The Endophthalmitis Vitrectomy Study (EVS) has been the cornerstone of post-cataract surgery bacterial endophthalmitis management for nearly three decades. [6] It recommended the presenting vision as a guide for primary treatment (TAP-INJ and VIT-INJ) and using two intravitreal antibiotics (vancomycin and ceftazidime). However, despite this well-designed study and its legacy, there are limitations related to the presenting vision-guided treatment choice and restricting vitrectomy to eyes with light perception or worse. [8, 9, 10, 11] Over the years, we have a better understanding of infection-associated inflammation in endophthalmitis. Also, with the newer technology of small gauge vitrectomy and superior fluidics, there is greater safety of vitreous surgery than in the 1990s when the EVS was designed.
The EMS is close to a real-world study that recruited people with postoperative endophthalmitis despite disease severity. It adopted three important modifications from the EVS: (1) vitrectomy as the primary treatment based on the presenting inflammation score; (2) a second choice of intravitreal antibiotic combination directed to (more virulent) gram-negative infection; (3) using newer molecular technology in microbiology work-up. These modifications were based on our earlier studies: that a significant proportion of gram-negative bacterial infection occurs in India and many parts of Asia, [12, 13] that inflammation score at presentation truly reflects the severity of endophthalmitis, [2] that there is increasing resistance of gram-negative infection to ceftazidime, [14] and that next-generation sequencing is superior in detecting microorganisms in culture-negative and mixed infection endophthalmitis. [4] Also, our decision to retain vancomycin and consider imipenem as an alternative to ceftazidime in the second intravitreal antibiotics combination was based on our earlier reports of antibiotic susceptibility trend over a quarter century [14] and susceptibility of imipenem to gram-negative bacterial infection, specifically to Pseudomonas species. [15] Incidentally, there was a higher proportion of gram-negative infections and a higher proportion of multi-drug resistant (MDR) bacterial infections in the current cohort. Both phenomena are possibly explained by the referral patterns to our facility. We had reported 5.2% MDR bacterial endophthalmitis in 2000–2007 from our facility, and in 78.6%, it was a gram-negative infection; [16] in the current cohort, there was 46% MDR bacterial infection, and it included 86.2% gram-negative infection.
Three lessons from the interim analysis of EMS in endophthalmitis management are:
-
An inflammation score could be a good indicator of endophthalmitis severity
-
Imipenem is not necessarily a better alternative to ceftazidime in gram-negative infection (in India)
-
Molecular techniques should be included in microbiology work-up (if cost permits)
These three discussed in more details below
-
An inflammation score is a better indicator of disease severity.
Like infection, inflammation plays an equally significant role in endophthalmitis. The inflammatory cascade activated by the specific toxic effects of the pathogen ultimately determines the final anatomical (structural) and functional (visual) outcome. [17] If not treated with anti-inflammatory drugs, inflammation could continue even after the infection is controlled. [18, 19] We have validated the inflammation score earlier. [2, 20] Our study showed that the eyes with a higher IS often had a gram-negative infection (Supplementary Table 3). The presenting vision did not differentiate between gram-positive and gram-negative infection.
-
Imipenem is not necessarily a better alternative to ceftazidime in gram-negative infections.
The current combination of intravitreal antibiotics, per the EVS, is directed against gram-positive cocci (vancomycin) and gram-negative bacilli (ceftazidime). [6] Gram-negative infection is more challenging to treat because of their inherent property of producing a variety of tissue-damaging enzymes and the rapid development of antibiotic resistance. [21]
Ceftazidime is a third-generation cephalosporin antibiotic that works against gram-negative and gram-positive aerobe bacteria with good stability against certain β-lactamase enzymes. Imipenem is a β -lactam antibiotic highly resistant to the β-lactamase enzymes produced by multiple drug-resistant gram-negative bacteria. In our study, the susceptibility-resistance pattern of ceftazidime and imipenem against gram-negative bacilli was similar. In the current cohort, gram-negative bacilli had better susceptibility to colistin. Colistin is a polymyxin class of drug. It interacts with the bacterial outer membrane causing a bactericidal effect. It is effective against several gram-negative bacilli, including multi-drug resistant infections. [22]
-
Molecular techniques should be included in the microbiology work-up.
Culture is the current gold standard of microbiology work-up in endophthalmitis. It allows for antimicrobial susceptibility testing and information on the minimum inhibitory concentration. But culture-negative endophthalmitis is not uncommon, and these are also not necessarily TASS (Toxic Anterior Segment Syndrome). We have reported increased instances of negative culture in endophthalmitis (51–69%) over 25 years. [14] In addition to possible overdiagnosis, other factors such as small sample size, sequestration of microorganisms in the capsular bag after an intraocular lens surgery, prior antibiotic therapy, and delay in processing the samples are the other reasons. The newer molecular methods such as polymerase chain reaction (PCR), s real-time or quantitative PCR (RT-PCR; q PCR), Sanger sequencing, and targeted NGS are the new additions. [23, 24]
In our study, culture positivity nearly doubled from 29–56%, with the addition of the PCR and NGS. The Sanger sequencing is superior to PCR because it makes species identification. The targeted NGS identifies the species normally not detected by conventional culture methods, and it is specifically valuable in polymicrobial infection. In this cohort, 46% (29 of 63) of microorganisms were multidrug resistant. At least half of the gram-negative bacilli susceptible to ceftazidime were resistant to imipenem, and the ones susceptible to imipenem were resistant to ceftazidime. While susceptibility to colistin was far superior to other antibiotics, 3 of 23 gram-negative ceftazidime-susceptible bacilli were also resistant to colistin.
Fungal endophthalmitis
The series of 15 microbiology-positive fungal infections did not allow us to make a definite conclusion though the current results suggested that primary vitrectomy is a required intervention. These patients should receive an intravitreal antifungal agent injection and vitrectomy as soon as a microbiology confirmation is available. [25, 26]
Interim analysis
Interim analysis, a standard practice in large and long duration clinical trials, evaluates the current data in an ongoing trial, in which the primary research question is addressed and which has the potential to modify the conduct of the study. [27, 28] The interim analysis in the current study showed that colistin could be a better alternative to imipenem and ceftazidime.
Weakness and strength
Weakness. It is an interim analysis; the final results could change once the remaining patients (n = 188) are recruited. Current recruitment includes only 5.6% (n = 14) people with non-cataract endophthalmitis and 6% (n = 15) with fungal infection; these are not enough in quantity to arrive at a definitive conclusion. Furthermore, these data are from a tertiary-level referral center; hence these do not reflect the causative microbiological spectrum of endophthalmitis in India.
Strength. By recruiting patients of any severity, the EMS is closer to the clinic population as one encounters in patient care. Measurement of the inflammation score allowed us to assess the disease severity of the affected eye.