Many hospitals rely heavily on aminoglycosides within their antimicrobial regimens. Locally we use Gentamicin as our go-to aminoglycoside of choice for sepsis (combined with Amoxicillin and Metronidazole) as well as for severe pyelonephritis amongst other things. In fact, it was one of my favourite choices for buying time to work out what is wrong with septic patients who cannot afford to wait for a diagnosis before treatment is started. ECIC (a.k.a. my wife) used to tease me that Gentamicin was the only Microbiology oncall advice needed and therefore our cat could do the oncall… how rude!
Genetically mutated ears?
Gentamicin is rapidly bactericidal (it kills bacteria) and has good activity against the common causes of sepsis e.g. Staphylococcus aureus, Escherichia coli, Klebsiella spp. therefore a stat dose of Gentamicin in a septic patient often stabilises the patient while you work out what is wrong with them! Gentamicin is also the recommended treatment of sepsis in neonates (up to 28 days after birth) in combination with Benzylpenicillin.
Other aminoglycosides include Amikacin, Tobramycin and Neomycin.
BUT there are some problems with aminoglycosides which can make Doctors reluctant to prescribe them. The first is nephrotoxicity (damage to kidneys), but this is usually reversible on stopping them, the second is ototoxicity (damage to hearing and balance) which, whilst much less common than nephrotoxicity, is irreversible.
The ototoxicity issue has recently become very topical as the Medicines and Healthcare products Regulatory Agency (MHRA) in the UK has issued guidance about screening for genetic mutations in patients who might be at higher risk for ototoxicity from the use of aminoglycosides.
How do aminoglycosides cause ototoxicity?
Aminoglycosides kill bacteria by irreversibly binding to part of the bacterial ribosome (where proteins are produced) causing protein formation to stop before the protein is fully formed. There is a specific site on the ribosome to which the aminoglycoside binds.
Weirdly, aminoglycosides concentrate within the inner ear where they are not well metabolised. These higher levels of aminoglycosides can eventually damage the hair cells of the inner ear causing permanent loss of hearing and balance. They damage the cells by targeting the mitochondria in the cells of the inner ear.
Mitochondria are the part of the cell which generates the cell’s energy. A mitochondrion has its own DNA, separate from the rest of the cell’s DNA in the nucleus, and which is actually quite similar to the DNA of bacteria. The mitochondrion also has its own ribosomes that produce proteins. Because the mitochondrion is so similar to a bacterium, when aminoglycoside levels get too high, they start to target these mitochondria because of their similarity. It’s a bit like a case of mistaken identity.
In medicine we try to reduce the risk of ototoxicity by monitoring the levels of aminoglycosides in blood and only using them for the minimum duration necessary to treat the infection. We try to stop the amounts getting to toxic levels.
What’s all this about genetic mutations predisposing to ototoxicity?
The MHRA are concerned about a mutation known by the catchy name of m.1555A>G …yep, that really is its name. The mutation is in mitochondrial DNA at the aminoglycoside binding site. We inherit our mitochondrial DNA from our mothers and so if there are any mitochondrial genetic mutations they are also inherited from our mothers. As a result, susceptibility to ototoxicity from aminoglycosides is genetic and often can run in families.
The mutation m.1555A>G in the mitochondrial DNA makes the mitochondrial ribosome look even more like a bacterial ribosome; it looks bacterial not human. As a result, aminoglycosides “see” the mitochondrial ribosome as a bacterial ribosome and so attack it, killing the mitochondrion in the process and thereby killing the cell. This is particularly a problem in the hair cell of the ear where the levels of aminoglycosides concentrate, and hence ototoxicity occurs.
How common is this problem?
To try and work out an approximate value for how common hearing loss due to aminoglycoside toxicity with mitochondrial mutations is there are 3 values we need to know:
There have been a couple of studies that have screened neonates for the genetic mutations that predispose to hearing loss. They have found the mutations to be uncommon, about 0.2% or 1 in 500. I’ll use 0.2% and call this number for “how common”, A.
It’s impossible to tell how many people are given aminoglycosides every year but I’m going to use the incidence of sepsis as a surrogate marker and say that this “is the number of people per year given aminoglycosides” (yep, it’s a crude estimate but it probably overestimates the number as not every person with sepsis will be given an aminoglycoside). The incidence of sepsis is said to be up to 0.2% of the population per year according to UK figures. I’ll use 0.2% and call this number for “use of aminoglycosides”, B.
A few studies have looked at the incidence of hearing loss in neonates who have the predisposing mitochondrial mutation and who have been given Gentamicin, and how many developed hearing losses. The highest rate of hearing loss I could find is 30%, although other studies suggest the value may be lower. I’ll use 30% and call this number for “mutation present and develops hearing loss”, C.
The neonate figures are just a useful subject group which can be used to extrapolate to the general population.
So, the incidence of hearing loss due to the use of aminoglycosides in people with a predisposing mitochondrial mutation is approximately = A x B x C = 0.012% of the population. To put that into whole numbers, 7,800 people from the UKs 65 million population will get hearing loss from aminoglycosides per year.
That may “sound” (no pun intended) like a lot of people but remember I have used the highest risk numbers I could find. In comparison it is said that 10% of 13-23 year-olds in Europe have been diagnosed with permanent hearing loss from listening to music through ear buds… now that is a scary statistic!
So what? I hear you ask (still no pun intended)
Okay, that’s a good question. Up to 7,800 people are going to get hearing loss in the UK from the use of aminoglycosides, can we do anything about it?
Well, if we knew who was at risk from hearing loss due to a specific antibiotic, we might be able to use a different antibiotic to treat them. Perhaps those with the specific mutation could carry a medical card or wear a medical alert bracelet saying, “don’t give me aminoglycosides”? It’s possible, but I have some worries about it.
So, what do I think we should do?
I don’t think it’s feasible to screen everyone for a mitochondrial mutation before giving an aminoglycoside antibiotic. When I want to give an aminoglycoside, it is urgent. I can’t wait for the results of the genetic test to come back before giving it. But I guess I could test people I either plan to give aminoglycosides to non-urgently or those I suspect I’ll need to give repeated courses to.
I would test:
Other than that, I wouldn’t test anyone else. Once I knew someone’s risk I would either try and find an alternative treatment next time they needed an antibiotic, or if the aminoglycoside was unavoidable (like in the treatment of M. abscessus in CF) then I would monitor the patients hearing closely during treatment to pick up damage to hearing early, and potentially stop/change the treatment.
So, in conclusion there does “app’ear” (oops, pun intended that time!) to be an increased risk of damage to hearing and balance in a small subset of the population with a specific genetic mitochondrial mutation when they are given an aminoglycoside BUT the guidance is clear - genetic testing should not delay urgently needed aminoglycoside treatment. If the patient is septic then start treating them as soon as possible, the risk of hearing loss is low but the risk of death from sepsis is high.
Other aminoglycosides include Amikacin, Tobramycin and Neomycin.
BUT there are some problems with aminoglycosides which can make Doctors reluctant to prescribe them. The first is nephrotoxicity (damage to kidneys), but this is usually reversible on stopping them, the second is ototoxicity (damage to hearing and balance) which, whilst much less common than nephrotoxicity, is irreversible.
The ototoxicity issue has recently become very topical as the Medicines and Healthcare products Regulatory Agency (MHRA) in the UK has issued guidance about screening for genetic mutations in patients who might be at higher risk for ototoxicity from the use of aminoglycosides.
How do aminoglycosides cause ototoxicity?
Aminoglycosides kill bacteria by irreversibly binding to part of the bacterial ribosome (where proteins are produced) causing protein formation to stop before the protein is fully formed. There is a specific site on the ribosome to which the aminoglycoside binds.
Weirdly, aminoglycosides concentrate within the inner ear where they are not well metabolised. These higher levels of aminoglycosides can eventually damage the hair cells of the inner ear causing permanent loss of hearing and balance. They damage the cells by targeting the mitochondria in the cells of the inner ear.
Mitochondria are the part of the cell which generates the cell’s energy. A mitochondrion has its own DNA, separate from the rest of the cell’s DNA in the nucleus, and which is actually quite similar to the DNA of bacteria. The mitochondrion also has its own ribosomes that produce proteins. Because the mitochondrion is so similar to a bacterium, when aminoglycoside levels get too high, they start to target these mitochondria because of their similarity. It’s a bit like a case of mistaken identity.
In medicine we try to reduce the risk of ototoxicity by monitoring the levels of aminoglycosides in blood and only using them for the minimum duration necessary to treat the infection. We try to stop the amounts getting to toxic levels.
What’s all this about genetic mutations predisposing to ototoxicity?
The MHRA are concerned about a mutation known by the catchy name of m.1555A>G …yep, that really is its name. The mutation is in mitochondrial DNA at the aminoglycoside binding site. We inherit our mitochondrial DNA from our mothers and so if there are any mitochondrial genetic mutations they are also inherited from our mothers. As a result, susceptibility to ototoxicity from aminoglycosides is genetic and often can run in families.
The mutation m.1555A>G in the mitochondrial DNA makes the mitochondrial ribosome look even more like a bacterial ribosome; it looks bacterial not human. As a result, aminoglycosides “see” the mitochondrial ribosome as a bacterial ribosome and so attack it, killing the mitochondrion in the process and thereby killing the cell. This is particularly a problem in the hair cell of the ear where the levels of aminoglycosides concentrate, and hence ototoxicity occurs.
How common is this problem?
To try and work out an approximate value for how common hearing loss due to aminoglycoside toxicity with mitochondrial mutations is there are 3 values we need to know:
- How common is the mutation within the population?
- How common is the use of aminoglycosides?
- How many of those with the mutation will develop hearing loss if given an aminoglycoside?
There have been a couple of studies that have screened neonates for the genetic mutations that predispose to hearing loss. They have found the mutations to be uncommon, about 0.2% or 1 in 500. I’ll use 0.2% and call this number for “how common”, A.
It’s impossible to tell how many people are given aminoglycosides every year but I’m going to use the incidence of sepsis as a surrogate marker and say that this “is the number of people per year given aminoglycosides” (yep, it’s a crude estimate but it probably overestimates the number as not every person with sepsis will be given an aminoglycoside). The incidence of sepsis is said to be up to 0.2% of the population per year according to UK figures. I’ll use 0.2% and call this number for “use of aminoglycosides”, B.
A few studies have looked at the incidence of hearing loss in neonates who have the predisposing mitochondrial mutation and who have been given Gentamicin, and how many developed hearing losses. The highest rate of hearing loss I could find is 30%, although other studies suggest the value may be lower. I’ll use 30% and call this number for “mutation present and develops hearing loss”, C.
The neonate figures are just a useful subject group which can be used to extrapolate to the general population.
So, the incidence of hearing loss due to the use of aminoglycosides in people with a predisposing mitochondrial mutation is approximately = A x B x C = 0.012% of the population. To put that into whole numbers, 7,800 people from the UKs 65 million population will get hearing loss from aminoglycosides per year.
That may “sound” (no pun intended) like a lot of people but remember I have used the highest risk numbers I could find. In comparison it is said that 10% of 13-23 year-olds in Europe have been diagnosed with permanent hearing loss from listening to music through ear buds… now that is a scary statistic!
So what? I hear you ask (still no pun intended)
Okay, that’s a good question. Up to 7,800 people are going to get hearing loss in the UK from the use of aminoglycosides, can we do anything about it?
Well, if we knew who was at risk from hearing loss due to a specific antibiotic, we might be able to use a different antibiotic to treat them. Perhaps those with the specific mutation could carry a medical card or wear a medical alert bracelet saying, “don’t give me aminoglycosides”? It’s possible, but I have some worries about it.
- Firstly, it might delay or prevent patients being treated properly for sepsis whilst their Doctors try and work out what to give them.
- Secondly it may result in higher risks from different, potentially life-threatening, complications such as Clostridium difficile associated disease due to the different antibiotics.
- Thirdly, genetic tests take days to do, they just cannot be done quickly enough to screen people BEFORE treatment; the patient would have to be screened before they became unwell, maybe as part of a screening program.
- And finally, the cost involved in screening would be horrendous! Even if each test was only £20 (and that may be an underestimate) then the cost to screen the entire population in one go would be £1.3 billion, and nearly £13 million to screen all babies born every year.
So, what do I think we should do?
I don’t think it’s feasible to screen everyone for a mitochondrial mutation before giving an aminoglycoside antibiotic. When I want to give an aminoglycoside, it is urgent. I can’t wait for the results of the genetic test to come back before giving it. But I guess I could test people I either plan to give aminoglycosides to non-urgently or those I suspect I’ll need to give repeated courses to.
I would test:
- Cystic fibrosis (CF) patients with Mycobacterium abscessus infection in whom I might have to use 1-3 months of IV aminoglycosides followed by months to years of nebulised aminoglycosides
- Patients with recurrent infections such as UTIs where either antibiotic resistance or allergies mean they will get repeated courses of aminoglycosides
- Patients with prosthetic valve infective endocarditis where I have to use a combination of aminoglycosides with other nephrotoxic drugs such as glycopeptide antibiotics and diuretics for 6 weeks duration
Other than that, I wouldn’t test anyone else. Once I knew someone’s risk I would either try and find an alternative treatment next time they needed an antibiotic, or if the aminoglycoside was unavoidable (like in the treatment of M. abscessus in CF) then I would monitor the patients hearing closely during treatment to pick up damage to hearing early, and potentially stop/change the treatment.
So, in conclusion there does “app’ear” (oops, pun intended that time!) to be an increased risk of damage to hearing and balance in a small subset of the population with a specific genetic mitochondrial mutation when they are given an aminoglycoside BUT the guidance is clear - genetic testing should not delay urgently needed aminoglycoside treatment. If the patient is septic then start treating them as soon as possible, the risk of hearing loss is low but the risk of death from sepsis is high.
RSS Feed
