Zinc has been tried as a remedy for a lost sense of smell since at least the 1970s, on the reasoning that the mineral is needed for the sense of smell to work and that damaged olfactory tissue might regenerate faster with more of it. A systematic review published on 28 October in the Brazilian Journal of Otorhinolaryngology has now gathered the trial evidence in one place, and the pooled result does not support the practice.
Six studies, 1,460 participants, and no statistically significant difference between zinc and control on validated smell tests: an odds ratio of 1.78 with a 95% confidence interval running from 0.63 to 5.03. The authors grade the evidence Level 2a and write that current evidence “does not show a consistent clinical benefit of zinc for olfactory dysfunction.” Their summary line is blunter still: not routinely recommended.
What the review actually pooled
The search covered PubMed, the Cochrane Library, Scopus and Web of Science up to April 2025, along with trial registries, and returned 731 records. Ten full texts were read; six met the criteria — three randomised trials and three observational cohorts published between 1976 and 2024. Together they contain 436 people given zinc and 1,024 who were not, followed for anywhere from one month to fifteen.
The population was not one condition. Included studies covered smell loss after infection, after head injury, during chemotherapy and of unknown cause, which matters because those are different problems with different odds of recovery.
The primary outcome was performance on validated instruments — the University of Pennsylvania Smell Identification Test, Sniffin’ Sticks, or a phenyl ethyl alcohol detection threshold. Heterogeneity across the pooled estimate was moderate at 63.2%. Patient-reported improvement, available in three of the six studies, was also null, at an odds ratio of 0.69 (0.25 to 1.93) with heterogeneity of 90.9%. Distorted smell — parosmia and phantosmia — was measured in a single study, giving an odds ratio of 0.45 (0.1 to 1.98) that the authors decline to treat as an answer.
Methodological housekeeping was conventional for a review of this size. It is registered on PROSPERO, retrospectively, under CRD420261321935, and risk of bias was assessed with the Cochrane RoB2 tool for the trials and ROBINS-I for the cohorts: two of the three trials carry “some concerns”, one is at low risk, and all three cohorts are at moderate risk.
The one positive signal, and the one pointing the other way
There is an exception in the table, and it is worth naming precisely. A 2015 trial of 145 people with smell loss after head trauma tested 30 mg a day of zinc gluconate and reported an odds ratio of 4.9 (1.7 to 13.9). It is the only positive result in the review, and the authors note it was “not consistent across other studies”. One trial in one subgroup is a hypothesis, not a finding to shop with.
The counterweight sits in the same table, and it is unusual to see it published. A 2024 retrospective cohort of 714 people with post-infectious loss, with the corresponding author of this review among its authors, found zinc associated with lower odds of self-reported improvement, at 0.26 (0.25 to 0.46). A retrospective cohort cannot settle the question either way. But a review that prints a result running against its own hypothesis is at least not shopping for an effect.
The dose column is its own warning
Across the six studies the dose runs from 17 mg to 300 mg a day, in three different salts — sulfate, gluconate, oxide — over one to fifteen months. The entries are not even expressed the same way: one 1976 trial specifies 100 mg of zinc ion, others give the amount of the compound. Zinc sulfate and zinc gluconate do not deliver the same quantity of elemental zinc, and the review itself says the heterogeneity in formulation and dose “likely contributes to the variability in outcomes”, leaving open whether the null result reflects no effect or simply noise.
For scale, MedlinePlus puts the recommended dietary allowance at 11 mg a day for men and boys aged 14 and over, and 8 mg for women 19 and over. Even allowing for the salt-versus-ion distinction, the upper end of that range sits well beyond what a multivitamin provides. MedlinePlus lists what large doses do: diarrhoea, abdominal cramps and vomiting, usually within three to ten hours of swallowing the tablets, resolving shortly after stopping; and over the long term, excess zinc can lead to copper, iron or magnesium deficiency.
Tolerability in the review is thin ground. Adverse events were “poorly reported overall”; only the 1976 trial described them, as mild and mostly gastrointestinal, and no study reported anything serious. That is a statement about reporting, not about safety at 300 mg for a year.
It is also worth remembering what a supplement label certifies. The FDA does not have authority to approve dietary supplements for safety and effectiveness, or their labelling, before they are sold, so the dose on the bottle has not been reviewed against an evidence standard the way a drug’s has.
Why a supplement can look like it worked
Most smell loss after a respiratory infection gets better by itself. MedlinePlus is direct about it: “If you have a loss of smell due to a recent upper respiratory infection, be patient. The sense of smell often returns to normal without treatment.” Some loss arrives with age, for which the same entry says there is no treatment; in most cases no clear cause is identified at all.
Set a supplement against that background and it will frequently coincide with recovery it did not cause. That is precisely why the pooled trial comparison, rather than the individual story, is the relevant measure — and why a null pooled result in a condition that self-resolves is a meaningful null rather than a failed study.
There is a further wrinkle. MedlinePlus’s entry on zinc notes that people who use nasal sprays and gels containing zinc may lose their sense of smell, and its entry on impaired smell lists “possibly zinc-based products” among the medicines that can change odour detection. Nasal zinc gels and swabs have been linked to lasting loss of smell — a safety point we covered when flu season opened. Oral zinc is a different route and this review is about oral supplements. The point is that “zinc for smell” describes at least two products with opposite risk profiles.
What the trials do support
Olfactory training — deliberately sniffing a small set of familiar odours, usually daily for several weeks — has the better record. A meta-analysis of ten randomised trials covering 628 patients, published in June in the Turkish Archives of Otorhinolaryngology, found a modest but statistically significant improvement on objective smell scores, with a standardised mean difference of 0.30 (0.08 to 0.51), and on quality of life at −0.40 (−0.65 to −0.15). Heterogeneity was low for both.
Modest is the operative word. A standardised mean difference of 0.30 is a small effect, the training protocols varied between trials, and MedlinePlus frames retraining as something that “can help improve the sense of smell in a minority of patients”. It involves no product and no prescription; it is also not a cure.
What the review cannot tell you
The limits are the review’s own, and they are worth stating rather than burying. The total sample is modest; results could not be pooled by cause of smell loss because the subgroup data were insufficient; the small number of trials ruled out tests for publication bias and any meta-regression on age, duration of loss or baseline score; quality of life was never measured systematically; and the studies span five decades, so the instruments and definitions are not the same.
Two consequences follow. “No demonstrated benefit” is not the same as “proven to do nothing” — a better-designed trial in a defined group, particularly people with measured zinc deficiency, could still change the picture. And nothing in this review speaks to prevention, or to how zinc performs alongside training rather than instead of it; its only look at distorted smell rests on one study’s 66 patients.
If smell loss continues or gets worse, or arrives with other unexplained symptoms, MedlinePlus advises contacting a health care provider. Sudden loss can accompany head injury or a neurological condition, and those are not self-treatment problems. This article reports what the cited sources say and is not medical advice — see our medical disclaimer — and questions about a supplement belong with a clinician, particularly where dose or a competing medication is involved, as with the label arithmetic in our census of high-dose B6 products.