CLINICAL SCIENCE Endophthalmitis and Concurrent or Delayed-Onset Rhegmatogenous Retinal Detachment Managed With Pars Plana Vitrectomy, Intravitreal Antibiotics, and Silicone Oil Vivek Pravin Dave, MD, FRCS; Avinash Pathengay, FRCS; Nidhi Relhan, MD; Pranjali Sharma, MBBS; Subhadra Jalali, MD; Rajeev Reddy Pappuru, MS; Mudit Tyagi, MS; Raja Narayanan, MD; Jay Chhablani, MD; Taraprasad Das, MD; Harry W. Flynn Jr., MD INTRODUCTION BACKGROUND AND OBJECTIVE: The current study describes the treatment outcomes in patients with endophthalmitis and concurrent or delayed-onset retinal detachment managed with pars plana vitrectomy, intravitreal antibiotics, and silicone oil. PATIENTS AND METHODS: In this noncomparative, retrospective case series, the medical records of patients diagnosed with endophthalmitis and retinal detachment from January 1991 through December 2014 at a tertiary eye care center in South India were reviewed. All patients received silicone oil for the management of retinal detachment either concurrently or during follow-up treatment. RESULTS: A total of 93 patients were included in the current study. Retinal detachment was diagnosed at presentation in 20 of 93 patients (21.5%) (concurrent group: Group 1) and during follow-up in the remaining 73 of 93 patients (78.5%) (delayed-onset group: Group 2). In Group 1, the initial treatment consisted of vitrectomy, intravitreal antibiotics, and silicone oil injection in 19 of 20 patients. In Group 2, patients did not receive silicone oil during initial treatment but underwent silicone oil injection during subsequent surgery for repair of retinal detachment. Rates of complete retinal reattachment and visual acuity of 20/400 or better were 73.7% and 30.0%, respectively, in Group 1 and 98.5% and 39.7%, respectively, in Group 2. The median visual acuity at last follow-up in 44 eyes undergoing silicone oil removal was 20/100 (logmar 0.7), whereas in the remaining 49 eyes that did not undergo silicone oil removal, median visual acuity was 20/2000 (logmar 2.0). CONCLUSION: In these patients with endophthalmitis with concurrent or delayed-onset retinal detachment, the use of silicone oil can be a useful adjunct. [Ophthalmic Surg Lasers Imaging Retina. 2017;48:546-551.] In patients with endophthalmitis, the occurrence of rhegmatogenous retinal detachment (RD) is uncommon and is generally associated with poor visual outcomes. Management of such patients can pose a surgical challenge. While managing the endophthalmitis, RD can be identified at the time of presentation, during vitrectomy, or in the postoperative period. Indirect ophthalmoscopy, echography, or direct visualization under microscope may confirm the diagnosis. Patients with endophthalmitis and concurrent or delayed-onset RD may have very poor anatomic and visual outcomes. 1-4 In the Endophthalmitis Vitrectomy Study (EVS), patients with RD at initial diagnosis of endophthalmitis were excluded. However, 35 of 420 patients overall (8.3%) developed RD during followup. The rate of RD among group of patients undergoing vitrectomy, needle vitreous aspiration, and mechanical vitreous biopsy were 7.8%, 11%, and 8%, respectively. In the EVS patients with available data, final visual acuity (VA) of 20/40 or greater was reported in eight of 30 patients with RD (27%) compared to 201 of 364 patients (55%) without RD. 5 The overall visual outcomes in this subgroup of the EVS were From Srimati Kannuri Santhamma Centre for Vitreoretinal Diseases, LV Prasad Eye Institute, KAR Campus, Hyderabad, India (VPD, NR, SJ, RRP, MT, RN, JC, TD); LV Prasad Eye Institute, GMR Varalakshmi Campus, Visakhapatnam, India (AP); and the Department of Ophthalmology, Bascom Palmer Eye Institute University of Miami, Miller School of Medicine, Miami (HWF, NR). 2017 Dave, Pathengay, Relhan, et al.; licensee SLACK Incorporated. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (https://creativecommons.org/licenses/by-nc/4.0). This license allows users to copy and distribute, to remix, transform, and build upon the article non-commercially, provided the author is attributed and the new work is non-commercial. Originally submitted January 13, 2017. Revision received February 14, 2017. Accepted for publication March 1, 2017. This study is supported in part by NIH Center Core Grant P30EY014801 and a Research to Prevent Blindness Unrestricted Grant. The funding organizations had no role in the design or conduct of this research. The authors report no relevant financial disclosures. Address correspondence to Avinash Pathengay, FRCS, Retina and Uveitis Service, GMR Varalakshmi Campus, LV Prasad Eye Institute, Visakhapatnam, India; email: avinash@lvpei.org. doi: 10.3928/23258160-20170630-05 546 Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina
TABLE 1 Demographic Profile and Baseline Characteristics of the Patients With Endophthalmitis and Concurrent or Delayed-Onset Retinal Detachment All Patients Group 1 (n = 20) Group 2 (n = 73) Age in Years 28.8 ± 18.8 32.2 ± 17.6 27.9 ± 19.1 (Mean ± SD) (median: 24; range: 2 to 79) Gender 76 males, 17 females 16 males, 4 females 60 males,13 females Lens Status at Presentation Pseudophakic 15 of 93 patients (16.1%) 8 of 20 patients (40.0%) 7 of 73 patients (9.6%) Aphakic 25 of 93 patients (26.9%) 10 of 20 patients (20.0%) 15 of 73 patients (20.5%) Phakic 53 of 93 patients (57.0%) 8 of 20 patients (40.0%) 45 of 73 patients (61.6%) Etiology of Endophthalmitis Posttraumatic 63 of 93 patients (67.7%) 15 of 20 patients (75.0%) 48 of 73 patients (65.8%) Postcataract 16 of 93 patients (19.4%) 3 of 20 patients (15.0%) 13 of 73 patients (17.8%) Endogenous 5 of 93 patients (5.4%) 1 of 20 patients (5.0%) 4 of 73 patients (5.5%) Others (associated with) Penetrating keratoplasty 2 0 2 Scleral buckle 3 1 2 Corneal ulcer 1 0 1 Intravitreal triamcinolone Chronic post-cataract surgery Comorbidities 1 0 1 2 0 2 Corneal laceration 5 1 4 Intraocular foreign body 18 11 7 Time to Develop RD (Weeks) 9.0 ± 41.9 (median: 2) At presentation 11.5 ± 47.1 (Mean ± SD) (range: 0 to 384) Visual Acuity at Presentation 1.73 logmar ± 0.41 logmar 1.93 logmar ± 0.28 logmar 1.77 logmar ± 0.44 logmar logmar (Snellen) (20/1,074) (20/1,700) (20/1,177) Group 1 (RD at presentation [concurrent RD]); Group 2 (RD during follow-up [delayed-onset RD]) RD = retinal detachment; SD = standard deviation reported to be poor, with more than half of patients with RD (16 of 30) achieving visual acuity worse than 5/200 despite a high anatomic success rate of 78% (18 of 23 patients). 5 In the setting of endophthalmitis, RD developing after vitrectomy surgery has been reported to range from 4.6% to 16% 1,5-7 (compared to 5% in vitrectomy performed for non-endophthalmitis cases 1,5,6 ). Risk factors associated with poor visual outcomes include infection with more virulent organisms, open globe injuries, retained intraocular foreign body, history of posterior capsular rupture, or choroidal detachment at the time of initial diagnosis. 6 The severity of endophthalmitis and media clarity may direct the management strategy (pars plana vitrectomy [PPV] + intravitreal antibiotics versus vitreous biopsy + intravitreal antibiotics) and timing of surgical intervention. In the current study, the demographic profiles, clinical features, causative organisms, and treatment outcomes are described in patients with endophthalmitis and RD. PATIENTS AND METHODS Using the diagnosis code endophthalmitis in the patient database of the Medical Records Department at the LV Prasad Eye Institute, Hyderabad, India, we July 2017 Vol. 48, No. 7 547
TABLE 2 Initial Treatment and Anatomical and Functional Outcomes for Group 1 Versus Group 2 Initial Treatment Group 1 (n = 20) Group 2 (n = 73) PPV + intravitreal antibiotics + SOI 19 of 20 patients (95%) 4 of 73 patients (5.5%) PPV + intravitreal antibiotics 1 of 20 patients (5%) 55 of 73 patients (75.3%) Vitreous tap + intravitreal antibiotics None 14 of 73 patients (19.2%) PPV + SOI (During Follow-Up) 1 of 20 patients 73 of 73 patients Recurrent RD 7 of 20 patients 22 of 73 patients SOR 9 of 20 patients 35 of 73 patients Anatomical Success a (Complete Retinal Attachment at Last Follow-Up) Functional Success b (Visual Acuity 20/400 at Last Follow-Up) 14 of 19 patients* (73.7%) 66 of 67 patients** (98.5%) 6 of 20 patients (30%) 29 of 73 patients (39.7%) Follow-Up Duration (Months) 36.8 ± 53.7 (Range: 4 to 168) 22.6 ± 31.6 (Range: 4 to 180) Group 1 (RD at presentation [concurrent RD]); Group 2 (RD during follow-up [delayed-onset RD]) a = P value of.001; b = P value of.59; * = data available for 19 patients; ** = data available for 67 patients PPV = pars plana vitrectomy; SOI = silicone oil injection; RD = retinal detachment; SOR = silicone oil removal identified patients with endophthalmitis between January 1991 and December 2014. The current retrospective study is a noncomparative, consecutive case series of patients diagnosed with both endophthalmitis and RD. There was no prospective study protocol for directing treatment, but the individual physician made the decision for initial and subsequent treatments. The study adhered to the tenets of the Declaration of Helsinki. An institutional review board approved the current study. An informed consent for the research was obtained from the patients at the time of first visit to the hospital. Endophthalmitis was defined by severe inflammation of the ocular tissues and fluids characterized clinically by combination of signs and symptoms including ocular pain, decreased vision, eyelid edema, conjunctival congestion, chemosis, anterior segment inflammation, hypopyon, vitritis, and decreased red reflex. 6 Etiologies of endophthalmitis were classified into postoperative, posttraumatic, and endogenous. Diagnosis of RD was made preoperatively (by fundus examination and/or echography) or intraoperatively under microscopic viewing. The patients were classified into one of two groups: Group 1 (patients with RD at presentation [concurrent RD group]) and Group 2 (patients who developed RD during follow-up [delayed-onset RD group]). Anatomical success was defined as retina attached completely at the last visit. Functional success was defined as the best-corrected VA of 20/400 or better at the last visit. The data entry into an excel spreadsheet and statistical analyses were performed using MedCalc version 12.2.1.0 statistical software (MedCalc, Ostend, Belgium). Parametric continuous data were analyzed using the paired t-test, non-parametric continuous data were analyzed by the Wilcoxon signed rank test, and the association between categorical variables were analyzed by the Chi-square test. A P value of less than.05 was assigned as statistically significant. Appropriate confidence intervals were calculated wherever deemed necessary. RESULTS In a database of 5,924 patients with endophthalmitis, 443 had concurrent RD. Of these 443 patients, 289 (65.2%) were deemed inoperable either due to no light perception at presentation, phthisis bulbi, or very poor visual prognosis. The remaining 154 patients (34.8%) underwent surgical management for RD while still receiving intravitreal antibiotics for endophthalmitis. Among these, 93 patients (21.0%) with endophthalmitis met the inclusion criteria (having received silicone oil for the management of RD immediately or during follow-up and follow-up duration of 4 months or longer). The current study included 93 patients (76 males, 17 females) who ranged in age from 2 years to 79 years (mean age: 28.8 years ± 18.8 years). RD was diagnosed at presentation in 20 of 93 patients (21.5%) (Group 1) and during follow-up in the 548 Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina
TABLE 3 Visual Acuity at Presentation and Last Follow-Up in Eyes With Endophthalmitis and Retinal Detachment Visual Acuity At Presentation At Last Follow-Up Worse than 5/200 Group 1 (n = 20) Group 2 (n = 73) Group 1 (n = 20) Group 2 (n = 73) 19 of 20 patients (95.0%) 65 of 73 patients (89.0%) 13 of 20 patients (65.0%) 40 of 73 patients (54.8%) 5/200 to 20/400 0 of 20 patients (0%) 0 of 73 patients (0%) 1 of 20 patients (5.0%) 4 of 73 patients (5.5%) Better than 20/400 1 of 20 patients (5.0%) 8 of 73 patients (10.9%) 6 of 20 patients (30.0%) 29 of 73 patients (39.7%) Mean BCVA 1.93 ± 0.28 1.50 ± 0.66 1.77 ± 0.44 1.35 ± 0.70 logmar (Snellen) (20/1700) (20/632) (20/1177) (20/447) Group 1 (RD at presentation [concurrent RD]); Group 2 (RD during follow-up [delayed-onset RD]) BCVA = best-corrected visual acuity; RD = retinal detachment remaining 73 of 93 patients (78.5%) (Group 2). Table 1 provides the demographic profile and baseline characteristics of these patients. The right eye was involved in 49 patients (53%). The mean ages at presentation in Group 1 and Group 2 were 32.2 years ± 17.6 years and 27.9 years ± 19.1 years, respectively. In Group 1 patients, the etiologies of endophthalmitis were exogenous in 19 of 20 patients (95.0%) (post-trauma: 75.0%, post-cataract surgery: 15.0%, and other: 5.0%) and endogenous in one of 20 patients (5.0%). In Group 2 patients, the etiologies of endophthalmitis were exogenous in 69 of 73 patients (94.5%) (post-trauma : 65.8%, postcataract surgery: 17.8%, and other: 11.0%) and endogenous in four of 73 (5.5%). In the delayed-onset RD group (Group 2), the mean interval between the initial endophthalmitis treatment and RD was 11.5 weeks ± 47.1 weeks. Depending on the severity of the endophthalmitis and decision by the vitreoretinal surgeon, varying primary surgical procedures were performed (Table 2). In Group 1, the initial treatment consisted of PPV + intravitreal antibiotics + silicone oil injection in 19 of 20 patients (one patient underwent initial PPV + intravitreal antibiotics). During the initial surgery, intravitreal antibiotics in full dose were injected after fluid-air exchange, before silicone oil injection in these 19 patients. In Group 2, the initial treatment consisted of vitreous biopsy and intravitreal antibiotics (14 of 73 patients; 19.2%), vitrectomy and intravitreal antibiotics (55 of 73 patients; 75.3%), or vitrectomy and silicone oil injection ± intravitreal antibiotics (four of 73 patients; 5.5%). The patients who did not receive silicone oil during initial surgery underwent silicone oil injection during subsequent surgery for persistent or recurrent detachment. Positive cultures were identified in 31 of 93 patients (33.3%) (Group 1: 3 of 20 patients; Group 2: 28 of 73 patients). The mean visual acuity at presentation and at the last visit in both groups is shown in Table 3. The majority (95.0% in Group 1 and 89.0% in Group 2) of patients had very poor VA at presentation (< 5/200). In Group 1, the anatomical and functional success was reported in 73.7% and 30% patients respectively (Table 2). In Group 2, the anatomical and functional success was reported in 98.5% and 39.7% patients respectively (Table 2). Anatomical success rates were significantly worse in Group 1 compared to Group 2 but among the two groups, the mean visual acuity at the last examination was not statistically different (1.77 logmar ± 0.44 logmar in Group 1 versus 1.35 logmar ± 0.70 logmar in Group 2; P =.5). Silicone oil removal (SOR) was performed in 44 of 93 patients (47.3%) by the last follow-up examination. For the remaining 49 patients, SOR was not performed for various reasons: lost to follow-up in 30 patients (61%), poor visual prognosis (due to corneal opacification/hypotony/optic atrophy/no light perception) in 19 (39%) patients. Among the 44/93 patients who underwent SOR, the mean interval between VR surgery and SOR was 4.5 months ± 2.3 months (median: 4 months; range: 0.5 months to 11 months). Mean duration of follow-up after SOR was 20.2 months ± 40.4 months (median: 7 months; range: 2 months to 120 months) in Group 1 and 18.0 months ± 21.0 months (median: 11 months; range: 1 month to 90 months) in Group 2. After SOR, complications were identified in 28 of 44 eyes including recurrent RD (nine patients), epiretinal membrane (six patients), glaucoma (four patients), RPE atrophic changes in the macula (four patients), squint (two pa- July 2017 Vol. 48, No. 7 549
TABLE 4 Comparison of the Current Study With the Previous Reported Case Series in the Literature Type of Endophthalmitis Exogenous (81%), endogenous (19%) Study Period April 1987 March 1992 Place Number of Cases With RD Foster et al. 6 Nelsen et al. 1 EVS 5 Present Series BPEI-Miami; WSU- Detroit Exogenous bacterial July 1978 December 1983 16 patients 9 of 55 patients (all delayed-onset) Post-cataract surgery All types 1990 1994 January 1991 December 2014 Multi-center United States LVPEI - Hyderabad, India 35 of 420 patients (all delayed-onset) Inoperable RD Cases 2 of 16 3 12 289 443 of 5,924 patients (concurrent: 20; delayed-onset: 73; information N/A: 350) RD Surgery Performed 14 of 16 6 of 9 23 of 35 154 of 443 Mean Age at Presentation (Range) Follow-Up Duration (Range) Visual Acuity at Last Follow-Up 20/400 58.70 years (5 to 88 years) 56.11 years (14 to 77 years) N/A 28.83 years (2 to 79 years) 1 month to 6 years 5 to 18 months N/A 4 to 180 months 4 of 16 patients (25.0%) < 20/400 5 of 16 patients (31.3%) LP/NLP 7 of 16 patients (43.7%) 6 of 9 patients (66.7%) Lost to follow-up: 2 patients; for remaining 21 patients: 14 of 21 (66.7%) Lost to follow-up: 61 patients; for remaining 93 patients: 35 of 93 (37.6%) None 7 of 21 patients 5 of 93 patients (5.4%) 3 of 9 (33.3%) None (33.3%) RD = retinal detachment; LP = light perception; NLP = no light perception; N/A = not available; EVS = Endophthalmitis Vitrectomy Study 53 of 93 patients (57.0%) tients), optic atrophy (two patients), and band keratopathy (one patient). The median visual acuity at last follow-up in 44 eyes undergoing silicone oil removal was 20/100 (logmar 0.7), whereas in the remaining eyes that did not undergo silicone oil removal was 20/2000 (logmar 2.0). DISCUSSION In this large study, incidence of concurrent or delayed-onset retinal detachment was 7.5% (443 of 5,924 patients), which is similar to the incidence rates of 4.6% to 16% reported in literature. 1,5-7 The age at presentation in the current study was the third decade (mean age: 28.8 years ± 18.8 years), which was consistent with trauma as the leading etiology. Endophthalmitis caused by more virulent organisms, poor presenting visual acuity, and posterior capsular rupture at the time of cataract surgery are reported to be associated with higher rates of RD. 5,6 In some series, a greater incidence of RD was reported in the vitrectomy with intravitreal antibiotic group as compared to the only intravitreal antibiotic group. 1,7 However, the current study included all the patients who had RD either at presentation or during followup who underwent vitrectomy surgery. The development of RD in these patients may occur at presentation or after surgery or during course of follow-up. 1,8,9 A comparison of various aspects of these cases with endophthalmitis and concurrent or delayed-onset RD among few reports in the literature are shown in Table 4. The dosage of intravitreal antibiotics in eyes undergoing silicone oil injection is controversial. An experimental animal study performed in 1999 by Hegazy et al. injected intravitreal antibiotics after vitrectomy, fluid-air exchange, and silicone oil injection. This study demonstrated retinal toxicity of a full dose of intravitreal antibiotics in silicone oil-filled rabbit 550 Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina
eyes. 10 However, these results may not be applicable to the human eyes. In the current study, the intravitreal antibiotics were used in full dose after fluid-air exchange followed by silicone oil injection. Mieler et al. used a half-dose of intravitreal antibiotics in 50% air filled eyes with traumatic endophthalmitis with RD. 11 In a case of endophthalmitis in a silicone oilfilled eye reported by Zimmer-Galler et al., silicone oil removal was performed followed by injection of a full dose of intravitreal antibiotics. 12 After waiting for 15 minutes with intravitreal antibiotics in the vitreous cavity, 80% fluid-gas exchange was performed followed by silicone oil-air exchange and the patient had a good visual outcome. The importance of these modifications of dose and procedures remains unproven. The antimicrobial activity of silicone oil has been reported in various studies. 13,14 Possible mechanisms of the antimicrobial activity include direct toxicity and nutritional deprivation for microorganisms. 14 This may provide added benefit for patients with RD associated with endophthalmitis. In a small case series of four patients, Aras et al. described the outcomes after silicone oil removal in cases of RD associated with endophthalmitis. 14 One patient had a final VA of 20/40, whereas the remaining patients had very poor VA outcomes. In a multicentered, retrospective, noncomparative clinical case series (2016), the visual outcomes of acute-onset endophthalmitis in 70 patients who underwent therapeutic PPV were reported. 15 In this retrospective study, there were 15 of 70 eyes (21.4%) with a RD either at the time of initial PPV or during follow-up. In this study, eight eyes underwent more than one PPV and seven eyes received silicone oil tamponade. VA outcomes were generally poor in this study (at the last follow-up, VA ranged from counting fingers to no light perception, except for one eye that ultimately underwent silicone oil removal and had final VA of 20/40 with an attached retina). The limitations of current study include its retrospective nature, lack of a defined protocol for treatment, and treatment by multiple vitreoretinal surgeons. Of the total 443 cases of RD with endophthalmitis noted, only 93 were deemed operable and were included in the detailed analysis. The large number of excluded cases indicate the complexity of the disease and poor visual prognosis of this condition. The current study does not address the larger number of patients with severe visual loss and more advanced clinical severe disease (endophthalmitis and RD) who were deemed unsalvageable and were not included in the current study. In conclusion, RD associated with endophthalmitis is associated with generally poor anatomic and visual outcomes. Based on the current study data, silicone oil can be a useful option in the management of concurrent or delayed-onset RD. Repeat intravitreal antibiotics can be utilized if necessary. REFERENCES 1. Nelsen PT, Marcus DA, Bovino JA. Retinal detachment following endophthalmitis. Ophthalmology. 1985;92(8):1112-1117. 2. Brinton GS, Topping TM, Hyndiuk RA, Aaberg TM, Reeser FH, Abrams GW. Posttraumatic endophthalmitis. Arch Ophthalmol. 1984;102(4):547-550. 3. Affeldt JC, Flynn HW Jr., Forster RK, Mandelbaum S, Clarkson JG, Jarus GD. Microbial endophthalmitis resulting from ocular trauma. Ophthalmology. 1987;94(4):407-413. 4. Vahey JB, Flynn HW Jr. Results in the management of Bacillus endophthalmitis. Ophthalmic Surg. 1991;22(11):681-686. 5. Doft BM, Kelsey SF, Wisniewski SR. Retinal detachment in the endophthalmitis vitrectomy study. Arch Ophthalmol. 2000;118(12):1661-1665. 6. Foster RE, Rubsamen PE, Joondeph BC, Flynn HW Jr, Smiddy WS. Concurrent endophthalmitis and retinal detachment. Ophthalmology. 1994;101(3):490-498. 7. Olson JC, Flynn HW Jr., Forster RK, Culbertson WW. Results in the treatment of postoperative endophthalmitis. Ophthalmology. 1983;90(6):692-699. 8. Landers JH, Chappell CW. Bilateral metastatic endophthalmitis. Retina. 1981;1(3):175-178. 9. Tolentino FI, Schepens CL, Freeman HM. The vitreous in peripheral retinal breaks and rhegmatogenous retinal detachment. In: Vitreoretinal Disorders: Diagnosis and Management. Philadelphia, PA: WB Saunders; 1976:396. 10. Hegazy HM, Kivilcim M, Peyman GA, et al. Evaluation of toxicity of intravitreal ceftazidime, vancomycin, and ganciclovir in a silicone oil-filled eye. Retina. 1999;19(6):553-537. 11. Mieler WF, Glazer LC, Bennett SR, Han DP. Favourable outcome of traumatic endophthalmitis with associated retinal breaks or detachment. Can J Ophthalmol. 1992;27(7):348-352. 12. Zimmer-Galler IE, Santos A, Haller JA, Campochiaro PA. Management of endophthalmitis in a silicone oil-filled eye. Retina. 1997;17(6):507-509. 13. Ozdamar A, Aras C, Ozturk R, Akin E, Karacorlu M, Ercikan C. In vitro antimicrobial activity of silicone oil against endophthalmitiscausing agents. Retina. 1999;19(2):122-126. 14. Aras C, Ozdamar A, Karacorlu M, Ozkan S. Silicone oil in the surgical treatment of endophthalmitis associated with retinal detachment. Int Ophthalmol. 2001;24(3):147-150. 15. Sridhar J, Yonekawa Y, Kuriyan AE, et al. Microbiologic spectrum and visual outcomes of acute-onset endophthalmitis undergoing therapeutic pars plana vitrectomy. Retina. 2017;37(7):1246-1251. July 2017 Vol. 48, No. 7 551