Although certainly not a new topic, postoperative endophthalmitis—its prevention, pathogenesis, and management— is the subject of this month’s column. Of importance is the rise of Methicillin-resistant Staphylococcus aureus (MRSA) as a source of postoperative corneal and intraocular infection. One center reported that as many as 30% to 40% of endophthalmitis cases were caused by MRSA.¹ Based on these data, it is paramount that surgeons pay close attention preoperatively to the potential risk factors for infection and postoperatively to the prevention of infection. The early identification of the signs and symptoms of infection is also important.

Before surgery, physicians must pay close attention to blepharitis, dry eyes, and the presence of punctal plugs. I pretreat patients who have any of these risk factors with topical azithromycin gel (Azasite; Inspire Pharmaceuticals, Inc.) q.d. for 1 week. For severe blepharitis, I also recommend daily lid scrubs starting 1 week before surgery and continued for a minimum of 2 weeks postoperatively. Intraoperatively, I add vancomycin 20 μg/mL to the irrigating solution for intracameral delivery during cataract surgery. Fourth-generation fluoroquinolones are started 30 minutes prior to surgery and continued immediately for at least 14 days postoperatively. Recent studies have demonstrated that most postoperative endophthalmitis is documented 7 to 9 days after surgery.

If the posterior capsule has been compromised, I advocate the use of preservative-free Kenalog 4 mg (Bristol-Myers Squibb Co.) mixed with 1 mL of balanced salt solution. Delivering this mixture to the anterior chamber aids in the identification of anteriorly prolapsed vitreous, which could remain incarcerated in the corneal wound and lead to postoperative endophthalmitis. Late-onset endophthalmitis is almost always the result of a leaking clear corneal wound in which it has been demonstrated that tear film fluids can migrate and inoculate the anterior chamber with microorganisms when the IOP dips below 12 mm Hg.² Of interest is the recent development of Biodendrimer adhesives (Hyperbranch Medical Technology, Inc., Raleigh, NC), which are currently being investigated by the FDA. The adhesives bind extremely well to the corneal surface and, unlike cyanoacrylate, are transparent once polymerized, smooth, and soft to touch. They are therefore well tolerated by patients without the need for a bandage contact lens. As surgeons’ concerns regarding postoperative infection rise, patients demand less postoperative management and fewer visits. It is my belief that Biodendrimer adhesives will assume a substantial role in preventing postoperative infections after intraocular surgery.

I hope you enjoy this installment of “Peer Review,” and I encourage you to seek out and review the articles in their entirety at your convenience.

—Mitchell C. Shultz, MD, section editor

ANALYZING ANTIBIOTIC-RESITANT BACTERIA
In a retrospective, consecutive, observational case series, 64 cases of acute endophthalmitis occurring within 6 weeks after cataract surgery were identified over a 3-year interval at one vitreoretinal practice. Cultures from 33 of the 64 eyes tested positive for MRSA. Six of the 33 cultures that tested positive demonstrated MRSA infections. Those six eyes were started on fluoroquinolone antibiotics 2 to 3 days postoperatively. Corneal abnormalities were found in five of the six cases. Visual acuity at the last follow-up visit was no light perception in two eyes, hand movements in two eyes, and 20/30 in two eyes. One eye had no light perception vision and underwent enucleation within 3 days of presentation. All six organisms were sensitive in vitro to gentamicin and vancomycin. None of the organisms was sensitive to fluoroquinolone antibiotics, but investigators noted that not all of them were tested against fluoroquinolones.³

In an observational, prospective study, investigators collected preoperative conjunctival cultures from 1,940 consecutive patients undergoing cataract surgery during a 1-year period. Of these cultures, 4,391 microorganisms were isolated to identify the presence of antibioticresistant conjunctival bacteria. Investigators stated that 94.23% of the isolated microorganisms were gram-positive, and 5.31% were gram-negative. Of the 1,940 cultures, the most prevalent conjunctival bacteria were coagulasenegative staphylococci, which were present in 88.3% of cultures. Diphtheroids were harbored in 58.1% of patients, propionibacteria in 31%, streptococci in 23.1%, S. aureus in 10.2%, haemophilus and gram-negative diplococci in 7.5%, other gram-negative rods in 4.5%, and enterococcus in 2%. The enterococci-staphylococci profile was the most resistant to antibiotics. Investigators concluded that the typical respiratory bacteria remained sensitive to chloramphenicol and ß-lactams, and other gram-negative rods were sensitive to aminoglycoside, quinolones, and certain β-lactams.⁴

In an in vitro laboratory investigation, researchers examined in vitro antibiotic susceptibility patterns of conjunctival bacterial flora isolated before surgery from patients undergoing refractive surgery. Of 105 patients (105 eyes) scheduled for refractive surgery, 67.6% underwent LASIK with a femtosecond laser, 22.9% underwent LASIK with an automated microkeratome, 7.6% underwent LASEK, and 1.9% received a phakic IOL. Preoperative conjunctival cultures were obtained on the day of surgery before topical antibiotics, anesthetics, and povidone-iodine were administered. Coagulasenegative staphylococci were identified in 85% of patients, S. aureus was identified in 2.3%, and Streptococcus pneumoniae was identified in 1.2%. Five gram-negative bacilli were isolated, but no fungi or mycobacteria were isolated. Researchers concluded that gemifloxacin, moxifloxacin, and gatifloxacin were the most effective against conjunctival bacteria isolated from refractive surgery patients, and resistance to ofloxacin and levofloxacin is increasing among methicillinsusceptible coagulase-negative staphylococci (Table 1).⁵

ANTIBIOTICS’ PREPARATION AND DELIVERY
Ten ophthalmic OR nurses from two hospitals in Glasgow, Lanarkshire, United Kingdom, performed different protocols to dilute cefuroxime for intracameral injection. Oven-dried, analytical-grade potassium chloride was used as a surrogate for cefuroxime. Each group of nurses prepared solutions of 1.0 mg in 0.1 mL according to a protocol with which they were familiar and one with which they were unfamiliar. Ten samples were collected from each group of nurses for each protocol. Ten analytical chemists also performed both dilutions to act as a nonmedical comparison group. A total of 30 samples were obtained for each protocol. The median dose after dilution was 1.17 mg for the unfamiliar protocol and 2.05 mg for the familiar protocol. For the familiar protocol, the median dose was significantly higher (P < .001), and there was greater variability. Investigators noted that inadequate mixing in a 1-mL syringe was likely the cause for inaccuracy in the familiar protocol, which indicates that small-volume syringes should not be used for mixing. They concluded that the mathematical accuracy of a dilution protocol does not ensure dosing accuracy in a clinical setting.⁶

To combat infections that arise from the bacterial colonization of a new IOL and the subsequent formation of an antibiotic-tolerant biofilm, researchers developed a polymeric hydrogel system that delivers specific levels of antibiotics over an extended period of time within the globe of the eye. Investigators loaded norfloxacin into cross-linked poly(2-hydroxyethyl methacrylate) gels, which were surface-modified with octadecyl isocyanate to produce a hydrophobic ratelimiting barrier controlling the release of norfloxacin. Octadecyl surface modification was characterized using scanning electron microscopy and X-ray photoelectron spectroscopy. A 15-minute modification reportedly led to a uniform surface coating and nearly a zero-order release of norfloxacin from the matrix. They concluded that norfloxacin released from coated, poly(2-hydroxyethyl methacrylate) kills Staphylococcus epidermidis in suspension and on a simulated medical implant’s surface.⁷

VANCOMYCIN
In a retrospective study, researchers analyzed 16,606 cataract surgeries that were performed at the Warrington Hospital in North West England over an 11-year period. The surgeries were divided into two time periods. Period A consisted of surgeries that were performed between January 1, 1998, and December 31, 2000, prior to the introduction of intracameral vancomycin at the end of surgery. Period B consisted of surgeries that were performed between January 1, 2001, and December 31, 2008, after the introduction of intracameral vancomycin at the end of surgery. The incidence of endophthalmitis per 1,000 cataract surgeries was 3.0 during period A and 0.08 during period B. This reduction was statistically significant (Chi-squared test; 36.6; P < .0001). The relative risk of developing endophthalmitis without intracameral vancomycin prophylaxis was 38, and the reduction in absolute risk was 292 cases of endophthalmitis per 100,000 cataract surgeries.⁸

Following routine cataract surgery with phacoemulsification and the implantation of a PCIOL, 19 patients received injections of vancomycin 1 mg/0.1 mL saline solution. Aqueous samples were obtained by inserting a cannula into the anterior chamber through the sideport incision. In nine patients, group 1, aqueous sampling took place 1 minute after intracameral injections of vancomycin. In 10 patients, group 2, aqueous sampling took place between 18 and 24 hours postoperatively. Fluorescence polarization immunoassay was used to calculate the aqueous vancomycin concentration. The mean and median concentrations of vancomycin were 5,385 mg/L and 5,458 mg/L in group 1, respectively. In group 2, the mean and median concentrations of vancomycin were 41.1 mg/L and 40.6 mg/L, respectively. The concentration of vancomycin exceeded its minimum inhibitory concentration by a factor of four for up to 26 hours postoperatively.⁹

In a prospective, controlled trial, 41 patients (50 eyes) with cataracts were randomized to receive 20 µg/mL of vancomycin and 8 µg/mL of gentamicin in the infusion fluid at the time of cataract surgery or no intracameral antibiotics. Follow-up was completed on 49 eyes. Optical coherence tomography was performed to measure macular thickness at 1 day and 1 and 5 weeks postoperatively. In patients who did not receive intracameral antibiotics, the central retinal thickness increased at 5 weeks from baseline by a mean of 12.3 ±27.5 µm (P < .001). In patients who received intracameral antibiotics, central retinal thickness increased at 5 weeks from baseline by a mean of 10 ±13 µm (P < .001). The mean contrast sensitivity was 1.26 in patients with increased macular thickness and 1.43 in those with no change (P = .001). Researchers noted no significant effect of intracameral antibiotics in the infusion fluid on macular thickness or visual function after cataract surgery.¹⁰

To evaluate the oxidative-stress parameters of intracameral cefuroxime and vancomycin in the cornea, the right eyes of 30 animal models were randomized into three groups: a cefuroxime group (n = 10), a vancomycin group (n = 10), and a control group (n = 10). Twenty eyes were injected with 0.1 mL of the two antibiotic agents, and 10 eyes were injected with 0.1 mL of balanced salt solution. Corneal thickness and clarity were measured before surgery and at 3 and 6 hours postoperatively. Corneal tissues were extracted and homogenized in 2 mL of physiological saline and kept at -80°C. Malondialdehyde and total thiol levels were measured with spectrophotometric methods, and a statistical analysis was conducted. Neither cefuroxime nor vancomycin caused corneal thickening or edema at 3 and 6 hours after injection, and no anterior chamber reactions were observed in either group. In the cefuroxime group, the level of total thiol significantly decreased (P = .001), and the level of malondialdehyde significantly increased (P < .001). There were no biochemical changes in the vancomycin group compared with the control group.¹¹

In a prospective, interventional, hospital-based study, 400 patients (400 eyes) who underwent phacoemulsification between January 2004 and 2006 were nonrandomly assigned to two groups. Group 1 comprised 180 patients who received topical 0.3% ciprofloxacin eye drops q.i.d. for 1 day preoperatively and did not receive vancomycin in the irrigating solution during cataract surgery. Group 2 comprised 220 patients who underwent cataract surgery with 20 μg/mL of vancomycin in the irrigating solution but did not receive topical antibiotics preoperatively. Anterior chamber aspirates were obtained from both groups at the end of surgery. Statistical analysis was performed with a Chi-square test. Bacteria were cultured in 21.1% of eyes in group 1 and in 7.7% of eyes in group 2. Coagulase-negative Staphylococcus was the most prevalent organism in both groups. Multiple organisms were identified in 2.2% of eyes in group 1, whereas none of the eyes in group 2 showed multiple organisms. None of the eyes in either group exhibited fungal contamination. One patient in group 1 developed endophthalmitis caused by Alcaligenes faecalis. All patients were analyzed at a follow-up visit between 6 and 14 months postoperatively.¹²

COMPLEX CASES
A 54-year-old man presented to the Cole Eye Institute in Cleveland with a complaint of severe pain and decreased vision in his left eye for a 2-day period. The patient had a BCVA of 20/25 in his right eye and 20/100 in his left eye. His left eye was swollen and displayed moderate eyelid erythema and diffuse injection. A slitlamp examination revealed a corneal ulcer associated with a loose suture from cataract surgery that had been performed 2 years ago. The broken suture was removed, and the patient was started on topical antibiotic treatment, alternating between 50 mg/mL of cefazolin and 14 mg/mL of gentamicin every 30 minutes. Cultures of the ulcer revealed MRSA. The patient’s treatment regimen was changed to include 50 mg/mL of vancomycin, but the ulcer continued to progress. Three days later, the ulcer perforated, and the patient underwent an emergency corneal patch graft and lateral tarsorrhaphy. Preoperative cultures were negative, and the patient was treated with Zymar (Allergan, Inc.) q.i.d. and bacitracin ointment. By 1 month postoperatively, the BCVA in the patient’s left eye was 20/30, and the ocular surface and patch graft were intact. Investigators reported that, to their knowledge, this was “the first reported case of suture-related MRSA keratitis after uncomplicated clear corneal cataract surgery.”¹³

A 31-year-old man was referred to the Cabarrus Eye Center in Concord, North Carolina. He had developed necrotizing community-based MRSA conjunctivitis that caused palpebral conjunctival ulceration and destruction of postseptal soft tissue with invasion of extraconal fat. The authors of the report noted that, although the patient’s infection was severe, the external signs of eye lid disease were modest. He did not have any apparent risk factors for MRSA colonization. The patient was treated for 7 days with intravenous vancomycin and tobramycin sulfate as well as oral rifampin and then for 7 days with oral trimethoprim sulfa double strength with clinical resolution. The report’s authors stated that physicians should consider MRSA in patients with atypical conjunctivitis showing patchy necrosis of the conjunctiva and investigate any dimpled or eroded areas for evidence of deeper invasion.¹⁴

Section Editor Mitchell C. Shultz, MD, is in private practice and is an assistant clinical professor at the Jules Stein Eye Institute, University of California, Los Angeles. He acknowledged no financial interest in the products or companies mentioned herein. Dr. Shultz may be reached at (818) 349-8300; izapeyes@gmail.com.

1. Mather R, Karenchak LM,Romanowski EG,Kowalski RP.Fourth generation fluoroquinolones:new weapons in the arsenal of ophthalmic antibiotics.Am J Ophthalmol.2002;133(4):463-466.
2. Taban M, Sarayba MA, Ignacio TS, et al.Ingress of India ink into the anterior chamber through sutureless clear corneal cataract wounds.Arch Ophthalmol.2005;123(5):643-648.
3. Deramo VA, Lai JC,Winokur J, et al.Visual outcome and bacterial sensitivity after methicillin-resistant Staphylococcus aureus-associated acute endophthalmitis.Am J Ophthalmol.2008;145(3):413-417.
4. Fernández-Rubio E, Urcelay JL, Cuesta-Rodriguez T.The antibiotic resistance of conjunctival bacteria:a key for designing a cataract surgery prophylaxis.Eye (Lond).2009;23(6):1321-1328.
5. Chung JL, Seo KY,Yong De.Antibiotic susceptibility of conjunctival bacterial isolates from refractive surgery patients.Ophthalmology.2009;116(6):1067-1074.
6. Lockington D, Flowers H,Young D,Yorston D.Assessing the accuracy of intracameral antibiotic preparation for use in cataract surgery.J Cataract Refract Surg.2010;36(2):286-289.
7. Anderson EM, Noble ML, Garty S, et al.Sustained release of antibiotic from poly(2-hydroxyethyl methacrylate) to prevent blinding infections after cataract surgery.Biomaterials.2009;30(29):5675-5681.
8. Anijeet DR, Palimar P, Peckar CO.Intracameral vancomycin following cataract surgery:an eleven-year study.Clin Ophthalmol.2010;4:321-326.
9. Murphy CC, Nicholson S, Quah SA, et al.Pharmacokinetics of vancomycin following intracameral bolus injection in patients undergoing phacoemulsification cataract surgery.Br J Ophthalmol.2007;91(10):1350-1353.
10. Ball JL,Barrett GD.Prospective randomized controlled trial of the effect of intracameral vancomycin and gentamicin on macular retinal thickness and visual function following cataract surgery.J Cataract Refract Surg. 2006;32(5):789-794.
11. Ozlem TY,Necati DM, Fatma YM,et al.Are cefuroxime and vancomycin really safe on the corneal endothelial cells? Graefes Arch Clin Exp Ophthalmol.2010;248(3):415-420.
12. Srinivasan R, Gupta A, Kaliaperumal S, et al.Efficacy of intraoperative vancomycin in irrigating solutions on aqueous contamination during phacoemulsification.Indian J Ophthalmol.2008;56(5):399-402.
13. Tarabishy AB, Steinemann TL.Suture-related keratitis following cataract surgery caused by methicillin-resistant Staphylococcus aureus.Clin Ophthalmol.2010;4:179-182.
14. Brown SM,Raflo GT, Fanning WL.Transconjunctival orbital invasion by methicillin-resistant Staphylococcus aureus.Arch Ophthalmol.2009;127(7):941-942.