See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/10717113

Transabdominal sonography before uterine

exploration as a predictor of retained placental
fragments
ARTICLE in JOURNAL OF ULTRASOUND IN MEDICINE: OFFICIAL JOURNAL OF THE AMERICAN INSTITUTE OF
ULTRASOUND IN MEDICINE JULY 2003
Impact Factor: 1.54 Source: PubMed

CITATIONS

READS

85

4 AUTHORS, INCLUDING:
Ori Shen

Ron Rabinowitz

Shaare Zedek Medical Center

Shaare Zedek Medical Center

82 PUBLICATIONS 438 CITATIONS

96 PUBLICATIONS 1,370 CITATIONS

SEE PROFILE

SEE PROFILE

Vered H Eisenberg
Sheba Medical Center
68 PUBLICATIONS 549 CITATIONS
SEE PROFILE

Available from: Vered H Eisenberg

Retrieved on: 10 November 2015

Article

Transabdominal Sonography
Before Uterine Exploration
as a Predictor of Retained
Placental Fragments
Ori Shen, MD, Ron Rabinowitz, MD, Vered H. Eisenberg, MD,
Arnon Samueloff, MD

Objective. To evaluate the diagnostic accuracy of sonography in postpartum patients thought to have
retained placental fragments. Methods. The study group consisted of 39 postpartum women in
whom inspection of the placenta brought up suspicion of retained placental fragments. All these
women underwent manual exploration of the uterine cavity. Before the procedure, all patients had
two-dimensional sonographic imaging, after which they were divided into 2 groups. The first group
comprised women who were judged to have had an empty uterus or nothing but intrauterine fluid
collection. The second group consisted of those in whom sonography showed echoes of what might
appear as residual trophoblastic tissue, that is, echogenic, hypoechoic, or mixed echo intracavitary patterns. The sonographic patterns were then correlated with the presence or absence of retained placental fragments as found on uterine manual exploration and pathologic examination. Results. In 18
patients, no suspected contents were observed on sonography. In 17 of these patients, the uterus was
empty on manual uterine exploration. One of these patients had residual trophoblastic tissue, which
was of minimal quantity and clinically unimportant. In 21 patients, sonography suggested retained placental tissue. In 15 of these patients, pathologic examination confirmed residual trophoblastic tissue,
and in the remaining 6, the uterus contained blood clots, decidua, or both. Conclusions. Sonography
is an effective tool for evaluating postpartum patients thought to have retained placental fragments.
Normal sonographic findings might obviate the need for manual exploration of the uterine cavity. A
questionable sonographic result is not an effective tool for distinguishing between placental fragments
and blood clots. Key words: placenta; sonography; third stage of labor.

R
Received December 24, 2002, from the Department
of Obstetrics and Gynecology, Shaare Zedek
Medical Center, Jerusalem, Israel; and Ben-Gurion
University of the Negev, Beer-Sheva, Israel. Revision
requested January 6, 2003. Revised manuscript
accepted for publication February 20, 2003.
Address correspondence and reprint requests to
Ori Shen, MD, Department of Obstetrics and
Gynecology, Shaare Zedek Medical Center, PO Box
3235, Jerusalem 91031, Israel; e-mail: orishen@
netvision.net.il.

esidual trophoblastic tissue is an infrequent complication of labor and delivery and might appear
as an early or late postpartum hemorrhage.1 It
might also predispose to puerperal infection.
This situation is suspected when routine examination of
the placenta reveals an incomplete placenta.
Because accurate diagnosis of retained placental fragments cannot always be made from examination of the
placenta, whenever this possibility is considered, manual exploration of the uterine cavity is performed. This
procedure requires general or regional anesthesia. In an
attempt to decrease the high false-positive rate of the
clinical evaluation, attempts have been made to use
sonography to enhance the predictive value, thus avoiding unnecessary interventions.

2003 by the American Institute of Ultrasound in Medicine J Ultrasound Med 22:561564, 2003 0278-4297/03/$3.50

Transabdominal Sonography for Retained Placental Fragments

Longitudinal studies have been performed2,3 to

study the appearance of the normal uterus in involution during puerperium. Several studies have
attempted to describe the appearance of the
postpartum and postabortion uterus in cases of
postpartum hemorrhage with the use of transabdominal sonography.4,5 Others have used transvaginal duplex Doppler sonography in similar
cases. On the basis of these studies and others,6 it
would appear that in patients with postpartum or
postabortion bleeding, an empty uterus or a uterus
containing only fluid, especially when Doppler
studies reveal absent or scanty high-resistance flow
in the endometrium, have a high negative predictive value for retained placental fragments.7,8 It has
been suggested that unnecessary interventions
might be avoided in such patients.
We conducted this study to examine the diagnostic accuracy of combined transabdominal and
transvaginal sonography in patients with no excessive bleeding but who were thought to have
incomplete placentas on postpartum examination. It was our intent to offer the clinician in the
delivery department an effective, simple tool that
would allow decisions regarding the need to perform uterine exploration in these cases.

Materials and Methods

This study was conducted in concordance with
departmental protocol, and other than the sonography, there was no change in patient treatment
from the standard protocol. The study was
approved by the local Research Committee.
The study group consisted of 39 patients in the
immediate postpartum period. All had vaginal
delivery of singleton neonates at 36 to 41 weeks
gestational age (mean, 38 weeks 6 days) between
March 2001 and October 2002 at Shaare Zedek
Medical Center. In all cases, the placenta was delivered spontaneously within 30 minutes of delivery.
The mean age SD for the study group was 28 4.2
years. Parity was 3.6 1.4. Twenty-five patients
(64%) received regional anesthesia during labor
and delivery.
In all cases, examination of the placenta by
2 examiners revealed suspected incomplete placentas. A decision to perform manual exploration
of the uterus was made by the staff on the basis of
placental appearance alone. None of the patients
had a postpartum hemorrhage. Cases in which it
was obvious that a cotyledon was missing were
excluded from the study.
562

A two-dimensional scan of the uterine cavity

was performed initially by an abdominal
approach (25 MHz, SonoAce 5500; Medison Co,
Ltd, Seoul, South Korea), and if the abdominal
scan showed no abnormalities, a transvaginal
scan (57.5 MHz) was performed to verify the initial finding. All scans were performed by an experienced sonographer (O.S. or R.R.) within 1 hour
of delivery. The sonographers did not examine
questionable placentas, nor did they discuss
their appearance with the staff. The scans were
categorized either as normal, signifying an
empty uterus or fluid only, or as having questionable uterine content, when there was a distinct
mass containing echogenic, hypoechoic, or
mixed echogenic patterns measuring greater
than 10 mm in thickness.
The clinicians performing the manual exploration were not informed of the sonographic
findings. Uterine exploration was performed
under general or epidural anesthesia. Any uterine content retrieved on exploration was sent for
pathologic examination.

Results
The uterus was judged to be empty in 18 cases:
17 with a midline echo measuring 10 mm or less
and 1 containing only fluid (negative sonographic findings). Figures 1 and 2 illustrate, respectively, 1 case with negative sonographic findings and
a second case with positive findings. In 21 cases,
Figure 1. Sonogram showing no questionable content.

J Ultrasound Med 22:561564, 2003

Shen et al

sonography revealed questionable uterine content (distinct masses with a thickness of >10
mm). In 7 of these, there was echogenic matter in
the uterine cavity, and in 14, a mixed echo pattern was observed. Table 1 shows a comparison
of the sonographic and clinicopathologic findings. In the single case with false-negative findings, the amount of chorionic tissue found on
pathologic examination was minimal (<5 cm3)
and unlikely to be of any clinical importance. In
this case, a midline echo of 10 mm was shown.
The sensitivity of sonography for predicting
retained products of conception was 93.8%.
Specificity was 73.9%. The measurement of
agreement between the 2 measurements was
0.646 and was statistically significant (P < .001).
In all cases with positive sonographic findings,
the questionable content was shown on transabdominal scans. In all cases with normal findings,
transvaginal scans were confirmatory, failing to
show additional pathologic entities. All patients
had benign courses and were discharged 48 to 72
hours postpartum. None was readmitted to our
institution for late febrile or hemorrhagic complications.

Discussion
Previous studies have attempted to predict
retained products of conception by sonography.
Using transabdominal sonography in 53 patients
with postpartum hemorrhage, Hertzberg and
Figure 2. Sonogram showing questionable uterine content.

Bowie,5 were able to predict an empty uterus in

24 patients with negative sonographic findings.
Achiron et al7 used a transvaginal approach in a
mixed group of postpartum and postabortion
patients with late postpartum or postabortion
hemorrhage and were able to predict a benign
clinical course when the uterus was judged to be
empty. No pathologic correlation was available,
because conservative treatment was selected in
these patients. They were able to distinguish
between residual trophoblasts and clots or decidual debris with Doppler studies. Wolman et al9
and Tal et al10 reported on their experience with
sonohysterography to predict retained tissue in
bleeding postabortion and postpartum patients.
We chose to examine a specific subset of
patients, that is, asymptomatic postpartum
patients, thought to have incomplete placentas.
In 41% of our patients, retained placental tissue
was recovered. Our results establish that for this
set of patients, a high negative predictive value
can be obtained with two-dimensional sonography with no need for specialized additional testing. Images obtained by the transabdominal
route of the uterine cavity have proved satisfactory for correctly predicting an empty uterus.
Our results indicate that transvaginal scanning
does not add useful information to an abdominal examination with normal findings. We speculate that this might be due to the proximity of
the enlarged uterus to the abdominal wall.
A scan with positive findings apparently lacks
sufficient accuracy to distinguish retained tissue
from blood clots or decidua. Because it was our
intention to offer a practical and readily available
technique that can be applied immediately in the
delivery room area, we chose not to use color
Doppler sonography or sonohysterography to
assist in this distinction, as described by others.710
Our study establishes the clinicopathologic
relationship in the specific clinical situation of
possible retained placental fragments immediately after birth in asymptomatic patients. The

Table 1. Correlation Between Sonographic and Clinicopathologic

Findings

Sonographic Findings

Negative (no questionable content)

Positive (questionable uterine content)

J Ultrasound Med 22:561564, 2003

Clinicopathologic Findings
Negative
Positive
(No Trophoblast) (Trophoblast Tissue)

17
6

1
15

563

Transabdominal Sonography for Retained Placental Fragments

transvaginal approach might be less agreeable to

a postpartum patient, and it might not be available in the delivery room setup. The transabdominal approach was found to be sufficiently
accurate for these patients. We found sonography to be insufficiently accurate in distinguishing retained trophoblast from blood clots or
decidua.
In this study, sonography was performed by
expert sonographers. The clinician would not
have access to such specialist consultation at all
times. Because we think it is feasible with little
effort to train residents to identify an empty
uterus in the postpartum patient according to
the described technique, we are now conducting
a study in which, on the same patient group, residents with minimal training in sonography will
perform the examinations to establish whether
these results are valid when the sonography is
performed by nonexperts.
When routine examination of the placenta raises suspicion of a missing placental fragment, the
standard of care is to perform uterine exploration
with anesthesia. Our data support the hypothesis
that if a transabdominal sonogram in such a case
suggests an empty uterus, invasive intervention
should be avoided. In our group, 46% of uterine
cavity explorations might have been avoided.
When abdominal or vaginal sonography or both
show a distinct intrauterine mass larger than 10
mm thickness, uterine cavity exploration is warranted.

5.

Hertzberg BS, Bowie JD. Ultrasound of the postpartum uterus: prediction of retained placental tissue.
J Ultrasound Med 1991; 10:451456.

6.

Herman A. Complicated third stage of labor: time to

switch on the scanner. Ultrasound Obstet Gynecol
2000; 15:8995.

7.

Achiron R, Goldenberg M, Lipitz S, Mashiach S.

Transvaginal duplex Doppler ultrasonography in
bleeding patients suspected of having residual trophoblastic tissue. Obstet Gynecol 1993; 81:507
511.

8.

Alcazar JL. Transvaginal ultrasonography combined

with color velocity imaging and pulsed Doppler to
detect residual trophoblastic tissue. Ultrasound
Obstet Gynecol 1998; 11:5458.

9.

Wolman I, Gordon D, Yaron Y, Kupferminc M,

Lessing JB, Jaffa AJ. Transvaginal sonohysterography
for the evaluation and treatment of retained products of conception. Gynecol Obstet Invest 2000;
50:7376.

10. Tal J, Timor-Tritsch I, Degani S. Accurate diagnosis of

postabortion placental remnant by sonohysterography and color Doppler sonographic studies. Gynecol
Obstet Invest 1997; 43:131134.

References

564

1.

Common complications of pregnancy. In:

Cunningham FG, MacDonald PC, Gant NF (eds).
Williams Obstetrics. 21st ed. Int ed. New York, NY:
McGraw-Hill; 2001:635652.

2.

Edwards A, Ellwood DA. Ultrasonographic evaluation of the postpartum uterus. Ultrasound Obstet
Gynecol 2000; 16:640643.

3.

Mulic-Lutvica A, Bekuretsion M, Bakos O, Axelsson

O. Ultrasonic evaluation of the uterus and uterine
cavity after normal, vaginal delivery. Ultrasound
Obstet Gynecol 2001; 18:491498.

4.

Lee CY, Madrazo B, Drukker BH. Ultrasonic evaluation of the postpartum uterus in the management of
postpartum bleeding. Obstet Gynecol 1981; 58:
227232.

J Ultrasound Med 22:561564, 2003