Medications linked to tinnitus

Gentamicin · Tobramycin · Streptomycin

The most seriously ototoxic medications in common use. ASHA's guidelines estimate roughly four million patients a year are potentially at risk of hearing loss from this class alone. Typically reserved for serious infections and given under hospital supervision with blood-level monitoring specifically because of this risk.

Evidence. Professional-body guidelines (ASHA) plus a review in Hearing Research. The monitoring protocols exist because the risk is not in question — and prospective hearing assessment is the only reliable way to detect the damage before someone notices it.

Cisplatin · Carboplatin

Among the highest rates on this page. Of 145 testicular cancer survivors treated with cisplatin, 74% reported some ototoxicity and 68% reported tinnitus specifically; a separate study of 243 survivors found tinnitus in 60.5%. For a substantial share of patients this is closer to an expected outcome than a rare complication.

Not uniform across the class. Genuinely dose-dependent: cumulative dose is a consistent risk factor, and risk is also associated with age and prior noise exposure. Cisplatin is retained in the cochlea indefinitely, which is why the effect does not clear the way most drug exposures do.

Timing. Affects the 4,000–8,000 Hz range first, sometimes progressing to conversational frequencies at higher cumulative doses.

This is not a reason to decline a cancer treatment. It is a reason to ask your oncologist about audiological monitoring as part of the treatment plan.

Evidence. Cohort studies of treated survivors with consistent findings across separate samples, plus official prescribing information describing the ototoxicity as cumulative and potentially severe.

Ibuprofen · Naproxen · COX-2 inhibitors · Acetaminophen

A separate question from acute overdose: what regular use over years does. Following 69,455 women for 22 years, frequent use of NSAIDs or acetaminophen was associated with close to a 20% higher risk of persistent tinnitus, and frequent COX-2 inhibitor use with 21% higher risk. Risk rose with frequency of use.

Not uniform across the class. Low-dose aspirin — the cardiovascular-protection dose — showed no elevated risk with frequent use. Moderate-dose aspirin was associated with higher risk specifically in women under 60, not in older women.

Evidence. A large longitudinal cohort (Nurses' Health Study II, over 1.1 million person-years). Observational — it establishes an association across a very large sample, not causation in an individual.

Bisoprolol · Nebivolol · Timolol (combination formulations)

Adverse-event reporting data shows a disproportionate tinnitus signal for specific beta-blockers: bisoprolol (reporting odds ratio 4.28), nebivolol (8.06), and timolol in combination formulations (23.29).

Not uniform across the class. This is the clearest case on the page. Carvedilol and labetalol have not shown the same disproportionate signal. “Beta-blocker” is not a single uniform risk, and an alternative within the same class may behave differently.

Timing. Case patterns describe onset within weeks of starting. One analysis found full recovery in about 65% of drug-related tinnitus reports overall, so a meaningful minority did not fully resolve.

Evidence. Pharmacovigilance disproportionality analysis of VigiBase, the WHO's global adverse-reaction database. A high reporting odds ratio flags a signal worth investigating — it is not a confirmed causal rate in the prescribed population.

Simvastatin · Atorvastatin · Fluvastatin

A 2025 analysis of 90,271 patients with hyperlipidaemia in the NIH All of Us database found statin use associated with higher odds of tinnitus (odds ratio 1.36) and sensorineural hearing loss (1.60) versus non-use.

Not uniform across the class. The statins did not behave alike. Simvastatin showed the strongest association with tinnitus (1.50), atorvastatin a more moderate one (1.21), and fluvastatin no statistically significant association at all (1.02).

Evidence. A large retrospective cohort study. The authors state it cannot establish causation: people prescribed statins already carry cardiovascular risk factors independently associated with tinnitus. Older, smaller studies found the opposite direction, which the article says plainly rather than omitting.

Sertraline · Fluoxetine · Paroxetine · Citalopram · Venlafaxine · Bupropion

Genuinely two-directional. Tinnitus is listed as a possible side effect in the prescribing information for nearly every SSRI — for citalopram, a documented incidence of 0.1% to 1%. The same drugs are also prescribed to reduce tinnitus-related distress, and a retrospective analysis of 37 patients found significant improvement in Tinnitus Severity Index scores.

Not uniform across the class. The evidence is not the same across the group. For SSRIs there is a pharmacovigilance signal — case reports plus disproportionality in two databases. For tricyclics the article establishes only that tinnitus appears in the package inserts for the whole class, which is a listing rather than a signal, and a weaker thing.

Timing. In most documented cases, within the first month of starting. One case report describes onset roughly five weeks after starting sertraline at a low 50 mg dose — so it is not necessarily dose-dependent.

Never stop an antidepressant abruptly. Discontinuation needs to be tapered under medical guidance, and tinnitus has also been documented after stopping, not only after starting.

Evidence. Pharmacovigilance case reports (around 30 in the Netherlands' Lareb database, with a disproportionality signal confirmed there and in the WHO database), plus individual case reports and one small retrospective series. No trial evidence either way.

Aspirin at high serum concentrations

The oldest and best-understood drug–tinnitus relationship. At high enough serum concentrations, salicylate produces temporary hearing loss of around 25 dB alongside tinnitus and hyperacusis. Symptoms resolve once the drug clears the body.

Not uniform across the class. A single normal dose is not the concern. Measurable ototoxicity is documented starting around 19.6 mg/dL serum salicylate — well above typical directed use.

Timing. With acute high dosing; resolves within days of stopping or reducing.

Evidence. A worked-out mechanism, not just an association: salicylate inhibits chloride binding on prestin, the motor protein in the cochlea's outer hair cells, and raises NMDA receptor activity.

Furosemide

Associated with tinnitus especially at high doses or when given quickly by IV. The mechanism is unlike the drugs above: rather than killing hair cells, loop diuretics disrupt the stria vascularis — the tissue maintaining the cochlea's electrical environment — and that disruption reverses.

Evidence. A mechanistic review in Journal of Otology. Permanent loss is rare and mainly reported in severe renal failure or alongside another ototoxic drug.

Quinine

Not only at high doses. Controlled studies in healthy volunteers found quinine produced measurable high-frequency hearing loss at ordinary therapeutic plasma concentrations, alongside tinnitus, with audiograms returning to normal afterwards.

Evidence. A controlled pharmacological study in British Journal of Clinical Pharmacology, plus a 2021 state-of-the-art review of antimalarial ototoxicity in Frontiers in Neurology which found most reported symptoms reversible.

Diazepam · Alprazolam · Lorazepam

The association runs the opposite way to the rest of this page: benzodiazepines are generally not associated with tinnitus while being taken. Tinnitus is documented as a withdrawal symptom after stopping, particularly after long-term use.

Timing. Most withdrawal symptoms resolve within about four weeks. Protracted cases are the exception: in one case series, tinnitus persisted 6 to 8 months in two of three patients before resolving.

Never stop a benzodiazepine abruptly after extended use. Abrupt discontinuation carries risks including seizure that are more urgent than withdrawal-related tinnitus.

Evidence. A case series plus a controlled discontinuation study of 40 long-term patients, in which the placebo group reported more withdrawal symptoms including tinnitus, and rated them more severe, than the group on a tapering dose.