Acupuncture in pregnancy and labour: the evidence for effectiveness

This paper presents a summary of the evidence for the effectiveness of acupuncture in the treatment of pregnancy- related conditions and other uses in obstetrics. The majority of the articles reviewed relate to pain relief in labour (22 studies) and induction/ duration of labour (13 studies). Others cover various conditions of pregnancy such as backache, pelvic pain, or breech presentation. There is a wide variability in the type of acupuncture and methodological design making it difficult to compare studies and develop overall conclusions. Nevertheless the available sources provide some evidence that acupuncture is an effective treatment for these conditions.

Download the complete article as a pdf pdf_button


Obstetrics is defined as the science of midwifery, that is to say, assistance at childbirth.

There are reports in the ancient texts of acupuncture being used to aid childbirth at least as far back as the Jin Dynasty (265-420) (Zheng, 1990).

In theory, acupuncture is ideally suited to obstetrics. There are restrictions on the use of drugs during pregnancy which may have harmful, teratogenic effects on the foetus. This has meant there was little to offer women for the minor ailments of pregnancy, which in some cases can be quite severe, even needing hospitalisation. Acupuncture has been used to treat a long list of conditions including morning sickness, migraine, backache and constipation. It has also been used to encourage version of the foetus in breech presentation, induction of labour, and pain relief in labour. After the birth it may be used to treat haemorrhoids, mastitis, depression and other problems associated with this period.

As "obstetrics" is not in itself a condition needing treatment, this paper relates to the disorders and complications of pregnancy. (For pregnancy sickness, see Briefing Paper 10, Obstetrics (1)).

Due to the wide variability and number of studies, the main features of the studies are presented in table form.

Studies that have been added since the original version of this briefing paper are as follows: Habek et al (2003) and (Neri et al 2004, 2007) on breech presentation; Elden et al (2008 x3) on pelvic pain; Martensson et al (2008), Fan Qu and Jue Zhou (2007), Ziaei and Hajipour (2006), Borup et al (2009), Hantoushzadeh  et al (2007) and Citkovitz et al (2009) on pain relief in labour; Harper et al (2006) and Gaudernack et al (2006) on augmentation of labour; Guerreiro da Silva (2007) on emotional complaints in pregnancy; and reviews by Betts (2006), Ee et al (2008), Coyle et al (2007), Van den Berg et al (2008), Lee and Ernst (2004), and Smith and Crowther (2007).


Many of the reviews from the original version, superceded by more recent ones, have been discarded:

Chez and Jones (1997), Beal (1999), Allaire (2001), Gentz (2001), Ewies and Olah (2002),  Young and Jewell (2002), Fugh-Berman and Kronenberg (2003).


Literature search

A search was made using the ARRC database, as well as further searches on AMED, BNI, EMBASE, NHS EVIDENCE, MEDLINE, CINAHL AND COCHRANE databases, using the key words "pregnancy (+obstetric etc), labour, induction", plus 'acupuncture'.

Additional articles were obtained through cross-referencing the literature cited in individual studies and reviews. After excluding those in a foreign language, and letters and commentaries, there were 54 original clinical studies remaining, plus 8 review papers. Various different types of methodology are represented in the studies: randomised controlled trials, case series, case reports and observational studies.


Malposition and breech presentation

Refer to table 1.

Of the ten studies reported, eight used a control group, either no intervention8,9,26,31,35,42 or the standard knee-chest position47 or sham acupuncture41. One case series7 showed a 61% success rate in foetuses turning to cephalic presentation. The remaining study43 compared the cardiovascular effects and fetal behaviour during three alternative treatment approaches: moxibustion, acupuncture or acupuncture plus moxibustion applied to point Bl67  for breech presentation. Version to cephalic presentation occurred in 56% of cases; of these, 80% for moxibustion, 28% for acupuncture and 57% for acupuncture plus moxibustion.

The controlled studies comprised four randomised trials (RCTs)9,26,35,42 , one cross-over41 and three with non-randomised matched groups8,31,47 . The interventions were variously moxibustion, acupuncture, combined acupuncture and moxibustion, electro-acupuncture (EA) or auricular seed pressing. All showed significantly positive effects except for the EA group in one study31. (In this large retrospective case control the moxibustion and EA groups produced very similar results, 92% and 89% version, but their respective controls differed by 10%: 73% and 83%. Hence the moxa intervention was deemed effective but not the EA. The apparent distinction may be no more than a chance sampling effect). The small cross-over study41 was notable in that the outcomes were changes in foetal heart rate and foetal movements rather than simply turning/not turning.


One further study has been published by Cardini and co-workers10. As this trial was interrupted with fewer than half the patients recruited it has not been summarised here.

Back and pelvic pain

Refer to table 2.

Of seven studies in this category, four17,20,24,65 were controlled trials, two22,58 were single case reports and one 57 a retrospective case series. In three of the controlled studies acupuncture was significantly more effective than medication24, physiotherapy65 or exercises17.  All showed a positive outcome for acupuncture though there were wide variations in the nature and frequency of the treatment.

The only study 17,18 which followed up patients in the post-natal period found approximately three-quarters of women were free of pain 3 weeks after delivery and 99% 12 weeks after. There were no differences in recovery between the 3 treatment groups. The same author19 later reported on adverse effects of standard treatment, acupuncture and stabilising exercises in his previous study and found that there were no serious events after any of the treatments and only minor adverse complaints from the acupuncture group (who received  strong needle stimulation) with women rating acupuncture favourably despite this. There were no observable severe adverse influences on the pregnancy, mother, delivery or the fetus/neonate.

This same author20 went on to publish a paper comparing acupuncture with non-penetrating sham acupuncture. This showed acupuncture had no significant effect on pain or on degree of sick leave compared with controls but there was some improvement in performing daily activities. Interestingly, de qi was reported in 93% of the acupuncture group and 28% of the sham group! The authors felt this may be due to the sham treatment not being totally inert. There were baseline differences in numbers on sick leave, therefore resting (more for the sham group) , which may have biased the trial results. Most women with pelvic pain in pregnancy find that their pain improves when they stop work and rest more.


Pain relief in labour

Refer to table 3

Eight articles refer to uncontrolled case series and in all of these1,30,34,40,46,61,66,67 there were success rates (i.e. adequate pain relief achieved) of 56-92%.. Of the eight, five used only EA, one used EA plus MA, one MA plus moxibustion and one EA or MA. In this last there was 60% success with EA but 0% with MA. Hence in none of these studies was MA alone shown to be effective.

In one of the controlled trials63 only 10% of patients reported adequate analgesia though this was set up primarily to compare manual acupuncture (MA) and EA.

For the other sixteen controlled trials nine were RCTs and the rest were contemporaneous or retrospective matched controls. In contrast to the case series most of them used MA. Seven 6,27,44,49,55,56,69 out of sixteen found acupuncture reduced the need for conventional medication and  four21,45,50,51 showed a reduced pain rating (though Ramnero et al49 achieved fewer epidurals without changing pain ratings and Ziaei and Hajipour69 similarly found no significant differences in pain rating but less medication required by the acupuncture group). Five13,38,39.54,68 found no significant differences in analgesic requirements between the acupuncture and control groups but most of them provided only weekly treatments from week 35 or 36 until delivery rather than intensive acupuncture during labour itself.

Five RCTs on pain relief in labour have been added to this update. One trial 69 using acupuncture on a small group of women during labour found no significant differences among the groups in pain and relaxation score, duration of labour or caesarean section rate but fewer women needed augmentation (increasing or enhancing contractions with an intravenous infusion of the hormone 'Syntocinon', a synthetic from of oxytocin) in the acupuncture group than the two control groups.

One study 21 compared an electro-acupuncture group to a control group that had no pain relief.  Beta-endorphins and 5HT levels were compared and found to be significantly higher in the acupuncture group which also recorded lower pain intensity and better relaxation.

A study on the effects of acupuncture during labour on nulliparous (first baby) women 27 found pain scores reduced more with real than sham acupuncture though differences in baseline levels could have biased the results. The main finding of this study was a reduction in duration of active phase and less need for oxytocin augmentation. There was a high willingness to use acupuncture again (95%) in the study group .

More recently, one prospective RCT 39 compared acupuncture with subcutaneous injections of sterile water as treatment for labour pain. There were no significant differences regarding the need for additional pain relief, and women given sterile water injection experienced less labour pain compared to women given acupuncture. (This technique is used in Scandinavian countries, the USA and Canada).

The largest RCT to date 6 compared acupuncture to TENS and traditional analgesia. They also looked at birth experience and obstetric outcome. Use of pharmacological and invasive methods was significantly lower in the acupuncture group, mean Apgar scores and cord blood pH (indicators of fetal wellbeing) were significantly higher in the acupuncture group.

A recently published case- controlled pilot study13 found the acupuncture group had significantly fewer caesarean sections (7% vs 20 % p = 0.004) This involved a small sample size but at a time when there is a strong move to try and reduce Caesarean rates this is an important finding. No significant differences were noted in other clinical endpoints but 87% of patients reported that acupuncture had helped them.

Given the substantial increase in evidence from RCTs in the last few years it is useful to summarise their results as a separate group in more detail. Four21,44,45,49 of the nine RCTs compared acupuncture to no (additional) intervention; the numbers of subjects ranged from 36 to 200; three were wholly positive and one partly positive for acupuncture in respect of the primary outcomes. Three27,51,69 compared acupuncture to sham acupuncture; the numbers of subjects ranged from 90 to 210; two were positive and one partly so. Two6,39 compared acupuncture to other interventions; the numbers of subjects were 128 and 607; one was positive (v. TENS or traditional analgesia) and one negative (v. sterile water).

Induction of labour and duration of labour

Refer to table 4.

Four case series34,59,60,67 used EA and successfully induced 66-100% of women at term. Acupuncture was much less effective pre-term and for mid-term abortions34,60. By contrast, inhibition of early labour may be possible60. The other two case series showed that acupuncture could reduce the duration of labour, measured as either induction to delivery interval46 or the speed of cervical dilation (an increase in which was also associated with more intense and frequent contractions)36.

Three controlled trials on induction of labour produced positive results for acupuncture compared with no treatment. In one33 all 35 women having EA started contractions within 25 minutes, and they were stronger and more frequent. In the second,38 success was measured by cervical ripening and in the time between due date and delivery. The third and more recent study 28 found mean time to delivery to be 21 hours sooner in the acupuncture group and Caesarean section was less likely. As stated above, any data on reduction of Caesarean rates at this time is welcome to all professionals, regulators and legislators and of course patients and will hopefully inspire further studies in this area.

Eight23,27,32,45,50,51,54,68,69 out of ten controlled trials have shown acupuncture to decrease length of labour, though in two of these this was only for primiparous women. By contrast Lyrenas et al37 found no such reduction and indeed longer gestations. In her recent review5 Betts discusses some possible reasons for these anomalous results. Ziaei69 found no significant effect on overall duration or the lengths of the different stages.

Again, faster cervical dilation in the first stage of labour was found to be associated with increases in both intensity and frequency of contractions33. One of the most recent studies 23 was on augmentation after spontaneous rupture of membranes at term and found a significant reduction in duration of labour and need for Oxytocin.

Miscellaneous conditions

Refer to table 5.

One report11 refers to acupuncture for placental retention, with a positive outcome compared to the control group. Another64 is not a specific study but a status report on the use and effectiveness of acupuncture analgesia for Caesarian Section in China during the 1970s and 80s. A third study 25 compared acupuncture to counselling and psychotherapy for women with mild to moderate emotional complaints in pregnancy with a positive outcome for the acupuncture group.



Swan and Cook53(2003) reviewed acupuncture in obstetric care and concluded that 'the available evidence is not of sufficient strength or quality to support the widespread introduction of acupuncture into obstetrics under the banner of evidence- based practice.'


Betts5 (2006) discussed selected acupuncture studies for pelvic pain, morning sickness, breech presentation, cervical ripening and pre-birth acupuncture. She wrote primarily from the perspective of an experienced user and teacher of traditional acupuncture, and a midwife, rather than from that of a medical researcher. Hence she was able to point out the possible benefits of research to the profession, as well as the dubious procedures that have been employed in some of the study protocols.


Back and pelvic pain in pregnancy

Ee et al's (2008) systematic review16 of acupuncture for back and pelvic pain in pregnancy included two small trials and one large trial. Acupuncture as an adjunct to standard treatment was superior to standard treatment alone and physiotherapy in relieving mixed back/pelvic pain. However, they concluded that limited evidence supported acupuncture in treating this condition and that additional high quality trials were needed to test the existing promising evidence.

Breech presentation

Coyle et al's (2007) systematic review14 of cephalic version by moxibustion for breech presentation included three trials involving a total of 597 women. The authors concluded there was insufficient evidence to support the use of moxibustion to correct a breech presentation, but it may be beneficial in reducing the need for ECV (external cephalic version) and decreasing the use of oxytocin before or during labour. Again there was a call for well-designed trials to properly evaluate this technique.


Van den Berg et al's (2008) systematic review62 looked at controlled trials on acupuncture-type interventions for breech presentation, including moxibustion, acupuncture or electro-acupuncture to point Bl67, compared to expectant management.  Six RCTs and three cohort studies were included in the review. They concluded that acupuncture type interventions were effective compared to expectant management, but that as some studies were of inferior quality, further RCTs of improved quality were needed.


Induction of labour

Smith and Crowther52 (2007) reviewed acupuncture for induction of labour, but found only one trial of 56 women which met the inclusion criteria. They therefore concluded that there was a need for well-designed RCTs in this area and for trials to assess clinically meaningful outcomes.

Pain relief in labour

Huntley et al's (2004) systematic review29 of complementary and alternative medicine (CAM) for labour pain, included only prospective randomised controlled trials. There were two on acupuncture meeting these criteria. Both scored 3 out of a possible 5 on the Jadad scale (for assessing methodological rigour) because of lack of double blinding. Although pain ratings were not greatly improved, the results suggested receiving a physical intervention like acupuncture does have an influence on a woman's pain management during labour. Again they stated that more research is warranted.


Lee and Ernst70 (2004) conducted a systematic review on acupuncture for labour pain and included three RCTs, all of good quality and comprising 496 participants. They concluded that the evidence for the use of acupuncture as an additional method of pain relief is promising, but that further research is warranted owing to the limited amount of data available.


Forbidden points

Although these are not research papers, it is important to mention the literature that exists on this topic.

There are some who prefer not to treat women in pregnancy with acupuncture, and this is often due to concerns over the safety aspects in relation to the so-called "forbidden points". Detailed discussion on this is beyond the remit of this briefing paper, but the reader may like to refer to articles such as those by Dale 15(1997), Chen 12(1998),  Forrester 22(2003) or Betts 4 (2005).



The majority of these studies show positive outcomes for acupuncture treatment of various conditions in pregnancy and labour. Despite the methodological deficiencies in many of the studies the verdict of the review papers seems over-cautious. Of the 34 controlled trials presented here (excluding two that just compared different types of acupuncture), only 5 did not produce positive results (in at least one of the primary outcomes) for acupuncture - though the size and significance of the benefit is still to be determined. There is a great difference in the choice of points, methods of stimulation and duration of treatment:  further studies would be required before specific treatment protocols could be recommended.

Reference list

1. Abouleish E and Depp R. Acupuncture in Obstetrics. Anesthesia and Analgesia 1975;54(1):83-88.

2. Allaire A. Complementary and Alternative Medicine in the Labor and Delivery Suite. Clinical Obstetrics and Gynecology 2001; 44(4):681-691

3. Beal M. Acupuncture and Acupressure, Applications to Women's Reproductive Health Care. Journal of Nurse-Midwifery. 1999; 44(3):217-230

4. Betts D. The use of acupuncture as a routine pre-birth treatment. Journal of Chinese Medicine 2005; 76: 5-8

5. Betts D. A review of research into the application of acupuncture in pregnancy. Journal of Chinese Medicine 2006; 80:50-55

6.Borup L, Wurlitzer W, Hedegaard M, Kesmodel U, and Hvidman L. Acupuncture as Pain relief During delivery: A Randomised Controlled Trial. Birth. 36:1 march 2009.

7. Cardini F, Basevi V, Valentini A, Martellato A 1991 Moxibustion and Breech Presentation, preliminary results. American Journal of Chinese Medicine 1991;19:105-14

8. Cardini F and Marcolongo A. Moxibustion for Correction of Breech Presentation: a clinical study with retrospective control. American Journal of Chinese Medicine 1993;21:133-8

9. Cardini F and Weixin H. Moxibustion for correction of Breech Presentation: a randomised controlled trial. Journal of the American Medical Association 1998 ;280: 1580-4

10. Cardini F, Lombardo P, Regalia AL, Regaldo G et al. A randomised trial of moxibustion for breech presentation. 2005; 112(6): 743-7

11. Chauhan P, Gasser F, Chauhan A. Clinical Investigation on the Use of Acupuncture for Treatment of Placental Retention. American Journal of Acupuncture, 1998;26(1):19-25.

12. Chen Ke-qin. Forbidden Points. International Journal of Clinical Acupuncture 1998 Vol9, No1

13.Citkovitz C, Klimenko E, Bolyai M, Applewhite L, Julliard K and Wieiner Z.

Effects of Acupuncture During Labor and Delivery in a U.S. Hospital Setting: A Case- Controlled Pilot Study. The Journal of Alternative and Complementary Medicine. Vol 15, No 5, 2009. Pp501-505.

14. Coyle M, Smith C and Peat B. Cephalic version by moxibustion for breech presentation. Cochrane Database of Systematic Reviews, 2007, no. 4.

15. Dale R. The Contraindicated (Forbidden) Points of Acupuncture for Needling, Moxibustion and Pregnancy. American Journal of Acupuncture, 1997;25(1):51-57.

16. Ee C, Manheimer E, Pirotta and White A.  Acupuncture for pelvic and back pain in pregnancy : a systematic review. American Journal of Obstetrics and Gynecology. March 2008. 254- 259.

17. Elden H, Ladfors L,  Olsen MF, Ostgaard H-C and Hagberg H Effects of acupuncture and stabilising exercises as adjunct to standard treatment in pregnant women with pelvic girdle pain: randomised single blind controlled trial. British Medical Journal. 2005; 330:761-765

18. Elden H, Hagberg H, Fagevik Olsen M, Ladfors L, Ostgaard H. Regression of pelvic girdle pain after delivery: follow-up of a randomised single blind controlled trial with different treatment modalities. Acta Obstetrica et Gynecologica. 2008; 87: 201-208

19. Elden H, Ostgaard H-C, Fagevik-Olsen M, Ladfors L and Hagberg H. Treatments of pelvic girdle pain in pregnant women: adverse effects of standard treatment, acupuncture and stabilising exercises on the pregnancy, mother, delivery and the fetus/neonate. BMC Complementary and Alternative Medicine 2008, 8:34.

20. Elden H, Fagevik-Olsen M, Ostgaard H, Stener-Victorin E, Hagberg H. Acupuncture as an adjunct to standard treatment for pelvic girdle pain in pregnant women: randomised double-blinded controlled trial comparing acupuncture with non-penetrating sham acupuncture. BJOG 2008; 115: 1655-1668

21. Fan Qu and Jue Zhou. Electro-Acupuncture in Relieving Labor Pain. Evidence-based Complementary and Alternative Medicine. 2007. March; 4 (1) : 125-130

22.. Forrester M. Low back pain in Pregnancy. Acupuncture in Medicine. 2003;21(1-2):36-41.

23. Gaudernack L, Forbord S and Hole E. Acupuncture administered after spontaneous rupture of membranes at term significantly reduces the length of birth and use of oxytocin. A randomized controlled trial. Acta Obstetrica et Gynecologica. 2006; 85: 1348-1353

24. Guerreiro da Silva JB, Nakamura MU, Cordeiro JA, Kulay L Jr. Acupuncture for Low Back Pain in  Pregnancy- a prospective, quasi-randomised, controlled study. Acupuncture in Medicine. 2004; 22(2):60-67.

25.Guerreiro da Silva . Acupuncture for mild to moderate emotional complaints in pregnancy- a prospective, quasi-randomised, controlled study. Acupuncture in Medicine. 2007; 25 (3): 65-71.

26. Habek D, Habek J and Jagust M. Acupuncture Conversion of Fetal Breech Presentation.

Fetal Diagnosis and Therapy. 2003; 18: 418-421

27. Hantoushzadeh S, Alhusseini N and Lebaschi A. The effects of acupuncture during labour on nulliparous women: A randomised controlled trial. Australia and New Zealand Journal of Obstetrics and Gynaecology. 2007; 47: 26-30

28. Harper et al. A randomised controlled trial of acupuncture for initiation of labour in nulliparous women. Journal of Fetal Neonatal Medicine. 2006 Aug; 19 (8): 465-70

29. Huntley A, Thompson Coon J, Ernst E. Complementary and alternative medicine for labor pain: a systematic review. American Journal of Obstetrics and Gynecology 2004; 191: 36-44.

30. Hyodo M and Osamu G. Use of Acupuncture Anesthesia for Normal Delivery. American Journal of Chinese Medicine. 1977;5(1):63-69.

31 Kanakura Y, Kometani K, Nagata T, Niwa K, Kamatsuki H, Shinzato Y, Tokunaga Y. Moxibustion Treatment of Breech Presentation. American Journal of Chinese Medicine. 2001;.29(1):37-45.

32. Kubista E, Kucera H. Acupuncture as a Method of Preparation in Obstetrics. American Journal of Chinese Medicine 1974;2(3):283-287.

33.. Kubista E, Kucera H, Muller-Tyl E. Initiating Contractions of the Gravid Uterus Through Electro-acupuncture. American Journal of Chinese Medicine. 1975;3(4):343-346.

34. Ledergerber C. Electroacupuncture in Obstetrics. Acupuncture and Electro-Therapeutics Research 1976;2:105-118.

35. Li Qinghua, Wang Lin. Clinical Observation on Correcting Malposition of Fetus by Electro-Acupuncture. Journal of Traditional Chinese Medicine. 1996;16(4): 260-262.

36. Lin Pei-chong. Acupuncture to Accelerate Labour. International Journal of Clinical Acupuncture. 1998 ;9(4):459-462.

37. Lyrenas S Lyrenas S, Lutsch H, Hetta J, Lindberg B. Acupuncture before delivery: effect on Labour. Gynecologic and Obstetric Investigation 1987;24:217-224

38. Lyrenas S, Lutsch H, Hetta J, et al. Acupuncture before delivery: effect on Pain Perception and the Need for Analgesics. Gynecologic and Obstetric Investigation. 1990; 29 (2): 118-124.

39. Martensson L, Stener-Victorin E and Wallin G. Acupuncture versus subcutaneous injections of sterile water as treatment for labour pain. Acta Obstetrica et Gynecologica. 2008; 87: 171-177

40. Martoudis S and Christofides K. Electroacupuncture for Pain Relief in Labour. Acupuncture in Medicine 1990;8(2):51-53.

41. Neri I, Fazzio M, Menghini S, Volpe A, Facchinetti F. Non-Stress Test Changes During Acupuncture Plus Moxibustion on BL67 Point in Breech Presentation. Journal of the Society of Gynecological Investigation 2002;9(3) :158-162

42. Neri I, Airola G, Contu G, Allais G, Facchinetti F and Benedetto C. Acupuncture plus moxibustion to resolve breech presentation: a randomised controlled study. The Journal of Maternal-Fetal and Neonatal Medicine. 2004; 15: 247-252.

43. Neri I, De Pace V, Venturini P and Facchinetti F. Effects of Three Different Stimulations (Acupuncture, Moxibustion, Acupuncture plus Moxibustion) of BL.67 Acupoint at Small Toe on Fetal bahviour of Breech Presentation. The American Journal of Chinese Medicine, 2007; 35(1): 27-33.

44. Nesheim B  Kinge R, Berg B, Alfredsson B, Allgot E, Hove G, Johnsen W, Jorsett I, Skei S, Solberg S. Acupuncture During Labour Can Reduce The Use of Meperidine: A Controlled Clinical Study. The Clinical Journal of Pain 2003;19(3): 187-191

45. Pei Deen and Huang Yuelan. Use of Acupuncture Analgesia During Childbirth. Journal of Traditional Chinese Medicine. 1985;5(4): 253-255

46. Perera W. Acupuncture in Childbirth. British Journal of Acupuncture 1979;2(1):12-14

47. Qin G and Tan H. 413 cases of abnormal fetal position corrected by auricular plaster therapy. Journal of Traditional Chinese Medicine. 1989;9:235-237.

48. Rabl M , Ahner R, Bitschnau M, Zeisler H, Husslein P et al. Acupuncture for cervical ripening and induction of labour at term-a randomised controlled trial. Wiener klinische wochenschrift (The Middle European Journal of Medicine.) 2001 113(23-24):942-946

49. Ramnero A, Hanson U, Kihlgren M. Acupuncture treatment during labour- a randomised controlled trial. British Journal of Obstetrics and Gynaecology 2002 ;109(6):637-44

50. Skelton I.F and Flowerdew MW. Acupuncture and labour- A summary of results. Midwives Chronicle and Nursing Notes. 1988; May:134-137.

51. Skilnand E, Fossen D and Heiberg E.  Acupuncture in the management of pain in labour. Acta Obstet Gynecol Scand 2002;81:943-948.

52. Smith C and Crowther C. Acupuncture for Induction of Labour. Cochrane Database of Systematic Reviews, 2007, no.4.

53. Swan P and Cook T. Acupuncture in Obstetric Care. Journal of the Association of Chartered Physiotherapists in Women's Health. 2003;92:63-68.

54. Tempfer C, Reisler H, Heinzl H, Hefler L, Husslain P, Kainz C. Influence of acupuncture on maternal serum levels of interleukin-8, prostaglandin F2alpha, and beta-endorphin: a matched pair study. Obstetrics and Gynecology 1998;92(2):245-8

55. Ternov N, Nilsson M, Lofberg L, Algotsson L, Akeson J Acupuncture for pain relief during childbirth. Acupuncture and Electrotherapeutics Research 1998a; 23:19-26

56. Ternov N, Buchhave P, Svensson G, Akeson J,Acupuncture During Childbirth Reduces Use of Conventional Analgesia Without Major Adverse Effects: A Retrospective Study. American Journal of Acupuncture 1998b; 26 (4):233-239

57. Ternov N, Grennert L, Aberg A, Algotsson L, Akeson J Acupuncture for Lower Back and Pelvic Pain in Late Pregnancy: a Retrospective Report on 167 Consecutive Cases. Pain Medicine 2001;2(3):204-207.

58. Thomas C and Napolitano P. Use of Acupuncture for managing chronic pelvic pain in pregnancy, a case report. Journal of Reproductive Medicine. 2000;45(11):944-6.

59. Tsuei J, Lai F, Lai Y. Induction of Labor by Acupuncture and Electrical Stimulation. Obstetrics and Gynecology. 1974;43(3): 337-342.

60. Tsuei J, Lai Y and Sharma S. The Influence of Acupuncture Stimulation During Pregnancy: The induction and inhibition of labor. Obstetrics and Gynecology  1977; 50(4): 479-480.

61. Umeh B. Sacral Acupuncture for pain relief in labour: initial clinical experience in Nigerian women. Acupuncture Electrotherapeutics Research. 1986;11:147-151.

62. Van den Berg I, Bosch J, Jacobs B, Bouman I, Duvekot J and Hunink M. Effectiveness of acupuncture-type interventions versus expectant management to correct breech presentation: A systematic review. Complementary Therapies in Medicine. (2008) 16, 92-100.

63. Wallis L , Shinder S, Palahniuk R, Spivey H. An Evaluation of Acupuncture Analgesia in Obstetrics. Anaesthesiology 1974;41(6):596-601

64. Wang DW and Jin YH. Present status of Cesarean Section under Acupuncture Anesthesia in China. Fukushima Journal of Medical Science.1989;35(2):45-52.

65. Wedenberg K, Moen B and Norling A. A prospective randomised study comparing acupuncture with physiotherapy for low-back and pelvic pain in pregnancy. Acta Obstet Gynecol Scand.  2000;79(5):331-5

66. Yanai N, Shalev E, Yagudin E, Zuckerman H. The Use of Electroacupuncture During Labour. American Journal of acupuncture 1987;15(4):311-312

67. Yip S, Pang J and Sung M Induction of Labor by Acupuncture Electro- Stimulation American Journal of Chinese Medicine 1976;4(3):257-265

68. Zeisler H, Tempfer C, Mayerhofer K, Barrada M,  Husslein P. Influence of acupuncture on duration of labor. Gynecologic and Obstetric Investigation 1998;46(1):22-5

69. Ziaei S and Hajipour L. Effect of Acupuncture on Labor. International Journal of Gynecology and Obstetrics. 2006 ; 92, 71-72.

70. Lee H and Ernst E. Acupuncture for labor pain management: a systematic review. American Journal of Obstetrics and Gynecology. 2004; 191 (5): 1573-1579.

71. Zheng J. Legends about Acupuncture Treatment of Difficult Labour. International Journal of Clinical Acupuncture, Vol. 1 (3), pp. 309-10


Tables of acupuncture studies in obstetrics

Numbers in brackets below the author details are the reference list numbers, for ease of matching the text summaries to the table entries.











Cardini 1991     [7]

Case Series


Moxibustion to  Bl67


30 minutes daily for 15 days

61% success rate

Qin and Tan




Case series with

matched control



Ear seeds to 7 auricular points. 3 groups at different gestations

Knee-chest position for 40 patients

Seeds stimulated pre-prandially for 4 days. Repeat once more if ineffective


Significant difference in version at all 3 gestations compared with control

Cardini 1993



Clinical study with retrospective control matched for parity and gestational age.


Moxibustion to Bl67

Routine care with no intervention

Daily for minimum 5 days

Significantly higher version rate in study group compared to control.

Li and Wang



Randomised controlled.


Electro-acupuncture or moxibustion to Bl67.

No treatment

Daily until correction for up to 6 days

81% in electro- group

75% in moxibustion and 16% control( p<0.005 between electro and control)

Cardini 1998




Randomised, controlled.


Moxibustion to Bl67

Routine care with no intervention

Daily for 7 days plus 7 days if still breech

75.% cephalic at 35 weeks v 48% of controls (p<.001)

75% cephalic at birth v 62%, but 24 in control group had External Cephalic Version

Kanakura  et al. 2001



Matched  retrospective control  groups - one for each of 2 interventions



Moxibustion or electro-stimulation to Bl67, Sp6, Kid1.

No intervention.

Daily for 30 minutes until correction or until deemed non-responsive

92% moxibustion group turned to cephalic, 73% in controls (p<0.0001). In electro group, 89% vs 83%


Neri et al 2002





Single- blind cross-over design. Sham acupuncture followed by true, 1-2 days later




and moxibustion bilaterally to Bl67.


Minimal (sham) acupuncture

20 minutes per session, twice per week, up to 5 sessions if required

Reduction in foetal heart rate and increase in foetal movements with true acupuncture. No significant changes in control

Habek et al 2003


Randomised, controlled


34 acup

33 control

Manual acupuncture to Bl67

No Intrevention

30 minutes per day twice a week from 34 weeks. Mean of 6 acup treatments

76.4% version in acup group vs 45.4% controls (p=0.001)

Neri et al.







112 acup

114 control


Bilateral acupuncture PLUS moxibustion to Bl67

No intervention

Needles in situ with De Qi plus continuous moxibustion for 20 minutes, twice weekly for 2 weeks.

53.6% version in acup group vs 36.7% controls, (p=0.001)

Caesarean section rate 52.3% acup vs 66.7% controls. (p=0.03)

Neri et al




3-group comparison: moxibustion, acupuncture or acupuncture + moxa


15 moxa, 10 acup, 14 bothp

Bilateral acupuncture to

Bl 67, or bilateral with moxa or bilateral moxa alone.

Comparison of 3 different acu-moxa interventions

20 minutes per session twice weekly with testing 20 mins before, during and after treatment.

80% version for moxa alone, 57% for acupuncture plus moxa and 28% for acupuncture alone. Moxa alone affected fetal movements, acup plus moxa reduced fetal heart rate and movements.
















Thomas &



Pelvic [58]


Case report


Auricular to Shenmen, Sympathetic, Abdomen 2 and Lumbar vertebrae.


Needles retained for 8 hours, 5 times in 1st week, then twice weekly until delivery

Discharged from hospital after 4th treatment and much less narcotic analgesia needed

Ternov  et al


Both [57]

Retrospective case series


Various points, for 45 minutes, stimulated at 15 minute intervals



Good or excellent analgesia in 72% of patients



Back [22]


Case Report


Various points at each treatment


7 treatments over 9 week period

Pain rated 20-55 pre-acup. and 5-10 after (by VAS)


et al. 2000



Randomised controlled.


Acupuncture to ear points followed by  various body points when needed


3 times weekly for 2 weeks,  then twice a week, total of 10 treatments in 1 month

VAS values better after acup than physio in morning (p<.02) and evening (p<.01). Disability rating index significantly better in acup group only


Guerreiro da

Silva et al


Back    (24]

Prospective, quasi-randomised, controlled



(27 acup

34 control)

Various points based on TCM and individualised

Standard treatment with Paracetamol and Hyoscine

Mostly once weekly, twice if severe, over 8 weeks.

Pain reduction 4.8 points in acupuncture group

vs -0.3 control (p<0.0001)

Elden et al




RCT, single blind

Acup or exercise as adjuncts to standard therapy


(125 acu

130 exer

131 std)


a)      local points (segmental & extra-segmental) - by ahshi & pain diagnosis

b)      general pain-relieving points (unspecified)

1) Std therapy (information, advice, pelvic belt, home exercises)

2) Standard plus stabilising exercises

Twice a week for 6 weeks

Acup group had largest improvement in pain intensity - both self- & independently assessed. (Most outcomes statistically significant)

Elden et al




RCT, double blind.

1.Standard treatment plus acup or 2. standard plus non-penetrating sham acup


58 in group 1, and 57 in group 2.


Unclear from paper but included trigger points and a selection of points including LI4, Bl 32,33, Kid 11 -

considered by some to be forbidden at this stage in gestation (12-29 weeks)

Sham  needling.

12 treatments of 30 mins each, manually stimulated every 10 mins to evoke de qi, twice weekly for 4 weeks and weekly for 4 weeks.

No significant effect on pain  (p= 0.493) .  Acupuncture group in regular work more (p= 0.041)  and had superior ability to perform daily tasks on DRI1 (p=0.001)

De Qi reported in 28% of sham group!













Abouleish  &

Depp  1975


Case series


Electro-acupuncture to various points, usually 8 for each patient.




66% experienced relief of pain, but comment made on impracticality.






Case series


Manual or electro-acupuncture to St44, St36, Sp6 and others


[Not clear in paper]

9 cases with electro-acup successful, 6 not.

None in manual group successful.

Hyodo and Osamu  1977


Observational study



Electro-acupuncture to St36, LI4 and Sp6


Continuous  from early labour until 3rd stage of labour

Definite subjective and objective relief of pain in 60% of primiparas and 90% of multiparas after acupuncture.

Perera   1979



Case series


Acupuncture to Du20, LI4, St44, Bl67 all on left side only. Electro to Neima (ex) and Sp6 only



92% effective. Induction to delivery interval also shortened.

Umeh.   1986



Case series


Sacral acupuncture to Bl32 with manual stimulation and moxibustion


[Not stated in paper]

63% had adequate pain relief on VAS, with 31% of these reporting no pain at all in average 8 hours of labour.


Yanai et al.



Observational study


Electro-acupuncture to LI4 and ear Shenmen


At start of active stage of labour

56% mild to good pain relief by patient assessment. Midwives assessed it as 94% and physicians 87%.

Martoudis & Christofides



Observational study


Electro-acupuncture to LI4 and ear Shenmen

bilaterally for 20 or 30 minutes


Dependent on duration of labour

Slight to very good benefit in 88% of cases, "failure rate" 12%.

Yip et al, 1976 - see section 5 - women treated with acupuncture needed less analgesia than usual


Wallis et al




Comparison between manual and electro-acupuncture



Points according to TCM diagnosis

Electro versus manual acupuncture

According to patient request for analgesia

19 out of 21 reported inadequate analgesia

Pei and Huang  1985



Randomised controlled


(100 acup  100 control)


Electro-acupuncture to either Bl32 or special local point or combined Bl32, Sp14 and St30 .


No intervention


94-97% in treatment group had adequate analgesia vs 0%  in control group, and acupuncture group had shorter course of labour.


Skelton and




Non-randomised, controlled


Electro-acupuncture to Sp6, St36 and Neima (ex).

Conventional analgesia.

Continuous until delivery. Entonox also available.


Only multiparous patients had less pain. Significantly shorter first stage for primigravids than control.

Lyrenas et al. 1990



Contemporan-eous matched control group.


Acupuncture   to St 36,  GB34, Sp6 and Bl62 bilaterally

No intervention

Once weekly until delivery from 36 weeks and for 30 mins each time

No lessening of labour pain, nor reduction in analgesic requirements, with acupuncture

Ternov et al 1998a.


Contemporan-eous matched control group.


(90 in each group)

Individualised to a variety of points


No intervention.

Variable but mostly continuous throughout labour.


Standard analgesia needed for  40% in acupuncture group vs 87% controls (p<0.0001)

Ternov et al. 1998b.





Retrospective comparison of 12 months records: before and after acupuncture introduced.


(1708 no acup 1609 with acup)

Individualized and including tangential needling.

No intervention.

Variable but mostly continuous throughout labour.

After introduction of acupuncture, significant reduction in use of nitrous oxide, IM Meperidine, local Bupivicaine and sterile water (p<0.01)

Ramnero et al   2002



Randomised controlled


Individualised acupuncture to various points.

No intervention

Needles left in situ for one to three hours

Acupuncture group: significantly fewer epidurals & more relaxed. No difference in pain or labour outcomes.

Skilnand et al   2002



Controlled, single-blind, randomised study



Individualised acupuncture to various points

Minimal sham acupuncture

Variable but mostly continuous with needles taped down

True acupuncture reduced labour pain scores by c.20% more than sham and decreased time to delivery by over an hour.

Nesheim et al.   2003




Controlled: part randomised, part matched group. Non-blinded.


(106 acup

184 controls)

Individualised choice of acupuncture points with deqi .

No intervention

(92 randomised

92 matched from register)

Most needles in situ 10 to 20 mins, some less, and some retained

Meperidine needed by 11% acupuncture group, 37% control group 1 (p=0.0001)and 29% control group 2

Ziaei and Hajipour




Randomised, controlled


30 acup

30 = control 1

30 = control 2

Points: GV20, Yintang, St 36, Sp6, LI4, LI3, CV2,3.

Control1: Needled 6 points normally used for vaccination . Control 2: No intervention

De Qi obtained. Needles taped and left in situ until delivery.

No significant differences in pain intensity or degree of relaxation but reduced need for augmentation (using intravenous oxytocin to stimulate contractions) in acupuncture group (p= 0.03 for this)



Fan Qu et al 2006




Randomised, controlled


Electro-acupuncture to LI4 and Sp6 bilat. 2-100 Hz dense-disperse. Strength increased gradually.

No intervention or pain relief!

20 mins then repeated when 7-8 cms of cervical dilatation reached.

Mann-Whitney Test used to assess pain scores. Beta-endorphins and 5HT levels measured by blood analysis. Lower pain intensity in acupuncture group and better relaxation (p=0.018 & 0.013) Significant difference in concentration of endorphin and 5HT (p=0.037 & 0.030)







Randomised, controlled


70 real acup

74 minimal acup

Manual acu to individualised points inc LI4, Bl32, UB60, Sp6, St36, Liv3, GB34, Ht 7.

De Qi obtained

Non-acupuncture points. Needles manually stimulated for 20 mins but patients not asked about De Qi!

From active phase of labour until delivery or when patient requested removal or effect terminated or no effect.

VAS scale used for pain assessment.

1y outcome=pain, duration and acceptability; 2y =oxytocin use.

Acup group better for pain score after 2 hours (p= <0.001), duration of labour (p<0.001) and oxytocin use (p=0.001 ). Willingness to have acup again 95% vs 74% controls


Martensson et al 2008



Prospective, randomised, controlled


62 acup

66 sterile water inj.

Manual to GV20, LI4, Sp6 plus  individualised local points from Bl23, 24, 54, Ex19, GB25-29 and Kid 11.

De Qi obtained.

5-8 subcutaneous injections of 0.5 ml sterile water in area of pain.






Needles in situ 40 mins with manual stimulation every 10 mins.

Sterile water injections provided greater pain relief than acupuncture

Borup et al. 2009



Prospective, randomised, controlled



(314 acup

144 TENS

149 traditional analgesia)

Individualised manual acupuncture chosen from 34 specified points according to woman's mobility and pain locality.

TENS to lower back or traditional analgesia

Variable from 30 minutes to 2 hours and could be repeated.

Use of pharmacological and invasive methods significantly lower in acupuncture group (vs traditional p<0.001; vs TENS p=0.031) Pain scores comparable. Mean Apgar score at 5 mins and cord pH significantly higher (i.e. better) in acupuncture group. No difference in duration of labour or use of Oxytocin

Citkovitz C, et al. 2009







Case- controlled pilot study

45 acup, 127 historical controls

Electro-acupuncture to individualised points chosen from study protocol, with all patients having ear Shenmen, Uterus, Endocrine points plus LI4 and Sp6.

Matching with 1-3 patients drawn in reverse chronological order on basis of matching parameters.

Electro-stimulation at continuous frequency of 10Hz, amplitude set to patient's comfort level between 3-6 mA. Duration not mentioned.

Acupuncture group had significantly fewer Cesarean sections (7% vs 20%, p= 0.004). No significant differences noted in other clinical endpoints but 87% of patients commented how acupuncture had helped them. On multiple occasions, increase in strength and/or regularity of contractions noted  after needle stimulation esp to BL67

Tempfer et al, 1998 - see section 5 - no significant difference in use of analgesics


Zeisler et al, 1998 - see section 5 - no significant difference in use of analgesics

















Tsuei et al 19744  1974





Induction of labour

Observational study

12 (4 still- births,

1 missed abortion,

7 post-dates)

Acupuncture to LI 4 and  Sp 6 bilaterally with electro-stimulation for 10 participants


Continuous throughout labour

Uterine contractions initiated in 10 out of 12 cases (83% success rate). Average induction to delivery time was 13.1 hours.

Lederger-ber   1976




Induction of labour

Case series


(12 at term, 5 pre-term)


Electro-stimulation and electro-acupuncture to Ren3 and Sp15.


Induction: stimulation every 3 mins for 15 secs, followed by needling if unsuccessful.


100% success in 12 patients at term;

of 5 pre-term patients, 3 were successful.

Yip et al 1976


Induction of labour

Case series


Electro-Stimulation to Sp6 and LI4 at 5 cycles per second.


Continuously throughout first stage of labour

21/31 were successful;

majority needed less analgesia than is usual.

Tsuei et al









Induction and inhibition of labour

Observational study

60 (41 induction at term,

7  mid-term abortion,

12  inhibition of premature labour)

For inhibition of labour: electro-acupuncture to Sp 4. Induction: electro-acupuncture to LI4 & Sp 6. For mid-term as above plus  GB34 & Ren1 on 2nd day.


For induction, max 8 hours, repeated next day if no response. If response, treat until delivery. For inhibition: twice daily for 1st 3 days then twice weekly up to 20 times.


For  induction  78% success  at term. 0% for mid-term abortion. For inhibition, success rate 92% .





To accelerate labours with abnormal progress.

Case series


LI4(reinforced) and Sp 6(reduced)


For 30 minutes

Average speed of cervical dilation increased after acupuncture (P=<0.001). Intensity and frequency of contractions also improved (P=<0.05)

Perera, 1979 - see section 3 - induction to delivery interval shortened




Kubista & Kucera





Preparation for labour from 37 weeks gestation in primiparae.

Contemporan-eous matched control group


(60 acup,

60 control)


St 36, Kid 8, GB34 and Bl62 . No manual manipulation or electric stimulation; even technique; deqi.


No intervention

At weekly intervals before due date, 3 to 4 times, for 20-25 minutes.

Subjective length of labour shorter in acup group (P<0.02) and active phase of labour shorter in acup group (P<0.1)


Kubista et al. 1975





Induction of labour with intact membranes

Contemporan-eous matched control group


(35 acup,

35 control)

Electro-acupuncture to Kid 8, St36, Ren 6, "Bachmann 25" point.

No intervention


For an average of 2 hours

All in treatment group experienced contractions within 25 minutes. 31 had statistically significant increase in contraction frequency and intensity (P<0.01). No significant change in controls.


Lyrenas et al.








Length of pregnancy and duration of labour

Controlled, not randomised-acupuncture group self-selected.


(56 acup, 112 control plus 36 in two  reference groups)

St36, Sp6, GB34 and Bl62. Manual needling; even technique; deqi.


Control group: no intervention

Ref grp 1: lumbar puncture and interview

Ref grp 2: no intervention.


Once weekly from week 36 until delivery.

Acupuncture group appeared to have longer gestations and duration of labour was not shortened (second stage was longer). More use of oxytocin in acupuncture group.


Tempfer et al. 1998








Duration of labour;

levels of maternal serum factors involved in cervical maturation


Matched pairs -

women giving birth in same time period, matched for age and parity



(40 acup,

40 control)


Du20, He7, P6 :

bilateral, manual needling with deqi





Once weekly for 4

weeks, starting week 35


Total labour: 136 mins less in

acup group (p<.001). First stage (3cm up to full dilation ): 139 mins less (p<.001). Second stage: 2 mins longer

No signif diffs in interleukin-8, prostaglandin F2α, β-endorphin.

No differences in analgesic use or maternal birth injuries


Zeisler et al    1998



Duration of labour.

First parity only.


Contempor-aneous matched control group

57 acup

63 control

Du20, He 7, P6: bilateral, manual needling with deqi

No acupuncture

Once weekly for 4 weeks, starting week 36

Median duration of first stage of labour was 196 mins in acup group v 321 in control. No difference in second stage.

Rabl et al. 2001



Cervical ripening and induction of labour at term

Randomised controlled


(25 acup,

20 control)

LI4 and Sp6, "neutral" needle technique, deqi obtained.

No intervention

For 20 minutes

Acupuncture helped cervical ripening (P=0.04) and shortened time interval between due date and actual time of delivery (5.0 vs 7.9 days) (P=0.03)


Gaudernack et al. 2006



Augment-ation in labour after spontaneous rupture of membranes at term.

Randomised controlled


43 acup

48 control


All had St36, Liv 3, and CV4.

Individualised points acc to TCM, from GB41,K6, K3, Sp6, LI4, San6,Lu7, Ht7


No acupuncture

For 20 minutes

Duration of labour significantly reduced in acup group (p=0.03) and significant reduction in need for oxytocin (p=0.018). Acup group who needed induction had significantly shorter active phase (p=0.002)


Harper et al.




Induction of labour in nulliparous women

Randomised controlled. Usual medical care vs usual care plus acupuncture


30 acup

26 control

LI4, Sp6

Electro to Bl 31 and 32

All points bilateral

Routine medical care

30 mins for 3 out of 4 consecutive days

Mean time to delivery 21 hours sooner in acup group (p=0.36); spontaneous labour 70% vs 50% favouring acup group (p=0.12);

Caesarean Section less likely (p=0.07)

Pei and Huang, 1985 - see section 3 - shorter course of labour in acupuncture group


Skelton & Flowerdew 1988 - see section 3 - significantly shorter first stage for primigravids in acupuncture group


Skilnand, 2002 - see section 3- time to delivery reduced by more than 1 hour in acupuncture group


Ziaei   (69) 2006 - see section 3 - no effect on duration of labour


Hantoushzadeh 2007 (27) - see section 3 - shorted duration of labour and less oxytocin
















Wang & Jin








Caesarian section

- acupuncture


Retrospective case series

a) 24271



b) 16649


Refer to paper for details


Refer to paper for details

a) 92% success

(i.e. sufficient analgesia achieved)

b) 99% success


Refer to paper for further details









Placental Retention

Retrospective comparison over 2 years



(30 - acup

45 - control)

Acupuncture bilaterally to Bl67 and/or Ren3

Manual removal of placenta

Acupuncture until delivery of placenta, up to 20 minutes, then considered non response.

83% in acupuncture group delivered placenta in 20 minutes. Complication

rate 20% in acupuncture group and 58% in manual removal group.


da Silva







Emotional complaints in pregnancy

Prospective, quasi-randomised, controlled


(28 acup, 23 control)

Standardised points including Ht7, PC6, Lu9, St36, Liv3, Yintang,GV20, CV17, plus up to 4 additional points to individualise treatment.

Counselling and/or phytotherapeutic agents (Passiflora edulis or Hypericum perforatum)


Weekly, occasionally twice over 8 weeks = minimum 8 trs, maximum 12 trs

Numerical Rating Scale scores of intensity of emotional distress decreased by at least half in 60% of the study group and 26% of control group, (p=0.013)



[1] DRI = disability rating index