Original Article
Occult Placenta Accreta: Risk Factors,
Adverse Obstetrical Outcomes, and
Recurrence in Subsequent Pregnancies
Clodagh Mullen, MD1 Ashley N. Battarbee, MD2 Linda M. Ernst, MD3 Alan M. Peaceman, MD4
1 Department of Obstetrics and Gynecology, University Hospitals Case Address for correspondence Clodagh Mullen, MD, MacDonald
Medical Center, MacDonald Women’s Hospital, Cleveland, Ohio Women’s Hospital, University Hospitals Cleveland Medical Center,
2 Department of Obstetrics and Gynecology, University of North 11100 Euclid Avenue, Mailstop 5034 Cleveland, OH 44106
Carolina School of Medicine, Chapel Hill, North Carolina (e-mail: [email protected]).
3 Department of Pathology and Laboratory Medicine, NorthShore
University HealthSystem, Evanston, Illinois
4 Department of Obstetrics and Gynecology, Northwestern University
Feinberg School of Medicine, Chicago, Illinois
Am J Perinatol
Downloaded by: University of Tennessee. Copyrighted material.
Abstract Objective To assess the risk factors, adverse obstetrical outcomes, and recurrence
risk associated with pathologically diagnosed occult placenta accreta.
Methods This was a retrospective observational study of clinically adherent placentas
requiring manual extraction that underwent pathological examination. Cases were
defined as those with histological evidence of placenta accreta, and controls were
defined as those without accreta. All subsequent pregnancies were evaluated to
determine the recurrence risk of occult accreta in future pregnancies.
Results Of 491 women with clinically adherent placentas, 100 (20.1%) with a
pathological diagnosis of occult accreta were compared with 391 (79.9%) without
occult accreta. In bivariable analysis, risk factors associated with occult accreta
Keywords included a history of previous cesarean (19 vs. 10.7%; p ¼ 0.03) and prior uterine
► retained placenta surgery (35 vs. 19.7%; p ¼ 0.001). Adverse obstetrical outcomes were more common
► focal accreta in women with occult accreta including postpartum hemorrhage (59 vs. 31.7%;
► postpartum p < 0.001) and peripartum hysterectomy (21 vs. 0.3%; p < 0.001). In 130 subsequent
hemorrhage pregnancies, there was an increased risk of retained placenta (42.9 vs. 19%; p ¼ 0.04)
► placental pathology and recurrence of occult accreta (29.6 vs. 6.8%; p ¼ 0.05).
► occult accretas Conclusion Occult accreta is associated with an increased risk of hemorrhagic
► histological accreta morbidity and recurrence of morbidly adherent placenta in subsequent pregnancies.
Occult, or histological, placenta accreta, defined as a segment ogy of occult placenta accreta is similar to placenta accreta,3
of myometrial invasion to the basal plate that is not appre- the risk factors for and clinical significance of the patholo-
ciated on gross pathological examination, has been described gical diagnosis are not fully understood. Occult placenta
in the pathology literature for the past three decades.1,2 The accreta has also been referred to as focal accreta, pathologi-
diagnosis of occult accreta has no predefined size or percen- cally diagnosed accreta, and histological accreta.1–4
tage of placenta affected, just so long as it is able to separate Small observational studies have demonstrated an asso-
from the uterus in the clinical setting. While the histopathol- ciation between occult placenta accreta and previous uterine
received Copyright © by Thieme Medical DOI https://doi.org/
November 19, 2017 Publishers, Inc., 333 Seventh Avenue, 10.1055/s-0038-1669440.
accepted after revision New York, NY 10001, USA. ISSN 0735-1631.
July 11, 2018 Tel: +1(212) 584-4662.
Occult Placenta Accreta Mullen et al.
intervention such as dilation and curettage (D&C) or cesar- Results
ean section. In these studies, none of the cases resulted in a
hysterectomy, but there was an increased risk of adverse During the study period, there were 1,798 cases of manual
maternal outcomes.3 Placental retention during the third placental extraction (2.4% of all deliveries). Of these, 1,036
stage of labor is a significant cause of maternal morbidity and women were excluded from the analysis due to a lack of
mortality5–8; however, it is unclear if outcomes with occult placental adherence as the indication for manual extraction,
placenta accreta differ from clinically adherent placenta that is, no documented evidence of adherence after a cord
without evidence of myometrial invasion on pathology. avulsion or preferred removal after twins or cesarean deliv-
Thus, the objective of this study was to define risk factors, ery. An additional 203 were excluded due to a lack of
adverse obstetrical outcomes, and the recurrence risk asso- placental pathology, 61 for incomplete documentation, and
ciated with occult placenta accreta. 7 with an antepartum diagnosis of placenta accreta, leaving
491 women who met inclusion criteria (►Fig. 1). Women
with a diagnosis of occult accreta on placental pathology
Methods
(100 cases) were then compared with those without accreta
This was a retrospective observational study of patients with (391 controls).
adherent placentas requiring manual extraction at Prentice Women with occult accreta were more likely to be older,
Women’s Hospital from January 2009 to December 2014. have a lower body mass index (BMI), and have conceived
Women were included in the analysis if there was documen- through in vitro fertilization (IVF) (►Table 1). Those with
tation of placental adherence to the uterine wall without occult accreta were also more likely to have a history of
Downloaded by: University of Tennessee. Copyrighted material.
easy separation within 30 minutes of delivery as the indica- previous cesarean, prior uterine surgery or D&C, and a
tion for manual extraction of the placenta, which is standard history of Asherman’s syndrome. Furthermore, the occult
practice at this institution. Request for subsequent patholo- accreta group was more likely to have placenta previa (12 vs.
gical examination of the placenta by the provider was also 3.2%; p < 0.001). There was no antenatal imaging suggestive
required for inclusion in this study. Women were excluded if of accreta for any included previa.
there was antepartum imaging consistent with or suggestive Obstetric hemorrhagic complications were more common
of placenta accreta or diagnosis of accreta, percreta, or in women with occult accreta, including higher frequency of
increta at the time of delivery. postpartum hemorrhage, transfusion, maternal intensive
Electronic medical records were individually reviewed for care unit (ICU) admission, and peripartum hysterectomy
all cases of manually extracted placenta, and data were (►Table 2). All ICU admissions were for ongoing hemody-
abstracted to identify predisposing risk factors, maternal namic monitoring in the setting of excessive blood loss. Only
and neonatal outcomes in the incident pregnancy, and the one patient in the nonaccreta group received a hysterectomy
results of subsequent pregnancies. Hemorrhage was defined (0.3%) compared with 21 in the occult accreta group (21%;
as >500 mL in a vaginal delivery or >1,000 mL in a cesarean p < 0.001). The occult accreta group had a higher cesarean
delivery. Pathology reports were reviewed after standar- section rate (49.5 vs. 26.3%; p < 0.001). There was no sig-
dized pathological examination was performed by one of nificant difference in the frequency of chorioamnionitis or
two pathologists with perinatal expertise. Occult placenta
accreta was defined using the diagnostic histological features
of accreta: the presence of myometrial fibers at the basal Manually extracted
placentas in viable
plate and in contact with chorionic tissue or Rohr’s fibrinoid pregnancies
without intervening decidua.9 1798
Women with a diagnosis of occult placenta accreta on
pathology (cases) were compared with those without evi-
dence of accreta (controls). The primary objective of this Not retained
study was to define risk factors, adverse obstetrical out- 1036
No placental
comes, and a secondary outcome of recurrence risk asso- pathology
ciated with occult placenta accreta. Chi-square test, 203
Incomplete
Student’s t-test, and Mann–Whitney U-test were used for documentation
bivariable analysis, as appropriate, to compare risk factors for 61
and outcomes associated with occult placenta accreta. Multi-
Retained placentas
variable logistic regressions were performed for the out- with pathology
comes that were significantly different between the two 498
groups in bivariable analysis. Covariates were entered into Ultrasound concern
the regression equations if they differed between groups in for accreta
7
bivariable analysis at a level of p < 0.05. Odds ratios with 95%
confidence intervals were estimated from the regressions. Focal accreta No focal accreta
All tests were two-tailed, and p < 0.05 was used to define 100 391
significance. Approval for this study was obtained from the
Northwestern University Institutional Review Board. Fig. 1 Acquisition of cases and controls.
American Journal of Perinatology
Occult Placenta Accreta Mullen et al.
Table 1 Patient characteristics and potential risk factors for Table 2 Obstetric complications and procedures
occult accreta
No accreta Occult p-Value
No accreta Occult p-Value (n ¼ 391) accreta
(n ¼ 391) accreta (n ¼ 100)
(n ¼ 100) Antepartum bleeding 32 (8.18) 10 (10) 0.56
Maternal age (y) 33.5 5.1 35.4 5.7 0.002 Preeclampsia 35 (8.95) 8 (8) 0.76
Body mass index 29.9 5.1 28.6 4.5 0.02 Chorioamnionitis 24 (6.1) 5 (5) 0.67
(kg/m2)
Cesarean delivery 102 (26.1) 47 (47) < 0.001
Race/Ethnicity
Indication for cesarean
Non-Hispanic 224 (57.3) 64 (64) 0.58
white Prior cesarean 27 (26.5) 15 (31.9) 0.60
Non-Hispanic 33 (8.4) 8 (8) Nonreassuring 33 (32.4) 9 (19.1)
black fetal status
Hispanic 58 (14.8) 9 (9) Arrest disordera 28 (27.5) 8 (17)
Other 33 (8.4) 7 (7) Placenta previa 10 (9.8) 6 (12.8)
Declined 43 (11) 12 (12) Other 4 (3.9) 9 (19.1)
Gestational age 38.6 39 0.15 Postpartum 124 (31.7) 59 (59) <0.001
at delivery (wk) (36.5–40) (36.4–39.5) hemorrhage
Downloaded by: University of Tennessee. Copyrighted material.
Nulliparity 212 (54.2) 44 (44) 0.07 Postpartum D&C 89 (22.8) 24 (24) 0.79
Multiple gestation 41 (10.5) 7 (7) 0.29 IR embolization 7 (1.8) 2 (2) 0.89
IVF pregnancy 77 (19.7) 35 (35) 0.001 Hysterectomy 1 (0.3) 21 (21) < 0.001
Placenta previa 12 (3.1) 12 (12) <0.001 Blood transfusion 20 (5.12) 30 (30) <0.001
History of cesarean 42 (10.7) 19 (19) 0.03 Maternal ICU 2 (0.5) 10 (10) <0.001
admission
History of D&C 88 (22.5) 37 (37) 0.003
History of uterine 77 (19.7) 35 (35) 0.001 Abbreviations: D&C, dilation and curettage; IR, interventional
surgery radiology; ICU, intensive care unit.
Note: All data are presented as n (%).
History of 9 (2.3) 7 (7) 0.02 a
Arrest disorder includes arrest of dilation and arrest of descent.
Asherman’s
syndrome
Interpregnancy 28.7 10.4a 33.1 14.9b 0.06 Table 3 Multivariable regression of obstetrical complications
interval (mo)
Abbreviations: D&C, dilation and curettage; IVF, in vitro fertilization. Occult accreta, aOR (95% CI)a
Note: All data are presented as mean standard deviation, n (%), or Cesarean delivery 1.9 (1.1–3.2)
median (25–75%).
a
n ¼ 99. Postpartum hemorrhage 2.5 (1.5–4)
b
n ¼ 35. Hysterectomy 123.7 (13.9–1103)
Blood transfusion 7.8 (4–15.3)
preeclampsia between the two groups (►Table 2). There
Maternal ICU admission 21.9 (4.3–110.9)
were also no differences between the two groups with regard
to gestational age at delivery, neonatal ICU admission, low Abbreviations: aOR, adjusted odds ratio; CI, confidence interval;
Apgar scores, or birth weight. In multivariable analysis, ICU, intensive care unit.
a
occult accreta remained associated with all maternal hemor- Adjusted for maternal age, body mass index, in vitro fertilization,
history of cesarean, history of dilation and curettage, history of uterine
rhagic morbidities when controlling for risk factors of age,
surgery, and previa.
BMI, IVF, previa, and history of prior cesarean section, uterine
surgery, or D&C (►Table 3).
Overall, there were 130 observed subsequent pregnancies two had no obstetric complications, one required a hyster-
at the same institution for this study population, with only ectomy, and one experienced a postpartum hemorrhage,
14 occurring in the occult accreta group. The rate of sub- which was conservatively managed with balloon tamponade
sequent pregnancy, excluding women with a hysterectomy but required a blood transfusion. Of the nonaccreta group,
during the index pregnancy, was higher in the group without there were 22 retained placentas out of 116 pregnancies
accreta (29.7 vs. 17.9%; p ¼ 0.03). None of these subsequent (19%), of which 8 were occult accretas.
pregnancies had suspected accreta on prenatal testing. Of
those with occult accreta in the index pregnancy (n ¼ 14),
Discussion
there was a higher frequency of retained placenta (n ¼ 6;
42%), of which four were recurrent occult accreta (28%). In this retrospective study, we found that the identification of
Among those women with recurrent occult accreta (n ¼ 4), occult placenta accreta among women with manual removal of
American Journal of Perinatology
Occult Placenta Accreta Mullen et al.
the placenta was associated with similar risk factors and surgery become more prevalent in our obstetrical popula-
adverse maternal outcomes, which, in turn, were known to tion. There should be a heightened awareness of possible
be associated with placenta accreta. History of prior cesarean, occult placenta accreta and its associated major hemorrhagic
prior uterine surgery, IVF, and placenta previa in the current morbidity for patients with a retained placenta and risk
pregnancy were all significantly more common in the occult factors for accreta such as prior uterine surgery. Pathological
accreta group. The association between IVF and accreta has examination of the placenta should be considered in these
been described by Kaser et al, who found cryopreserved scenarios, as the results may impact management of future
embryo transfer as an independent risk factor.10 This may be pregnancies. Patients with occult placenta accreta should
a reflection of both uterine factors underlying the etiology of be counseled on the increased likelihood of recurrence and
infertility and lower endometrial thickness at the time of increased hemorrhagic morbidity in subsequent pregnan-
transfer. The outcomes with occult placenta accreta were cies. Providers should prepare for possible excessive blood
significantly more likely to have a major hemorrhagic morbid- loss and hysterectomy at the time of delivery with planned
ity including postpartum hemorrhage, blood transfusion, hys- delivery at a tertiary hospital with blood products available.
terectomy, and maternal ICU admission. Furthermore, in
subsequent pregnancies, we found an increased risk of mor- Condensation
bidly adherent placenta following index pregnancies with Occult placenta accreta is associated with significant hemor-
occult placenta accreta compared with a retained placenta rhagic morbidity and recurrence in future pregnancies.
without accreta (42 vs. 19%).
Our findings are consistent with that of Roeca et al who Note
Downloaded by: University of Tennessee. Copyrighted material.
found that the recurrence risk of pathologically diagnosed This paper was presented in a poster format at the 2017
accreta after an index pregnancy with placenta accreta was Annual Clinical and Scientific Meeting of the American
29.6%,4 which was significantly higher than those with a College of Obstetricians and Gynecologists, San Diego, CA,
retained placenta without occult accreta in our population May 6 to 9, 2017.
(6.8%). In this study of 39 subsequent pregnancies after an
index pregnancy with placenta accreta, 54% had either major Conflict of Interest
or minor morbidity in the index pregnancy, including post- None.
partum hemorrhage, blood transfusion, or hysterectomy, simi-
lar to the high rates of hemorrhagic morbidity in our study.
Those with morbidity in the index pregnancy had a 24% rate of
References
recurrence of accreta in the subsequent pregnancy, also con-
1 Jacques SM, Qureshi F, Trent VS, Ramirez NC. Placenta accreta:
sistent with our results. Our study, however, differs in design as mild cases diagnosed by placental examination. Int J Gynecol
it investigated the potential risk factors, pregnancy outcomes, Pathol 1996;15(01):28–33
and subsequent pregnancies associated with occult placenta 2 Gorodeski IG, Bahari CM, Holzinger M, Schachter A, Neri A.
accreta compared with retained placenta without accreta. Placenta previa with focal accretion. Isr J Med Sci 1982;18(02):
The strengths of our study include individual review of 277–280
3 Stanek J, Drummond Z. Occult placenta accreta: the missing link
medical records to validate our cohort of women with
in the diagnosis of abnormal placentation. Pediatr Dev Pathol
clinically adherent placentas and to abstract detailed mater- 2007;10(04):266–273
nal and neonatal outcomes. Standardized pathological exam- 4 Roeca C, Little SE, Carusi DA. Pathologically diagnosed placenta
ination of the placenta by one of two pathologists with accreta and hemorrhagic morbidity in a subsequent pregnancy.
perinatal expertise at our institution strengthened the con- Obstet Gynecol 2017;129(02):321–326
5 Urner F, Zimmermann R, Krafft A. Manual removal of the placenta
sistency of the diagnosis of occult placenta accreta; however,
after vaginal delivery: an unsolved problem in obstetrics. J Pregnancy
there is the limitation of an incomplete examination that 2014;2014:274651
may bias results toward the null. One important limitation to 6 Ashwal E, Melamed N, Hiersch L, Wiznitzer A, Yogev Y, Peled Y.
note, however, is the possibility of ascertainment bias at the The incidence and risk factors for retained placenta after vaginal
time of placental examination based on the knowledge of delivery - a single center experience. J Matern Fetal Neonatal Med
risk factors associated with placenta accreta. While the study 2014;27(18):1897–1900
7 Belachew J, Cnattingius S, Mulic-Lutvica A, Eurenius K, Axelsson O,
design allowed us to evaluate both factors predictive of occult
Wikström AK. Risk of retained placenta in women previously
accreta and adverse outcomes associated with a diagnosis of delivered by caesarean section: a population-based cohort study.
occult accreta, our power to detect a significant difference BJOG 2014;121(02):224–229
was diminished in the subsequent pregnancy analysis. 8 Combs CA, Laros RK Jr. Prolonged third stage of labor: morbidity
Furthermore, although we adjusted for all significant differ- and risk factors. Obstet Gynec 1991;77(06):863–867
ences between groups in multivariable analyses, we cannot 9 Linn RL, Miller ES, Lim G, Ernst LM. Adherent basal plate myometrial
fibers in the delivered placenta as a risk factor for development of
exclude the possibility of unmeasured confounding.
subsequent placenta accreta. Placenta 2015;36(12):1419–1424
This study helps to further our understanding of occult 10 Kaser DJ, Melamed A, Bormann CL, et al. Cryopreserved embryo
accreta in the setting of retained placenta, which will likely transfer is an independent risk factor for placenta accreta. Fertil
increase in frequency as the risk factors of IVF and uterine Steril 2015;103(05):1176–1184
American Journal of Perinatology
