Postoperative Endophthalmitis in Immediate Sequential Bilateral Cataract Surgery: A Nationwide Registry Study.

PURPOSE: To report the incidence of postoperative endophthalmitis (PE) after immediate sequential bilateral cataract surgery (ISBCS) in Sweden. DESIGN: Retrospective cohort registry study. PARTICIPANTS: Patient data from 1 457 172 cataract extractions, including 1 364 934 unilateral surgeries and 92 238 ISBCSs. METHODS: Endophthalmitis cases reported to the Swedish National Cataract Register (NCR) during a 16-year period (2002-2017) were analyzed in comparison to all control cases with regard to patient characteristics, surgical technique, and capsule complication. MAIN OUTCOME MEASURE: Incidence and determinants for PE in ISBCS compared with unilateral surgeries. RESULTS: A total of 422 cases of PE were identified in 1 457 172 cataract extractions, yielding an overall incidence of 0.029% (95% confidence interval [CI], 0.0262-0.0317). For unilateral procedures, the rate was 0.0299% (95% CI, 0.0270-0.0328) or 408 cases in 1 364 934 operations, whereas that for ISBCS was 0.0152% (95% CI, 0.0072-0.0231) or 14 incidents in 92 238 operations (P = 0.01). In a logistic regression model including all cataract procedures, nonuse of intracameral (IC) antibiotics (ABs), capsule complication, age 85 years or more, male gender, and ocular comorbidity were found to be independent risk factors for PE. All these parameters were less frequent in ISBCS. Notwithstanding, in the same multivariate analysis, ISBCS in itself was associated with a significantly lower risk for PE. At follow-up, 5 of the 14 PE cases in the ISBCS cohort had a visual acuity (VA) of 20/200 or worse. Of these, one 93-year-old ISBCS patient developed bilateral infection. CONCLUSIONS: After ISBCS in Sweden, PE occurred once in 6600 surgeries. The risk of sustaining a final VA of 20/200 or less was 1 incident in 18 000 operated eyes. When counseling potential ISBCS patients about the risk of PE, it seems reasonable to state that the reported risk in the literature is lower than that with unilateral surgery but not negligible. Precautions remain necessary.

CLINICAL CHARACTERISTICS AND VISUAL OUTCOMES IN ENDOPHTHALMITIS AFTER KERATOPROSTHESIS IMPLANTATION.

PURPOSE: To describe the presentation, microbiology, management, and prognosis of eyes with endophthalmitis after Boston keratoprosthesis implantation. METHODS: Retrospective case series with history, diagnostics, management, and outcomes data in endophthalmitis after keratoprosthesis implantation presenting to a tertiary center between 2009 and 2020. RESULTS: Of 137 keratoprosthesis-implanted eyes, 7 eyes of 7 patients (5%) developed endophthalmitis. On presentation, 6 (86%) reported decreased visual acuity, and only 1 (14%) reported pain. Peripheral corneal ulcers were present in 2 eyes (29%). Seidel testing was negative in all cases. Six eyes (86%) had retroprosthetic membranes. One (14%) underwent initial pars plana vitrectomy with mechanical vitreous biopsy, whereas 6 (86%) received a needle vitreous tap-half of which were dry. Organisms were isolated after vitreous tap in two eyes: Streptococcus intermedius and Mycobacterium abscessus. The mean visual acuity preendophthalmitis, at presentation, and at 6 months were 20/267, 20/5,944, and 20/734, respectively. The visual acuity improved 9.08 ± 11.78 Early Treatment Diabetic Retinopathy Study lines from presentation to 6 months. Six-month visual acuity was correlated with preendophthalmitis visual acuity (r = 0.92, P = 0.003) but not presenting visual acuity (P = 0.838). CONCLUSION: Visual acuity at 6 months is correlated with preendophthalmitis visual acuity, not presenting visual acuity. Endophthalmitis should be considered in the differential diagnosis of painless intraocular inflammation any time after keratoprosthesis implantation, even if Seidel negative.

C-X-C Chemokines Influence Intraocular Inflammation During Bacillus Endophthalmitis.

Purpose: The purpose of this study was to explore the C-X-C chemokines CXCL2 and CXCL10 as potential anti-inflammatory targets for Bacillus endophthalmitis. Methods: Bacillus endophthalmitis was induced in C57BL/6J, CXCL2-/-, and CXCL10-/- mice. At specific times postinfection, eyes were analyzed for Bacillus, retinal function, and inflammation. The efficacies of intravitreal anti-CXCL2 and anti-CXCL10 with or without gatifloxacin in B. cereus endophthalmitis were also assessed using the same techniques. Results: Despite similar Bacillus growth in eyes of C57BL/6J, CXCL2-/-, and CXCL10-/- mice, retinal function retention was greater in eyes of CXCL2-/- and CXCL10-/- mice compared to that of C57BL/6J mice. Neutrophil migration into eyes of CXCL2-/- and CXCL10-/- mice was reduced to a greater degree compared to that of eyes of C57BL/6J mice. Infected CXCL2-/- and CXCL10-/- mouse eyes had significantly less inflammation compared to that of C57BL/6J eyes. Retinal structures in infected eyes of CXCL2-/- mice were preserved for a longer time than in CXCL10-/- eyes. Compared to untreated eyes, there was less inflammation and significant retention of retinal function in eyes treated with anti-CXCL2 and anti-CXCL10 with or without gatifloxacin. Conclusions: For Bacillus endophthalmitis, the absence of CXCL2 or CXCL10 in mice resulted in retained retinal function and less inflammation. The absence of CXCL2 led to a better clinical outcome than the absence of CXCL10. The use of anti-CXCL2 and anti-CXCL10 limited inflammation during B. cereus endophthalmitis. These results highlight the utility of CXCL2 and CXCL10 as potential targets for anti-inflammatory therapy that can be tested in conjunction with antibiotics for improving treating Bacillus endophthalmitis.

Hypopyon: Is-it Infective or Noninfective?

Hypopyon usually corresponds to the sedimentation of white blood cells, and it signifies severe intraocular inflammation. This key clinical sign may occur in association with a wide variety of infectious, inflammatory, and neoplastic conditions that may be sight- and, occasionally, life-threatening. A careful history and thorough clinical examination are the cornerstones for orienting the differential diagnosis, identifying the causative agent, and initiating prompt and appropriate treatment. This review outlines the clinical characteristics and management of hypopyon in relation with the underlying causative infectious or noninfectious ocular or systemic diseases.

Bilateral Endogenous Methicillin-Resistant Staphylococcus aureus Endophthalmitis in a Young Athlete: A Story of Full Recovery

Endogenous endophthalmitis (EE) is an ophthalmological emergency. We report the long-term outcome of bilateral methicillin-resistant Staphylococcus aureus EE in a 23-year-old healthy immunocompetent athlete who presented with EE secondary to pelvic abscess and remained with excellent vision.

Visual function deficits in eyes with resolved endophthalmitis.

To evaluate the changes in functional vision in patients with resolved endophthalmitis. This was a cross-sectional study. The study included 20 patients with resolved endophthalmitis and best-corrected visual acuity of 20/100 or better. Visual acuity (VA), contrast threshold (CT), red/green (RG) and yellow/blue (YB) colour vision and 15 Hz flicker modulation threshold (FMT) were assessed using standard psychophysical techniques. The median age was 54 years. The median visual acuity was 0.27 (~ 20/40-Snellen Equivalent) ((interquartile range [IQR]), 0.30) logMAR). The median log contrast threshold (CT) was - 1.13 (IQR, 0.36) log units (normative value for age-matched CT: - 1.61 log units). The median red/green (RG) and yellow/blue (YB) thresholds were 11.52 (IQR, 26.19) and 9.45 (IQR, 16.20) CAD units respectively, which were at least 5 times higher than age-matched normative RG and YB thresholds. The median central cone- mediated FMT was 17.64% (IQR, 23.40%), which was much higher compared to age-matched FMT (5.48% [IQR, 3.47]). Linear regression revealed significant relationship between contrast thresholds and foveal thickness (y = 0.001x-1.47, R2 = 0.20, p = 0.048). Though endophthalmitis may resolve with a good visual acuity, deficits in visual functions like chromatic discrimination, cone-mediated flicker and contrast sensitivity persist.

VISUAL OUTCOME OF EARLY VITRECTOMY AND INTRAVITREAL ANTIBIOTICS IN ACUTE POSTSURGICAL AND POSTINTRAVITREAL INJECTION ENDOPHTHALMITIS: European Vitreo-Retinal Society Endophthalmitis Study Report Two.

PURPOSE: To evaluate the visual outcome associated with intravitreal antibiotics (IVA) and pars plana vitrectomy (PPV) for acute postprocedure endophthalmitis. METHODS: Data from 237 eyes presenting with acute postprocedure endophthalmitis were collected from 57 retina specialists in 28 countries. All eyes were treated with IVA on the day of presentation. We classified eyes according to the method of treatment used as IVA and early PPV (IVA + PPV within 1 week of presentation) groups. RESULTS: After exclusion of ineligible eyes, data from 204 eyes were analyzed. The mean (SD) age of patients was 62.7 (21.8) years and 69.3 (12.7) years in the IVA and PPV groups, respectively (P = 0.18). Endophthalmitis secondary to cataract, intravitreal injections, PPV, and other intraocular procedures represented 64.2%, 16.2%, 13.7%, and 5.9% of cases, respectively. Intravitreal antibiotics alone were administered in 55 eyes (27.0%), and early PPV was performed in 149 eyes (73.0%). No difference was found between groups in the final visual acuity of ≥20/60 (43.6%, 65 eyes vs. 34.5%, 19 eyes) and ≤counting fingers (30.9%, 46 eyes vs. 36.4%, 20 eyes) for IVA versus early PPV groups, respectively. Vision of light perception (odds ratio = 12.2; 95% confidence interval: 2.0-72.6) and retinal detachment (odds ratio = 7.7; 95% confidence interval: 1.5-409) at baseline were predictive of vision of ≤counting fingers. Retinal detachment at baseline (odds ratio = 20.4; 95% confidence interval: 1.1-372.1) was predictive of final retinal detachment status. CONCLUSION: The current retrospective multicenter cohort of eyes with acute postprocedure endophthalmitis reports similar outcomes after treatment with IVA alone when compared with IVA and early PPV within 1 week of presentation.

Pars Plana Vitrectomy for Endophthalmitis: Microbiologic Spectrum and Clinical Outcomes.

Purpose: To report visual outcomes, microbiologic spectrum, and antibiotic resistance in endophthalmitis patients undergoing pars plana vitrectomy (PPV).Methods: Retrospective study of 32 patients who underwent PPV with microbial analysis. Linear mixed models were utilized to compare visual acuity (VA).Results: Streptococcal species and coagulase-negative staphylococcus (CoNS) were the most common organisms. No resistance to vancomycin or fluoroquinolones was observed. Culture-negative individuals had better VA and lower incidences of retinal detachment (RD) and hypotony and better VA than culture-positive group at post-surgical month 3 (p = .025) and marginally at month 12 (p = .098). CoNS endophthalmitis (final logMAR VA 0.80) was associated with better VA than Streptococcal endophthalmitis (final logMAR VA 2.36) (p = .001). Secondary RD was observed in 33.3% of non-cataract endophthalmitis.Conclusion: No organisms were resistant to vancomycin or fluoroquinolones. Culture-negative endophthalmitis had better VA and lower rates of RD and hypotony than culture-positive group. A high rate of RD was observed in non-cataract endophthalmitis.

Post-cataract Surgery Fungal Endophthalmitis: Management Outcomes and Prognostic Factors.

Purpose: To describe the clinical and microbiological features, management outcomes, and prognostic factors of visual improvement in post-cataract surgery fungal endophthalmitis.Methods: We evaluated 17 patients with culture-confirmed fungal endophthalmitis following cataract surgery. Data were collected regarding intravitreal antibiotics and antifungals, pars plana vitrectomy (PPV), intraocular lens (IOL) explantation, and scleral-fixated IOL (SFIOL) implantation. Clinical outcomes were recorded in terms of visual acuity and management strategies.Results: PPV was performed as a primary procedure at first presentation in three eyes. Fourteen eyes (82.3%) received intravitreal voriconazole. IOL explantation was performed in four eyes, followed by an SFIOL implantation in three eyes. Six eyes additionally presented with keratitis. At the final follow-up, cases having visual acuity ≥6/60 doubled from 17.6% to 35.3%. Eyes receiving antifungals had a trend toward worse visual outcome.Conclusions: Corneal involvement in addition to endophthalmitis and the presence of Aspergillus terreus were found to be poor prognostic markers.

Venom Ophthalmia and Ocular Complications Caused by Snake Venom.

Little is known about the detailed clinical description, pathophysiology, and efficacy of treatments for ocular envenoming (venom ophthalmia) caused by venom of the spitting elapid and other snakes, as well as ocular complications caused by snake venom injection. In this paper, we review clinical information of case reports regarding venom ophthalmia and snake venom injection with associated ocular injuries in Asia, Africa, and the United States. We also review the literature of snake venom such as their compositions, properties, and toxic effects. Based on the available clinical information and animal studies, we further discuss possible mechanisms of venom ophthalmia derived from two different routes (Duvernoy's gland in the mouth and nuchal gland in the dorsal neck) and the pathophysiology of snake venom injection induced ocular complications, including corneal edema, corneal erosion, cataract, ocular inflammation, retinal hemorrhage, acute angle closure glaucoma, as well as ptosis, diplopia, and photophobia. Finally, we discuss the appropriate first aid and novel strategies for treating venom ophthalmia and snake envenoming.

Clinicomicrobiological profile, visual outcome and mortality of culture-proven endogenous endophthalmitis in Taiwan.

This is a retrospective study in consecutive cases with cultured-proven endogenous endophthalmitis (EE) treated at the largest tertiary medical center in middle Taiwan in the past 10 years. 83 eyes of 70 patients were enrolled. The mean interval between systemic diseases to the diagnosis of EE was 8.84 ± 6.94 days. The mean initial visual acuity (VA) in the logarithm of minimal angle of resolution (logMAR) was 1.63 ± 0.87. Type 2 diabetes mellitus was the most common predisposing medical illness (N = 53, 63.86%). The most common infectious sources were intra-abdominal abscess (N = 36, 43.37%), and the second most reason was urinary tract infection. The causative pathogen was Gram-negative predominant (N = 64, 77.11%). After aggressive treatment, 34.94% of eyes regain useful vision, and only six eyes underwent enucleation or evisceration. The binary multivariate logistic regression model revealed that female gender (95% CI 1.002-19.036, p = 0.05, OR 4.37), initial VA logMAR (95% CI 0.089-0.550, p = 0.01, OR 0.22), and more intravitreal injections (95% CI 0.368-0.927, p = 0.023, OR 0.58) were independent risk factors influencing final outcomes. Based on the results mentioned above, early diagnosis is recommended to gain better outcomes. The mean interval between systemic diseases to the diagnosis of EE was 8.84 ± 6.94 days in our sample population, clinicians should maintain a higher index of suspicion during this period when encountering patients with bacteremia or fungemia.

Case Report: Beyond the Blood-retina Barrier: Intravitreal Caspofungin for Fungal Endophthalmitis.

SIGNIFICANCE: Two fungal endophthalmitis cases demonstrate safety and efficiency of intravitreal caspofungin as a new therapy option in fungal endophthalmitis. PURPOSE: The purpose of this study was to evaluate the intravitreal application of caspofungin for the treatment of fungal endophthalmitis because rising resistance to voriconazole and amphotericin B leads to a need for new antifungal therapy options. CASE REPORT: Initially, both patients with fungal endophthalmitis underwent pars plana vitrectomy. Microbiological analysis revealed Aspergillus terreus and Candida dubliniensis, which both possess atypical resistance patterns. Caspofungin has a low bioavailability in the eye when given systemically. It was injected intravitreally into the eyes affected by fungal endophthalmitis. An injection of 100 µg of caspofungin in a volume 0.1 mL was applied repeatedly. Clinical parameters were recorded. Both eyes were stabilized by the treatment. Finally, the intraocular infections with atypical mycotic agents were eliminated. Visual acuity improved to 0.4 logMAR (20/50 Snellen) in the first case and to 1.0 logMAR (20/200 Snellen) in the second case. During the treatment course, we have not seen any toxic effects or damage of intraocular structures related to the intravitreal administration of caspofungin. CONCLUSIONS: In summary, intravitreal caspofungin was effective and well tolerated in both cases. Therefore, caspofungin seems to be a safe and effective intravitreal alternative to voriconazole and amphotericin B in fungal endophthalmitis.

Clinical Decision Making and Visual Outcomes in Endophthalmitis Treated with Systemic Corticosteroids.

PURPOSE: To characterize practice patterns and compare visual outcomes in patients with endophthalmitis who did or did not receive systemic corticosteroid therapy. DESIGN: Retrospective, nonrandomized comparative trial. PARTICIPANTS: Eyes diagnosed with endophthalmitis at the Duke Eye Center between January 1, 2009, and January 1, 2018, with at least 6 months of follow-up from the time of initial diagnosis. METHODS: Retrospective chart review identified 133 eyes of 130 consecutive patients meeting inclusion criteria. Clinical presentation, initial management decisions, and subsequent outcomes and complications were assessed. Visual acuity (VA) at presentation with endophthalmitis and at 6 months were assessed for each patient. Eyes treated with systemic corticosteroid therapy were identified for further analysis. MAIN OUTCOME MEASURES: Factors associated with oral steroid use on binary logistic regression analysis and improvement in VA 6 months after endophthalmitis. RESULTS: Of 133 eyes with endophthalmitis, 33 (25%) received oral steroids. Oral steroid use was associated with culture-positive endophthalmitis (odds ratio [OR] 2.7; 95% confidence interval [CI], 1.2-6.2), hypotony (OR, 4.2; 95% CI, 1.3-13.6), conjunctival hyperemia (OR, 2.6; 95% CI, 1.02-6.5), and anterior chamber fibrin on examination (OR, 2.7; 95% CI, 1.1-6.3). Eyes with endogenous endophthalmitis were less likely to receive oral steroids (OR, 0.2; 95% CI, 0.05-0.92). Eyes treated with oral steroids were more likely to show VA improvement of 3 lines or more after endophthalmitis (OR, 2.8; 95% CI, 1.1-6.7) and to exhibit greater improvement from presentation to month 6 (-1.102 logarithm of the minimum angle of resolution [logMAR] vs. -0.655 logMAR; P = 0.024). CONCLUSIONS: Systemic corticosteroid therapy was associated with improved visual outcomes in endophthalmitis. The decision to administer oral steroids was significantly associated with hypotony, conjunctival hyperemia, and fibrinous reaction on presentation. Culture-positive cases were more likely to receive oral steroids, whereas patients with endogenous endophthalmitis were less likely to receive oral steroids. A prospective, randomized trial of systemic corticosteroid use in endophthalmitis may be warranted.

The role of pars plana vitrectomy in the management of fungal endogenous endophthalmitis.

PURPOSE: To identify the causative microorganism of fungal endogenous endophthalmitis in our tertiary referral uveitis center and review the therapeutic role of pars plana vitrectomy in patients with fungal endogenous endophthalmitis. METHODS: Seven eyes of six cases were identified as fungal endogenous endophthalmitis through positive cultures of ocular fluids and clinical presentations. The final anatomical and functional results were evaluated. RESULTS: Four women (66.7%) and two men (33.3%) underwent vitrectomy. Control of infection was achieved early on in all cases. Candida (71.4%) and Aspergillus (28.6%) species were identified as causative fungi in patients with fungal endogenous endophthalmitis. Two patients were reoperated due to reinfection and retinal detachment, respectively. Visual acuity improved in six eyes (85.7%) and worsened in one eye (14.3%). At the final examination, the retina was flat in all cases. No eye developed phthisis bulbi. CONCLUSION: Candida species are the most common causative organisms of fungal endogenous endophthalmitis in this study. Pars plana vitrectomy in fungal endogenous endophthalmitis may enhance the treatment of infection by removing fungal elements in the vitreous and aid in diagnosis. Vitrectomy may also be an important tool in the management of vision-threatening post-infectious sequelae such as retinal detachment and reinfections.

Effect of time to primary repair on final visual outcome after open globe injury.

BACKGROUND/AIM: Historic data suggest that open globe injuries should be repaired within 12-24 hours to reduce the risk of endophthalmitis. However, endophthalmitis is uncommon when systemic antibiotic prophylaxis is given. It is not clear whether delayed primary repair impacts visual outcomes in other ways or what is the optimum time to repair. We aimed to examine the effect of time to primary repair on visual outcomes. METHODS: This is a retrospective comparative case series including all open globe injuries presenting to the Birmingham Midland Eye Centre between 1 January 2014 and 15 March 2016. Presenting features, mechanism of injury, visual acuity at 6-12 months and demographic data were examined. RESULTS: 56 open globe injuries were repaired, of which sufficient data for analysis were available on 52 cases. The mean time to primary repair was 1 day after injury (range 5 hours to 7 days). Final visual acuity at 6-12 months was related to the presenting visual acuity and the Ocular Trauma Score and to the time between injury and primary repair, with a reduction in predicted visual acuity of logarithm of the minimum angle of resolution of 0.37 for every 24 hours of delay (95% CI 0.14 to 0.6). DISCUSSION: Open globe injuries should be repaired promptly. Presenting visual acuity remains the strongest predictor of outcome; however, delay to primary repair also reduced final visual acuity, and any significant delay from injury to repair is likely to negatively impact final visual outcome.

LONG-TERM VISUAL OUTCOMES AND CLINICAL FEATURES AFTER ANTI-VASCULAR ENDOTHELIAL GROWTH FACTOR INJECTION-RELATED ENDOPHTHALMITIS.

PURPOSE: To determine long-term visual outcomes in patients who developed endophthalmitis after intravitreal anti-vascular endothelial growth factor injections and to correlate visual outcomes with clinical features. METHODS: This is a retrospective, multicenter, consecutive case series of patients diagnosed with anti-vascular endothelial growth factor injection-related endophthalmitis who were treated at Mid Atlantic Retina, the Retina Service of Wills Eye Hospital, Philadelphia, PA, and the University of Southern California Roski Eye Institute, Los Angeles, CA. Patients were included if they had at least 1 year of follow-up. Primary outcome was to evaluate long-term visual outcomes up to 5 years of follow-up. The secondary outcome was to determine clinical features (e.g., culture results) that may predict long-term visual acuity outcomes. RESULTS: A total of 56 cases of endophthalmitis from 168,247 anti-vascular endothelial growth factor injections were identified (0.033%, 1/3,004 injections), from which 51 eyes met inclusion criteria. Mean follow-up period was 3.3 years (median 4 years; range 1-5 years). A total of 24 patients (47%) reached a maximum final follow-up of 5 years. Mean Snellen visual acuity at the causative injection visit was 20/102 and decreased to counting fingers at diagnosis (P < 0.001). At 6-month follow-up, mean visual acuity improved to 20/644 (P < 0.001) and remained stable up to 5 years (20/480, P = 0.003) follow-up compared with diagnosis. At the final follow-up, 20 eyes had visual acuity that returned to within one line of baseline visual acuity (visual recovery group), whereas 31 patients' visual acuity was at least one line worse than initial visual acuity (visual deterioration group). The cultures for the visual recovery group were more likely to grow coagulase-negative Staphylococcus, whereas the visual deterioration group primarily grew Streptococcus species, Staphylococcus aureus, and Enterococcus faecalis (P = 0.002, comparing organisms isolated in the visual recovery and deterioration group). CONCLUSION: Visual outcomes after anti-vascular endothelial growth factor injection-related endophthalmitis seem to reach peak improvement by 6 months and remain stable up to a median of 4-year follow-up. Patients who develop culture-negative endophthalmitis or endophthalmitis secondary to coagulase-negative Staphylococcus are more likely to regain baseline visual acuity compared with cases secondary to Streptococcus species.

ENDOGENOUS KLEBSIELLA PNEUMONIAE ENDOPHTHALMITIS IN NORTHERN CALIFORNIA.

PURPOSE: To report the clinical features, treatment modalities, and visual outcomes in 12 eyes with endogenous Klebsiella pneumoniae endophthalmitis (EKPE). METHODS: The medical records of all patients diagnosed with EKPE at Stanford Hospital (Palo Alto, CA) and Santa Clara Valley County Hospital (Santa Clara, CA) from January 2000 to March 2017 were retrospectively reviewed. RESULTS: A total of 10 patients (12 eyes) were diagnosed with EKPE. The median age at presentation was 56, 80% were male, and 30% were non-Asian. Presenting visual acuities ranged from 20/20 to no light perception. Of the 12 eyes 10 received a tap and injection (range, 1-33 injections per eye), 2 eyes underwent primary enucleation or evisceration, and 1 patient underwent pars plana vitrectomy after tap and injection. Final visual acuities ranged from no light perception (six eyes) to 20/300 or better (five eyes). Five patients eventually underwent evisceration or enucleation. All cases were associated with positive blood and/or vitreous cultures and had concurrent systemic infection. CONCLUSION: Endogenous Klebsiella pneumoniae endophthalmitis is a rare, but devastating, ocular infection. Most cases in this series resulted in light perception vision or worse, and almost half required enucleation or evisceration. In light of the virulence of EKPE, early diagnosis and treatment should be initiated in all suspected cases.