The 2020 Cochrane review of acupuncture and acupressure for pain in labour, by Caroline Smith and colleagues at Western Sydney University, included 28 randomised trials with 3,960 women between them: 13 of acupuncture and 15 of acupressure. Against a sham, it found that acupuncture may make little or no difference to how much pain women felt, but may increase their satisfaction with pain relief and probably reduces their use of drug pain relief. Most of its other findings rest on evidence it rated very low in certainty, and no trial in it was at low risk of bias in every domain. NICE’s intrapartum care guideline for England tells staff not to offer acupuncture or acupressure in labour and to support a woman who wants to use them. This page reports what each source found.
The Cochrane review
The review searched to February 2019 for trials in women in spontaneous or induced labour, whether first births or not, comparing acupuncture or acupressure with a sham, with usual care, with no treatment or with another non-drug method. Pain was measured on 0 to 10 or 0 to 100 scales. The review rates each finding with GRADE, the standard scale of how far evidence can be trusted, from high to very low.
| Comparison | Trials and women | What the review found | Certainty |
|---|---|---|---|
| Acupuncture against sham acupuncture | 2 trials, 325 women (pain); 1 trial, 150 women (satisfaction); 2 trials, 261 women (drug pain relief) | Little or no difference in pain intensity. More than twice as many women satisfied with their pain relief. A quarter fewer used drug pain relief (risk ratio 0.75). | Low (pain); moderate (satisfaction, drug pain relief) |
| Acupuncture against usual care | 4 trials, 495 women (pain); 6 trials, 1,059 women (drug pain relief) | Uncertain whether acupuncture reduces pain or the use of drug pain relief; the trials disagreed widely. | Very low |
| Acupuncture against no treatment | 1 trial, 163 women | Uncertain. | Very low |
| Acupressure against a sham | 6 trials, 472 women (pain); 4 trials, 313 women (caesarean) | Uncertain whether acupressure reduces pain. Fewer than half as many caesarean sections (risk ratio 0.44). | Very low (pain); moderate (caesarean) |
| Acupressure against usual care | 8 trials, 620 women | Uncertain whether acupressure reduces pain. | Very low |
| Acupressure against a combined control (sham or usual care) | 2 trials, 322 women | Probably slightly less pain. | Moderate |
Acupuncture may have little or no effect on rates of caesarean or assisted vaginal birth, with forceps or ventouse. No trial reported on women’s sense of control in labour, and only one on their satisfaction with the birth. The authors called for more trials with sham controls and usual-care comparisons that measure those outcomes. An earlier 2012 Cochrane overview of every method of pain relief in labour, by Jones and colleagues at the University of Liverpool, had placed acupuncture among the methods that “may work”, with evidence drawn mainly from single trials.
Trials
A 2009 Danish trial by Borup and colleagues randomised 607 healthy women in labour at term to acupuncture, to TENS (transcutaneous electrical nerve stimulation, a skin-surface device) or to conventional pain relief. Women in the acupuncture group used fewer drug and invasive methods of pain relief than either other group, but reported similar pain scores, and acupuncture did not change the length of labour. A 2014 Swedish trial by Vixner and colleagues at two hospitals randomised 303 women having their first baby to 40 minutes of manual acupuncture, electroacupuncture or standard care without acupuncture. Pain scores did not differ between the groups. Fewer women in the electroacupuncture group had an epidural: 46%, against 61% with manual acupuncture and 70% with standard care. The authors noted that every woman in the trial had free access to other pain relief.
The context differs in a 2010 trial by Hjelmstedt and colleagues at a public
hospital in India, where neither social support in labour nor epidurals were available. It
randomised 212 women having their first baby to 30 minutes of pressure on
SP06 (Sanyinjiao) on both legs during contractions, to
light touch at the same point, or to standard care. Pain fell most in the acupressure group,
immediately after treatment, and the authors described the effect as small.
Acupressure
A 2021 review by Chen and colleagues of 13 acupressure trials with 1,586 women found less pain when acupressure was added to standard care than with standard care alone or with sham acupressure, from immediately after treatment to at least 60 minutes later. It also found a shorter first and second stage of labour against untreated controls, no difference in the use of oxytocin or painkillers, and no conclusive effect on caesarean rates. No adverse events were reported. The authors rated the evidence moderate and called for higher-quality trials.
Acupressure also appears in antenatal education. In the 2016 Complementary Therapies for Labour and Birth trial in Sydney, 176 women expecting their first baby were randomised to a two-day antenatal course plus standard care, or standard care alone. The course taught six techniques together: acupressure, visualisation and relaxation, breathing, massage, yoga, and partner support. Epidural use was 23.9% in the course group and 68.7% with standard care. Because the six techniques were taught as a package, the trial cannot say how much of the difference came from acupressure. Acupressure describes the technique.
What NICE says
Recommendation 1.6.8 of NICE’s intrapartum care guideline (NG235), published in September 2023, reads: “Do not offer acupuncture, acupressure or hypnosis during labour. If a woman wants to use any of these techniques, support her choice.” It is tagged [2007, amended 2023]. NICE’s update information explains that tag: the recommendation last had an evidence review in 2007, and was changed in 2023 by committee consensus without a new one. The recommendation’s evidence base therefore predates both the 2011 Cochrane review and its 2020 update. The guideline treats TENS separately: the NHS does not provide TENS devices, but staff should support a woman who wants to use one. Clinical guidelines compares NICE’s positions on acupuncture with other guideline bodies’.
Points and pregnancy
A 2015 review by David Carr of the safety of obstetric acupuncture lists SP06
among the points most often cited as forbidden in pregnancy before 37 weeks, with
LI04 (Hegu), BL67
(Zhiyin) and GB21 (Jianjing) among the others. Pregnancy
and the forbidden points explains why points avoided while a pregnancy is meant to continue are
the ones chosen at term, and how AcuiQ separates the two readings.
What the reviews could not settle
The Cochrane review could not say whether acupuncture or acupressure reduces the intensity of pain in labour, as opposed to how women rate their pain relief or how much medication they use. Its sham-controlled acupuncture evidence comes from two trials, and the trials comparing acupuncture with usual care disagreed too much to give a clear answer. The trials were also run in settings as different as a Swedish labour ward with epidurals on request and an Indian public hospital without them, and a method that changes little where epidurals are available may change more where they are not. None of the reviews could compare points, timing or styles of stimulation with any confidence.
AcuiQ’s labour pain page lists the protocols individual studies prescribed, each with its citation; the childbirth pain, anxiety in labour and prolonged labour pages hold related studies.