The incidence of endophthalmitis after cataract surgery may differ depending on multiple factors including patient demographics, surgeons, instruments, techniques, and unknown factors. Large series have reported rates of acute-onset postoperative endophthalmitis ranging from 0.03 to 0.2% [1, 6, 7]. Published data from the US Centers for Medicare and Medicaid in 2003–2004 estimated the rate at 0.1% [8]. During this timeframe, the vast majority of US centers did not use intracameral antibiotics. A contemporaneous study (2002–2004) performed in Sweden when most patients received intracameral antibiotics reported the rate at 0.048% [9]. A study using data from the American Academy of Ophthalmology’s Intelligent Research in Sight (IRIS) Registry and US Centers for Medicare and Medicaid from 2010 to 2014 reported an incidence of endophthalmitis after cataract surgery of 0.14% [6].

The European Society of Cataract and Refractive Surgeons (ESCRS) designed a multicenter RCT to identify the risk factors and report on the incidence of postoperative endophthalmitis after cataract surgery [10]. The study included over 16,000 patients that were randomized to (1) no antibiotics, (2) postoperative topical levofloxacin, (3) intracameral cefuroxime, or (4) intracameral cefuroxime and postoperative topical levofloxacin. In this study, intracameral cefuroxime was associated with an approximate fivefold decrease in the rates of postoperative endophthalmitis. In addition, clear corneal incision surgery was associated with an approximate sixfold increased risk, and silicone intraocular lenses were associated with an approximate threefold increased risk.

There are several important criticisms of this study. The relatively high rates of endophthalmitis in eyes not randomized to receive intracameral cefuroxime (approximately 0.2%) may have exaggerated the apparent benefit of these antibiotics. The study design allowed multiple different surgical techniques, which may have introduced bias [11]. The choice of levofloxacin as the postoperative topical antibiotic, rather than the more efficacious fourth-generation fluoroquinolones, might have influenced the results [12].

Despite these criticisms, following the publication of the ESCRS trial, the use of intracameral antibiotics became more common in many parts of the world. Multiple nonrandomized series, most of which were retrospective, were subsequently reported in the UK, Spain, France, Singapore, the USA, Sweden, Japan, Portugal, Ireland, Israel, and other nations (Table 29.2) [13–32].

When evaluating these results, it is important to consider that:

Intracameral antibiotics may be associated with dilution errors, cystoid macular edema, toxic anterior segment syndrome (TASS), and selection of resistant organisms [35]. Intracameral vancomycin is reported to be associated with hemorrhagic occlusive retinal vasculitis and severe visual loss [36]. Intracameral aminoglycosides are associated with retinal vascular toxicity and severe visual loss [37]. Intracameral antibiotics are associated with fungal infections: in one report, seven consecutive patients developed endophthalmitis caused by Fusarium species following the use of intracameral cefuroxime [38].

Aprokam (Thea Pharmaceuticals, Clermont-Ferrand, France), a prepackaged formulation of cefuroxime for intracameral use, is approved by the European Medicines Agency for cataract surgery. This formulation of cefuroxime decreases concerns about compounding. However, this agent is not available in many nations, including the USA, and its use is variable even in nations where it is available. For example, intracameral antibiotics are used almost universally in Sweden, very commonly in France, but much less commonly in the UK, Spain, Germany, Belgium, Italy, the Netherlands, Poland, and Japan [39, 40]. A survey performed in Europe reported that 26% of surgeons were not using intracameral antibiotics routinely, and the main reason reported was the belief that intracameral antibiotics were unnecessary [41].

The American Society of Cataract and Refractive Surgery (ASCRS) reported a poll in which 1147 members participated [42]. Intracameral antibiotics were injected at the conclusion of surgery by 36% of all respondents. Further:

The incidence of endophthalmitis after intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents may differ depending on various factors. Large series have reported rates ranging from 0.02 to 0.3% per injection [43]. A 2016 series of 503,890 injections reported an overall rate of 0.036%, with no significant differences reported between aflibercept, bevacizumab, and ranibizumab [44]. However, the cumulative risk to each individual patient developing endophthalmitis is typically much higher because most patients may receive a series of injections. The Comparison of Age-Related Macular Degeneration Treatments Trials (CATT) reported a cumulative rate of 0.9% at 2 years [45].

Currently, no RCTs have evaluated alternative approaches for intravitreal injections. Therefore, only “guidelines” based on expert committees have been published [46, 47]. Current guidelines recommend the routine use of povidone-iodine on the ocular surface, reducing aerosolized droplets containing oral isolates and deferring treatment when active external ocular infections are present (Fig. 29.2). However, there is no consensus regarding the routine use of face masks, surgical drapes, eyelid speculums, conjunctival displacement, and the location of injection.

The ideal setting in which intravitreal treatments are performed remains unresolved. In the USA, most intravitreal injections are performed in the outpatient clinic. However, in some European nations, intravitreal injections are performed in an operating room or under similar aseptic conditions. A retrospective series, which directly compared patients injected in a clinic versus a different group of patients injected in an operating room, reported no significant differences in endophthalmitis rates [48].

The use of pre- or postinjection topical antibiotics varies widely [49]. Many retrospective series have reported no statistically significant difference in the rates of endophthalmitis with or without the use of topical antibiotics around the time of intravitreal injection (Table 29.3) [50–59, 61].