Aminoglycoside ototoxicity has never behaved like a dose-dependent toxicity. Nephrotoxicity was largely tamed by once-daily dosing and therapeutic drug monitoring; hearing loss was not, and it still occurs at concentrations that monitoring would report as therapeutic. The reason for a substantial share of it is genetic. Three mitochondrial DNA variants in MT-RNR1 — m.1555A>G, m.1494C>T and m.1095T>C — carry a strong association with aminoglycoside-induced hearing loss, and they are not rare: about 1 in 330 people across populations.
The guideline from the UK Centre of Excellence in Regulatory Science and Innovation in Pharmacogenomics recommends avoiding aminoglycosides at any detectable variant level. Because mitochondrial DNA is heteroplasmic, 'any detectable level' is a deliberate threshold rather than a cautious one — the authors are not setting a percentage below which exposure is safe.
The operationally important sentence is the one about urgency. Where testing is unavailable and the clinical situation is pressing, the guideline says aminoglycoside treatment should not be delayed. That matters because the settings where the drug is most needed — neonatal sepsis, undifferentiated severe infection — are exactly the settings where a laboratory turnaround cannot be waited for. Point-of-care testing exists in some English centres and early health economic modelling suggests it may be cost-saving in neonates by preventing lifelong deafness, but that is an argument for building the service, not for withholding the antibiotic while you wish you had one.
- Ask about hearing loss on the mother's side — mitochondrial inheritance is maternal, and a family history is the cheapest available screen
- Record any prior aminoglycoside exposure and its outcome; a previous course without hearing loss does not exclude a variant
- Where use is planned rather than emergent — cystic fibrosis, mycobacterial regimens, elective surgical prophylaxis — there is time to test before the first dose
- Do not read a therapeutic trough as reassurance about hearing; the toxicity is not concentration-dependent in carriers
- In sepsis with no result available, give the drug
Why it matters
Therapeutic drug monitoring has been doing a job it cannot do: it protects the kidney and tells you nothing useful about the ear.
Don't overread it
This is a guideline built on an established genotype–phenotype association, not a trial showing that a testing programme reduces deafness at population level.
The statistics, in plain English
A carrier frequency of about 1 in 330 sounds small until it is set against how widely aminoglycosides are given. The figure that shapes services is the other one: only around 20% of aminoglycoside use is predictable in advance, so a pre-emptive testing programme can reach roughly a fifth of exposures at best. The remaining four-fifths are emergency prescriptions, which is why point-of-care testing rather than laboratory testing is the part of this that would change outcomes.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for clinical pharmacology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free