Ultrasound Obstet Gynecol 2007; 30: 81–85
Published online 14 May 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.4027
Pelvic floor function in elite nulliparous athletes
J. A. KRUGER*, H. P. DIETZ† and B. A. MURPHY*
*Department of Sport and Exercise Science, University of Auckland, Auckland, New Zealand and †Western Clinical School, Nepean
Clinical School, University of Sydney, Penrith, NSW, Australia
K E Y W O R D S: 3D ultrasound; pelvic floor; translabial ultrasound; women athletes
ABSTRACT
Objective There is preliminary evidence linking longterm participation in high-impact exercise with poor
performance in labor and increased incidence of stress
urinary incontinence, which may be due to altered pelvic
floor function. Recent work has shown that HIFIT
(high-impact, frequent intense training) athletes have an
increased cross-sectional area of the levator ani muscle
group as visualized using magnetic resonance imaging
(MRI). The aim of this study was to further characterize
pelvic floor muscle function and pelvic organ descent in a
nulliparous athletic population and compare it with nonathletic controls matched for age and body mass index,
using three-dimensional/four-dimensional (3D/4D) pelvic
floor ultrasound imaging.
Methods In this prospective comparative study translabial ultrasound imaging was used to assess pelvic floor
anatomy and function in 46 nulliparous female volunteers
(aged 19–39 years), 24 HIFIT and 22 controls. Twodimensional (2D) and 3D translabial ultrasonography
was performed on all subjects, after voiding and in
the supine position. Descent of the pelvic organs was
assessed on maximum Valsalva maneuver, whilst volume
datasets were acquired at rest, during pelvic floor muscle
contraction and during a Valsalva maneuver. Participants
performed each maneuver at least three times and the
most effective was used for evaluation.
Results HIFIT athletes showed a higher mean diameter of
the pubovisceral muscle (0.96 cm vs. 0.70 cm, P < 0.01),
greater bladder neck descent (22.7 mm vs. 15.1 mm,
P = 0.03) and a larger hiatal area on Valsalva maneuver
(21.53 vs. 14.91 cm2 , P = 0.013) compared with the
control group. There were no significant differences in
hiatal area at rest or on maximal voluntary contraction
between the two groups.
Conclusion HIFIT athletes show significant differences
in several of the measured parameters for both function
and anatomy of the pelvic floor. Further research into
the impact of this altered function on childbirth and
continence mechanisms is needed. Copyright  2007
ISUOG. Published by John Wiley & Sons, Ltd.
INTRODUCTION
The role of pelvic floor muscles in supporting the pelvic
organs, continence and childbirth has been investigated
in several studies1 – 3 . Changes in the function and
morphology of the pelvic floor following vaginal delivery,
incontinence and prolapse have been demonstrated using
magnetic resonance imaging (MRI) and, more recently,
three-dimensional (3D) ultrasound imaging4 – 7 . Several
authors have suggested that high impact (landing) sport
may lead to the development of stress incontinence in
nulliparous women who have competed for prolonged
periods8 – 11 , suggesting that there may be a functional
change in the pelvic floor muscles of highly athletic
women. Additionally there is preliminary evidence to
suggest that some women undergoing high-impact,
frequent intense training (HIFIT) have unexpected
difficulties during delivery, which may be attributable
to changes in pelvic floor muscles12 . It is possible that
repeated high impact landing leads to alterations in
pelvic floor morphology and function. Sapsford and
Hodges13 have shown that the muscles of the pelvic
floor are co-activated with the abdominal muscles,
particularly transversus abdominis, during exercise or
increases in intra-abdominal pressure. The abdominal
muscles contract to stabilize the torso during limb
movements and landing. Therefore, it is highly probable
that the pelvic floor would be activated during landing in
athletes involved in high-impact sport.
‘Levator ani’ is the collective term used to describe the
deep muscles of the pelvic floor. The levator ani consists
primarily of the striated muscles pubococcygeus, puborectalis and iliococcygeus14 . As there is some controversy
about the nomenclature for the muscles of the pelvic floor,
for the purposes of this paper we have adopted the terms
defined by DeLancey1 , whereby ‘pubovisceralis muscle’ is
Correspondence to: J. A. Kruger, Department of Sport and Exercise Science, Tamaki Campus, Auckland University, P.O. Box 92019,
Auckland, New Zealand (e-mail: [email protected])
Accepted: 25 January 2007
Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.
ORIGINAL PAPER
82
used to represent both the musculus pubococcygeus and
musculus puborectalis, as these two muscles cannot be
differentiated when using ultrasound or MRI. The levator
hiatus is defined as the area bounded by the pubovisceralis
muscle laterally and dorsally and the inferior pubic rami
anteriorly.
Although it is known that long-term resistance training
causes hypertrophy of skeletal muscle15 , very little
research has been done on the effects of long-term, highimpact training on the function and morphology of the
muscles of the pelvic floor in elite female athletes.
We have previously demonstrated that there is an
increase in the cross-sectional area of the levator ani
muscle as measured by MRI in women who have participated in long-term high-impact sport and exercise16 .
What is not known is whether this altered morphology
results in altered function. Recent developments in 3D
ultrasonography have allowed for characterization of the
function and morphology of the pelvic floor in nulliparous women17 – 20 . This study utilizes this technique to
compare morphology and function of the pelvic floor
in HIFIT nulliparous women to an age-matched control
group. Our hypothesis was that the HIFIT group would
have increased size and altered function of the levator ani
complex as compared with the control group.
METHODS
Translabial three-dimensional and four-dimensional
(3D/4D) ultrasonography was used to assess pelvic
floor function in 46 nulliparous female volunteers (aged
19–39 years); 24 were HIFIT athletes and 22 were controls. Recruitment was primarily from advertisement at
local university campuses, whereby the participants were
matched according to age and body mass index (BMI)
scores. All the participating athletes were active in their
various sports and at the time of testing had been training
for a minimum of 5 years and had reached a national or
international level of competition. All training and sports
had to include a large component of high-impact (repetitive landing/jumping) and high-intensity training, as it
was this aspect of the sport that was thought to lead to
pelvic floor muscle change. Sports which were considered
appropriate and applicable included sport aerobics, running, netball, basketball, squash, tennis and gymnastics.
The participants who acted as controls did not exercise
at all, or if they were active it was not more than three
times a week in general ‘stretching classes’ at the gym,
or walking or swimming. They did not participate in any
high-impact activities.
Based on previous work in which the mean hiatal area
on voluntary Valsalva maneuver in a group of normal
nulliparous women was 14.05 ± 5.87 cm2 , a large effect
size of 0.8, an alpha of 0.01 (corrected for multiple
comparisons) and a power of 0.8, a power calculation
(G-power) indicated that 19 subjects in each group would
be needed to show any significant differences between the
measured parameters.
Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.
Kruger et al.
The participants were interviewed prior to the ultrasound scan to determine symptoms of incontinence, bowel
dysfunction or prolapse as well as questions on amenorrhea and performance of pelvic floor muscle exercises.
Joint hypermobility, a potential confounder to some
of the measurements, such as bladder neck descent, was
assessed using the Beighton scale, which has been modified
from the Wilkinson–Carter measures, and includes five
distinct characteristics: passive extension of the 5th finger
past 90◦ ; passive apposition of the thumb to the forearm;
hyperextension of the elbow past 190◦ ; hyperextension of
the knee past 10◦ ; and trunk flexion to allow palms flat on
the floor21 . Scores > 6 on the Beighton scale are considered
diagnostic for hypermobility. Hypermobility syndrome is
characterized by an increase in joint laxity and a decrease
in tissue stiffness, due to a change in the ratio of Type I
to Type III collagen. Type III collagen is more dominant
in people with hypermobility syndrome21 . Athletes who
fall into the hypermobile classification may have a natural
predisposition to perform well at sports such as aerobics
and gymnastics, where enhanced flexibility is seen as an
asset, and as such, the HIFIT group could have an overly
high incidence of hypermobile participants. In order to
ensure that hypermobility was not the reason for any
difference in pelvic floor function between athletes and
controls, both groups were assessed retrospectively using
the Beighton scale.
Imaging was performed using GE Kretz Voluson
730/730 Expert systems (GE Kretztechnik GmbH, Zipf,
Austria) with 8–4-MHz transducers. All subjects were
imaged supine, after voiding. Imaging was performed in
the mid-sagittal plane with the angle of acquisition set at
85◦ . Volume datasets were acquired at rest, on pelvic
floor muscle contraction and on Valsalva maneuver.
The most effective of at least three maneuvers was
used for evaluation. Previously published parameters of
pelvic floor function and anatomy were used in both
groups20 . Descent of the pelvic organs was assessed on
Valsalva maneuver. This included bladder neck descent,
cervical descent and descent of the rectal ampulla on
Valsalva maneuver. The levator hiatus was measured
at rest, on Valsalva maneuver and on maximum pelvic
floor contraction. Figure 1 shows the plane used to
determine levator hiatal area. The plane of minimal hiatal
dimensions is identified in the mid-sagittal plane as the
minimal distance between the posterior aspect of the
symphysis pubis and the anorectal angle, or the anterior
border of the pubovisceral muscle (Figure 1a). Once the
plane had been defined in the mid-sagittal plane, this crosssection of the volume was displayed in the axial plane
during post processing. The thickness of the pubovisceral
muscle was determined by moving the plane cranially
until the maximum thickness of the muscle could be seen
in the axial plane. This plane was reached approximately
1 cm above the levator hiatus. A detailed description of
the methodology is given elsewhere20 .
Data acquisition and analysis were undertaken by two
of the authors (J. K. and H. P. D.). Images were saved on
DVD, which allowed for later analysis using the software
Ultrasound Obstet Gynecol 2007; 30: 81–85.
83
Pelvic floor function in athletes
Figure 1 Translabial ultrasound image showing identification of the plane of minimal dimensions (represented by the diagonal line) in the
mid-sagittal plane (a), and the angled axial plane of the levator hiatus (b). Dotted line indicates area measurement of the hiatus at rest.
4D View V 2.1 (GE Medical Kretztechnik). Ethical
approval was granted by the University of Auckland
Human Participants Ethics committee (ref 2004/205).
Statistical analysis was performed using SPSS V 12
for Windows (SPSS Inc., Chicago, IL, USA). t-tests
were used to determine significant differences (with
P < 0.05 taken as significant) between the groups in
the measured parameters, following normality testing.
Intraclass correlation coefficients (ICC) were used to
establish agreement.
RESULTS
All the participants were nulligravid and none had
a history of pelvic surgery. The results from the
questionnaire indicated that all were asymptomatic for
prolapse, three of the HIFIT group (12.5%) reported
stress incontinence and two (8.3%) urge incontinence
more than once a month. Two (9.1%) of the controls
reported both stress and urge incontinence more than
once a month. Seven (29.2%) of the HIFIT women
reported that they thought that they were exercising
pelvic floor muscles during training, particularly when
performing ‘core stability work’ (defined as the voluntary
contraction of the abdominal muscles and the pelvic floor
muscles in an effort to increase spinal stability). The
general characteristics of the HIFIT and control groups
are summarized in Table 1.
Initial measurements were done by the first author
(J. K.) and validated by the second (H. P. D.).
Interobserver reliability indices were calculated for the
biometric parameters of the levator hiatus on a series
of 46 volumes. ICC values ranked between 0.57 and
0.81, with the best agreement for levator hiatal area
on Valsalva. As the intraclass correlations were good to
excellent for all measures of the levator hiatus, the ratings
Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.
Table 1 General characteristics of participants
Group
HIFIT (n = 24)
Control (n = 22)
Mean
age
(years)
28.5
27.6
BMI
(kg/m2 )
Hypermobility
( n) *
Stress
incontinence
(%)
Urge
incontinence
(%)
22.3
23.2
3 >6
None > 6
12.5
9.1
8.3
9.1
*Defined as a score > 6 on the Beighton scale. BMI, body mass
index; HIFIT, high-impact, frequent intense training.
of the more experienced assessor (H. P. D.), who was
also blinded to the groupings, were used for statistical
analysis.
The results of bladder neck descent, uterine descent and
descent of the rectal ampulla relative to the symphysis
pubis on maximum Valsalva are summarized in Table 2.
Mean bladder neck descent for the HIFIT group was
22.7 ± 7.85 mm, compared to 15.1 ± 10.2 mm in the
control group (P = 0.03). Uterine and rectal descent on
Valsalva were more pronounced in the HIFIT group
(lower numbers signifying lower organ position on
Valsalva), but these differences did not reach significance.
The area of the levator hiatus on maximum Valsalva
measured 21.53 ± 9.98 cm2 and 14.91 ± 7.18 cm2 for
the HIFIT and control groups, respectively (P = 0.013).
The spread of data for the HIFIT group was wide, as
demonstrated in Figure 2. With respect to hypermobility,
22 of the 24 participants in the HIFIT group were able
to be contacted and three of those 22 were found to
score greater than 6 on the Beighton scale. None of the 22
control subjects scored greater than 6. The individual data
were examined for the three hypermobile participants.
Two of these three participants showed an increased size
of the levator hiatus on Valsalva, as measured using a
Ultrasound Obstet Gynecol 2007; 30: 81–85.
84
Kruger et al.
Table 2 Biometric indices of the levator hiatus, pelvic organ descent and pubovisceral muscle
Levator hiatal area at rest (cm2 )
Levator hiatal area on PFMC (cm2 )
Levator hiatal area on Valsalva (cm2 )
Pubovisceral muscle diameter (cm)
Bladder descent on Valsalva (mm)
Uterine position on Valsalva (mm)
Rectal position on Valsalva (mm)
HIFIT group
(mean ± SD)
(n = 24)
Controls
(mean ± SD)
(n = 22)
12.71 ± 2.49
10.59 ± 1.71
21.53 ± 9.98
0.96 ± 0.17
22.70 ± 7.85
22.70 ± 17.15
1.04 ± 17.38
12.77 ± 2.43
9.72 ± 2.11
14.91 ± 7.18
0.70 ± 0.11
15.10 ± 10.20
28.70 ± 16.90
5.37 ± 11.36
P
0.727
0.201
0.013*
< 0.010*
0.030*
0.308
0.348
*Statistically significant. HIFIT, high-impact, frequent intense training; PFMC, pelvic floor muscle contraction.
DISCUSSION
Hiatal area on Valsalva (cm2)
60
50
40
30
20
10
0
HIFIT group
Controls
Figure 2 Differences in hiatal area between high-impact, frequent
intense training (HIFIT) group and control group on Valsalva
maneuver. Boxes represent data between the 25th and 75th
percentiles, and the internal line is the 50th percentile. Whiskers
indicate the 5th and 95th percentiles and filled circles are the
outliers.
3.0
Stretch ratio
2.5
2.0
1.5
1.0
0.5
0
5
10
15 20 25 30 35
Subject number (n = 42)
40
45
50
Figure 3 Stretch ratio for control group (!;
, mean value)
,
and high-impact, frequent intense training group ( ;
mean value), showing hypermobility. , hypermobile subjects.
Only 42 of the 46 subjects are represented because the stretch ratio
could not be calculated in all cases due to artifacts on imaging.
ž
°
‘stretch ratio’, which is the ratio of the hiatal area on
Valsalva to the hiatal area at rest (Figure 3).
Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.
This study illustrates a number of functional and
morphological differences in the pelvic floor muscles of
HIFIT women and a matched control group as imaged
using 3D/4D pelvic floor ultrasonography. The results
confirmed our previous MRI findings that the HIFIT
group has significantly higher diameters of the levator
ani muscle. Despite this, HIFIT athletes were able to
markedly increase the area of the levator hiatus during
a voluntary Valsalva maneuver. There may be several
explanations for this, including a probable increased
kinesthetic awareness in the HIFIT group, and hence
the ability to recruit task-specific muscles, as well as
an increased abdominal strength22 and the concomitant
increase in intra-abdominal pressure that high-impact
athletes are able to develop.
Previous studies using 3D ultrasound imaging and MRI
have evaluated functional and morphological parameters
of young nulliparous women. Dietz et al.20 found the
hiatal area in their series of 52 nulligravid women to be
11.25 cm2 at rest and 14.05 cm2 during Valsalva, while
the diameter of the pubovisceral muscle was 0.73 cm.
These findings are remarkably similar to those in the
control group of this study. Tunn et al.23 found the hiatal
area at rest as measured using MRI to be 12.8 cm2
in continent nulliparous women (mean age 30 years),
whereas Goh et al.24 , also using MRI, found the pelvic
hiatal area to be 20.06 cm2 at rest and 27.83 cm2 on
straining in a group of 25 continent women. This group,
however, included parous women and had a mean age
of 34 years. Differences between hiatal area estimates on
3D/4D ultrasound imaging and MRI are not surprising,
as the MRI studies used a different plane of measurement
from the plane of minimal dimensions on Valsalva used by
the ultrasound studies. Inevitably, differences in the hiatal
measurements will result when different planes are used.
Two-dimensional (2D) ultrasonography has been used
to quantify anatomical changes in the pelvic organ mobility of women suffering from stress incontinence and
prolapse7,25,26 . In some of the HIFIT group, ultrasound
measurements of organ descent reached similar values to
those of symptomatic women in terms of descent of the
rectal ampulla, cervix and bladder neck7 . Women with
stress incontinence commonly (although not invariably)
Ultrasound Obstet Gynecol 2007; 30: 81–85.
Pelvic floor function in athletes
show an increase in bladder neck mobility and descent
below the level of the symphysis pubis on Valsalva maneuver. Mobility of the bladder neck in the HIFIT group and
the levator hiatal area were significantly higher than in
our control group and in studies of bladder neck mobility
of nulliparous women27 , yet most of the HIFIT women
were asymptomatic for incontinence. One of the HIFIT
women developed a significant enterocele on Valsalva
but was asymptomatic for clinical symptoms of prolapse.
Intuitively one would expect women with increased pelvic
organ descent to have ‘weak’ pelvic floor muscles. However, this may not be the case, as our HIFIT group showed
a combination of thicker muscle and greater muscle distensibility. The differences observed may be due to differences
in connective tissue or muscle biomechanics that may
predate or be the consequence of high-impact training.
Although there is a known correlation between
hypermobility syndrome and some connective tissue
disorders, including increased risk of pelvic organ
prolapse21 , a recent study by Dietz et al. found no
association between hypermobility and pelvic organ
mobility in a group of nulliparous young women27 . These
findings are similar to those found in our group of HIFIT
women, where there was no correlation between Beighton
score and biometric indices of hiatal size or pelvic organ
mobility.
The demands of improved performance throughout
competitive sport and the concomitant increased training
schedules seem to increase for every generation. Owing
to the benefits of regular body exercise on the whole
body, women cannot and should not be discouraged
from exercising. Nonetheless, an increase in the general
population of women who engage in long-term highimpact sports such as running or aerobics for a significant
number of years prior to childbirth means that it is
important to understand the consequences of long-term
exercise on pelvic floor function. It has been previously
suggested that repetitive high impact may overload the
muscles of the pelvic floor, which may result in stress
injuries to the fascia, muscles or ligaments11 . In light of the
findings from this current study further research into the
biomechanical properties of the pelvic floor and the effects
of frequent, intense high-impact training is warranted.
REFERENCES
1. DeLancey JO. The anatomy of the pelvic floor. Curr Opin
Obstet Gynecol 1994; 6: 313–316.
2. Dannecker C, Anthuber C. The effects of childbirth on the pelvic
floor. J Perinat Med 2000; 28: 175–184.
3. Bo K. Pelvic floor muscle training is effective in treatment of
female stress urinary incontinence, but how does it work? Int
Urogynecol J 2004; 15: 76–84.
4. DeLancey JOL, Kearney R, Chou Q, Speights S, Binno S. The
appearance of levator ani muscle abnormalities in magnetic
resonance images after vaginal delivery. Obstet Gynecol 2003;
101: 46–53.
5. Fielding JR. Practical MR imaging of female pelvic floor
weakness. Radiographics 2002; 22: 295–304.
Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.
85
6. Dietz HP. Levator function before and after childbirth. Aust N
Z J Obstet Gynaecol 2004; 44: 19–23.
7. Dietz H, Haylen B, Broome J. Ultrasound in the quantification
of female pelvic organ prolapse. Ultrasound Obstet Gynecol
2001; 18: 511–514.
8. Thyssen H, Clevin L, Olesen S, Lose G. Urinary incontinence in
elite female athletes and dancers. Int Urogynecol J 2002; 13:
15–17.
9. Nygaard I, Thompson F, Svengalis B, Albright J. Urinary incontinence in elite nulliparous athletes. Obstet Gynecol 1994; 84:
183–187.
10. Bo K, Borgen JS. Prevalence of stress and urge urinary
incontinence in elite athletes and controls. Med Sci Sports Exerc
2001; 33: 1797–1802.
11. Eliasson K, Larsson T, Mattsson E. Prevalence of stress incontinence in nulliparous elite trampolinists. Scand J Med Sci Sports
2002; 12: 106–110.
12. Kruger J, Murphy B. Pelvic floor alterations in elite nulliparous
athletes: implications for labour and delivery. 5th Interdisciplinary World Congress on Low Back and Pelvic Pain,
Melbourne, Australia, 2004.
13. Sapsford RR, Hodges PW. Contraction of the pelvic floor
muscles during abdominal maneuvers. Arch Phys Med Rehab
2001; 82: 1081–1088.
14. Wall N. The muscles of the pelvic floor. Clin Obstet Gynaecol
1993; 36: 911–923.
15. MacDougall J, Sale D, Elder G, Sutton J. Ultrastructural properties of human skeletal muscle following heavy resistance
training & immobilization. Med Sci Sports Exerc 1976; 8:
A72.
16. Kruger J, Murphy B, Heap S. Alterations in levator ani in elite
nulliparous athletes: a pilot study. Aust N Z J Obstet Gynaecol
2005; 45: 42–47.
17. Yang JM, Yang SH, Huang WC. Biometry of the pubovisceral
muscle and levator hiatus in nulliparous Chinese women.
Ultrasound Obstet Gynecol 2006; 28: 710–716.
18. Oerno A, Dietz H. Levator co-activation is an important
confounder of pelvic organ descent on Valsalva. ICS Annual
Scientific Meeting, Christchurch, New Zealand, 2006.
19. Braekken I, Majida M, Engh M, Umek WH, Dietz H, Bo K.
Can 3D ultrasound be used in measurement of pelvic floor
thickness and dimensions of levator hiatus? Int Urogynecol J
2006; 17: S137–S138.
20. Dietz H, Shek C, Clark B. Biometry of the pubovisceral muscle
and levator hiatus by 3D pelvic floor ultrasound. Ultrasound
Obstet Gynecol 2005; 25: 580–585.
21. Russek L. Hypermobility syndrome. Phys Ther 1999; 79: 591.
22. Rivera M, Rivera-Brown A, Frontera W. Health related physical fitness characteristics of elite Puerto Rican athletes. J Strength
Conditioning Res 1998; 12: 199–203.
23. Tunn R, DeLancey J, Howard D, Ashton-Miller J, Quint L.
Anatomic variations in the levator ani muscle, endopelvic fascia,
and urethra in nulliparas evaluated by magnetic resonance
imaging. Am J Obstet Gynecol 2003; 188: 116–121.
24. Goh V, Halligan S, Kaplan G, Healy J, Bartram C. Dynamic
MR imaging of the pelvic floor in asymptomatic subjects. AJR
Am J Roentgenol 2000; 174: 661–666.
25. Schaer G, Koechli O, Schuessler B, Haller U. Perineal ultrasound for evaluating the bladder neck in urinary stress incontinence. Obstet Gynecol 1995; 85: 220–224.
26. Bernstein IT. The pelvic floor muscles: muscle thickness in
healthy and urinary-incontinent women measured by perineal
ultrasonography with reference to the effect of pelvic floor
training. Estrogen receptor studies. Neurourol Urodyn 1997;
16: 237–275.
27. Dietz HP, Eldridge A, Grace M, Clarke B. Pelvic organ descent
in young nulligravid women. Am J Obstet Gynecol 2004; 191:
95–99.
Ultrasound Obstet Gynecol 2007; 30: 81–85.