All articlesScience

96.8% Improved on Steroids, 96.6% on Placebo: What Meniere's Trials Actually Show

96.8% of people who got steroid injections for Meniere's said their vertigo improved. In the placebo group it was 96.6%. That is from a 2023 Cochrane review: two small trials, 60 people in total.

The same questions keep coming up in the Meniere's community: does salt matter, does betahistine work, are the injections worth it. So I read the guideline US ENTs work from (AAO-HNS, 2020) and the Cochrane reviews behind it. This is a summary of what those documents say.

Not medical advice. This is a summary of public documents, not a recommendation. Do not stop or change any treatment without talking to your ENT. Seek urgent care for sudden hearing loss in one ear, or for vertigo with a severe headache, double vision, slurred speech, weakness or numbness.

Steroid injections: almost everyone improved

Vertigo improvement: steroid injections vs placebo96.8 percent of people in the steroid groups and 96.6 percent in the placebo groups reported their vertigo improved.Said their vertigo improvedA difference of 0.2 percentage points0%25%50%75%100%Steroid96.8%Placebo96.6%
People reporting improved vertigo 6 to 12 months after intratympanic corticosteroid injections or placebo. Risk ratio 1.00 (95% CI 0.92 to 1.10). Cochrane review, Webster et al. 2023 (CD015245): 2 trials, 60 participants, low-certainty evidence.

Nearly everyone in both groups said their vertigo was better. With so few people studied and such a high improvement rate on placebo, the trials cannot tell a treatment that works from one that does not. The same review found that steroid injections may slightly reduce how often attacks happen (3 trials, 472 people), but that evidence is low-certainty too.

Salt: barely tested

The guideline lists one small randomized trial of salt restriction, and it found no effect. Cochrane looked for placebo-controlled trials of salt or caffeine restriction and found none. So it is not proven and not disproven. It has barely been tested.

The guideline still recommends talking to patients about diet changes, because the risk is low. If you want to see where your own intake sits, the sodium calculator and the low-sodium diet guide are a good start.

Betahistine: placebo did as well

BEMED trial: attacks fell in every groupThree lines for placebo, low-dose and high-dose betahistine all fall over nine months and end close together.Attacks went down in every groupIncluding the group that took a placebomorefewermonth 0month 9PlaceboBetahistine, low doseBetahistine, high dose
Schematic of the shape of the result, not the published curves. In the BEMED trial (221 people, 9 months of treatment), attack incidence did not differ between placebo, low-dose and high-dose betahistine (P = 0.759). Adrion et al., BMJ 2016;352:h6816.

In the best trial so far (BEMED), attacks went down over 9 months in the betahistine groups and in the placebo group, with no significant difference between them.

The line that explains a lot

Up to 6 in 10 people on placeboTen circles, six of them filled, showing up to 60 percent vertigo control in placebo groups.Up to 6 in 10 people on placebohad their vertigo controlled in trials
The AAO-HNS guideline notes that vertigo control of up to 60% has been seen in placebo groups. Meniere's attacks come and go on their own, which makes it hard for any treatment to show a clear win over placebo. Basura et al., Clinical Practice Guideline: Ménière's Disease, 2020.

The guideline says up to 60% vertigo control has been seen in placebo groups. Meniere's comes in waves and often quiets down on its own. People also tend to start something new when their symptoms are at their worst, and symptoms often settle afterwards anyway. That makes it very easy to credit, or blame, the wrong thing.

What this does not mean

None of this means stop what you are doing. Something with weak trial evidence can still be what works for you, and that is a conversation for your ENT. "Not well tested" and "does not work" are different statements, and most of what is listed here is the first one.

Testing one trigger at a time

If trials cannot settle it, a careful personal experiment is the next best thing: change one thing, keep everything else steady, log it, and give it long enough to see past a lucky month. The caffeine and alcohol guide walks through a four-week version.

For those who tested one trigger at a time: did the result match what you believed before?

Sources: Webster KE et al., Intratympanic corticosteroids for Ménière's disease, Cochrane Database of Systematic Reviews (2023, CD015245); Webster KE et al., Lifestyle and dietary interventions for Ménière's disease, Cochrane Database of Systematic Reviews (2023, CD015244); Adrion C et al., Efficacy and safety of betahistine treatment in patients with Meniere's disease (BEMED), BMJ (2016); Basura GJ et al., Clinical Practice Guideline: Ménière's Disease, Otolaryngology–Head and Neck Surgery (2020). Always confirm specifics with a clinician familiar with your case.