Secondary Logo

Journal Logo


Aqueous level abatement profiles of intracameral antibiotics: A comparative mathematical model of moxifloxacin, cefuroxime, and vancomycin with determination of relative efficacies

Arshinoff, Steve A. MD, FRCSC1,2,*; Felfeli, Tina MD3; Modabber, Milad MD, MSc, FRCSC4

Author Information
Journal of Cataract & Refractive Surgery: November 2019 - Volume 45 - Issue 11 - p 1568-1574
doi: 10.1016/j.jcrs.2019.06.009
  • Free


In a recent paper, Arshinoff and Modabber1 reviewed the evidence for intracameral (IC) antibiotic prophylaxis in cataract surgery and derived a mathematical model for the abatement rate and thus the duration of efficacy of moxifloxacin. Mathematical models require a number of assumptions, and with the emergence of new evidence, we have been able to improve upon our constants and assumptions in our model. This paper expands on the previous study by objectively evaluating the three most commonly used IC antibiotics—moxifloxacin, cefuroxime, and vancomycin—with updated mathematical analysis. Herein, we sought to demonstrate the relative efficacy of the three drugs in the context of postoperative endophthalmitis (POE) prophylaxis, and to produce graphs that can be used to select the antibiotic with the greatest likelihood for success versus local pathogens.

POE is an infrequent (˜1:1000 cases globally), yet potentially devastating cause of permanent and severe vision loss in about one third of cases.2 Despite universal aseptic surgical protocols and topical antibiotic prophylaxis for cataract surgery, culture-confirmed POE continues to be a serious problem.3,4 Adequate draping, well-sealed incisions, and preoperative topical povidone–iodine are important proven methods in reducing infection rates.4,5 Yet, preoperative and postoperative topical antibiotics, despite their common use, remain unproven in their efficacy and fail to attain intraocular bactericidal levels.6–11 In its landmark study,12 the European Society of Cataract and Refractive Surgery (ESCRS), conducted a prospective multicentered randomized placebo-controlled trial that demonstrated a 5-fold decrease in endophthalmitis incidence with prophylactic IC cefuroxime. No additional benefit of topical postoperative levofloxacin was observed. Multiple studies have since confirmed the findings of this pivotal study of the benefit of prophylactic IC antibiotics.1 Consequently, the use of IC antibiotics in POE prophylaxis has been steadily rising globally.

Arshinoff and Modabber1 developed a mathematical model to highlight the optimal dose, concentration, and method of administration of moxifloxacin for IC endophthalmitis prophylaxis. They concluded that injection of 0.4 mL of 150.0 μg/0.1 mL moxifloxacin as an exchange of the anterior chamber (AC) volume at the end of surgery offered the safest and most effective IC administration approach for POE prophylaxis. This method is also more likely than alternative protocols to leave the desired dose in the AC. This is opposed to the practice of injecting only 0.1 mL in the AC, which can be difficult to do consistently with accuracy,1 or performing an AC washout with very low dose moxifloxacin, leaving insufficient drug concentration postoperatively to be effective.13

When choosing an IC antibiotic, it is important to understand antibiotic kinetics. Moxifloxacin, a fourth-generation fluoroquinolone, is a dose-dependent agent inhibiting DNA gyrase (a topoisomerase II) and topoisomerase IV. Both cefuroxime and vancomycin are time-dependent agents that act by inhibiting bacterial cell wall synthesis. In a dose-dependent antibiotic, at sufficiently high concentrations above the minimum inhibitory concentration (MIC) for a given bacterium, referred to as mutant prevention concentration (MPC) (MPC = 8 to 10 times the MIC for fluoroquinolones), the rate of killing is maximal, and the opportunity for resistance development is minimized.14–16 The MPC is defined as the MIC of the least susceptible single-step mutant of a given bacteria, a concept most applicable to fluoroquinolones.17 Moreover, because of the post-antibiotic effect, dose-dependent antibiotics continue to have a bactericidal effect even after the AC concentration has fallen below the MIC. The post-antibiotic effect, first described for quinolones and aminoglycosides, is defined as persistent suppression of bacterial growth after a brief exposure (1 to 2 hours) of bacteria to an antibacterial agent even in the absence of host defense mechanisms. The duration of the post-antibiotic effect is increased by the time and dose of exposure.18 This extended effect varies for different bacteria and drugs; however, it is typically around 3 hours for fluoroquinolones.19

In contrast, both cefuroxime, a second-generation cephalosporin, and vancomycin, a complex glycopeptide, exhibit time-dependent antibacterial activity. As such, they depend upon exceeding a minimum time duration (usually 2 to 3 hours) of drug exposure at exceeding the MIC to be effective. The efficacy of time-dependent antibiotics is optimized by dosing strategies that maximize the duration of drug exposure at a given level, usually frequent intervals of repeated low doses (orally or intravenously), which is an unrealistic method for IC administration. For time-dependent antibiotics, approximately 4 to 6 times for the MIC is considered to be the optimal concentration at which killing is maximal and the risk for resistance development is lowest, whereas no benefit is gained by exceeding these concentrations.20

During the first three postoperative days, the most commonly identified causative organisms of POE are methicillin-resistant coagulase-negative Staphylococcus (MRCoNS) and methicillin-sensitive coagulase-negative Staphylococcus (MSCoNS) 62%, methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive Staphylococcus aureus (MSSA) (11%), and Streptococcus species (11%).21 Recently, a few reports of highly resistant isolates to systemic levels of cephalosporins and fluoroquinolones have appeared.22 The highest reported moxifloxacin MICs are 32.0 μg/mL for MRSA and 64.0 μg/mL for CoNS.23 Although methicillin resistance remains prevalent among staphylococcal isolates from ocular infections, cefuroxime has a reported MIC90 (MIC of 90% of strains tested) of 1.5 μg/mL for MRSA and 2.0 μg/mL for CoNS.24 The most recent Antibiotic Resistance Monitoring in Ocular Microorganisms (ARMOR) study did not measure the MIC90 of isolates to cefuroxime, but for moxifloxacin, listed the MIC90 of MSCoNS at 16.0 μg/mL, and of MRCoNS at 64.0 μg/mL, whereas the MIC90 of MSSA was 0.06 μg/mL and MRSA was 32 μg/mL. The ARMOR study25 found that vancomycin exhibits a MIC90 of 1.0 μg/mL and 2.0 μg/mL against MRSA and CoNS, respectively, and demonstrates high efficacy against all staphylococcal isolates.25 The MIC90s of cefuroxime and vancomycin are highest and identical for CoNS. Therefore, in preparing the graphs in this study, they were plotted as a single line to represent both.

Materials and methods

In this study, it was assumed that the clearance rates from the AC of all three antibiotics are similar, and that the chief cause of concentration decline is simple progressive dilution through aqueous circulation, yielding a classic picture of exponential decay mathematics. This assumption is only a first-order approximation, and there are limitations to this approach that somewhat offset each other. The listed serum half-life of cefuroxime is only about one fifth that of vancomycin and moxifloxacin (which are similar). Aqueous secretion of 2.0 μL/min might be transiently altered during the postoperative inflammatory state, but in turn, it might be partially balanced by altered postoperative trabecular outflow resistance, which can decrease flow and cause transient postoperative spikes in intraocular pressure. Aqueous flow through the newly pseudophakic aqueous volume of 500.0 μL causes continuous dilution of the IC antibiotic; however, the capsular bag is a relatively sequestered reservoir decreasing the overall effective dilution rate. Clearly, many of these postoperative factors somewhat counterbalance each other, and the fine points can thus be neglected in a first-order approximation. The injected doses of all three drugs considerably exceed what is required to be bactericidal to common endophthalmitis pathogens in the early postoperative period. Because the concern is with the duration of effect of each antibiotic for comparative purposes, the precipitous concentration decline can be acknowledged and then somewhat ignored within the first hour to 25% of the injected levels, as measured by Montan et al.,26 without hypothesizing on potential causes and thus focusing on the longer-term abatement rates.

To calculate the abatement rates of IC antibiotic concentrations, the injection dose of moxifloxacin was 600.0 μg in 0.4 mL into the newly aphakic 500.0 mL space, which yields an AC concentration of 1200.0 μg/mL immediately after injection.1 For cefuroxime and vancomycin, the standard injection doses of 1.0 mg in 0.1 mL was used for all calculations, yielding an initial AC concentration of 2000.0 μg/mL. These calculations are in keeping with the clinical measurements of Montan et al.,26 who measured the immediate post-injection IC level of cefuroxime injected into soft eyes (to prevent loss) to be 2742.0 μg/mL (indirectly suggesting that the AC volume in their soft eyes averaged 0.365 mL [1.0 mg injected ÷ 2.742 mg/L = 0.365 mL]), which is consistent with this study’s calculations of a full 0.5 mL AC when the newly pseudophakic eye is not soft.

The rate of abatement of an antibiotic in the AC can be calculated by initially using the Montan et al.26 objective measurement of declining to one fourth of the initial value over the first hour.

This abatement is likely caused by early postoperative stabilization of the eye attributable to volume expansion of the initially soft eyes, early incision leakage, tissue absorption, and other pharmacokinetic issues.

After the first-hour post-cataract surgery, the eye is expected to reach a state of equilibrium with an estimated aqueous turnover rate of 2.0 μL/min. Given the total newly aphakic or pseudophakic AC volume of 500.0 μL, the calculated standard exponential decay rate (exchange is 2.0 μL of 500.0 μL every hour) becomes 0.7866 each hour. This yields the expected half-life of IC antibiotics to be 2.89 hours in the AC, which is consistent with previous studies27–29 that estimated the half-life of vancomycin in the AC to be 2 to 2.67 hours. For the purposes of this study, it was assumed that the washout rates of all three antibiotics from the AC are similar. The calculated abatement rates that yield the expected IC concentrations at any given time postoperatively have only an incomplete correlation to the relative potency of the different drugs (which is also a function of their MICs, time dependence or dose dependence, and other factors).29 The assumed constant for the abatement rate can be gradually refined as more objective data becomes available in the literature in the future.


The generated data were used to draw three figures. Figure 1 shows the calculated expected abatement rate of moxifloxacin in the AC on the background of the MIC90s for the indicated bacteria from the ARMOR study. When 600.0 μg of moxifloxacin was injected into the newly aphakic AC of 0.5 mL volume, the initial aqueous concentration achieved was 1200.0 μg/mL. Over the first hour, this level dropped by a factor of 4, to 300 μg/mL, which was still 5000 times higher than the MIC90 for MSSA published by the ARMOR study, and 500 times higher than the MPC (10 times MIC90) for the same strains. Even after 14 hours, the concentration of moxifloxacin in the AC was 13.2 μg/mL, still exceeding the MPC of 0.6 μg/mL by more than 22 times. Only at 27 hours after injection did the IC moxifloxacin dilute to below the MPC, and at 37 hours, to below the MIC90 of MSSA. However, because of the mechanism of action of moxifloxacin, rare mutant strains might occasionally surface with unusually high MICs. In the ARMOR study, rare strains of MRSA and CoNS were reported to have MICs of 32.0 μg/mL and 64.0 μg/mL, respectively. In such cases, moxifloxacin would still have been effective for 10.5 and 7.5 hours, respectively, as shown in Figure 1 by the horizontal lines representing those MIC90s.

Figure 1
Figure 1:
Abatement rate of IC moxifloxacin 600.0 μg in 0.4 mL against the background of the MIC90s for the indicated strains from the ARMOR study. The calculated abatement of the concentration of moxifloxacin in the AC shows that the AC level will not fall below the ARMOR-reported MIC90 of MSSA, the most frequent pathogen, for nearly 37 hours, or below its mutant prevention concentration for 27 hours. Even for the most resistant strains ever reported, ARMOR CoNS MIC90 = 64.0 mg/L and ARMOR MRSA MIC90 = 32.0 mg/L, the level of moxifloxacin exceeds those MICs for 7.5 hours and 10.5 hours, respectively (AC = anterior chamber; ARMOR = Antibiotic Resistance Monitoring in Ocular Microorganisms study22,25; CoNS = coagulase-negative Staphylococcus; IC = intracameral; MIC = minimum inhibitory concentration; MIC90 = minimum inhibitory concentration of 90% of strains tested; Moxi = moxifloxacin; MPC = mutant prevention concentration [10 × MIC for dose-dependent fluoroquinolones]; MRSA = methicillin-resistant Staphylococcus aureus; MSSA = methicillin-sensitive Staphylococcus aureus).

Figure 2 shows the comparative abatement rates of the three antibiotics—moxifloxacin, cefuroxime, and vancomycin—depicted against the background resistances of the targeted bacteria. Vancomycin and cefuroxime are indicated on the same abatement line as their injected doses, which are identical (1.0 mg), and their MIC90s for CoNS (highest MICs) are also identical. Both cefuroxime and vancomycin were injected at doses of 1000.0 μg/0.1 mL, which is 1.67 times the injected dose of moxifloxacin, and this ratio persisted over time in this model. The initial concentrations of cefuroxime and vancomycin were identical at 2000.0 μg/mL, and then declined to 25% or 500 μg/mL after the first hour. By 17 hours post injection, the concentrations of cefuroxime and vancomycin, at 10.74 μg/mL, fell below 6 times the ARMOR MIC90 for CoNS (12 μg/mL, the optimal killing concentration for time-dependent antibiotics). By hour 24, the concentrations of cefuroxime and vancomycin fell to 2.0 μg/mL—the ARMOR MIC90 level of CoNS—and were therefore no longer effective. In fact, because of their time dependence, they had not been effective for the preceding approximately 2.5 to 3 hours. It is clear that the separation (indicated by vertical double-ended arrows between the moxifloxacin abatement line, and the resistance lines in blue), exceeded the corresponding separation for vancomycin and cefuroxime (red vertical double-ended arrows next to the corresponding blue arrows).

Figure 2
Figure 2:
Abatement rates of IC moxifloxacin 600.0 μg in 0.4 mL and cefuroxime–vancomycin 1.0 mg in 0.1 mL against the background of the MIC90s for the indicated strains from the ARMOR study. The calculated abatement rates of moxifloxacin (blue) and cefuroxime–vancomycin (red) in the AC shows that the separation between the AC drug concentration and target MICs is greater for moxifloxacin (double-ended blue arrows) than for vancomycin or cefuroxime (adjacent double-ended red arrows) (AC = anterior chamber; ARMOR = Antibiotic Resistance Monitoring in Ocular Microorganisms study22,25; Cef = cefuroxime; CoNS = coagulase-negative Staphylococcus; IC = intracameral; MIC = minimum inhibitory concentration; MIC90 = minimum inhibitory concentration of 90% of strains tested; Moxi = moxifloxacin; MPC = mutant prevention concentration [10 × MIC for dose-dependent fluoroquinolones]; MSSA = methicillin-sensitive Staphylococcus aureus; Vanc = vancomycin).

Figure 3 shows the “bactericidal efficacy” of all three antibiotics calculated by shifting the abatement lines shown in Figure 2 in opposite directions to account for the post-antibiotic effect of the dose-dependent moxifloxacin of 3 hours as well as the extra 3 hours required for cefuroxime and vancomycin because of their time dependence. This was done by holding the dose of moxifloxacin level from hour 1 (300.0 μg/mL) for 3 hours before recommencing further abatement, because, due to the post-antibiotic effect, the effective bactericidal concentration at any given time would be the does-dependent antibiotic’s concentration 3 hours earlier. Alternatively, the levels of cefuroxime and vancomycin were diminished by an extra 3 hours dilution from the first to second hours (ie, from 500.0 μg/mL at hour 1 to 191.4 μg/mL at hour 2), due to the extra time required for a given dose to have its bactericidal effect, because a time-dependent antibiotic’s effective bactericidal capacity at a given time is a function of its declining concentration 3 hours later. Other than for these adjustments to the curves, the same abatement rate of 0.7866/hour was used throughout. The corrections for dose-dependent and time-dependent antibiotics increase the disparity of the separation of the moxifloxacin curves compared with the cefuroxime–vancomycin curves. Extension of these graphs illustrates that an IC injection of moxifloxacin 600.0 μg remains bactericidal, above the MPC of MSSA for over 29 hours, finally falling below the MIC90 of 0.06 μg/mL at 40 hours post injection. Conversely, for cefuroxime and vancomycin, the “effective dose” fells below 6 times the CoNS MIC90 (its optimal level) at 14 hours, and below the MIC90 at 21 hours. Therefore, the bactericidal efficacy of cefuroxime and vancomycin last only about half as long as that of moxifloxacin.

Figure 3
Figure 3:
Abatement rates of IC moxifloxacin 600.0 μg in 0.4 mL and cefuroxime–vancomycin 1.0 mg in 0.1 mL, considering the time-dependent delayed antibiotic effect and the dose-dependent post-antibiotic effect. The abatement rates of IC moxifloxacin and cefuroxime–vancomycin corrected for their bactericidal effect difference because of dose-dependence of moxifloxacin and time-dependence of cefuroxime and vancomycin. The resulting times over which moxifloxacin will exceed the MPC and MIC90 of ARMOR MSSA are 30 hours and 40 hours, respectively; whereas the time that cefuroxime or vancomycin remain above the most effective 6 times the MIC90 of CoNS is 14 hours, and above the MIC90 is 21 hours. IC moxifloxacin therefore remains bactericidal in the AC for about twice as long as cefuroxime or vancomycin. If the efficacy of one of the above drugs in the AC is desired to be determined against any local bacterium, a straight line need only be drawn at the log of its MIC. The MIC90 of the most moxifloxacin-resistant ARMOR MRCoNS (64.0 μg/mL) have been plotted, indicating that moxifloxacin AC levels will exceed the MIC90 for 10 hours (AC = anterior chamber; ARMOR = Antibiotic Resistance Monitoring in Ocular Microorganisms study22,25; Cef = cefuroxime; CoNS = coagulase-negative Staphylococcus; DD = dose-dependent; IC = intracameral; MIC = minimum inhibitory concentration; MIC90 = minimum inhibitory concentration of 90% of strains tested; Moxi = moxifloxacin; MPC = mutant prevention concentration [10 × MIC for dose-dependent fluoroquinolones]; MRCoNS = methicillin-resistant coagulase-negative Staphylococcus; MSSA = methicillin-sensitive Staphylococcus aureus; PAE = post-antibiotic effect; TD = time-dependent; Vanc = vancomycin).

In addition, as shown in Figure 3, the most resistant ARMOR MRCoNS with MIC90 of 64.0 μg/mL was plotted to illustrate the reduction in efficacy of moxifloxacin when the most resistant strains of MRCoNS were encountered. Here, the AC moxifloxacin concentration fell below the ARMOR MRCoNS MIC90 at just after 10 hours. However, this should be viewed in comparison to the presence of time-dependent resistant strains (Entrococci for cefuroxime and gram-negatives for vancomycin), which are not depicted at all and are never included in resistance calculations in the literature. In those cases, bacterial resistance to these time-dependent antibiotics is absolute, with no dose possibly high enough to be bactericidal. These pathogens are simply considered resistant and are therefore not included in the calculation of the MIC90 of sensitive strains because their inclusion would yield MIC90s of infinity. Unlike cefuroxime and vancomycin, none of the bacteria commonly implicated in endophthalmitis are completely resistant to moxifloxacin. Therefore, the impression gained by examination of MICs of rare strains most resistant to moxifloxacin out of the clinical context of actual “failures” of prophylaxis to prevent endophthalmitis by any bacterium is misleading.


The evidence for the benefit of IC antibiotics in POE prophylaxis is mounting. In November 2017, Chang et al.30 published a study of more than 1 000 000 eyes from the Aravind Eye Care System. They observed a 3-fold reduction in POE rates with IC moxifloxacin prophylaxis compared with those who did not receive IC antibiotics (0.02% vs 0.07%; P < .0001) in manual small-incision cataract surgery cases, and they found a 6-fold reduction in cases of phacoemulsification (0.01% vs 0.07%; P < .001).31 This data is consistent with our model for moxifloxacin and the broadly reported literature.1

Our experience with IC moxifloxacin, spanning more than 9000 cases since November 2004, has also been very encouraging. To date, we have experienced no deleterious side effects from the administration of the moxifloxacin hydrochloride (Vigamox) preparation for IC antibacterial prophylaxis at the end of cataract surgery. On postoperative day 1, we have not seen any cases of observable fibrin in the AC since we began using prophylactic IC moxifloxacin in 2004, suggesting that IC moxifloxacin is not toxic to the interior of the eye, and that postoperative day 1 fibrin might be attributable to subclinical infection. In vitro studies have reported relatively low toxicity of moxifloxacin, with Kernt et al.32 finding no deleterious effect with doses up to 150.0 μg/mL. Haruki et al.33 also studied cultured corneal cell endothelial membrane damage with exposure to moxifloxacin, cefuroxime, and levofloxacin, and they found that moxifloxacin caused damage to endothelial cell membranes only at levels exceeding 500.0 μg/mL for 6 to 24 hours or more, whereas decreased cell viability was found with levels exceeding 1000.0 μg/mL for 24 hours (2000.0 μg/mL for cefuroxime). The AC moxifloxacin concentration achieved at surgery can be calculated because we know that the total dose injected almost immediately dilutes into the 0.5 mL newly aphakic AC. Surgeons generally inject 500.0 to 600.0 μg of moxifloxacin. So, the final achieved initial concentrations are between 1000.0 to 1200.0 μg/mL, which falls to 400.0 to 452.0 μg/mL by 2 hours postoperatively. Despite millions of eyes having been injected with these doses to date, no one to our knowledge, has ever published a case of observed corneal toxicity from moxifloxacin. We want to maximize the injected dose so that the injection is bactericidal to the most resistant strains reported; however, increasing it beyond what has already been commonly used and reported should be done cautiously in light of the aforementioned in vitro toxicity studies.

The real bactericidal test of our mathematical model lies in assessing whether it is consistent with literature reports of global experience with IC antibiotics. We have chosen to study and compare moxifloxacin, cefuroxime, and vancomycin because they are the three most commonly used IC antibiotics for POE prophylaxis globally.

In 2013, Asena et al.34 evaluated the ocular pharmacokinetics, safety and efficacy of IC moxifloxacin 0.5% solution in a rabbit model. After IC injection of 500.0 μg moxifloxacin in 0.1 mL, the measured half-life in the rabbit AC was found to be 2.2 hours, similar to our calculations. However, at 0.5 hours post injection, the measured aqueous concentration of moxifloxacin was only 15.88 μg/mL ± 1.86 (SD), that is, only about 1% of the injected dose, whereas Montan et al.26 had found 25% in humans. Although protein binding and other pharmacokinetic issues could be contributory, leakage of the IC antibiotic after injection remains the most likely cause of this very significant loss. This highlights the difficulty of injecting 0.1 mL accurately, with no loss. Alternatively, dilution of the antibiotic to allow exchange of the AC volume at the end of surgery thus leaving 500.0 to 600.0 μg in the AC appears to be a more reliable means of achieving the desired antibiotic dose retained in the AC.

Libre and Matthews35 conducted laboratory in vitro studies, whereby they incubated bacteria isolated from cases of endophthalmitis along with vancomycin, cefuroxime, moxifloxacin, and combinations of the three antibiotics. The antibiotic levels tested were high, corresponding to clinical maximum IC doses (IC injections of 1.0 mg for vancomycin and cefuroxime; 0.5 mg for moxifloxacin), or low (one third of the “high” doses). Their conclusions were: (1) Low concentrations were generally ineffective, except against Streptococci and Propionibacterium. (2) Only moxifloxacin was effective against all isolates. (3) Vancomycin and cefuroxime were inactive versus Pseudomonas aeruginosa. (4) Elimination of Staphylococcus and Pseudomonas required the maximum doses. They concluded that moxifloxacin gave the best coverage of the studied IC antibiotics, but that it should be dosed at 0.5 mg or higher.

Most recently, Bowen et al.11 published a detailed meta-analysis of studies of the past 20 years, which included 17 studies and more than 925 793 eyes, regarding the safety and efficacy of the same three IC antibiotics. They found the overall POE rates were: IC cefuroxime = 0.0332, IC moxifloxacin = 0.0153 and IC vancomycin = 0.0106. No additional benefit of topical antibiotics was found. They expressed concern about dilution errors, toxic anterior segment syndrome, macular toxicity, and a risk for allergy with cefuroxime, as well as risk for endophthalmitis with resistant bacteria such as Pseudomonas. Their concern with vancomycin was the lack of gram-negative coverage and the recent reports of hemorrhagic occlusive retinal vasculitis. Overall, moxifloxacin was found to have the best safety profile and the broadest antibacterial spectrum. They commented that a 500.0 μg injection dose seems preferable to 100.0 μg.

The graphs we have generated are consistent with every paper we could find dealing with IC antibiotic prophylaxis of POE. They show a visual representation of the amalgamation of a large volume of information, and to the best of our knowledge, they are accurate at present. They may be relied upon to ascertain the relative efficacy of each of the three antibiotics to local pathogens, as long as the doses injected are the same as we used for the calculations and the MICs of the pathogens are known for the three antibiotics. These data permit the drawing of horizontal resistance lines (logs of MICs) on Figure 3, illustrating local resistance and suggesting which antibiotic would be best in a given situation. This has been very useful for us, and hopefully will be for the reader.

Our study is limited in that it is a first-order approximation of the kinetics of abatement of IC antibiotic levels. Over time, the abatement rates of each antibiotic will become clearer, as they have since our first study1 2 years ago, and our models will be able to be improved. However, the greatest limitation herein is the data we relied upon to generate the horizontal bacterial resistance lines on each graph. Bacterial resistance levels change as new mutations are discovered. Mutation is the mechanism that forms the basis of evolution and the process will continue. MIC values will thus likely change over time. Hopefully, new antibiotics will appear that will yield better IC prophylaxis in the never-ending war with pathogenetic bacteria.

Moreover, the difference in behavior of time-dependent and dose-dependent antibiotics cannot be overstated. Each bacterium has its own unique behavior to each unique drug and the classification into behavior types is only an approximation of a very complicated process. With time, we will gain a deeper understanding of each antibiotic’s abatement rate within the AC and the resistance mechanisms of the bacteria that we seek to eliminate. Abatement rate graphs will require periodic reassessment and updating. Nevertheless, models, such as the ones we have created, not only help us now but serve to point out areas where our knowledge is incomplete and requires further investigation.

In summary, we were able to derive first-order mathematical IC concentration abatement graphs for the three most commonly used prophylactic antibiotics (cefuroxime, moxifloxacin, and vancomycin). The resulting curves are consistent with the literature on IC antibiotics for POE prophylaxis. For optimum utility of the graphs, one must inquire of local laboratories for the MICs of local endophthalmitis pathogens to the three drugs, calculate their logs, and plot them on printouts of Figure 3. One can then observe the anticipated comparative efficacy profile of the three antibiotics to local endophthalmitis pathogens.

Pharmacotherapy of antibacterials is quite complicated and depends on issues that are not stagnant. For now, our preferred prophylactic agent is moxifloxacin 600.0 μg in 0.4 cc, prepared by adding 7.0 mL of a balanced salt solution to the 3.0 mL of moxifloxacin contained in a bottle of Vigamox, and administered as the very last step of surgery. However, for the longer term, in a world of constantly evolving bacterial resistance, the search for ideal antiinfective IC therapy to be used at the end of surgery is never over. Models that reflect current understanding will also require periodic updating.

What Was Known

  • Intracameral (IC) administration of prophylactic antibiotics significantly reduces the rate of postoperative endophthalmitis, irrespective of the background rate. Concerns have been raised about dilution errors, mostly with cefuroxime, and the lack of commercially prepared prediluted single-use preparations of all of three antibiotics (cefuroxime, moxifloxacin, and vancomycin).
  • No case of dilution error with moxifloxacin leading to a detrimental effect has been reported in the literature. Recent concern over vancomycin has occurred because of the emergence of hemorrhagic occlusive retinal vasculitis.

What This Paper Adds

  • Using updated methodology for the abatement rate determination methodology with respect to moxifloxacin, graphs were created to compare results between moxifloxacin, cefuroxime, and vancomycin. These graphs provide a useful tool to understand the current literature comparing the relative efficacy of the three antibiotics, and for surgeons to use to understand which of the three antibiotics has the greatest likelihood of success in their local environment.


1. Arshinoff SA, Modabber M. Dose and administration of intracameral moxifloxacin for prophylaxis of postoperative endophthalmitis. J Cataract Refract Surg 2016;42:1730-1741.
2. Gower EW, Lindsley K, Nanji AA, Leyngold I, McDonnell PJ. Perioperative antibiotics for prevention of acute endophthalmitis after cataract surgery. Cochrane Database Syst Rev 2013; 7:CD006364
3. Slean GR, Shorstein NH, Liu L, Paschal JF, Winthrop KL, Herrinton LJ. Pathogens and antibiotic sensitivities in endophthalmitis. Clin Exp Ophthalmol 2017;45:481-488.
4. Gentile RC, Shukla S, Shah M, Ritterband DC, Engelbert M, Davis A, Hu DN. Microbiological spectrum and antibiotic sensitivity in endophthalmitis: a 25-year review. Ophthalmology 2014;121:1634-1642.
5. Olson RJ, Braga-Mele R, Chen SH, Miller KM, Pineda R 2nd, Tweeten JP, Musch DC. Cataract in the adult eye preferred practice pattern. Ophthalmology 2017;124:P1-P119.
6. Yamada M, Mochizuki H, Yamada K, Kawai M, Mashima Y. Aqueous humor levels of topically applied levofloxacin, norfloxacin, and lomefloxacin in the same human eyes. J Cataract Refract Surg 2003;29:1771-1775.
7. Fukuda M, Yamada M, Kinoshita S, Inatomi T, Ohashi Y, Uno T, Shimazaki J, Satake Y, Maeda N, Hori Y, Nishida K, Kubota A, Nakazawa T, Shimomura Y. Comparison of corneal and aqueous humor penetration of moxifloxacin, gatifloxacin and levofloxacin during keratoplasty. Adv Ther 2012;29:339-349.
8. Cagini C, Piccinelli F, Lupidi M, Messina M, Cerquaglia A, Manes S, Fiore T, Pellegrino RM. Ocular penetration of topical antibiotics: study on the penetration of chloramphenicol, tobramycin and netilmicin into the anterior chamber after topical administration. Clin Exp Ophthalmol 2013;41:644-647.
9. Yoshida J, Kim A, Pratzer KA, Stark WJ. Aqueous penetration of moxifloxacin 0.5 % ophthalmic solution and besifloxacin 0.6 % ophthalmic suspension in cataract surgery patients. J Cataract Refract Surg 2010;36:1499-1502.
10. Relhan N, Forster RK, Flynn HW Jr. Endophthalmitis: then and now. Am J Ophthalmol 2018;187:xx-xxvii.
11. Bowen RC, Zhou AX, Bondalapati S, Lawyer TW, Snow KB, Evans PR, Bardsley T, McFarland M, Kliethermes M, Shi D, Mamalis CA, Greene T, Rudnisky CJ, Ambati BK. Comparative analysis of the safety and efficacy of intracameral cefuroxime, moxifloxacin and vancomycin at the end of cataract surgery: a meta-analysis. Br J Ophthalmol 2018;102:1268-1276.
12. Endophthalmitis Study Group, European Society of Cataract & Refractive Surgeons. Prophylaxis of postoperative endophthalmitis following cataract surgery: results of the ESCRS multicenter study and identification of risk factors. J Cataract Refract Surg 2007;33:978-988.
13. Matsuura K. Advantages of intracameral injection with higher volume diluted moxifloxacin. J Cataract Refract Surg 2017;43:709-711.
14. Hesje CK, Tillotson GS, Blondeau JM. MICs, MPCs and PK/PDs: a match (sometimes) made in hosts. Expert Rev Respir Med 2007;1:7-16.
15. Levison ME. Pharmacodynamics of antimicrobial drugs. Infect Dis Clin North Am 2004;18:451-465.
16. Lode H, Borner K, Koeppe P. Pharmacodynamics of fluoroquinolones. Clin Infect Dis 1998;27:33-39.
17. Smith HJ, Nichol KA, Hoban DJ, Zhanel GG. Stretching the mutant prevention concentration (MPC) beyond its limits. J Antimicrob Chemother 2003;51:1323-1325.
18. Sharma KK, Sangraula H, Mediratta. Some new concepts in antibacterial drug therapy. Indian J Pharmacol 2001;34:390-396.
19. Drusano G, Labro MT, Cars O, Mendes P, Shah P, Sörgel F, Weber W. Pharmacokinetics and pharmacodynamics of fluoroquinolones. Clin Microbiol Infect 1998;4(Suppl 2):S27-S41.
20. Drusano GL. Antimicrobial pharmacodynamics: critical interactions of ‘bug and drug’. Nat Rev Microbiol 2004;2:289-300.
21. Yannuzzi NA, Si N, Relhan N, Kuriyan AE, Albini TA, Berrocal AM, Davis JL, Smiddy WE, Townsend J, Miller D, Flynn HW Jr. Endophthalmitis after clear corneal cataract surgery: outcomes over two decades. Am J Ophthalmol 2017;174:155-159.
22. Sahm DF, Morris TW. Antibiotic resistance among ocular pathogens in the united states five-year results from the antibiotic resistance monitoring in ocular microorganisms (ARMOR) surveillance study. JAMA Ophthalmol 2015;133:1445-1454.
23. Asbell PA, Colby KA, Deng S, McDonnell P, Meisler DM, Raizman MB, Sheppard JD Jr, Sahm DF. Ocular TRUST: nationwide antimicrobial susceptibility patterns in ocular isolates. Am J Ophthalmol 2008;145:951-958.
24. Sueke H, Kaye S, Neal T, Murphy C, Hall A, Whittaker D, Tuft S, Parry C. Minimum inhibitory concentrations of standard and novel antimicrobials for isolates from bacterial keratitis. Investig Ophthalmol Vis Sci 2010;51:2519-2524.
25. Asbell PA, Mah FS, Sanfilippo M, Decory HH. Antibiotic susceptibility of bacterial pathogens isolated from the aqueous and vitreous humor in the Antibiotic Resistance Monitoring in Ocular Microorganisms (ARMOR) surveillance study. J Cataract Refract Surg 2016;42:1841-1843.
26. Montan PG, Wejde G, Setterquist H, Rylander M, Zetterström C. Prophylactic intracameral cefuroxime: evaluation of safety and kinetics in cataract surgery. J Cataract Refract Surg 2002;28:982-987.
27. Arshinoff SA, Modabber M. Response: advantages of intracameral injection with higher volume diluted moxifloxacin. J Cataract Refract Surg 2017;43:709-713.
28. Mendívil AS, Mendívil MP. The effect of topical povidone–iodine, intraocular vancomycin, or both on aqueous humor cultures at the time of cataract surgery. Am J Ophthalmol 2001;131:293-300.
29. Murphy CC, Nicholson S, Quah SA, Batterbury M, Neal T, Kaye SB. Pharmacokinetics of vancomycin following intracameral bolus injection in patients undergoing phacoemulsification cataract surgery. Br J Ophthalmol 2007;91:1350-1353.
30. Relhan N, Schwartz SG, Grzybowski A, Flynn HW Jr. Re: Haripriya et al.: Endophthalmitis reduction with intracameral moxifloxacin prophylaxis: an analysis of 600 000 surgeries (Ophthalmology. 2017; 124:768–775). Ophthalmology 2017;124:e77-e78.
31. Haripriya A, Chang DF. Intracameral antibiotics during cataract surgery: evidence and barriers. Curr Opin Ophthalmol 2018;29:33-39.
32. Kernt M, Neubauer A, Liegl RG, Lackerbauer CA, Eibl KH, Alge CS, Ulbig MW, Kampik A. Intracameral moxifloxacin: in vitro safety on human ocular cells. Cornea 2009;28:553-561.
33. Haruki T, Miyazaki D, Matsuura K, Terasaka Y, Noguchi Y, Inoue Y, Yamagami S. Comparison of toxicities of moxifloxacin, cefuroxime, and levofloxacin to corneal endothelial cells in vitro. J Cataract Refract Surg 2014;40:1872-1878.
34. Asena L, Akova YA, Goktaş MT, Bozkurt A, Yaşar U, Karabay G, Demiralay E. Ocular pharmacokinetics, safety and efficacy of intracameral moxifloxacin 0.5% solution in a rabbit model. Curr Eye Res 2013;38:472-479.
35. Libre PE, Mathews S. Endophthalmitis prophylaxis by intracameral antibiotics: in vitro model comparing vancomycin, cefuroxime, and moxifloxacin. J Cataract Refract Surg 2017;43:833-838.


None of the authors has a financial or proprietary interest in any material or method mentioned.

© 2019 by Lippincott Williams & Wilkins, Inc.