Uterine Evacuation for Second-Trimester

Fetal Death and Maternal Morbidity

Andrea G. Edlow, MD, MSc, Melody Y. Hou, MD, MPH, Rie Maurer, MA, Carol Benson, MD,
Laurent Delli-Bovi, MD, and Alisa B. Goldberg, MD, MPH

OBJECTIVE: To estimate maternal morbidity associated abortion group (Nⴝ1) (Pⴝ.07). Induction of labor was
with uterine evacuation for second-trimester fetal de- more morbid than dilation and evacuation after adjusting
mise compared with that associated with induced sec- for confounders (OR 5.36; 95% CI 2.46 –11.69), primarily
ond-trimester abortion. as a result of increased odds of infection requiring
METHODS: This retrospective cohort study compared intravenous antibiotics. Gestational age of 20 weeks or
the maternal outcomes of two cohorts: 1) women diag- greater was significantly associated with maternal mor-
nosed with fetal demise between 14 and 24 weeks who bidity (OR 2.59; 95% CI 1.39 – 4.84).
subsequently underwent dilation and evacuation or in- CONCLUSION: In the second trimester, uterine evacu-
duction of labor; and 2) women undergoing induced ation for fetal demise was not significantly associated
abortion between 14 and 24 weeks by either dilation and with maternal morbidity compared with induced abor-
evacuation or induction of labor. The primary outcome tion. Induction of labor was more morbid than dilation
was major maternal morbidity. Assuming morbidity rates and evacuation as a result of an increased risk of pre-
of 11% for fetal demise and 1% for induced second- sumed infection.
trimester abortion, 94 patients were needed per group to (Obstet Gynecol 2011;117:307–16)
detect significant difference in maternal morbidity (80% DOI: 10.1097/AOG.0b013e3182051519
power, 5% alpha). LEVEL OF EVIDENCE: II
RESULTS: We identified 121 women with fetal demise
and 121 women who underwent induced abortion for
inclusion. There were no maternal deaths. In crude and
adjusted analyses, treatment for fetal demise was not
I ntrauterine fetal demise has historically been asso-
ciated with higher rates of maternal morbidity and
mortality compared with other pregnancy outcomes.
associated with increased maternal morbidity (25 of 121)
Many of the studies examining maternal complica-
compared with induced abortion (27 of 121) (adjusted
odds ratio [OR], 1.15; 95% confidence interval [CI],
tions of intrauterine fetal demise date back to the
0.57–2.32). There were more blood transfusions in the 1950s and 1960s when intrauterine fetal demise went
fetal demise group (Nⴝ7) compared with the induced- undetected for long periods of time as a result of few
prenatal visits and infrequent clinical use of ultra-
From the Departments of Obstetrics and Gynecology and Radiology and the sonography.1–3 The recognition of an increased risk of
Center for Clinical Investigation, Brigham and Women’s Hospital, Boston, coagulopathy in cases of intrauterine fetal demise
Massachusetts; the Department of Obstetrics and Gynecology, Boston Medical delivered remote from demise, as well as the devel-
Center, Boston, Massachusetts; and Women’s Health Services, Brookline,
Massachusetts. opment of new and safer techniques for removal of a
Supported by an Expanding the Boundaries Research Grant awarded by the
demised fetus in the second trimester, has led to active
Department of Obstetrics and Gynecology, Brigham and Women’s Hospital. uterine evacuation soon after recognition of intrauter-
Presented as an oral abstract at the National Abortion Federation Annual ine fetal demise.
Meeting, April 26, 2010, Philadelphia, Pennsylvania. Since the shift toward active management, few
Corresponding author: Alisa B. Goldberg, MD, MPH, Department of Obstetrics studies have examined maternal outcomes in cases of
and Gynecology, Brigham and Women’s Hospital, 75 Francis Street, Boston, intrauterine fetal demise.4 –7 Some recent analyses
MA 02115; e-mail: agoldberg@pplm.org.
suggest that despite the shift toward active manage-
Financial Disclosure
The authors did not report any potential conflicts of interest.
ment, intrauterine fetal demise may still be associated
with higher rates of maternal morbidity7 and mortal-
© 2011 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins. ity.8 The maternal morbidity associated with induc-
ISSN: 0029-7844/11 tion of labor compared to dilation and evacuation for

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 OBSTETRICS & GYNECOLOGY 307
management of intrauterine fetal demise is unknown. same day, or next closest day, as the procedure for
Data comparing induction of labor and dilation and intrauterine fetal demise. The rationale for selecting
evacuation for second-trimester induced abortion are the first eligible “control” patient in this manner was
also scarce.9 –11 to achieve standardization of clinical procedures as
Our primary objective was to estimate whether much as possible and to ensure a similar health care
uterine evacuation for second-trimester intrauterine provider pool. The Family Planning Clinic is an
fetal demise, defined as pregnancy loss between 14 ambulatory clinic within Brigham and Women’s Hos-
and 24 weeks of gestation, is associated with an pital dedicated to providing outpatient gynecologic
increased risk of maternal morbidity compared with procedures, including uterine evacuation for first- and
induced second-trimester abortion. Our secondary second-trimester induced and spontaneous abortions.
objective was to compare the maternal morbidity of The Family Planning Clinic schedule also includes
second-trimester dilation and evacuation with that of patients who, as a result of medical complexity,
induction of labor in the second trimester. history of anesthesia issues, or other considerations,
underwent dilation and evacuation in the main hos-
MATERIALS AND METHODS pital operating room; and 2) for each woman with an
We conducted a retrospective cohort study with intrauterine fetal demise who chose induction of
approval by the Partners Healthcare Institutional labor, control subjects were selected in a similar
Review Board. To investigate maternal morbidity and fashion from the population of all women who under-
mortality in women found to have intrauterine fetal went induction–termination in the labor and delivery
demise in the second trimester, we compared the unit at Brigham and Women’s Hospital during the
maternal outcomes of two cohorts: 1) women diag- study period. Indication for pregnancy termination
nosed with a fetal demise by ultrasonography be- was not a criterion for selection of control subjects.
tween 14 and 24 weeks of gestation at Brigham and The indications for induced abortion in the in-
Women’s Hospital from January 1, 2003, to May 8, duction of labor and dilation and evacuation control
2009, who subsequently underwent uterine evacua- cohorts included: undesired pregnancy with no
tion by dilation and evacuation or induction of labor known fetal or maternal conditions informing deci-
at Brigham and Women’s Hospital or at a Brigham sion for termination, fetal anomaly or fetal condition
and Women’s Hospital-affiliated private outpatient incompatible with life, preterm premature rupture of
facility dedicated to providing dilation and evacuation membranes, and severe maternal medical conditions
services (Women’s Health Services); and 2) women in which pregnancy could jeopardize the mother’s
undergoing induced abortion for any indication at health or life.
Brigham and Women’s Hospital between 14 and 24 Exclusion criteria were the same for both cohorts
weeks of gestation by either dilation and evacuation and included: 1) multiple gestations; 2) placentation
or induction of labor during the same time period. abnormalities including placenta previa, accreta, in-
The intrauterine fetal demise cohort was identi- creta, and percreta; 3) any maternal condition that
fied through a comprehensive radiology database that would preclude eligibility for either dilation and
was queried for all diagnoses of “demise” from 14 –24 evacuation or induction; and 4) uterine evacuation
weeks of gestation during the study period of interest. that was not completed at either Brigham and Wom-
To be included in the intrauterine fetal demise cohort, en’s Hospital or Women’s Health Services, so that
demise had to be diagnosed before labor induction accurate and complete documentation of the proce-
and could not be an incidental finding after induction dure and all complications was available. Patients
of labor was started for another indication. Similarly, were not excluded from either the intrauterine fetal
iatrogenic demises after feticidal injection as part of an demise or the control cohort for maternal conditions
induced abortion were not included in the intrauter- including preterm premature rupture of membranes,
ine fetal demise cohort. We identified the induced severe preeclampsia, abruption, and chronic medical
abortion cohort by procedure type and date of pro- conditions, including hypertension, systemic lupus
cedure using the following methods: 1) for each erythematosus or other immunologic conditions,
woman with an intrauterine fetal demise who under- thrombophilias or bleeding disorders, renal dysfunc-
went dilation and evacuation as the mode of uterine tion, immunosuppression, or diabetes.
evacuation, the schedule of Brigham and Women’s Estimated gestational age for both cohorts was
Hospital Family Planning Clinic was queried for the based on earliest documented ultrasonography. For
first eligible patient undergoing induced second-tri- many women, the earliest documented ultrasonogra-
mester abortion by dilation and evacuation on the phy was at the time of presentation with intrauterine

308 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity OBSTETRICS & GYNECOLOGY
fetal demise or for induced abortion. Ultrasono- most commonly used misoprostol protocol involved the
graphic dating of gestational age was based on bipa- initial administration of 400 micrograms of vaginal
rietal diameter corrected for occipitofrontal diameter misoprostol followed by 200 micrograms of vaginal
in the majority of cases; if only biparietal diameter misoprostol every 3– 4 hours until delivery. For some
was available, this measurement was used to estimate patients, a high-dose oxytocin regimen was used at the
gestational age.12 Gestational age was based on femur discretion of the health care provider. Laminaria were
length in situations in which biparietal diameter or often, but not uniformly, placed 12–24 hours before
biparietal diameter corrected for occipitofrontal diame- misoprostol or oxytocin induction, depending on gesta-
ter was either unavailable (eg, anencephaly) or of ques- tional age and health care provider preference.
tionable accuracy (eg, partial collapse of skull or over- For labor induction terminations for anomalies,
lapping sutures noted on ultrasonographic report). preterm premature rupture of membranes, or hemo-
Both Brigham and Women’s Hospital and Wom- lysis, elevated liver enzymes, low platelets syndrome,
en’s Health Services provide abortion services up to feticidal injection was performed on the day before
24 weeks gestational age. Abortion services are rou- induction, most commonly with intraamniotic
tinely offered into the early part of the 23rd week of digoxin, but occasionally with fetal intracardiac po-
gestation to ensure that with standard error in ultra- tassium chloride. With the exception of feticidal in-
sonographic dating, procedures remain under 24 jection, patients undergoing induction of labor for
weeks. At both Brigham and Women’s Hospital and reasons other than intrauterine fetal demise received
Women’s Health Services, procedures are performed the same drug regimen as patients with intrauterine
by obstetrics and gynecology residents supervised by fetal demise. Patients who developed fever during the
family planning fellows or attendings or by family induction process were treated for presumed chorio-
planning fellows and attendings without resident in- amnionitis at the discretion of the health care pro-
volvement. There is significant overlap in the health vider based on the health care provider’s suspicion for
care provider pool at these two sites. At both Brigham true chorioamnionitis and not prostaglandin-induced
and Women’s Hospital and Women’s Health Ser- fever. Typically patients are not treated for chorioam-
vices, patients undergoing dilation and evacuation for nionitis during misoprostol induction unless fever is
a fetus of 14-week size or greater routinely received more than 101°F.
laminaria for cervical ripening on the day before the Data collection included maternal demographic
procedure. All patients routinely received antibiotics information; obstetric and substance use history; size
to presumptively treat Chlamydia (either 7 days of 100 of pregnancy at the time of diagnosis of demise by
mg doxycycline twice daily or a single dose of 1 g radiologic criteria (biparietal diameter corrected for
azithromycin) starting on the day of laminaria place- occipitofrontal diameter or crown rump length or
ment. Patients between 18 and 24 weeks may have femur length if applicable); gestational age by best
received 2 days of cervical ripening with laminaria at obstetric dating (eg, gestational age based on earliest
the discretion of the health care provider. Starting in ultrasonography) at the time of diagnosis of demise if
November 2006, as a result of a cluster of infectious different from size of pregnancy by radiographic
complications, all patients receiving 2 days of lamina- criteria; maternal tobacco use; presence of fetal anom-
ria received ceftriaxone (or levofloxacin for known alies; obstetric complications of current pregnancy;
penicillin or cephalosporin allergy) prophylaxis in maternal medical comorbidities; induction agent;
addition to the aforementioned oral regimen for laminaria administration, feticidal injection, or both
treatment of Chlamydia. Starting in 2007, at Brigham when relevant; estimated blood loss when available;
and Women’s Hospital, patients receiving 2 days of fetal autopsy findings when available; and length of
laminaria also received intraamniotic digoxin injec- hospital stay.
tion. Paracervical vasopressin was routinely adminis- The following events were defined as major ma-
tered at both Brigham and Women’s Hospital and ternal complications: death; need for transfusion;
Women’s Health Services before dilation and evacua- infection requiring treatment with intravenous antibi-
tions at 14 –24 weeks, and uterotonic agents were ad- otics (not prophylaxis); intensive care unit stay; sepsis,
ministered during or after the procedure as needed. All shock, or systemic inflammatory response syndrome;
procedures were performed with ultrasonographic guid- uterine perforation; unplanned procedure (including
ance and conducted under intravenous general anesthe- same day or interval suction curettage, hysterotomy,
sia by a staff anesthesiologist or nurse anesthetist. hysterectomy, laparotomy, uterine artery emboliza-
For patients who underwent an induction of labor tion, or other unplanned procedure); and thrombotic
in the second trimester for intrauterine fetal demise, the event (amniotic fluid embolus, stroke, deep venous

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity 309
thrombosis, or pulmonary embolus) diagnosed within with spontaneous placental delivery in the hospital,
6 weeks of dilation and evacuation or induction of and one woman underwent spontaneous vaginal de-
labor. A dichotomous outcome variable for any of livery at home and required dilation and evacuation
these major maternal complications was created. for retained placenta. The two women who delivered
Computerized medical records and paper charts spontaneously were included in the induction of labor
were examined for both the intrauterine fetal demise cohort for analysis, because they had indicated intent
and control cohorts. Data were abstracted from charts to pursue induction of labor. The control cohort
by two individuals, including the primary author included 121 women who underwent induced abor-
(A.G.E.), using a standardized data sheet. The pri- tion (82 by dilation and evacuation and 39 by induc-
mary author reviewed every completed data sheet to tion of labor).
ensure accuracy and uniform abstraction technique. The intrauterine fetal demise and induced abor-
Numeric variables were compared using either tion cohorts did not vary significantly with respect to
Student’s t tests or Wilcoxon rank-sum tests. Categor- race, parity, prior cesarean delivery, obstetric compli-
ical variables were compared using either Fisher’s cations in the current pregnancy, chronic medical
exact tests or chi square tests. Data were missing for conditions, or drug use (Table 1). The intrauterine
more than 5% of records for the following variables: fetal demise cohort had a greater incidence of preterm
race (16%), drug use (12%), smoking (10%), and premature rupture of membranes and chronic hyper-
history of cesarean delivery (7%). The pattern of tension. Women in the intrauterine fetal demise co-
missing data appeared to be random, so we applied hort were significantly older, less likely to smoke, had
the multiple imputation method for analysis.13 Five a lower gestational age, and were less likely to have
imputed data sets were created for multiple imputa- fetal anomalies than were women in the induced
tion. Maternal age (30 years old or older compared abortion cohort. The median (quartile 1, quartile 3)
with younger than 30 years old) and gestational age number of days from diagnosis of intrauterine fetal
(20 weeks or greater compared with less than 20 demise to initiation of procedure was 1 (0, 3).
weeks) were dichotomized. There were no maternal deaths in the study
Univariable logistic regression was performed to population. Maternal morbidity associated with uter-
assess the group effect and procedure effect on ma- ine evacuation for intrauterine fetal demise compared
ternal morbidity. Multivariable logistic regression was with induced abortion is depicted in Table 2. Evacu-
also performed adjusting for potential confounders, ation of second-trimester intrauterine fetal demise was
including maternal age, parity, gestational age, and not associated with increased risk of maternal mor-
smoking status. There were no significant baseline bidity compared with second-trimester induced abor-
differences between groups with respect to race, tion on crude analysis (25 of 121 women undergoing
obstetric complications, and maternal medical comor- uterine evacuation for intrauterine fetal demise had
bidities, so the multivariable logistic regressions did associated maternal morbidity compared with 27 of
not adjust for these variables. We did choose to adjust 121 women undergoing induced abortion; OR 0.91;
for baseline differences that were not statistically 95% CI 0.49 –1.68; P⫽.75). Multivariable logistic re-
significant between groups but that we believed might gression was performed to adjust for potential con-
be associated with morbidity. A P value, odds ratio founders, including maternal age, parity, gestational
(OR), and its 95% confidence intervals (CIs) were age, and smoking status (Table 3). However, the
reported for each finding. SAS 9.1 was used for the association between uterine evacuation for intrauter-
analysis. Assuming maternal morbidity of 11% for ine fetal demise and maternal morbidity remained
intrauterine fetal demise and 1% for induced second- nonsignificant (adjusted OR 1.15; 95% CI 0.57–2.32).
trimester abortion,10 at least 94 patients were needed Blood transfusions were increased in the intrauterine
in each group to maintain 80% power and a two-sided fetal demise group (n⫽7) compared with the induced
␣ of 0.05. abortion group (n⫽1), although this association did
not achieve significance (P⫽.07).
RESULTS Although the numbers of women with any indi-
Of 253 patients diagnosed with fetal demise on vidual marker of morbidity were too small to ade-
Brigham and Women’s Hospital ultrasonography at quately power statistical comparisons, there were no
14 –24 weeks of gestation, 121 women met inclusion cases of thrombotic events; intensive care unit stay; or
criteria. Of these, 82 underwent dilation and evacua- sepsis, shock, or systemic inflammatory response syn-
tion, 37 underwent induction of labor, one woman drome among women having second-trimester in-
underwent spontaneous vaginal delivery at home duced abortion, whereas there were eight instances of

310 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity OBSTETRICS & GYNECOLOGY
Table 1. Demographic Summary of Intrauterine Fetal Demise and Second-Trimester Induced
Abortion Cohorts
Second-Trimester
IUFD (nⴝ121) Abortion (nⴝ121) P*

Age (y) 32⫾7 29⫾8 ⬍.001

30 or older 89 (74) 56 (46)
Race .89
African American 19 (15.7) 19 (15.7)
White 57 (47.1) 40 (33.0)
Hispanic 22 (18.2) 18 (14.9)
Unknown 8 (6.6) 30 (24.8)
Drug use .29
Yes 2 (1.7) 6 (5.0)
Unknown 23 (19.0) 5 (4.1)
Current Smoking .001
Yes 7 (5.7) 27 (22.3)
Unknown 21 (17.4) 4 (3.3)
Reproductive history
Primiparous 32 (26.5) 45 (37.2) .07
Among multiparous women .74
History of cesarean delivery 22/89 (24.7) 21/76 (27.6)
Unknown 16/89 (18.0) 0/76 (0.0)
Gestational age (d) 125.8⫾19.1 (17.9 wk) 134.0⫾17.9 (19.1 wk) ⬍.001
20 wk or greater 30 (24.8) 57 (47.1)
Anomalies 32 (26.5) 60 (49.6) ⬍.001
†
Obstetric complications 10 (8.3) 5 (4.1) .18
PPROM 10 (8.3) 3 (2.5) .08
Preeclampsia or gestational hypertension 0 (0) 2 (1.7) .50
HELLP 0 (0) 2 (1.7) .50
Chronic medical conditions 15 (12.4) 11 (9.1) .41
Thrombophilia 4 (3.3) 1 (0.8) .37
Chronic hypertension 9 (7.4) 2 (1.7) .06
Bleeding disorder 1 (0.8) 2 (1.7) 1.0
Immune suppression 4 (3.3) 2 (1.7) .68
Renal disease 4 (3.3) 2 (1.7) .68
Lupus 3 (2.5) 2 (1.7) 1.0
Diabetes 1 (0.8) 3 (2.5) .62
IUFD, intrauterine fetal demise; PPROM, premature preterm rupture of membranes; HELLP, hemolysis, elevated liver enzymes, low
platelet count.
Data are mean⫾standard deviation or n (%) unless otherwise specified.
Categorical variables are presented with frequency counts (%).
* Missing or unknown data excluded in P value calculation.
†
Numbers below do not add up to 5 because some patients had more than one obstetric comorbidity.

these morbidities in the intrauterine fetal demise 127.8⫾18.3 days (18.3 weeks) (P⬍.001). Gestational
cohort, representing six patients. In a post hoc anal- age 20 weeks or greater was associated with signifi-
ysis excluding intravenous antibiotic use as an indica- cantly greater odds of having any maternal morbidity
tion of major maternal morbidity, the rate of morbid- compared with gestational age less than 20 weeks (OR
ity for the induced abortion cohort fell to 7.4%, the 2.59; 95% CI 1.39 – 4.84; P⫽.002). A stratified analysis
rate for the intrauterine fetal demise cohort fell to examining maternal morbidity associated with intrauter-
19.8%, and intrauterine fetal demise became associ- ine fetal demise compared with induced abortion by
ated with a significant increase in maternal morbidity gestational age 20 weeks or greater or less than 20 weeks
(crude OR 3.08; 95% CI 1.37– 6.94; P⫽.005). found that at gestational ages 20 weeks or greater, 18 of
Higher gestational age was significantly associ- 57 women undergoing induced abortion had maternal
ated with maternal morbidity. The mean gestational morbidity compared with 10 of 30 undergoing a proce-
age among women who had any maternal morbidity dure for intrauterine fetal demise (OR 1.08; 95% CI
was 137.6⫾19.3 days (19.6 weeks), whereas the mean 0.42–2.78; P⫽.87). At gestational ages less than 20
gestational age among women with no morbidity was weeks, 9 of 64 women undergoing induced abortion had

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity 311
Table 2. Maternal Morbidity and Mortality: only the presence of severe chorioamnionitis, funisi-
Uterine Evacuation for Intrauterine tis, or both was significantly associated with the
Fetal Demise Compared With presence of any maternal morbidity (P⬍.001).
Second-Trimester Induced Abortion Demographic characteristics of patients in the
Second- induction of labor and dilation and evacuation co-
Trimester horts (pooling women with intrauterine fetal demise
IUFD Abortion
(nⴝ121) (nⴝ121) P* and women who underwent a second trimester in-
duced abortion) are listed in Table 4. Overall, induc-
Transfusion 7 (5.8) 1 (0.8) .07 tion of labor was more morbid than dilation and
IV antibiotic use 13 (10.8) 19 (15.7) .27
evacuation in both crude and adjusted analyses (crude
ICU stay 2 (1.7) 0 (0) .50
Sepsis, shock, SIRS 3 (2.5) 0 (0) .12 OR 6.27; 95% CI 3.23–12.17; OR adjusted for ma-
Uterine perforation 0 (0) 0 (0) NA ternal age, gestational age, parity, and smoking status,
Unplanned procedure 9 (7.4) 9 (7.4) 1.0 5.36; 95% CI 2.46 –11.69). Among patients undergo-
Peri- or postoperative 3 (2.5) 0 (0) .25 ing induction of labor, there was no increased mor-
thrombotic event
Death 0 (0) 0 (0) NA bidity for women with an intrauterine fetal demise
Any of the above
†
25 (20.7) 27 (22.3) .75 (OR 0.43; 95% CI 0.17–1.07). Among patients under-
IUFD, intrauterine fetal demise; IV, intravenous; ICU, intensive going dilation and evacuation, there was no signifi-
care unit; SIRS, systemic inflammatory response syndrome; cant increase in morbidity for those with an intrauter-
NA, not applicable. ine fetal demise (OR 2.17; 95% CI 0.77– 6.10).
Categorical variables are presented with frequency counts (%).
* P values were obtained by excluding missing data. Maternal morbidity associated with dilation and evac-
†
The numbers above do not add up to 25 because some patients uation compared with induction of labor is depicted
had multiple morbidities or complications.
in Table 5. The results of the multivariable logistic
regression analysis for maternal morbidity by proce-
maternal morbidity compared with 15 of 91 undergoing dure type (dilation and evacuation compared with
a procedure for intrauterine fetal demise (OR 1.21; 95% induction of labor) are depicted in Table 6.
CI 0.49 –2.96; P⫽.68). The increased morbidity observed in the induc-
We examined the relationship between the pres- tion of labor cohort can be attributed to increased
ence of maternal morbidity and the placental and odds of intravenous antibiotic treatment. We per-
fetal pathology findings at autopsy for the intrauterine formed a post hoc analysis excluding intravenous
fetal demise group. Placental and fetal pathologic antibiotic use from the definition of morbidity and
findings included presence or absence of abruption; found that 11 of 78 patients in the induction of labor
placental infarcts; severe chorioamnionitis, funisitis, group had any morbidity compared with 13 of 164
or both; anomalies; and cord accident. We found that patients in the dilation and evacuation group. This

Table 3. Unadjusted and Adjusted Odds Ratios for Procedure Indication and Maternal Morbidity
Unadjusted OR Adjusted
Covariate (95% CI) OR (95% CI) Adjusted P

Indication .69
IUFD 0.91 (0.49–1.68) 1.15 (0.57–2.32)
Induced abortion Reference Reference
Maternal age (y) .49
30 or older 1.09 (0.58–2.05) 1.28 (0.63–2.62)
Younger than 30 Reference Reference
Gestational age (wk) .01
20 or greater 2.59 (1.39–4.84) 2.49 (1.27–4.85)
Less than 20 Reference Reference
Current smoking .66
Yes 1.61 (0.61–4.20) 0.79 (0.27–2.28)
No Reference Reference
Parity .003
Primiparous 2.73 (1.45–5.12) 2.75 (1.40–5.42)
Multiparous Reference Reference
OR, odds ratio; CI, confidence interval; IUFD, intrauterine fetal demise.

312 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity OBSTETRICS & GYNECOLOGY
Table 4. Demographic Summary by Procedure Type
IOL (nⴝ78) D&E (nⴝ164) P*

Age (y) 32⫾7 30⫾8 .04

Race .07
African American 10 (12.9) 28 (17.1)
White 34 (43.6) 63 (38.4)
Hispanic 11 (14.1) 29 (17.7)
Unknown 9 (11.5) 29 (17.7)
Drug use 1.0
Yes 2 (2.6) 6 (3.7)
Unknown 12 (15.4) 16 (9.8)
Current smoking .01
Yes 4 (5.1) 30 (18.3)
Unknown 10 (12.8) 15 (9.1)
Reproductive history
Primiparous 34 (43.6) 43 (26.2) .01
Among multiparous women .65
History of cesarean delivery 11/44 (25.0) 32/121 (26.4)
Unknown 2/44 (4.5) 14/121 (11.6)
Gestational age (d) 144.0⫾15.9 (20.6 wk) 132.2⫾16.4 (18.9 wk) ⬍.001
Anomalies 49 (62.8) 43 (26.2) ⬍.001
Obstetric complications 6 (7.7) 9 (5.5) .51
PPROM 6 (7.7) 7 (4.3) .27
Preeclampsia or gestational hypertension 0 (0) 2 (1.2) 1.0
HELLP 0 (0) 2 (1.2) 1.0
Chronic medical conditions 7 (9.0) 19 (11.6) .54
Thrombophilia 1 (1.3) 4 (2.4) 1.0
Chronic hypertension 2 (2.6) 9 (5.5) .51
Bleeding disorder 0 (0) 3 (1.8) .55
Immune suppression 3 (3.9) 3 (1.8) .39
Renal disease 2 (2.6) 4 (2.4) 1.0
Lupus 1 (1.3) 4 (2.4) 1.0
Diabetes 2 (2.6) 2 (1.2) .60
IOL, induction of labor; D&E, dilation and evacuation; PPROM, premature preterm rupture of membranes; HELLP, hemolysis, elevated
liver enzymes, low platelet syndrome.
Data are mean⫾standard deviation or n (%) unless otherwise specified.
Categorical variables are presented with frequency counts (%).
* P values were obtained by excluding missing data.

result was not significant (P⫽.13). Thirty-two of the 52 labor. Thirteen of 39 women (33%) who underwent
patients who had any morbidity were treated with induction of labor experienced maternal morbidity
intravenous antibiotics for presumed infection. Of compared with 12 of 82 (15%) of women who under-
these, 11 had evidence of severe histologic chorio- went dilation and evacuation. The association was
amnionitis or funisitis on the placental pathology statistically significant (OR 2.92; 95% CI 1.18 –7.21;
report. Two of the patients treated with intravenous P⫽.02). In the induced abortion cohort, 21 of 39
antibiotics had sepsis, shock, or systemic inflamma- women (54%) with induction of labor experienced
tory response syndrome, and both of these patients maternal morbidity compared with six of 82 women
also had evidence of severe histologic chorioamnio- with dilation and evacuation (7%). The odds of mor-
nitis. One patient underwent dilation and evacua- bidity were 14.78 times higher undergoing induction
tion and the other induction of labor. of labor for induced abortion compared with dilation
We performed an analysis to examine the effect and evacuation (95% CI 5.21– 41.92; P⬍.001).
of procedure on morbidity within the intrauterine
fetal demise and the second-trimester induced abor- DISCUSSION
tion cohorts. Within the intrauterine fetal demise Uterine evacuation for second-trimester intrauterine fe-
cohort, including intravenous antibiotic use in the tal demise is not significantly associated with increased
definition of morbidity, odds of morbidity were 2.92 overall maternal morbidity compared with induced
times higher in women who underwent induction of abortion in our study population. However, when

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity 313
Table 5. Maternal Morbidity and Mortality: outcomes in the setting of active management of
Induction of Labor Compared With intrauterine fetal demise suggest persistence of in-
Dilation and Evacuation in the Second creased maternal morbidity and mortality.5,7,8,14 A
Trimester recent analysis suggested that maternal mortality rates
IOL D&E in the setting of intrauterine fetal demise may be as
(nⴝ78) (nⴝ164) P* high as 96.3 per 100,000 outcomes compared with an
Transfusion 3 (3.9) 5 (3.1) .72 overall maternal mortality rate of 5.59 per 100,000.8
IV antibiotic use 26 (33.3) 6 (3.7) ⬍.001 Prior studies have been limited by their largely de-
ICU stay 0 (0) 2 (1.2) 1.0 scriptive or cross-sectional nature, lack of control
Sepsis, shock, SIRS 1 (1.3) 2 (1.2) 1.0 groups, and potential overestimation of maternal
Uterine perforation 0 (0) 0 (0) NA
morbidity by including primarily intrauterine fetal
Unplanned procedure 9 (11.5) 9 (5.5) .09
Peri or postoperative 0 (0) 3 (1.8) .55 demises in the third trimester when options for re-
thrombotic event moval of a demised fetus may be limited to induction
Death 0 (0) 0 (0) NA of labor or abdominal delivery.4 –7 Gestational age in
Any of above 34 (43.6) 18 (11.0) ⬍.001 prior studies ranged from 24 to 40 weeks, and uterine
IOL, induction of labor; D&E, dilation and evacuation; IV, evacuation was achieved by either vaginal delivery or
intravenous; ICU, intensive care unit; SIRS, systemic
inflammatory response syndrome; NA, not applicable.
hysterotomy in all studies except one.7 The compli-
Categorical variables are presented with frequency counts (%). cations seen in these studies, including perineal and
* P values were obtained by excluding missing data. lower genital tract lacerations, reflect the higher ges-
tational age and mode of evacuation.
we excluded intravenous antibiotic use as an indi- Evidence comparing maternal morbidity associ-
cation of major maternal morbidity, we found that ated with induction of labor compared with dilation
intrauterine fetal demise was associated with signif- and evacuation for the management of intrauterine
icantly more morbidity than induced abortion. The fetal demise is sparse. Data comparing induction of
higher transfusion rate in the intrauterine fetal labor and dilation and evacuation using modern
demise cohort is also concerning. Induction of techniques for second-trimester induced abortion are
labor was more morbid than dilation and evacua- also limited.9 –11 One retrospective study found that
tion in the second trimester for the overall cohort as dilation and evacuation was associated with signifi-
well as within each group. Increased use of intravenous cantly lower complication rates than induction of
antibiotics to treat presumed infection in the induction of labor for second-trimester induced abortion.11 Rates of
labor cohort comprised the majority of morbidity, but the chorioamnionitis and endometritis observed in the in-
rate of serious infection was low overall. duction of labor cohort of two small retrospective studies
Our findings should be viewed in the context of were much lower than in our study, at 9%10 and 1.4%.11
prior studies. The few studies examining maternal Given the difficulty with randomization to dilation and

Table 6. Unadjusted and Adjusted Odds Ratios for Procedure Type and Maternal Morbidity
Unadjusted OR Adjusted OR
Covariate (95% CI) (95% CI) Adjusted P

Procedure type ⬍.001

IOL 6.27 (3.23–12.17) 5.36 (2.46–11.69)
D&E Reference Reference
Maternal age (y) .58
30 or older 1.09 (0.58–2.05) 1.23 (0.59–2.58)
Younger than 30 Reference Reference
Gestational age (wk) .76
20 or greater 2.59 (1.39–4.84) 1.13 (0.53–2.43)
Less than 20 Reference Reference
Current smoking .84
Yes 1.61 (0.61–4.20) 1.12 (0.36–3.49)
No Reference Reference
Parity .02
Primiparous 2.73 (1.45–5.12) 2.34 (1.15–4.75)
Multiparous Reference Reference
OR, odds ratio; CI, confidence interval; IOL, induction of labor; D&E, dilation and evacuation.

314 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity OBSTETRICS & GYNECOLOGY
evacuation or induction of labor for second-trimester trimester induced abortion most often have dilation
induced abortion,9 retrospective and nonrandomized and evacuation procedures in an outpatient setting
studies such as this one are likely to provide the best data with very low rates of morbidity, whereas women
available about maternal morbidity. with a second-trimester intrauterine fetal demise may
Our study has several limitations. Retrospective receive more heterogeneous care. In this study, we
data collection may have led to ascertainment bias estimated the morbidity of each group to help assess
both in patient identification and in data acquisition. whether these patients can and should be managed
Selection bias could have contributed to a difference similarly.
in morbidity between induction of labor and dilation Although our initial analysis found that dilation
and evacuation, particularly because the induction of and evacuation was associated with less morbidity
labor group had a higher mean gestational age. We do
than induction of labor for all women requiring
not believe that this difference was the result of health
second-trimester uterine evacuation regardless of in-
care provider avoidance of dilation and evacuation at
dication, when intravenous antibiotic use for treat-
higher gestational ages. Rather, there may have been
ment of fever and presumed infection was excluded
more routine offering of, and easier access to, induction
of labor for patients with intrauterine fetal demise, fetal as a criterion for morbidity, there was no difference in
anomalies, or pregnancy complications, which are typ- morbidity between procedures. Thus, it remains un-
ically conditions diagnosed later in pregnancy. To min- clear whether it is better for a woman with an
imize selection bias, all women included in the study intrauterine fetal demise to delay uterine evacuation
needed to be clinically eligible for either dilation and to access a dilation and evacuation procedure or
evacuation or induction of labor. whether she should proceed with induction of labor if
Although our expected rate of maternal morbid- that service is more readily available. Perhaps in
ity in the second-trimester induced abortion group situations in which induction of labor is more readily
was 1%, our observed event rate was 22%. Much of available and intravenous antibiotics are accessible,
the observed morbidity in the induced abortion co- patients should not delay uterine evacuation to access
hort was attributable to the use of intravenous antibi- a dilation and evacuation. Future larger studies should
otics for treatment of presumed infection. Because explore specific aspects of morbidity, including the
intravenous antibiotic use may not truly represent need for transfusion of blood products, to help health
serious morbidity, and because some of these diag- care providers select the safest setting (ie, outpatient
nosed infections may have been prostaglandin-in- clinic or hospital-based operating room) for the man-
duced fevers instead of true infections, we performed agement of women with second-trimester intrauterine
a post hoc analysis excluding the use of intravenous fetal demise.
antibiotics from the composite maternal morbidity
score. With this exclusion, the rate of maternal mor-
bidity observed in the second-trimester abortion co- REFERENCES
1. Pritchard JA. Fetal death in utero. Obstet Gynecol 1959;14:
hort fell to 7.4%, a rate consistent with previously
573– 80.
reported complication rates for second-trimester
2. Pritchard JA. Hematological problems associated with deliv-
abortions performed in a hospital setting.10 After ery, placental abruption, retained dead fetus and amniotic fluid
making this adjustment, the complication rate among embolism. Clin Hematol 1973;2:563– 86.
women undergoing evacuation for intrauterine fetal 3. Tricomi V, Kohl SG. [Fetal death in utero.] Am J Obstet
demise was significantly higher than for women un- Gynecol 1957;74:1092–7.
dergoing second-trimester induced abortion. 4. Magann EF, Chauhan SP, Bofill JA, Waddell D, Rust OA,
A strength of this study is the presence of a Morrison JC. Maternal morbidity and mortality associated
with intrauterine fetal demise: five-year experience in a tertiary
rigorously selected control group. We did not exclude referral hospital. South Med J 2001;94:493–5.
patients from either cohort based on maternal condi-
5. Maslow AD, Breen TW, Sarna MC, Soni AK, Watkins J, Oriol
tions, which makes it likely that patients in both NE. Prevalence of coagulation abnormalities associated with
cohorts were drawn from a similar population. Select- intrauterine fetal death. Can J Anaesth 1996;43:1237– 43.
ing the induced abortion cohort by procedure type 6. Tempfer CB, Brunner A, Bentz EK, Langer M, Reinthaller A,
and date helped ensure standardization of methods Hefler LA. Intrauterine fetal death and delivery complications
associated with coagulopathy: a retrospective analysis of 104
and health care provider pool. Our selection of cases. J Womens Health (Larchmt) 2009;18:469 –74.
women undergoing second-trimester induced abor-
7. Ifnan F, Jameel MB. Maternal morbidity and mortality associ-
tion as control participants also has clinical relevance. ated with delivery after intrauterine fetal death. J Coll Physi-
In the United States, women undergoing second- cians Surg Pak 2006;16:648 –51.

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity 315
 8. Grimes DA. Estimation of pregnancy-related mortality risk by 11. Autry AM, Hayes EC, Jacobson GF, Kirby RS. A comparison
pregnancy outcome, United States, 1991 to 1999. Am J Obstet of medical induction and dilation and evacuation for second-
Gynecol 2006;194:92– 4. trimester abortion. Am J Obstet Gynecol 2002;187:393–7.
9. Grimes DA, Smith MS, Witham AD. Mifepristone and miso- 12. Benson CB, Doubilet PM. Sonographic prediction of gesta-
prostol versus dilation and evacuation for midtrimester abor- tional age: accuracy of second- and third-trimester fetal mea-
tion: a pilot randomised controlled trial. BJOG 2004;111: surements. AJR Am J Roentgenol 1991;157:1275–7.
148 –53.
13. Allison PD. Missing data. Thousand Oaks (CA): Sage; 2001.
10. Turok DK, Gurtcheff SE, Esplin MS, Shah M, Simonsen SE,
Trauscht-Van Horn J, et al. Second trimester termination of 14. Calderon-Margalit R, Friedlander Y, Yanetz R, Deutsch L,
pregnancy: a review by site and procedure type. Contraception Manor O, Harlap S, et al. Late stillbirths and long-term
2008;77:155– 61. mortality of mothers. Obstet Gynecol 2007;109:1301– 8.

316 Edlow et al Intrauterine Fetal Demise and Maternal Morbidity OBSTETRICS & GYNECOLOGY