The Journal of Maternal-Fetal & Neonatal Medicine

ISSN: 1476-7058 (Print) 1476-4954 (Online) Journal homepage: https://www.tandfonline.com/loi/ijmf20

Uterine exteriorization versus intraperitoneal

repair in primary and repeat cesarean delivery: a
randomized controlled trial

Aya Mohr-Sasson, Elias Castel, Irina Lurie, Sigal Heifetz, Salim Kees & Eyal
Sivan

To cite this article: Aya Mohr-Sasson, Elias Castel, Irina Lurie, Sigal Heifetz, Salim Kees & Eyal
Sivan (2020): Uterine exteriorization versus intraperitoneal repair in primary and repeat cesarean
delivery: a randomized controlled trial, The Journal of Maternal-Fetal & Neonatal Medicine, DOI:
10.1080/14767058.2020.1720638

To link to this article: https://doi.org/10.1080/14767058.2020.1720638

Published online: 02 Feb 2020.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ijmf20
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE
https://doi.org/10.1080/14767058.2020.1720638

ORIGINAL ARTICLE

Uterine exteriorization versus intraperitoneal repair in primary and repeat

cesarean delivery: a randomized controlled trial
Aya Mohr-Sassona,b , Elias Castela,b, Irina Luriea, Sigal Heifetza,b, Salim Keesa,b and Eyal Sivana,b
a
Department of Obstetrics and Gynecology, Sheba Medical Center, Ramat Gan, Israel; bSackler School of Medicine, Tel Aviv
University, Tel Aviv, Israel

ABSTRACT ARTICLE HISTORY

Purpose: The aim of this study is to evaluate the effect of uterine exteriorization versus intraper- Received 8 November 2019
itoneal repair, in first compared to repeat cesarean delivery. Accepted 21 January 2020
Methods: A prospective randomized control single-blinded trial conducted in a single tertiary
KEYWORDS
center between March 2014 and March 2015, including 32 and 63 women in first and recurrent
Intraperitoneal repair;
cesarean sections, respectively. Inclusion criteria were elective operation and gestational age primary cesarean delivery;

37 weeks. Operative outcomes were compared between the groups including mean operative repeat cesarean delivery;
time, blood loss, hypotension, perioperative nausea and pain. Post-operative outcomes were fur- uterine exteriorization;
ther compared, including post-operative analgesia demand, first recognized bowel movement, uterine insicition
nausea, length of hospital stay, fever, endometritis surgical site infection rate, and total
satisfaction.
Results: During the study period, 45 and 50 women were designated for uterine exteriorization
and intraperitoneal uterine repair, respectively. Mean blood loss was 452 cc (±10.44) for the
extraperitoneal compared to 540 cc (±29.83) for the intraperitoneal uterine repair group
(p ¼ .004). No other significant differences in either intraoperative or postoperative complications
were demonstrated in and between the groups.
Conclusion: Intraperitoneal repair of uterine incision is associated with higher operative blood
loss compared to uterine exteriorization. No other differences in operative and postoperative
complication rates were found between the groups.

Introduction in primary versus repeat cesarean delivery. The aim of

this study is to evaluate the effects of uterine exterior-
Cesarean delivery (CD) is one of the most frequently
ization versus intraperitoneal repair, in primary and
performed surgical procedures in women, with an
increasing rate of operations worldwide [1–3]. The repeat cesarean delivery.
World Health Organization (WHO) has recommended a
maximum CD rate of 10–15% [4] in order to reduce Materials and methods
maternal and neonatal morbidity and mortality [5];
nevertheless, rates have been reported up to more This is a prospective randomized control single-
than 50%, especially in developing countries [6,7]. blinded trial conducted in a single tertiary center
Different techniques have been practiced in order between March 2014 and March 2015. Women were
to reduce morbidity during and after cesarean deliv- recruited in the preoperative clinic. After signing
ery. The techniques vary depending on both the clin- informed consent, randomization was conducted using
ical situation and the preferences of the operator, and the block randomization method and was carried out
mainly due to limited information available concerning separately for women in primary and for women in
the most appropriate surgical technique to adopt [8]. repeat cesarean delivery. Inclusion criteria were elect-
Uterine placement during the closure of the uterine ive operation and gestational age
37 weeks. Patients
incision after delivering the fetus and removing the with abnormal placentation, chorioamnionitis, previous
placenta has been specifically addressed in numerous other abdominal surgery, past complication during the
studies [9–12]; nevertheless, findings are not unani- first CD on operation report, bleeding disorder, or
mous, and no reference differentiates the techniques urgent operation were excluded.

CONTACT Aya Mohr-Sasson mohraya@gmail.com Department of Obstetrics and Gynecology, Sheba Medical Center, Ramat Gan, Israel
ß 2020 Informa UK Limited, trading as Taylor & Francis Group
2 A. MOHR SASSON ET AL.

All operations were conducted using the same sur- Primary outcome was defined as any operative
gical steps and performed by perinatology experts complication including excessive blood loss (>500 cc),
that are highly skilled in both techniques. During pri- intraoperative nausea, vomiting, and intraoperative
mary cesarean delivery, a transverse skin incision was pain. Secondary outcomes included postoperative
used (Pfannenstiel type) using scalpel. In repeat cesar- analgesia demand, first recognized bowel movement,
ean delivery, skin incision was carried out on top of nausea, length of hospital stay, fever, endometritis sur-
the previous uterine scar. Subcutaneous tissue layer gical site infection rate, and total satisfaction.
was cut by electrocautery that was also used if fibrotic The study protocol was approved by the “Sheba
tissue was seen during repeat operations. Fascial layer Medical Center” Institutional Review Board (ID494–13-
was incised by a small transverse incision usually SMC) on 8 December 2013 and was supported by the
made medially with scalpel and then extended lat- National Institutes of Health (NCT02373501).
erally with scissors or by the use of electrocautery.
Rectus muscles were separated bluntly. Similarly, peri- Statistical analysis
toneum was opened when possible using fingers in
order to minimize the risk of inadvertent injury to Normality of the data was tested using the Shapiro–Wilk
bowel, bladder, or other organs. Adhesions observed or Kolmogorov–Smirnov tests. Data are presented as
in the surgical field, more often seen during repeat median and interquartile range (IQR). Comparison
between unrelated variables was conducted with
operation, were separated using electrocautery.
Student’s t-test or Mann–Whitney U test, as appropriate.
Bladder flap was made based on surgeon decision.
The chi-square and Fisher’s exact tests were used for
Low transverse incision was made along the lower
comparison between categorical variables. Significance
uterine segment. If uterine scar was detected, incision
was accepted at p < .05. Statistical analyses were con-
was made if possible on top of it. After delivery of the
ducted using the IBM Statistical Package for the Social
fetus and the placenta, the uterus was sutured extra-
Sciences (IBM SPSS v19; IBM Corporation Inc, Armonk,
abdominal or intraperitoneal based on allocation.
NY, USA). Analysis was made by in tension to treat.
Uterus was closed in a two layer, continuous full-thick-
ness closure with delayed absorbable synthetic
braided suture (polyglactin 910). Peritoneum was left Results
opened. Fascia was closed using a continuous non- During the study period, 44 and 86 women were
locking closure with slowly absorbable #1 braided recruited for the study in the primary and repeat
suture (polyglactin 910) or with monofilament, (poly- cesarean delivery, respectively. Of them, 32 (72.72%)
dioxanone) in primary and repeat cesarean deliveries, and 63 (73.25%) women were finally included in the
respectively. Subcutaneous adipose layer was sutured analysis. Reason for exclusion after recruitment was
with interrupted delayed absorbable sutures if the due to transition to urgent/emergent cesar-
layer was
2 cm thick. Skin was reapproximated ean delivery.
using staples. In the primary CD group, more women were oper-
Operative outcomes were compared between the ated using the intraperitoneal technique (n ¼ 19, 60%)
groups including mean operative time in minutes, esti- compared to those that underwent uterine exterioriza-
mated blood loss in milliliters, hypotension reported tion (n ¼ 13, 40%). Women in the repeat CD group,
by the anesthesiologist as a sudden drop of blood were equally divided. Altogether, a total of 50 and 45
pressure (usually more than 20 mmHg), perioperative women, in each of the groups, were designated for
nausea reported by the woman, and pain by the intraperitoneal and uterine exteriorization, respectively
Visual Analog Scale (VAS) assessing pain [13]. (Figure 1).
Postoperative outcomes were further compared Comparing women’s demographics and characteris-
including excessive analgesia demand (frequently than tics, demonstrated no statistical difference in age, ges-
the accepted ward protocol), day of first recognized tational week at operation, body mass index (BMI),
bowel movement, nausea, length of hospital stay, gravidity, and parity (Table 1). Women in both of the
fever above 38  C, suspected endometritis, surgical site groups had a history of only one CD in the
infection rate, vascular thromboembolic events, and past (p ¼ .35).
total satisfaction evaluated on day 3 post operation by Operative and postoperative characteristics are
scale of 1–10 (while 1 stand for not satisfied at all and described in Table 2. Intraperitoneal uterine repair was
10 for highly satisfied). found to be shorter [median time 24 (IQR 20–28) versus
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 3

Figure 1. Study population (by intention to treat).

Table 1. Patients’ demographics and characteristics.a

Intraperitoneal (n ¼ 50) Extra-abdominal (n ¼ 45) p Value
Age (years) 35 (31–38) 34 (30–37) .38
Gestational week 38.3 (38.2–39.5) 38.3 (38.1–39.5) .44
BMI (kg/m2) 23.29 (20.57–28.67) 22.98 (19.95–25.71) .24
Gravidity 3 (2–4) 3 (2–4) .87
Parity 1 (0–1) 1 (0–2) .54
Past cesarean delivery 1 (0–1) 1 (0–2) .35
BMI: body mass index.
Data are presented as median and interquartile range (IQR).

Table 2. Operative and postoperative characteristics. with no superiority of one over the other. No cases of
Intraperitoneal Extra-abdominal fever, vascular thromboembolic events, endometritis, or
(n ¼ 50) (n ¼ 45) p Value
wound infection were reported.
Operation duration (min) 24 (20–28) 27 (21–40) .07
Blood loss (ml)a 540 (±29.83) 452 (±10.44) .004 Subanalyzing intraperitoneal technique and uterine
Adhesionsb 20 (40%) 17 (37.8%) .35 exteriorization separately for primary and repeat cesar-
Nauseab 16 (32%) 17 (37.8%) .55
Hypotensionb 2 (4%) 1 (2.2%) .62
ean delivery revealed no differences for operative and
Pain (1–10 scale)b 5 (10%) 2 (4.5%) .31 postoperative complications (Table 3). Higher rates of
Post operation pain (1–10) 8 (7–10) 8 (6–10) .92 adhesions, as expected, and intraoperative nausea
Time until walking (hours) 8 (6–16) 6 (6–13) .44
Bowel movement (day) 2 (2–3) 2 (2–3) .35 reported by the women were observed in the repeat
Satisfaction (1–10 scale) 9 (7–10) 10 (7–10) .63 compared to the primary CD group, however, with no
Data are presented as median and interquartile range (IQR).
a
Data are presented as mean and standard deviation (SD).
statistical difference between the techniques.
b
Data are presented as n, (%). Additional analysis based on actual exteriorization of
the uterus versus intraperitoneal repair was made.
Matching was much lower in the intraperitoneal [n ¼ 8/
27 (IQR 21–40) min], with higher rates of hypotension 19 (42.1%) and n ¼ 21/31 (67.7%) for primary and
[2 (4%) versus 1 (2.2%)], and perceived as more painful repeat CD, respectively] compared to the uterine exter-
[5 (10%) versus 2 (5.5%)] compared to uterine exterior- iorization repair groups [n ¼ 24/13 (184%) and n ¼ 42/
ization, all nonreaching statistical significance. Mean 29 (131%) for primary and repeat CD, respectively].
blood loss was 452 cc (±10.44SD) for the extraperito- Table 4 describes operative and postoperative char-
neal compared to 540 cc (±29.83SD) for the intraperito- acteristics based on analysis by groups. In this analysis,
neal uterine repair group, the only operative difference postoperative pain, assessed by VAS, was higher in the
that was found to be significant (p ¼ .004). Estimation uterine exteriorization [median 9 (IQR 7–10)] compared
of day one postoperative pain by VAS was 8 (p ¼ .92), to intraperitoneal uterine repair [median 7 (IQR 6–7)]
and median time for bowel movements was on the group, in primary CD (p ¼ .03). All other parameters
second day (p ¼ .35), in both of the groups. The rate of were comparable between the groups, in concordance
satisfaction was found to be high in both of the groups with primary analysis by intention to treat.
4 A. MOHR SASSON ET AL.

Table 3. Operative and postoperative characteristics – analysis by intention to treat.

Primary CD (n ¼ 32) Repeat CD (n ¼ 63)
Intraperitoneal Extra-abdominal Intraperitoneal Extra-abdominal
Operation type (n ¼ 19) n (%) (n ¼ 13) n (%) p Value (n ¼ 31) n (%) (n ¼ 32) n (%) p Value
Adhesions 0 (0.0) 1 (7.7) .22 12 (41.1) 10 (37) .10
Nausea 5 (26.3) 3 (23.1) .83 11 (35.5) 14 (43.8) .50
Hypotension 1 (5.3) 0 (0) .55 1 (3.2) 1 (3.1) .52
Pain 3 (15.8) 2 (6.5) .28 1 (8.3) 1 (3.1) .98
Operation duration (min)a 25 (19–28) 31 (19–44) .24 24 (20–30) 27 (22–33) .17
Blood loss (ml)a 500 (462–600) 475 (412–500) .06 500 (500–600) 500 (450–500) .03
Post operation pain 7 (7–10) 9 (8–10) .23 8 (6–10) 8 (6–10) .43
(1–10 scale)a
Time until walking (hours)a 12 (6–18) 12 (6–19) .90 7 (6–13) 6 (6–12) .54
Bowel movement (day)a 3 (2–3) 2 (2–3) .26 2 (2–3) 2 (2–3) .72
Satisfaction (1–10 scale)a 8 (7–10) 9 (6–10) .92 9 (8–10) 10 (7–10) .64
a
Data are presented as median and interquartile range (IQR).

Table 4. Operative and postoperative characteristics – analysis by group.

Primary CD (n ¼ 32) Repeat CD (n ¼ 63)
Intraperitoneal Extra-abdominal Intraperitoneal Extra-abdominal
Operation Type (n ¼ 8) n (%) (n ¼ 24) n (%) p Value (n ¼ 21) n (%) (n ¼ 42) n (%) p Value
Adhesions 0 (0.0) 1 (4.2) .56 9 (42.8) 20 (47.6) .57
Nausea 4 (50.0) 4 (16.7) .05 8 (38.1) 17 (40.5) .86
Hypotension 0 (0.0) 1 (4.2) .56 1 (4.8) 1 (2.4) .61
Pain 2 (25) 2 (8.7) .24 2 (9.5) 1 (2.4) .25
Operation duration (min)a 20 (19–25) 26 (18–40) .33 26 (20–30) 24 (22–38) .72
Blood loss (ml)a 475 (345–600) 500 (450–500) .81 500 (500–575) 500 (450–500) .17
Post operation pain 7 (6–7) 9 (7–10) .03 8 (6–10) 8 (6–10) .44
(1–10 scale)a
Time until walking (hours)a 7 (6–15) 12 (6–18) .33 6 (6–21) 7 (6–12) .69
Bowel movement (day)a 3 (2–3) 2 (2–3) .12 2 (2–3) 2 (2–3) .93
Satisfaction 8 (7–10) 8 (7–10) .90 9 (8–10) 10 (7–10) .88
(1–10 scale)a
a
Data are presented as median and interquartile range (IQR).

Note of transparency – Preliminary analysis based inconsistent with regard to blood loss when comparing
on earlier work was conducted and presented at the both techniques [15]. The Cochrane published in 2000
38th Annual meeting 2018 SMFM conference [14]. meta-analysis including two trials that involved 486
women, concluded that exteriorization made no signifi-
cant difference to blood loss [16]. Nevertheless, Orji
Discussion et al. [15] found 8 years later, in a prospective study
Main findings including 210 women, significant reductions in intrao-
perative blood loss (p < .05) in the uterine exterioriza-
Our study revealed the following findings: (1) Higher
tion group. Meta-analysis performed using Zaphiratos
blood loss was reported during intraperitoneal com- et al. [12] in 2015 reported uterine repair by exterioriza-
pared to extraperitoneal repair of uterine incision. tion may reduce blood loss and the associated decrease
(2) No differences were observed in all other operative in hemoglobin; however, he concluded that this differ-
and postoperative parameters comparing intraperito- ence is not of clinically relevance.
neal to exteriorization of the uterus and in subanalysis In our study, 66/95 (69.47%) of the women under-
for primary compared to repeat operations. (3) In add- went actual exteriorization of the uterus although
itional analysis by grouping, a difference in the VAS only 45/95 (47.36%) were allocated to this group pri-
score for pain assessment was found higher for the marily. Due to this observation, we assumed that our
exteriorization, in the primary CD group. finding did have clinical influence as opposed to
Zaphiratos [12]: it might have been that due to higher
blood loss in the intraperitoneal repair group, and in
Interpretation
order to improve visualization of the incision and
Our study revealed significantly higher blood loss dur- achieve better blood loss control, a transition from
ing intraperitoneal compared to extraperitoneal repair intraperitoneal to extraperitoneal uterine incision
of uterine incision (þ88 cc, p < .004). Published data are repair was decided.
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 5

In all other operative and postoperative parameters, Clinical implications

no difference was observed comparing intraperitoneal
The implications of our findings are concerning the
to exteriorization of the uterus (Table 2). Furthermore,
selection of operative technique based on women’s
we report no postoperative differences between tech-
past surgery. Women undergoing either first or repeat
niques in subanalysis for primary compared to repeat
CD might benefit from exteriorization of the uterus due
operation (Table 3). Our findings correlate to previous to less bleeding and most probably better observation
studies that are not consistent in their postoperative of uterine incision. This recommendation is while taking
differences [16–18]. A possible explanation for our into consideration a slight elevation in postoperative
results could be the relatively small study groups cre- pain, in the primary CD group. It is reasonable, in
ated due to the subanalysis. Larger randomized pro- women undergoing repeat cesarean delivery, due to
spective investigation need to be done to strengthen similar operative and postoperative results, to consider
these finding. intraperitoneal repair in expectance of adhesions.
In analysis by grouping, we did observe a difference
in the VAS score for pain assessment that was higher
for the exteriorization of the uterus by 2 points (9 ver- Conclusion
sus 7 cm), in the primary CD group; however, both Our study revealed that intraperitoneal repair of the uter-
scores were in the “severe pain” reference [19]. This ine incision at CD was associated with higher blood loss
finding might be explained by the routine examination compared to exteriorization. Women undergoing extrap-
of the gutters for blood clots while performing exter- eritoneal repair are more prone to suffer from postopera-
iorization of the uterus that might create additional tive pain after primary cesarean delivery; nevertheless, in
abdominal discomfort. This intervention during oper- all other operative and postoperative characteristics,
ation is not performed during intraperitoneal repair. both techniques were comparable for primary and
This difference was not observed in the repeat CD repeat cesarean delivery. Tailoring the proper surgical
group, whereas VAS score for pain was similar for both technique should be based mainly on operator skills
techniques. It is possible that in multigravida women, when both techniques are equally feasible to perform.
tissues are more flexible and therefore less irritated by These findings are of interest to physicians and patients
the discomfort created during this maneuver. alike. Further larger randomized prospective investigation
need to be done to strengthen these finding.

Strength and limitations

Author’s contribution
The strength of this study is in it being a prospective
Aya Mohr-Sasson – conception, carrying out, plan-
study with randomized allocation to the study groups.
ning, analyzing and writing up
Moreover, to the best of our knowledge, this is the
Elias Castel – conception, carrying out
first study comparing uterine scar repair in both tech-
Irina Luria – conception, carrying out
niques and separately for primary and repeat cesarean
Sigal Heifetz – carrying out
deliveries. Finally, analysis was carried out both by
Sailm Kees – carrying out
intention to treat and by treatment carried out, in
Eyal Sivan – planning, analyzing and writing up
order to overcome the mismatch between the groups,
mainly observed in the intraperitoneal group.
The study has several limitations. Although random- Disclosure statement
ization was performed, 28% of the women that were No potential conflict of interest was reported by
recruited were excluded between randomization and the author(s).
actual cesarean delivery. The reason was due to
urgent/emergent CD performed prior to the date ORCID
planned for the elective cesarean delivery.
Aya Mohr-Sasson http://orcid.org/0000-0001-9818-3041
Furthermore, matching between allocation and final
surgical technique was much lower in the intraperito-
neal compared to the extra-abdominal group, making References
analysis by intention to treat less reliable. [1] Belizan JM, Althabe F, Cafferata ML. Health conse-
Nevertheless, in additional analysis by grouping, simi- quences of the increasing caesarean section rates.
lar findings were observed. Epidemiology. 2007;18(4):485–486.
6 A. MOHR SASSON ET AL.

[2] Tollånes MC. Increased rate of caesarean sections – [12] Zaphiratos V, George RB, Boyd JC, et al. Uterine exter-
causes and consequences. Tidsskriftet. 2009;129(13): iorization compared with in situ repair for cesarean
1329–1331. delivery: a systematic review and meta-analysis. Can J
[3] Shabila NP. Rates and trends in cesarean sections Anesth/J Can Anesth. 2015;62(11):1209–1220.
between 2008 and 2012 in Iraq. BMC Pregnancy [13] Haefeli M, Elfering A. Pain assessment. Eur Spine J.
Childbirth. 2017;17(1):22. 2006;15(S1):S17–S24.
[4] Betran AP, Torloni MR, Zhang JJ, et al. WHO state- [14] Available from: https://www.sciencedirect.com/sci-
ment on caesarean section rates. BJOG: Int J Obstet
ence/article/pii/S0002937817318513?via%3Dihub.
Gy. 2016;123(5):667–670.
[15] Orji EO, Olaleye AO, Loto OM, et al. A randomised
[5] Stanton CK, Holtz SA. Levels and trends in cesarean
controlled trial of uterine exteriorisation and non-
birth in the developing world. Stud Fam Plann. 2006;
37(1):41–48. exteriorisation at caesarean section. Aust N Z J Obstet
[6] Althabe F, Belizan JM. Caesarean section: the paradox. Gynaecol. 2008;48(6):570–574.
Lancet. 2006;368(9546):1472–1473. [16] Wilkinson C, Enkin MW. Uterine exteriorization versus
[7] Molina G, Weiser TG, Lipsitz SR, et al. Relationship intraperitoneal repair at caesarean section. Cochrane
between cesarean delivery rate and maternal and Database Syst Rev. 2000;2(2):CD000085.
neonatal mortality. JAMA. 2015;314(21):2263–2270. [17] Jacobs-Jokhan D, Hofmeyr G. Extra-abdominal versus
[8] Dodd JM, Anderson ER, Gates S, et al. Surgical techni- intra-abdominal repair of the uterine incision at cae-
ques for uterine incision and uterine closure at the sarean section. Cochrane Database Syst Rev. 2004;
time of caesarean section. Cochrane Database Syst 4(4):CD000085.
Rev. 2014;7(7):CD004732. [18] Edi-Osagie EC, Hopkins RE, Ogbo V, et al. Uterine
[9] Ozbay K. Exteriorized versus in-situ repair of the uter- exteriorisation at caesarean section: influence on
ine incision at cesarean delivery: a randomized con- maternal morbidity. BJOG. 1998;105(10):1070–1078.
trolled trial. Clin Exp Obstet Gynecol. 2011;38(2): [19] Hawker GA, Mian S, Kendzerska T, et al. Measures of
155–158. adult pain: Visual Analog Scale for Pain (VAS Pain),
[10] Siddiqui M, Goldszmidt E, Fallah S, et al.
Numeric Rating Scale for Pain (NRS Pain), McGill Pain
Complications of exteriorized compared with in situ
Questionnaire (MPQ), Short-Form McGill Pain
uterine repair at cesarean delivery under spinal anes-
thesia: a randomized controlled trial. Obstet Gynecol. Questionnaire (SF-MPQ), Chronic Pain Grade Scale
2007;110(3):570–575. (CPGS), Short Form-36 Bodily Pain Scale (SF-36 BPS),
[11] Gode F, Okyay RE, Saatli B, et al. Comparison of uter- and Measure of Intermittent and Constant
ine exteriorization and in situ repair during cesarean Osteoarthritis Pain (ICOAP). Arthritis Care Res. 2011;
sections. Arch Gynecol Obstet. 2012;285(6):1541–1545. 63(Suppl 11):S240–S252.