Intended for healthcare professionals


Pelvic girdle pain in pregnancy

BMJ 2005; 331 doi: (Published 28 July 2005) Cite this as: BMJ 2005;331:249
  1. R William Stones (r.w.stones{at}, senior lecturer in obstetrics and gynaecology,
  2. Kathleen Vits, clinical specialist physiotherapist
  1. University of Southampton School of Medicine, Southampton SO16 5YA
  2. Department of Obstetrics and Gynaecology, Princess Anne Hospital, Southampton

    Exercises may help, and evidence is increasing that acupuncture reduces pain

    Musculoskeletal pain in the pelvic area is common during pregnancy and can cause substantial distress and disruption of function. The lack of any standard definitions of such pain, however, makes it difficult to compare reports of prevalence, treatments, and outcomes. Useful terms for different clinical subgroups include pregnancy related pelvic girdle pain and pregnancy related low back pain.1 Authors of British review articles and case reports often use the term symphysis pubis dysfunction to describe the pain, but others consider that such dysfunction is more often a secondary problem coexisting with lumbar or sacroiliac pain.

    A systematic review of 28 studies that used the two terms pregnancy related pelvic girdle pain and pregnancy related low back pain found that prevalence ranged from 3.9% to 89.9% (mean 45.3%).1 This wide range illustrates the problems of definition, identification, and classification. The authors found that estimates of prevalence depended on the inclusion or exclusion of patients with coexisting pain higher in the back and the definition(s) of musculoskeletal pelvic pain used to select patients.

    Pelvic instability in pregnancy or the puerperium has been widely publicised in the media. This may have led to unnecessary medicalisation of pelvic musculoskeletal pain associated with normal pregnancy.2 On the other hand, many women do experience considerable impairment of function; around a third report that the pain disturbs their sleep.1 An Italian study rated functional impairment due to back pain in pregnancy using a validated questionnaire and found that six of 76 women had severe impairment.3 Intriguingly, impairment scores were higher in women carrying boys than in those carrying girls (P = 0.0007 in a multivariate model).

    Given that pregnant women experience troublesome pain, however defined, it is important to understand what might cause the pain. The main factors are probably mechanical, due to the alteration in posture required to carry the increasing mass in the abdomen, and hormonal, through changes in the pelvic ligaments. However, the hormone responsible is unclear. Although relaxin acts on human uterine tissue by regulating the expression of metalloproteinases in the matrix,4 it does not seem to generate musculoskeletal problems. A longitudinal study of 35 women assessed in the first and third trimesters found no association between changes in relaxin concentrations and either the measured laxity of the wrist joints or the onset of pelvic pain.5 Furthermore, ultrasonography shows an association between the width of the symphysis pubis and pain at that site, irrespective of serum relaxin concentrations.6 Pregnancy may compromise the inherent stability of bones and ligaments in both the spine and pelvis, requiring muscular activity to maintain stability of the associated joints.

    In clinical practice attempts to reduce pain in the lower back and pelvis in pregnant women typically include early education, advice, and exercise prescribed by a physiotherapist. Despite some agreement on definitions of pelvic instability, no pathophysiological subgroups have been identified as a basis for treatment. A practical approach to physiotherapy is to assess individuals and then treat groups of women with similar distribution of pain.7

    In the United Kingdom, recommended treatment for symphysis pubis pain includes limiting hip abduction to within the range that does not induce pain, reducing other activities that induce pain, and often suggesting the use of a support belt. By contrast, Röst, a Dutch physiotherapist, advocates on the basis of observational data a much more active approach including exercising to increase the range of hip abduction and overcoming functional limitations.8 Many pregnant women seeking specialist advice for pelvic girdle pain have already tried paracetamol or codeine and found them ineffective. However, some will appreciate advice that paracetamol and weak opioids are safe in pregnancy although non-steroidal anti-inflammatory drugs should be avoided.

    Two systematic reviews should also guide practice for pregnant women with non-specific pain in the pelvis or lower back. A Cochrane review found water gymnastics, acupuncture, and use of a specially shaped pillow for sleeping to be beneficial, and its authors said of physiotherapy “There is some measurable reduction in pain with both [acupuncture and physiotherapy], more so with acupuncture but this may be a reflection more of the personal care given by the acupuncturist compared with the group therapy from the physiotherapist.”9 The second systematic review considered nine trials but was not able to extend the conclusions of the Cochrane review because of heterogeneity in the design of the trials.10

    Elden and colleagues have now reported a controlled trial of acupuncture and stabilising exercises for women with well defined pelvic girdle pain, in which each control participant was offered advice, a pelvic belt, and muscle strengthening exercises.11 The authors analysed treatment effects for subgroups with different patterns of pain on a visual analogue scale. After treatment, pelvic pain was reduced significantly in the group who had stabilising exercises compared with controls (median difference 9 points (P = 0.0312) for pain in the morning; 13 points (P = 0.0245) in the evening), but the reduction in pain was even greater for those who had acupuncture (12 in the morning and 27 in the evening, both P < 0.001). Those caring for women with pregnancy related pelvic pain now need to press for increased availability of acupuncture, and researchers need to assess the potential benefits of combining acupuncture with stabilising exercises.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    View Abstract

    Log in

    Log in through your institution


    * For online subscription