Scale back to fight superbugs: A look at antimicrobial usage in your veterinary patients
Carla Johnson, DVM
Carla Johnson, DVM, practices emergency medicine at Berkeley Dog and Cat Hospital in Berkeley, California, and general practice at Cameron Veterinary Hospital in Sunnyvale, California. Her nonveterinary loves are writing, Dressage with her Iberian warmblood mare, Synergy; watercolor painting on yupo; vinyasa yoga; and running with her dog Tyson. Try as she might, her curly-coated Scottish Fold, Hootie, refuses to go jogging with her.
When it comes to bacterial strains like MRSA affecting humans, trying to fight off the resistance isn't futile. Here are questions you can ponder before every case to keep antibiotics going strong in pets.
Hold on tight for an important discussion. (Photo: Abode Stock)
The moment we've all been dreading is here. At a recent Fetch dvm360 conference, Dawn Boothe, DVM, MS, PhD, DACVIM, DACVCP, discussed antimicrobial resistance and stressed that it's time for veterinarians to collectively make big changes in antimicrobial usage. We must at least begin learning how, why, when and what to adjust in our treatment and diagnostic protocols, both to make them more effective and to help make the world safer with regard to antibiotic resistance.
Did you know?
MRSA (methicillin-resistant Staphylococcus aureus) is more dangerous than MRSP (methicillin-resistant Staphylococcus pseudintermedius) because it's both virulent and resistant.
In 2016, the AVMA established an antimicrobial stewardship task force that has been working to assist in preserving the effectiveness and availability of antibiotics in veterinary medicine,1 and this year the California Veterinary Medical Association added mandatory continuing education on responsible antimicrobial use for veterinary licensing and renewals.2
Why? The issue has become critical in the realm of human medicine with the persistence of deadly methicillin-resistant Staphylococcus aureus (MRSA) and the evolution of superbugs like colistin-resistant Escherichia coli.3 It's estimated that 10 million people will die annually of resistant superbugs by 2050, compared with 700,000 in 2015.4
Is resistance a deadly threat in pets?
It appears that veterinary medicine has thus far been spared the deadly strains (although this may be due to less collective reporting), but we are using the same drugs and getting similar resistance, so we are not far behind. Dr. Boothe says that in veterinary laboratories over the last 20 years, bacterial resistance to antibiotics has increased progressively and dramatically, particularly documented for Staphylococcus species and E. coli.
Our ability to empirically predict what “bug” we can expect in any particular infection, and more importantly to “guess” which antibiotic will successfully cure that infection, has plummeted. Some we can still take a stab at, but in many cases predictability has dropped from 90% accuracy to 50%, says Dr. Boothe. A 50-50 chance of the medication working was not likely what you were hoping for when you sent home that amoxicillin-clavulanate for your patient's suspected UTI. If dangerous resistance to the antibiotics that we share with human medicine worsens, another outcome is that we might be in danger of losing drug access, as we have already with certain fluoroquinolones in food animals.
The crux of the matter
Inherent resistance for the bacterial classes has not changed much. It's the acquired resistance that's posing a worldwide problem. We can continue to develop stronger antibiotics, but the bacteria will continue to develop new forms of resistance. Dr. Boothe says, “It's unwise to underestimate an adversary that has had a 3 billion-year evolutionary head start.”
If we can't win this war with better drugs, we must change something else. Even if we ignore the universal nature of the problem and focus directly on our patients, our antibiotic protocols are no longer working the way we think. We might have encyclopedia volumes of empirical evidence in our heads that all scream successful usage, but many times it was really just our healthy average patient's immune system that cured the problem, not us.
So what happens, then, to our really ill or immunosuppressed patients that cannot cure their infection without an accurate antimicrobial regimen? What about the patients with Cushing's disease or sepsis or the geriatric patient? What about those cases that go downhill mysteriously? Are our skills up to the task of both preventing resistant infections and addressing them when they arise? The Centers for Disease Control and Prevention (CDC) reports that the greatest risk factor for the development of antibiotic resistance is the use of antibiotics,5 so we must focus on how we think about and use antibiotics. We are, in fact, putting these drugs not just into animals, but out into the environment as well.
If changing how we choose and use antibiotics feels like a monumental issue, it's because it is. The use of these drugs is ingrained into our practice habits and inseparable from our personal understanding of pathology. There are also many non-world health factors that affect our day-to-day use of these drugs, such as cost, convenience, client compliance and client expectations. Even if we can make a difference, where in the world do we start and how do we do this without hurting our patients and practices? Dr. Boothe makes an easy first proposal: We first need to change the way we think about treating infections. Our way of thinking may be the hardest habit to change, but we need to try. “We need to do a better job,” she says. “At the very least, if we aren't part of the problem, then we need to be part of the solution.”
Resistance in a nutshell
Basically, antibiotics trigger resistance by coming in contact with bacteria without killing them. Multidrug resistance (MDR) is the most concerning form of resistance, but whether resistance stays limited to only one class of bacteria (within class) or impacts other classes (MDR) is a huge issue, says Dr. Boothe.
The way bacteria fight antibiotics and how they confer resistance is multifactorial and too complicated for a brief synopsis, but in general Dr. Boothe says it can happen in just a couple of days or even in minutes. Bacteria can spread resistance either directly to their offspring or to other bacteria through shared genetic material like plasmids (small circular DNA molecules that readily move into normal bacteria and code for resistance). Shared DNA is a favorite of E. coli and a very dangerous weapon because plasmids can cross freely into both gram-negative and gram-positive bacteria, both pathologic organisms and normal flora. With most plasmid-conferred resistance, ceasing antibiotics will generally cause the plasmid, and thus the resistance, to just go away.
E. coli develops resistance to fluoroquinolones through altered mutation in proteins that impact bacterial DNA when the bacterial drug target-site mutates and becomes less susceptible to destruction, causing in-class resistance, says Dr. Boothe. However, with continuous exposure, organisms rapidly turn on cell wall efflux pumps, a type of defense in the bacterial wall that actively expels the antibiotic from bacteria. Many antimicrobials are a target for these pumps, resulting in MDR. Dr. Boothe says the pumps are strategically placed close to a common type of defense that can also result in MDR-cell wall porins. These porins, found particularly in gram-negative organisms, cause the organism to become less permeable to antibiotics. However, if drugs do get through the porins, efflux pumps can turn on after just 10 minutes if an antibiotic concentration at the site of infection is not high enough to start killing bacteria.
Pseudomonas species has always had porins, which is why it is an inherently MDR organism, says Dr. Boothe. Organisms can also simply destroy many antibiotics. The cephalosporin and penicillin antibiotics are common targets for beta-lactamase production from bacterial cell walls, and their use is notorious for stimulating even more versions of the beta-lactamases; there are now over 1,000 different types. Beta-lactamases can actually destroy beta-lactam antibiotics before they reach the site of infection. The most benign beta-lactamase resistance targets lower tier beta-lactam drugs, such as amoxicillin or cephalexin. However, third-generation cephalosporins can trigger a much more dangerous type of beta-lactamase in gram-negative organisms by turning on the production of extended-spectrum beta-lactamases (ESBLs). The genes that code for this resistance are shared horizontally with other classes of bacteria, making them resistant to higher tier drugs that otherwise could target Enterobacteriaceae. Everyone using Convenia (cefovecin) in their practice should be aware that it is a third-generation cephalosporin capable of inducing ESBLs and that these enzymes will target higher-tier cephalosporins that are critically important in human medicine. More recently, they also are targeting carbapenems such as meropenem, which have been one of the last group of antibiotics effective against Enterobacteriaceae.
In her series of lectures on antimicrobial resistance at Fetch dvm360, Dr. Boothe detailed the scope of the problem, how to understand and interpret minimum-inhibitory concentrations (MICs) and how to design the best antibiotic treatment plan. In addition, she provided a plethora of insight. She hit on many questions that veterinarians need to be asking themselves when reaching for antibiotics. Before considering cost and convenience, ask yourself these questions:
1. Do I really need to treat this patient with antibiotics right now?
If you get a positive bacterial culture result, say on a routine urine screen that you happen to have cultured, but the patient is asymptomatic, as a general rule you do not need to treat, says Dr. Boothe. Indeed, you should not. This is especially true if the microorganism shows evidence of multidrug resistance (MDR). It seems counterintuitive, but many E. coli are nonvirulent, and these are often the resistant ones. In fact, when they acquire resistance they usually have to drop their virulence genes to make room. A resistant E. coli UTI that is asymptomatic may self-resolve before it drops its resistance gene and reacquires its virulence gene. If it regains virulence, it may have dropped its resistance genes. If clinical signs appear again, Dr. Boothe says to re-culture to see if lower tier drugs are now potentially effective.
2. Do I have enough proof that a pathologic infection is present?
In patients with complicated infections, or for which the accuracy of predicting the infecting microbe is in doubt, treatment ideally will be based on culture prior to initiation of antimicrobial therapy. Dr. Boothe says to culture early and often, before exposing the bacteria to resistance-triggering antimicrobials. Interpret those results using the patient's “big picture.” Corroborate with additional evidence such as exam findings, cytology and history. Consider whether the test results fit with the patient. If you have a situation that does not make sense, such as fecal bacteria in a urinalysis with no pyuria, or a urine culture with anaerobes or fecal bacteria in a patient with no clinical signs, could you have tipped the gastrointestinal tract with needle during the cystocentesis? If yes, then don't treat, and re-culture if signs develop. Make sure you're convinced that this patient has a bacterial infection.
Third-generation cephalosporins are “uniquely qualified to induce [extended-spectrum] beta-lactamase resistance” in response to antibiotics, says Dr. Boothe. Convenia (cefovecin) is a third-generation antibiotic. It's a great drug for Staphylococcus species and maybe even E. coli but is not even very broad-spectrum. Yet it triggers a multidrug and multiclass form of bacterial resistance. This is a great example of inadvertent irresponsible use of an antibiotic in our profession. The MDR it triggers crosses to other species of bacteria and stimulates resistance in normal flora. It should be used only for what it has been intended: cat bite wounds, or when a culture supports its use over any other.
3. Do I really need to treat this patient with antibiotics at all?
Dr. Boothe says to ask yourself, “Am I sure that systemic antibiotics are the only effective way to treat this infection? Can I treat with something else or rely on local treatment?” Proper wound care is essential to decontaminate and reduce the population of bacteria as much as possible, either to avoid antibiotic use altogether or to allow systemic antibiotics to work effectively. The body has a massive defense system to fight infection. Often our antimicrobial regimen seems to work but, really, the patient would have had the exact same outcome without it. Can you try a nonsystemic modality such as a shampoo or topical medication? Can you just wait it out and recheck the patient in a few days?
Surgical infections that include hardware will not usually resolve with antibiotics until the hardware is removed. Formation of the biofilm on the hardware surface presents a barrier to antibiotic penetration. Organisms within the biofilm are generally much more resistant to antimicrobials. As a general rule, they must be removed if the infection is to be cured.
4. Did I do a thorough workup to properly evaluate for underlying causes?
Paramount to success with recurrent infections is identifying the underlying cause. With recurrent UTIs, for example, you need to do imaging to look for an underlying nidus for infection, such as calculi or a mass. If an underlying cause cannot be found or corrected, Dr. Boothe says it's likely that repetitive antimicrobial therapy will cause bacteria to develop increasing resistance until you are faced with a multidrug-resistant organism. Treat the underlying cause as soon as possible.
5. Do I trust my culture and susceptibility results?
“Your culture data is only as good as the sample you've collected,” says Dr. Boothe. Poor samples and poor sample collection techniques notoriously cause misleading or useless results. Wound swabs are a great example of this. Swabs of infected wounds typically culture only the surface-colonizing organisms that may actually be assisting healing; the bad bugs are deeper. Despite your best swabbing efforts, Dr. Boothe says, 30% or more of present organisms may be missed depending on the type of the wound.
Ear swabs, submission of drain tubes or unprotected endotracheal tube tips, and many urine sampling measures will often produce misleading isolates. Dr. Boothe recommends that you send in tissue samples of infected wounds for culture whenever possible, and use very sterile sampling techniques with all other submissions. Think about where your infection lives, how many bacteria-laden environments you will need to cross to get there, and the best way to minimize exposure to these while accessing the infection site. Sample handling is important as well. Obligate anaerobes can die after 10 minutes of exposure to oxygen. Temperature and humidity are also important and the lab should be consulted regarding the need for transport on ice.
Many E. coli and staphylococci have turnover rates of less than 30 minutes. If samples containing these organisms are not properly refrigerated, within seven hours the size of the population in the sample will expand, making contamination look like an infection, or infection look more severe.
You might assume, like me, that you can just scoop up a smattering of bacteria from your very contaminated and infected wound (or ear) and that your culture will tell you about all of the bugs. Then you can simply formulate a plan to kill them all, contaminants and pathogens alike. It's just not that simple, says Dr. Boothe. The dangerous strains may not survive to show up in your culture and MIC. When you treat all the other bugs, you may successfully reduce the population and (for now) resolve clinical signs but leave the environment unguarded, to be solely inhabited by a superbug like Pseudomonas, a common sequelae to the treatment of recurrent otitis in dogs.
6. Do I know how to interpret the MIC?
For various reasons, the process of interpretation is complex and fraught with exceptions when it comes to “taking this in vitro data, and apply[ing] it to our patient,” Dr. Boothe says. The Clinical Laboratory Standards Institute (CSLI) works hard to assess published data regarding organisms infecting patients and their susceptibility to different drugs. They publish protocols for testing and guidelines for interpreting the results. They publish breakpoint drug data that helps guide whether an isolate MIC qualifies as S, R, or even I. However, despite their best efforts to keep up, the rapid pace at which microbes change makes it difficult to stay current and accurate for certain drugs or bacteria (and often for animals it is still based on human data). Amoxicillin-clavulanic acid is an example of a drug that has been recently updated in response to E. coli's changing susceptibility as well as our understanding of how antimicrobials work. CLSI has provided new breakpoints for this drug and, as a result, interpretation of MIC will change from lots of Ss to many more Rs.
More MIC explanation
If you tested all of the CFUs infecting a patient, you would get a spectrum of MICs, says Dr. Boothe. The MIC from the lab is thus most likely correct for that drug dose 50% of the time (MIC50). The mutant prevention concentration (MPC) is a newer term emerging that describes the highest MIC (MIC100) in that population prior to exposure to any antibiotic. This would be the MIC to target, but labs cannot measure that. Even if you cannot culture a patient, Dr. Boothe says there is some information in the literature. A reasonable target for designing dosing regimens is the MIC90 of the drug for an organism. If 100 dogs with a UTI had an E. coli cultured, and the MIC was determined for each, the MIC90 would be the 90th percentile-that is, the concentration that would inhibit 90% of the 100 organisms. For example, the MIC90 for E. coli and amoxicillin-clavulanic acid is 8 µg/ml. The 90th percentile for the organism (an MIC90) might be published for that drug (for example, on package inserts). In fact, there is a mountain of information out there to help you learn to use calculations to translate all of the above nuances and more into drug protocols, but a mountain it is. So, like me, you just might need to ask for help with interpreting difficult MICs.
Among the problems with culture is that bugs are adaptive and tricky. In the culture medium the bacteria are isolated from each other in a fluid, but “in the patient, these organisms are smart,” and they are determined to survive; they can communicate with each other, Dr. Boothe explains. They live on a flat surface and build communities designed to protect one another. Biofilm is a frequent and formidable barrier to successful therapy. We realize that the way they act in the lab versus in the serum versus in the infected tissue is going to be different, but this is one of the reasons that the CLSI works as hard as it does to come up with protocols and guidelines for culture and susceptibility testing. Because of this, only the culture and susceptibility data collected using their guidelines may be credible.
To add to the muddle, even when you properly culture a sample, the data is limited. For example, the lab is testing only one to several single colony-forming units (CFUs), yet the infecting population is composed of tens to hundreds of thousands of CFU. The larger the infecting population, the more drug molecules are needed. Not only are there more bugs to inhibit, but the bugs will be producing more destructive enzymes and it's likely that at least one CFU will have developed resistance by chance alone, says Dr. Boothe. This is the main reason why “decontaminating” the site may increase the chances of success.
Not all Ss are alike, nor are they absolute. Dr. Boothe says it's important to understand that an S next to a drug does not mean that isolate has not developed any resistance to the drug. It means that CLSI thinks effective concentrations can be achieved for that organism at the dose upon which the testing is based. That dose is not always known. For example, for amoxicillin and E. coli, the dose is 13 mg/kg every 8 hrs. For enrofloxacin, the dosing range is 5 to 20 mg/kg once daily, but as the MIC gets higher, it's important to use that high dose of 20 mg/kg.
7. Is my drug choice, dose or drug interval adequate for what I see on my MIC and for what I know about the drug?
Design the drug treatment plan to leave no survivors. A dead bug cannot be a resistant bug, and “dead bugs don't mutate,” says Dr. Boothe. First, learn to read your MICs for the development of resistance. For example, if any bacteria is I or R to any of the fluoroquinolones, it's on its way to MDR, no matter how many other fluoroquinolones have Ss. However, you still might be able to cure the infection if you treat it immediately and appropriately. Next, “always start with the best drug, but the lowest tiered,” Dr. Boothe advises. For example, Convenia is a third-generation cephalosporin, therefore a “high-tiered” drug, as are the fluoroquinolones. Penicillins and amoxicillin drugs are examples of lowest-tiered drugs.
Quick explanation of tiering
There are several reasons a drug might be a high-tier drug-for example, if the drug is the best at getting a select population of organisms, particularly important pathogens. Imipenems, aminoglycosides and vancomycin are tiered higher because they target Enterobacteriaceae and, for vancomycin, methicillin-resistant staphylococci.
A drug that is potentially toxic is also tiered higher. Aminoglycosides are nephrotoxic, and increasingly we see different toxicities to fluoroquinolones. For Dr. Boothe, if the resistance is multidrug, this is a reason to tier a drug higher. Fluorinated quinolones and higher-generation cephalosporins should be higher tier because of the type of resistance left behind if therapy fails. And importance to human medicine is a reason for veterinarians to tier a drug higher. The World Health Organization has indicated which drugs are critically important to human medicine and veterinarians should be careful when using them.
Once a decision is made to use a drug, “get in quick, hit hard and get out quick,” says Dr. Boothe. The more at risk your patient is for therapeutic failure, the more important it is to hit hard with a higher dose, or more frequent dosing regimen, depending on the drug type. “Never use a fluoroquinolone unless you are willing to use it at as high a dose that you can,” Dr. Boothe pleads. In most cases, we are not using high enough doses of the concentration-dependent drugs like aminoglycosides and fluoroquinolones. These drugs usually need to reach a minimum “wipe-out” concentration at the site of an infection, one that will kill all of the bacteria, but generally need only to be dosed once a day. You may have to double the dose of a fluoroquinolone just to make it effective: if the organism has a high MIC, Dr. Boothe does not hesitate to add a second high dose (10 to 20 mg/kg for enrofloxacin in dogs) to increase the chance of success. For aminoglycosides, the risk of toxicity will be increased if a second dose is added, so occasionally she will add a different antibiotic to a short course of an aminoglycoside.
For time-dependent drugs (basically everything else), particularly penicillins and cephalosporins, Dr. Boothe says we are often not keeping the drug concentration above MIC for long enough at the tissue site. Because penicillins have very short half-lives (one to two hours), it's very hard to maintain concentrations above the MIC unless you are willing to dose the patient every eight hours or more. Cephalosporins have a longer half-life in general, and thus it may be easier to use them at 12-hour dosing intervals. But for organisms with high MIC, the dosing needs to be more frequent. Also, look at your inoculum size: Dr. Boothe says the more CFUs present in your culture, the more drug you will need to cure it. You may need to use multiple drugs to adequately treat an infection if you can't safely increase the dose or increase the frequency safely. If you have access, consult with a microbiologist or clinical pharmacologist for advice on your drug plans whenever you do a culture and MIC, especially if the results are resistant or unexpected.
Did you know?
Culture and susceptibility data is based on plasma drug concentrations, but the infection is in the tissue, says Dr. Boothe. Many drugs only make it to the infection site at 50% of the plasma concentration. Water-soluble drugs such as beta-lactam antibiotics and aminoglycosides have an even harder time penetrating into infection sites.
And did you know ... ?
In dogs and cats, enrofloxacin is partially converted to ciprofloxacin in blood. Ciprofloxacin has very poor bioavailability in dogs-40%, says Dr. Boothe-so you need at least a double dose compared with enrofloxacin. In cats, the bioavailability is 0% to 20%, which is completely useless.
8. Do you understand mechanisms of toxicity for each of your antibiotics?
If not, research them, consult with a specialist, or both. The mechanism of action of the drug is not going to be the same as the mechanism of toxicity, so there are often things that can be done to increase effectiveness without causing toxicity. And there are often some things that can be done to reduce toxic effects, says Dr. Boothe. If you are going to use higher-than-labeled dosages of a drugs, make sure you understand which patients are at higher risk, what you can do to lower those risks, and whether you really need to use that dose or that drug. For example, you can increase the effectiveness and reduce the renal toxicity of aminoglycosides by giving the entire 24-hour dose of the drug (with t.i.d. drugs, give the total day's dose once in the morning for dogs, once in the evening for cats) and give saline-containing fluids subcutaneously at the same time to flush it out the kidneys.
Never use ciprofloxacin in cats; it is extremely retinotoxic.
9. How do we know when to use prophylactic antibiotics?
Dr. Boothe would advise as little prophylactic antibiotic-use as possible. In a patient with really severe dental disease, a short course of antibiotic prior to the cleaning procedure is reasonable. Also, some orthopedic procedures and ophthalmologic surgeries are routinely done with prophylactic antibiotics, as the risk of complications tends to overshadow the risk of resistance. For non-contaminated elective surgeries in other categories, antibiotics should not be used. Guidelines for everything in-between still appears to be a work in progress or an individual case-based decision.
Blinded by the light: The coolest fact
Dr. Boothe says the reason enrofloxacin causes retinal blindness in cats is that they are missing one of the p-glycoprotein-like efflux pumps in their retinas. The retina can't kick the fluoroquinolones out, and they accumulate. When the feline retina is exposed to light radiation, phototoxicity occurs. Keeping cats away from light while on the drug can prevent this. Marbofloxacin at 11 times the recommended dose for two weeks did not cause phototoxicity in cats, so we could just use marbofloxacin instead of enrofloxacin in cats, says Dr. Boothe. Orbifloxacin and pradofloxacin are also safer than enrofloxacin for feline eyes.
Where to go from here
The beauty and burden of our profession is that we are forced to think for ourselves. Antibiotics are essential tools for our practice, but we need to use them more wisely. If nothing else, we need to understand them better, because there are no black-and-white guidelines for us at this time, and there may never be. Our widespread comfort with using cefovecin for convenience is telling for our lack of understanding. It has demonstrated efficacy against streptococci and Staphylococcus pseudintermedius, but the MIC90s for this drug and E. coli (and even Staphylococcus aureus) are at least four times higher, making it harder to maintain concentrations above the MIC for an adequate period of time.
Because of this, cefovecin may not be the best drug for treating these other organisms. Since cefovecin resistance is likely to involve ESBL, it is also a drug that should not be used routinely. Unfortunately, for many of us, this is the drug of choice for nearly every cat that requires antibiotics, is fractious or has a noncompliant owner, and it has seemed benign. We can do better and we can become more informed. At a minimum, we choose the antimicrobial CE lectures at our conferences, or take online courses focusing on antimicrobial use, and get more informed. Soon, more definitive dosing and drug-choice guidelines should become available to us. In the meantime, if you do not need to use antibiotics, do not use them. If you are going to use them, make sure you gather enough information about why you are going to use them and how.
1. AVMA establishes antimicrobial policy for veterinary medicine. Am Vet Available at: https://www.americanveterinarian.com/news/avma-establishes-antimicrobial-policy-for-veterinary-medicine.
2. New continuing education requirement-use of medically important antimicrobial drug. Available at: https://www.vmb.ca.gov/licensees/antimicrobial_ce.shtml
5. CDC. Antibiotic/antimicrobial resistance (AR/AMR). Available at: https://www.cdc.gov/drugresistance/about.html