Sheared Epidural Catheter
During an Elective Procedure
Karen Noblett, MD, Amanda McKinney, MD,
and Richard Kim, MD
BACKGROUND: Epidural catheter fracture and retention is uncommon. Imaging techniques identifying catheter location are limited, and clear management guidelines are lacking. Radiculopathy and lumbar stenosis are
recognized complications.
CASE: During epidural placement in an elderly woman
undergoing gynecologic surgery, a portion of the catheter
was shorn off in the epidural space. Imaging, including
plain X-ray and computed tomography scan had limited
ability to identify the catheter. The initial neurosurgical
consultation recommended leaving the catheter in place.
However, the patient became symptomatic with radicular pain; 16.5 cm of the catheter were subsequently
surgically removed.
CONCLUSION: Retained epidural catheter fragments
can be associated with long-term sequelae. Immediate
removal is facilitated by a blood-clot tract and may
present the best time for removal. Current imaging
techniques have limited use in these situations.
(Obstet Gynecol 2007;109:566–8)

E

pidural catheter fracture and retention is uncommon
considering the large number of cases performed
each year. Although epidural catheters are labeled as
being radiopaque, imaging techniques for identifying
catheter location have been limited. Additionally, clear
management guidelines are lacking. The most commonly described scenario involves only the tip of the
catheter being sheared off, with retention in the epidural
space. In this situation, most recommend leaving the
fragment in place given the limited ability to visualize
the catheter radiologically, and the invasiveness of surgical removal. However, complications such as radiculopathy, radicular pain, and lumbar stenosis are recognized complications, and can occur at a time remote
from the incident. We present a case where more than
16 cm of catheter was sheared off, yet was still met with
some controversy over management.

From the Irvine Medical Center, University of California, Orange, California.
Corresponding author: Karen L. Noblett, MD, University of California, Irvine,
101 City Drive, Dept. OB/GYN Rt. 56, Orange, CA 92868; e-mail:
knoblett@uci.edu.
© 2007 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/07

566

Noblett et al

CASE
A healthy 78-year-old woman presented for elective surgery involving a posterior repair with graft augmentation
and midurethral sling for her symptomatic pelvic prolapse
and urinary incontinence. Her surgical history was significant for an abdominal hysterectomy 30 years prior for
uterine fibroids. Given her options, the patient elected to
have epidural anesthesia primarily to avoid postoperative
nausea. A 36-inch, 20-gauge, radiopaque polytetrafluoroethylene (Teflon, E.I. du Pont de Nemours and Company,
Wilmington, DE) epidural catheter was placed through a
17-gauge Tuohy needle at the level of L-3–L-4. Initially,
resistance was met passing the catheter, but with slight
alteration of the angle, the catheter was eventually threaded
into the epidural space. During removal of the Tuohy
needle, before it exited the level of the skin, the patient
suddenly shifted position, jerking upward, and the catheter
was shorn off at the tip of the needle, leaving, at that point,
an undetermined portion of the catheter in the epidural
space.
Assessment at the time estimated that 12 cm of the
catheter was retained. Subsequently, a second epidural
catheter was placed at the level of L-2–L-3 without incident,
in an effort to proceed with the planned surgery and to
provide a comparative reference for later imaging and
removal of the retained catheter. However, upon further
consideration, given the uncertainty of the location of the
broken catheter, and concern that the required dorsal
lithotomy positioning for the surgery may cause migration
of the catheter, the decision was made to postpone the
surgery.
In the operating room, plain radiographs were obtained
in the anterior–posterior and lateral positions. Unfortunately, these images did not provide definitive visual confirmation of the catheter location, and the second epidural
catheter was subsequently removed and a computed tomography (CT) scan was obtained. The CT scan identified
approximately 1 cm of the catheter at the level of L-3 (Fig.
1). The patient was admitted for observation and a neurosurgery consultation was requested. Upon transfer to the
floor, the patient reported moderate back pain at the site of
the epidural, but denied any motor or sensory deficits of the
lower extremities. Neurologic examination at that time
confirmed the patient’s claim of no focal neurologic deficit.
Per consultation with Baxter Pharmaceutical (the manufacturer of the epidural catheter) and a neuroradiologist, a
second set of plain radiographs were taken in the lateral
position. The X-rays were obtained after imaging an identical catheter placed on a towel and adjusting the contrast
of the X-ray to enhance the ability to visualize the catheter.
Using this technique, several centimeters of the catheter
were visualized (Fig. 2). The recommendation by the neurosurgeon initially consulted was to leave the catheter
fragment in place, citing the patient’s symptoms not being
significant and the risk of surgery outweighing the benefit of
removing it. The team managing the patient’s care felt that
the patient’s clinical status warranted a second opinion.

Retained Epidural Catheter During Surgery

OBSTETRICS & GYNECOLOGY

Fig. 1. Axial view of a portion of the epidural catheter
(arrow), appearing as just a fragment.
Noblett. Retained Epidural Catheter During Surgery. Obstet
Gynecol 2007.

The opinion of the second neurosurgeon was that the
catheter should be removed based on the patient’s clinical
status of escalating pain as well as his view that the surgery
carried minimal risk. After a lengthy discussion with the
patient and her family regarding the risks and benefits of
leaving the catheter in place compared with removing it,
the patient ultimately decided for catheter removal. Shortly
after her decision, the patient developed severe radicular
pain in her right lower extremity. Again, no motor deficit
was noted.
The patient was taken back to surgery and underwent a
partial laminotomy between L-3 and L-4. The neurosurgeon
indicated that identifying the location of the catheter was
facilitated by formation of a blood clot tract that led to its
point of entrance into the epidural space. Approximately
16.5 cm of epidural catheter was removed from the lumbar
epidural space. (Fig. 3). The procedure was uncomplicated
and was completed in approximately 25 minutes. The
patient was discharged to home on postoperative day 1,
with resolution of her radicular pain and improvement of
her low back pain. At 5 weeks postoperatively, the patient
remained symptom free and able to carry out all her
activities of daily living.

COMMENT
The incidence of epidural catheter fracture and retention is a relatively uncommon occurrence considering
the large numbers of epidurals that are placed each
year. There are no current data reporting the incidence of this complication. The usual recommendation is to leave the fragment in place, because com-

VOL. 109, NO. 2, PART 2, FEBRUARY 2007

plications are rare. However, this recommendation
generally applies only when a small fragment of the
catheter is involved, which is the most common
scenario.1 Removal is recommended when the catheter is at or near the level of the skin, secondary to its
potential for causing infection, or when the patient is
symptomatic from the retained fragment. There have
been several cases in the literature where larger
fragments have been left in place, ultimately resulting
in the development of symptoms that eventually
required removal of the catheter. Radiculopathy, radicular pain, and lumbar stenosis are all recognized
complications of retained catheter fragments.2– 4
Although the catheter is labeled as radiopaque,
this claim was not validated by our imaging studies.
Our CT scan only revealed a small portion of the
catheter; however, other reports in the literature
demonstrated greater success with this modality.5,6
Additionally, it was only with significant modification
of our plain films that we were finally able to determine that more than just a small fragment of the
catheter had been retained. Even with these modifications, visualization was less than optimal. The fact
that our imaging studies did not definitively confirm
the amount of catheter retained led to the controversy
as to whether the catheter should be removed or left
in place.
This case report brings up several important
points, particularly for obstetrician– gynecologists
who are involved in a high number of cases where
epidurals are commonly used. Because of the rare
occurrence of this type of complication, there are
insufficient data regarding criteria for intervention.
However, what data exists suggest that retention of
larger fragments warrants immediate removal secondary to delayed sequelae. The decision of our patient
for immediate removal supports this conclusion, because she began experiencing significant radicular
pain that immediately resolved after surgery. Additionally, the success of immediate removal was facilitated by the presence of the blood clot tract that led
to the location of the catheter. The blood clot tract
would not be present had the surgery been delayed
and would have made any eventual need for removal
of the catheter more complicated. Another important
point that should be gleaned from this experience is
that when managing a patient who requires specialty
care, obtaining a second opinion when the management is in doubt is not only appropriate but is
mandated. Because there is such limited data on what
to do in these situations, it is to be expected that there
would be differing opinions in management. Fortunately, this patient received the appropriate care in

Noblett et al

Retained Epidural Catheter During Surgery

567

Fig. 2. A. Sagittal view of the
retained portion of the epidural
catheter under plain view. Arrows
indicate the location of the catheter when viewed under magnification. B. Sagittal section of the epidural catheter highlighted for
easier identification.
Noblett. Retained Epidural Catheter
During Surgery. Obstet Gynecol
2007.

ufacturers. This would facilitate removal of the catheter at the time of surgery, particularly if removal is
done at a time remote from the original incident.
REFERENCES
1. Bromage PR, Epidural analgesia. Philadelphia (PA): Saunders;
1978. p. 664–6.
2. Perkins WJ, Davis DH, Huntoon MA, Horlocker TT. A
retained Racz catheter fragment after epidural neurolysis:
implications during magnetic resonance imaging. Anesth
Analg 2003;96:1717–9.

Fig. 3. Retained portion of the epidural catheter after
removal.
Noblett. Retained Epidural Catheter During Surgery. Obstet
Gynecol 2007.

this situation and had an excellent outcome. Finally,
improved visualization of the epidural catheter with
imaging techniques must be addressed with the man-

568

Noblett et al

3. Blanchard N, Clabeau JJ, Ossart M, Dekens J, Legars D,
Tchaoussoff J. Radicular pain due to a retained fragment of
epidural catheter. Anesthesiology 1997;87:1567–9.
4. Staats PS, Stinson MS, Lee RR. Lumbar stenosis complicating
retained epidural catheter tip. Anesthesiology 1995;83:1115–8.
5. Ugboma S, Au-Truong X, Kranzler LI, Rifai SH, Joseph NJ,
Salem MR. The breaking of an intrathecally-placed epidural
catheter during extraction. Anesth Analg 2002;95:1087–9.
6. Moore DC, Artru AA, Kelly WA, Jenkins D. Use of computed
tomography to locate a sheared epidural catheter. Anesth
Analg 1987;66:795–6.

Retained Epidural Catheter During Surgery

OBSTETRICS & GYNECOLOGY

Occurrence of Postoperative
Hematomas After Prolapse
Repair Using a Mesh
Augmentation System
Christine A. LaSala, MD,
and Megan O. Schimpf, MD
BACKGROUND: Mesh-augmented repair kits are increasingly being used for anterior compartment defect
repair and can be associated with complications.
CASES: Two cases are presented of postoperative pelvic
hematomas confirmed on computerized tomography
scan after anterior repair with a mesh-augmented anterior vaginal prolapse repair system. Both resolved without requiring drainage, although one patient did require
a blood transfusion, and the other required intravenous
antibiotics to treat an infected hematoma.
CONCLUSION: Significant complications may arise from
this new technology. Further studies are needed to
determine whether the benefits of this technology outweigh its risk of use.
(Obstet Gynecol 2007;109:569–72)

I

n the past year, several mesh augmentation systems
for pelvic reconstructive surgery have been introduced to the market. These systems are increasingly
being used for anterior compartment defect repair.
Initial reports from the manufacturer have included a
2.5% risk of postoperative complications, including a
1.75% risk of hematoma.1
We report two cases of large pelvic hematomas that
occurred after an anterior mesh augmentation procedure, which were successfully managed conservatively.

discussion of potential risks of the procedure, including
injury to adjacent organs, hemorrhage requiring blood
transfusion, mesh erosion, and failure of the procedure. She
underwent an uncomplicated anterior repair using a mesh
augmentation system (Prolift, Gynecare, Ethicon, Johnson &
Johnson, Somerville, NJ), posterior repair and cystourethroscopy. She received antibiotic prophylaxis at the time of
surgery. Bupivacaine with epinephrine and lidocaine were
infiltrated into the anterior vaginal wall to assist with
hydrodissection and hemostasis. A 3– 4-cm full-thickness
midline longitudinal vaginal incision was made approximately 3 cm from the urethral meatus, extending to 2–3 cm
from the vaginal cuff. The vaginal muscularis was not
dissected off or split from the vaginal wall. The retropubic
space was entered bluntly and sharply, allowing palpation
of the ischial spine and visualization of the arcus tendineus
fascia pelvis. The mesh was soaked in bacitracin solution
before placement and irrigated copiously after insertion.
The vagina was packed with dry plain gauze, which was
removed on postoperative day 1. Antibiotic coverage continued intravenously for 24 hours.
Hematocrit on the evening of surgery was 26.3. She had
low urine output of 50 mL over an 8-hour shift that
improved after she received a 500-mL intravenous fluid
bolus and 10-mg intravenous furosemide. Hematocrit on
postoperative day 1 was 23.0, and a suprapubic mass was
palpable on bimanual pelvic examination, consistent with a
hematoma. Computed tomography (CT) scan confirmed a
7⫻7– cm right pelvic retropubic hematoma with a moderate mass effect on the bladder (Fig. 1).
Repeat hematocrit 6 hours later was 19.3. The patient was
asymptomatic with stable vital signs but was transfused two
units of packed red blood cells because of her advanced age
and cardiac disease. Posttransfusion hematocrit was 28.5.
She voided spontaneously and completely on postoperative day 2. She was discharged to home on postoperative

CASES
Patient 1 was a postmenopausal woman with a symptomatic
stage II anterior vaginal wall prolapse using the Pelvic Organ
Prolapse Quantification staging. The anterior vaginal wall
prolapse at its greatest extent was 2 cm beyond the hymenal
ring. Multichannel urodynamics failed to show any potential
urinary incontinence with the prolapse reduced.
The patient signed informed consent after a thorough

From the Division of Urogynecology, Department of Obstetrics and Gynecology,
Hartford Hospital, Hartford, Connecticut.
Corresponding author: Christine A. LaSala, MD, 85 Seymour Street, Suite 525,
Hartford, CT 06106; e-mail: clasala@harthosp.org.
© 2007 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/06

VOL. 109, NO. 2, PART 2, FEBRUARY 2007

Fig. 1. Case 1 pelvic computed tomography scan reveals retropubic hematoma (H) compressing bladder (B). L, left; R, right.
LaSala. Pelvic Hematoma After Mesh-Augmented Surgery. Obstet
Gynecol 2007.

LaSala and Schimpf

Pelvic Hematoma After Mesh-Augmented Surgery

569

day 3 and had an otherwise uncomplicated postoperative
course. Follow-up pelvic exams in the office confirmed the
slow resolution of the hematoma, with complete resolution
by 6 months after surgery. The repair was not compromised
by the hematoma.
Patient 2 was a postmenopausal woman with a stage III
anterior wall prolapse, stage II uterine prolapse, and a stage I
rectocele by Pelvic Organ Prolapse Quantification staging.
She had used a pessary in the past, with resultant stress urinary
incontinence. She elected to undergo definitive surgery.
An anterior repair using mesh augmentation with Prolift
(Gynecare), cystourethroscopy, and posterior repair was
performed. A midurethral, retropubic tension-free vaginal
tape (TVT, Gynecare, Johnson & Johnson, Somerville, NJ)
for her confirmed stress urinary incontinence. Informed
consent was obtained as with patient 1.
Surgery was performed without difficulty in a manner
similar to patient 1. Hematocrit on the evening of surgery
was 38.5 and was stable the next morning at 36.7. The
patient had an uncomplicated hospital course and was
discharged to home on postoperative day 3. She was sent
home with a urethral catheter in place because of urinary
retention. She was given nitrofurantoin 100 mg daily for
prophylaxis.
She returned to the office for voiding trials on postoperative day 7 reporting a temperature of 38.9°C with general
malaise, although she denied shortness of breath, chest
pain or worsening post-operative pain. Urine dipstick analysis was negative. Pelvic examination confirmed a
4⫻4 – cm palpable retropubic hematoma. She had perianal,
bilateral, nontender ecchymoses, presumably from the posterior repair. There were no palpable posterior vaginal wall
masses. The remainder of her physical examination was
unremarkable.
The patient was admitted with a diagnosis of presumed
infected hematoma and given intravenous moxifloxacin
and metronidazole. White blood cell count was 22.9, and
hematocrit was 31.3, which remained stable on later studies. Computed tomography scan confirmed a heterogeneous collection between the bladder and cervix, measuring 8.3⫻3.2 cm, which was consistent with a hematoma or
possibly an abscess (Fig. 2). Her temperature varied between 37.5–38.9°C over the next 3 days, although her
white blood cell count decreased to 15. Magnetic resonance imaging was performed in an attempt to differentiate
between hematoma and abscess. This showed two focal
collections in the pelvis: a 6⫻2– cm mass between the
bladder and cervix as previously seen on CT scan as well as
a second collection consistent with a hematoma measuring
3⫻6 cm in the left obturator internus muscle (Fig. 3). After
remaining afebrile on intravenous antibiotics for 48 hours,
she was changed to oral metronidazole and levofloxacin.
She was discharged home on hospital day 6. At no time did
the patient have increased vaginal bleeding or discharge.
The hematoma was no longer palpable 6 weeks postoperatively. The postoperative examination showed no
prolapse.

570

LaSala and Schimpf

Fig. 2. Case 2 pelvic computed tomography scan reveals
hematoma (H) lying between bladder (B) and cervix (C). L,
left; R, right.
LaSala. Pelvic Hematoma After Mesh-Augmented Surgery. Obstet
Gynecol 2007.

Fig. 3. Case 2 pelvic magnetic resonance imaging reveals
two hematomas (H) lateral to the bladder (B). L, left; R,
right.
LaSala. Pelvic Hematoma After Mesh-Augmented Surgery. Obstet
Gynecol 2007.

COMMENT
As with any surgical innovation, caution with new
technology is warranted not only in the interest of
patient safety but also to assure that the product is at
least as efficacious, less invasive, or offers reduced
morbidity as compared with currently used techniques. Despite limited evidence-based medicine concerning these procedures, they are being marketed
widely, and sometimes to surgeons who may not be
completely familiar with the pertinent anatomy.
A MEDLINE search of the English literature
from January 1966 to September 2006 using the terms

Pelvic Hematoma After Mesh-Augmented Surgery

OBSTETRICS & GYNECOLOGY

“extraperitoneal,” “colpopexy,” “hematoma,” “Prolift,” “mesh,” “anterior repair,” and “graft” failed to
find any reports of pelvic hematomas resulting from
use of one of the mesh augmentation systems. There
are case reports regarding hematomas and infected
hematomas associated with transobturator midurethral slings.2,3 The superficial strap passage of the
Anterior Prolift (Gynecare) is similar to the passage of
the transobturator sling, and it is reasonable that
similar complications with these devices may occur.
Previously published articles reviewing surgical
outcomes and complications of different techniques in
the repair of prolapse of the anterior vaginal compartment are few. The vaginal–paravaginal repair is the
nearest comparison to the anterior mesh augmentation procedures with respect to anatomic dissection.
Young et al4 reported on 100 women who underwent
vaginal–paravaginal repair with a 16% (16/100) transfusion rate. No hematomas or surgical-site infections
were reported in that study.4
In a prospective, randomized, controlled study
evaluating the efficacy of three different techniques
for anterior colporrhaphy, including the use of polyglactin absorbable mesh, Weber et al5 concluded that
the use of mesh did not improve surgical outcome.
Overall, the anatomic cure rate of anterior colporrhaphy approximated 45%.5 A review by Maher et al6 on
the surgical management of anterior vaginal wall
prolapse found that the overall success rates for
anterior colporrhaphy vary between 37% and 100%.
A Cochrane Database review on the surgical management of the anterior vaginal wall found that standard
anterior repair was associated with more recurrent
cystoceles than a polyglactin mesh augmented repair
(relative risk 1.39, 95% confidence interval 1.02–
1.90), but data on morbidity and other clinical outcomes were too few to compare.7
The role of graft augmentation in an attempt to
improve surgical success rates is a reasonable concept
and preliminary results seem favorable. However,
graft augmentation would ideally be associated with
minimal morbidity and complications compared with
more traditional approaches.
The Anterior Prolift (Gynecare) procedure involves placing a polypropylene mesh between the
anterior vaginal wall and the bladder in attempt to
mimic the native anterior endopelvic fascia as it
extends between the right and left arcus tendineus
fascia pelvis (‘white line”). It does not rely on any
native tissue for its attachment points, but rather
passes behind the obturator muscle and through the
obturator membrane.

VOL. 109, NO. 2, PART 2, FEBRUARY 2007

Abstracts reporting on the use of the Prolift
(Gynecare) mesh augmentation systems report a less
than 3% overall complication rate and a less than 2%
incidence of hemorrhage of 687 women evaluated in
one study and 180 in the other.1,8
Although the cause of pelvic organ prolapse may
be multifactorial, the poor integrity of the endopelvic
fascia is thought to be one of the contributing factors.
Reliance on endopelvic fascial integrity is, therefore,
avoided when using this mesh graft.
The procedure is done blindly with finger-guidance throughout each trocar pass. The mesh is placed
through trocars that are inserted through the obturator foramen and behind the obturator internus muscle. Pertinent structures in the anatomic plane include
the obturator vessels and nerves as well as the plexus
of veins within the endopelvic fascia. Venous oozing
is common and often controlled with tamponade.
However, the retropubic, retroperitoneal, and pararectal spaces may accommodate significant occult
bleeding. The obturator vessels traverse within 2 cm
of the anteromedial edge of the obturator foramen.
This is particularly important during the superficial
passes of the Prolift (Gynecare) mesh.
We present the cases of two patients of the 39
patients who have undergone this procedure who
experienced pelvic hematomas after a mesh-augmented anterior repair. Patient 1 had a retropubic
hematoma that likely occurred after completion of the
procedure, because hemostasis had been assured. The
origin of the bleeding is not known. The obturator
vein, or one of its branches, may have been lacerated
with the superior trocar placement. Patient 2 apparently had two separate hematomas. The one between
the bladder and cervix likely originated from a vein
underneath the anterior vaginal wall that may have
initially been in vasospasm from the infiltration of the
local anesthetic mixture. It is difficult to determine the
origin of the hematoma along the obturator muscle,
because it may have originated from the obturator
vessels or the endopelvic venous plexus. Patient 2 did
not have a retropubic hematoma so we do not think
the hematoma was due to the TVT (Gynecare)
placement.
Of the 39 anterior vaginal prolapse repairs we
have done using mesh augmentation kits, two had
significant complications, for a rate of 5%. Both cases
presented were managed conservatively, without reoperation. This may be the optimal management of
this complication as long as the patient remains stable.
Both patients were doing well at the time of this
report and are pleased with their outcomes. It is too
soon to know the surgical outcomes of this procedure.

LaSala and Schimpf

Pelvic Hematoma After Mesh-Augmented Surgery

571

We are in need of randomized, long-term studies
comparing the traditional approach to anterior colporrhaphy with the mesh-augmented approach with
respect to surgical complications and outcomes.
REFERENCES
1. Cosson M, Caquant F, Collinet P, Rosenthal C, Clave H,
Debodinance P, et al. Prolift (Mesh Gynecare) for pelvic organ
prolapse surgical treatment using the TVM Group Technique:
a retrospective study of 687 patients. Available at: http://
www.jnjgateway.com/public/USENG/PROLIFT_Clinical_
Pelvic_Organ_Prolapse.pdf. Retrieved November 21, 2006.
2. Game X, Mouzin M, Vaessen C, Malavaud B, Sarramon JP,
Rischmann P. Obturator infected hematoma and urethral
erosion following transobturator tape implantation. J Urol
2004;171:1629.
3. Goldman HB. Large thigh abscess after placement of synthetic
transobturator sling. Int Urogynecol J Pelvic Floor Dysfunct
2006;17:295–6.

Postpartum Rupture of an
Intracranial Aneurysm
Nisha Kanani, MD, and Eric Goldszmidt, MD
BACKGROUND: The relative risk of intracerebral hemorrhage during pregnancy and 6 weeks postpartum is
higher than that of the nongravid population.
CASE: A 37-year-old multiparous, previously healthy
woman appeared to have had a seizure on emergence
from general anesthesia for cesarean delivery. Subsequent neuro-imaging revealed a giant unruptured internal carotid artery aneurysm. The day after discharge from
the hospital, with planned outpatient neurosurgery follow-up, she re-presented with a subarachnoid hemorrhage. This ultimately resulted in her death.
CONCLUSION: Given the significant morbidity and
mortality associated with cerebral aneurysms in pregnancy, they should be considered in all cases of acute
neurological deterioration. Also, given the increased risk
of aneurysmal rupture in the gravid patient, expedited
management of newly discovered cerebral aneurysms
should be contemplated.
(Obstet Gynecol 2007;109:572–4)

From the Department of Anesthesia and Pain Management, Mount Sinai
Hospital, University of Toronto, Toronto, Ontario, Canada.
Corresponding author: Dr. Eric Goldszmidt, 600 University Avenue #1514,
Toronto, Ontario, Canada, M5G 1X5; e-mail: e.goldszmidt@utoronto.ca.
© 2007 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/07

572

Kanani and Goldszmidt

4. Young SB, Daman JJ, Bony LG. Vaginal paravaginal repair:
one-year outcomes. Am J Obstet Gynecol 2001;185:1360–6.
5. Weber A, Walters M, Piedmonte M, Ballard L. Anterior
colporrhaphy: a randomized trial of three surgical techniques.
Am J Obstet Gynecol 2001;185:1299–304.
6. Maher C, Baessler K, Glazener CM, Adams EJ, Hagen S.
Surgical management of pelvic organ prolapse in women.
Cochrane Database Syst Rev 2004 Oct 18;4:CD004014.
7. Maher C, Baessler K. Surgical management of anterior vaginal
wall prolapse: an evidence based literature review. Int Urogynecol J Pelvic Floor Dysfunct 2006 17:195–201.
8. Lucente V, Lacquetin B, Miller D, Berrocal J, Clave H,
Cosson M, et al. Transvaginal mesh (TVM): an innovative
approach to placing synthetic mesh transvaginally for surgical correction of pelvic support defects—peri-operative
safety results. Available at: http://www.jnjgateway.com/
public/USENG/PROLIFT_Clinical_TVM.pdf. Retrieved
November 21, 2006.

S

ubarachnoid hemorrhage is the third leading cause
of nonobstetric maternal mortality.1 The incidence
of intracerebral hemorrhage in the puerperium is estimated to range between 3 and 11 per 100,000 pregnancies, well above that of the non-gravid population.2,3 The
increased tendency for aneurysms to develop or rupture
during pregnancy suggests that hemodynamic, hormonal, and other physiologic changes of pregnancy
may have a role in their formation and rupture. We
present the case of a 37-year-old gravida two, para one
woman who was discovered to have a left internal
carotid aneurysm after a seizure on emergence from
general anesthesia for a cesarean delivery. She was
discharged from hospital with an appointment for outpatient neurosurgical follow-up but suffered a fatal subarachnoid hemorrhage 1 week postpartum.

CASE
A 37-year-old multiparous, previously healthy woman, presented to labor and delivery in active labor at 3 cm of cervical
dilation. She subsequently received epidural analgesia for
labor. The patient eventually required an urgent cesarean
delivery for failure to progress. The epidural was topped-up
with a total volume of 25 mL of 2% lidocaine with 1:200,000
epinephrine. She continued to feel pain when tested with an
Allis forcep. Ultimately, the epidural was deemed to be
inadequate, and a general anesthetic was chosen.
A rapid sequence induction was done using 200 mg of
propofol and 120 mg of succinylcholine. Before emergence,
the patient was reversed with neostigmine and glycopyrrolate,
and four twitches were noted on a train of four stimulation at
80 mA. Immediately after emergence and extubation, the
patient was quite agitated and disoriented. She was thrashing
all four limbs and began to desaturate. As oxygen was being

Postpartum Rupture of an Aneurysm

OBSTETRICS & GYNECOLOGY

applied, the patient arched back in a tonic posture, her rolled
eyes back, and she lost consciousness. She began to foam at
the mouth, and manual ventilation with the circuit was
attempted with difficultly. The brief tonic phase was followed
by a period of flaccid paralysis in all four limbs. At this stage
she did not demonstrate any peripheral or corneal reflexes.
Pupils were equal and reactive to light. An urgent head
computed tomography (CT) scan was arranged. Twenty minutes after the apparent seizure, the patient had not regained
consciousness. She was reintubated using 100 mg of succinylcholine and was transferred to the CT scanner, followed by
transfer to the intensive care unit. She was maintained on a
propofol infusion. The head CT with intravenous contrast
showed an area of hypoattenuation in the inferiomedial aspect
of the left frontal lobe most suggestive of an acute infarct. No
evidence of hemorrhage was noted.
Overnight in the intensive care unit, the patient was kept
intubated and sedated with a propofol infusion. She later
regained consciousness and self-extubated. The propofol
was discontinued, and the patient was found to be fully
oriented with no neurological deficits. Her last memory was
of emergence from anesthesia and being instructed to
breathe. The neurology team was consulted that morning
and a brain magnetic resonance imaging (MRI) and electroencephalogram were arranged.
The MRI was done several days later and showed a giant
unruptured left internal carotid aneurysm (2.3⫻2.1⫻1.9
cm) with mass affect on the frontal lobe. There was
indentation of the inferior aspect of the left frontal lobe
without evidence of associated edema. The EEG showed a
subtle nonspecific dysrhythmia in the left frontotemporal
region. The patient did not have any further seizures during
her hospital stay.
Given the MRI findings, a neurosurgeon from the neurosciences center was consulted. Upon reviewing the clinical history and scans it was felt that she was fit for
discharge with follow-up in the aneurysm clinic in 4 weeks.
Although there are no practice guidelines for timing an
intervention for an incidentally discovered aneurysm, the
arranged management plan was consistent with routine
practices at this center. The peripartum state is not usually
taken into special consideration. She was discharged from
the hospital 6 days after the date of her admission.
The day after her discharge from the hospital, the patient
presented to the emergency department of the neurosciences
center with a severe headache and a Glasgow coma scale
score of 15 of 15. Shortly thereafter, she had another seizure
and required intubation. A CT scan confirmed a subarachnoid
hemorrhage inferior to the left Sylvian fissure extending to the
basal cisterns and the fourth ventricle. Later that day, the
patient experienced a sudden tachycardia and hypertension,
with a heart rate of 180 and a blood pressure of 240/110
mmHg. A repeat CT scan showed ventricular extension of the
subarachnoid hemorrhage. The patient lost all brain stem
reflexes other than the respiratory reflex and was extubated.
She died 3 days later. Post mortem examination confirmed a
subarachnoid hemorrhage.

VOL. 109, NO. 2, PART 2, FEBRUARY 2007

COMMENT
Subarachnoid hemorrhage is the third leading cause
of nonobstetric-related maternal death.1 Most aneurysms are diagnosed at the time of leak or rupture.
Interestingly, the initial presentation of the aneurysm
in our case report was not hemorrhage but rather a
seizure on emergence from anesthesia. The mass
effect of the aneurysm was sufficient to create a
seizure focus. The seizure itself may have been precipitated by hypoxia after extubation.
There is a general consensus in the literature
that management for a ruptured aneurysm in the
pregnant population should be no different from in
the nonpregnant population. Discovery of an unruptured aneurysm in pregnancy or the immediate
postpartum period poses an entirely different dilemma. There is some evidence to suggest that
aneurysms are more likely to develop and rupture
during the puerperium. In particular, it has also
been suggested that the time when aneurysms are
most likely to rupture, is in the late stages of
pregnancy and up to 6 weeks postpartum.2,4,5 Further to this, one study suggests that the mortality
from a subarachnoid hemorrhage in pregnant patients is higher than in nonpregnant patients, 85%
versus 30%.6 This study may suggest some urgency
in the management of an unruptured aneurysm in
the obstetric population. A more recent study
showed that the maternal mortality rate in aneurysmal rupture was 35%, similar to the non-gravid
population, but that the mortality from a ruptured
arterio-venous malformation in pregnancy was
higher than in the non-gravid population, 28%
versus 10%.4
There is a physiologic basis to support the findings of an increased tendency for aneurysms to rupture late in pregnancy and in the early postpartum
period. The pregnant state has been associated with
hyperplasia of vessel wall smooth muscle, fragmentation of the internal elastic lamina, and loss of normal
elastic fiber alignment.7 One study showed a doubling
of plasma levels of matrix metalloproteinase-2 in the
last half of pregnancy.8 This is a protein capable of
degrading elastin and fibrillar collagen in vessel walls.
There are also multiple cardiovascular changes capable of contributing to arterial rupture. Pregnancy is a
high-volume, high-output, low-resistance state capable of producing hemodynamic stress on a possibly
weakened arterial wall. Blood volume increases by
50% above prepregnancy levels by the third trimester.
Cardiac output also increases by 40%, mainly due to
an increase in stroke volume. These physiologic

Kanani and Goldszmidt

Postpartum Rupture of an Aneurysm

573

changes may explain why cerebral, aortic, and splenic
aneurysms all present with increased frequency during pregnancy. It also may explain why 50% of all
aneurysmal ruptures in women below the age of 40
years are pregnancy related.9
The quoted incidence of subarachnoid hemorrhage from aneurysmal rupture in pregnancy has
ranged from 3 to 11 per 100,000 pregnancies.2,3 This
poses a relative risk greater than that of the nonpregnant population. A recent study by Kittner et al2
evaluated girls and women aged 15 to 44 years who
had a diagnosis of cerebral infarction or intracerebral
hemorrhage in central Maryland and Washington
DC. They showed that the relative risk of intracerebral hemorrhage during pregnancy and 6 weeks
postpartum was 5.6 times that of the nonpregnant
patient with the greatest risk occurring in the postpartum period. Other studies have shown the highest rate
of aneurysmal rupture occurred during the late stages
of pregnancy itself.4,5
In summary, the possibly increased incidence of
cerebral aneurysmal rupture in pregnancy and the significant mortality associated with rupture should be
considered in a patient presenting with acute central
neurological abnormalities during the puerperium. Also,
when faced with a peripartum patient with an inciden-

tally discovered unruptured aneurysm, the possibility of
expedited management should be considered.

Pregnancy Outcome After
Intra-Abdominal Bleeding Due
to Placenta Percreta at 14
Weeks of Gestation

will rise concomitantly resulting in a greater incidence of
severe complications.

Annemaaike E. Roeters, MD,
Martijn A. Oudijk, MD, PhD,
Roger Heydanus, MD, PhD, and
Hein W. Bruinse, MD, PhD
BACKGROUND: With the rising rate of cesarean deliveries, the rate of placenta previa and placenta percreta

From the Department of Obstetrics, University Medical Center, Utrecht, the
Netherlands, and Department of Obstetrics and Gynecology, Amphia Hospital,
Breda, the Netherlands.
Corresponding author: Hein W. Bruinse, MD, PhD, Department of Obstetrics,
University Medical Center Utrecht, KE 04.123.1/P.O. Box 85090, 3508 AB,
Utrecht, the Netherlands; e-mail: H.W.Bruinse@umcutrecht.nl.
© 2007 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/07

574

Roeters et al

REFERENCES
1. Visscher HC, Visscher RD. Indirect obstetric deaths in the
state of Michigan 1960-1968. Am J Obstet Gynecol 1971;109:
1187–96.
2. Kittner SJ, Stern BJ, Feeser BR, Hebel R, Nagey DA, Bucholz
DW, et al. Pregnancy and the risk of stroke. N Engl J Med
1996;335:768–74.
3. Sharshar T, Lamy C, Mas JL. Incidence and causes of strokes
associated with pregnancy and puerperium: a study in public
hospitals of Ile de France. Stroke in Pregnancy Study Group.
Stroke 1995;26:930–6.
4. Dias MS, Sekhar LN. Intracranial hemorrhage from aneurysms and arteriovenous malformations during pregnancy and
the puerperium. Neurosurgery 1990;27:855–65.
5. Hunt HB, Schifrin BS, Suzuki K. Ruptured berry aneurysms
and pregnancy. Obstet Gynecol 1974;43:827–37.
6. Miller HJ, Hinkley CM. Berry aneurysms in pregnancy: a 10
year report. South Med J 1970;63:279.
7. Manalo-Estrella P, Barker AE. Histopathologic findings in
human aortic media associated with pregnancy. Arch Pathol
1967;83:336–41.
8. Zucker S, Lysik RM, Gurfinkel M, Zarrabi M, Stetler-Steveson
W, Liotta LA, et al. Immunoassay of type IV collagenase/
gelatinase (MMP-2) in human plasma. J Immunol Methods
1992;148:189–98.
9. Barrett JM, Van Hooydonk JE, Boehm FH. Pregnancy-related
rupture of arterial aneurysms. Obstet Gynecol Surv 1982;37:
557–66.

CASE: This case report describes a pregnancy with a
massive intra-abdominal bleeding due to placenta percreta at 14 weeks of gestation. Several management
options were discussed, and finally continuation of pregnancy was chosen. No further complications occurred,
and in the 35th week, an elective cesarean delivery and
hysterectomy were performed. A healthy male newborn
of 2,400 g was born.
CONCLUSION: While the outcome of pregnancy was
favorable in this case, it does not rule out the possibility
of severe complications with this management.
(Obstet Gynecol 2007;109:575–6)

P

lacenta percreta is a rare cause of intra-abdominal
bleeding during pregnancy. Placenta percreta is
characterized by penetration of the chorionic villi
through the myometrial and serosal layer of the
uterus. A recent analysis reported an incidence of
placenta creta in 1 in 533 deliveries.1 The incidence of
placenta accreta, increta and percreta, defined by
myometrial invasion, appears to be rising, due to

Pregnancy Outcome After Placenta Percreta

OBSTETRICS & GYNECOLOGY

increasing cesarean rates.1,2 Other independent risk
factors are placenta previa, which results in a 5%
chance of trophoblastic invasion in the myometrium
in an unscarred uterus,3 and increased maternal age.1,2
In case of placenta previa and a history of only one
cesarean delivery, the cumulative risk of placenta
creta is as high as 25%.3 Patients with placenta percreta are at significant risk of massive blood loss,
disseminated intravascular coagulation, infection, and
maternal and fetal morbidity and mortality. We
present a case in which pregnancy was complicated
by massive intra-abdominal bleeding due to placenta
percreta at gestational week 14. The management
chosen in this case will be described.

CASE
A 34-year-old woman, gravida 2, para 1, was referred to
our university medical center. The previous pregnancy
was complicated by preeclampsia and the hemolysis,
elevated liver enzymes, low platelets (HELLP) syndrome
and was delivered at 33 weeks of gestation by an
uncomplicated cesarean birth. A healthy girl of 1,590 g
was born. In her current pregnancy, the patient presented
at a general hospital with acute abdominal pain at 14
weeks of gestation. Examination was suspicious for an
intra-abdominal bleeding. An initial laparoscopy was
converted to a median laparotomy because of a hemorrhage originating from the lower uterine segment at the
site of the previous uterine incision resulting in an
estimated blood loss of 4,000 mL. Perioperative blood
pressure and oxygen saturation were kept within normal
ranges. The hemoglobin level at arrival at the emergency
room was 9.2 g/dL and decreased to a minimum of 4.8
g/dL during operation. After transfusion of 8 units of
packed cells and 6 units of fresh frozen plasma, it
normalized at 12.6 g/dL. Tissue protruding through the
bleeding defect was removed and the defect was closed.
The tissue obtained was reported to be trophoblast by the
pathologist. This confirmed the suspected diagnosis of
intra-abdominal hemorrhage due to placenta percreta.
Postoperatively she did well and on ultrasound examination the fetus was alive with normal movements.
The couple was informed about the diagnosis and
possible risks for mother and fetus. These risks included
recurrence of a life-threatening intra-abdominal bleeding,
invasion by the growing placenta in the surrounding tissues
with its concomitant complications, as well as possible
hypoxic damage to the fetus caused by massive bleeding. A
literature search on management of this clinical presentation was not helpful. Four management strategies were
considered and discussed with the couple: first, a hysterectomy after embolization of uterine or internal iliac arteries
to prevent a potential life-threatening hemorrhage without
consideration of fetus or uterus; second, embolization and
a hysterotomy with removal of fetus and placenta hoping to

VOL. 109, NO. 2, FEBRUARY 2007

preserve the uterus; third, a feticide combined with methotrexate possibly resulting in a spontaneous vaginal abortion or, if necessary, vaginal or abdominal removal; and
finally, a “wait and see” policy within a clinical setting with
all precautions of adequate intervention in case of recurrent
bleeding. The latter strategy with regular evaluation of the
fetal development by ultrasonography was chosen in accordance with the couple who had decided never to
become pregnant again, even if the treatment would result
in preservation of the uterus. She was admitted into the
general hospital for surveillance considering the risk of
recurrent bleeding. In the following 12 weeks, no further
complications occurred. An extensive scan at 20 weeks of
gestation revealed no fetal congenital or acquired anomalies. The patient was transferred to our university medical
center at 26 weeks of gestation, in vicinity of a Neonatal
Intensive Care Unit. An emergency protocol was prepared
in case of an acute hemorrhage, which implied an immediate laparotomy by experienced obstetric surgeons in
cooperation with the departments of anesthesiology, urology, pediatrics, and the blood transfusion department.
Evaluation by ultrasound and magnetic resonance
imaging showed a placenta previa totalis, covering almost the whole anterior uterine wall. Several areas of
penetration of the placenta through the lower uterine
wall at the level of the previous uterine incision and also
at the site of the cervix were present. There were no signs
of invasion of placental tissue in the bladder or the
surrounding tissue (Fig. 1).
Fetal ultrasound scans were performed every 2 weeks
and showed normal growth. At 33 weeks of gestation, the
patient received a 2-day course of betamethasone to enhance fetal lung maturity. At 34 1/7 weeks of gestation, an
elective cesarean delivery was performed.
At this point, two strategies of management were
considered: hysterectomy or conservative treatment with
methotrexate. The final decision would be made at the
time of operation. Hysterectomy was the treatment of
choice; however, if placental tissue was found to invade
surrounding tissues, conservative treatment with methotrexate was the second option. Both options were discussed with the couple. They emphasized that they had
no wish to preserve fertility and preferred a hysterectomy
in view of the possible complications of conservative
treatment. Preoperatively catheters were inserted in both
internal iliac arteries, and ureteral catheters were placed
to facilitate identification of the ureters. The patient was
given epidural and general anesthesia. An abdominal
midline incision was made, after which the uterus was
inspected. The placenta was visible at the site of the
previous uterine incision with modest elaboration in the
parametrium (Fig. 2). A transverse incision was made in
the uterine fundus and a male newborn, weighing 2,400
g with Apgar scores of 5, 8, and 9 at 1, 5, and 10 minutes,
respectively, was born. There was no bleeding from the
undisturbed placental site, and after cord ligation the
uterine incision was closed. Subsequently, embolization

Roeters et al

Pregnancy Outcome After Placenta Percreta

575

Fig. 2. Inspection of uterus after abdominal midline incision. The placenta was visible at the site of the previous
uterine incision. Placental tissue (A).
Roeters. Pregnancy Outcome After Placenta Percreta. Obstet
Gynecol 2007.

Fig. 1. Placenta clearly growing in the uterus wall. Placental
tissue (A), uterine wall (B). L, left.
Roeters. Pregnancy Outcome After Placenta Percreta. Obstet Gynecol 2007.

of the anterior part of both internal ileac arteries was
performed. Because there was no placental invasion in
the surrounding tissues, a total hysterectomy was performed without complications. Blood loss was estimated
at 1,000 mL. The hemoglobin count was 12.1 g/dL
preoperatively and 8.1 g/dL postoperatively.
The patient had an uncomplicated postoperative course
and was discharged on day 7 in good health. Histopathological examination confirmed a placenta previa-percreta
with a uterus lacking myometrium at the sites of the
invaded placental tissue.
The male infant developed a mild respiratory distress
syndrome but recovered well and was discharged on day
25 in good condition.

weeks of gestation and the therapeutic dilemma
that evolved.
With the rising cesarean delivery rates, the rate of
placenta previa and creta will rise concomitantly, and
obstetricians might encounter exceptional cases like
the one described more often. While the outcome was
favorable, this does not itself vindicate our management. Several life-threatening complications could
have occurred, with serious consequences.
REFERENCES
1. Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 2005;192:
1458–61.

COMMENT

2. Miller DA, Chollet JA, Goodwin TM. Clinical risk factors for
placenta previa-placenta accreta. Am J Obstet Gynecol 1997;
177:210–4.

This case of placenta percreta is exceptional in view
of the massive intra-abdominal bleeding at 14

3. Clark SL, Koonings PP, Phelan JP. Placenta previa/accreta and
prior cesarean section. Obstet Gynecol 1985;66:89–92.

576

Roeters et al

Pregnancy Outcome After Placenta Percreta

OBSTETRICS & GYNECOLOGY