It was a cold November morning in my second year of residency when a young immunosuppressed patient, Sara, came in for a Foley catheter problem. It was a complicated exchange, whether from strictures, difficult anatomy, or operator error, and it took a few attempts to get the new catheter in place. Soon enough, however, she was discharged.
It took only a few hours for Sara (not her real name) to return, floridly septic and rapidly decompensating. The prolonged manipulation of the urinary tract had clearly seeded the bloodstream, and we were witnessing the fast progression of deadly gram-negative sepsis. In just minutes, she was started on broad-spectrum antibiotics, fluids, and vasopressors, had central and arterial lines placed, and was intubated. We watched, helpless, as she died.
We all strive to be evidence-based, but we often operate in spheres devoid of data or develop clinical idiosyncrasies based on experience. I despise Ortho-Glass, for instance, and insist on plaster as the final word in fracture fixation. Because of various failures and foibles, I prefer Xeroform for wound dressings, always think about PEs when patients mention shortness of breath at the top of stairs, and administer haloperidol and lorazepam for agitated patients myself. When Foley placements are complicated or traumatic, I have a low threshold for a prophylactic dose of ceftriaxone. I could scavenge sources for an evidence-based justification for these practices, but the truth is they're things I do because I've been burned. I think we all have practices like these, and I have a closely held, evidence-free confession: I use aminoglycosides.
The Eagle Effect
Sara didn't go home that day, but I did. I spent years developing a framework for decompensating septic patients that rests on the critical stalwarts we know are of paramount importance in treating sepsis—fluid administration, source control, early broad-spectrum antibiotics—and includes procedural interventions that facilitate life-sustaining treatment.
As time passed, I learned about the Eagle effect and the niche role of aminoglycosides, and I transitioned from the ideals of evidence-based medicine into the real and pragmatic world of evidence-informed medicine. It is at this precipice of real-world practice where studies and statistics often do not exist, are not of high quality, or have modest-to-marginal effect sizes that we change that we push forward the boundaries of care and not only develop experience but define the next advances in EM. (West J Emerg Med. 2019;20:573; https://bit.ly/2Ufgo2m.)
Harry Eagle, who was studying the bactericidal activity of penicillin, described a puzzling phenomenon in 1948 for some strains of bacteria, later called the Eagle effect, a paradoxical nonreaction of disseminated infections to antibiotics, particularly those most effective against actively dividing bacteria. In patients with a significant enough infectious load, some bacteria persist in the stationary growth phase, protecting those microbes from antibiotics that act by inhibiting synthesis or binding actively expressed binding proteins. The Eagle effect has been described for a remarkable range of gram-positive and -negative bacteria and mycobacteria exposed to different antibiotic classes with diverse chemical structures, cellular targets, and sites of action. (Trends Microbiol. 2019;27:339.)
We use the Eagle effect routinely in clinical practice, though we may not be aware of it, such as using clindamycin for disseminated or necrotizing soft tissue infections or adding tobramycin to the end of a VAP bundle. It's not too heroic a mental jump, I have reasoned, to think that there may be a terminal spectrum of sepsis where the bacterial inoculum is just too great to respond to doses of Zosyn and vancomycin, where rapidly progressing gram-negative sepsis might best be addressed by an antibiotic able to have an effect on any stage of the bacterial cell life cycle.
Make a Difference
Aminoglycosides—gentamicin, amikacin, tobramycin—irreversibly bind to the 30S subunit of bacterial ribosomes, inhibiting and interrupting protein translation. They are particularly active against aerobic, gram-negative bacteria and act synergistically against certain gram-positive organisms. We don't seem to use them much in emergency medicine, enamored as we are with the broad-spectrum bazookas that meet our needs the overwhelming majority of the time, but aminoglycosides have long been deployed alongside beta-lactams or vancomycin, harnessing synergistic transport across bacterial cell walls.
Admittedly, a systematic review of randomized clinical trials has not shown an advantage to adding an aminoglycoside to a beta-lactam antibiotic for treating most clinical infections (Int J Antimicrob Agents. 2011;37:491), especially when newer broad-spectrum antipseudomonal beta-lactam antibiotics are used, though a notable exception includes infective endocarditis, where gentamicin is routinely recommended due to evidence of synergy and improved outcomes. (Circulation. 2015;132:1435; https://bit.ly/2JVt4JK.)
What remaining clinical evidence exists is weak at best and generally points away from significant clinical benefit in most patients (Int J Antimicrob Agents. 2020;55:105839; Clin Infect Dis. 2017;64:1731), but good arguments can be made that few, if any, studies have thoroughly examined the population of interest—rapidly deteriorating and critically ill ED patients refractory to other interventions.
Aminoglycosides certainly shouldn't be given without due consideration for risks, particularly the specter of nephrotoxicity or the possibility of sterilizing an unknown source prior to control, but the potential advantages to combination antimicrobial therapy may ultimately increase the likelihood that the infective pathogen will be susceptible to at least one of the components of an empiric combination regimen, perhaps sidestepping an extrapolated Eagle effect and may harness synergistic antibiotic effects. (Clin Microbiol Rev. 2012;25:450; https://bit.ly/36lo4FL.)
There are without a doubt two ways to interpret this evidence. Absent definitive data, bedside approaches will vary based on each clinician's deductions and experience. Since the day Sara died, I've worked to parse the literature, searching for an approach that might have made a difference for her. I've read, reasoned, and prescribed, all prompted by that day. Careful and thoughtful use of aminoglycosides may have made a difference for Sara, and may have a place in treating the dying and decompensating septic patient in the ED.
Dr. Pescatoreis the chief physician for the Delaware Division of Public Health and an emergency physician at Albert Einstein Health Network in Philadelphia. He is also the host with Ali Raja, MD, of the podcast EMN Live, which focuses on hot topics in emergency medicine:http://bit.ly/EMNLive. Follow him on Twitter@Rick_Pescatore, and read his past columns athttp://bit.ly/EMN-Pescatore.