Blackwell Publishing IncMalden, USAJSMJournal of Sexual Medicine1743-6095© 2006 International Society for Sexual MedicineSeptember/October 200636979987Original ArticleNerve Distribution

in the VaginaPauls et al.

979

ORIGINAL RESEARCH—BASIC SCIENCE

A Prospective Study Examining the Anatomic Distribution of Nerve

Density in the Human Vagina

Rachel Pauls, MD,* George Mutema, MD,† Jeffrey Segal, MD,‡ W. Andre Silva, MD,§
Steven Kleeman, MD,* Vicki Dryfhout, MA,¶ and Mickey Karram, MD*
*Division of Urogynecology and Reconstructive Pelvic Surgery, Good Samaritan Hospital, Cincinnati, OH, †Department of
Pathology, Good Samaritan Hospital, Cincinnati, OH, ‡Saint Barnabas Medical Center—Center for Urogynecology,
Livingston, NJ, §Pacific Northwest Urogynecology—Urogynecology, Federal Way, WA, ¶Hatton Institute for Research and
Education, Good Samaritan Hospital, Cincinnati, OH, USA

DOI: 10.1111/j.1743-6109.2006.00325.x

ABSTRACT

Introduction. Women possess sufficient vaginal innervation such that tactile stimulation of the vagina can lead to
orgasm. However, there are few anatomic studies that have characterized the distribution of nerves throughout the
human vagina.
Aim. The aim of this prospective study was to better characterize the anatomic distribution of nerves in the adult
human vagina. A secondary aim was to assess whether vaginal innervation correlates with the subject’s demographic
information and sexual function.
Methods. Full-thickness biopsies of anterior and posterior vagina (proximal and distal), cuff, and cervix were taken
during surgery in a standardized manner. Specimens were prepared with hematoxylin and eosin, and S100 protein
immunoperoxidase. The total number of nerves in each specimen was quantified. Enrolled patients completed a
validated sexual function questionnaire (Female Sexual Function Index, FSFI) preoperatively.
Main Outcome Measures. A description of vaginal innervation by location and an assessment of vaginal innervation
in association with the subject’s demographic information and sexual function.
Results. Twenty-one patients completed this study, yielding 110 biopsy specimens. Vaginal innervation was some-
what regular, with no site consistently demonstrating the highest nerve density. Nerves were located throughout
the vagina, including apex and cervix. No significant differences were noted in vaginal innervation based on various
demographic factors, including age, vaginal maturation index, stage of prolapse, number of vaginal deliveries, or
previous hysterectomy. There were no correlations between vaginal nerve quantity and FSFI domain and overall
scores. Fifty-seven percent of the subjects had female sexual dysfunction; when compared to those without dysfunc-
tion, there were no significant differences in total or site-specific nerves.
Conclusions. In a prospective study, vaginal nerves were located regularly throughout the anterior and posterior
vagina, proximally and distally, including apex and cervix. There was no vaginal location with increased nerve density.
Vaginal innervation was not associated with demographic information or sexual function. Pauls R, Mutema G,
Segal J, Silva WA, Kleeman S, Dryfhout V, and Karram M. A prospective study examining the anatomic
distribution of nerve density in the human vagina. J Sex Med 2006;3:979–987.
Key Words. Anatomy (Gross and Microscopic); Central Nervous System Control; Neurophysiologic Studies of
Sexual Function

Introduction canalization, and resorption. In the XX fetus,

absence of müllerian-inhibiting factor and test-
he human vagina develops in a stepwise fash- osterone leads to elongation of the uterovaginal
T ion via the processes of elongation, fusion, primordium. This primordium meets the distal

© 2006 International Society for Sexual Medicine J Sex Med 2006;3:979–987

980
urogenital sinus to form the vaginal plate. Central
erosion occurs to create a vaginal lumen, which is
complete by the 20th week of gestation [1].
Although controversial, most studies report that
the müllerian ducts form the upper third of the
vagina whereas the lower two-thirds derive from
the urogenital sinus [2,3].
The neurophysiology of the vagina is poorly
understood. Studies using animal models have
demonstrated that fibers from the pudendal, pel-
vic, and hypogastric nerves innervate the pelvic
organs [4]. The pudendal nerve provides innerva-
tion for the perineum, clitoris, and urethra, and
pelvic nerve sensory fibers innervate the vagina.
These fibers have the greatest concentration in
the vaginal fornix [5]. In the rat vagina, large num-
bers of nerves demonstrating branching toward
the epithelium have been documented distally,
whereas no such fibers were observed in the prox-
imal vagina. Rich innervation of various struc-
tures, especially vaginal arteries, has been noted
[6]. Estrogen levels may play a role in this inner-
vation. Some studies have shown that increased
estrogen leads to expansion of the size and sensi-
tivity of pudendal perineal nerves [4], and causes
alterations in rat vaginal mucosa morphology and
contractility [7]. However, others reported no
change in nerve density based on estradiol admin-
istration, ovariectomy, or estrous cycle in rat vag-
inal tissue [6,8].
Few studies are available using women. Initial
work by Krantz suggested that “ganglion cells”
were in the adventitia surrounding the vagina and
along the lateral walls next to the blood supply [9].
They were seen in the upper third of the vagina
below the bladder. However, only sparse free
nerve endings associated with pain were seen in
the epithelium and muscularis [9]. In the 1960s,
Owman and colleagues [10] evaluated the vaginal
cuff of adult women undergoing hysterectomy.
They described few adrenergic nerve terminals in
these cuff specimens [10]. Later work utilizes
immunohistochemical stains to improve visualiza-
tion of vaginal nerves. Hilliges et al. [11] evaluated
six sites in the vaginal mucosa using protein gene
product 9.5 staining. They reported increased
innervation in the distal and anterior vaginal wall
[11]. A more recent study outlined an analysis of
female fetal specimens. They demonstrated that
branches of the inferior hypogastric plexus extend
on the lateral walls of the vagina, are most dense
on the mid and proximal vagina, and travel supe-
riorly to cover the proximal anterior vaginal wall
[12].
J Sex Med 2006;3:979–987
Pauls et al.
The purpose of vaginal innervation is unclear.
These nerves may be necessary for cervical dila-
tion and subsequent vaginal delivery; altered
cervical innervation in late pregnancy has been
reported [13]. After childbearing is complete, the
vagina functions as an organ for sexual intimacy
and intercourse. Vaginal eroticism has been docu-
mented, with both the anterior and posterior
vagina noted to elicit orgasmic responses [14–16].
Some have commented on a “vaginal pacemaker”
that exists at the upper vagina, is responsible for
vaginal contractile activity, and may represent an
area of erotic sensitivity [17]. Intact neural path-
ways are essential for sexual responses, such as
arousal and lubrication, and subjects with neuro-
logical disorders often have problems with sexual
function [18,19]. Based on these reports, it is pos-
sible that vaginal sensation and sexual responses
are mediated by nerve fibers terminating in the
vaginal mucosa; however, this has not been
described previously. Indeed, although a relation-
ship between pudendal nerve integrity and sexual
function has been suggested [20], there is no sci-
entific evidence that vaginal innervation is corre-
lated with symptoms of sexual function.
The aim of this study was to provide a descrip-
tion of nerve distribution and density throughout
the vagina. A secondary aim was to determine
whether there was an association between vaginal
innervation and the subject’s demographic infor-
mation, such as age, vaginal estrogenization, pre-
vious hysterectomy, and sexual function.
Material and Methods
This was a prospective study of patients undergo-
ing vaginal surgery for prolapse and incontinence
between February and August 2004. Eligibility
criteria included sexual activity (defined as sexual
intercourse in a stable, heterosexual relationship)
in the prior 4 weeks and planned surgery involving
at minimum an anterior and posterior colporrha-
phy. Subjects were excluded if they had diabetes
or any neurological condition, or if surgery was
converted to an abdominal approach. During
these months, nearly 60 patients were scheduled
for this surgery, and approximately half were sex-
ually active. All eligible patients were advised of
the study, and all agreed by written informed con-
sent. A Pelvic Organ Prolapse––Quantification
(POP-Q) examination was performed in the stan-
dard fashion [21], and preoperative urodynamic
testing was completed in subjects with symptoms
of incontinence. As a method of characterizing
Nerve Distribution in the Vagina
the population, the Female Sexual Function Index
(FSFI) was completed by all the subjects preoper-
atively. Institutional Review Board approval was
obtained for this study.
The FSFI is a brief instrument for assessment
of sexual function that consists of 19 questions and
has been validated based on DSM-IV diagnoses of
desire disorder, arousal disorder, and orgasmic
dysfunction [22,23]. Questions are scored for
domains of libido, arousal, lubrication, orgasm,
satisfaction, and pain. Although other validated
questionnaires are available and some validated for
use in subjects with prolapse and/or incontinence
[24], we chose the FSFI because it is relatively
short, is part of our initial patient history forms,
and has been utilized by others in studies of this
population [25]. For this study, female sexual dys-
function (FSD) was defined as a total score of 26
or less (maximum possible score of 36) [26]. In
those subjects with a total score of 26 or less,
additional information was obtained from their
chart to strengthen the diagnosis of sexual dys-
function. In all cases, subjects noted problems with
desire, lubrication, orgasm, or pain elsewhere in
their medical history.
Prior to surgical draping, a vaginal smear was
obtained for maturation index. Maturation index
specimens were taken from the posterior fornix
and labeled with subject identifier.
Full-thickness biopsies were obtained in a stan-
dardized manner during surgical repair by one of
five investigators. To ensure good sampling tech-
nique and adequacy of the stains, specimens were
initially taken from three cadaver specimens and
analyzed in the same fashion. Six potential biopsy
sites were identified: proximal and distal anterior
wall, proximal and distal posterior wall, apex, and
cervix. These sites were chosen to maximize sam-
pling of the entire vagina. Distal specimens were
taken 3 cm inside the hymenal ring, and proximal
specimens halfway between the hymen and vaginal
cuff. Apical specimens were from the vaginal cuff
in patients without a uterus and from the vagina
adjacent to the cervix in subjects undergoing a
hysterectomy. Ectocervical tissue was obtained
after the specimen had been assessed by the
pathology department and was determined to be
free of malignancy. The specimens were taken
from the midline after colporrhaphy incisions
were made, but prior to lateral dissection, using
Mayo scissors. All specimens were color coded for
location and labeled with subject identifier. The
majority had five biopsies with a range of 3–6; 110
biopsies were obtained as described.
981
Vaginal biopsy specimens were fixed in 10% neu-
tral buffered formalin and embedded in paraffin
blocks using conventional histopathologic tech-
niques. Ten horizontal 5-mcm-thick levels were
obtained from each block. Multiple levels were
obtained in an attempt to reduce sampling error
and ensure that the maximum number of nerves
was generated from each specimen. Levels 1, 3,
and 6 were prepared with hematoxylin and
eosin stain for routine microscopy. Levels 2 and 5
were processed for immunohistochemistry accord-
ing to the avidin-biotin-peroxidase complex
method using the Ventana Benchmark-automated
immunostainer (Ventana Medical Systems, Tucson,
AZ) and antibodies against S100 protein (Pre dilute;
Ventana Medical Systems). A negative control
(omission of the primary antibody) and a positive
control (a Schwannoma) were carried out with each
immunohistochemical reaction. Vaginal smears
were fixed in ThinPrep solution and processed with
the ThinPrep 2000 Processor (Cytyc Corp., Box-
borough, MA). The tissue was processed in only
two batches in order to reduce the variances
between procedures. All tissue processing was per-
formed in a clinical lab in conjunction with routine
histology cases. The lab has a daily quality assurance
program as guided by the College of American
Pathologists. All specimens were evaluated concur-
rently by two investigators, one of whom is a Board-
Certified Pathologist, using a double-headed
Olympus BX41 microscope (Center Valley, PA).
Maturation index cytology specimens were
evaluated under high (40×) and low (10×) power
using multiple fields. Subjects with 40% or more
superficial cells were classified as having premeno-
pausal estrogenization, 10–39% superficial cells
were termed intermediate, and those with 9% or
fewer superficial cells in conjunction with para-
basal cells were considered to have no estrogen
effect [27].
S100 immunoperoxidase is a general nerve stain
that has been used in studies of heart, gut, and
vagina [28–30]. In an effort to ensure that the
maximum nerve estimate was generated for each
specimen, a methodology for counting nerves was
devised. Both S100 immunostained levels were
examined under low power to obtain a general
description and locate five fields with the greatest
number of nerves. If required, the hematoxylin
and eosin stains were also utilized to characterize
nerve location. Under high power, these 10 fields
were evaluated and the nerves quantified. Both
longitudinal cross sections and radially oriented
nerves were counted. It was immediately evident
J Sex Med 2006;3:979–987
982 Pauls et al.

Table 1 Demographic characteristics (N = 21)

Age: mean ± SD 50 ± 12.8 years
Race: Caucasian 21 (100%)
Marital status
Married 20 (95%)
Divorced 1 (5%)
Smoker
Yes 1 (5%)
No 20 (95%)
Menopausal status
Premenopausal 10 (48%)
Postmenopausal 11 (52%)
Hormone replacement 10 (91%)
Vaginal maturation index
Premenopausal 6 (29%)
Intermediate 8 (38%)
No estrogen effect 7 (33%)
Previous hysterectomy 9 (43%)
Abdominal 6 (67%)
Vaginal 3 (33%)
Figure 1 S100 immunostaining; large nerve in center of Previous prolapse repair/anti-incontinence 3 (14%)
photograph, small nerves indicated by arrow. procedure
Stress incontinence 12 (57%)
Fecal incontinence 1 (5%)
Vaginal deliveries: mean 2.7
that there were two types of nerves: small and <3 11 (52%)
large. Because the clinical significance of these two ≥3 10 (48%)
types of nerves was unknown, these were assessed Prolapse stage; maximum descent of any compartment
Stage 2 19 (90%)
separately (see Figure 1). An average number of Stage 3 2 (10%)
large, small, and total nerves per high-power field
were then calculated for each specimen. Excluding
cervical biopsies, one investigator was blinded to
vaginal location of the sample and patient charac- have an anterior and posterior colporrhaphy at the
teristics. The blinded investigator was responsible time of their repair, two opted for nonsurgical
for selection of the fields for quantification. treatments, and one had surgery outside the
Statistical analysis was performed with SPSS approved sites. The remaining 21 subjects com-
13.0 for Windows (SPSS Inc., Chicago, IL). pleted the study (Table 1). The mean age of the
Descriptive statistics, including means and fre- subjects was 50 years. All the subjects were parous
quencies, are reported in Tables 1 and 2. Bivariate and Caucasian. Ten subjects were premenopausal
analyses include independent t-tests, chi-squared with normal monthly menstrual cycles, and 11
test or Fisher exact test, Wilcoxon rank sum test postmenopausal. Twelve patients had urodynamic
(Mann–Whitney test), and Pearson’s r correlation. stress incontinence, whereas the majority had
Effect size calculation was completed using n- stage 2 prolapse. Ten subjects underwent a hyster-
Query Version 6.0 (Statistical Solutions Ltd, Los ectomy at the time of their surgery.
Angeles, CA). Vaginal nerve density was assessed at each ana-
tomic location to compare sites to each other, as
well as to determine the site of highest innervation
Results
(Table 2). One subject was noted to have 250
Twenty-nine patients agreed to participate in this nerves noted at the cuff site only; this specimen
study. Of them, eight were excluded: five did not was excluded from analysis as an outlier. Vaginal

Table 2 Vaginal innervation by location and comparisons of anatomic sites

Mean number Comparison of mean number
Location Subject N of nerves ±SD of nerves at location sites
Distal anterior (DA) 21 21.22 ±10.34 PP > DP > DA > PA > Cervix > Cuff
Proximal anterior (PA) 19 21.19 ±7.30
Distal posterior (DP) 21 21.61 ±9.94
Proximal posterior (PP) 20 24.7 ±6.22
Cuff 18 18.21 ±8.12
Cervix 10 22.45 ±12.75

J Sex Med 2006;3:979–987

Nerve Distribution in the Vagina
innervation was relatively uniform, with nerves
noted throughout the vagina. Overall, the vaginal
cuff had the least innervation. Sites of the vagina
were then compared in each subject who had sam-
ples taken from all the five vaginal sites (N = 17).
There was no site in the vagina that consistently
demonstrated the most innervation: four (23%)
had the highest innervation in the distal anterior
vagina, three (18%) had the most nerves at the
proximal anterior specimen, three (18%) at the
distal posterior, one (6%) subject had the highest
nerve density at the cuff, and proximal posterior
specimens bore the greatest number of nerves in
six (35%) subjects.
An overall description of the nerves and their
location was documented for all specimens. Small
nerves were noted to be thinner and occurred
in each field, whereas large nerves were thicker,
often appeared bundled, and were less abundant
(Figure 1). There were no nerves noted in the
epithelium. However, nerves appeared in both the
superficial and deep submucosa, with relatively
equal distribution. Perivascular innervation, often
in the deep submucosa, was noted in all anatomic
sites. Large nerves tended to be located in the
deeper submucosa, whereas the small nerves were
noted both superficially and deep. Few locore-
gional differences were noted except for cervical
specimens. At this site, the majority of the nerves
were in the deep submucosa.
Nerve quantity and distribution were then
examined in association with the subject’s demo-
graphic and medical history. There were no dif-
ferences in nerve quantity based on age (P =
0.910), maturation index (P = 0.901), previous
hysterectomy (P = 0.660), or number of vaginal
deliveries (P = 0.299). The total number of nerves
was similar based on stage of prolapse (P = 0.384)
and in those with stress urinary incontinence (P =
0.900). Nerve location was then assessed based on
subject age, maturation index, menopausal status,
previous hysterectomy, parity, stage of prolapse,
and stress incontinence. There were no significant
relationships noted between depth of nerves or
perivascular innervation and these factors.
Twelve (57%) subjects had FSD based on the
FSFI. Total and domain scores were significantly
different in subjects with dysfunction compared
to those without in all domains except orgasm
(Table 3). Comparison of demographic informa-
tion was performed for subjects with and without
dysfunction. Age was the only factor that was
significant; younger age was associated with FSD
(P = 0.012).
983
Table 3 Mean FSFI domain scores (N = 21)
FSD No FSD
Domain (N = 12) (N = 9) P value
Desire 2.65 4.13 0.004**
Arousal 2.85 4.70 0.002**
Lubrication 3.48 5.23 0.026*
Orgasm 3.43 4.84 0.185
Satisfaction 3.10 5.29 0.001**
Pain 2.73 5.87 0.000***
Total score 18.24 30.07 0.000***
*Indicates significance at 0.05 using Wilcoxon rank sum test; **indicates sig-
nificance at 0.01 using Wilcoxon rank sum test; ***indicates significance at
0.001 using Wilcoxon rank sum test.
FSFI = Female Sexual Function Index; FSD = female sexual dysfunction.
Innervation was then assessed with respect to
the patient’s sexual function. There was no corre-
lation between total nerves and desire (r = 0.092,
P = 0.691), arousal (r = 0.131, P = 0.571), lubrica-
tion (r = 0.223, P = 0.332), orgasm (r = 0.121,
P = 0.602), satisfaction (r = 0.163, P = 0.480), pain
(r = −0.069, P = 0.768), or overall (r = 0.127, P =
0.584) scores. Large, small, and total nerves were
subsequently analyzed in subjects with FSD com-
pared to subjects without (Figures 2 and 3). There
were no differences in total or site-specific nerves
between the two groups.
Discussion
The primary purpose of this study was to charac-
terize vaginal innervation in women using a large
sample size, age range, and multiple vaginal biop-
sies. Previous research has been restricted by these
factors, or the use of animal or fetal specimens,
leading to a limited understanding of nerve distri-
bution in the adult vagina. Whereas studies have
suggested presence of nerves in the distal vagina
[11], there is a perception that the upper vagina is
deplete in innervation. We report consistent nerve
distribution throughout the vagina, with the vag-
inal cuff showing slightly fewer nerves than the
anterior and posterior walls. This distribution
could be attributed to the embryological origin of
these structures, with the vagina derived from the
urogenital sinus bearing greater nerves than that
of the müllerian ducts.
What remains uncertain is the purpose of the
vaginal nerves we described. Vaginal eroticism has
been reported, with stimulation of both anterior
and posterior vagina leading to orgasm [14,15].
Central and peripheral neurological disorders lead
to problems with orgasm, genital sensation,
decreased lubrication, and sexual desire [19],
implying that adequate innervation is important to
these functions. New studies on mechanisms of
J Sex Med 2006;3:979–987
984 Pauls et al.

3.5
P = .208
3.0
P = .338
2.5
Mean Nerves

P = .906
2.0
P = .394
P = .827
1.5
P = .564
1.0
0.5
P = .348
0.0
e

ge

s
ite
rg

rg

rg

rg

rg
ar
La

La

La

La

La

lS
fL

Al
r

x
uf
rio

rio

io

io

vi

of
C
er

er

er
te

te

st

st

e
An

An

g
Po

Po

Figure 2 Mean large nerves in

ra
l

al
ta

e
l

al
im

ta

Av
is

im
is

patients with FSD vs. no FSD.

D

ox

e
D

ox

rg
Pr

Pr

FSD = female sexual dysfunction.

FSD No FSD

arousal show that central input is important in La ween groups. Therefore, we postulate that alth-
blood flow changes to the vagina [31], and ana- ough immunohistochemical staining yields a
tomic studies suggest a role for innervation in vag- description of nerve density, it does not provide
inal blood flow and capillary permeability [32]. information of nerve behavior. Nerve function
Finally, a recent study using the rat model dem- may be different in subjects with FSD compared
onstrated a link between the vagina and central to those without. Unfortunately, vaginal nerve
nervous system sites involved in female sexual activity is not easily assessed using conventional
responses [33]. We hypothesized that sexual surface electromyogram or pudendal nerve motor
responses, such as arousal, lubrication, orgasm, latency methodology.
and pain, may be mediated by nerve fibers termi- Previous anecdotal evidence suggests the exist-
nating in the vaginal mucosa, and consequently ence of an erogenous zone on the anterior vaginal
the nerve number would be associated with sexual wall at the bladder neck, termed the “Gräfenberg,”
dysfunction. However, we report no differences in or “G-spot” [16,35]. Presence of such an area has
total or site-specific innervation in patients with been widely accepted by the general public; how-
dysfunction compared to those without. A post- ever, anatomic proof for this site is inconclusive
hoc effect size calculation was performed and [36,37]. Research has suggested that the paraure-
revealed a large effect of 2.18 [34]. While this thral glands are the homolog to the male prostate
result could be biased due to our small sample, and [38], and some think these glands may be contrib-
therefore should be interpreted with caution, it utory to this zone, whereas others feel that it is the
suggests that the results are supported by adequate bulbs of the clitoris [39] draping down on the
power to determine significant differences bet- anterior vagina that leads to eroticism at this site.

P = .088
P = .466
25 P = .558
P = .951 P = .983
P = .937 P = .829
P = .757
20
Mean Nerves

15

10

0
l

s
al

al

al

al

al

al

it e

ite
m

Sm

lS

lS
rS

rS

rS

rS

fS

Al

Al
x
uf
io

io

rio

rio

vi

of

of
C
er

er
te

te

te
nt

ge

ge
An

os

os
lA

ra

ra
lP

P
al
ta

ve

Figure 3 Mean small and total nerves

al
m

ta

Av
is

im

lA
xi

is
D

s
ox

al
ve
Pr

Sm

in patients with FSD vs. no FSD.

Pr

er
lN

FSD No FSD
FSD = female sexual dysfunction.
ta
To

J Sex Med 2006;3:979–987

Nerve Distribution in the Vagina
We did not document a corresponding increase in
innervation in the anterior vagina. However, we
do not claim that this is proof that the G-spot does
not exist. Our study did not specifically set out to
find the G-spot; we were limited by a single biopsy
in the distal anterior vagina and only sampling
vaginal mucosa. Thus it is possible that we may
have missed this area.
A secondary aim of this study was to quantify
innervation based on various demographic factors,
including vaginal estrogenization and previous
hysterectomy. There are conflicting reports about
the effects of estrogen on nerve size [4,6,8,32],
with a recent study showing that testosterone
treatment, rather than estrogen, led to increases in
vaginal nerve fibers [8]. We did not note a differ-
ence in innervation based on estrogen status in the
vagina, similar to other reports [6,8,32]. This sug-
gests that estrogen may not be a mediator of nerve
quantity or size in the vagina. Recently, there has
been interest in development of surgical tech-
niques to preserve pelvic autonomic nerves and
reduce potential negative impacts of hysterectomy
on sexual function [19]. We found no difference in
nerve quantity or distribution in our subjects who
had undergone prior hysterectomy. Although our
findings cannot predict function or type of inner-
vation, vaginal nerve quantity and location appears
to be unaffected by hysterectomy. Further studies
are needed to establish the impact of hysterectomy
on pelvic nerve and sexual function.
One limitation of our work is that all the sub-
jects had prolapse. Prolapse has been associated
with subjectively decreased vaginal innervation
[28], making this a potential confounder. Al-
though we did not note a significant difference
in our subjects based on stage of prolapse, it is
possible that subtle changes were present in
these specimens. We were aware of this limita-
tion, but thought that this was the best way to
obtain multiple biopsies from the vagina with
minimal risk to the patient. Because patients
with prolapse constituted the entire sample, it
was felt that comparisons within and between
subjects with respect to sexual function and
nerve location would be acceptable. A second
limitation exists in the use of histological sec-
tions to quantify nerve number. Although mea-
sures were taken to maximize the number of
nerves estimated from each biopsy, by selecting
only the 10 highest density fields for analysis, we
may have misrepresented the true density of the
specimen. Random selection of fields for quanti-
fication may be a means to reduce this potential
985
bias. Finally, it is important to note that nerves
on histology may not represent type and mode
of tissue innervation. It is not possible to know
from immunohistochemical stains the nature of
an organ’s gross innervation, which limits the
findings described here.
Nerve quantity alone cannot be predictive
of sexual function or dysfunction. While we
explored a potential relationship between sexual
function and nerve density, we were aware that this
is only one factor. Sexual function is complex,
impacted by multiple issues in the context of a
woman’s life, and many variables may have been
responsible for the differences seen here. Never-
theless, our findings have advanced our knowledge
regarding vaginal anatomy and sexuality. Future
investigation should attempt to assess nerve
behavior in relation to sexual function, rather than
nerve density.
Conclusions
In summary, a prospective study in women involv-
ing biopsies of proximal and distal anterior and
posterior portions of the vagina, including apex
and cervix, was performed. Vaginal innervation
was regularly identified in all locations, with no
site of increased density. There was no correlation
between vaginal nerve quantity and the patient’s
demographic information or sexual function.
Acknowledgments
The author (R.P.) would like to acknowledge the Tri-
Health Hatton Research Department for providing
funding for this project.
Corresponding Author: Rachel Pauls, MD, Good
Samaritan Hospital—Division of Urogynecology and
Reconstructive Pelvic Surgery, 375 Dixmyth Ave
Seton Center, 8th Floor, Cincinnati, OH 45220, USA.
Tel: +1-513-872-1658; Fax: +1-513-872-4498; E-mail:
rachel_pauls@trihealth.com
Conflict of Interest: None declared.
References
1 Croak AJ, Gebhart JB, Klingele CJ, Lee RA, Ray-
burn WF. Therapeutic strategies for vaginal Mulle-
rian agenesis. J Reprod Med 2003;48:395–401.
2 Capenter SEK, Rock JA. Pediatric and adolescent
gynecology, 2nd edn. Lippincott: Williams &
Wilkins; 2000.
3 Sobel V, Zhu YS, Imperato-McGinley J. Fetal hor-
mones and sexual differentiation. Obstet Gynecol
Clin North Am 2004;31:837–56 x–xi.
J Sex Med 2006;3:979–987
986
4 McKenna KE. The neurophysiology of female sex-
ual function. World J Urol 2002;20:93–100.
5 Giraldi A, Marson L, Nappi R, Pfaus J, Traish AM,
Vardi Y, Goldstein I. Physiology of female sexual
function: Animal models. J Sex Med 2004;1:237–53.
6 Giraldi A, Alm P, Werkstrom V, Myllymaki L,
Wagner G, Andersson KE. Morphological and
functional characterization of a rat vaginal smooth
muscle sphincter. Int J Impot Res 2002;14:271–82.
7 Onol FF, Ercan F, Tarcan T. The effect of ovariec-
tomy on rat vaginal tissue contractility and histo-
morphology. J Sex Med 2006;3:233–41.
8 Pessina MA, Hoyt RF Jr, Goldstein I, Traish AM.
Differential effects of estradiol, progesterone, and
testosterone on vaginal structural integrity. Endo-
crinology 2006;147:61–9.
9 Krantz KE. Innervation of the human vulva and
vagina; A microscopic study. Obstet Gynecol 1958;
12:382–96.
10 Owman C, Rosenbren E, Sjoberg NO. Adrenergic
innervation of the human female reproductive
organs: A histochemical and chemical investigation.
Obstet Gynecol 1967;30:763–73.
11 Hilliges M, Falconer C, Ekman-Ordeberg G,
Johansson O. Innervation of the human vaginal
mucosa as revealed by PGP 9.5 immunohistochem-
istry. Acta Anat (Basel) 1995;153:119–26.
12 Yucel S, De SA Jr, Baskin LS. Neuroanatomy of
the human female lower urogenital tract. J Urol
2004;172:191–5.
13 Stjernholm Y, Sennstrom M, Granstrom L, Ekman
G, Johansson O. Protein gene product 9.5-immu-
noreactive nerve fibers and cells in human cervix of
late pregnant, postpartal and non-pregnant women.
Acta Obstet Gynecol Scand 1999;78:299–304.
14 Alzate H, Londono ML. Vaginal erotic sensitivity.
J Sex Marital Ther 1984;10:49–56.
15 Alzate H. Vaginal eroticism: A replication study.
Arch Sex Behav 1985;14:529–37.
16 Goldberg DC, Whipple B, Fishkin RE, Waxman H,
Fink PJ, Weisberg M. The Grafenberg spot and
female ejaculation: A review of initial hypotheses. J
Sex Marital Ther 1983;9:27–37.
17 Shafik A, El Sibai O, Shafik AA, Ahmed I, Mostafa
RM. The electrovaginogram: Study of the vaginal
electric activity and its role in the sexual act and
disorders. Arch Gynecol Obstet 2004;269:282–6.
18 Basson R, Leiblum S, Brotto L, Derogatis L, Four-
croy J, Fugl-Meyer K, Graziottin A, Heiman JR,
Laan E, Meston C, Shover L, vanLankveld J,
Schultz WW. Revised definitions of women’s sexual
dysfunction. J Sex Med 2004;1:40–8.
19 Nappi R, Salonia A, Traish AM, van Lunsen RH,
Vardi Y, Kodiglu A, Goldstein I. Clinical biologic
pathophysiologies of women’s sexual dysfunction. J
Sex Med 2005;2:4–25.
20 Connell K, Guess MK, La Combe J, Wang A, Pow-
ers K, Lazarou G, Mikhail M. Evaluation of the role
of pudendal nerve integrity in female sexual func-
J Sex Med 2006;3:979–987
Pauls et al.
tion using noninvasive techniques. Am J Obstet
Gynecol 2005;192:1712–7.
21 Bump RC, Mattiasson A, Bo K, Brubaker LP,
DeLancey JO, Klarskov P, Shull BL, Smith AR.
The standardization of terminology of female pelvic
organ prolapse and pelvic floor dysfunction. Am J
Obstet Gynecol 1996;175:10–7.
22 Meston CM. Validation of the Female Sexual Func-
tion Index (FSFI) in women with female orgasmic
disorder and in women with hypoactive sexual
desire disorder. J Sex Marital Ther 2003;29:39–46.
23 Rosen R, Brown C, Heiman J, Leiblum S, Meston
C, Shabsigh R, Ferguson D, D’Agostino R Jr. The
Female Sexual Function Index (FSFI): A multidi-
mensional self-report instrument for the assessment
of female sexual function. J Sex Marital Ther
2000;26:191–208.
24 Rogers RG, Kammerer-Doak D, Villarreal A,
Coates K, Qualls C. A new instrument to measure
sexual function in women with urinary incontinence
or pelvic organ prolapse. Am J Obstet Gynecol
2001;184:552–8.
25 Salonia A, Zanni G, Nappi RE, Briganti A, Deho F,
Fabbri F, Colombo R, Guazzoni G, DiGirolamo V,
Rigatti P, Montorsi F. Sexual dysfunction is com-
mon in women with lower urinary tract symptoms
and urinary incontinence: Results of a cross-
sectional study. Eur Urol 2004;45:642–8.
26 Wiegel M, Meston C, Rosen R. The female sexual
function index (FSFI): Cross-validation and devel-
opment of clinical cutoff scores. J Sex Marital Ther
2005;31:1–20.
27 DeMay RM. The art and science of cytopathology.
Chicago: American Society of Clinical Pathologists,
1996.
28 Boreham MK, Wai CY, Miller RT, Schaffer JI,
Word RA. Morphometric properties of the poste-
rior vaginal wall in women with pelvic organ pro-
lapse. Am J Obstet Gynecol 2002;187:1501–8.
29 Nemeth L, Puri P. The innervation of human bowel
mucosa and its alterations in Hirschsprung’s disease
using a whole-mount preparation technique. Pedi-
atr Surg Int 2000;16:277–81.
30 Park AM, Armin S, Azarbal A, Lai A, Chen PS,
Fishbein MC. Distribution of cardiac nerves in
patients with diabetes mellitus: An immunohis-
tochemical postmortem study of human hearts.
Cardiovasc Pathol 2002;11:326–31.
31 Brotto LA, Basson R, Gorzalka BB. Psychophysio-
logical assessment in premenopausal sexual arousal
disorder. J Sex Med 2004;1:266–77.
32 Hoyle CH, Stones RW, Robson T, Whitley K,
Burnstock G. Innervation of vasculature and
microvasculature of the human vagina by NOS and
neuropeptide-containing nerves. J Anat 1996;188:
633–44.
33 Marson L, Murphy AZ. Identification of neural cir-
cuits involved in female genital responses in the rat:
A dual virus and anterograde tracing study. Am J
Nerve Distribution in the Vagina
Physiol Regul Integr Comp Physiol 2006;29(2):19–
28.
34 Cohen J. Statistical power analysis for the behavioral
sciences, 2nd edn. Hillsdale, NJ: Lawrence Earl-
baum Associates; 1988. Epub February 16, 2006.
35 Addiego F, Belzer E, Comolli J, Moger W, Perry J,
Whipple B. Female ejaculation: A case study. J Sex
Res 1981;17:13–21.
36 Alzate H, Hoch Z. The “G spot” and “female ejac-
ulation”: A current appraisal. J Sex Marital Ther
1986;12:211–20.
987
37 Hines TM. The G-spot: A modern gyneco-
logic myth. Am J Obstet Gynecol 2001;185:
359–62.
38 Zaviacic M, Ablin RJ. The female prostate and pros-
tate-specific antigen. Immunohistochemical local-
ization, implications of this prostate marker in
women and reasons for using the term “prostate” in
the human female. Histol Histopathol 2000;15:131–
42.
39 O’Connell HE, Sanjeevan KV, Hutson JM. Anat-
omy of the clitoris. J Urol 2005;174:1189–95.
J Sex Med 2006;3:979–987