A C TA Obstetricia et Gynecologica

AOGS RE V I EW AR TIC LE

Treatments for pregnancy-related lumbopelvic pain: a

systematic review of physiotherapy modalities
€ 4, SARA POUSETTE5 &
ANNELIE GUTKE1,2,*, CAROLA BETTEN3, KRISTINA DEGERSKAR
MONIKA FAGEVIK OLSEN  1,6

1
Department of Health and Rehabilitation, Unit of Physiotherapy, Institute of Neuroscience and Physiology, Sahlgrenska
Academy, University of Gothenburg, Gothenburg, 2Division of Physiotherapy, Department of Medical and Health Sci-
ences, Linko€ping University, Linko€ping, 3Primary Health Care, Skurup, Region Sk
ane, 4Primary Health Care, Lund, Region
 5 6
Skane, Private Clinic ‘Friskispraktiken’, Stockholm, and Department of Physical and Occupational Therapy, Sahlgrenska
University Hospital, Gothenburg, Sweden

Key words Abstract

Pelvic girdle pain, low back pain, pregnancy,
post-partum, physiotherapy Objective. To explore the effect of physiotherapeutic interventions on preg-
nancy-related lumbopelvic pain. Material and methods. Data sources: MED-
Correspondence LINE, Cochrane Central Register of Controlled Trials, PEDro, CINAHL,
Annelie Gutke, Department of Health and AMED, and SCOPUS databases were searched up to December 2014 for studies
Rehabilitation, Unit of Physiotherapy, Institute
written in English, French, German or Scandinavian languages that evaluated
of Neuroscience and Physiology, Sahlgrenska
physiotherapeutic modalities for preventing and treating pregnancy-related lum-
Academy, University of Gothenburg, Box 430,
405 30 Gothenburg, Sweden. bopelvic pain. Results. For lumbopelvic pain during pregnancy, the evidence
E-mail: [email protected] was strong for positive effects of acupuncture and pelvic belts. The evidence
was low for exercise in general and for specific stabilizing exercises. The evi-
Conflict of interest dence was very limited for efficacy of water gymnastics, progressive muscle
The authors have stated explicitly that there relaxation, a specific pelvic tilt exercise, osteopathic manual therapy, craniosa-
are no conflicts of interest in connection with
cral therapy, electrotherapy and yoga. For postpartum lumbopelvic pain, the
this article.
evidence was very limited for clinic-based treatment concepts, including spe-
Please cite this article as: Gutke A, Betten C, cific stabilizing exercises, and for self-management interventions for women
Degersk€ar K, Pousette S, Fagevik Ols en M. with severe disabilities. No specific adverse events were reported for any inter-
Treatments for pregnancy-related lumbopelvic vention. No meta-analysis could be performed because of study heterogeneity.
pain: a systematic review of physiotherapy Conclusions. The levels of evidence were strong for a positive effect of acu-
modalities. Acta Obstet Gynecol Scand 2015; puncture and pelvic belts, but weak for an effect of specific exercises. Caution
94: 1156–1168.
should prevail in choosing other interventions for pregnancy-related lumbopel-
vic pain.
Received: 12 May 2015
Accepted: 21 May 2015
Abbreviations: CCT, controlled clinical trial; LBP, low back pain; PEDro,
DOI: 10.1111/aogs.12681 Physiotherapy Evidence Database; PGP, pelvic girdle pain; RCT, randomized
controlled trial; TENS, transcutaneous electrical nerve stimulation.

Key Message
Introduction
Pregnancy-related lumbopelvic pain frequently neces-
More than 50% of women experience pain in the lumbo- sitates leave from work and is associated with a risk
pelvic area during pregnancy (1). The prevalence of pain for long-term back pain. Evidence was found for
located in the pelvis increases with the duration of preg- positive effects of acupuncture and the use of pelvic
nancy (1), but the prevalence of pain in the lumbar spine belts during pregnancy, but only weak evidence was
is the same for pregnant and non-pregnant populations found for physical exercise.
(2,3). Pelvic girdle pain (PGP) is located between the

1156 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

posterior iliac crests and the gluteal folds, and it may it. In eight previous reviews, quality assessments of pre-
radiate down the thighs and (rarely) into the lower legs. vious studies have been performed, including from three
PGP can occur with or without pain in the pubic sym- up to 26 studies, and the latest updated Cochrane
physis (1). Low back pain (LBP) is defined as pain local- review was published in 2013 (1,18–24). However,
ized below the ribs, but above the gluteal folds, with or among these reviews, two included publications only up
without radiation down the legs (4). The term lumbopel- to 2006 (1,18), one focused only on acupuncture (20),
vic pain is used when no distinction is made between another on manipulative therapy and included only one
PGP and LBP (5). Clinical experience has shown that dif- randomized controlled trial (RCT) (22), and one was
ferentiating between pain in the pelvic girdle and that in only concerned with treatment in pregnancy (24) and
the lumbar spine is important in selecting a targeted one with postpartum treatment (21). One of the identi-
treatment. A reliable classification system to make this fied recent reviews only included studies with functional
differentiation has been presented (6). outcome measures and therefore only four studies were
Previous studies have shown that among women with assessed (23), which made it difficult to apply the find-
lumbopelvic pain, 50% have been classified with PGP, ings.
17% with lumbar pain, and 33% with combined PGP Although functional outcomes are important, we per-
and LBP (2); thus, the majority of women with preg- ceived a need for a review with wider criteria, which
nancy-related lumbopelvic pain have PGP. Among would provide a broad base for drawing conclusions.
women with lumbopelvic pain, 45% reported “mild,” Moreover, several recently published RCTs, which were
30% “moderate,” and 25% “severe” pain-related problems not included in the above-mentioned reviews, have evalu-
(5). For the majority of women, PGP started during ges- ated common treatments for PGP, including exercises,
tational weeks 12–24 (7,8). The wide range in the pelvic belts, and electrotherapy. To provide recommenda-
reported prevalence (45–73%) of lumbopelvic pain tions for future studies, we aimed to identify all published
(2,9,10) has been ascribed to the different criteria used to studies on interventions used by physiotherapists for
classify types and severity of pain. pregnancy-related lumbopelvic pain; consequently, we did
The prognosis of PGP is relatively good; in one study not limit our search to RCTs. The purpose of this review
the recovery rate was 67% (11), but in another study was to explore the prevailing knowledge of physiothera-
43% reported persistent pain (7% recurrent pain and peutic interventions for women with pregnancy-related
36% constant pain) at 6 months postpartum (12). Three lumbopelvic pain, to perform a quality assessment of the
years after pregnancy, 20% of women with lumbopelvic RCTs, and to rate the levels of evidence for different
pain during pregnancy report persistent complaints (13). treatment effects.
Persistent pain after delivery is associated with a high
level of pain during pregnancy, pain in all three pelvic
Material and methods
joints, several positive pain provocation tests, and early
onset of pain (11,14). Pregnancy-related lumbopelvic pain Previously, we performed a search with the aim of defin-
negatively impacts daily functioning and well-being ing national guidelines (published on http://fysioterapeu-
(2,15). It can potentially prevent women from performing terna.se/Professionsutveckling/Riktlinjer/). For that study,
almost all daily activities (7,15). Common tasks, such as we performed a computer-aided literature search for
walking or standing, increased the pain in 80% of cases, clinical studies which evaluated treatment for pregnancy-
and the pain was triggered within 30 min (16). In gen- related lumbopelvic pain in the following databases:
eral, pregnant women experience decreased health-related Medline/PubMed, the Cochrane Central Register of Con-
quality of life, but the decrease is greater for women with trolled Trials in the Cochrane Library, the Physiotherapy
pregnancy-related lumbopelvic pain (17). Women with Evidence Database (PEDro), Cumulative Index to Nurs-
combined LBP and PGP and women with pain in all ing and Allied Health Literature (CINAHL), Allied and
three pelvic joints experienced the greatest negative complementary Medicine (AMED), and Scopus. At that
impact on health and function (2,11). time, we searched for studies published from the incep-
Non-invasive treatment is preferred to avoid medica- tion of physiotherapy for back pain up to February 2011.
tion in pregnancy, and surgery is not an option. Physio- For the present review, we included the articles identified
therapists use several treatments for pregnancy-related in that previous study, and we added a complementary
lumbopelvic pain, including passive treatments such as search up to December 2014.
mobilization, and active treatments such as exercises (1). For these searches we used the *truncation to include
It is also common to provide education about the pain all linguistic forms of the root search terms in the OVID
and discuss the benefits of a pelvic belt (1). However, databases (PubMed, EMBASE, AMED, and CINAHL) and
there is no consensus about what to use or when to use in Scopus. We also used Boolean operators “or” and

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1157
Treatments for pregnancy-related lumbopelvic pain A. Gutke et al.

“and” between inclusion criteria. Searches were limited to by the Swedish Council on Technology Assessment in
full-length articles written in English, French, German, or Health Care (http://www.sbu.se/en/). We defined the lev-
any Scandinavian language. The search terms were cus- els of evidence as strong (≥two studies of high quality
tomized for use in each of the aforementioned databases and no factors that weakened the overall assessment),
and included (for example in PubMed): pelvic girdle pain moderate (one study of high quality, ≥two studies of
OR pelvic pain OR low back pain AND pregnan* OR moderate quality, and no factors that weakened the over-
postpartum AND physiotherapy OR physical therapy OR all assessment), low (≥two studies with moderate quality,
acupuncture OR training OR exercis* OR mobilization and no factors that weakened the overall assessment), or
OR manipulation OR belt OR school. very limited (one study of at least moderate quality).

Study selection Results

The selection procedure was conducted in three phases. A total of 58 articles were identified (Figure 1). Two
The first two phases were performed by the first author, studies were excluded; of these, one evaluated a non-com-
and the third phase by the first and last authors. The mercially available pillow (26) and the other did not
phases included (i) reviewing all titles from the search, report specific results of the treatment (27). The 34 iden-
and proceeding with the relevant or uncertain; (ii) tified RCTs from the 58 articles were rated and included
reviewing the abstracts, and proceeding with the relevant in the level of evidence assessment. The remaining 22
or uncertain articles; and (iii) reviewing full-text articles articles included eight CCTs (including one conducted
to determine relevance for quality assessment. Our review postpartum), three long-term follow-up studies (Tables 1,
of the reference lists of identified articles followed the S1 and S2), two observational studies, four observational
same procedure and was performed by all the authors. retrospective studies, one experimental case study (post-
We included all studies that evaluated treatment effects of partum), one case series, and three pilot studies.
physiotherapy interventions for women with pregnancy-
related lumbopelvic pain. We also included studies that
Acupuncture
described interventions with modalities (such as acupunc-
ture) provided by physiotherapists in some countries and Acupuncture was evaluated in seven RCTs (28–34) and
modalities performed by other professionals (such as a one long-term follow-up study (35) (Table 1). The results
chiropractors) in other countries. indicated that acupuncture treatment during pregnancy
Randomized controlled trials were evaluated for inter- reduced pain and increased function (28–30,34) and the
nal validity and statistical information based on the 10 ability to work (31). One study indicated that acupunc-
individual criteria from PEDro (25), with each item ture was as effective as a combination of stabilizing train-
scored 1 for “positive” or 0 for “negative/missing.” Qual- ing, massage, and stretching for the regression of pain
ity was defined as low (scores 0–3), moderate (scores 4– postpartum (35). No serious adverse events were
6) or high (scores ≥7). According to the PEDro scale, the recorded, either for the fetus or for the woman (36,37).
risk of bias was examined across studies by analyzing The level of evidence was strong for the effect of acu-
biases of selection, performance, detection, and attrition. puncture treatment during pregnancy. Another CCT
Additionally, we analyzed biases of provider, reporting, study, not included for the level of evidence assessment,
and publication. This process ensured that the systematic indicated improved function and less pain after acupunc-
review was conducted according to the PRISMA protocol. ture (38).
Although we decided not to include controlled clinical
trials (CCT) in the systematic review, we scored the CCTs
Pelvic belt/girdle
with the PEDro scale, and present the results separately.
Each author independently assessed each trial for a The use of a pelvic belt during pregnancy was examined
quality score. When disagreement occurred, a consensus in three RCTs (39–41) (Table S2). For women with PGP,
was reached by group discussion. The studies were using the belt significantly decreased the pain intensity
assessed for the following treatments: acupuncture, pelvic and disability compared with exercises and compared
belt/girdle, physical exercises performed in groups or with education with only general information (41). For
individually (during pregnancy, after pregnancy, in women with lumbopelvic pain, two different non-rigid
water), manual therapies, electrotherapy, yoga, patient belts significantly reduced pain intensity. Although there
education, and self-management interventions. was no significant difference in the improvement of pain
The level of evidence was based on the RCTs only, and intensity between the two belts, one tended to improve
it was determined according to the information provided function more than the other (40). For women with

1158 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

Records identified through database Additional records identified through

searches reference lists
(n = 602) (n = 6)

Records after duplicates were

removed
(n = 230)

Records screened Records excluded

(n = 230) (n = 172)

Full-text articles assessed Full-text articles excluded,

for eligibility due to evaluation of a non
(n = 58) available method (n = 1)
specific results of
treatment missing (n = 1)

Studies included in
qualitative synthesis
(n = 56)

Studies included in
quantitative synthesis
(n = 34)

Figure 1. Flow chart of the systematic search process.

symphyseal pain, the average pain intensity was reduced performed both at home and in a group. In many stud-
in the group that received exercise-only and in the group ies, exercises were combined with other physiotherapy
that used a rigid belt; however, addition of a belt to exer- modalities, such as manual therapies, pelvic belt, and
cises had no apparent effect (39). The level of evidence information. Mostly, those studies reported positive
was strong for the effect of a pelvic belt. Another non- effects on pain and disability compared with a control
RCT pilot study, not included in the evidence shown, group or to standard care alone (30,44,47,50), except one
indicated that women who used a belt had significantly study (51). However, when the standard care included
higher activity levels than women in a control group home exercises combined with patient education, pelvic
(42). belt, hot packs, and massage, the intervention group who
received acupuncture had lower pain intensity and their
activity levels increased (34). Comparing different combi-
Group and/or individual physical exercises
nations of physiotherapy modalities, all groups showed
The effects of exercise during and after pregnancy were improvements (46).
evaluated in 19 RCTs (30,34,41,43–58) and one long-term Only providing information in a group appeared to be
follow-up study (35) (Table S1). Fifteen RCTs (30,34,41, insufficient for alleviating pain (41), but doing something
43–51,53,54,56) evaluated the effects of exercises during was typically more successful than doing nothing
pregnancy; of these, five RCTs investigated exercises per- (43,47,49,54). The identified long-term follow-up showed
formed in groups, nine RCTs investigated exercises per- no difference between groups (35). No major side-effects
formed individually, and one RCT included exercises of stabilizing exercises were reported, although three

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1159
Treatments for pregnancy-related lumbopelvic pain A. Gutke et al.

Table 1. Randomized controlled trials on acupuncture included.

Author, publ. year, PEDro Study Number Outcome Significant results in

(reference no) score design of women Diagnosis When Treatment measurements RCT or CCT studies

Ekdal, 2010 (33) 2 RCT 40 LBP and P Acupuncture from Pain-O-Meter, Less pain with
PGP gestational week 20 or Pain (VAS), treatment later in
26, 8 treatments for Short form pregnancy
30 min McGill
questionnaire,
HRQL (SF-36)
Elden, part 1, 8 RCT 386 PGP P Standard treatment Pain (VAS), Less pain after
2005 (30) (advice, information, assessment of stabilizing exercises
home exercise program severity or acupuncture
and pelvic belt) alone or (examination) compared with
plus 12 treatments for standard treatment
30 min acupuncture or alone. Less pain
plus 6 h supervised after acupuncture
stabilizing exercises vs. stabilizing
exercises
Elden, part 2, 6 RCT 328 PGP PP See above, Elden part 1 Pain (VAS), Fewer positive
2008 (35) Follow-up postpartum assessment of provocation tests
severity after stabilizing
(examination) exercises than the
controls
Elden, part 3, 8 RCT 383 PGP, P See above Elden part 1, Adverse events No significant
2008 (36) Adverse effects differences between
the groups
Elden, 2008 (31) 8 RCT 115 PGP P Acupuncture vs. non- HRQL (EQ-5D), Lower sick leave and
penetrating sham Pain (VAS), disability scores
acupuncture both 12 disability (ODI), after penetrating
treatments for 30 min sick-leave, acupuncture vs.
twice/week for 4 weeks assessment of sham acupuncture
and once/week for severity
4 weeks (examination)
Guerreiro da Silva, 6 CCT 71 LBP P Acupuncture (8–12 Pain (scales), Less pain and less
2004 (38) PGP treatments for 25 min) Functional pain medication
vs. conventional capacity (scales) after acupuncture.
treatment with Increased capacity
paracetamol and to perform general
hyoscine activities, to walk
and to work after
acupuncture
Kvorning, 3 RCT 72 LBP and P Acupuncture twice/week Pain (VAS), Less pain and less
2004 (28) PGP during 2 weeks, influence of pain impact of pain
thereafter once/week in on physical during activity after
gestational weeks 24– activity (scales), treatment with
37 vs. control group delivery data, acupuncture
adverse events
Lund, 2006 (32) 6 RCT 70 LBP and P Superficial or deep Pain (VAS), health No significant
PGP acupuncture, each 10 (NHP) differences between
treatments for 30 min the groups
Wang, 2009 (29) 7 RCT 159 LBP and P Ear-acupuncture, Sham Pain (VAS), Less pain and
PGP ear-acupuncture or disability DRI improved functional
control group. ability in
Treatment for 1 week acupuncture vs.
sham treatment or
control group

1160 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

Table 1. Continued

Author, publ. year, PEDro Study Number Outcome Significant results in

(reference no) score design of women Diagnosis When Treatment measurements RCT or CCT studies

Wedenberg, 4 RCT 60 LBP and P Acupuncture for 30 min Pain (VAS), Less pain after
2000 (34) PGP three times/week disability (DRI), acupuncture
during 2 weeks, overall effect compared to
thereafter twice/week and adverse physiotherapy
for 2 weeks vs. events
physiotherapy
(information, pelvic
belt, training, massage)
one to two times/week,
50 min each, totally 10
treatments in 6–
8 weeks

CCT, controlled clinical trial; DRI, disability rating index; EQ-5D, Euroqol 5Dimensions; HRQL, health-related quality of life; LBP, low back pain;
NHP, Nottingham health profile; ODI, Oswestry disability index; P, pregnancy; PEDro, Physiotherapy Evidence Database; PGP, pelvic girdle pain;
PP, postpartum; RCT, randomized controlled trial; VAS, visual analogue scale.

women experienced increased pain intensity during exer- Five CCTs that evaluated exercises were not included
cises (36). in the level of evidence assessment. Three CCTs indicated
In three RCTs the focus was on preventive training improvements in sick leave and usefulness (59), pain
(45,48,53). Of these, one study focused on pelvic floor (60), and interference with daily activities (61). Two
exercises; it was found that exercise reduced the incidence CCTs reported no significant difference between groups
of lumbopelvic pain in late pregnancy and increased (62,63). Other non-RCT studies were found that were not
physical function, but did not change the rate of sick included in the level of evidence assessment. One was an
leave compared with controls (53). In another study, a experimental case study that showed a trend (n = 10
12-week exercise program, which included home exercises women) for stabilizing exercises to reduce pain intensity
and weekly group sessions, was compared with standard and ameliorate limitations in activity levels (64), one ret-
antenatal care. In the exercise group, a lower proportion rospective observational study reported less severe back
of individuals claimed sick leave due to lumbopelvic pain pain after back-care classes (65) and one article reported
than those in the control group, but the groups were not the side-effects of treatments (36).
different in the prevalence of lumbopelvic pain (45).
Another study compared lumbopelvic pain in groups that
Water gymnastics
performed exercises either in water or on land; the
authors showed that exercise in water was associated with Water gymnastics was evaluated in two RCTs (48,56). Both
lower pain intensity and less sick leave than exercise on studies reported reduced pain intensity and reduced sick
land (48). leave after water gymnastics. One study (48) only showed
Based on the above-described results from exercise significant effects in the group with lumbar pain, not in
studies, the majority of studies showed a positive effect women with PGP. Therefore, the level of evidence was very
on lumbopelvic pain during pregnancy. However, there limited in support of the effects of water gymnastics.
was a low level of evidence supporting the efficacy of any
specific intervention or type of exercise because only a
Exercise during and after pregnancy
few studies were conducted on each type of exercise.
Moreover, the effects differed among studies on specific No RCT was found that tested the effect of exercise dur-
outcome variables. Thus, the level of evidence was low in ing and after pregnancy and thus no study on this was
support of the notion that exercise, in general, can reduce included. One CCT evaluated the impact of patient edu-
pain intensity during pregnancy. More specifically, the cation and home exercises after video instruction (63) in
levels of evidence were low for supporting specific stabi- late pregnancy, continued until 8 weeks after birth. The
lizing exercises in reducing pain intensity during preg- results showed that the intervention had no effect on
nancy; very limited for supporting a specific pelvic tilt in back pain.
reducing pain intensity; and no evidence supported the Three RCTs evaluated individual exercises for post-
efficacy of preventive exercises. partum PGP (52,55,57), and one study exercises for

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1161
Treatments for pregnancy-related lumbopelvic pain A. Gutke et al.

postpartum back pain (58). No significant between-group passive and active lumbar spine mobilizations (71) and the
differences were observed between a control group and other results after a diagnosis-based clinical assessment
two types of exercises; videotape-instructed stabilizing schedule (72). Furthermore, two retrospective studies
exercises of the diagonal trunk muscles, and exercises for (73,74), a pilot RCT (70), and a case series (75) reported
the longitudinal trunk muscles (52); however, 25% of the results of lumbar and sacroiliac joint mobilization.
women in the study stopped exercising due to increased
pain intensity during exercise. An individual treatment
Electrotherapy
concept, which included stabilizing exercises, significantly
decreased pain intensity and disability, and increased Transcutaneous electrical nerve stimulation (TENS) was
health-related quality of life compared with conventional examined in an RCT study (43) (Table S2). The study
treatment (57); that effect had persisted at the 2-year fol- showed a significantly greater decrease in pain and
low-up (66). A home-based program of specific stabiliz- increase in function in the TENS group compared with
ing exercises decreased pain intensity but not more than the exercise and acetaminophen groups. However, since
that achieved in the natural clinical course (55). Core sta- there was only one RCT on TENS, the evidence for the
bilizing exercises and postural correction initiated at the efficacy of electrotherapy was very limited.
maternity ward reduced back pain better than strengthen-
ing exercises (58). The level of evidence was very limited
Yoga
in support of a postpartum, individualized program that
included specific stabilizing exercises. There was no evi- One RCT study (76) evaluated yoga therapy (Table S2).
dence to support the home-based program of specific sta- The study showed that yoga decreased pain compared
bilizing exercises, core stabilizing exercises with postural with posture orientation. The level of evidence for the
correction, or videotape-instructed, diagonal or longitudi- efficacy of yoga was very limited.
nal, trunk muscle exercises.
Patient education
Manual therapies
An RCT study by Haugland et al. (77) (Table S2) evalu-
One RCT evaluated osteopathic manual therapy (soft tis- ated patient education. They demonstrated no differences
sue, myofascial release, range-of-motion, muscle energy) between treatment and control groups. No evidence was
(67) (Table S2). Manual therapies significantly decreased found for the efficacy of patient education in combina-
pain intensity and disability compared with a general tion with exercise. Three non-RCT studies, not included
treatment with or without sham-ultrasound (67). The in the level of evidence, were CCTs on patient education
level of evidence was very limited for the efficacy of oste- (8,60,78). These studies dealt with multimodal approaches
opathic manual therapy in pregnancy. which included education and reported that women expe-
Craniosacral therapy was evaluated in one RCT study rienced less discomfort (78), decreased pain intensity
that showed less pain intensity in the morning and less (60), and decreased sick leave (8) compared with control
deterioration of function after craniosacral therapy in women. We also found a pilot study which reported that
conjunction with standard treatment compared with stan- pain improved with prenatal education and exercises for
dard treatment alone (68) (Table S2). However, the clini- preventing lumbopelvic pain (79).
cal relevance of the study result was questionable and
therefore evidence for this treatment was very limited.
Self-management intervention
The result of another RCT that examined the effects of
massage in preventing lumbopelvic pain, showed no Self-management intervention was assessed in an RCT
between-group difference (69) (Table S2). Thus, no evi- which focused on the patient–therapist relationship, educa-
dence has shown that massage can prevent pregnancy- tion, and encouraging an active lifestyle (80) (Table S2).
related lumbopelvic pain. This intervention appeared to provide a small improve-
No RCT was identified that tested the effects of mobili- ment in function measured at 12 weeks postpartum for
zation. Therefore, no study was included concerning the women with severe symptoms [defined as a rating ≥13
level of evidence for mobilization. However, other studies points on the Roland Morris Disability Questionnaire
on mobilization were identified from which we can learn (RMDQ)] (80). The authors stated that changes over time
for future studies. A pilot study compared spinal manipu- were too small to be clinically relevant. Therefore, the
lation, a neuro-emotional technique, and exercise with evidence was very limited supporting the short-term
improvement of function in all three groups (70). In two efficacy. There was no evidence for the long-term effective-
observational studies (71,72), one reported the effects of ness (81).

1162 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

that exercises were effective in preventing lumbopelvic

Progressive muscle relaxation
pain; however, studies showed contradictory results in
Progressive muscle relaxation exercises were evaluated in specific outcomes. This discrepancy gave rise to the risk
an RCT study (82) (Table S2). The study showed signifi- of detection bias and other difficulties that might impede
cant decrease in pain and increased health-related quality drawing a firm conclusion. Furthermore, there was very
of life in the intervention group compared with the con- limited evidence supporting postpartum exercises as an
trols. However, since there was only one RCT on progres- individual treatment concept, including specific stabilizing
sive muscle relaxation, the level of evidence for its exercises. The levels of evidence were also very limited for
efficacy was very limited. manual therapies, such as osteopathic manual therapy
Women from all over the world were represented in the (67) or cranio-sacral therapy (68), electrotherapy (43),
reviewed studies, but the majority of the data were col- progressive muscle relaxation (82), and yoga (76) due to
lected from European populations (Table S3). Most stud- the small number of RCTs.
ies did not report ethnicity. Tables 1, S1 and S2 reveal the A limitation of the study was that gestational age and
large variations in the treatments provided. Likewise, a parity varied between study participants. This may have
wide range of outcome measures were used, even for the increased the risk of selection bias because it is known
same outcome variables, such as pain and disability. that pain becomes worse with multiparity and with pro-
gression of pregnancy. Also, because most studies did not
report ethnicity we could not determine whether the
Discussion
results might be more or less relevant for any specific
Many women experience LBP and/or PGP during preg- populations. Only one reviewer initially selected studies
nancy of sufficient severity to interfere with daily living, in the first two phases of the search, which may have
reduce the ability to be active, and limit work perfor- introduced selection bias. However, all authors reviewed
mance. At the societal level, a large part of the sick leave all the reference lists of the identified studies. Further-
cost during pregnancy is due to back pain (83). Because more, our study may be limited by the heterogeneity with
the societal burden of long-term pain is high (84), inter- regard to the studied disorders. Some studies only
ventions against pregnancy-related lumbopelvic pain are included participants with PGP (30,31,41,46,52,57,68,77),
needed to prevent or limit persistent pain during and others included participants with lumbopelvic pain
after pregnancy. This review is complementary to previ- (8,28,32–34,38,40,48,53,55,60), and some included only
ous recommendations, because we included more treat- women with symphysis pubis pain (39). Some studies
ment modalities, as well as additional RCTs (1,18–24). used the term LBP (49,54,63,67,71) but defined it as lum-
Acupuncture and pelvic belt treatments were supported bopelvic pain. This suggested that the results should be
by a strong level of evidence. The seven RCTs that assessed generalized to women with pain in the lumbopelvic area.
acupuncture were homogeneous; each RCT evaluated acu- There was also a risk of performance bias because the
puncture that targeted the pain location area (one study studies were conducted at different time points (ranging
used ear-acupuncture) and applied a similar dose (eight to from early pregnancy to postpartum), for different dura-
12 sessions in 6/7 trials). Different types of pelvic belts tions (1–20 weeks), and with different intervention fre-
reduced pain intensity for women with lumbopelvic pain quencies (once, twice, three times per week or daily).
(40) but only a rigid belt reduced pain for women with Moreover, treatments were often combined, which made
symphyseal pain (39). For women with PGP, a non-rigid it difficult to determine which component was important
belt was effective over the short term; this suggested that for the outcome. Confounding factors, for example in
the pelvic belt may be a first treatment choice to stabilize group interventions, were seldom discussed. In addition,
the pelvis before exercise treatment has taken effect (41). there may be differences in how the treatments were per-
Exercise therapy is a common intervention, and accu- formed in different studies, by individuals with different
mulated evidence has suggested that exercises can reduce professional backgrounds. The treatment providers were
pain intensity and improve function. However, it is diffi- mostly a physiotherapist or a midwife.
cult to recommend which type of exercise may be the The large majority of studies reported outcomes in
most effective because the exercise types varied widely pain intensity and function; however, because different
between studies. Exercises included individual or group measurements were used, there was a risk of detection
programs, conducted in water or on land, and located at bias. Currently, no consensus exists for outcome measures
home or in a clinic. The results did not conclusively in studies of lumbopelvic pain. However, a questionnaire
indicate which type of exercise should be recommended, was recently published that was developed for patients
or whether some exercises were more appropriate for a with PGP (85). Only two studies (68,81) discussed the
particular group of women. Several studies have shown minimal clinically important difference between groups;

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1163
Treatments for pregnancy-related lumbopelvic pain A. Gutke et al.

thus, there was little guidance for judging the relevance of

Funding
a specific modality.
There may have been attrition bias. Almost all studies No financial or material support was received for this
reported attrition but they did not discuss its influence study.
on the results. A substantial attrition can limit the possi-
bility of drawing conclusions because the power of the
References
study may be lost. Likewise, there was a risk of dilution
bias, but in most studies, this was neither controlled for 1. Vleeming A, Albert HB, Ostgaard HC, Sturesson B, Stuge
nor discussed. There was also a risk of reporting bias, B. European guidelines for the diagnosis and treatment of
because we only included studies published in English, pelvic girdle pain. Eur Spine J. 2008;17:794–819.
French, German, and Scandinavian languages. Various 2. Gutke A, Ostgaard HC, Oberg B. Pelvic girdle pain and
terms were used for pregnancy-related lumbopelvic pain, lumbar pain in pregnancy: a cohort study of the
but there may have been a bias in choosing the search consequences in terms of health and functioning. Spine
terms. We considered the risk of publication bias to be (Phila Pa 1976). 2006;31:E149–55.
low because, although the reviewed studies were mainly 3. Gutke A, Lundberg M, Ostgaard HC, Oberg B. Impact
of postpartum lumbopelvic pain on disability, pain
published in obstetrics and musculoskeletal scientific
intensity, health-related quality of life, activity level,
journals, the searches were conducted in databases that
kinesiophobia, and depressive symptoms. Eur Spine J.
represented all types of journals.
2010;20:440–8.
There are several lists of criteria for assessing methodo-
4. van Tulder M, Becker A, Bekkering T, Breen A, del Real
logical quality in research studies. The PEDro scale which MT, Hutchinson A, et al. Chapter 3. European guidelines
we used, included only internal validity criteria that refer for the management of acute nonspecific low back pain in
to characteristics of the study that might be related to primary care. Eur Spine J. 2006;15(Suppl 2):S169–91.
selection, performance, detection, and attrition biases 5. Wu WH, Meijer OG, Uegaki K, Mens JM, van Dieen JH,
(86). Many scales are designed to evaluate the effects of Wuisman PI, et al. Pregnancy-related pelvic girdle pain
drugs in double-blind, fully randomized, controlled fash- (PPP). I: terminology, clinical presentation, and
ions. However, in clinical trials on physiotherapy, it is prevalence. Eur Spine J. 2004;13:575–89.
nearly impossible to blind the provider and participant; 6. Gutke A, Kjellby-Wendt G, Oberg B. The inter-rater
thus, the full score on the scale cannot be achieved. reliability of a standardised classification system for
Our primary purpose in the present study was to grade pregnancy-related lumbopelvic pain. Man Ther.
the level of evidence based on RCTs and to summarize 2009;15:13–8.
the prevailing knowledge on interventions for pregnancy- 7. Mens JM, Vleeming A, Stoeckart R, Stam HJ, Snijders CJ.
related lumbopelvic pain. In addition, we aimed to iden- Understanding peripartum pelvic pain. Implications of a
tify needs for future studies. Our search results suggested patient survey. Spine (Phila Pa 1976). 1996;21:1363–9;
that more homogeneous studies of high quality are discussion 9–70.
required for all the interventions included in this review. 8. Noren L, Ostgaard S, Nielsen TF, Ostgaard HC.
In particular, only one RCT was identified for each of the Reduction of sick leave for lumbar back and posterior
different manual therapy interventions; thus, the evidence pelvic pain in pregnancy. Spine (Phila Pa 1976).
was very limited, which suggested an urgent need for 1997;22:2157–60.
9. Robinson HS, Eskild A, Heiberg E, Eberhard-Gran M.
more studies. We found no RCT where manipulation and
Pelvic girdle pain in pregnancy: the impact on function.
mobilization were evaluated, though we did find a pilot
Acta Obstet Gynecol Scand. 2006;85:160–4.
study (70), two observational studies (71,72), some retro-
10. Mogren IM, Pohjanen AI. Low back pain and pelvic pain
spective studies (70,73,74), and a case series study (75)
during pregnancy: prevalence and risk factors. Spine (Phila
on mobilization therapies that reflected clinical practice. Pa 1976). 2005;30:983–91.
The results from those pilot studies, and taking into 11. Albert H, Godskesen M, Westergaard J. Prognosis in four
account the limitations of the study designs, only provide syndromes of pregnancy-related pelvic pain. Acta Obstet
a basis for planning studies on manual therapies in the Gynecol Scand. 2001;80:505–10.
field and cannot be used as a basis for clinical advice. 12. Mogren IM. BMI, pain and hyper-mobility are
Future studies would benefit from including homoge- determinants of long-term outcome for women with low
neous groups and using a single set of criteria for back pain and pelvic pain during pregnancy. Eur Spine J.
selecting study populations. Similarly, a single set of out- 2006;15:1093–102.
come measures for well described therapeutic modalities 13. Noren L, Ostgaard S, Johansson G, Ostgaard HC. Lumbar
should be used to compare the effectiveness of different back and posterior pelvic pain during pregnancy: a 3-year
interventions. follow-up. Eur Spine J. 2002;11:267–71.

1164 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

14. Ostgaard HC, Roos-Hansson E, Zetherstrom G. Regression 29. Wang SM, Dezinno P, Lin EC, Lin H, Yue JJ, Berman MR,
of back and posterior pelvic pain after pregnancy. Spine et al. Auricular acupuncture as a treatment for pregnant
(Phila Pa 1976). 1996;21:2777–80. women who have low back and posterior pelvic pain: a
15. Kristiansson P, Svardsudd K, von Schoultz B. Back pain pilot study. Am J Obstet Gynecol. 2009;201:e1–9.
during pregnancy: a prospective study. Spine (Phila Pa 30. Elden H, Ladfors L, Olsen MF, Ostgaard HC, Hagberg H.
1976). 1996;21:702–9. Effects of acupuncture and stabilising exercises as adjunct
16. Rost CC, Jacqueline J, Kaiser A, Verhagen AP, Koes BW. to standard treatment in pregnant women with pelvic
Pelvic pain during pregnancy: a descriptive study of signs girdle pain: randomised single blind controlled trial. BMJ.
and symptoms of 870 patients in primary care. Spine 2005;330:761.
(Phila Pa 1976). 2004;29:2567–72. 31. Elden H, Fagevik-Olsen M, Ostgaard HC, Stener-Victorin
17. Olsson C, Nilsson-Wikmar L. Health-related quality of life E, Hagberg H. Acupuncture as an adjunct to standard
and physical ability among pregnant women with and treatment for pelvic girdle pain in pregnant women:
without back pain in late pregnancy. Acta Obstet Gynecol randomised double-blinded controlled trial comparing
Scand. 2004;83:351–7. acupuncture with non-penetrating sham acupuncture.
18. Stuge B, Hilde G, Vollestad N. Physical therapy for BJOG. 2008;115:1655–68.
pregnancy-related low back and pelvic pain: a systematic 32. Lund I, Lundeberg T, Lonnberg L, Svensson E. Decrease of
review. Acta Obstet Gynecol Scand. 2003;82:983–90. pregnant women’s pelvic pain after acupuncture: a
19. Pennick VE, Liddle SD. Interventions for preventing and randomized controlled single-blind study. Acta Obstet
treating pelvic and back pain in pregnancy. Cochrane Gynecol Scand. 2006;85:12–9.
Database Syst Rev. 2013;(8):CD001139. 33. Ekdahl L, Petersson K. Acupuncture treatment of pregnant
20. Ee CC, Manheimer E, Pirotta MV, White AR. women with low back and pelvic pain – an intervention
Acupuncture for pelvic and back pain in pregnancy: a study. Scand J Caring Sci. 2010;24:175–82.
systematic review. Am J Obstet Gynecol. 2008;198:254–9. 34. Wedenberg K, Moen B, Norling A. A prospective
21. Ferreira CW, Alburquerque-Sendi NF. Effectiveness of randomized study comparing acupuncture with
physical therapy for pregnancy-related low back and/or physiotherapy for low-back and pelvic pain in pregnancy.
pelvic pain after delivery: a systematic review. Physiother Acta Obstet Gynecol Scand. 2000;79:331–5.
Theory Pract. 2013;29:419–31. 35. Elden H, Hagberg H, Olsen MF, Ladfors L, Ostgaard HC.
22. Khorsan R, Hawk C, Lisi AJ, Kizhakkeveettil A. Regression of pelvic girdle pain after delivery: follow-up of
Manipulative therapy for pregnancy and related a randomised single blind controlled trial with different
conditions: a systematic review. Obstet Gynecol Surv. treatment modalities. Acta Obstet Gynecol Scand.
2009;64:416–27. 2008;87:201–8.
23. Richards E, van Kessel G, Virgara R, Harris P. Does 36. Elden H, Ostgaard HC, Fagevik-Olsen M, Ladfors L,
antenatal physical therapy for pregnant women with low Hagberg H. Treatments of pelvic girdle pain in pregnant
back pain or pelvic pain improve functional outcomes? A women: adverse effects of standard treatment, acupuncture
systematic review. Acta Obstet Gynecol Scand. and stabilising exercises on the pregnancy, mother, delivery
2012;91:1038–45. and the fetus/neonate. BMC Complement Altern Med.
24. van Benten E, Pool J, Mens J, Pool-Goudzwaard A. 2008;8:34.
Recommendations for physical therapists on the treatment 37. Ternov NK, Grennert L, Aberg A, Algotsson L, Akeson J.
of lumbopelvic pain during pregnancy: a systematic Acupuncture for lower back and pelvic pain in late
review. J Orthop Sports Phys Ther. 2014;44:464–73, A1– pregnancy: a retrospective report on 167 consecutive cases.
15. Pain Med. 2001;2:204–7.
25. PEDro. PEDro physiotherapy evidence database. Sydney: 38. Guerreiro da Silva JB, Nakamura MU, Cordeiro JA, Kulay
the George Institute. 1999. Available online at: http:// L Jr. Acupuncture for low back pain in pregnancy – a
www.pedro.org.au/english/downloads/pedro-scale/ prospective, quasi-randomised, controlled study. Acupunct
(accessed December 23, 2014). Med. 2004;22:60–7.
26. Thomas IL, Nicklin J, Pollock H, Faulkner K. Evaluation 39. Depledge J, McNair PJ, Keal-Smith C, Williams M.
of a maternity cushion (Ozzlo pillow) for backache and Management of symphysis pubis dysfunction during
insomnia in late pregnancy. Aust N Z J Obstet Gynaecol. pregnancy using exercise and pelvic support belts. Phys
1989;29:133–8. Ther. 2005;85:1290–300.
27. Berg G, Hammar M, M€ oller-Nielsen J, Linden U, Thorblad 40. Kalus SM, Kornman LH, Quinlivan JA. Managing back
J. Low back pain during pregnancy. Obstet Gynecol. pain in pregnancy using a support garment: a randomised
1988;71:71–5. trial. BJOG. 2007;115:68–75.
28. Kvorning N, Holmberg C, Grennert L, Aberg A, Akeson J. 41. Kordi R, Abolhasani M, Rostami M, Hantoushzadeh S,
Acupuncture relieves pelvic and low-back pain in late Mansournia MA, Vasheghani-Farahani F. Comparison
pregnancy. Acta Obstet Gynecol Scand. 2004;83:246–50. between the effect of lumbopelvic belt and home based

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1165
Treatments for pregnancy-related lumbopelvic pain A. Gutke et al.

pelvic stabilizing exercise on pregnant women with pelvic pregnancy: a randomized, controlled clinical trial. J
girdle pain; a randomized controlled trial. J Back Rehabil Med. 2010;42:929–35.
Musculoskelet Rehabil. 2013;26:133–9. 56. Kihlstrand M, Stenman B, Nilsson S, Axelsson O. Water-
42. Carr CA. Use of a maternity support binder for relief of gymnastics reduced the intensity of back/low back pain in
pregnancy-related back pain. J Obstet Gynecol Neonatal pregnant women. Acta Obstet Gynecol Scand.
Nurs. 2003;32:495–502. 1999;78:180–5.
43. Keskin EA, Onur O, Keskin HL, Gumus II, Kafali H, 57. Stuge B, Laerum E, Kirkesola G, Vollestad N. The efficacy
Turhan N. Transcutaneous electrical nerve stimulation of a treatment program focusing on specific stabilizing
improves low back pain during pregnancy. Gynecol Obstet exercises for pelvic girdle pain after pregnancy: a
Invest. 2012;74:76–83. randomized controlled trial. Spine (Phila Pa 1976).
44. George JW, Skaggs CD, Thompson PA, Nelson DM, 2004;29:351–9.
Gavard JA, Gross GA. A randomized controlled trial 58. Chaudry S, Rashid F, Shah SIH. Effectiveness of core
comparing a multimodal intervention and standard stabilization exercises along with postural correction in
obstetrics care for low back and pelvic pain in pregnancy. postpartum back pain. Rawal Med J. 2013;38:256–9.
Am J Obstet Gynecol. 2013;208:e1–7. 59. Ostgaard HC, Zetherstrom G, Roos-Hansson E, Svanberg
45. Stafne SN, Salvesen KA, Romundstad PR, Stuge B, B. Reduction of back and posterior pelvic pain in
Morkved S. Does regular exercise during pregnancy pregnancy. Spine (Phila Pa 1976). 1994;19:894–900.
influence lumbopelvic pain? A randomized controlled trial. 60. Shim MJ, Lee YS, Oh HE, Kim JS. Effects of a back-pain-
Acta Obstet Gynecol Scand. 2012;91:552–9. reducing program during pregnancy for Korean women: a
46. Nilsson-Wikmar L, Holm K, Oijerstedt R, Harms-Ringdahl non-equivalent control-group pretest-posttest study. Int J
K. Effect of three different physical therapy treatments on Nurs Stud. 2007;44:19–28.
pain and activity in pregnant women with pelvic girdle pain: 61. Yan CF, Hung YC, Gau ML, Lin KC. Effects of a stability
a randomized clinical trial with 3, 6, and 12 months follow- ball exercise programme on low back pain and daily life
up postpartum. Spine (Phila Pa 1976). 2005;30:850–6. interference during pregnancy. Midwifery. 2014;30:412–9.
47. Garshasbi A, Faghih Zadeh S. The effect of exercise on the 62. Dumas GA, Reid JG, Wolfe LA, Griffin MP, McGrath MJ.
intensity of low back pain in pregnant women. Int J Exercise, posture, and back pain during pregnancy. Clin
Gynaecol Obstet. 2005;88:271–5. Biomech (Bristol, Avon). 1995;10:104–9.
48. Granath AB, Hellgren MS, Gunnarsson RK. Water aerobics 63. Oh HE, Lee YS, Shim MJ, Kim JS. Effects of a postpartum
reduces sick leave due to low back pain during pregnancy. back pain relief program for Korean women. Taehan
J Obstet Gynecol Neonatal Nurs. 2006;35:465–71. Kanho Hakhoe Chi. 2007;37:163–70.
49. Kashanian M, Akbari Z, Alizadeh MH. The effect of 64. Gustafsson J, Nilsson-Wikmar L. Influence of specific
exercise on back pain and lordosis in pregnant women. Int muscle training on pain, activity limitation and
J Gynaecol Obstet. 2009;107:160–1. kinesiophobia in women with back pain post-partum – a
50. Kluge J, Hall D, Louw Q, Theron G, Grove D. Specific ‘single-subject research design’. Physiother Res Int.
exercises to treat pregnancy-related low back pain in a 2008;13:18–30.
South African population. Int J Gynaecol Obstet. 65. Mantle MJ, Greenwood RM, Currey HL. Backache in
2011;113:187–91. pregnancy. Rheumatol Rehabil. 1977;16:95–101.
51. Eggen MH, Stuge B, Mowinckel P, Jensen KS, Hagen KB. 66. Stuge B, Veierod MB, Laerum E, Vollestad N. The efficacy
can supervised group exercises including ergonomic advice of a treatment program focusing on specific stabilizing
reduce the prevalence and severity of low back pain and exercises for pelvic girdle pain after pregnancy: a two-year
pelvic girdle pain in pregnancy? A randomized controlled follow-up of a randomized clinical trial. Spine (Phila Pa
trial. Phys Ther. 2012;92:781–90. 1976). 2004;29:E197–203.
52. Mens JM, Snijders CJ, Stam HJ. Diagonal trunk muscle 67. Licciardone JC, Buchanan S, Hensel KL, King HH, Fulda
exercises in peripartum pelvic pain: a randomized clinical KG, Stoll ST. Osteopathic manipulative treatment of back
trial. Phys Ther. 2000;80:1164–73. pain and related symptoms during pregnancy: a
53. Morkved S, Salvesen KA, Schei B, Lydersen S, Bo K. Does randomized controlled trial. Am J Obstet Gynecol.
group training during pregnancy prevent lumbopelvic 2010;202:43.e1–8.
pain? A randomized clinical trial. Acta Obstet Gynecol 68. Elden H, Ostgaard HC, Glantz A, Marciniak P, Linner
Scand. 2007;86:276–82. AC, Olsen MF. Effects of craniosacral therapy as adjunct
54. Suputtitada A, Wacharapreechanont T, Chaisayan P. Effect to standard treatment for pelvic girdle pain in pregnant
of the “sitting pelvic tilt exercise” during the third women: a multicenter, single blind,
trimester in primigravidas on back pain. J Med Assoc randomized controlled trial. Acta Obstet Gynecol Scand.
Thai. 2002;85(Suppl 1):S170–9. 2013;92:775–82.
55. Gutke A, Sjodahl J, Oberg B. Specific muscle stabilizing as 69. Field T, Hernandez-Reif M, Hart S, Theakston H,
home exercises for persistent pelvic girdle pain after Schanberg S, Kuhn C. Pregnant women benefit from

1166 ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167
A. Gutke et al. Treatments for pregnancy-related lumbopelvic pain

massage therapy. J Psychosom Obstet Gynaecol. intervention for pregnancy-related pelvic girdle and/or low
1999;20:31–8. back pain after delivery: short-term results of a
70. Peterson CD, Haas M, Gregory WT. A pilot randomized randomized clinical trial [ISRCTN08477490]. BMC
controlled trial comparing the efficacy of exercise, spinal Musculoskelet Disord. 2006;7:19.
manipulation, and neuro emotional technique for the 81. Bastiaenen CH, de Bie RA, Vlaeyen JW, Goossens ME,
treatment of pregnancy-related low back pain. Chiropr Leffers P, Wolters PM, et al. Long-term effectiveness and
Man Therap. 2012;20:18. costs of a brief self-management intervention in women
71. McIntyre IN, Broadhurst NA. Effective treatment of low with pregnancy-related low back pain after delivery. BMC
back pain in pregnancy. Aust Fam Physician. 1996;25(9 Pregnancy Childbirth. 2008;8:19.
Suppl 2):S65–7. 82. Akmese ZB, Oran NT. Effects of progressive muscle
72. Murphy DR, Hurwitz EL, McGovern EE. Outcome of relaxation exercises accompanied by music on low back
pregnancy-related lumbopelvic pain treated according to a pain and quality of life during pregnancy. J Midwifery
diagnosis-based decision rule: a prospective observational Womens Health. 2014;59:503–9.
cohort study. J Manipulative Physiol Ther. 2009;32:616–24. 83. Sydsjo A, Sydsjo G, Kjessler B. Sick leave and social
73. Daly JM, Frame PS, Rapoza PA. Sacroiliac subluxation: a benefits during pregnancy – a Swedish-Norwegian
common, treatable cause of low-back pain in pregnancy. comparison. Acta Obstet Gynecol Scand. 1997;76:748–54.
Fam Pract Res J. 1991;11:149–59. 84. Woolf AD, Erwin J, March L. The need to address the
74. Diakow PR, Gadsby TA, Gadsby JB, Gleddie JG, Leprich burden of musculoskeletal conditions. Best Pract Res Clin
DJ, Scales AM. Back pain during pregnancy and labor. J Rheumatol. 2012;26:183–224.
Manipulative Physiol Ther. 1991;14:116–8. 85. Stuge B, Garratt A, Krogstad Jenssen H, Grotle M. The
75. Lisi AJ. Chiropractic spinal manipulation for low back pelvic girdle questionnaire: a condition-specific instrument
pain of pregnancy: a retrospective case series. J Midwifery for assessing activity limitations and symptoms in people
Womens Health. 2006;51:e7–10. with pelvic girdle pain. Phys Ther. 2011;91:1096–108.
76. Martins RF, Pinto ESJL. Treatment of pregnancy-related 86. van Tulder M, Furlan A, Bombardier C, Bouter L.
lumbar and pelvic girdle pain by the yoga method: a Updated method guidelines for systematic reviews in the
randomized controlled study. J Altern Complement Med. cochrane collaboration back review group. Spine (Phila Pa
2014;20:24–31. 1976). 2003;28:1290–9.
77. Haugland KS, Rasmussen S, Daltveit AK. Group
intervention for women with pelvic girdle pain in
pregnancy. A randomized controlled trial. Acta Obstet
Supporting information
Gynecol Scand. 2006;85:1320–6. Additional Supporting Information may be found in the
78. Mantle MJ, Holmes J, Currey HL. Backache in pregnancy online version of this article:
II: prophylactic influence of back care classes. Rheumatol
Rehabil. 1981;20:227–32. Table S1. Included randomized controlled trials on physi-
79. Ciardi S, Gozzo V, Wilmarth M. Pregnant women’s cal exercises performed in groups or individually.
response to a prenatal body mechanics and exercise Table S2. Included randomized controlled trials on pelvic
program for the prevention of low back pain: report on a belts, manual therapies, patient education, self-manage-
pilot study. J Sect Women’s Health. 2002;26:17–22. ment interventions, electrotherapy, and yoga.
80. Bastiaenen CH, de Bie RA, Wolters PM, Vlaeyen JW, Table S3. External sources of bias.
Leffers P, Stelma F, et al. Effectiveness of a tailor-made

ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 1156–1167 1167