Aminoglycosides: an update on indications, dosing and monitoring

[Music] Welcome to the Australian Prescriber podcast. An independent, no-nonsense podcast for busy health professionals.

I'm Dhineli Perera, your host for this episode, and I'm unashamedly wrapped to be chatting to Dr Carly Hughes about aminoglycosides. Aminoglycosides have made a bit of a rockstar comeback, all thanks to its rapid bactericidal activity and relatively low rates of resistance in Australia.

Dr. Carly Hughes is an infectious diseases physician and clinical microbiologist at Townsville University Hospital, Queensland. She's also a senior lecturer at the School of Medicine at James Cook University Townsville in Queensland. Carly, together with her formidable team, including Hazel Moore, Daniel Yeoh, Ed Raby, and Indy Sandaradura have gifted us with a really beautiful summation of the changes to aminoglycoside dosing and monitoring recommendations in the Therapeutic Guidelines: Antibiotic, which were implemented earlier this year. Welcome to the program, Carly.

Thanks so much for having me. Excited to be here.

So Carly, aminoglycosides. We were in favour of their use and then we fell out of favour, and now they're back in the good books. Why has there been so much back and forth when it comes to prescribing aminoglycosides?

Yeah, absolutely. Aminoglycosides have had a really long and convoluted history in Australia and there definitely still is some institutional variation when it comes to their use. So when we're talking about aminoglycosides, obviously, we're referring to gentamicin, tobramycin and amikacin being the ones we use most commonly.

I think people are sometimes put off by their use because of their narrow therapeutic range that you have to screen the patients for risk factors for toxicity, then you've got to look for the toxicity. You need to individualise their dosing, you need to individualise their monitoring. And then there was this new guidance which we'll get to for some breakpoint-setting bodies, CLSI [Clinical and Laboratory Standards Institute] and EUCAST [European Committee on Antimicrobial Susceptibility Testing], about avoiding monotherapy. But on the flip side, they're still highly effective agents for Gram-negative infections. We do have increasing concern about resistance to other agents that are used to treat Gram-negatives where aminoglycosides largely retain activity. So that's why, in my opinion anyway, they're back in favour.

So there's no avoiding it really. We've got to get back to comfort and using them again comfortably. So with this back and forth in mind, Carly, are there any other advantages and disadvantages to take into consideration when prescribing aminoglycosides?

Yeah. So we'll start with the advantages. Most community- and healthcare-associated Gram-negative pathogens are susceptible to aminoglycosides. We know that they are associated with rapid control of Gram-negative infections and they're bactericidal agents with that peak. But then the added benefit is the postantibiotic effect where the bacteria keep being killed for many hours after the plasma concentration is undetectable, and then that means that we could dose them in a convenient once-daily dosing.

Also, when you combine them with cell-wall–active drugs like a beta-lactam, they can be synergistic for difficult-to-treat infections like Enterococcus and streptococcal endocarditis. They rarely cause allergy or hypersensitivity reactions. And the other advantage is that they are rarely associated with Clostridioides difficile, known as C. diff infection.

The disadvantages are well known. Primarily, they cause nephrotoxicity, which is usually associated with prolonged treatment. So talking about longer than 5 to 7 days and pre-existing kidney impairment. The nephrotoxicity that we see is generally reversible, though not always. And then the other one is that they cause ototoxicity, so vestibular and, less commonly, auditory toxicity. And that's, again, mostly associated with prolonged courses. And unfortunately, those vestibular and auditory toxicities are often irreversible.

Right. So those lists are in the article and I would definitely recommend listeners to check them out if you're not already familiar with some of those advantages and disadvantages that Carly has just mentioned. Carly, there does seem to be controversy around when you can and can't use aminoglycosides as monotherapy. Could you clear this up for us a bit?

Yeah, so this has stemmed from some recent updates by 2 of our leading international susceptibility testing and reporting guidelines. They are CLSI, which is the American body, and EUCAST, which is the European. Australian laboratories use either or sometimes a combination of both of those breakpoint guidelines. And what they have both done in the last few years is reviewed their aminoglycoside breakpoints.

When these bodies do that, they look at preclinical data, they look at laboratory data, they look at PK/PD [pharmacokinetic/pharmacodynamic], they look at clinical data, outcome data, and they look at the MIC [minimum inhibitory concentration] distributions. And although there hasn't been a lot of new literature in this field, when they've re-reviewed the data that they had, they have recommended that breakpoints for aminoglycosides, and when I say breakpoints, I mean how we determine whether we report an isolate as susceptible standard dose, susceptible increased exposure or resistant, that they are only to be interpreted into those categories if they are used either in combination with another active agent from another class or if you have undergone a source control procedure. So that's where not recommending use in monotherapy comes from.

The exception there is infections that originate in the urinary tract because we know that aminoglycosides are heavily concentrated in the renal tissue and in the urine. So, in infections originating from the urinary tract, monotherapy is generally deemed acceptable.

In the Therapeutic Guidelines: Antibiotic, we usually recommend aminoglycosides for infections outside the urinary tract together with another antibiotic or with source control. For example, in the intra-abdominal infection chapter, source control is really recommended in most scenarios. However, we do recognise that sometimes the second agent is unlikely to have Gram-negative cover. So the key example there is the ampicillin–gentamicin–metronidazole therapy in, again, intra-abdominal infections.

Ampicillin at the moment only covers about 50% of E. coli isolates and much less for other Gram-negatives. So, in that scenario, the aminoglycoside is essentially being used as monotherapy. We did extensively discuss that and we have continued to recommend aminoglycosides because amongst the expert committee, they felt that there had not been a signal that clinical failures were occurring and there is substantial clinical experience with their use. So that's why they've gone down that path.

If there is microbiological results that suggest that you have an organism that is sensitive to the aminoglycoside that you're using and it's not in the urinary tract, is that a time that you can use that as monotherapy?

Yeah, so I think ideally in that scenario, once you have other susceptibilities available, that would be the time that you're switching to an alternative agent, just to avoid the toxicities of prolonged aminoglycosides use or potential toxicities. So, let's say for example, you had a pseudomonas in blood and its susceptible increased exposure to piperacillin/tazobactam, that would be the time that you would switch to that.

The other option is source control. So I think if you were happy that you had really adequate source control, then potentially monotherapy if that was the only option available would be reasonable. But again, once you've had appropriate source control after a couple of days, it's reasonable to be considering an oral switch at that point. And a lot of the guidelines are pushing for early oral switch. So I think there are limited scenarios where you would consider using aminoglycosides monotherapy unless you really didn't have any other options for infections outside of the urinary tract.

So a really resistant organism perhaps might be that time that you're sort of stuck in the corner and you have to use it, but outside of that, there's usually a safer alternative.

Yeah, absolutely.

Thank you for clarifying that. Your article also neatly delineates 3 types of categories for aminoglycoside indications. Do you mind stepping us through these?

Yeah, absolutely. So the 3 categories that we talk about using aminoglycosides are empirical therapy. That's where we're primarily recommending them in the guidelines. The second category is that of initial directed therapy when you're waiting for the results of the susceptibility testing to come back. And then the third category is that of directed therapy.

So the first category, the empirical therapy is when we're using short-term therapy empirically for serious infections. When you suspect a Gram-negatives at play, for example, intraabdominal infections, urinary tract infections, sepsis of unclear source, and then we're recommending that they're used for a short duration and then rationalise to directed therapy once you know what you're dealing with.

For the directed therapy pending results of susceptibility, that's only in the specific scenario where you know you've got a Gram-negative infection, but you suspect that other empirical antibiotics may be resistant. So, for example, you know that the patient is colonised with a multidrug-resistant Gram-negative and you choose to use an aminoglycoside while you're waiting for formal susceptibility results. It's also recommended as part of combination therapy for serious Pseudomonas aeruginosa infections.

The last category is that of directed therapy and that's pretty uncommon, but if you have a confirmed Gram-negative infection that's resistant to antibiotics that are more appropriate for longer term use, sometimes we do recommend combination therapy with aminoglycosides, for example, brucellosis, nontuberculosis mycobacterial infection. And then the other category is synergistic therapy, so where you use it with another antimicrobial and that's largely for treatment of streptococcal and enterococcal and less commonly Bartonella endocarditis.

That again is beautifully presented in the article. So the readers that are interested in really understanding those 3 categories a bit better, definitely check that out. So Carly then, what are our alternatives to aminoglycosides, and when is it appropriate to use them?

So we want to be having a hard think about who has precautions and contraindications to use of aminoglycosides. And in those patients, a broad-spectrum beta-lactam, for example, piperacillin+tazobactam, amoxicillin+clavulanic acid, ceftriaxone, cefotaxime, they can all be used in that situation. Which one you choose depends on what you're trying to treat. So the indication for the antibiotics, but also what is the expected pathogens.

We have also, in this edition of the Antibiotic guidelines and also in the paper, outlined that beta-lactams can be used upfront initially in empiric therapy for those who you think the empiric therapy needs to continue for more than 72 hours. Now, that can be a pretty difficult decision to make at the patient's bedside sometimes, but for example, if you've got a young patient in the ED [emergency department] with appendicitis and they're on the E-board and you know they're going to have their surgery done within 72 hours, then it is reasonable to start them on an aminoglycoside-containing regimen.

Whereas if you have a more complex patient, they're unlikely to get source control, you might have to leave them on intravenous antibiotics for a longer time, and that's the sort of patient that you might start on the non-aminoglycoside–containing regimen, so the beta-lactam regimens that I've just discussed that avoids the need to switch. Some clinicians were a bit uncomfortable with that switching after 72 hours to a different regimen because we don't recommend empiric aminoglycosides generally after 72 hours of use.

The other category is the carbapenems like meropenem, generally as part of antimicrobial stewardship, we really try to reserve those for scenarios where they are really needed upfront. For example, someone with sepsis or septic shock from a urinary source and the patient is at risk of multidrug-resistant Gram-negatives, they might be someone who had been hospitalised or in a long-term facility with a known outbreak of multidrug-resistant Gram-negatives or a country that has a high prevalence of multidrug-resistant Gram-negatives. They might be the person who you would use a carbapenem in upfront, but I think that scenario is less likely.

So there are some exceptions to it, but as you've described, it is quite clear. And worst case scenario, you could use your aminoglycoside for those first 72 hours. And if unexpectedly the patient is continuing empiric therapy, then there is the option to change at that point?

Yeah, absolutely.

So I was also hoping you could break down the information surrounding the optimal management for your Enterobacterales versus your pseudomonal infections and just take-home messages for listeners: which aminoglycoside can we use, and why?

The first consideration is you have to think about it in your setting. So what are the pathogens that you see? What's your local antibiogram like, what aminoglycoside do you have to hand? I know some hospitals in Queensland where I work have gentamicin in their after-hours stock, whereas some have tobramycin.

So what do you have access to? And also, on that same side of the coin, what therapeutic drug monitoring do you have access to? And then lots of places have local guidelines, state guidelines, national guidelines, and they should be followed where possible. However, gentamicin and tobramycin are the 2 that are most commonly used for treatment of Gram-negative infections, and that's because amikacin has got a broader spectrum and, generally, we're trying to reserve it for Gram-negative infections that are resistant to gentamicin and tobramycin. It does have some Gram-positive cover. So for example, we use it in things like nontuberculous mycobacterial infections. So for Enterobacterales, as I mentioned, this is the E. coli, the Klebsiella, and the whole host of other Enterobacterales, either gentamicin or tobramycin can be used.

The resistance data across Australia really varies and ESBL [extended-spectrum beta-lactamase]–producing E. coli for example, the prevalence of it ranges very widely across Australia, but we know that about 70% of them are gentamicin susceptible. So gentamicin or tobramycin for Enterobacterales is fine depending on your local data and availability.

Pseudomonas is a bit of a different story. Tobramycin is preferred over gentamicin for treatment of pseudomonas infections. We know that in pseudomonas isolates, tobramycin has got a 2-fold lower MIC than gentamicin. And because they've got such similar pharmacokinetic and pharmacodynamic profiles and similar adverse events, tobramycin is preferred. EUCAST and CLSI, they've actually removed gentamicin breakpoints for pseudomonas. So if laboratories are following those guidelines, they might not even release a gentamicin result for pseudomonas and they will only release tobramycin. So that's something else to take into consideration.

At this point, it is probably worth highlighting that these changes haven't really impacted how we treat the nonTB [tuberculosis] mycobacterial infections and synergistic treatment of Gram-positive infective endocarditis. Have I got that right?

Yeah, that's correct. So as I mentioned, amikacin, we do use for nontuberculous mycobacteria because it's got activity whereas gentamicin and tobramycin don't. There is one exception called Mycobacterium chelonae where we do use tobramycin, but by and large, it would be amikacin if we need it. And the other thing is synergistic therapy of Gram-positive endocarditis, specifically streptococci and enterococci most commonly. In certain circumstances, we do still recommend that synergistic gentamicin, and it is just gentamicin in that scenario because there's more evidence for its use and more clinical experience. So we do recommend gentamicin for synergistic dosing.

So now to the dose, which dose and which weight is recommended for initial dose calculations by the guidelines now in adults for aminoglycosides?

So it's 7 mg/kg for adults for gentamicin and tobramycin, and it's 30 mg/kg for amikacin. Now, we do need to undertake dose adjustments for adults with kidney impairment if they're not septic and don't require ICU [intensive care unit] support. If they are septic or do require ICU support, then the 7 and the 30 mg/kg dose is fine.

For kids, more than 1 month old, the dose of gentamicin and tobramycin similarly is 7 mg/kg, but the amikacin dose is 20 mg/kg. Neonates, it varies according to gestation and chronological age. Doses should generally be followed by therapeutic drug monitoring, but those with altered pharmacokinetics, the ICU patient who's critically ill, augmented renal clearance, those on dialysis, they may need dose modification, and the calculators do address that somewhat. That dosing is obviously not applicable to synergistic therapy or NTM [nontuberculous mycobacteria] treatment where the dosing is different.

In terms of weight metrics, as we know, the dose depends on your kidney function and the volume of distribution of the drug, and they're related to body weight. Therapeutic Guidelines has decided to use, and this is a bit of a change from previously, decided to use lean body weight to calculate doses for all patients, including underweight patients, including obese patients. And that's based on a study where it performed better than total and ideal body weight in estimating gentamicin volume of distribution.

As I mentioned, there are now guidelines that include calculators in the Therapeutic Guidelines for initial dosing of all the 3 aminoglycosides, but you do need expert input from your ID [infectious diseases] pharmacist or ID physician or microbiologist for dosing in adults if they are obese, greater than 35 kg/m² BMI [body mass index]. In kids, fat-free mass is the most accurate. But for practicality, the Therapeutic Guidelines just recommends actual body weight in kids who are not obese. And again, in kids with obesity, expert guidance is recommended.

So all those calculators are available and worth checking out if you haven't already. You did touch on therapeutic drug monitoring. When do you believe that it should be utilised and how do you recommend listeners go about interpreting these levels?

Yeah, so there are some local and statewide guidance and protocols for aminoglycoside therapeutic drug monitoring. So wherever possible, they should be followed. Generally, you don't need to do TDM or therapeutic drug monitoring if the aminoglycoside is going to be stopped within those 72 hours. However, you should consider it even after the first dose if the patient's kidney function is changing rapidly or in patients with altered pharmacokinetics like obese patients.

It is recommended from the first dose, however, if you suspect that the therapy is going to exceed 48 hours, because that enables you to adjust it quickly based on your therapeutic drug monitoring.

We've now swapped to AUC [area under the concentration–time curve] monitoring preferred over trough concentration monitoring because it more accurately predicts the efficacy and the toxicity. And we're going for an AUC between 80 to 100 for gentamicin and tobramycin for our empirical Gram-negative treatment. But once you've got the MIC of the pathogen, you can adjust your targets for directed therapy.

And there's a couple of different ways that you can get your AUC. Increasingly, I think people are using these model-informed precision dosing using Bayesian dosing software, and there's a number of different options for that. The alternative is manual calculation methods, which has the slight downside that it needs at least 2 plasma concentrations, whereas the Bayesian dosing software's predictions can be made with just a single concentration.

And the other thing is that the trough concentrations aren't suitable for efficacy, but they can indicate accumulating drug, and they can be used to minimise toxicity. So for those targets, we're aiming for less than 1 mg/L for gentamicin and tobramycin, less than 4 for amikacin. Generally, the Therapeutic Guidelines is recommending multiple-daily dosing for synergy for gentamicin. So then we're recommending trough concentrations and we're aiming for less than one.

That's a great summary of some of the TDM recommendations. And so then leading from that, how does this fit in with monitoring for aminoglycoside toxicity? You did touch on how AUC can inform us on that. How does it all tie in?

Yeah. And again, there's a box in our paper addressing this, but as I mentioned, the 2 major toxicities that we're worried about are nephrotoxicity and vestibular and auditory toxicity. In terms of nephrotoxicity, ideally, we should have a kidney function before we're starting in the aminoglycosides. The exception to that is the critically unwell patient in ED where you don't have time to potentially wait for renal function measurement to come back. They need antibiotics imminently, but most patients, you know their renal function before you're starting.

And then during therapy, we recommend monitoring kidney function 2 to 3 times weekly or more frequently if it's changing. And in fact, clinically what we see sometimes is when we are using our Bayesian monitoring and the dose is going down, that's sort of an early indication that their kidneys are going off and they're the patients who you might measure renal function more frequently.

In terms of the vestibular and auditory toxicity, it's really critical to have a chat to your patients and educate them about what to look for. And there are some patient handouts available for that as well. But critically, we're asking them to report either balance problems or hearing issues, and important that they know that that can continue even post-treatment. So regularly assessing them for gait ataxia, imbalance, oscillopsia, which is when the objects jump around in your field of vision, or blurred vision during head movement, and then hearing loss is the other one.

To monitor the vestibular function, so that sort of balance, you can compare the visual acuity at rest, and then you do it during passive sinusoidal head rotation and it's got to be at 2 Hz. I generally just consult a YouTube video about how to do that. And a loss of more than 2 lines on your visual acuity suggests that there's some hyperfunction there. And in someone who you're worried about on those screening tests, they're the people you'd be referring for formal vestibular function testing. And there's both audiologists and vestibular physiotherapists around who can help you with that.

If you are needing to use aminoglycosides for a longer period of time, so we mentioned the toxicities are often in those who need longer courses, so more than 5 days, then they're the people in whom formal testing of vestibular function and high-frequency audiometry should be considered because the high frequencies are the ones that go early.

And if you're using what we would call prolonged therapy, so more than 2 weeks, then baseline audiometry and then periodic audiometry, both during and after the duration of aminoglycosides, is recommended. And occasionally, potentially people who need repeated courses like people living with cystic fibrosis, for example, can do sort of periodic monitoring if they're needing repeated courses of aminoglycosides would be the other population to consider.

Wonderful. Well, that's unfortunately all the time we've got for this episode. A really big thank you for joining us today, Carly.

Thanks for having me.

[Music]

Dr Hughes' article, 'Aminoglycosides: an update on indications, dosing and monitoring', is available on the Australian Prescriber website. The views of the host and guests on the podcast are their own and may not represent Australian Prescriber or Therapeutic Guidelines. The authors acknowledged that they were members of the expert groups for Therapeutic Guidelines: Antibiotic Version 17. I'm Dhineli Perera, and thanks for joining us on the Australian Prescriber Podcast.