MATERNAL AND NEONATAL HEALTH

Analysis of pelvic floor musculature function during pregnancy and

postpartum: a cohort study
(A prospective cohort study to assess the PFMS by perineometry and digital vaginal palpation
during pregnancy and following vaginal or caesarean childbirth)

Adriana de Souza Caroci, Maria Luiza Gonzalez Riesco, Wesllanny da Silva Sousa, Ana Carolina
Cotrim, Edinéia Maria Sena, Nelly Lima Rocha and Cintia Najara Caroci Fontes

Aim. To compare the measurements of women’s pelvic floor musculature strength (PFMS) during pregnancy and postpartum
period.
Background. Pregnancy and childbirth can have an influence on the muscles and pelvic floor and can cause morbidities of
women’s genito-urinary tract.
Design. A prospective cohort study.
Methods. There were included 226 primigravidae women, attended by community health services in the city of Ita-
pecerica da Serra, Sao Paulo, Brazil. The participants were followed in four stages: (1) within 12 weeks of pregnancy; (2)
between 36–40 weeks of pregnancy; (3) within 48 hours after childbirth; (4) 42-60 days after childbirth. Data were
collected from February 2007–August 2008. The pelvic floor musculature strength was evaluated by perineometry and
digital vaginal palpation in stages 1, 2 and 4. The final sample included 110 women who completed all four stages of
the study.
Results. The pelvic floor musculature strength of the women did not change significantly during pregnancy or after
delivery (ANOVA : p = 0Æ78). In all three examined stages, a low-intensity pelvic floor musculature strength was prevalent
(in mmHg: stage 1 = 15Æ9; stage 2 = 15Æ2, stage 4 = 14Æ7), with scores from 0–3 on the Oxford scale. The pelvic floor
musculature strength did not differ in relation to maternal age, skin colour, conjugal status, dyspareunia, stool char-
acteristics, type of delivery, or conditions of the perineum. An interaction between maternal nutritional state and
newborn’s weight may affect the pelvic floor musculature strength (MANOVA : p = 0Æ04).
Conclusion. Pregnancy and childbirth did not reduce significantly pelvic floor musculature strength. The perineometry and
digital vaginal palpation used to assess the pelvic floor musculature strength were well accepted by the women.
Relevance to clinical practice. In clinical practice, digital vaginal palpation is effective for supporting the diagnosis of
urinary, intestinal and sexual dysfunctions. Perineometry use is particularly important together with the performance of
perineal exercises with biofeedback in the treatment these disorders.

Authors: Adriana de Souza Caroci, MSc, PhD, RNM, Lecturer,
Adventist University Center of São Paulo; Maria Luiza Gonzalez
Riesco, MSc, PhD, RNM, Associate Professor, School of Nursing of
Universidade de São Paulo; Wesllanny da Silva Sousa, RN,
Midwifery Advanced Program Student, Adventist University Center
of São Paulo and Nurse, Family Health Program of Iaciara, Goiás;
Ana Carolina Cotrim, RN, Nursing Advanced Program Student,
Educational Institute of São Paulo; Edinéia Maria Sena, RN,
Midwifery Advanced Program Student, Adventist University Center
of São Paulo; Nelly Lima Rocha, RN, Midwifery Advanced Program
Student, Adventist University Center of São Paulo; Cintia Najara
Caroci Fontes, RN, Midwifery Advanced Program Student, Adventist
University Center of São Paulo and Nurse, Renal Care Medical
Center, São Paulo, Brazil
Correspondence: Maria Luiza Gonzalez Riesco, Associate Professor,
School of Nursing, Universidade de São Paulo, São Paulo, Brazil.
Telephone: +55-11-30617607.
E-mail: riesco@usp.br

2424  2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
doi: 10.1111/j.1365-2702.2010.03289.x
Maternal and neonatal health A cohort study of pelvic floor musculature function during pregnancy and postpartum
Key words: childbirth, lacerations, midwifery, nursing, pelvic floor, perineum
Accepted for publication: 7 April 2010
Introduction
A loss or reduction in perineal muscle tone is very common
among women who have given birth vaginally (Hay-Smith
et al. 2002). This change can lead to genito-urinary and anal
dysfunction, which has a negative effect on women’s sexual,
physical, social and psychological health (Wyman et al. 1990,
Norton et al. 2006). Pregnancy and vaginal birth can cause
stress urinary incontinence (UI) and damage to the pelvic floor
(PF). The main studies in this area have concentrated on
investigating the mechanisms that cause PF trauma and on
appropriately evaluating the symptoms contributing to this
clinical picture (Fitzpatrick & O’Herlihy 2001). Longitudinal
studies have long been recommended to evaluate the effects
of pregnancy and childbirth on UI and PF function (Kolbl
2001).
Evaluation of pelvic floor musculature strength (PFMS)
after childbirth forms a key part of the evidence used by
studies and research. Such studies were enabled by the
development of methods and equipment that have improved
this evaluation and, consequently, the accuracy of diagnosis
(Fitzpatrick & O’Herlihy 2001). PF musculature can be
classified into different functional levels, and PFMS evalua-
tion should be conducted based on a gynaecological exam-
ination using different methods, such as electromyography,
perineometry, digital vaginal palpation, vaginal cones, ultra-
sonography and manometry (Fitzpatrick & O’Herlihy 2001,
Kolbl 2001, Fleming et al. 2003, Herbison et al. 2002,
Dumoulin et al. 2004, Menta & Schirmer 2006, Örnö et al.
2008, Braekken et al. 2009). Besides pregnancy and vaginal
birth, parity, perineal trauma and other mechanic, endocrine
and neural factors can contribute to the reduction in the
PFMS (Fitzpatrick & O’Herlihy 2001, Coletti et al. 2005,
Norton et al. 2006, Zanetti et al. 2007, Castro et al. 2008,
Örnö et al. 2008, Voorham-van der Zalm et al. 2008,
Braekken et al. 2009). To summarise, the hypothesis inves-
tigated in this study was that pregnancy and childbirth reduce
women’s PFMS.
Aim
The aim of this study was to compare the measurements of
women’s PFMS during pregnancy and postpartum period
based on the following factors: maternal age, skin colour,
conjugal status, dyspareunia, nutritional state, stool charac-
 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
teristics, type of delivery, condition of the perineum and the
newborn’s weight.
Methods
Design
A prospective cohort study examining women’s PFMS
during pregnancy and following vaginal or caesarean child-
birth was conducted. PFMS was evaluated by perineometry
and digital vaginal palpation.
Setting
Data concerning pregnant women, childbirth and postpartum
period were collected in five community health services
(CHS) and in two public hospitals in Itapecerica da Serra,
Brazil. In CHS, nurse-midwives initiate prenatal care and
subsequent consultations involve doctors. Nurse-midwives
are responsible for uncomplicated births, while doctors
attend to deliveries involving dystocia and caesareans. These
services offer coverage to a population of approximately
187,000 inhabitants.
Ethical considerations
This study was approved by the Committee for Research
Ethics and authorised by the appropriate authorities at the
locations where it was carried out. Women were admitted to
the study after receiving oral and written information from
the researchers and giving voluntary written consent. The
study’s outline was reviewed with participants before they
signed consent forms. The researchers had no ties to the
manufacturers or distributors of the equipment used in this
study.
Population and sample size
The inclusion criteria were to be nulliparous in the first
12 weeks of pregnancy. The exclusion criteria were as
follows: multiple pregnancy; previous abdominal or urogen-
ital surgery; use of hormone therapy; diseases that might
interfere with PFMS; resistance to digital vaginal palpation or
insertion of a perineometer into the vagina; difficulty in
understanding Portuguese or in communicating.
2425
A de Souza Caroci et al.
The sample size was calculated based on the study of
Menta and Schirmer (2006). Assuming 5% probability for
the occurrence of Type I error and a test power of 80%, it
was determined that 136 pregnant women were needed.
Taking into consideration that this was a longitudinal study
with different stages of data collection, the possibility that
some women might not participate in the full course of the
study was considered. Thus, to obtain the estimated sample
size, the recruitment of participants was increased by 50%,
with the aim being to recruit 204 women.
To guarantee the estimated sample size, 226 pregnant
women were enrolled. As the number of women who
dropped out of the study was higher than expected, the
researchers performed a comparative analysis between
women included in the final sample and those who quit the
study. The homogeneity of both groups suggests that the
effect of the loss did not greatly influence the results of
the final sample of the 110 women in the cohort.
Data collection
Data were collected in four stages. Stages 1, 2 and 4 involved
consultations, interviews, physical examinations and PFMS
measurements. The same procedures were performed in stage
3, except for omission of the PFMS measurement. Stages 1–4
were as follows:
• Stage 1 was conducted with women during the first
12 weeks of pregnancy. A subsequent doctor’s appoint-
ment and the next return visit with the researcher for the
second stage of the study were established.
• Stage 2 was conducted when the women were between 36–
40 weeks pregnant.
• Stage 3 was conducted with postpartum women up to
48 hours after childbirth. The researcher was in daily
contact with the hospitals to find out when participants of
the study were admitted. PFMS was not evaluated because
of the discomfort this might have produced after childbirth
and because this data could be evaluated at a later time. An
appointment was made for the mother to return to the
CHS.
• Stage 4 was conducted with the women in the period of
42–60 days after childbirth.
The researcher made her telephone number available to
these women, in case they needed it and so they could
advise her when they were admitted to hospital. Two days
before the date scheduled for stages 2 and 4, participants in
the study were contacted by phone or by telegram to
confirm their attendance at the CHS. If they missed their
appointments, the researcher made a home visit to collect
the data. After all of the data were collected, the women
2426  2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
were given instructions about perineal exercises to strength-
en PF musculature and were encouraged to perform them;
they were taken to referral services in the event of
complications. The trial profile shows the study’s sequence
(Fig. 1).
Procedures used to measure PFMS
The measurement of PFMS was performed by only one
researcher, in all participants and all stages. Before starting
data collection, the researcher received specialised training by
nurses, physiotherapists, doctors and manufacturers of the
equipment used. Two methods of measuring PFMS were used
on all participants: perineometry and digital vaginal palpa-
tion. For the perineometry, PFMS was measured with an
electronic perineometer. For digital vaginal palpation, PFMS
was classified according to the Oxford scale (Bø & Larsen
1990).
The sequence for applying these methods was randomised,
by means of an electronically produced random numeral
table. Opaque envelopes were used; these were opened only
at the moment that PFMS was measured and contained the
sequence of methods to be used. Thus, digital vaginal
palpation might precede perineometry or vice versa to avoid
any trending in the data.
The Oxford classification scale
• Grade 0 – no contractions;
• Grade 1 – flicker of muscular contraction;
• Grade 2 – low intensity but sustained contraction;
• Grade 3 – moderate contraction, with an increase in
intravaginal pressure, squeezing of fingers and with a slight
elevation of the vaginal wall;
• Grade 4 – satisfactory contraction, which pushes against
the examiner’s fingers, with elevation of the vaginal wall in
the direction of the pubic symphysis;
• Grade 5 – strong contraction, firm compression of the
examiner’s fingers with a positive movement in the direc-
tion of the pubic symphysis.
Equipment
The study used a Quark electronic perineometer, model
Perina 996-2 (Quark Medical Products, Piracicaba, SP,
Brazil), which registers the potential action of PF muscle
contractions and translates their intensity from visual signals
based on a graduated numerical scale that ranges from
0–46Æ4 mmHg. This model is a biofeedback device to be
inserted into a woman’s vagina and it measures the PFMS. It
Maternal and neonatal health A cohort study of pelvic floor musculature function during pregnancy and postpartum
Data concerning pregnancy, childbirth and postpartum period were collected in
5 CHS and in 2 public hospitals in Itapecerica da Serra, Brazil
Pregnant women between 36 and 40 weeks n = 128
• Pacient record
• Interview
• Physical
examination
42 to 60 days after childbirth n = 124, stage 4
Figure 1 Trial profile study’s sequence.
has an electronic display, a pressure transducer and a vaginal
probe. When the contraction varies from 1Æ6–14Æ4 mmHg,
the bar of LEDs is illuminated in green, in orange from 16Æ0–
30Æ4 mmHg and in red from 32Æ0–46Æ4 mmHg.
The vaginal probe is elastic and inflatable, allowing greater
adaptation, perception and comfort. The surface of the tube
with rubber rings favours its keeping in the vagina. The
perineometer selected for this study met all the required
safety norms for electromedical equipment and is registered
with the Brazilian Ministry of Health.
This apparatus registers, indistinctly, all increases in
intracavity pressure, meaning that it does not differentiate
between muscle contractions of the perineum and of the
abdomen. Therefore, a surface electromyograph (Lynx
brand, model Bio-ADS1200; Lynx Electronic Technology,
São Paulo, SP, Brazil) was used to monitor abdominal
relaxation during PFMS measurement. The electromyograph
has the function of indicating the differentiation between
these muscles and it detects, through external electrodes, the
electrical activity of the muscle during rest and contraction,
using computer-linked software. The computer screen shows
muscular activity graphically. The PFMS, measured by the
perineometer, was recorded only when the electromyograph
register indicated abdominal musculature activity compati-
 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
Inclusion criteria: primigravidae women with
no more than 12 weeks of pregnancy
Exclusion criteria: multiple pregnancy; previous
abdominal or urogenital surgery; the use of hormone
therapy; diseases that might interfere with PFMS;
resistance to digital vaginal palpation or the insertion of
a perineometer into the vagina; difficulty in
understanding Portuguese or in communicating
Pregnant women were enrolled n = 226, stage 1
stage 2
• Pacient record
up to 48 hours after • Interview
childbirth n = 134, stage 3, hospital
• Physical examination
• PFMS
Women that finished the four stages of the
study n = 110
ble with rest (on the electromyograph scale of between
0–10 lV).
Procedures used to prepare women for measuring PFMS
by two methods
1 The woman was placed in a gynaecological position, with
the genital region and the lower limbs uncovered but
protected with a sheet;
2 The electromyograph’s four electrodes were connected
over the reto-abdominal muscles (two electrodes on the
right side and two on the left, between the upper border of
the pubis and the umbilical region);
3 The operator put on surgical gloves;
4 The operator taught each woman how to make contrac-
tions as if she were ‘holding in’ urine, using only the PF
muscles and avoiding the contraction of the muscles of the
abdomen, thighs and buttocks.
Perineometry
5 The pressure level of the scale was zeroed;
6 The elastic probe was covered with an unlubricated, dis-
posable condom;
7 Lubricate the condom with lubricant;
2427
A de Souza Caroci et al.
8 The probe was introduced 3-4 cm into the vagina;
9 The woman was instructed to relax her PF muscles;
10 The woman was asked to hold and maintain, for as long
as possible, a voluntary contraction of the perineal mus-
cles around the vaginal probe, in a sequence of three
sessions, with an inter-session interval of 15 seconds;
11 The vaginal probe was maintained in position intra-
vaginally during the PFMS measurement and intervals;
12 The greatest force of the voluntary contraction by the PF
muscles was recorded;
13 The woman was allowed to rest for one minute before
beginning digital vaginal palpation (in case this had not
already been performed, in accordance with the ran-
domisation).
Digital vaginal palpation
5 The two distal phalanges of the pointer and middle fingers
were introduced into the vagina, using lubricant;
6 The woman was asked to make a voluntary contraction of
the perineal muscles around the fingers of the examiner and
to hold it as long as possible, in a sequence of three ses-
sions, with an inter-session interval of 15 seconds;
7 The examiners fingers were kept in the vagina during the
PFMS measurement and intervals;
8 The highest classification level for contraction reached was
recorded, using the Oxford scale;
9 The woman was allowed to rest for one minute before
beginning perineometry (if this had not been performed
previously, in accordance with the randomisation).
The study’s variables
The following variables were examined: PFMS (PF muscle
tone, evaluated during contraction); maternal age; skin
colour; conjugal status; weight; height; body mass index;
classification of nutritional state (low weight, appropriate
weight, overweight, obesity) (Atalah et al. 1997); gestational
age; dyspareunia; stool characteristics; type of delivery;
newborn’s weight; condition of the perineum during delivery
(episiotomy, intact perineum or first to forth degree perineal
lacerations).
Data analysis
Descriptive and comparative statistical analyses were carried
out. Two-tailed tests (chi-square and Student’s t-test) were
used, and the significance level was set at 5% (p < 0Æ05).
Two-way analyses of variance (ANOVA ) was used for multiple
comparisons of the PFMS measurements in stages 1, 2 and 4,
regarding the variables of age, skin colour, conjugal status,
2428  2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
dyspareunia and stool characteristics. Regarding the condi-
tion of the perineum and newborn’s weight, a comparison of
the PFMS measurements in stage 4 was made, using one-
factor ANOVA (Costa Neto 1992).
A multivariate analysis of variance (MANOVA ) was per-
formed to verify whether the variables, when combined,
show interactions or differences not found in the univariate
analysis. The statistical F associated with Wilks’ Lambda
values was adopted (Hair Junior et al. 1998, Ferrán Aranz
2001). When the combination of variables showed an
interaction (p < 0Æ05), ANOVA and Tukey’s HSD test were
performed to identify how this interaction occurred and
discriminate the differences in the means of PFMS (Dawson
& Trapp 2003). All tests were performed using the statistical
packages SAS System for Windows V9 (SAS Institute Inc.,
Cary, NC, USA) and SPSS for Windows v13.0 (SPSS Inc.,
Chicago, IL, USA).
Results
Of the 226 women enrolled, 116 failed to participate to
completion. Thus, the sample of women who finished the
four stages of the study was 110. The average age of the
women was 21Æ4 (SD 5Æ1). The women began prenatal care at
an average gestational age of 9Æ4 weeks (SD 2Æ2). The women
reached stage 2 of the study at an average gestational age of
36Æ5 weeks (SD 0Æ8). First sexual intercourse took place at an
average age of 17Æ3 years (SD 4Æ0). The average length of
active sex life was 3Æ1 years (SD 2Æ8). The average frequency
of sexual activity was 2Æ5 times per week (SD 1Æ6). At the
start of pregnancy, 25 (22Æ7%) women were low weight and
65 (59Æ1%) presented appropriate weight. At the end of
pregnancy, 18 (16Æ4%) of the women were low weight and
57 (51Æ8%) presented appropriate weight.
PFMS can be classified according to intensity as follows: no
contraction (0Æ0); slight contraction (1Æ6–16Æ0 mmHg); mod-
erate contraction (17Æ6–3Æ0 mmHg); normal contraction
(33Æ6–46Æ4 mmHg) (Barbosa et al. 2005). The ANOVA indi-
cated that the perineometry measurements in stages 1
(15Æ9 mmHg), 2 (15Æ2 mmHg) and 4 (14Æ7 mmHg) did not
differ between the stages (p = 0Æ78). Low-intensity (slight)
muscular contractions were most prevalent from the start of
pregnancy through the postpartum period. The study also
verified that the frequency of moderate intensity contractions
declined over time. It should be noted that in those cases
where the PFMS value was higher than 0Æ0, but the surface
electromyograph indicated contractive activity of the abdom-
inal muscle, an absence of contraction was scored.
Among all the women, 73 (66Æ4%) had vaginal deliveries
and 37 (33Æ6%) delivered by caesarean section. The
Maternal and neonatal health A cohort study of pelvic floor musculature function during pregnancy and postpartum

frequency of perineal trauma (episiotomy or spontaneous
laceration) among the women with vaginal deliveries was
93Æ2%. All women who suffered perineal trauma received
perineal sutures. Neonates’ birth weights varied from 2245–
4650 g, with a mean birth weight of 3205 g (SD 466Æ3).
The two-way ANOVA with one of the factors correspond-
ing to each of stages 1, 2 and 4 and the other corresponding
to maternal age, skin colour, conjugal status, dyspareunia
and stool characteristics revealed no significant differences
between the factors or between the stages and the factors. In
analysing each stage on its own, the average perineometry did
not show any statistically significant differences among the
women according to their age group, white or non-white skin
colour, whether or not they lived with their partners, whether
or not they complained of dyspareunia and whether they
regularly presented dried out stool or not. Table 1 synthesises
the descriptive data for the factors analysed, as well as the
respective p-value for each factor.
The comparison of perineometry averages with the nutri-
tional state of the women did not reveal any statistically
significant differences (Table 2). There was no statistically
significant difference in the comparison of perineometry
averages after childbirth among women who had vaginal
deliveries and caesarean sections or among those with
episiotomy, intact perineum or perineal laceration. Equally,
the average PFMS after delivery did not differ statistically
regarding the newborn’s weight (Table 3).
The results of MANOVA showed no significant interactions
among the means of the perineometry for most of the
analysed combinations, considering that p < 0Æ05. Never-
theless, the combination between the newborn’s weight and
the mother’s nutritional state may influence the PFMS

Table 1 Means of perineometry of stages 1, 2 and 4

Perineometry (mmHg)

Stage 1 Stage 2 Stage 4

Variable n (%) mean (SD) mean (SD) mean (SD) p-value*

Age (years) Age Stage · age

14–20 50 (45Æ5) 16Æ2 (11Æ1) 16Æ2 (12Æ3) 17Æ2 (10Æ9) 0Æ12 0Æ82
20–30 49 (44Æ5) 15Æ8 (10Æ9) 15Æ1 (12Æ4) 14Æ4 (11Æ0)
30–39 11 (10Æ0) 14Æ1 (1Æ1) 12Æ5 (12Æ2) 11Æ3 (10Æ4)
Skin colour Skin colour Stage · skin colour
White 29 (26Æ4) 16Æ2 (11Æ4) 16Æ5 (12Æ6) 16Æ3 (11Æ2) 0Æ88 0Æ37
Not white 81 (73Æ6) 15Æ7 (11Æ0) 14Æ6 (12Æ3) 13Æ7 (11Æ0)
Conjugal status Conjugal status Stage · conjugal status
Living with partner 67 (60Æ9) 16Æ1 (11Æ0) 14Æ2 (12Æ4) 13Æ4 (10Æ9) 0Æ37 0Æ60
Not living with partner 43 (39Æ1) 15Æ4 (10Æ9) 16Æ5 (12Æ3) 15Æ9 (11Æ0)
Dyspareunia Dyspareunia Stage · dyspareunia
Yes 26 (23Æ6) 14Æ8 (10Æ9) 11Æ9 (12Æ5) 12Æ3 (10Æ8) 0Æ07 0Æ77
No 84 (76Æ4) 16Æ2 (10Æ9) 16Æ1 (12Æ3) 15Æ0 (11Æ0)
Stool characteristics Stool characteristics Stage · stool characteristics
Dried out 40 (36Æ4) 16Æ4 (11Æ0) 15Æ0 (12Æ4) 13Æ4 (11Æ0) 0Æ68 0Æ64
Not dried out 70 (63Æ6) 15Æ6 (11Æ0) 15Æ2 (12Æ3) 15Æ0 (10Æ9)

*ANOVA .

Table 2 Means of perineometry of stages 1 and 2

Perineometry (mmHg) Stage

Stage 1 Stage 2 1·2

Nutricional state n (%) Mean (SD) n (%) Mean (SD) p-value*

Low weight 25 (22Æ7) 14Æ6 (11Æ3) 18 (16Æ4) 16Æ7 (12Æ3) 0Æ58

Appropriate weight 65 (59Æ1) 16Æ4 (10Æ9) 57 (51Æ8) 14Æ2 (12Æ3) 0Æ29
Overweight 14 (12Æ7) 18Æ0 (11Æ2) 20 (18Æ2) 13Æ4 (12Æ3) 0Æ24
Obesity 6 (5Æ5) 10Æ4 (11Æ4) 15 (13Æ6) 19Æ2 (12Æ8) 0Æ10

*Student’s t-test.

 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433 2429
A de Souza Caroci et al.

Table 3 Means of perineometry of stage 4 Discussion

Perineometry (mmHg) Stage 4 There was a high percentage of loss of the study population
Variable n (%) mean (SD) p-value (51Æ3%), which can be attributed to the longitudinal nature
of the study. However, as the variables analysed when
Type of delivery
Spontaneous 73 (66Æ4) 13Æ7 (10Æ1) 0Æ16*
comparing the group of dropouts with the study sample
Caesarean 37 (33Æ6) 16Æ9 (13Æ0) indicated homogeneity between both groups, the likelihood
Condition of the perineum that the results were skewed by those who dropped out of the
Episiotomy 48 (65Æ8) 13Æ2 (10Æ7) 0Æ67 cohort is reduced.
Intact and 1st degree 18 (24Æ6) 13Æ7 (9Æ0) To obtain a more precise PFMS variation, it is important to
laceration
follow the women before, during and after the pregnancy,
2nd degree laceration 7 (9Æ6) 16Æ9 (9Æ0)
Newborn’s weight evaluating parity, type of delivery, perineal conditions,
<2500 6 (5Æ4) 16Æ0 (12Æ5) 0Æ65 genito-urinary and intestinal tract symptoms, as well as sex
2500–3500 76 (69Æ1) 15Æ0 (11Æ0) life. When evaluating the effects of pregnancy on perineal
>3500 28 (25Æ5) 12Æ5 (10Æ9) muscle function by measuring PFMS, some studies have
*Student’s t-test. concluded that pregnancy is associated with a reduction in

ANOVA . PFMS (Fleming et al. 2003). This reduction may occur
because during pregnancy progesterone increases, smooth
Table 4 Multivariate analysis of variance of the means of perine- muscle relaxes, conjunctive tissue is altered, the uterus raises
ometry (mmHg) of stages 1, 2 and 4
abdominal pressure and because of prior anatomical defects
Wilks’ (Valadares 2002).
Perineometry and combined variables Lambda F p-value* The average PFMS of the participants in this study was
Age · nutritional state (stage 2) 0Æ87 0Æ93 0Æ54 between 14Æ7–15Æ9 mmHg, which can be classified as a slight
Age · type of delivery 0Æ98 0Æ44 0Æ85 contraction. These results showed that PFMS decreased,
Age · condition of the perineum 0Æ87 0Æ57 0Æ89 slightly, over the course of pregnancy, and this reduction was
Type of delivery · nutritional 0Æ90 1Æ24 0Æ27 maintained up to the end of the postpartum period. However,
state (stage 2)
there were no statistically significant differences among the
Type of delivery · condition of the 1Æ00 – –
stages, probably because the participants were primigravidae
perineum
Type of delivery · newborn’s weight 0Æ97 0Æ90 0Æ45 women. If there is a reduction in PFMS in each pregnancy,
Condition of the perineum · nutritional 0Æ74 0Æ68 0Æ88 even one of low intensity, over the course of various
state (stage 2) pregnancies, this reduction could become exacerbated, caus-
Condition of the perineum · newborn’s 0Æ83 1Æ35 0Æ22 ing morbidities.
weight
Given the above results, we can suppose that if PFMS does
Newborn’s weight · nutritional state 0Æ80 1Æ86 0Æ04
(stage 2) not recover fully, these women might present morbidity in the
genito-urinary tract in future. Therefore, it is fundamental
*MANOVA .

that health professionals who treat women ask them to bring
ANOVA (p < 0Æ02); Tukey’s HSD (p < 0Æ02).
up any genito-urinary tract complaints. On the other hand,
(p = 0Æ04). In this situation, obese women with babies over the parameters adopted to classify PFMS intensity in women
3500 g showed higher means of perineometry than those without clinical symptoms may be open to debate (Fischer &
with appropriate weight and babies over 3500 g (p < 0Æ02) Linde 1997, Fitzpatrick & O’Herlihy 2001, Cammu et al.
(Table 4). 2004, Diokno et al. 2004, Belo et al. 2005, Castro et al.
As indicated in Table 5, PFMS evaluated in stage 4 by 2008). Further studies of PFMS measurement are necessary
digital vaginal palpation did not show any statistically and important to establish a profile of PFMS during
significant variation in terms of delivery type, presence of pregnancy and after childbirth.
episiotomy, or birth weight of around 3500 g. In this stage, The fact that PFMS in the majority of women involved in
the majority of the women presented vaginal contraction this study was classified as slight or moderate from the start
varying from none to low intensity (grades 0 to 2). PFMS was of pregnancy, regardless of the influence of pregnancy or
considered satisfactory or strong (grades 4 or 5) in 13Æ7 and delivery, can also be attributed to their difficulty in identi-
24Æ3% of the women with vaginal delivery and caesarean fying the perineal muscles and to a lack of practice in
sections, respectively. exercising them (Oliveira et al. 2007).

2430  2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
Maternal and neonatal health A cohort study of pelvic floor musculature function during pregnancy and postpartum

Table 5 Vaginal digital palpation of stage 4

Vaginal digital palpation (stage 4)

Oxford scale grade

0–2 3 4 and 5 Total

Variable n % n % n % n % p-value*

Type of delivery
Spontaneous 39 53Æ4 24 32Æ9 10 13Æ7 73 100 0Æ38
Caesarean 17 46Æ0 11 29Æ7 9 24Æ3 37 100
Total 56 50Æ9 35 31Æ8 19 17Æ3 110 100
Condition of the perineum
With episiotomy 25 52Æ1 16 33Æ3 7 14Æ6 48 100 0Æ98
Without episiotomy 14 56Æ0 8 32Æ0 3 12Æ0 25 100
Total 39 53Æ4 24 32Æ9 10 13Æ7 73 100
Newborn’s weight (g)
£3500 41 50Æ0 27 33Æ0 14 17Æ0 82 100 0Æ91
>3500 15 53Æ6 8 28Æ6 5 17Æ8 28 100
Total 56 50Æ9 35 31Æ8 19 17Æ3 110 100

*Chi-square test.

This being the case, women should be taught the factors
related to PFMS reduction and a supervised programme of
exercises to strengthen the PF should be implemented
during prenatal consults so that they can know how to
properly identify their PF muscles (Zanetti et al. 2007). It is
worth pointing out that all the women evaluated in stage 4
of the study were instructed, verbally and in writing, about
how to perform a programme of perineal exercises.
The average PFMS in this study may be different from
that of other similar studies (Barbosa et al. 2005, Menta &
Schirmer 2006, Oliveira et al. 2007) because it used surface
electromyography to evaluate abdominal muscle activity and
not clinical inspection. This procedure was essential, as the
women had difficulty in distinguishing between contractions
of the muscles of the abdomen and those of the perineum
(Nagib et al. 2005). During data collection, one researcher
proceeded with PFMS measurement, while another read the
results on the electronic visor of the perineometer and the
electromyograph screen. In this way, we sought to avoid
the possibility that the results of the perineometry might
influence the researcher when performing digital vaginal
palpation.
In this study, as in some others, the maternal age, skin
colour and conjugal status had no influence on the PFMS
values during pregnancy or postpartum (Menta & Schirmer
2006, Oliveira et al. 2007), in spite of some authors having
suggested that these variables could interfere with PFMS
(Allen et al. 1990, Klein et al. 1997). Regarding the nutri-
tional state and the newborn’s birth weight, the interaction
of these variables can affect the PFMS, as indicated in this
study. In addition, third- and fourth-degree perineal lacer-
ations can damage the anal sphincter muscles and the nerves
of the pudenda (Fitzpatrick & O’Herlihy 2001); however,
none of the women in this study suffered such severe
degrees of laceration. Although some studies associate UI
and anal incontinence with PF damage inflicted during
vaginal delivery (Dannecker & Anthuber 2000, Voorham-
van der Zalm et al. 2008), more longitudinal studies are
needed to evaluate the effects of pregnancy and delivery on
genito-urinary tract incontinence and on PF function (Kolbl
2001).
Postpartum PFMS showed no statistically significant asso-
ciation with the conditions of the perineum during vaginal
delivery, independent of whether the perineum remains
intact, or whether there are first- and second-degree lacera-
tions or episiotomy. These results are consistent with other
findings (Menta & Schirmer 2006). However, Leeman et al.
(2007) contended that PFMS six weeks after delivery was
reduced in women with second-degree lacerations, compared
to those who had an intact perineum (53 against 28%,
p = 0Æ03). The results of this study showed that type of
delivery did not significantly disrupt average PFMS or Oxford
scale. Other studies, by contrast, have shown that vaginal
delivery can cause PF damage (Dannecker & Anthuber 2000,
Fitzpatrick & O’Herlihy 2001, Hay-Smith et al. 2002, Souza
2002, Norton et al. 2006).
A study that analysed the influence of delivery type on
PFMS using digital vaginal palpation and perineometry in a
cohort of 30 women who had never been pregnant and 64
first-time mothers (32 postcaesarean and 32 postvaginal

 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433 2431
A de Souza Caroci et al.
delivery) between four and six months after delivery
concluded that vaginal and caesarean delivery both reduced
PFMS; however, PFMS was lower among women who had
vaginal births in comparison with those who had caesareans
or who had never been pregnant. The authors suggested that
routine episiotomy and not just vaginal delivery could
decrease PFMS, because all women who had vaginal births
underwent episiotomy (Barbosa et al. 2005).
One important aspect that should be mentioned is the fact
that the participants in this study readily accepted both
methods of PFMS evaluation. Of all of the women enrolled
in our study, only four (1Æ8%) refused to continue in the
study. The methods used to evaluate PFM were selected
because they are the most widely used in clinical practice
(Coletti et al. 2005).
To summarise, the results of PFMS evaluation during
pregnancy and postpartum vary rather reasonably among
the studies. We consider that this variation is mainly because
of the different inclusion criteria, population following
along the study and the PFMS measuring methods.
It is worth reiterating that the sample estimate was not
reached because of loss of follow-up of women. Thus, this
limitation should be considered when interpreting the results
of this study. Although our sample size is greater than that of
similar studies, a larger n would allow more accurate
analysis of the interaction between the variables that
influence the PFMS. Therefore, the researchers of this study
recommend that more studies be performed that follow
women for a longer period after delivery to evaluate genito-
urinary tract symptoms, sexual dysfunction and PFMS
variation.
Conclusion
The results obtained by this study lead to the conclusion that
there was no significant variation among PFMS in early and
late pregnancy and postpartum. Furthermore, maternal age,
skin colour, conjugal status, dyspareunia and stool charac-
teristics did not affect PFMS during pregnancy or postpar-
tum. Neither did delivery type (vaginal or caesarean) or
condition of the perineum (intact, spontaneous perineal
laceration, or episiotomy) influence postpartum PFMS.
However, the interaction between nutritional state and
newborn’s weight may affect the values of perineometry.
Relevance to clinical practice
The perineometry and digital vaginal palpation are valid
methods to assess the PFMS, with good acceptance by
women. Digital vaginal palpation is a simple method, which
2432  2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
does not require special equipment, but the professional, who
uses it, must be adequately prepared to assess the PFMS. In
clinical practice, this method is effective to support the
diagnosis of urinary, intestinal and sexual dysfunctions.
Regarding perineometry, its use is particularly important
together with the performance of perineal exercises with
biofeedback in the treatment of these disorders.
Acknowledgements
This work was supported by the State of São Paulo Research
Foundation (FAPESP), São Paulo (SP), Brazil. We also thank
School of Nursing of the Universidade de Sao Paulo and
Adventist University Center of Sao Paulo.
Source of support
This study was partially funded by the State of São Paulo
Research Foundation (FAPESP), São Paulo (SP), Brazil. This
funding involved resources for the purchase of equipment and
supplies, transportation, statistical analysis and technical
training scholarships.
Contributions
Study design and manuscript preparation: ASC, MLGR; data
collection and analysis: ASC, MLGR, WSS, ACC, EMS,
NRL, CNCF.
Conflict of interest
None.
References
Allen RE, Hosker GL, Smith ARB & Warrell DW (1990) Pelvic floor
damage and childbirth: a neurophysiological study. British Journal
of Obstetrics and Gynaecology 97, 770–779.
Atalah ES, Castillo LC, Castro SR & Aldea PA (1997) Propuesta de
un nuevo estandar de evaluacion nutricional en embarazadas.
Revista Medica de Chile 125, 1429–1436.
Barbosa AMP, Carvalho LR, Martins AMVC, Calderon IMP &
Rudge MVC (2005) Efeito da via de parto sobre a força muscular
do assoalho pélvico. Revista Brasileira de Ginecologia e Obstetrı́-
cia 27, 677–682.
Belo J, Francisco E, Leite H & Catarino A (2005) Reeducação do
pavimento pélvico com cones de Plevnik em mulheres com incon-
tinência urinária. Acta Medica Portugesa 18, 117–122.
Bø K & Larsen S (1990) Classification and characterisation of
responders to pelvic floor muscle exercise for female stress
urinary incontinence. Neurourology and Urodynamics 9, 395–
396.
Maternal and neonatal health A cohort study of pelvic floor musculature function during pregnancy and postpartum
Braekken IH, Majida M, Engh ME & Bø K (2009) Test-retest reli-
ability of pelvic floor muscle contraction measured by 4D ultra-
sound. Neurourology and Urodynamics 28, 68–73.
Cammu H, Van Nylen M, Blockeel C, Kaufman L & Amy jj (2004)
Who will benefit from pelvic floor muscle training for stress uri-
nary incontinence? American Journal of Obstetrics and Gynecol-
olgy 191, 1152–1157.
Castro RA, Arruda RM, Zanetti RD, Santos PD, Sartori MGF &
Girão MJBC (2008) Single-blind, randomised, controlled trial of
pelvic floor muscle training, electrical stimulation, vaginal cones
and no active treatment in the management of stress urinary
incontinence. Clinics 63, 465–472.
Coletti SH, Haddad JM & Barros JPF (2005) Avaliação Funcional do
Assoalho Pélvico. In Reabilitação do Assoalho Pélvico nas Funções
Urinárias e Anorretais (Amaro JL, Haddad JM, Trindade JCS &
Ribeiro RM eds). Segmento Farma, São Paulo, Brazil, pp. 67–75
Costa Neto PLO (1992) Estatı́stica. Edgard Blücher, São Paulo,
Brazil.
Dannecker C & Anthuber C (2000) The effects of childbirth on the
pelvic floor. Journal of Perinatal Medicine 28, 175–184.
Dawson B & Trapp R (2003) Bioestatı́stica básica e clı́nica.
McGraw-Hill, São Paulo, Brazil.
Diokno AC, Sampselle CM, Herzog AR, Raghunathan TE, Hines S,
Messer KL, Karl C & Leite MC (2004) Prevention of urinary
incontinence by behavioral modification program: a randomized,
controlled trial among older women in the community. Journal of
Urology 171, 1165–1171.
Dumoulin C, Gravel D, Bourbonnais D, Lemieux MC & Morin M
(2004) Reliability of dynamometric measurements of the pelvic
floor musculature. Neurourology and Urodynamics 23, 134–142.
Ferrán Aranz M (2001) SSPS para Windows: análisis estadı́stico.
Mc Graw-Hill, Madri, Spain.
Fischer W & Linde A (1997) Pelvic floor findings in urinary incon-
tinence: results of conditioning using vaginal cones. Acta Obste-
tricia et Gynecologica Scandinavica 76, 455–460.
Fitzpatrick M & O’Herlihy C (2001) The effects of labour and
delivery on the pelvic floor. Best practice & Research: Clinical
Obstetrics & Gynaecology 15, 63–79.
Fleming N, Newton ER & Roberts J (2003) Changes in postpartum
perineal muscle function in women with and without episiotomies.
Journal of Midwifery & Women’s Health 48, 53–59.
Hair Junior JF, Anderson RE, Tathan RL & Black WC (1998) Multi-
variate Data Analysis, 5th edn. Prentice Hall, New Jersey, USA.
Hay-Smith J, Herbison P & Mørkved S (2002) Physical therapies for
prevention of urinary and faecal incontinence in adults. Cochrane
Database of Systematic Reviews Issue 2, Art. No.: CD003191.
Herbison P, Plevnik S & Mantle J (2002) Weighted vaginal cones
for urinary incontinence. Cochrane Database of Systematic
Reviews Issue 1, Art. No.: CD002114.
 2010 Blackwell Publishing Ltd, Journal of Clinical Nursing, 19, 2424–2433
Klein MC, Janssen PA, MacWilliam L, Kaczorowski J & Johnson B
(1997) Determinants of vaginal-perineal integrity and pelvic floor
functioning in childbirth. American Journal of Obstetrics and
Gynecology 176, 403–410.
Kolbl H (2001) Pregnancy, childbirth and the pelvic floor. Zen-
tralblatt Gynäkologie 123, 666–671.
Leeman LM, Rogers RG, Greulich B & Albers LL (2007) Do unsu-
tured second-degree perineal lacerations affect postpartum func-
tional outcomes? Journal of American Board of Family Medicine
20, 451–457.
Menta SS & Schirmer J (2006) Relação entre a pressão muscular
perineal no puerpério e o tipo de parto. Revista Brasileira de
Ginecologia e Obstetrı́cia 28, 523–529.
Nagib ABL, Guirro ECO, Palauro VA & Guirro RRJ (2005) Ava-
liação da sinergia da musculatura abdomino-pélvica em nulı́paras
com eletromiografia e biofeedback perineal. Revista Brasileira de
Ginecologia e Obstetrı́cia 27, 210–215.
Norton C, Hosker G & Brazzelli M (2006) Biofeedback and/or
sphincter exercises for the treatment of faecal incontinence in
adults. Cochrane Database of Systematic Reviews Issue 3, Art.
No.: CD002111.
Oliveira C, Lopes MAB, Longoe Pereira LCL & Zugaib M (2007)
Effects of pelvic floor muscle training during pregnancy. Clinics 62,
439–446.
Örnö AK, Marsál K & Herbst A (2008) Ultrasonographic anatomy
of perineal structures during pregnancy and immediately following
obstetric injury. Ultrasound in Obstetrics and Gynecology 32,
527–534.
Souza ELBL (ed) (2002) A Reeducação da Musculatura do Assoalho
Pélvico como Método de Tratamento Conservador da Inconti-
nência Urinária. In Fisioterapia Aplicada à Obstetrı́cia: Aspectos
de Ginecologia e Neonatologia, 3rd edn. MEDSI, Rio de Janeiro,
Brazil, pp. 391–401.
Valadares JD (2002) Adaptações Fisiológicas da Gestação. In Fisio-
terapia Aplicada à Obstetrı́cia: Aspectos de Ginecologia e Neo-
natologia, 3rd edn (Souza ELBL ed.). MEDSI, Rio de Janeiro,
Brazil, pp. 19–32.
Voorham-van der Zalm PJ, Lycklama à Nijeholt GAB, Elzevier HW,
Putter H & Pelger RCM (2008) Diagnostic investigation of the
pelvic floor: a helpful tool in the approach in patients with com-
plaints of micturition, defecation and/or sexual dysfunction.
Journal of Sexual Medicine 5, 864–871.
Wyman JF, Harkins SW & Fantl JA (1990) Psychosocial impact of
urinary incontinence in the community-dwelling population.
Journal of American Geriatrics Society 38, 282–288.
Zanetti MRD, Castro RA, Rotta AL, Santos PD, Sartori M & Girão
MJBC (2007) Impact of supervised physiotherapeutic pelvic floor
exercises for treating female stress urinary incontinence. Sao Paulo
Medical Journal 125, 265–269.
2433