Purpose
To report the clinical settings, antibiotic susceptibilities, and outcomes of endophthalmitis caused by Streptococcus species.
Design
Retrospective, observational case series.
Methods
Single-center study evaluating all patients with culture-positive endophthalmitis caused by Streptococcus species between January 1, 2000, and December 31, 2011.
Results
Study criteria were met by 63 patients. The most common clinical settings were bleb associated (n = 17; 27%), after intravitreal injection (n = 16; 25%), and after cataract surgery (n = 13; 21%). The isolates were Streptococcus viridans (n = 47; 71%), Streptococcus pneumoniae (n = 13; 21%), and β-hemolytic Streptococci (n = 5; 8%). Sixty (95%) of 63 isolates were susceptible to vancomycin, 47 (98%) of 48 isolates were susceptible to ceftriaxone (third-generation cephalosporin), and 57 (93%) of 61 isolates were susceptible to levofloxacin (third-generation fluoroquinolone). Between the first and second half of the study, the minimal inhibitory concentration of antibiotics required to inhibit 90% of isolates increased by 1.5-fold for ceftriaxone and 2-fold for levofloxacin and remained the same for vancomycin. Initial treatment was vitreous tap (49; 78%) or pars plana vitrectomy (14; 22%); all received intravitreal antibiotics. Visual acuity outcomes were variable: best-corrected visual acuity was 20/400 or better in 16 (25%) patients and worse than 20/400 in 47 (75%) patients. Evisceration or enucleation was performed in 16 (25%) patients.
Conclusions
Streptococcus isolates generally had high susceptibility rates to commonly used antibiotics. Higher antibiotic minimal inhibitory concentrations were required to inhibit 90% of isolates in vitro in the second half of the study period compared with the first half. Despite prompt treatment, most patients had poor outcomes.
Endophthalmitis is a severe sight-threatening ocular infection caused by a variety of microbes. Streptococcus is a common cause of endophthalmitis after glaucoma filtering surgery and endogenous bacterial endophthalmitis. Streptococcus is also the second most common genus of organisms identified in endophthalmitis after cataract surgery and after intravitreal anti–vascular endothelial growth factor injection. In previous reports, endophthalmitis cases caused by Streptococcus species have poorer visual outcomes compared with cases caused by coagulase-negative Staphylococcus species, the most common cause of postoperative endophthalmitis overall.
A previous study from 1977 through 1990 from our institution reported clinical settings, antibiotic susceptibilities, and treatment outcomes for endophthalmitis caused by Streptococcus species. Since then, there has been a dramatic rise in the number of intravitreal injections performed. Additionally, Enterococcus faecalis , which formerly was classified as a group D Streptococcus species, has been reclassified under the distinct Enterococcus genus. There have also been reports of increased antibiotic nonsusceptibility among Streptococcus species and the spread of virulence factors between the species. The purpose of this study was to provide an update on the clinical settings, antibiotic susceptibilities and minimal inhibitory concentrations (MICs), management strategies, and visual acuity (VA) outcomes in a more recent series of culture-proven streptococcal endophthalmitis cases from our institution. Based on a PubMed literature search, the current study is the largest series of culture-positive endophthalmitis cases caused by streptococcal species.
Methods
The study protocol for a retrospective review of medical and microbiology records for all patients treated at Bascom Palmer Eye Institute with vitreous fluid culture-proven endophthalmitis caused by Streptococcus species between January 1, 2000, and December 31, 2011, was approved by the Institutional Review Board of the University of Miami Miller School of Medicine Medical Sciences Subcommittee for the Protection of Human Subjects. Isolates were identified using standard microbiological procedures. Shifting trends in in vitro MIC (measured in micrograms per milliliter) were analyzed using the E test (Biomeriuex, Raleigh, North Carolina, USA). The treatment strategies were determined by the individual treating physicians and did not follow a standardized protocol.
Statistical analysis was performed using Statistical Package for the Social Sciences software (SPSS, Inc, Chicago, Illinois, USA). Snellen best-corrected visual acuity (BCVA) was converted to logarithm of minimal angle of resolution (logMAR) equivalents and BCVA of counting fingers, hand movements, light perception, and no light perception were assigned logMAR values of 1.85, 2.3, 2.7, and 3.0, respectively, as previously described. The logMAR BCVA is presented as mean ± standard deviation. BCVA measurements obtained at presentation and last follow-up visit were analyzed based on the clinical setting and Streptococcus species group using 1-way analysis of variance with Tukey post hoc analysis. The Student t test was used to compare the VA outcomes between patients who received different initial treatment regimens. The Fisher exact test was used to compare the number of additional treatments and enucleations or eviscerations between different initial treatment regimens. The Fisher exact test also was used to compare the findings of the current study and the previously reported study on endophthalmitis resulting from Streptococcus from our institution. A P value of less than .05 was considered statistically significant.
Results
Patient Demographics, Ocular and Medical History, and Presentation
During the 12-year study period, 66 patients with streptococcal endophthalmitis met the study entry criteria. Three patients were excluded from the current study because of missing medical records. Also included in this study were 11 patients who were described previously as part of a case series detailing an outbreak of Streptococcus viridans resulting from contaminated bevacizumab intravitreal injections. The mean age of the patients was 66.9 years (median, 73 years; range, 3 to 92 years). There were 37 (58.7%) males and 23 (37%) right eyes. Thirty-five (56%) patients had a history of hypertension and 13 (21%) patients had a history of diabetes mellitus. Six (10%) patients were immunocompromised as a result of chronic systemic corticosteroids (n = 3), chronic systemic chemotherapeutic agents (n = 1), asplenia (n = 1), or genetic abnormalities (n = 1). Patients had an ocular history of glaucoma (n = 26), age-related macular degeneration (n = 17), proliferative diabetic retinopathy (n = 1), Fuchs endothelial dystrophy (n = 1), Peter anomaly (n = 1), and thyroid eye disease (n = 1).
The clinical setting for each case is summarized in Table 1 . The most common clinical scenario of streptococcal endophthalmitis in the current series was bleb associated (n = 17; 27%). The first case of endophthalmitis occurring after intravitreal injection in the current series occurred in 2007. One patient each from the cataract surgery and penetrating keratoplasty or keratoprosthesis surgery groups demonstrated endophthalmitis 1 day after removal of a suture. One patient in the penetrating keratoplasty or keratoprosthesis group had a suture abscess.
| Clinical Scenario | No. of Patients (%) | Age Range (y) | Time between Event and Diagnosis | No. of Streptococcus Organisms (% of Clinical Scenario) | ||
|---|---|---|---|---|---|---|
| β-Hemolytic | Pneumoniae | Viridans | ||||
| Bleb associated | 17 (27) | 27 to 92 | 2 to 480 m | 2 (12) | 4 (23) | 11 (65) |
| After intravitreal injection | 16 (25) | 59 to 88 | 1 to 8 d | 0 (0) | 1 (6) | 15 (94) |
| After cataract surgery | 13 (21) | 68 to 91 | 3 d to 1.5 m | 1 (8) | 4 (31) | 8 (61) |
| After PKP/K-Pro surgery a | 8 (13) | 3 to 87 | 1 d to 157.5 m | 2 (25) | 1 (12) | 5 (63) |
| Ruptured globe | 4 (6) | 13 to 51 | 1 d to 2 d | 0 (0) | 0 (0) | 4 (100) |
| Miscellaneous b | 3 (5) | 4 to 63 | 6 d to 57 m | 0 (0) | 3 (100) | 0 (0) |
| Endogenous | 2 (3) | 58 to 67 | — c | 0 (0) | 0 (0) | 2 (100) |
| Total | 63 (100) | 3 to 92 | 1 d to 480 m | 5 (8) | 13 (21) | 45 (71) |
a Includes 6 PKP and 2 K-Pro patients.
b Includes 2 postoperative glaucoma drainage device patients and 1 perforated corneal ulcer patient.
c Time between the causative event and diagnosis of endophthalmitis could not be determined.
At the presenting visit, 60 (95%) of the 63 patients reported pain and 50 (79%) had a hypopyon. A view of the posterior pole was unobtainable in 57 (91%) of 63 patients because of severe anterior segment inflammation and media opacities. Vitritis was noted in all patients in whom a view of the posterior pole could be obtained.
Clinical Management
The initial and subsequent clinical management details of patients are summarized in Table 2 . Initial treatment consisted of a vitreous tap and intravitreal antibiotics in 49 (78%) of 63 patients and pars plana vitrectomy (PPV) with intravitreal antibiotics in 14 (22%) of 63 patients. The highest rates of additional therapeutic interventions were among the following clinical settings: after intravitreal injection (14 of 16; 89%), after penetrating keratoplasty or keratoprosthesis (6 of 8; 75%), and the miscellaneous group (2 of 3; 67%). Of the 49 patients who underwent a vitreous tap and intravitreal antibiotics as initial treatment, 5 (10%) underwent additional intravitreal antibiotics only, 7 (14%) underwent additional intravitreal antibiotics followed by a PPV on a different date, and 15 (31%) underwent a PPV with intravitreal antibiotics only. Two (14%) of 14 patients who underwent initial treatment of PPV with intravitreal antibiotics underwent subsequent vitreous tap and intravitreal antibiotics on a different date. Evisceration or enucleation was performed in 14 (29%) of 49 patients initially treated with a vitreous tap and intravitreal antibiotics compared with 2 (14%) of 14 patients initially treated with PPV and intravitreal antibiotics ( P = .49). Of the patients who received initial treatment with vitreous tap and intravitreal antibiotics, 34 (69%) of 49 patients had additional interventions, compared with 4 (29%) of 14 who were treated initially with PPV and intravitreal antibiotics ( P = .01).
| Clinical Scenario | No. of Initial Treatment (%) | No. of Additional Treatment (%) | ||||
|---|---|---|---|---|---|---|
| Vitreous Tap + Antibiotics | PPV + Antibiotics | Dexamethasone | Vitreous Tap + Antibiotics | PPV + Antibiotics | Enucleation or Evisceration | |
| Bleb association (n = 17) | 12 (71) | 5 (29) | 16 (94) | 2 (12) | 6 (35) | 2 (12) |
| After intravitreal injection (n = 16) | 16 (100) | 0 (0) | 15 (94) | 7 (44) | 10 (63) | 7 (44) |
| After cataract surgery (n = 13) | 10 (77) | 3 (23) | 12 (92) | 5 (39) | 1 (8) | 2 (15) |
| After PKP/K-Pro surgery a (n = 8) | 5 (63) | 3 (37) | 8 (100) | 4 (50) | 2 (25) | 1 (13) |
| Ruptured globe (n = 4) | 2 (50) | 2 (50) | 3 (75) | 0 (0) | 1 (25) | 1 (25) |
| Miscellaneous b (n = 3) | 2 (67) | 1 (33) | 2 (67) | 0 (0) | 0 (0) | 2 (67) |
| Endogenous (n = 2) | 2 (100) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 1 (50) |
| Total (n = 63) | 49 (78) | 14 (22) | 56 (89) | 18 (29) | 20 (32) | 16 (25) |
a Includes 6 PKP and 2 K-Pro patients.
b Includes 2 postoperative glaucoma drainage device patients and 1 perforated corneal ulcer patient.
Vancomycin was used for intravitreal antibiotic treatment in all patients and a second intravitreal antibiotic (ceftazidime or amikacin) was used in 61 (97%) of 63 patients. The only 2 patients who did not receive a second antibiotic were part of an outbreak of vancomycin-susceptible S. viridans because of contaminated bevacizumab intravitreal injections and presentation after identification of the causative organism in other patients. Additionally, 56 (89%) of 63 patients were started on intravitreal dexamethasone as part of their initial treatment. All patients were started on topical antibiotic drops: 51 (81%) of 63 on fortified vancomycin and a second antibiotic (fortified tobramycin, fluoroquinolone, cephalosporin, or amikacin), 4 (6.3%) on fortified vancomycin alone, 4 (6.3%) on fortified tobramycin alone, 2 (3.2%) on a fluoroquinolone and cephalosporin, and 2 (3.2%) on fortified tobramycin and another antibiotic. A topical steroid drop was started within 48 hours of the initial treatment in 59 (94%) of 63 patients.
Microbiology and Antibiotic Susceptibility
The microbiology findings and antibiotic susceptibilities are summarized in Table 3 . In the current study, Streptococcus species were grouped as previously described by Mao and associates: (1) S. viridans , (2) Streptococcus pneumoniae , and (3) β-hemolytic Streptococci . The S. viridans group (45 of 63; 71%) was the most common organism isolated overall and in all the clinical scenarios, except for the miscellaneous group in which S. pneumoniae was isolated in all 3 patients. A single Streptococcus species was identified in 58 (92%) of 63 patients. The 5 polymicrobial cultures included: (1) an additional Streptococcus species, (2) an additional Streptococcus species and Bacteroides distasonis , (3) Staphylococcus aureus , (4) a coagulase-negative Staphylococcus species, and (5) a coagulase-negative Staphylococcus species.
| Vancomycin | Ceftriaxone | Levofloxacin | |
|---|---|---|---|
| Streptococcus organism | |||
| No. of isolates susceptible to antibiotic/no. of isolates tested (%) | |||
| β-hemolytic streptococci | 5/5 (100) | 3/3 (100) | 5/5 (100) |
| Streptococcus pneumoniae | 13/13 (100) | 9/10 (90) | 13/13 (100) |
| Streptococcus viridans group | 42/45 (93) | 38/38 (100) | 39/43 (91) |
| Total | 60/63 (95) | 50/51 (98) | 57/61 (93) |
| MIC (μg/mL) of antibiotic required to inhibit 90% of streptococcal isolates (range) | |||
| Time period | |||
| 2000 through 2005 | 1 (0.25 to 1.0) | 0.5 (0.012 to 0.60) | 1.5 (0.38 to 2.0) |
| 2006 through 2011 | 1 (0.38 to 1.5) | 0.75 (0.016 to 1.5) | 3 (0.25 to 32) |
Sixty (95%) of 63 streptococcal isolates were susceptible to vancomycin; 47 (98%) of 48 were susceptible to ceftriaxone, a third-generation cephalosporin; and 57 (93%) of 61 were susceptible to levofloxacin, a third-generation fluoroquinolone ( Table 3 ). The MIC required to inhibit 90% of streptococcal isolates for vancomycin was 1 μg/mL (range, 0.25 to 1 μg/mL) during the first 6 years (2000 through 2005) and 1 μg/mL (range, 0.38 to 1.5 μg/mL) during the second 6 years (2006 through 2011) of the study period. Between the first and second halves of the study period, the MIC required to inhibit 90% of isolates increased by 1.5-fold for ceftriaxone and 2-fold for levofloxacin ( Table 3 ).
Clinical Outcomes
The median follow-up period was 15.7 months (range, 1 day to 9.8 years). The BCVA at presentation and final follow-up visit for the different clinical settings and streptococcal isolates are summarized in Table 4 . There was no difference in BCVA at presentation among the different clinical settings ( P = .23) or among the different Streptococcus species groups ( P = .36). The bleb-associated patients had better VA outcomes compared with those who had received an intravitreal injection ( P = .001) and the miscellaneous group patients ( P = .008). Additionally, the cataract patients had better VA outcomes compared with the intravitreal injection ( P = .03) and the miscellaneous group ( P = .01) patients. There was no difference in the visual acuity outcomes among the 3 Streptococcus organism groups ( P = .87).
| Clinical Scenario | Presenting VA | VA Outcomes | ||||
|---|---|---|---|---|---|---|
| Mean VA ± SD (logMAR) | Mean Snellen VA Equivalent | ≥20/400, No. Patients (%) | <20/400, No. Patients (%) | Mean VA ± SD (logMAR) | Mean Snellen VA Equivalent | |
| Bleb associated (n = 17) | 2.18 ± 0.27 | 20/3000 | 8 (47) | 9 (53) | 1.74 ± 0.97 | 20/1100 |
| After intravitreal injection (n = 16) | 2.44 ± 0.49 | 20/5500 | 1 (6) | 15 (94) | 2.54 ± 0.79 | 20/6900 |
| After cataract surgery (n = 13) | 2.43 ± 0.27 | 20/5400 | 4 (30) | 9 (70) | 1.94 ± 1.17 | 20/1700 |
| After PKP/K-Pro surgery a (n = 8) | 2.54 ± 0.56 | 20/6900 | 2 (25) | 6 (75) | 2.15 ± 0.94 | 20/2800 |
| Ruptured globe (n = 4) | 2.40 ± 0.20 | 20/5000 | 1 (25) | 3 (75) | 2.25 ± 0.76 | 20/3600 |
| Miscellaneous b (n = 3) | 2.70 ± 0.00 | 20/10,000 | 0 (0) | 3 (100) | 3.00 ± 0.00 | 20/20,000 |
| Endogenous (n = 2) | 2.50 ± 0.28 | 20/6300 | 0 (0) | 2 (100) | 2.85 ± 0.21 | 20/14,000 |
| Streptococcal isolates | ||||||
| β-hemolytic Streptococci (n = 5) | 2.25 ± 0.66 | 20/3600 | 2 (40) | 3 (60) | 2.16 ± 1.19 | 20/2900 |
| Streptococcus pneumoniae (n = 13) | 2.22 ± 0.62 | 20/3300 | 5 (38) | 8 (62) | 2.04 ± 1.05 | 20/2200 |
| Streptococcus viridans group (n = 45) | 2.45 ± 0.47 | 20/5600 | 9 (24) | 36 (76) | 2.20 ± 0.94 | 20/3200 |
| Total (n = 63) | 2.39 ± 0.51 | 20/4900 | 16 (25) | 47 (75) | 2.16 ± 0.97 | 20/2900 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
