The Latest Updates and Management of Endophthalmitis

Postoperative endophthalmitis (PE), although rare, is undoubtedly the most serious complication in ocular surgery, with less than half of the affected patients able to achieve a final visual acuity of 20/40 or better.¹ The highest risk is observed in cataract surgery with intraocular lens (IOL) implantation and the lowest in pars plana vitrectomy (PPV).² Recent estimates of the rate of endophthalmitis after cataract surgery have ranged between 0.08% and 0.68%.³ In this editorial, we will review the latest information related to the respective roles of disinfection using povidone-iodine (PVI) before surgery and the use of preoperative, intracameral, and postoperative antibiotics in the prevention of PE. Latest developments in management of endophthalmitis will also be discussed.

The European Society of Cataract and Refractive Surgeons (ESCRS), in its first randomized multicenter study on antimicrobial prophylaxis, reported an incidence of endophthalmitis ranging from 0.049% to as high as 0.345% in the control group.⁴ The preferred practice pattern for prophylaxis of endophthalmitis has seen some major crests and troughs in recent periods, as the ESCRS group failed to reduce the rate of PE (0.074%) significantly less than expected, despite using intracameral cefuroxime in the study population.⁵

The usual concentration of PVI used before cataract surgery is between 5% and 10%, with an exposure time of 2 to 5 minutes. There has been a trend toward irrigating and applying PVI on the ocular surface for a contact period of at least 5 minutes before cataract surgery to reduce the risk of PE. Although there may be multiple factors, the application of PVI to the ocular adnexa and conjunctival surface is probably the single most determining factor in reducing the risk of PE after cataract surgeries.⁶ Although in vitro studies have shown a bactericidal effect of PVI even at 30 seconds, we still believe that these variable and inconsistent PVI regimens are most likely responsible for the more than expected incidence of endophthalmitis after cataract surgery and intravitreal injections. To date, there is no report of bacterial resistance against PVI as compared with antibiotics.⁷ Application of 10% PVI to the eyelids and periorbital area and 5% PVI on the ocular surface resulting in exposure for 5 minutes can prevent the growth of most postcataract surgery endophthalmitis bacterial isolates. Although there is no large randomized study to support the addition of topical antibiotics preoperatively or postoperatively to reduce the rate of postcataract surgery or intravitreal injection endophthalmitis, the practice is widely prevalent in most clinical settings.⁸ The study by Miller et al⁹ has been referred to in several published works stating that lidocaine gel may increase the risk of acute postoperative endophthalmitis as it was thought to act as a barrier to asepsis. However, a report by Lad et al¹⁰ showed little effect of lignocaine gel on asepsis. We support the use of additional PVI before and after the application of the gel before intraocular procedures as proposed by an expert panel in a recent guideline.⁸

The rationale for preoperative topical antibacterial therapy is to sterilize the ocular surface and achieve therapeutic intraocular concentrations. Fluoroquinolones are the most popular class of antibiotics used due to their broad-spectrum microbial coverage, good penetration into the anterior chamber, and low adverse effect profile, with 93% of surveyed members of the American Society of Cataract and Refractive Surgery (ASCRS) favoring their use.¹¹ It should be noted that reports doubting the role of preoperative antibiotics before cataract surgery do exist. Rudnisky et al⁵ found a 0.1% difference in the rate of endophthalmitis due to usage and nonusage of preoperative topical antibiotic. Even the large prospective ESCRS study did not show significant benefits of using preoperative levofloxacin an hour before surgery on the overall risk of PE. However, in recent studies, preoperative antibiotics have shown good prospect in infection prophylaxis. Topical moxifloxacin 0.5% achieved significantly higher aqueous concentrations than the known minimum inhibitory concentration (MIC) for Staphylococcus epidermidis and showed reduction of similar colony-forming unit when applied 4 times every 15 minutes 2 hours before surgery compared with 1 day before surgery (4 times a day with an additional drop 2 hours before operation).¹² In another retrospective analysis of endophthalmitis rates after the prophylactic use of third- versus fourth-generation fluoroquinolones in cataract surgery, Jensen et al¹³ found a statistically significantly lower rate of PE with the use of fourth-generation fluoroquinolones. Contrary to the common practice in a preoperative antibiotic regime for 3 days, the recommendation of using topical 0.5% moxifloxacin on the day of surgery seems prospective, but further studies are warranted to determine its exact role in reducing the risk of postcataract endophthalmitis.¹²

Intraoperative intracameral (IC) delivery is an efficient method to quickly attain high intraocular antibiotic concentrations to eliminate probable pathogens introduced into the eye during cataract surgery. The in vitro half-life of IC antibiotics is approximately 2 hours, with levels of MIC lasting up to 8 hours.¹⁴ The ESCRS data first supported the use of IC cefuroxime for the prevention of PE. In this multicenter European study of 13,698 patients, IC infusion of 1 mg cefuroxime in 0.1 mL isotonic sodium chloride solution was found to have a 5-fold reduction in the risk of endophthalmitis compared with placebo or preoperative topical levofloxacin.⁴ This study could not change the universal practice pattern, as it still had a significantly higher rate of endophthalmitis (0.074%) than expected. This can partly be attributed to its limited coverage against gram-negative bacteria, gaps in coverage of methicillin-resistant Staphylococcus aureus (MRSA), compounding considerations, hypersensitivity, and time-dependent killing.¹⁵ In the search for more effective antibiotics, IC fluoroquinolones were slowly entered into evaluation. Moxifloxacin is a fourth-generation fluoroquinolone with a broad spectrum of coverage against both gram-positive and gram-negative bacteria along with anaerobes. Unlike gatifloxacin that contains benzalkonium chloride preservative, moxifloxacin is commercially formulated without preservatives and thus has less chance of ocular toxicity, and its bactericidal efficacy is concentration dependent. Espiritu et al¹⁶ estimated that injecting 0.5 mg of moxifloxacin in 0.1 mL would achieve MIC around 300 times for most endophthalmitis pathogens. Multiple recent observational studies from the United States,¹⁷ India,¹⁸ and Iran¹⁹ involving 636,819 patients demonstrated a reduced likelihood of PE with the use of IC antibiotics at the time of cataract surgery. These new studies together with previously published ESCRS data represent more than 1.3 million patients in whom IC antibiotics were associated with a reduced incidence of PE.²⁰ Recent publications strongly support use of IC moxifloxacin prophylaxis after cataract and other intraocular surgery. This practice has already been adopted in Europe and is going to become indispensible in the ophthalmic community in the future.^17–19

The superior role of subconjunctival antibiotic injections, especially vancomycin and ceftazidime, after cataract surgery has been shown in some studies.²¹ However, they only provide temporary therapeutic antibiotic levels in the anterior chamber and do not adequately penetrate the vitreous.²² The results of these studies should be applied with caution, as indiscriminate use of these drugs can affect bacterial resistance, which in turn will reduce our choices in the treatment of proven endophthalmitis.

The use of postoperative antibiotics seems relevant to the increasing popularity of sutureless clear corneal incisions, which are thought to allow ocular surface fluids into the eye. In a large retrospective study of 20,013 patients, the overall endophthalmitis rate was 0.07% in uncomplicated cataract surgery patients who received topical gatifloxacin or moxifloxacin for prophylaxis.²³ Another cross-sectional study of 29,276 patients showed a reduction in the rate of PE from 0.197% to 0.056% after changing from early-generation fluoroquinolones to fourth-generation fluoroquinolones.¹⁰ The emergence of antibiotic resistance because of increasing and inappropriate use of perioperative antimicrobial prophylaxis is a serious concern. Although new groups of antibiotics prove to be better in reducing the risk of postoperative infections, it is better to give postoperative antibiotics for a full course of 4 times a day for 5 days rather than a half-hearted approach to reduce the risk of bacterial resistance. Susceptibility testing of endophthalmitis isolates shows a rise in resistant organisms such as MRSA and fourth-generation fluoroquinolone-resistant MRSA strains.²⁴ As long as new antibiotics are added to keep up with resistant bacteria, the concept of planned antibiotic restriction or cycling to alter the selective pressure of resistant strains is necessary to combat microbial resistance.²⁵

Patients with acute endophthalmitis should be differentiated from toxic anterior segment syndrome, where there is significant anterior segment involvement within 1 day of surgery and classically, patients have minimal or no pain and clear vitreous. When in doubt, it is always better to err on the side of caution and manage the patient as per the endophthalmitis treatment plan. In presumed endophthalmitis, anterior chamber aqueous tap and vitreous tap or biopsy is planned and the material is sent for complete microbiological testing including polymerase chain reaction for eubacterial and panfungal genome. Real-time polymerase chain reaction is a valuable diagnostic tool for rapid diagnosis of bacterial endophthalmitis with higher sensitivity than routine bacterial culture.²⁶

The high susceptibility rates and excellent safety profile of intravitreal vancomycin and ceftazidime within the therapeutic window make them the first-line choice as empiric antibiotics in the treatment of acute endophthalmitis. Gentamycin is particularly avoided because of its macular toxicity, and even other aminoglycosides like amikacin and tobramycin have also been reported to have retinal toxicity and are best avoided for intravitreal injections.²⁷ Other antibiotics worth mentioning are linezolid, which has been shown to have excellent intraocular penetration and anti gram-positive activity,²⁸ and fluoroquinolones.²⁹ Recently, intravitreal imipenem has been shown to be effective as opposed to its use as a systemic adjuvant in experimental models of endophthalmitis.³⁰ Voriconazole is the drug of choice in fungal endophthalmitis with a broader spectrum of action against filamentous and nonfilamentous fungi as compared with amphotericin, which works best against nonfilamentous yeasts and has the potential for retinal toxicity.³¹

Systemic antibiotics can also play a role by increasing the intravitreal concentration of the antibiotic, as the blood aqueous barrier has been broken down by inflammation. A more recent study reported similar findings related to the benefits of PPV and found that oral antibiotics significantly contribute to improved final visual acuity.³² The reason intravenous amikacin and ceftazidime did not show good results in the Endophthalmitis Vitrectomy Study (EVS) was probably because they both have poor intraocular penetration compared with fluoroquinolones.²² Given their favorable characteristics of broad coverage, good tolerability, and ease of oral administration, fourth-generation fluoroquinolones are promising adjunct therapies in the management of PE.

The EVS and a few subsequent studies have shown no visual benefit of PPV in endophthalmitis with initial visual acuity better than light perception.³³ The EVS data have their limitations, as more severe cases with no light perception and cases of endophthalmitis after procedures other than cataract surgery or secondary IOL implantation were excluded from the study. In the EVS, no or late surgery increased the risk of phthisis or enucleation (6%), delayed visual rehabilitation (53% of eyes with ≥20/40), and had a greater reoperation rate (6%) in the nonsurgical group as compared with 0% in the surgical group. With the advent of sophisticated microvitrectomy systems and instrumentation, PPV has evolved immensely in terms of safety and maneuverability in the past decade. This has led retinal surgeons worldwide to lower their threshold for vitrectomy in endophthalmitis.³⁴ Vitrectomy definitely helps in pathogen identification, removal of infecting organisms, and reduction of cells and inflammatory mediators and debris in endophthalmitis. This has helped in planning early surgical intervention, which is always better than operating in only light perception eyes, where along with surgical difficulties the prognostic outcome also remains uncertain. In the setting of gross anatomical alteration in endophthalmitis, vitrectomy needs judicious adjustments to reduce the risk of iatrogenic retinal tears and is better managed with the modern vitrectors with excellent duty cycle and wide-angle viewing systems. We recommend managing endophthalmitis based on the clinical appearance and course. Intravitreal antibiotics remain the mainstay of treatment in managing acute endophthalmitis, and early vitrectomy should be considered in less than expected clinical recovery after the injection of intravitreal antibiotics.