P
elvic inflammatory disease (PID) is a common, and the most serious,
complication of sexually transmitted diseases. The disease includes
infection and inflammation of the upper genital tract involving the
endometrium, fallopian tubes, ovaries, and adjacent pelvic
spaces.
The importance of early diagnosis and prompt and successful
treatment of this disease lies in the sequelae of this inflammatory
process that leads to chronic pelvic pain and inflammation,
scarring and adhesions with subsequent infertility, and increased
risk of ectopic pregnancy.
1
PID is a major medical problem in young women. Sexually active
female teenagers are at particularly high risk. The magnitude of
the problem of PID is so important in terms of the epidemiologic
and economic consequences of this disease that it may be concluded
that this is the most significant complication of all sexually
transmitted diseases in America.
Epidemiology
The annual incidence of PID in women 15 to 39 years of age seems
to be 10 to 13 per 1000 women with a peak incidence of about 20 per
1000 women in the age group 20 to 24 years.
2
An estimated 800,000 to 1 million cases of PID are reported in the
United States each year.
2,3
More than 275,000 patients are hospitalized annually for this
disease and more than 100,000 surgical procedures are performed.
The estimated cost to treat almost 1 million women annually in the
United States is greater than $3 billion and projections of PID
costs show that by the year 2000, the total annual cost of PID will
exceed $9 billion, assuming no change in PID incidence and a
constant rate of medical inflation.
3,4
Pathophysiology and etiology factors
PID is predominantly an ascending infection from the lower to
the upper genital tract. Microbial agents may reach the upper
genital tract by three routes:
1) ascending infections from the cervix and endometrium caused
by sexually transmitted microorganisms;
2) secondary to direct spread from nearby pelvic organs with
inflammatory process, such as appendicitis; and
3) hematogenous.
The cervix represents a natural barrier to ascending infections
from the vagina. The mucous plug is a mechanical barrier in
association with the protecting mechanism of cervical secretions.
There are situations where the cervical protecting factor is absent
or decreased, such as in menses.
Extension of columnar epithelium from the endocervical canal
into the vaginal portion of the cervix (cervical ectopia) has been
described as a factor of increased risk for PID in sexually active
adolescents. Microorganisms adhere better to columnar epithelium
than to squamous epithelium present in the post-adolescent.
There is also a correlation between promiscuous sexual relations
with multiple sexual partners, increased incidence of sexually
transmitted disease, and increased PID. The risk of developing PID
increases by a factor of 5 in these patients. A form of
"iatrogenic" PID is common. Opening of the cervical canal or
introduction of foreign material into the uterine cavity facilities
ascending spread of infection from the cervix. These procedures
include cervical dilatation, abortion, curettage, tubal
insufflation, hysterosalpingography, and IUD insertion.
Introduction of an IUD increases the risk of PID by 1.5 to 5 times,
depending upon the type of IUD employed.
5
Previous PID increases the risk for recurrent PID due to
increased susceptibility to bacterial colonization in previously
damaged tubal epithelium.
Oral contraceptives have accounted for a double effect in PID.
There is increased risk due to delay recession of cervical ectopia
and may enhance Chlamydia infection, particularly cervicitis by a
poorly understood mechanism. The protected effect of oral
contraceptive usage might be explained by the maintenance of the
cervical mucus barrier throughout the menstrual cycle, the short
scant menstrual bleeding from an inactive endometrium, and a
decreased muscular activity of the myometrium.
2
Reports from the last decades have clearly demonstrated that PID
has a multimicrobial etiology.
Neisseria gonorrhea
and
Chlamydia trachomatis
are the most common pathogens in PID.
6
Other aerobic and anaerobic agents, such as
Mycoplasma hominis
, peptostreptococcus, and Bacteroides species, account for 25% to
50% of PID.
Clinical findings
The symptoms of PID vary in large scale and may be atypical,
simulating other nongynecologic problems. The most common
presentation of PID is pelvic pain. On physical exam, patients with
PID present with tenderness to cervical motion and adnexal
tenderness. Purulent discharge may be a common finding. Additional
signs of infection such as elevated temperature, palpable adnexal
complex, leukocytosis, elevated erythrocyte sedimentation rate or
Creactive protein, and positive tests for either
N. gonorrhea
or
C. trachomatis
will improve the overall specificity of the clinical diagnosis.
The clinical diagnosis of PID on the basis of symptoms and signs
is often inaccurate. Only 20% of patients with proven PID by
laparoscopy have the classic clinical findings and clinical
diagnosis identifies only 65% of women with PID.
7
Laparoscopy has been the standard of reference in the diagnosis of
PID but it requires general anesthesia and hospitalization, is
expensive, and is not readily available.
8
The endometrial biopsy has been described as an office test for the
outpatient diagnosis of PID. It is less invasive than laparoscopy
but the results are not readily available.
9
Ultrasound
Transabdominal ultrasound is useful in identifying complicated
pelvic inflammatory disease and tubo-ovarian abscess.
10,11
Transvaginal ultrasound is a relatively simple procedure and can be
performed with an empty bladder.
Transvaginal ultrasound allows for improved resolution and
visualization of pelvic organs with the use of high frequency
transducers, and demonstrates various stages of tubo-ovarian
inflammation not previously seen with transabdominal sonography.
This information was obtained previously only with laparoscopy.
9,11
A wide spectrum of sonographic findings have been described.
Uterus and endometrium
-
Sonographic manifestations vary according to the extent, duration,
and site of the disease. The exam may be normal in early cases or
the uterus may be slightly enlarged. The margins of the uterus may
be ill defined. This feature has been termed "indefinite uterus" by
Bowie
12
and this is probably related to the presence of pelvic exudate
and/or adhesions (figure 1).
FIGURE 1.
Pelvic inflammatory disease. A transabdominal longitudinal
sonogram of the uterus [U] shows slightly enlarged uterus with
poorly defined margins. [B] Bladder.
Endometritis may not present specific sonographic features.
13
The endometrium may be thickened to more than 12 to 14 mm.
Endometrial echo texture may be increased or decreased and poorly
defined with hypoechoic areas (figure 2A) and fluid is sometimes
present within the endometrial cavity (figure 2B). Endometrial
findings are better depicted by transvaginal ultrasound.
11,14
FIGURE 2.
Endometritis. (A) Coronal endovaginal sonograrn of the uterus.
The endometrium is thickened (arrows) with ill-defined hypoechoic
areas. (B) Sagittal endovaginal sonogram demonstrates fluid
within the endometrial cavity (arrows).
Fallopian tubes
-
Often, the normal fallopian tube is not clearly identified with
transvaginal ultrasonography due to the tube's thin diameter (1 to
4 mm). The tubes are usually visible when sufficient intraluminal
or surrounding fluid is present.
Early in the course of acute salpingitis, the ultrasound exam
may be normal. Laparoscopy demonstrates only erythema and serosal
edema. With progression of the disease, purulent exudate accumulate
within the lumen of the tube. The inflammatory reaction may occlude
the ostium and the tube dilates with an elongated convoluted or
club-shaped configuration.
11,15
Purulent material within the distended tube constitutes a
pyosalpinx. This is manifested by internal echoes within the
fluid-filled tube, fluid debris, or layered pus may sometimes be
present
15
(figure 3). In the course of time with treated infections, the
infecting organism may disappear, the pus undergoes proteolysis,
and it is substituted by a thin serous fluid to produce a
hydrosalpinx.
FIGURE 3.
Pyosalpinx.(A) Sagittal and (B) coronal endovaginal ultrasound
shows distended left fallopian tube with thickened wall (arrows)
and low amplitude internal echoes, related to purulent material.
Echogenic fluid is adjacent to the tube. [F] Fluid.
Hydrosalpinx has been characterized as having four distinct
features: a tubular shape, folded configuration, well-defined
echogenic wall, and short linear echoes protruding into the lumen
(figure 4). The linear intraluminal echoes have been described as
possibly due to the wrinkled nature of the tubal epithelium.
15
The fluid of hydrosalpinx is anechoic in comparison with the
purulent and debris component on the pyosalpinx.
FIGURE 4.
Hydrosalpinx. (A) Longitudinal and (B) coronal transvaginal
sonograms in two different patients show characteristic folded
tubular configuration of distended fallopian tube filled with
fluid and well-defined echogenic thickened wall. (C) Longitudinal
transvaginal sonogram on a different patient with dilated
fluid-filled tube depicts nodular and linear echoes protruding
into the lumen (arrow).
Differential diagnosis of dilated fallopian tubes should be made
with distended pelvic veins and bowel loops. The use of color
Doppler easily distinguishes hydrosalpinx from dilated pelvic
veins. Peristaltic motion is the main feature for differentiating
bowel loops from dilated tubes.
Ovaries
-
The ovaries are relatively resistant to infection, however,
spillage of purulent material can eventually cause involvement of
the ovary.
In oophoritis, the ovaries are enlarged and globular with
multiple cysts (polycystic-like ovaries) and in early stages the
only sonographic finding may be ill-defined margins
9,14
(figure 5).
FIGURE 5.
Oophoritis. (A) Transverse transabdominal ultrasound shows that
the right ovary is enlarged and globular. (B) Transverse and (C)
sagittal endovaginal sonograms show an enlarged right ovary with
multiple cysts (polycystic-like appearance). The left fallopian
tube is dilated and fluid-filled. Fluid is surrounding the right
adnexa (arrows) corresponding to periovarian exudate . [B]
bladder, [U] uterus, [O] right ovary, [T] left fallopian
tube.
Oophoritis probably increases the volume of the ovaries by
producing inflammatory exudate and edema in the vascular pole. The
thickened ovarian capsule might prevent normal follicular growth
causing multifollicular degeneration.
9
Persistent spillage of purulent material into the pelvic results in
peritoneal irritation and adhesions. When the inflammatory process
progresses, the dilated convoluted tubes are not distinguishable
from the contiguous ovary constituting the tubo-ovarian complex.
7,11
Sonographic pattern with more extensive and severe inflammation has
been described as nonspecific adnexal pelvic mass
9-13
(figure 6).
FIGURE 6.
Tubo-ovarian complex. Transverse endovaginal sonogram of the
right adnexa. Complex mass encompassing ovary and tube. Some
fluid (arrows) is present and ill-defined margins most likely
related to the inflammatory process and adhesions.
Progression of the inflammatory process in patients with
suboptimal treatment or poor response to antibiotics may result in
the formation of tubo-ovarian abscesses. This serious complication
occurs in as many as one-third of patients hospitalized for acute
salpingitis.
Tubo-ovarian abscesses have a variable appearance on sonography.
Typically they are unilocular or multilocular complex masses with
irregular borders, thickened walls, and fluid debris levels
11,15
(figure 7). The hypoechoic loculations usually represent
collections of pus. Gas-containing abscesses are uncommon in PID
and, when present, the gas may produce bright echoes with acoustic
shadowing. Tubo-ovarian abscesses may produce gross distortion of
pelvic anatomy.
FIGURE 7.
Tubo-ovarian abscess (A) Sagittal endovaginal sonogram of the
left adnexa. Complex mass with thickened walls, internal echoes,
and dilated fluid-filled tube (arrow). (B) Transverse endovaginal
sonogram of the left adnexa of a different patient demonstrates a
multilocular complex fluid collection with thickened wall,
septations, and layering internal echoes related to purulent
material.
The appearance of tubo-ovarian abscesses is not pathognomonic
and differential diagnosis must be established with other entities
such as malignant ovarian neoplasms, endometriosis, and abscesses
of nonogynecologic origin related to diverticulosis or
appendicitis.
Transabdominal sonography may better evaluate the extent of
large tubo-ovarian abscesses than transvaginal sonography.
Cul-de-Sac
-Fluid may be identified in the cul-de-sac, either secondary to
purulent spilling from the tubes or as a peritoneal response to
inflammation. Fluid may show low level echoes indicating pus or
blood. Free pelvic fluid alone has lower specificity than other
ultrasound findings.
9
Doppler ultrasound
Color Doppler ultrasound has been utilized by some clinicians to
evaluate adnexal flow characteristics in benign conditions such as
flow in the corpus luteum, extrauterine pregnancy, and in ovarian
neoplasms.
Vasodilatation mediated by the inflammation and angioneogenesis
result in increased capillary flow and decrease in vascular
resistance during the acute phase of infection (figure 8).
FIGURE 8.
Tubo-ovarian abscess. Transverse endovaginal sonogram of the
right adnexa shows complex multiloculated fluid collection with
internal debris. Duplex ultrasound ofthe wall demonstrates
increased flow with low-resistance characteristics.
Low resistance blood flow with decreased values of the
resistance index (RI) and pulsatility index (PI) has been
documented by several authors.
16-18
RI value <0.5 has been found in the acute stage at the margin
of the infectious complex. During the healing process, the changes
in vascular flow return to normal.
16,17
RI and PI normalize as the infection subsides with adequate
treatment. The increase in vascular resistance returns to normal
before infectious parameters such as body temperature, erythrocyte
sedimentation rate, C-reactive protein, and leukocyte counts do.
16
Computed tomography
Computed tomography (CT) is also helpful in evaluating the
extent of the inflammatory process, particularly in patients with
poor response to antibiotic treatment in which surgical or
percutaneous drainage is contemplated.
The CT findings of tubo-ovarian abscess include thick-walled,
fluid density mass in an adnexal location with internal septations;
anterior displacement of the mesosalpinx; and thickening of the
uterosacral ligaments. Rectosigmoid involvement manifested by
luminal narrowing, infiltration of perirectal fat, and indistinct
borders between the pelvic mass and bowel has been also described
(figure 9). The ureters may be dilated due to compression or spasm
and para-aortic lymphadenopathy may be present.
19-21
FIGURE 9.
Tubo-ovarian abscess. Computed tomogram of the pelvis. Large
complex pelvic fluid collection displaces bladder and rectum.
Rectal wall is thickened.
CT findings are, however, not specific for tubo-ovarian
abscesses. In patients with poor response to antibiotic therapy,
percutaneous drainage of the collections may be guided by
ultrasound or CT.
22-24
Magnetic resonance
Magnetic resonance (MR) shows a variety of findings in
laparoscopically verified PID cases, including tubo-ovarian
abscesses, massively dilated to slightly dilated fluid-filled
tubes, and polycystic-like ovaries with free pelvic fluid. In a
recent study, MR was more accurate than transvaginal ultrasound in
the diagnosis of PID when comparing with laparoscopy.
25
Although at this time, transvaginal ultrasound is far more cost
effective than MR imaging, MR may have the potential to reduce the
need for diagnostic laparoscopy.
25
AR