Prenatal Massage

What the sources reviewed establish

The sources reviewed for this page do not establish when or where massage adapted for pregnancy first appeared, and no dated origin is claimed here. Touch and massage around pregnancy and birth are described in many cultures, but the verified sources opened for this draft are modern professional and research documents, not historical or anthropological studies of earlier practice. A reviewer with access to scholarship in the history of midwifery would be needed to add a sourced account of earlier traditions.

Professional teaching in recent decades

A 2011 article in the Massage Therapy Journal of the American Massage Therapy Association describes pregnancy massage as a specialty within massage practice. The article quotes practitioners and educators, cites no research studies, and is best read as an account of professional opinion. It states that advanced training in pregnancy massage is strongly recommended.

Controlled research on massage in pregnancy in the sources reviewed begins with trials from a research group led by Tiffany Field, published in 2007 and 2011, which tested massage in women with depression during pregnancy. Later systematic reviews, published in 2020 and after, pooled this and other work.

Names used in the literature

The practice appears under several names. The review by Hall and colleagues uses 'antenatal' for massage given before birth, while a survey of Australian massage therapists covers 'preconception, antenatal or postnatal' massage as separate service periods. 'Prenatal' and 'pregnancy' massage are used interchangeably in trial reports and professional articles. This page uses prenatal massage for massage given to a pregnant person before labor begins and treats labor massage separately.

Related terms that are not the same practice

Perineal massage, in which the tissue around the vaginal opening is massaged in late pregnancy or during the second stage of labor, is studied for perineal tears and is a distinct intervention. It is outside the scope of this page. Foot reflexology and aromatherapy massage also appear in pregnancy trials, and results from those practices are not carried over to prenatal massage here.

Positioning and session setting

The AMTA article describes side-lying as the main recommended position, and says that by the second trimester all massage should be given this way to keep pressure off the vena cava, the large vein returning blood to the heart. It suggests lying on the left side to support venous return and describes cushions, pillows and bolsters for comfort. The article presents these as practitioner guidance, not as findings from tested protocols. It also notes that some pregnancy-specific points are considered contraindicated but does not list them, referring readers to advanced training.

Training of therapists

A 2020 survey of 99 Australian massage therapists who offered preconception, antenatal or postnatal massage found that 72.7% had pregnancy-specific training. The authors reported that no regulatory body enforced minimum educational standards for massage therapists in Australia and identified gaps in training, including up-to-date curriculum, and fragmented industry training requirements. They suggested this may affect the safety and benefits of pregnancy massage. The AMTA article similarly recommends advanced training. No agreed international training standard was found in the sources reviewed.

Session formats used in trials

Trials differ in who gives the massage and how often. In a 2007 trial of 47 women with depression in pregnancy, partners gave massage twice weekly from 20 weeks of gestation until the end of pregnancy. In a 2011 trial of 84 women with depression, a therapist or study protocol delivered 20-minute sessions twice weekly for 12 weeks. A Hong Kong labor trial taught partners a massage technique in a 2-hour class at 36 weeks. These are research protocols, not standards of practice.

Outcomes that trials have measured

Prenatal massage is offered for pregnancy-related discomfort and stress, and the outcomes tested in the sources reviewed reflect this. Trials by the Field group measured back pain, leg pain, depression, anxiety and anger in women with depression during pregnancy. Systematic reviews have pooled anxiety and depression scores, and a separate set of reviews has examined massage for pain during labor.

Pregnancy-related back and pelvic pain

Low-back and pelvic pain are common in pregnancy. The 2015 Cochrane review of interventions for these conditions included 34 randomized trials with 5,121 women. Its summary of findings highlights exercise, osteomanipulative therapy, acupuncture, craniosacral therapy and a multimodal package of manual therapy, exercise and education. Massage is not among the interventions with findings reported in its abstract, so that review does not by itself show whether massage reduces these pains.

Anxiety and depression in pregnancy

A 2020 systematic review by Hall and colleagues included eight studies, seven of them randomized trials, of massage for anxiety or depression in pregnant women over 18. A meta-analysis of four studies found lower scores on the Center for Epidemiologic Studies Depression Scale than with usual care (mean difference -5.95, 95% confidence interval -8.11 to -3.80), and a meta-analysis of five studies found lower anxiety (standardized mean difference -0.59, 95% confidence interval -1.06 to -0.12, with substantial heterogeneity, I-squared 75%).

The authors noted that none of the trials had a low risk of bias, outcomes were self-reported, and the findings may be prone to bias. They stated that more research on safety and effectiveness is needed before practice recommendations can be made.

Trials by the Field group

In the 2007 trial, 47 women with depression were randomized to partner-given massage twice weekly from 20 weeks or to a control group. Self-reported leg pain, back pain, depression, anxiety and anger fell more in the massage group. In the 2011 trial, 84 women with depression were randomized to yoga, massage therapy or standard prenatal care for 12 weeks. Both active groups showed larger decreases in depression, anxiety, and back and leg pain than controls, and the active groups had greater gestational age and birth weight at delivery than the control group.

Both trials share authors, were small, enrolled women who were already depressed, and relied on self-report. The abstracts do not describe blinding, and a participant cannot be unaware of receiving massage. Findings from these trials should not be extended to women without depression.

Massage during labor

A 2024 meta-analysis of sacral massage during labor included nine articles, eight in the meta-analysis. It found lower labor pain intensity, with very high heterogeneity (I-squared 97%) and moderate certainty of evidence for pain, and no significant change in the length of the first stage or total duration of labor, with very low certainty. A larger 2026 network meta-analysis of 41 studies reported that massage reduced first-stage pain compared with routine care (standardized mean difference -0.69, 95% confidence interval -1.14 to -0.23), but also noted that most included studies were of low quality with high risk of bias.

In a Hong Kong randomized trial of 479 low-risk first-time mothers, a 2-hour childbirth massage class for partners did not significantly change use of epidural analgesia or pethidine (12.0% versus 15.9%). More women in the massage group had labor without analgesia (29.2% versus 21.5%). The finding is mixed, and it concerns partner-delivered massage during labor, not prenatal massage sessions.

Quality of the evidence

Across the sources reviewed, trials of massage in pregnancy are small, rely on self-reported symptoms, and cannot blind participants to the intervention. Hall and colleagues judged that none of the trials in their review had a low risk of bias, and Dong and Li reported high risk of bias in most labor studies. Comparisons with usual care cannot separate any effect of massage touch from the effect of time, attention and relaxation, and the sources reviewed did not include sham-controlled trials.

Gaps in what was found

The searches run for this draft found no systematic review of massage specific to pregnancy-related low-back or pelvic pain, and no large trial of massage for sleep or swelling. Searches run were PubMed queries combining massage, pregnancy, back pain, anxiety, depression, labor and adverse events, and other databases were not searched. Professional claims for these uses therefore have no verified support in this draft, and the absence of a review is not evidence of either benefit or lack of benefit.

A reported case of fatal pulmonary embolism

A 2020 case report described a 25-year-old woman at 25 weeks of gestation who had leg and foot massage at a traditional massage shop in Thailand. Shortly afterward she developed sudden breathlessness and then collapsed with cardiac arrest. Imaging showed massive pulmonary embolism. She was resuscitated but remained in a coma, and the fetus died in utero. The authors concluded that leg massage can dislodge a clot in deep vein thrombosis, that it should be contraindicated in that situation, and that traditional massage in pregnant women should be contraindicated unless they are shown to have no such risk.

This is a single case report. It does not show how often such events occur, and the authors themselves state that serious adverse events from traditional massage are very rare. It also describes traditional massage in a shop setting, which may differ from a prenatal session given by a trained therapist. The case is relevant because pregnancy raises the likelihood of an undetected clot, as the authors state.

Other harms reported in trials and reviews

The Cochrane review of back and pelvic pain interventions reported that adverse effects, when reported, were minor and transient, but that statement covers all interventions in the review, not massage specifically. The 2026 network meta-analysis reported a higher incidence of emergency cesarean section in the massage group than in routine care in its labor studies (36%), and its authors advised caution in interpreting this because of high risk of bias. The abstract does not establish that massage caused this difference. Hall and colleagues state that more evidence on safety is required.

Suitability and care boundaries

Sources reviewed here do not provide a validated list of contraindications. Practitioner guidance in the AMTA article advises side-lying positioning and advanced training, and the case report above describes leg massage as contraindicated where a clot is present. Pain, swelling or redness in one leg, sudden breathlessness, bleeding, or concern about a pregnancy complication are matters for urgent assessment by a doctor or midwife and are not matters for massage. Massage does not replace prenatal care, and the sources reviewed do not support delaying care in favor of massage.