantation of the intraocular lens and therapeutic keratoplasty. The literature regarding this rare infection is reviewed. Candida glabrata, an uncommon ocular pathogen, is being reported with increasing frequency and with a notable predilection for post-keratoplasty eyes. In the 10 reported cases, there is 100% concordance between host and donor tissue cultures. In half the cases, there was a latent period of several months. These infections can be difficult to treat because C glabrata is often resistant to the antifungal agents commonly used to treat Candida albicans.