Dry mouth after radiotherapy

Radiotherapy to the head and neck predictably damages the salivary glands, and the 2017 Cochrane review of drugs to prevent it uses two terms for the result: xerostomia, the sensation of a dry mouth, and salivary gland hypofunction, a measurable fall in the saliva the glands produce. Acupuncture trials ask two separate questions about it: whether needling during radiotherapy prevents the dryness, and whether needling months or years afterwards eases it. Two multicentre trials with a sham arm, Garcia 2019 and Cohen 2024, one for each question and with 339 and 258 people, compared true acupuncture with sham acupuncture, a procedure made to resemble it, and with standard care. Both found true acupuncture ahead of standard care; against sham, the difference was smaller and depended on the site or the statistical adjustment. Cochrane’s review covering acupuncture dates from 2013 and predates both.

What the trials measure

Most of the trials below score symptoms with the Xerostomia Questionnaire (XQ), on which a higher score means a drier mouth; Garcia 2019 counted a score above 30 as clinically significant xerostomia. They also measure saliva directly, as unstimulated flow, collected at rest, and stimulated flow, collected while the person chews or tastes something. The two kinds of outcome can disagree. The authors of the 2013 Cochrane review noted that dry mouth can be troublesome even when saliva production rises, and called it a reporting bias that only two of the acupuncture trials they found reported symptoms at all.

The Cochrane reviews

The Cochrane review of non-pharmacological interventions for dry mouth by Susan Furness and colleagues, published in September 2013 with searches to April 2013, is the most recent Cochrane review that includes acupuncture. Five small trials, 153 people with dry mouth after radiotherapy, compared acupuncture with placebo; four were at high risk of bias. Two of them, 70 people, reported dry mouth in a form that could be pooled, and found no difference between acupuncture and placebo (standardised mean difference −0.34, 95% CI −0.81 to 0.14; low-quality evidence). Unstimulated saliva rose by 0.02 ml a minute at the end of treatment (3 trials, 71 people) and stimulated saliva by 0.19 ml a minute (3 trials, 71 people), both low-quality evidence, and both differences were still present at 12 months in two trials of 54 people. Acupuncture brought more adverse effects than placebo, tiny bruises and tiredness, which were mild and short-lived. The authors suggested the null result on symptoms could reflect too few participants, or some benefit from the placebo procedure itself.

The 2017 Cochrane review by Philip Riley and colleagues on preventing dry mouth during radiotherapy covered pharmacological interventions only, 39 trials and 3,520 people, so it has no acupuncture comparison. A PubMed search of the Cochrane Database of Systematic Reviews on 11 October 2026 found no later Cochrane review of interventions for dry mouth.

Later reviews

None of the reviews below is a Cochrane review. Some pool trials against sham with trials against standard care, oral care education or pilocarpine, a drug that stimulates saliva; where a review kept them apart, the table says so.

A standardised mean difference (SMD) expresses the gap between groups in units of the spread of scores. Heterogeneity, given as I², measures how much the trials disagree beyond what chance explains. Tsai and colleagues, in Taiwan, also reported that Egger’s test suggested publication bias (p = 0.078). Dörfler and colleagues, in Germany, concluded that no clinically meaningful effect of acupuncture on xerostomia had been proven so far, while the favourable findings on symptoms called for further study.

Trials

• Prevention, against standard care, 2012. A trial by Zhiqiang Meng and colleagues at Fudan University Shanghai Cancer Center randomised 86 people with nasopharyngeal carcinoma to acupuncture three times a week on radiotherapy days, or to standard care. XQ scores were lower with acupuncture from week 3 of radiotherapy through 6 months afterwards.
• Treatment, against oral care education, 2013. The ARIX trial, led by Richard Simcock in seven UK cancer centres, recruited 145 people with dry mouth more than 18 months after radiotherapy and gave them, in a crossover design, two group sessions of oral care education and eight of group acupuncture. Acupuncture was followed by fewer reports of severe dry mouth (odds ratio 2.01, P = 0.031), sticky saliva, and waking at night to drink. Stimulated and unstimulated saliva did not change.
• Treatment, nerve stimulation against pilocarpine, 2015. The RTOG 0537 trial, reported by Raimond Wong and colleagues, randomised 148 people to 24 sessions over 12 weeks of acupuncture-like transcutaneous electrical nerve stimulation (ALTENS), which applies current through skin pads at acupuncture points, or to pilocarpine. Only 96 could be evaluated at 9 months, and the change in symptom burden did not differ between the groups. Adverse events occurred in 20.8% with ALTENS and 61.6% with pilocarpine.
• Prevention, against sham and standard care, 2019. The trial by M. Kay Garcia, Lorenzo Cohen and colleagues ran at MD Anderson Cancer Center in Houston and Fudan University Shanghai Cancer Center, enrolling people with oropharyngeal or nasopharyngeal cancer between 2011 and 2015. Of 399 randomised, 339 were analysed. True or sham acupuncture was given three times a week through a 6 to 7 week course of radiotherapy. The sham combined a real needle at a real point not indicated for dry mouth, real needles at sham points and non-penetrating placebo needles at sham points. One year after radiotherapy, adjusted XQ scores were 26.6 with true acupuncture, 31.3 with sham and 34.8 with standard care: true acupuncture differed from standard care (P = .001) but not significantly from sham (P = .06). Clinically significant xerostomia affected 34.6%, 47.8% and 55.1%. The two sites disagreed about the sham. In Shanghai, true acupuncture scored 8.2 points below sham (P = .001) and sham did not differ from standard care; in Houston, sham scored 10.5 points below standard care (P = .002) and true acupuncture did not differ from sham. One adverse event was related to acupuncture, pain at one needle site in the ear.
• Treatment, against sham and standard oral hygiene, 2024. The US multicentre trial by Cohen, Suzanne Danhauer, Garcia and colleagues recruited through the Wake Forest National Cancer Institute Community Oncology Research Program Research Base: 258 people at 33 community sites in 13 states, with moderate to severe dry mouth 12 months or more after radiotherapy and a mean of 4.2 years since diagnosis. All kept to standard oral hygiene, and they were randomised in equal thirds to oral hygiene alone or to true or sham acupuncture twice a week for 4 weeks, with 4 more weeks for those who had a minor response. True acupuncture used three points on each ear, CV24 (Chengjiang) on the chin, LU07 (Lieque) on each forearm, a point on each hand and KD06 (Zhaohai) on each leg. The sham used non-penetrating needles at inactive body sites and real needles at three points on the rim of each ear. At week 4, the primary time point, XQ scores were 50.6 with true acupuncture, 55.0 with sham and 57.3 with oral hygiene alone. True acupuncture differed from oral hygiene by 6.67 points (95% CI −11.08 to −2.27); its 4.41-point difference from sham (95% CI −8.62 to −0.19, P = .04) did not hold after adjustment for multiple comparisons, and sham did not differ from oral hygiene. By week 26 only the sham group still scored lower than oral hygiene. No serious adverse events occurred; in the true acupuncture group one person had bruising, one a headache and one high blood pressure.

The 2012, 2019 and 2024 trials share investigators: Garcia, Cohen, the radiation oncologist David Rosenthal and the dental oncologist Mark Chambers of MD Anderson are authors of all three. Sham acupuncture sets out what penetrating and non-penetrating shams can and cannot rule out.

What guidelines say

The ISOO/MASCC/ASCO guideline on salivary gland hypofunction and xerostomia after cancer treatment, published in July 2021 by Vanessa Mercadante and colleagues for the International Society of Oral Oncology, the Multinational Association of Supportive Care in Cancer and the American Society of Clinical Oncology, searched trials published to June 2020. It says acupuncture may be offered during radiotherapy for head and neck cancer to reduce the risk of xerostomia (evidence quality intermediate, strength of recommendation moderate), resting on two trials of 109 people and the 339-person Garcia trial, and adds that there is no evidence it increases salivary flow. It says acupuncture may be offered after radiotherapy (evidence quality low, recommendation weak), noting that the evidence is one study, ARIX; ALTENS and other transcutaneous electrostimulation get the same low, weak rating. The 2024 trial postdates its search, and AcuiQ found no update to it in Europe PMC by October 2026.

A 2024 MASCC/ISOO clinical practice statement by Catherine Hong and colleagues, a short practical manual rather than a graded guideline, lists acupuncture among interventions tested in a few randomised trials and calls a recommendation for routine clinical use premature, because the evidence conflicts and treatment protocols vary. The 2022 Society for Integrative Oncology–ASCO guideline on integrative medicine covers pain and does not address dry mouth. Nausea from chemotherapy and Cancer-related fatigue cover other cancer treatment effects.

What the reviews could not settle

Whether true acupuncture differs from sham is unresolved. The pooled comparisons disagree: Ni 2020 found a difference on symptoms, Gu 2025 did not, and Cochrane’s 2013 pooling of 70 people found none. In Garcia 2019 the sham behaved differently in Houston and Shanghai, and in Cohen 2024 the difference at week 4 did not survive adjustment for multiple comparisons. Both shams placed some real needles, and the 2024 authors wrote that their sham cannot be considered a true placebo. Symptoms and saliva also diverge: several trials and reviews found change in one without the other. Most trials measured symptoms for weeks or months; in the 2024 trial the difference from oral hygiene had gone by week 26, and its authors noted that no maintenance treatment was given. The guideline that recommends acupuncture predates the 2024 trial, and Cochrane has not revisited the question since 2013. Dry eye disease covers a condition with the same split between symptoms and measured secretion.

AcuiQ’s xerostomia page lists the protocols individual studies prescribed, each with its citation.