Pediatric Dorothy I. Bulas, Anna R. Nussbaum

MD

Promise MD

A. Ahlstrom, MD #{149}Carlos Regina M. O’Donnell, BA

#{149}

Blask,

Transabdominal (TA) and transvaginat (TV) sonograms (n 116) were obtained in 84 patients aged 12-21 years (mean, 16.2 years) with the dinical diagnosis of acute pelvic inflammatory disease (PID). The studies were compared for image quality and unique diagnostic information. TV sonography demonstrated superior resolution of 25 dilated fallopian tubes. Heterogeneous pelvic masses, described as tubo-ovanian abscesses on TA sonograms, could be separated on TV sonograms into various stages of PID including pyosalpinx, hydrosalpinx, tubo-ovarian complex, and tubo-ovarian abscess. Thirty-one TA and TV studies were normal despite patients fulfilling strict clinical critena for PID. The level of severity of PID, as determined at TA sonography, was altered in 28 cases, with medical therapy changed in 23 cases because of additional TV sonographic findings. TV sonography provided superior anatomic detail in the evaluation of patients with PID, demonstrating abnormalities that were not seen at TA sonography in 71% of patients. terms:

Fallopian #{149} Ovary, abscess,

853.217 gans,

abscess,

85.1298,

85.217

tubes, abscess, 852.217 #{149}Pelvic #{149}Pelvic organs, US,

or-

85.12989

Radiology

1992;

183:435-439

From the Departments of Diagnostic Imaging and Radiology (D.I.B., C.J.S., A.R.N.B.) and Pediatrics (D.l.B., P.A.A., C.J.S., A.R.N.B.), and the Office of the Research Director (R.M.O.), Children’s National Medical Center, 111 Michigan Ave NW, Washington, DC 20010; and the George Washington University School of Mcdicine and Health Sciences, Washington, DC I

(D.I.B., P.A.A., C.J.S., A.R.N.B.,

R.M.O.).

From

1991 RSNA scientific assembly. Received Dccember 2, 1991; revision requested January 3, 1992;

revision

January RSNA,

received

27. Address 1992

January

reprint

22;

requests

Sivit,

MD

#{149}

Pelvic Inflammatory Disease Comparison ofTransabdominal Transvaginal Sonographic

Index

J.

Radiology

accepted

to D.I.B.

the

P ally ally risk

ELVIC

in the Adolescent: and Evaluation’

inflammatory

is a serious transmitted active female (1). Although,

disease

complication infections, teenagers to our

(PID)

of sexuwith sexuat highest knowledge,

optimal therapy has not yet been established, it has been noted that patients with severe disease require prolonged administration of broad spectrum

antibiotics

intravenously

and possible surgical intervention 6). The severity of PID is difficult assess

clinically

Thus, the use of TV sonography in this population merits special attention. The purpose of this study was to determine prospectively if TV sonography was well tolerated in the adolescent population and whether it provided additional information in the diagnosis and management of acute PID.

(7).

The

finding

(2to of a

pelvic mass at transabdominal (TA) sonography has been used as evidence of severe tubo-ovarian inflammation (7,8). Findings from previous series have suggested the need for more individualized therapy, depending on the stage of acute salpingitis (2,3). The use of TA sonography to differentiate among various stages of PID, howeven, is limited (9,10). Swayne et al (11) attempted to correlate TA sonographic findings with laparoscopic findings, but only three sonographic patterns emerged: endometritis, focal

MATERIALS

AND

Over a 14-month period, 106 consecutive female patients aged 12-21 years (mean, 16.3 years) underwent 137 pelvic sonographic

examinations

pected PID. the adolescent cal

and

All

were

pelvic

examinations.

in urine,

blood

admission Twenty-two

when

studies

that

used

transvaginal (TV) sonography, however, have been promising. Its use in identifying dilated fallopian tubes has been established (12). Pyosalpinx, ovarian thickening, and small fluid collections have been identified with TV sonography in patients with PID (13). The role of TV sonography in the evaluation of pelvic infections, however, has yet to be established. Adolescents represent an especially high-risk group because of the high incidence of multiple sexual partners and the prevalence of Neisseria gonorrhoeae and Chlamydia trachomatis. Whether TV sonography is an acceptable examination to perform on adolescents

has

yet

to be established.

were

obtained

at

excluded

from the study because they did not fulfill clinical criteria for the diagnosis of acute ND. Clinical criteria included the presence

20 mm/h,

of preliminary

enythro-

were

possible. patients

sults

Pyo-

and

rates

of the following:

distortion.

A sample

cell counts,

cyte sedimentation

or more

pelvic

to

physi-

from the endocervix was cultured for N gonorrhoeae and C trachomatis. Measurements of human chonionic gonadotropin

salpmnx, hydmosalpinx, acute or chronic salpingitis, tubo-ovarian complex, and tubo-ovarian abscess could not be differentiated on a TA sonogram. Re-

total

of sus-

admitted

unit and underwent

abdominal adnexal

and

because

patients

of lower tenderness,

mass,

METHODS

pain, cervical tenderness,

motion plus one

temperature

of

38#{176}C or greater, white blood cell count 10,500/mm3 (10.5 x 109/L) or greaten,

vated

sedimentation

rate of greaten gonorrhoeae

of N

evidence

of ele-

than and/

or C trachomatis in the endocervix, or pelvic abscess on inflammatory complex at bimanual

examination

or sonography

(14-

17). The remaining

84 patients underwent TA and TV sonography (116 sonograms were obtained) for the initial evaluation and follow-up of acute PID. Within sonogram

72 hours was

of admission,

obtained

with

a pelvic both

TA

and

TV transducers. The purpose of the TV approach was explained to each patient. A consent

form

was

not

used,

since

most

patients were minors and it was difficult to locate legal guardians. Thus, with approval from the Institutional Review

Abbreviations: disease, TA

=

PID = pelvic transabdominal,

inflammatory TV = transvagi-

nal.

435

Board,

only

dune

was

verbal

consent

obtained.

of the

One

proce-

Comparison

sonographer

of TA and TV Sonographic

would perform both TA and TV portions of the examination. A 3-MHz sector transducer and, when possible, a 5-MHz linear transducer

(Acuson,

were

if)

used

Mountain

View,

for the TA portion

amination,

with

for optimal the patient

evaluation voided,

the

patient’s

performed

with

Finding

full

Normal study Uterus Enlarged

After

Endometnial

of the cxbladder

and

both

adnexa.

At the

and

technique

more

All

studies

dently

she

than

the other.

were

At least

unaware

course

and

grams

were

preferred

compared

(D.I.B.,

one

of the

of the TA

compared

34

33

7

0

8

14

5

24

tube/normal

I

5

49

20

0

14

0

15

17

32

(tubo-ovarovary

Pelvic fluid Cul-de-sac Periovarian Abdominal

C.J.S., on

TV

0 2

7 1

radiologists

patient’s

outcome.

one

indepen-

by two radiologists

A.R.N.B.). was

whether

TA

(tubo-

ian complex)

Inflamed

of Findings

tube

No recognizable structures ovarian abscess) Inflamed tube and ovary

conclusion

of both examinations, the patient was asked whether she had experienced any discomfort

fluid

Inflamed fallopian Pyosalpinx Hydrosalpinx Adnexal mass

TV sonography was a 5-MHz, 19-mm-diame-

ten, end-view transducer with a sector angle of 30#{176} (Acuson). Standard views included sagittal and coronal images of the uterus

of PID (116 Sonograms) No.

Cal-

of the organs.

Findings

clinical

and

for

TV

sono-

uterine

mor-

phologic characteristics; ovarian size, contour, and internal architecture; fallopian tube appearance; presence on absence of pelvic

masses;

and

presence

and

contents

of pelvic fluid collections. Ovaries were considered enlarged if ovarian volume was greaten than 14 cm3 (18,19). Fallopian tubes

were

considered

dilated

when

a tu-

bular structure with echogenic walls and folded configuration was identified (12). Doppler was used when necessary to distinguish

between

pelvic

lated fallopian The readers severity

of PID,

studies tem was

grading

separately

rated

such

moderate

were

sys-

PID

or

identified.

but

was

a.

PID

PID

findings

margins

noted

present.

TV

a scoring

ovarian

were

was

the

and

if additional

as ill-defined

mass

TA

and severe. no abnormalities

ovaries

al thickening

a di-

to evaluate

by using

only enlarged

and

the

of mild, moderate, rated mild when

was

veins

tube. were asked

or tub-

no

adnexal

rated

severe

b.

Figure fined

1. with

bladder sonogram

Dilated fallopian tube is clearly delineated with TV sonography and poorly conventional TA sonography. (a) Transverse TA sonogram obtained through (B) demonstrates the uterus (Li) and an amorphous left adnexal mass (arrows). demonstrates a fluid-filled tube (T) containing septations.

Percentage of agreement between readers’ grading of PID severity was

tubo-ovarian

ported metny

and/or

tubo-ovar-

ian abscess was identified within an adnexal mass at TV examination (Table). Once the patient was admitted, cefoxitin, doxycycline, intravenously rated

for

severe, 14 days

obtained

and clindamycin were administered. If PID was

antibiotics were with 26 follow-up

before

completion

PID was rated damycin

mild or

was

discontinued

and doxycycline venously

for

were 7-10

oral doxycycline ditional

7-10

continued sonograms

administered After

Clinical

for an ad-

follow-up

was

evaluated through the analysis of medical records and follow-up studies including one laparoscopy and two laparotomies. Descriptive

summaries

sonographic frequency

findings counts

and

of TV

were

tinuous variables such and patient age were reported plus

or minus

range. 436

#{149} Radiology

I standard

and

reported

percentages. as ovarian deviation

using

the

TA

as Convolume

as means and/or

test

Paired

were

measured

phy. were

Both left and right ovarian included in this analysis. tests

used

McNemar

pairs.

ume

were

to compare

two-sided,

and

were considered significant levels of less than .05.

the reof sym-

Student

ovarian

at TV versus

vol-

TA sonogrameasures All statistifindings

at probability

RESULTS

intra-

discharge,

was continued days.

If

by

for matched

tests

cal

of therapy. moderate, dinand cefoxitin

days.

(b) TV

if

a heterogeneous adnexal mass was identifled at TA examination on a pyosalpinx, complex,

dethe

Patient

Comfort

Fifty-seven believed that as

comfortable

percent of the TV sonography as TA

patients was

sonography.

Twenty-eight percent of patients found the distended bladder required for TA sonogmaphy uncomfortable and thus actually preferred the TV examination. The time to perform the additional TV examination averaged 8.6 minutes (range, 4-21 minutes).

Fallopian Nineteen

four

cases

tified

solely

Tubes cases of pyosalpinx of hydnosalpinx were

with

TV sonography

1). Five cases of pyosalpmnx case of hydrosalpinx were

at TA examination, and wall thickness

and iden-

(Fig

and one recognized

but tubal contents were more clearly

delineated on TV sonograms. Echogenic material and septa within dilated fallopian tubes were demonstrated transvaginally in five cases. Mildly thickened fallopian tubes were identified transvaginally in four cases.

In three be useful veins

cases, TV Doppler in differentiating from dilated fallopian

proved pelvic tubes.

to

Ovaries Ovarian

follicles

were

more

cleanly

delineated transvaginally in 80% of cases, and ovarian margins were more sharply defined in 50% of cases. There

May

1992

Uterus

The uterus appeared normal in 85% of the studies. Endometrial fluid was present in 14 cases (12%), eight of which were identified only on TV sonograms sonography

(Fig 3). Additionally, demonstrated one

intrauterine at TA

gestation

that

TV early

was

missed

sonognaphy.

The smaller field of view of the TV transducer made accurate longitudinal measurement of the uterus difficult. Thus, seven enlarged uteri were identified only at TA sonognaa.

phy.

b.

Figure

2.

TV sonographic

onstrates

depiction of a tubo-ovanian pelvic mass (arrowheads)

abscess. (a) Initial with no recognizable

after

3 weeks

therapy

Tube

(T) remains

a heterogeneous

(b) TV sonogram ovarian

obtained

margins

(arrows).

of antibiotic thickened

and

shows filled

TV sonogram dempelvic structures.

the sharply

with

Pelvic

TV sonognaphy fluid collections

fluid.

and the the

pelvis beyond TV transducer.

the

Fluid

delineated

focal

zone

seven

periovarian

tions that sonography.

of

fluid

were

fled

Inflammation

The

a sharply manginated cases, both fallopian Figure

rows) trial

TV sonogram

3.

demonstrates

of the uterus

fluid

within

(ar-

the

endome-

cavity.

could be recognized transvaginally, but the margins between the structures

were

complete

no

ence

statistically

significant

the

between

the measurement volume:

At TV

sonography,

the

in vol-

16.9 cm3 ± 21.7; TA was 17.3 cm3 ± 22.4; paired Student t test resulted in t = .680, P = .4972. However, 18 ovaries that measured oven 14 cm3 at TA examination were within ume

was

normal

limits

ovaries sonography transvaginally. size

transvaginally,

that

24

were measured This

believed

was

whereas

normal size at TA over 14 cm3 discrepancy in

to be secondary

improved definition gins at TV examination. Twelve enlarged

to

of ovarian ovarian

marcysts

that

appeared to contain internal echoes at TA examination were echofree, translucent cysts at TV sonognaphy. These echoes

most

likely

reverberation

were

artifact

caused

that

present at TV sonography. material within ovarian more cleanly demonstrated

amination tion

of three

Volume

cysts

in 12 additional

TA sonography

183

enabled

ovaries Number

#{149}

located 2

by

was

not

Echogenic was at TV cx-

cases. identifica-

high

it difficult structures

examination

differ-

two modalities of mean ovarian

ill defined.

in

(Fig 2). In these

sonography often strated septations

but no ovarian fled. Intemreaden TV sonographic ity was evaluated.

tissue

could

TV

be identiof TA and

grading of PID sevenIntemreader agree-

for TA sonographic

tions

was

significantly

agreed

interpretalow,

with

the

on only

77%

= .0002). There was closer agreement (90%) between readens’ grading of PID severity when evaluating TV sonognams (P = .3916).

more

echoes

easily were

fluid

nec-

identi-

transvagi-

cases.

at TV sonography was better delineated with TA sonognaphy because its extension into the abdomen.

of

Findings

Of the

initial

acute

PID,

for

sonognaphic

90 studies

performed

31 (34%)

evidence

revealed

no

of pelvic

in-

flammation. All these patients met the clinical criteria for PID, including 26 (84%) with positive cervical cultures. Seventy-two percent of patients who

met agreement

ment

level of severity of the time (P

cases,

better demonand debris levels,

collec-

One patient had abdominal fluid in the Morrison pouch, which was identified only at TA sonognaphy. A large loculated pelvic fluid collection noted

of the adnexa to identify any necogeither at TV on TA

were

cul-de-sac

in five

Clinical

In 20 cases,

distortion

made nizable was

ovary. In 14 tube and ovary

fluid

visualized at TA contents within the

internal

within

nally

TA sonography demonstrated 49 heterogeneous adnexal masses. In 15 pelvic masses noted at TA examination, TV sonography could help identify a dilated fallopian tube as well as

not

collections

ognized;

Tubo-ovarian

demonstrated 15 within the cul-de-sac

the TV sonographic

severe tures.

PID

had

positive

TV sonographic

criteria

for

cervical

cul-

findings

were

con-

firmed surgically in two cases. Ovanian tissue with poorly defined margins and a dilated fluid-filled tube was identified on a TV sonogram in

the

first

patient.

ic salpingitis sions and

At lapanotomy,

with fibrosis

chron-

tubo-ovarian was confirmed.

The level of severity of PID, as graded with TA sonograms, was altered in 28 cases (24%) because of additional TV sonographic findings. TV findings suggested more severe tuboovarian inflammation in 11 cases rated mild or moderate at TA examination. In 17 cases, TV sonographic

second

patient

dilated,

echogenic

findings 23 cases, because

tamed 2 weeks later revealed velopment of a hydrosalpinx

the dewith an

adjacent

sharply

manginated

ovary

(Fig 4d).

Because

of persistent

suggested less severe PID. In medical therapy was altered of additional TV sonographic

findings. No treatment failures noted in this group of patients.

were

was

noted

adheThe

tube

to have and

a

peniova-

nan fluid collections (Fig 4a, 4b). Ovarian margins were ill defined on follow-up TV sonograms (Fig 4c). Laparoscopic findings confirmed the presence of periovarian adhesions and

pyosalpmnx.

a lapanotomy

TV

was

sonograms

ob-

performed.

Radiology

pain, The

#{149} 437

ovary appeared cally inflamed brous obliteration was resected.

normal, but a chronifallopian tube with fiof the distal lumen

DISCUSSION PID refers genital tract tubes and/or

to infection involving contiguous

the

in the upper fallopian structures.

Adolescents

represent

a high-risk

group because of the high prevalence of multiple sexual partners and of N gonorrhoeae and C trachomatis in this patient population (1). Sequelae of PID include ectopic pregnancy, infertility, and chronic pelvic pain (10). PID is a syndrome that may include endometnitis, salpingitis, oophonitis, pelvic peritonitis, and tubo-ovanian abscess. The mechanism of tubal infection is not well understood. It likely begins as a cervical infection with N gonorrhoeae and/or C trachomatis altering the cervicovagmnal environment and resulting in an ovengrowth of endogenous flora. As the infection ascends, the fallopian tubes become hyperemic, thickened, and filled with purulent exudate. Peritoneal spillage may occur. Tubal occlusion may develop, resulting in a pyosalpinx. If the infection subsides at this

stage,

a hydrosalpinx

If peritonitis adhesions

may

progresses, may

result

with

tube and complex. may cause

these organs a tubo-ovarian

to break abscess

Clinically,

acute

discharge,

fusion

ing

prove

the

diagnosis

of

salpingitis

includes

fever,

pelvic

sensitivity

a

and

of the

published to im-

clinical

A clinical

inpatient

diagnosis

therapy

PID has been advocated lems with noncompliance lescent

population.

for

due to probin the ado-

Patients

with

se-

vere disease have a substantially greater recovery of aerobes and anaerobes from culdocentesis fluid (3).

Salpingitis

tubo-ovanian ian abscess 438

with

tubal

occlusion,

complex, and tubo-ovarappear to require pro-

Radiology

#{149}

of PID.

(a) TV

sonogram

(b) TV sonogram

material.

shows

demonstrates

dilated,

convoluted

ovarian

tissue

tube

(cursors)

(T) containadjacent to the

A small amount of periovarian fluid is present (arrows). (c) TV sonogram obtained 1 later shows low-level echoes that remain within the dilated tube and the ill-defined margin between the ovary (0) and the tube (T) (arrows). UT = uterus. (d) TV sonogram obtamed during follow-up examination performed 2 weeks later for persistent pelvic pain shows a thin, dilated fluid-filled tube (arrow). Laparotomy enabled the identification of a chronically inflamed hydrosalpinx.

longed

antimicrobial

against

anaerobes

therapy plus

close

effective follow-up

for possible surgical drainage (2,4,6). Knowledge of the severity of PID is useful for optimal management. The clinical presentation of patients with salpingitis versus those with more severe disease, however, is similar. Physical

plus a positive culture allow a correct diagnosis in up to 87% of patients, but the physician may miss up to 33% of patients with PID (16). Aggressive

echogenic

Progression

down, creating (2,5,20,21).

have been modified

(14-16).

4.

tube. week

tenderness. There is a high degree of inaccuracy, however, in the clinical diagnosis of PID (1,16). When exammed laparoscopically, only 60%-70% of adult women with the presumptive diagnosis of acute salpingitis actually had the disease (22). Criteria for the diagnosis of PID and subsequently

d.

Figure

follow.

ovary, called Further inthe walls of

b.

C.

peniovarian

the inflamed tubo-ovanian flammation

vaginal

a.

examination

may

result

in an

overestimate or an underestimate of the presence of a pelvic mass (23). Up to 70% of adnexal masses identified at sonography may be missed at physical examination (7). Thus, TA sonography became the modality of choice to exclude the presence of a tuboovarian mass. Although the finding of a pelvic mass

on

a TA

sonogram

is often

me-

femred to as a tubo-ovarian abscess, laparoscopic findings may range from pyosalpinx, to tubo-ovarian complex, to tubo-ovanian abscess. The theory that patients may benefit from more individualized therapy, depending on the severity of the disease, suggests

a need stages

The

to further differentiate of PID (2,3).

introduction

allows for visualization

Patten

the

of TV transducers

improved resolution of pelvic organs

et al (19) correlated

and (24,25).

TV sono-

graphic and laparoscopic 16 women with PID. TV

findings sonography

in

enabled the identification inflamed fallopian tubes periovanian inflammation.

of 93% of and 90% of Ovaries

that appeared enlarged, were indistinct, and had peniovarian fluid collections corresponded to lapanoscopic findings of peniovanian exudate and adhesions. Poorly defined adnexal masses seen at TV sonography corre-

lated

with

amorphous

abscess pathologically. by Patten et al (19),

ment

and

periutermne

tubo-ovarian In the series uterine enlarge-

inflammatory

changes were identified in only 25% of cases. However, because all of these patients had additional adnexal abnormalities, the low sensitivity in identifying uterine inflammatory

May

1992

changes did not alter the for the diagnosis of PID. A limiting factor of our

sensitivity study

was

the lack of surgical confirmation in all but two cases. Our TV sonographic findings, however, are similar to the observations made by Patten et al (19). Our study confirmed that these findings

are

unique

to the

TV

exami-

nation. The improved visualization tubal and ovarian inflammation vaginally

various copy.

allows

stages The

one

to differentiate

of PID

ability

of trans-

without

laparos-

to distinguish

a pyo-

salpinx from a tubo-ovanan complex or tubo-ovamian abscess may allow for more individualized medical and interventional therapy. TV fine-needle aspiration abscess

or drainage of tubo-ovamian has already been reported to

be a successful

and

safe

procedure

(26).

Theme are two possible explanations for the finding of 31 normal TA and TV sonograms in patients who met the clinical criteria for PID. Because of the lack of lapanoscopic confirmation, these patients may represent cases of a false-positive

clinical

diagnosis.

with

either

TV

or TA

several

abdominal

before

fluid

We thank Linda Catena, Brown-Jones, RDMS, and Debra

Carin,

for their

RDMS,

Number

#{149}

2

12.

Tessler

FN,

Perrella

Endovaginal

1.

2.

3.

13.

Lande

tubes.

IM, Hill

Adnexal

5.

approaches to pelvic infections in women. J Reprod Med 1988; 33:159-163. Peterson HB, Galaid El, Zenibnan JM.

SN, Mercer

U, Ismail

MA.

MC, Cosco

FE, Kator

NN.

abnormalities:

sonography.

Radiology

1988;

166:325-332.

14.

15.

16.

17.

18.

19.

20.

Washington AE, Sweet RL, Shafer MB. Pelvic inflammatory disease and its sequelae in adolescents. J Adolesc Health Care 1985; 6:298-310. Hager WD, Eschenbach DA, Spence MR, et al. Criteria for diagnosis and grading of salpingitis. Obstet Cynecol 1983; 61:113114. Hadgu A, Westrom L, Brooks CA, Reynolds GH, Thompson SE. Predicting acute pdvic inflammatory disease: a multivariate analysis. Am J Obstet Gynecol 1986; 155: 954-960. Sweet

RC.

Pelvic

inflammatory

in women.

disease

Infect

Dis Clin

North Am 1987; 1:199-215. Neiman HL, Mendelson EB. Ultrasound evaluation of the ovary. In: Callen PW, ed. Ultrasonography in obstetrics and gynecology. 2nd ed. Philadelphia: Saunders, 1988;

nosis

21.

23.

24.

Fayle J.

Sonography

in acute pelvic inJ Reprod Med 1982;

disease.

10.

Bernstine disease:

11.

Swayne

R. Acute pelvic a clinical follow-up.

a review

in the diag-

of pelvic

inflammatory

and discussion.

Int J Fertil

1986; 31:341-351. Timor-Tnitsch IE, Rottem S. Transvaginal sonography. 2nd ed. New York: Elsevier, 1991; 138-141. Jacobson L, Westrom L. Objectivized diagnosis of acute pelvic inflammatory disease. Am J Obstet Gynecol 1969; 105:10881098. Spaulding

LB.

Gelman

SR,

Wood

SD,

Monif CC. The role of ultrasonography in the management of endometntis/salpingitis/peritonitis. Obstet Gynecol 1979; 53: 442-446. Mendelson EB, Bohm-Velez M,Joseph N, Nelman HL. Cynecologic imaging: com-

and transvagi1988;

166:321-

324. 25.

Tessler

FN,

Sutherland

Schiller

VL,

ML, Grant

nal versus

endovaginal

prospective

study.

Perrella

EC. pelvic

Radiology

RR,

Transabdomisonography:

1989; 170:

553-556.

26.

vanSonnenberg C, Coodacre US-guided

abscesses 1991;

27:312-320.

Laparoscopy

treatment

panison of transabdominal nal sonography. Radiology

N, Neuhoff 5, Cohen H. Pelvic inflammatory disease in adolescents. J Pc-

flammatory

and

disease:

7. Golden

diatr 1989; 114:138-143. Taylor KJ, Wasson JF, Dc Graaff C, Rosenfield AT, Andnole VT. Accuracy of greyscale ultrasound diagnosis of abdominal and pelvic abscesses in 220 patients. Lancet 1978; 1:83-84. Spirtos NJ, Bernstine RL, Crawford WL,

Patten RM, Vincent LM, Wolner-Hanssen I’, Thorpe E. Pelvic inflammatory disease: endovaginal sonography with laparoscopic correlation. J Ultrasound Med 1990; 9:681689. Binstock M, Muzsnai D, Apodaca L, Cold-

man N, Keith L.

Surgical

Pelvic inflammatory disease: review of treatment options. Rev Infect [Ms 1990; 12: 656-664. Landers DV, Sweet RL. Current trends in the diagnosis and treatment of tuboovanan abscess. Am J Obstet Gynecol 1985; 151: 1098-1110.

1987;

et al. of di-

AJR 1989; 153:523-

and cul-de-sac

transvaginal

22.

15:103-107.

9.

AC,

diagnosis

423.

Westrom L. Incidence, prevalence and trends of acute PID and its consequences in industrialized countries. Am J Obstet Cynecol 1980; 138:880-892. Monif CRC. Clinical staging of acute bactenial salpingitis and its therapeutic ramifications. Am J Obstet Gynecol 1982; 143: 489-495. Kirshon B, Faro 5, Phillips LE, Pruett K. Correlation of ultrasonography and bacteriology of the endocervix and posterior cul-de-sac of patients with severe pelvic inflammatory disease. Sex Trans Dis 1988; Hajj

8.

Fleischer

525.

assistance.

4.

6.

RR,

sonographic

lated fallopian

References

logic 755.

183

themsonogra-

RDMS, Cathy

Acknowledgments:

E, D’Agostino HB, Casola BW, Sanchez RB, Taylor B. transvaginal drainage of pelvic

and fluid collections.

Radiology

181 :53-56.

inflammatory IntJ Fertil

32:229-232.

LC, Love

inflammatory

Volume

TV

and infertility

col-

lections, enlarged uteri, and laterally displaced ovaries missed at TV sonography, these findings did not add to the sensitivity in the diagnosis of acute PID. It is tempting to state that TA sonography is needed only as a backup examination. However, our sonographers had the benefit of the

to orient

performing

phy. A study in which TV and TA sonograms are obtained independently is needed to conclude that only TV sonograms are necessary for the evaluation of acute PID. In summary, TV sonography is a well-tolerated method for the evaluation of adolescents with PID. It provides superior anatomic detail of inflamed adnexa by demonstrating various stages of tubo-ovanian inflammation not previously seen at TA sonography. This additional information, obtained previously only with laparoscopy, may allow for more successful individualized therapy. I

sonogra-

phy. Thus, although TV sonography appears sensitive in the evaluation of PID, its specificity has yet to be determined. Although TA sonography helped identify

TA examination

selves

Pre-

vious series, however, found that up to 87% of patients meeting our clinical criteria had PID when they underwent laparoscopic examination (16). A more satisfying explanation may be that these patients have early or mild tubal erythema, in which case TV sonographic findings might be cxpected to be minimal. Correlation with clinical data remains critical. From our original series of 137 sonograms, five hemorrhagic cysts could not be distinguished from ovarian abscess

initial

correlation.

MB,

disease:

Karasick

SR.

Pelvic

sonographic-patho-

Radiology

1984;

151:751-

Radiology

#{149} 439