No pill or supplement has been shown to protect hearing well enough to rely on: eleven noise-prevention trials could not be pooled, and N-acetylcysteine failed its clearest test, against cisplatin hearing loss.
The question comes in two versions. Someone is going to a concert, or works somewhere loud, and wants to know whether taking something beforehand will protect their ears. Someone else is about to start cisplatin chemotherapy, has been told it can damage hearing, and wants to know whether a supplement can prevent that. The answer to both is the same, and it is not what the supplement market implies.
Everything that has been tried for noise
A 2021 systematic review in Otology & Neurotology gathered the trials of pharmacological prevention of noise-induced hearing loss: eleven articles, 701 patients receiving something before or during noise exposure.
The agents tested were:
- Alpha-lipoic acid
- Ambient oxygen and carbogen (a carbon dioxide and oxygen mixture)
- Beta-carotene
- Ebselen
- Magnesium aspartate
- N-acetylcysteine
- Vitamins C, E and B12
A number of the studies did show statistically significant reduction in noise-induced hearing loss. Two found significantly better hearing with carbogen or ebselen than with placebo at 4 kHz — which is the right place to look, because that is where the noise notch characteristically appears on an audiogram.
And the review could not pool any of it. The agents and the methods varied so much that a meta-analysis was not possible, which is a serious limitation rather than a technical footnote: eleven small trials of nine different substances, each measured its own way, cannot be added together into an answer.
Its conclusion names what looked promising — magnesium aspartate, carbogen, vitamin B12, alpha-lipoic acid — and then says the clinical significance of these remains unclear.
That is the state of the field. Not "nothing works", which would overclaim in the other direction. Rather: several things have produced signals in small studies, nobody has run the trial that would settle it, and no agent has evidence supporting a recommendation.
The leading candidate, tested properly, failed
N-acetylcysteine is the supplement most often sold and discussed for this purpose. It has a plausible mechanism — it supports glutathione, part of the cochlea's defence against oxidative stress — and it is cheap and widely available.
It has also now been tested in the setting where the damage is most predictable. A 2025 systematic review and meta-analysis in Otolaryngology–Head and Neck Surgery pooled seven studies and 217 patients taking NAC to prevent cisplatin-induced hearing loss.
- Systemic NAC against placebo — risk ratio 0.80 (95% CI 0.54–1.19), P = .28
- Transtympanic NAC against placebo — risk ratio 0.89 (0.51–1.54), P = .67
Neither reached significance, and no differences appeared anywhere across 0.5 to 8 kHz — the entire standard audiogram range.
The authors' conclusion: regardless of administration route, the published evidence does not show NAC is effective in preventing cisplatin-induced hearing loss in the standard clinical audiogram range.
One qualification, reported because it is in the paper. A qualitative tendency toward protection appeared at ultra-high frequencies — 10 and 12 kHz. That is above what a standard audiogram tests, and it was a tendency rather than a significant pooled effect.
It is worth noticing rather than acting on. Ultra-high-frequency hearing is damaged early by ototoxic drugs, so a real effect there would be consistent with the mechanism. It is also the range whose loss you are least likely to notice, and a protective effect that exists only at 10 kHz and above is not the protection anyone is looking for. What it justifies is a better-designed trial, not a purchase.
Why the framing matters
Both reviews are about trying to make the cochlea more resistant to a dose of damage. That is a hard problem, which is why the work so far has produced signals rather than treatments.
The measures that do work attack the other side of the equation — they reduce the dose:
- Hearing protection in predictably loud environments. Which earplugs, and when.
- Distance from the source, since intensity falls sharply with it.
- Duration, because damage is a function of level and time together.
None of those are exciting and all of them are supported. Noise-induced tinnitus covers the exposure side in full.
If you are weighing a supplement
- If a supplement is being sold to you as hearing protection, this is the evidence behind that claim: eleven small trials that could not be pooled, and the leading candidate failing its clearest test.
- If you are facing cisplatin, ask your oncology team about ototoxicity monitoring rather than about supplements. The prevention question is live in research — the chemotherapy article covers what has and has not worked, including a 2026 trial that did prevent tinnitus and left a third of patients with a perforated eardrum.
- If you are going somewhere loud, the intervention with evidence costs a few pounds and goes in your ears.
- If you already have tinnitus, none of this is aimed at you. These are prevention trials in people with hearing to protect, not treatments for a sound that has already started — for that, the treatment comparison is the relevant page.