DELAYED-ONSET BLEB-ASSOCIATED ENDOPHTHALMITIS (19962008)

Causative Organisms and Visual Acuity Outcomes

THEODORE LENG, MD, MS,* DARLENE MILLER, DHSC, HARRY W. FLYNN, JR, MD, DAVID J. JACOBS, MD, STEVEN J. GEDDE, MD
Purpose: The purpose of this study was to report the clinical features, organisms, and outcomes of patients with delayed-onset bleb-associated endophthalmitis. Methods: Retrospective consecutive case series. Patients who were treated for delayed-onset bleb-associated endophthalmitis between January 1, 1996, and July 1, 2008, at a single institution were included. Information on visual acuities, clinical characteristics, causative organisms, and treatment outcomes were collected. Infections within 1 month of glaucoma ltering surgery, inadvertent ltering blebs after cataract surgery, and patients with glaucoma drainage devices were excluded. Results: A total of 71 eyes from 68 patients were identied. An adjunctive antibrotic agent was used in 48 eyes (68%). The mean time between surgery and endophthalmitis was 4.8 years (range, 0.116; standard deviation, 3.6). The average follow-up time after initial treatment was 37 months (range 1144; standard deviation, 41). At presentation, 17 eyes (24%) had a bleb leak. Fifty-seven eyes (83%) were culture positive. The most common causative organisms were Streptococcus species in 20 eyes (30%), gram-negative organisms in 19 eyes (28%), and coagulase-negative Staphylococcus in 12 eyes (18%). All gram-positive isolates were sensitive to vancomycin. Nine eyes (13%) eventually underwent evisceration or enucleation secondary to pain and/or poor vision. The main outcome measure was bestcorrected visual acuity at the last follow-up examination. Final visual acuities in the initial tap/inject group (n = 45) versus the initial vitrectomy group (n = 24) were as follows: $20/40 (29% vs. 4.2%), 20/50 to 20/400 (36% vs. 29%), and ,5/200 (36% vs. 62%). Conclusion: Streptococcus species and gram-negative organisms were the most common causative isolates identied in this case series of delayed-onset bleb-associated endophthalmitis. Despite treatment of the infection, visual outcomes were generally poor. RETINA 31:344352, 2011

ndophthalmitis after glaucoma ltering surgery is a rare, but potentially devastating complication with a reported incidence of 0.2% to 9.6%.110 Risk factors for its development include the use of an antibrotic agent, an inferior bleb location, bleb
From the *Department of Ophthalmology, Stanford University School of Medicine, Stanford, California; and Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida. Supported in part by Research to Prevent Blindness, New York, NY, and the Heed Ophthalmic Foundation, Cleveland, OH. The authors do not have any nancial interests or conicting interests concerning the material in this article to disclose. Reprint requests: Theodore Leng, MD, MS, California Vitreoretinal Center, Stanford University, 1225 Crane Street, Suite 202, Menlo Park, CA 94025; e-mail: tedleng@stanford.edu

leakage, and a history of bleb infection.2,4,1124 Other potential associations include a diagnosis of primary open-angle glaucoma, blepharitis, nasolacrimal duct obstruction, hypotony, laser suture lysis, bleb needling, compression sutures, autologous blood injection, contact lens use, younger age, black race, and absence of a posterior lens capsule.3,1214,16,2527 Given the possible changes in south Florida microbial distribution and increasing amount of methicillin-resistant Staphylococcus aureus in recent years,28 this study sought to investigate the clinical characteristics, causative organisms, treatments, and visual acuity (VA) outcomes of patients who presented to a large medical center with delayed-onset blebassociated endophthalmitis.
344

BLEB-ASSOCIATED ENDOPHTHALMITIS  LENG ET AL

345

Patients and Methods The study protocol was approved by the Institutional Review Board of the University of Miami Miller School of Medicine. A search was conducted in the hospitals coding database to identify all patients with any diagnostic code of glaucoma and any diagnostic code of endophthalmitis who presented between January 1, 1996, and July 1, 2008. All charts were carefully reviewed, and only patients who had both previous glaucoma ltering surgery and a diagnosis of bleb-associated endophthalmitis were included. Because our hospital is a tertiary referral center, some patients were referred for treatment of the infection and had ltering surgery performed elsewhere; however, all follow-up care was performed at our institution until all inammation had subsided. The diagnosis of bleb-associated endophthalmitis was dened as a ltering bleb-associated intraocular infection requiring treatment with intravitreal antibiotics. Bleb infections that did not require intravitreal antibiotics were excluded, as were infections occurring within 1 month of ltering surgery, patients with glaucoma drainage devices, and patients with inadvertent ltering blebs after cataract surgery. The clinical history, clinical characteristics at presentation, culture results, organism sensitivities, treatments, and VA outcomes were retrospectively collected from the medical records of 19 different treating physicians. Cultures were taken from the bleb, the anterior chamber, and the vitreous cavity. Both Gram stain and Giemsa stain were performed and the samples cultured on blood agar, on chocolate agar, and in thioglycolate broth. As performed in similar studies,4,20,29,30 treatment success was dened as best-corrected VA of $20/400 at the nal follow-up. Results Seventy-one eyes from 68 patients with delayed-onset bleb-associated endophthalmitis were identied during the study period. Thirty-ve eyes (49%) were from male patients and 36 (51%) from female patients. The mean age at presentation was 74 years (range, 3399; standard deviation [SD], 13). Ten eyes (14%) were from patients with diabetes mellitus. There were no patients with other forms of immunosuppression, such as human immunodeciency virus and cancer, or on immunosuppressant therapy. The mean time between surgery and the diagnosis of endophthalmitis was 4.8 years (range, 0.116; SD, 3.6). The average follow-up time after initial treatment was 37 months (range, 1144; SD, 41). At the time of glaucoma surgery, an antibrotic agent was used in 48 of the eyes (68%). Thirty-eight eyes

(54%) received mitomycin C and 10 (14%) received 5-uorouracil. Sixteen eyes (23%) had undergone a combined cataract extractiontrabeculectomy surgery; 2 of which also previously had a penetrating keratoplasty. Overall, 42 eyes (59%) had intraocular surgery before the trabeculectomy procedure. The clinical characteristics of the initial examination are summarized in Table 1. Additionally, signicant blepharitis was noted in 3 eyes (4.2%), and lower eyelid ectropion was seen in 1 eye (1.4%). Eleven eyes (15%) received bleb manipulations before the development of endophthalmitis (Table 2). These included autologous blood injection, bleb needling, compression sutures, and bleb revision. More than 1 bleb manipulation was documented in 3 eyes, and the mean time between the nal manipulation and endophthalmitis diagnosis was 27.6 months (range, 0.4192; SD, 56). At the time of diagnosis, 1 eye (1.4%) underwent primary evisceration because of no light perception vision, severe pain, and patient request. A second eye received no initial treatment because of patient refusal. Twenty-four eyes (34%) underwent initial pars plana vitrectomy (PPV) with intravitreal antibiotics, and 45 eyes (63%) underwent initial vitreous tap and intravitreal injection of antibiotics without PPV. Of the latter 45 eyes, 18 (40%) subsequently underwent PPV for persistent inammation and/or the presence of

Table 1. Clinical Features of the Initial Ocular Examination in Eyes with Delayed-Onset Bleb-Associated Endophthalmitis Intraocular pressure (mmHg) Mean SD Range Bleb characteristics, n (%) Inferior bleb Purulent bleb Bleb leak History of bleb manipulation Anterior chamber composition, n (%) Fibrin Hypopyon Lens status, n (%) Phakic Aphakic PCIOL ACIOL View to fundus, n (%) Good Hazy Poor/none 21 14 055 6 54 17 11 (8.5) (76) (24) (15)

42 (59) 49 (69) 14 2 53 2 (20) (2.8) (75) (2.8)

7 (9.9) 17 (24) 47 (66)

ACIOL, anterior chamber intraocular lens; PCIOL, posterior chamber intraocular lens.

346

RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2011  VOLUME 31  NUMBER 2 Table 3. Isolated Organisms in Eyes with Delayed-Onset Bleb-Associated Endophthalmitis (n = 67) Gram-positives Streptococcus species S. oralis S. mitis S. viridans Beta-hemolytic Streptococcus, Group G S. sanguis S. agalactiae S. pneumoniae S. intermedius Peptostreptococcus species Streptococcus species, unspecied Staphylococcus species Coagulase-negative Staphylococcus S. aureus Enterococcus species Corynebacterium species Propionibacterium acnes Gram-negatives Moraxella species Pseudomonas aeruginosa Haemophilus inuenzae Serratia species Proteus mirabilis Enterobacter species n (%) 20 5 4 3 2 1 1 1 1 1 1 20 12 8 5 2 1 (30) (7.5) (6.0) (4.5) (3.0) (1.5) (1.5) (1.5) (1.5) (1.5) (1.5) (30) (18) (12) (7.5) (3.0) (1.5)

Table 2. Bleb Manipulations Performed in 11 Eyes with Bleb-Associated Endophthalmitis Autologous blood injection Bleb needling Compression sutures Bleb revision 3 1 4 7

More than one manipulation was documented in three eyes.

visually signicant vitreous opacities. Five eyes (21%) that received initial PPV with intravitreal antibiotics later underwent a second PPV for persistent inammation and/or the presence of visually signicant vitreous opacities. Sixty-two eyes (87%) initially received an intravitreal injection of vancomycin (1 mg/0.1 mL) with ceftazidime (2.25 mg/0.1 mL). Two eyes (2.8%) received monotherapy with intravitreal vancomycin. Two eyes (2.8%) received intravitreal gentamicin in addition to vancomycin and ceftazidime, 1 eye (1.4%) received intravitreal amikacin and ceftazidime, 1 eye (1.4%) received intravitreal vancomycin and gentamicin, and 1 eye (1.4%) received intravitreal cefazolin and gentamicin. Fifty-six eyes (79%) received intravitreal dexamethasone in combination with the antibiotics. Topical and subconjunctival antibiotics were administered in all eyes. Cultures were taken from 69 of the 71 eyes that presented with delayed-onset bleb-associated endophthalmitis (1 eye that was primarily eviscerated did not have cultures taken before or during surgery, and the second eye had no procedures performed at presentation because of patient refusal). Fifty-seven eyes (83%) were culture positive, and a total of 67 separate organisms were isolated from the intraocular specimens (Table 3). Eight eyes (12%) had polymicrobial infections with 2 organisms and 1 eye (1.4%) with 3 organisms. Streptococcus species (30%) and gramnegative organisms (28%) were the most common isolates cultured from the eyes. Coagulase-negative Staphylococcus represented 12 isolates (18%), and S. aureus was isolated in 8 eyes (12%). Enterococcus species represented 5 isolates (7.5%), and Moraxella species was the most represented gram-negative organism with 7 isolates (10%). All organisms were sensitive to the antibiotics clinically administered, with all the gram-positives being susceptible to vancomycin. After all treatments were administered, no eye showed evidence of intraocular infection or inammation. At presentation, 7 of the 45 eyes (16%) that initially underwent vitreous tap and intravitreal injection of antibiotics had light perception or no light perception vision. However, 16 of the 24 eyes (67%) that underwent initial PPV presented with light

19 (28) 7 4 3 3 1 1 (10) (6.0) (4.5) (4.5) (1.5) (1.5)

perception or no light perception vision. The VA outcomes are summarized in Tables 4 and 5. A nal best-corrected VA of $20/400 was achieved in 37 eyes (52%). An additional 2 eyes did not achieve a nal VA $20/400 but did return to their baseline vision, which was worse than 20/400 because of preexisting optic nerve disease. Eight eyes (40%) with Streptococcus species had a nal VA of $20/400. Seven eyes (58%) with coagulase-negative Staphylococcus and 5 eyes (62%) with S. aureus had a nal VA of $20/400. Only 1 eye (20%) with Enterococcus species had a nal VA of $20/400. Overall, 9 eyes (47%) with gram-negative isolates had a nal VA of $20/400, whereas among Moraxella species, there were 4 eyes (57%) with a nal VA of $20/400. A total of 9 eyes (13%) were eventually eviscerated or enucleated. One was eviscerated at the time of
Table 4. Visual Acuity Outcomes in Eyes with Bleb-Associated Endophthalmitis Initial Therapy Final VA (% patients) N $20/40 20/50 to 20/400 16/45 (36) 7/24 (29) ,5/200 16/45 (36) 15/24 (62)

Tap/inject 45 13/45 (29) Vitrectomy 24 1/24 (4.2)

Two patients were excluded (one because of primary evisceration and one because of refusal of treatment).

Table 5. Visual Acuity Outcomes of Delayed-Onset Bleb-Associated Endophthalmitis by Organism Age Case No. Sex Eye (years) 18 14 38 41 1 49 27 33 56 23 22 66 26 17 61 28 35 37 63 67 70 71 53 5 8 10 32 50 54 11 15 M M M F M F M F F M M M F F F M F F M F M M M F F F F F F M F OS OD OS OS OD OS OS OS OD OD OD OS OS OD OD OD OD OS OD OS OD OS OD OS OS OS OD OS OS OD OS 78 69 67 33 67 79 84 83 77 75 58 83 73 71 70 90 78 60 79 65 80 85 71 78 79 40 88 83 63 87 66 Organism Streptococcus oralis S. oralis S. oralis S. oralis S. mitis S. mitis S. mitis S. mitis S. viridans S. viridans Beta Streptococcus, Group G Beta Streptococcus, Group G S. sanguis S. pneumoniae S. intermedius Streptococcus sp. unspecied Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus S. aureus S. aureus S. aureus S. aureus S. aureus S. aureus Enterococcus faecalis E. faecalis Initial Intervention Tap/inject Vitrectomy Vitrectomy Vitrectomy Tap/inject Tap/inject Vitrectomy Vitrectomy Tap/inject Vitrectomy Tap/inject Tap/inject Vitrectomy Tap/inject Tap/inject Vitrectomy Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Vitrectomy Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Vitrectomy Tap/inject Tap/inject Baseline VA 20/30 Unavailable 20/200 20/40 20/40 20/25 20/30 20/60 20/25 20/20 20/70 20/400 20/50 20/400 20/50 20/60 20/70 20/50 20/30 20/50 20/30 20/25 20/50 20/25 20/300 20/400 20/25 20/30 20/60 20/25 20/30 Presenting 1-Month 3-Month 6-Month 1-Year VA VA VA VA VA CF LP LP LP HM HM LP LP 20/40 LP LP HM CF HM HM 20/200 HM HM HM CF 20/200 HM LP HM 20/400 1/200 CF 20/400 LP LP LP 20/50 20/70 20/50 20/60 20/100 20/60 20/400 3/200 20/60 3/200 LP CF 20/400 CF 20/50 20/100 LP 20/60 NLP CF 6/200 20/30 20/50 20/30 20/30 20/30 20/400 HM HM LP 20/70 4/200 HM 20/25 3/200 NLP HM CF 20/400 3/200 20/50 2/200 20/100 LP NLP HM NLP 10/200 HM CF 20/400 20/60 20/50 20/400 20/25 20/50 20/30 20/30 NLP NLP HM NLP Final VA Follow-up (months) 89.4 13.3 41.4 73.2 7.0 18.3 92.6 107.9 3.7 142.0 2.1 3.7 5.4 3.2 102.9 116.9 63.4 144.3 23.4 1.0 60.1 16.7 43.9 1.6 40.7 23.6 17.0 67.2 62.2 6.5 2.4

20/60 NLP NLP NLP NLP 20/100 20/200 HM 20/60 HM 20/200 NLP NLP 20/30 20/400 5/200 NLP HM HM 20/400 20/400 20/50

BLEB-ASSOCIATED ENDOPHTHALMITIS  LENG ET AL

20/30 20/50

20/50

20/200 20/400 20/30 20/30 20/70 20/30 20/30 20/50 20/40 CF 6/200 20/25 20/40 20/60 NLP 20/100

347

(continued on next page)

348

Table 5. (continued ) Age Case No. Sex Eye (years) 48 45 60 2 12 20 52 46 64 39 58 65 6 36 43 13 29 4 7 30 31 44 55 9 34 51 M M F F M F M F M F F M M M F M F F M M M M F F F M OS OS OD OS OD OD OD OD OD OD OS OD OS OD OD OD OS OS OS OS OD OD OS OS OD OS 72 69 78 72 63 67 81 90 62 65 81 68 71 51 85 77 93 99 57 81 81 90 79 86 79 39 Organism E. faecalis E. faecalis Corynebacterium sp. Moraxella sp. Moraxella sp. Moraxella sp. Moraxella sp. Moraxella sp. Pseudomonas aeruginosa P. aeruginosa P. aeruginosa P. aeruginosa Haemophilus inuenzae H. inuenzae Enterobacter aerogenes Serratia marcescens S. marcescens S. oralis, S. marcescens S. aureus, Moraxella sp. S. agalactiae, Proteus mirabilis Moraxella sp., coagulasenegative Staphylococcus H. inuenzae, coagulasenegative Staphylococcus Coagulase-negative Staphylococcus, Propionibacterium acnes E. faecalis, Corynebacterium sp. S. viridans, S. aureus Peptostreptococcus prevotii, 2 different species of coagulase-negative Staphylococcus No growth No growth No growth No growth No growth No growth No growth No growth Initial Intervention Tap/inject Vitrectomy Vitrectomy Tap/inject Tap/inject Tap/inject Tap/inject Vitrectomy Tap/inject Vitrectomy Vitrectomy Vitrectomy Tap/inject Tap/inject Tap/inject Tap/inject Vitrectomy Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Vitrectomy Vitrectomy Vitrectomy Baseline VA HM 20/60 20/30 20/80 HM 20/100 unavailable 20/40 20/30 20/70 4/200 20/100 20/50 20/25 20/40 20/80 20/100 20/50 20/25 4/200 CF 20/60 20/200 Presenting 1-Month 3-Month 6-Month 1-Year VA VA VA VA VA HM 20/80 HM CF LP 20/200 HM LP 20/100 LP LP LP 20/80 20/200 LP HM LP LP 3/200 HM LP CF 6/200 HM 20/70 20/50 20/70 HM 20/200 HM HM 20/60 LP LP LP 20/25 20/40 HM LP 2/200 LP 6/200 HM LP 20/200 HM HM NLP LP HM HM 20/40 HM 20/100 20/200 LP LP LP 20/25 20/200 LP NLP LP 20/100 20/40 HM 20/25 20/80 HM Final VA HM 1/200 20/60 20/25 HM 20/100 20/50 HM 20/25 NLP LP LP 20/25 20/30 20/200 NLP NLP NLP 20/70 LP HM 20/80 HM HM NLP LP Follow-up (months) 3.4 26.5 121.1 23.0 29.4 3.7 70.6 1.2 68.8 11.8 13.1 63.4 1.5 102.0 10.7 3.6 88.1 22.0 87.6 56.9 13.2 3.4 4.7 1.9 3.9 2.3 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2011  VOLUME 31  NUMBER 2

20/30 LP LP LP 20/200 NLP LP 20/200 NLP LP 20/80 LP LP

20/400 CF 20/30 LP Unavailable HM

NLP

3 19 25 40 42 57 68 69

F M F F M M M M

OS OS OS OD OD OS OS OS

52 76 68 78 54 75 81 92

Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject Tap/inject

20/30 20/40 20/60 20/80 1/200 20/30 20/400 20/80

HM HM HM HM HM 20/50 20/200 HM

CF 2/200 HM LP 2/200 20/400

CF HM 1/200 20/400 20/50

HM

HM

2/200 20/80 20/80 NLP LP 20/30 20/400 20/80

6.4 7.9 13.0 92.6 111.5 11.4 5.1 42.9

BLEB-ASSOCIATED ENDOPHTHALMITIS  LENG ET AL Follow-up (months)

349

Presenting 1-Month 3-Month 6-Month 1-Year VA VA VA VA VA

presentation, with the other 8 eventually undergoing evisceration or enucleation because of no light perception vision in painful eyes after a mean interval of 39 weeks (range, 3.9109; SD, 41) from the time of presentation. The mean intraocular pressure in eyes not eviscerated or enucleated at the nal follow-up was 12 mmHg (range, 032; SD, 6.5). The average number of glaucoma medications was 0.9 (range, 04; SD, 1.2). In 5 of the 63 eyes (7.9%) not enucleated, intraocular pressures were not controlled (.21 mmHg). During the study period, 14 eyes (20%) had inadequate intraocular pressure control on maximal medical therapy and underwent placement of a Baerveldt glaucoma drainage device. Only 1 of those eyes had an uncontrolled intraocular pressure at the nal follow-up (.21 mmHg). Discussion In this study, despite administration of appropriate intravitreal antibiotics, the VA outcomes were generally poor. In contrast to a previous study,31 the 24 eyes initially treated with PPV and intravitreal administration of antibiotics fared worse than those eyes that initially received initial vitreous tap with intravitreal injection of antibiotics, with only 8 PPV eyes (33%) achieving a nal VA $20/400. Conversely, the 45 eyes that received initial vitreous tap with intravitreal injection of antibiotics fared better with 29 (64%) having a nal VA $20/400; however, there was a selection bias uncovered in the study because those patients with worse presenting visions, and likely more virulent infections, being treated with initial PPV (67% of eyes initially treated with PPV vs. 16% of eyes initially treated with vitreous tap and intravitreal injection of antibiotics were light perception or no light perception at the initial presentation). The reported incidence of bleb-associated endophthalmitis after the use of mitomycin C in trabeculectomy surgery ranges from 2.1% to 2.6% per patient-year4,12 and 1.7% per patient-year in eyes that received 5-uorouracil.27 Additional studies show similar rates.3,10,16,17 In the current study, 68% of eyes that presented over a 12.5-year period with delayedonset bleb-associated endophthalmitis had been treated with either mitomycin C or 5-uorouracil at the time of ltering surgery. The current study was not designed to calculate the incidence of delayed-onset bleb-associated endophthalmitis or to determine if antibrotic use was associated with a greater rate of delayed-onset bleb-associated endophthalmitis. The spectrum of causative organisms in blebassociated endophthalmitis has been reported to differ from that of other associated forms of endophthalmitis.

Final VA

20/25 1/200 HM 20/50 NLP 20/40 NLP 20/30 1/200 HM 20/70 NLP

2.3 1.5 1.2 78.9 5.2

NLP 59
M, male; F, female; CF, counting ngers; LP, light perception; HM, hand motion; NLP, no light perception.

HM LP LP 5/200 NLP Vitrectomy Vitrectomy Vitrectomy Vitrectomy Evisceration 69 61 90 74 90 F M F M F 16 24 47 62 21 OS OD OS OS OS 20/30 1/200 20/70 20/40 20/60

Baseline VA

Initial Intervention

Organism

Age Case No. Sex Eye (years)

Table 5. (continued )

OD

83

No growth No growth No growth No growth Cultures not performed Cultures not performed

None

20/80

NLP

NLP

NLP

NLP

NLP

18.8

350

RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2011  VOLUME 31  NUMBER 2

Table 6. Organisms from Culture-Positive Cases of Delayed-Onset Bleb-Associated Endophthalmitis in Larger Reported Series from Bascom Palmer Eye Institute Culture Results Streptococcus species, n (%) Staphylococcus aureus, n (%) Coagulase-negative Staphylococcus, n (%) Gram-negative organisms, n (%) Culture negative, n (%)
*Signicantly different (x2 = 6.315, P = 0.0120).

Mandelbaum et al (19691984)30 17/30 (57)* 2/30 (7) 0 9/30 (30) 6/36 (17)

Kangas et al (19891995)29 15/31 0 7/31 7/31 1/32 (48) (23) (23) (3)

Current Study (1996 to July 2008) 20/67 8/67 12/67 19/67 12/69 (30)* (12) (18) (28) (17)

In endophthalmitis after cataract surgery, the less virulent Staphylococcus epidermidis species is the most common organism cultured, and VA outcomes are generally good.5,32,33 In delayed-onset bleb-associated endophthalmitis, more virulent organisms such as Streptococcus species and gram-negative organisms are more common, often with poor VA outcomes.12,29,30,34 In the current study, Streptococcus species and the gram-negatives were the most common organisms isolated (Table 3) and the VA outcomes were generally poor. All isolates were sensitive to either vancomycin and/or ceftazidime. Thus, continued use of these intravitreal antibiotics as the primary antimicrobial agents in the treatment of delayed-onset blebassociated endophthalmitis is supported. Comparing data from our institution with published reports dating from 1969 to the present (Table 6), a chi-square omnibus test comparing the 3 time periods found a signicant difference between the number of Streptococcus species infections (x2 = 7.199, P = 0.0273). No signicant difference was found when the 1969 to 1984 time period was compared with the 1989 to 1995 time period (x2 = 0.419, P = 0.5174) and the 1989 to 1995 time period was compared with the 1996 to 2008 time period (x2 = 3.172, P = 0.0749). However, when 1969 to 1984 was compared with 1996 to 2008, there was a signicant difference (x2 = 6.315, P = 0.0120), ultimately signifying that there were signicantly fewer Streptococcus species infections at our institution between 1969 and 1984 and the time period in the current study.

Nevertheless, the rates of endophthalmitis from S. aureus and gram-negative organisms have remained relatively stable. This downward trend in Streptococcus species infections may represent regional changes in microbial distribution over time.28 Whereas Streptococcus species was the predominant gram-positive organism isolated, Enterococcus species represented 7.5% of isolates. Even though all Enterococcus species were sensitive to vancomycin, there has been at least one case report of blebassociated endophthalmitis by vancomycin-resistant Enterococcus.35 Emergence of vancomycin-resistant organisms is a current and future concern. The most common gram-negative organism isolated in this study was Moraxella species (10%). In fact, the incidence of infections by this organism may be even higher, given that six of seven culture-negative cases of bleb-associated endophthalmitis in a recent study were actually conrmed to be Moraxella species by polymerase chain reaction analysis of aqueous humor samples.36 Although the culture-positive rate in this study was comparable to other larger studies (Table 7), novel ways are needed to increase the diagnostic speed, sensitivity, and specicity for these potentially devastating infections. The use of real-time polymerase chain reaction could potentially decrease the time to organism identication and could potentially help with the recognition of resistant organisms.37 During the study period, 52% of eyes achieved nal VAs $20/400. This treatment success rate is slightly

Table 7. Reported Series of Delayed-Onset Bleb-Associated Endophthalmitis Author Mandelbaum et al30 Ciulla et al20 Higginbotham et al4 Kangas et al29 Current study Total Number of Number of Positive Streptococcus Final VA $20/400, Streptococcus Species with Cases Cultures, n (%) Species, n (%) n (%) Final VA $20/400, n (%) 36 32 6 32 71 177 30/36 16/32 5/6 31/32 57/69 139/175 (83) (50) (83) (97) (83) (79) 17/30 5/16 2/5 15/31 20/67 59/149 (57) (31) (40) (48) (30) (40) 11/34 7/32 4/6 15/32 37/71 74/175 (32) (22) (67) (47) (52) (42) 3/16 (19) 0/5 (0) Data not available 6/15 (40) 8/20 (40) 17/56 (30)

BLEB-ASSOCIATED ENDOPHTHALMITIS  LENG ET AL

351

higher than a study performed at the same institution (47%) from 1989 to 1995.29 Other previously published studies (Table 7) demonstrated nal VAs $20/400 ranging from 22% to 67%.4,20,30 Overall, the generally poor VA outcomes in the current study supports the treatment outcomes in the current literature; however, little is known about the potential regional variation in microbial distributions and antibiotic sensitivities at other institutions. Thus, these data may not be applicable to other areas of the United States or the globe at large. Other limitations of this study include its retrospective nature. Additionally, there was no masking or randomization of treatment. A prospective trial with randomization to vitreous tap and intravitreal injection of antibiotics versus PPV and intravitreal injection of antibiotics would add to the ndings of this current descriptive study. In summary, delayed-onset bleb-associated endophthalmitis is a potentially devastating complication of ltering glaucoma surgery. Despite treatment with appropriate antibiotics in this study, VA outcomes are generally poor. Key words: bleb-associated endophthalmitis, endophthalmitis, glaucoma ltering surgery, intravitreal antibiotics, vitrectomy. References
1. Ciulla TA, Baker AS. Endophthalmitis following glaucoma ltering surgery. Int Ophthalmol Clin 1996;36:8796. 2. DeBry PW, Perkins TW, Heatley G, Kaufman P, Brumback LC. Incidence of late-onset bleb-related complications following trabeculectomy with mitomycin. Arch Ophthalmol 2002;120: 297300. 3. Hattenhauer JM, Lipsich MP. Late endophthalmitis after ltering surgery. Am J Ophthalmol 1971;72:10971101. 4. Higginbotham EJ, Stevens RK, Musch DC, et al. Bleb-related endophthalmitis after trabeculectomy with mitomycin C. Ophthalmology 1996;103:650656. 5. Katz LJ, Cantor LB, Spaeth GL. Complications of surgery in glaucoma. Early and late bacterial endophthalmitis following glaucoma ltering surgery. Ophthalmology 1985;92: 959963. 6. Muckley ED, Lehrer RA. Late-onset blebitis/endophthalmitis: incidence and outcomes with mitomycin C. Optom Vis Sci 2004;81:499504. 7. Phillips WB II, Wong TP, Bergren RL, Friedberg MA, Benson WE. Late onset endophthalmitis associated with ltering blebs. Ophthalmic Surg 1994;25:8891. 8. Sharan S, Trope GE, Chipman M, Buys YM. Late-onset bleb infections: prevalence and risk factors. Can J Ophthalmol 2009;44:279283. 9. Shigeeda T, Tomidokoro A, Chen YN, Shirato S, Araie M. Long-term follow-up of initial trabeculectomy with mitomycin C for primary open-angle glaucoma in Japanese patients. J Glaucoma 2006;15:195199. 10. Kuang TM, Lin YC, Liu CJ, Hsu WM, Chou CK. Early and late endophthalmitis following trabeculectomy in a Chinese population. Eur J Ophthalmol 2008;18:6670.

11. Ashkenazi I, Melamed S, Avni I, Bartov E, Blumenthal M. Risk factors associated with late infection of ltering blebs and endophthalmitis. Ophthalmic Surg 1991;22:570574. 12. Greeneld DS, Suner IJ, Miller MP, Kangas TA, Palmberg PF, Flynn HW Jr. Endophthalmitis after ltering surgery with mitomycin. Arch Ophthalmol 1996;114:943949. 13. Mochizuki K, Jikihara S, Ando Y, Hori N, Yamamoto T, Kitazawa Y. Incidence of delayed onset infection after trabeculectomy with adjunctive mitomycin C or 5-uorouracil treatment. Br J Ophthalmol 1997;81:877883. 14. Soltau JB, Rothman RF, Budenz DL, et al. Risk factors for glaucoma ltering bleb infections. Arch Ophthalmol 2000;118: 338342. 15. Susanna R Jr, Takahashi W, Nicolela M. Late bleb leakage after trabeculectomy with 5-uorouracil or mitomycin C. Can J Ophthalmol 1996;31:296300. 16. Akova YA, Bulut S, Dabil H, Duman S. Late bleb-related endophthalmitis after trabeculectomy with mitomycin C. Ophthalmic Surg Lasers 1999;30:146151. 17. Parrish R, Minckler D. Late endophthalmitisltering surgery time bomb? Ophthalmology 1996;103:11671168. 18. Caronia RM, Liebmann JM, Friedman R, Cohen H, Ritch R. Trabeculectomy at the inferior limbus. Arch Ophthalmol 1996; 114:387391. 19. Brown RH, Yang LH, Walker SD, Lynch MG, Martinez LA, Wilson LA. Treatment of bleb infection after glaucoma surgery. Arch Ophthalmol 1994;112:5761. 20. Ciulla TA, Beck AD, Topping TM, Baker AS. Blebitis, early endophthalmitis, and late endophthalmitis after glaucomaltering surgery. Ophthalmology 1997;104:986995. 21. Ayyala RS, Bellows AR, Thomas JV, Hutchinson BT. Bleb infections: clinically different courses of blebitis and endophthalmitis. Ophthalmic Surg Lasers 1997;28: 452460. 22. Jampel HD, Quigley HA, Kerrigan-Baumrind LA, Melia BM, Friedman D, Barron Y. Risk factors for late-onset infection following glaucoma ltration surgery. Arch Ophthalmol 2001; 119:10011008. 23. Lehmann OJ, Bunce C, Matheson MM, et al. Risk factors for development of post-trabeculectomy endophthalmitis. Br J Ophthalmol 2000;84:13491353. 24. Poulsen EJ, Allingham RR. Characteristics and risk factors of infections after glaucoma ltering surgery. J Glaucoma 2000;9: 438443. 25. Beyer TL, Vogler G, Sharma D, ODonnell FE Jr. Protective barrier effect of the posterior lens capsule in exogenous bacterial endophthalmitisan experimental primate study. Invest Ophthalmol Vis Sci 1984;25:108112. 26. Waheed S, Ritterband DC, Greeneld DS, Liebmann JM, Seedor JA, Ritch R. New patterns of infecting organisms in late bleb-related endophthalmitis: a ten year review. Eye 1998; 12(pt 6):910915. 27. Wolner B, Liebmann JM, Sassani JW, Ritch R, Speaker M, Marmor M. Late bleb-related endophthalmitis after trabeculectomy with adjunctive 5-uorouracil. Ophthalmology 1991; 98:10531060. 28. Major JC Jr, Engelbert M, Flynn HW Jr, Miller D, Smiddy WE, Davis JL. Staphylococcus aureus endophthalmitis: antibiotic susceptibilities, methicillin resistance, and clinical outcomes. Am J Ophthalmol 2010;149:278283. 29. Kangas TA, Greeneld DS, Flynn HW Jr, Parrish RK II, Palmberg P. Delayed-onset endophthalmitis associated with conjunctival ltering blebs. Ophthalmology 1997;104: 746752.

352

RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2011  VOLUME 31  NUMBER 2
34. Song A, Scott IU, Flynn HW Jr, Budenz DL. Delayed-onset bleb-associated endophthalmitis: clinical features and visual acuity outcomes. Ophthalmology 2002;109:985991. 35. Tang CW, Cheng CK, Lee TS. Community-acquired blebrelated endophthalmitis caused by vancomycin-resistant enterococci. Can J Ophthalmol 2007;42:477478. 36. Cornut PL, Chiquet C, Bron A, et al. Microbiologic identication of bleb-related delayed-onset endophthalmitis caused by moraxella species. J Glaucoma 2008;17: 541545. 37. Oechsler RA, Feilmeier MR, Ledee DR, et al. Utility of molecular sequence analysis of the ITS rRNA region for identication of Fusarium spp. from ocular sources. Invest Ophthalmol Vis Sci 2009;50:22302236.

30. Mandelbaum S, Forster RK, Gelender H, Culbertson W. Late onset endophthalmitis associated with ltering blebs. Ophthalmology 1985;92:964972. 31. Busbee BG, Recchia FM, Kaiser R, Nagra P, Rosenblatt B, Pearlman RB. Bleb-associated endophthalmitis: clinical characteristics and visual outcomes. Ophthalmology 2004;111: 14951503; discussion 1503. 32. Lalwani GA, Flynn HW Jr, Scott IU, et al. Acute-onset endophthalmitis after clear corneal cataract surgery (1996 2005). Clinical features, causative organisms, and visual acuity outcomes. Ophthalmology 2008;115:473476. 33. Puliato CA, Baker AS, Haaf J, Foster CS. Infectious endophthalmitis. Review of 36 cases. Ophthalmology 1982; 89:921929.