A sham-controlled trial tested ear acupressure for rosacea — and reported a real signal

A randomised trial has tested one of the older ideas in traditional medicine — pressing a seed against a precise point on the outer ear — against a sham version for the chronic facial skin condition rosacea, and reported that the real points outperformed the fake ones [s1]. Published in JAMA Dermatology, the study is small, single-centre and comes with important caveats, but its sham-controlled design makes it more informative than the testimonials that usually accompany such remedies.

Rosacea causes persistent facial redness, flushing, visible vessels and sometimes acne-like bumps. It is managed, not cured, with topical agents, oral antibiotics, laser treatment and trigger avoidance, and response is often incomplete [s2]. That gap between available treatment and satisfied patients is why alternative approaches keep drawing interest — and why a rigorous test of one of them is worth reading carefully.

What the trial did

Researchers at the Second Affiliated Hospital of Xi'an Jiaotong University enrolled 204 patients with rosacea, all of whom were receiving standard first-line treatment, between October 2025 and May 2026 [s1]. Participants were randomly assigned 1:1, 102 to each group, in a single-centre, sham-controlled, assessor-masked design — meaning the clinicians rating the skin did not know who was in which group [s1]. The active group was a mean age of 32.51 years and 84.3% women; the sham group averaged 30.82 years and was also 84.3% women [s1].

The intervention was deliberately low-tech. The active group pressed a seed of the plant Vaccaria segetalis against the cavum conchae, a hollow in the outer ear; the sham group pressed an identical seed against the superior margin of the auricular tubercle, a point not expected to be therapeutic [s1]. Both groups pressed three times a day for 30 seconds over eight weeks, with extra pressing permitted during flare-ups [s1]. The primary outcome was the proportion reaching an Investigator's Global Assessment (IGA) score of 0 or 1 — clear or almost clear skin — at week 8 [s1].

The result

On the primary outcome, 58 of 102 patients (56.9%) in the active group reached clear or almost clear skin at week 8, versus 36 of 102 (35.3%) in the sham group [s1]. That works out to a relative risk of 1.61 (95% CI 1.18–2.20; P=0.003) and a number needed to treat of 5 — meaning about five patients would need the active acupressure for one additional person to reach the target over sham [s1].

Secondary findings pointed the same way. An immediate vascular improvement was seen in 59.3% of the active group versus 7.4% of the sham group (P<0.001), and the authors reported that blood markers of inflammation and rosacea biology — LL-37, CGRP, TNF-alpha and IL-6 — fell more in the active group, though the abstract did not give the numbers [s1]. At week 20, the active group maintained significantly lower IGA and clinical erythema scores, again without numeric detail in the abstract [s1]. No serious adverse events were reported [s1].

The biomarker choices are not arbitrary. LL-37, an antimicrobial peptide, and CGRP, a nerve-signalling molecule, are both implicated in the flushing and inflammation of rosacea, and TNF-alpha and IL-6 are general markers of inflammatory activity. A plausible biological story — that stimulating a point on the ear dampens this cascade — makes the result easier to believe, but a plausible mechanism is not evidence of one, and without the actual values the biomarker claims have to be taken on trust. The twelve-week persistence of benefit to week 20 is the more clinically interesting secondary signal, since a treatment whose effect fades the moment patients stop would be of limited practical use.

How much to trust it

Several features strengthen the result: random assignment, a genuine sham comparator, masked assessors and a clinically meaningful primary endpoint all guard against the usual weaknesses of acupressure studies. The effect size is not trivial, and a number needed to treat of five would be respectable for a mainstream therapy.

The limits are equally real. This was a single centre in one country, with a mostly young, mostly female sample, over a short horizon; single-centre trials of novel treatments frequently report larger effects than later multi-centre replications. Everyone was already on standard treatment, so the trial tests acupressure as an add-on, not a replacement. A sham-controlled design reduces but does not eliminate placebo effects, particularly for a self-administered ritual performed several times a day, and patients necessarily knew they were pressing something. The biomarker and week-20 claims cannot be evaluated without the actual figures. And publication of a positive novel-therapy trial says nothing about the file-drawer of negative ones.

What it means

The honest summary is that a well-designed small trial found a real, statistically significant benefit for auricular acupressure added to standard rosacea care — enough to justify a larger, multi-centre replication, not enough to recommend it as established treatment. Patients curious about trying it have little to lose from a cheap, low-risk practice alongside their prescribed therapy, but should not drop proven treatments for it. Anyone citing this study as proof that acupressure cures rosacea is overreading a single, preliminary result.