Racial differences in pelvic morphology among asymptomatic ...

American Journal of Obstetrics and Gynecology (2005) 193, 2035–40
www.ajog.org
Racial differences in pelvic morphology among
asymptomatic nulliparous women as seen on
three-dimensional magnetic resonance images
Lennox Hoyte, MD, MSEECS, a John Thomas, MD, b Raymond T. Foster, MD, b
Susan Shott, PhD, c Marianna Jakab, MSEECS, a Alison C. Weidner, MD b
Harvard Medical School, a Boston, MA; Duke University Medical Center, b Durham, NC; Rush University, c
Chicago, IL
Received for publication January 6, 2005; revised May 15, 2005; accepted June 14, 2005
KEY WORDS
Magnetic resonance
imaging
Levator ani
Pelvic morphology
Racial differences
Three-dimension
reconstruction
Objective: Compare pelvic morphology between asymptomatic African-American and white
nulliparous women.
Study design: Resting supine T2-weighted magnetic resonance (MR) images were obtained in 12
African-American (AA) and 10 white American (WA) women without pelvic floor dysfunction.
Three-dimensional models were reconstructed from the MR images by a masked investigator, and
predefined bony and soft tissue pelvic floor parameters were measured and compared.
Nonparametric statistics were used, with significance considered at P ! .05.
Results: Subjects were similar in age and body mass index. Levator ani volume was significantly
greater in the AA versus the WA group (mean = 26.8 vs 19.8 cm 3 , P = .002). The levatorsymphysis
gap was smaller in the AA (left-18.2, right-18.8 mm) versus the WA group (22.4, 22.6
mm, P = .003, .048) on the left and right. Significant differences were seen in bladder neck
position, urethral angle, and the pubic arch angle.
Conclusion: The increased muscle bulk and closer puborectalis attachment seen among the
African-American nulliparous women may impact the development of pelvic floor dysfunction.
These findings need further study.
Ó 2005 Mosby, Inc. All rights reserved.
Female pelvic floor dysfunction (PFD) includes urinary
incontinence, pelvic organ prolapse (POP), fecal
incontinence, and defecatory dysfunction. PFD is
thought to stem from pelvic muscle, nerve, and connective
tissue damage sustained during vaginal childbirth. 1,2
Supported by NIH R01 HD38661 and NIH: P01 CA67165-06.
Presented at the 31st Annual Meeting of the Society of Gynecologic
Surgeons, April 4-6, 2005, Rancho Mirage, CA.
Reprints not available from the author.
Notably, approximately 75% of the 4 million annual US
births are delivered vaginally. 3,4
There is epidemiologic evidence of a relationship
between childbirth and PFD. 5 However, not all women
who undergo vaginal childbirth develop PFD. 6 Further,
not all nulliparous women are free of PFD. 7 In addition,
some have observed anecdotally that urinary incontinence
and POP may occur less often in women of
African descent, when compared with white women. 8,9
Work by Sze et al 10 and Handa et al 11 suggested that
0002-9378/$ - see front matter Ó 2005 Mosby, Inc. All rights reserved.
doi:10.1016/j.ajog.2005.06.060

2036 Hoyte et al
Figure 1 A, Reconstructed dorsal lithotomy view of the bony
and soft tissue pelvis. White: Pelvic bones; pink: vagina;
yellow: urethra; brown: levator; gray: symphysis coccyx;
dark blue: anal sphincter complex; light blue: rectum. B,
Bony pelvic measures: C-F: Pubococcygeal line; B-D: intertuberous
distance; A-E: interspinous distance. The angle (B-C-
D) is the pubic arch angle. The arrows point to the acetabulae,
which are the boundaries of the interacetabular distance.
Figure 2 A, 3D soft tissue measures: The levator hiatus
height is shown by the vertical white line. The levator hiatus
width is shown by the horizontal black line. B, The levator
symphysis gap (LSG). The right LSG is given by the white line,
and the left by the black line.
bony pelvic shape, rather than race, may be a factor in
the development of PFD among parous women. Therefore,
we sought to determine whether there were bony
and soft tissue differences in pelvic morphology between
well-characterized, asymptomatic nulliparous white and
African-American women. Magnetic resonance-based 3-
dimensional (3D) reconstruction techniques were used
as the imaging modality because of the superior tissue
definition and high resolution of magnetic resonance
imaging (MRI), coupled with the improved visualization
of spatial relationships offered by 3D techniques.
MRI has greatly improved our ability to study the
organs and tissues of the pelvis in women. Standard 2D
MRI has been used to assess the anatomy of the female
pelvic floor in cadavers and living women. 12,13 Threedimensional
reconstruction offers additional advantage
over 2D MRI because it affords better visual appreciation
of key structures, and superior assessment of volumetric
and plane geometry. 14 Further, 3D reconstruction
can minimize the artifacts possible because of variations
in MR slice acquisition angle across subjects. 15
Material and methods
This is an Institutional Review Board-approved, prospective
observational cohort study, examining MRI
data from 2 groups of nulliparous women: 12 African-
American and 10 white American. All were premenopausal,
and had stage 0 or 1 pelvic support as defined by
the POPQ system, 16 by examination in the supine
straining position. To be eligible for inclusion, each
subject had to deny pelvic floor symptoms of chronic

Hoyte et al 2037
Figure 3 A, Reconstructed side view, illustrating the pubococcygeal line (PCL, black line), the bladder neck to PCL (white line
with double arrows), Bladder neck to symphysis (white line). B, Reconstructed side view, showing the urethral angle, formed by the
symphysis and the long axis of the urethra (white lines).
pain, urinary incontinence, prolapse, or defecatory dysfunction.
Age, height, and weight were recorded.
Before MR scanning, subjects were asked to empty
their bladders. Supine MRI studies were performed as
follows: 3D Fast Spin Echo T2-weighted axial source
images were obtained with the use of a 1.5T magnet
(General Electric Medical Systems, Milwaukee, WI) and
a torso phased-array coil wrapped around the pelvis.
The following imaging parameters were used: repetition
time = 4200 ms, time to echo = 108 ms, 128 phase
encodes, 24-cm field of view, 2-mm slice thickness, no
gap. Pixel dimensions were 0.625 ! 0.625 mm. Total
scan time was 13 minutes per subject.
The MR data were transferred to a Dell Dimension
8250 computer with 2 GB of RAM (Dell Inc., Round
Rock, TX), and an ATI-RADEON 9800 graphics processor.
The 3D Slicer software (www.slicer.org) was
used to display the gray scale images and segmented
label maps. A combination of manual and semiautomatic
techniques were used to segment the axial gray
scale images into anatomically significant organs, specifically
bladder, urethra, levator complex, bony pelvis,
symphysis, and coccyx, and the 3D reconstructions were
made in a manner described in our previous work. 14 The
reconstructed models were then displayed and the
parameters measured onscreen. All MR segmenting
and subsequent analysis of the resulting 3D images
was performed by 1 individual, L.H., who was masked
to subject grouping throughout these procedures.
Bony pelvis parameters
The bony pelvic parameters were measured on the
reconstructed 3D models. Intertuberous distance was
measured from midpoint to midpoint of the iscial tuberosities.
The interspinous distance was measured between
the medial-most tips of the iscial spines. The interacetabular
distance was measured between the medial-most
aspects of the femoral sockets. The pubo-coccygeal line
(PCL) distance was measured as the distance between the
inferior-most aspect of the symphysis pubis and the tip of
the coccyx in the midline. The pubic arch angle was

2038 Hoyte et al
Table I Comparison of bony pelvic parameters between the groups
PCL (mm) PUBArch (deg) InterTub (mm) InterSpi (mm) InterAcet (mm)
AA
Mean 91.4 99.6 119.5 105.2 117.3
Range 77.3-111.5 84.6-125.2 97.7-132.9 92.3-119.9 97.4-127.6
WA
Mean 91.6 86.4 118.0 108.3 122.7
Range 79.5-100.6 78.5-98.2 103.2-134.5 97.4-115.9 110.2-132.5
P NS .006 NS NS NS
PCL is the anteroposterior distance from symphysis to tip of coccyx. It is a measure of the depth of the pelvic outlet. PUBArch is the angle of the pubic
arch. InterTub is the transverse distance between the middle if the iscial tuberosities. It is a measure of the width of the pelvic outlet. InterSpi is the
distance between the tips of the iscial spines. It is a marker for the width of the midpelvis. InterAcet is the distance between the medial-most aspects of
the acetabular sockets. It is a marker for the width of the pelvic inlet. AA, African American; WA, White American.
Table II Comparison of levator muscle parameters between the groups
LHH (mm) LHW (mm) LevVOL (cm 3 ) LSGL (mm) LSGR (mm)
AA
Mean 56.8 31.0 26.8 18.2 18.8
Range 44.2-72.7 20.1-36.1 19.0-33.9 14.6-22.4 11.3-21.9
WA
Mean 63.9 31.2 19.8 22.4 22.6
Range 56.4-70-2 24.3-36.8 13.5-29.1 18.6-30.1 18.6-30.5
P .056 NS .002 .003 .048
LHH is the distance from the inferior symphysis to the levator median raphe. LHW is the maximal transverse distance of the levator hiatus. LevVOL is the
volume of the levator ani complex. LSG is the distance from the inferior mid pubic symphysis to the nearest occurrence of the puborectalis muscle in the
left (LSGL), and right (LSGR). AA, African American; WA, White American.
Table III Urethral and bladder parameters
BNPCL (mm) BNSym (mm) UreAng (deg) Urethra (mm) BVOL (cm 3 )
AA
Mean 18.11 19.0 49.2 31.6 37.7
Range 7.8-28.2 11.5-22.0 40.6-62.9 23.0-40.3 12.3-138.5
WA
Mean 9.7 16.5 56.4 26.2 39.6
Range 6.8-15.0 11.6-21.1 43.7-71.7 18.1-31.8 4.9-113.3
P .001 .048 .036 .036 NS
BNPCL is the perpendicular distance from bladder neck to pubococcygeal line-positive values lie superior to the PCL, negative values lie inferior to the
PCL. BNSym is the distance between the bladder neck and the pubic symphysis. UreAng is the angle formed by the long axes of the pubic symphysis and
the urethra. Urethra is the length of the urethra along its long axis. BVOL is bladder volume. All linear measures are in millimeters, volumes are in cubic
centimeters, and angles are in degrees. AA, African American; WA, White American.
measured as the angle subtended by the inferior pubic
rami. The standard diagonal and obstetric conjugates
were not available for measurement on the 3D reconstructions
because the source axial scans did not reach the
sacral promontory, disallowing reconstruction of that
landmark. A dorsal lithotomy view of a reconstructed
bony pelvis and soft tissues is shown in Figure 1, A. Bony
measurements are illustrated in Figure 1, B.
Soft tissue parameters
The levator hiatus height was measured in the midline as
the distance from the inferior-most aspect of the pubic
symphysis to the inner aspect of the levator median
raphe. The levator hiatus maximum width was measured
as the maximal transverse distance of the innermost
surface of the anterior levator opening. The levatorsymphysis
gap was measured as the distance from the
middle of the inferior symphysis to the nearest aspect of
the puborectalis muscle on the right and left. The levator
measures are illustrated in Figure 2, A and B. The
bladder neck to PCL distance (BNPCL) was measured
as the perpendicular distance from the posterior aspect
of the bladder neck (ie, the junction of the urethra and
bladder) to the PCL. Negative values of BNPCL imply a
position inferior (caudad) to PCL, positive values of

Hoyte et al 2039
BNPCL imply positions superior (cephalad) to PCL.
The bladder neck to symphysis was measured as the
distance from the anterior aspect of the bladder neck to
the closest point on the symphysis. The bladder neck
parameters are illustrated in Figure 3, A. The urethral
angle was measured as the angle subtended by the long
axis of the symphysis pubis, and the long axis of the
urethra, as illustrated in Figure 3, B. The urethral length
was measured as the linear distance from distal to
proximal tips of the urethra. Bladder volume, urethral
volume, and levator volume were computed as the sums
of the area on each slice multiplied by the interslice
distance. Levator shape was defined as a ‘‘U’’ if the
levator moved laterally, then superiorly from its low
point at the median raphe in the midline. Levator shape
was defined as a ‘‘V’’ if it moved superiorly then outward
from its low point at the median raphe in the midline.
The SPSS statistical package was used for all statistical
analysis (SPSS, Chicago, IL). Nonparametric Mann-
Whitney tests were applied after review of frequency
histograms that demonstrated non-normal data. Twotailed
statistical significance is considered at P ! .05.
Results
Subjects did not differ by age; African-American women
had a mean age GSD of 30.13 G 6.3 years versus white-
American women at 26.4 G 4.4 years, P = NS. Similarly,
although AA subjects had a slightly higher body
mass index (BMI) at 26.5 (7.5) kg/m 2 versus WA
subjects at 23.9 (3.3) kg/m 2 , this difference was not
statistically significant.
Of the bony pelvic measures, only the pubic arch
angle was found to be significantly different between the
groups. The pubic arch angle was larger by 13 degrees in
AA women versus WA women. Bony pelvic comparisons
are detailed in Table I.
Levator ani volume was significantly higher in AA
women, who also had a significantly shorter levatorsymphysis
gap bilaterally, compared with the WA
women. The bladder neck was closer to the symphysis
in the AA women, and the angle of the urethra to the
symphysis was also smaller in this group. These differences
were statistically significant. Pelvic soft tissue
comparisons are detailed in Tables II and III.
Figure 4 shows representative 3D reconstructions
from an AA and WA nulliparous women, illustrating
the soft tissue differences between the 2 groups.
Comment
Our study data shows increased levator ani muscle bulk
among AA nulliparous women when compared with
age-matched WA nulliparous women. In addition, the
arms of the puborectalis muscle were carried closer to
Figure 4 Representative reconstructions from an African-
American (upper panel), and a white American (lower panel)
subject. Note the bulkier levator, carried closer to the symphysis
(on the left), and the smaller urethral angle (on the
right) in the African-American subject.
the superior pubic rami bilaterally, suggestive of a
longer, denser attachment of the levator muscle to the
arcus tendineus levator ani. The bladder neck is held
higher and closer to the symphysis in the AA women
than in the WA women, and the pubic arch is slightly
wider among the AA women. Taken collectively, these
differences suggest a levator ani complex in AA subjects
that is more intimately associated with its connective
tissue and bony attachments.
Previous investigators have reported striking differences
in the prevalence and incidence of PFD among
different races. 8,17-22 There is, however, no consensus in
the medical literature on exactly how race impacts the
development of PFD. Our data in this preliminary study
suggest certain hypotheses: it is possible that the kind of
musculoskeletal configuration we have observed in AA
women, coupled with a favorably wide pubic arch, is
protective against pelvic floor obstetric injury. For
instance, perhaps a bulkier levator muscle means that
an AA woman can have a more extensive denervation
injury before becoming symptomatic. Alternatively,

2040 Hoyte et al
denser and longer muscle-to-connective tissue attachments
may tolerate the same length of tear injury before
resulting in symptoms in AA parturients. It is interesting
that we found a wider pubic arch in our AA subjects,
which is inconsistent with the classic characterization of
an anthropoid pelvic shape in such women. However,
the clinical significance of the 13-degree difference seen
in the pubic arch is unclear. We speculate that our
observations are in keeping with those of Howard et al 23
and Baragi et al 24 who noted more dynamic muscular
attachments and a smaller overall pelvic floor area in
AA women. A smaller posterior pelvis was responsible
for the latter finding. Both of these characteristics would
tend to be protective against PFD by providing a
stronger muscle traversing a smaller pelvic floor, which
would be less vulnerable to the stresses of daily life and
function in our bipedal position. It may appear counterintuitive
to propose a smaller pelvic floor as being
protective against PFD, because vaginal delivery is
perhaps the most consistent risk factor associated with
the disease, until one considers that AA neonates are
consistently of smaller average birth weight than CA
neonates (http://www.marchofdimes.com/peristats).
Morphologic differences in the bony pelvis between the
sexes and among different races have been described 25 and
are useful in archeology and forensic science. If one
assumes, based on epidemiologic evidence, 8,17-22 that race
is a predisposing factor for PFD, then bony and soft tissue
pelvic morphology may be critical factors that promote
PFD in one race more than another. Previous investigations
have used modern imaging techniques to compare
pelvic morphology in women with and without PFD. 10,11
These studies suggest bony parameters that place women
at risk for PFD independent of race.
Although our pilot data failed to show many significant
differences in bony pelvic morphology between the
African-American and white groups, the differences
noted in soft tissue characteristics may help to account
for some of the differences in PFD prevalence between
parous African-American and white women. We are
presently conducting a larger study to confirm these
findings.
References
1. Snooks SJ, Swash M, Mathers SE, Henry MM. Effect of vaginal
delivery on the pelvic floor: a 5-year follow-up. Br J Surg 1990;77:
1358-60.
2. 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.
3. Menacker F, Curtin SC. Trends in cesarean birth and vaginal birth
after previous. Natl Vital Stat Rep 2001 2001;49:1-16.
4. Martin JA, Park MM, Sutton PD. Births: preliminary data for
2001. Natl Vital Stat Rep 2002 2002;50:1-20.
5. Hendrix SL, Clark AM, Nygaard I, Aragaki A, Barnabei V,
McTiernan A. Pelvic organ prolapse in the Women’s Health
Initiative: Gravity and gravidity. Am J Obstet Gynecol 2002;186:
1160-6.
6. Rortveit G, Daltveit AK, Hannestad YS, Hunskaar S. Urinary
incontinence after vaginal delivery or cesarean section. N Engl J
Med 2003;348:900-7.
7. Buchsbaum GM, Chin M, Glantz C, Guzick D. Prevalence of
urinary incontinence and associated risk factors in a cohort of
nuns. Obstet Gynecol 2002;100:226-9.
8. Bump RC. Racial comparisons and contrasts in urinary incontinence
and pelvic organ prolapse. Obstet Gynecol 1993;81:421-5.
9. Sampselle CN, Harlow SD, Skurnick J, Brubaker L, Bondarenko
I. Urinary incontinence predictors and life impact in ethnically
diverse perimenopausal women. Obstet Gynecol 2002;100:1230-8.
10. Sze EHM, Kohli N, Miklos JR, Roat T, Karram MM. Computed
tomography comparison of bony pelvis dimensions between
women with and without genital prolapse. Obstet Gynecol 1999;
93:229-32.
11. Handa VL, Pannu HK, Siddique S, Gutman R, VanRooyen J,
Cundiff G. Architectural differences in the bony pelvis of women
with and without pelvic floor disorders. Obstet Gynecol 2003;
102:1283-90.
12. Strohbehn K, Ellis J, Strohbehn JA, Delancey JOL. Magnetic
resonance imaging of the levator ani with anatomic correlation.
Obstet Gynecol 1996;87:277-85.
13. Tunn R, Delancey JOL, Howard D, Ashton-Miller J, Quint LE.
Anatomic variations in the levator ani muscle, endopelvic fascia,
and urethra in nulliparas evaluated by magnetic resonance imaging.
Am J Obstet Gynecol 2003;188:116-21.
14. Hoyte L, Schierlitz L, Zou K, Flesh G, Fielding JR. Two and 3
dimensional MRI comparison of levator ani structure, volume and
integrity in women with stress incontinence and prolapse. Am J
Obstet Gynecol 2001;185:11-9.
15. Hoyte L, Ratiu P. Linear measurements in 2-dimensional pelvic
floor imaging: the impact of slice tilt angles on measurement
reproducibility. Am J Obstet Gynecol 2001;185:537-44.
16. Bump RC, Mattiasson ABK, Brubaker LP, DeLancey JO,
Klarskov P, Shull BL, et al. The standardization of terminology
of female pelvic organ prolapse and pelvic floor dysfunction. Am J
Obstet Gynecol 1996;175:10-7.
17. Duong TH, Korn AP. A comparison of urinary incontinence
among African American, Asian, Hispanic, and white women. Am
J Obstet Gynecol 2001;184:1083-6.
18. Grodstein F, Fretts R, Lifford K, Resnick N, Curhan G. Association
of age, race, and obstetric history with urinary symptoms
among women in the Nurses’ Health Study. Am J Obstet Gynecol
2003;189:428-34.
19. Hendrix SL, Clark A, Nygaard I, Aragaki A, Barnabei V,
McTiernan A. Pelvic organ prolapse in the women’s health initiative:
gravity and gravidity. Am J Obstet Gynecol 2002;186:1160-6.
20. Klingele CJ, Carley ME, Hill RF. Patient characteristics that are
associated with urodynamically diagnosed detrusor instability
and genuine stress incontinence. Am J Obstet Gynecol 2002;186:
866-8.
21. Knobel J. Stress incontinence in the black female. S Afr Med J
1975;49:430-2.
22. VanDongen L. The anatomy of genital prolapse. S Afr Med J
1981;60:657-9.
23. Howard D, Delancey JO, Tunn R, Ashton-Miller JA. Racial
differences in the structure and function of the stress urinary
continence mechanism. Obstet Gynecol 2000;95:713-7.
24. Baragi RV, DeLancey JOL, Caspari R, Howard DH, Ashton-
Miller JA. Differences in pelvic floor area between African-American
and European American women. Am J Obstet Gynecol
2002;187:111-5.
25. Patriquin ML, Loth SR, Steyn M. Sexually dimorphic pelvic
morphology in South African whites and blacks. Homo
2003;53:255-62.