The Journal of Maternal-Fetal & Neonatal Medicine

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/ijmf20

Placenta accreta spectrum: a hysterectomy can

be prevented in almost 80% of cases using a
resective-reconstructive technique

José Miguel Palacios-Jaraquemada, Angel Fiorillo, Jorge Hamer, Marcelo

Martínez & Claudio Bruno

To cite this article: José Miguel Palacios-Jaraquemada, Angel Fiorillo, Jorge Hamer, Marcelo
Martínez & Claudio Bruno (2022) Placenta accreta spectrum: a hysterectomy can be prevented
in almost 80% of cases using a resective-reconstructive technique, The Journal of Maternal-
Fetal & Neonatal Medicine, 35:2, 275-282, DOI: 10.1080/14767058.2020.1716715

To link to this article: https://doi.org/10.1080/14767058.2020.1716715

View supplementary material

Published online: 26 Jan 2020.

Submit your article to this journal

Article views: 1201

View related articles

View Crossmark data

Citing articles: 47 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ijmf20
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE
2022, VOL. 35, NO. 2, 275–282
https://doi.org/10.1080/14767058.2020.1716715

ORIGINAL ARTICLE

Placenta accreta spectrum: a hysterectomy can be prevented in almost 80%

of cases using a resective-reconstructive technique
Jos
e Miguel Palacios-Jaraquemadaa,b , Angel Fiorilloa, Jorge Hamerc, Marcelo Martınezd and
Claudio Brunod
a
Department of Obstetrics and Gynaecology, Cemic University Hospital, Buenos Aires, Argentina; bMRI Department, Fundacion
Cientıfica del Sur, Buenos Aires, Argentina; cUltrasound Department, CEGYR Reproductive Medicine, Buenos Aires, Argentina; dCYMSA
Clınica y Maternidad Suizo Argentina, Buenos Aires, Argentina

ABSTRACT ARTICLE HISTORY

Objective: To describe the use of surgical repair (One-step resective-conservative surgery) in all Received 9 October 2019
cases of placenta accreta spectrum. Revised 11 December 2019
Study design: Multicentre retrospective case series from tertiary referral hospitals in Argentina. Accepted 13 January 2020
A total of 452 patients were accepted from 12 hospitals presenting suspicion of invasive pla-
KEYWORDS
centa by auxiliary methods (ultrasound, Doppler and MRI). At the time of the surgery, placenta Conservative technique;
accreta spectrum was classified according to invasion topography (specific blood supply) and parametrial invasion;
local features (proximity to other structures, adhesion process, and multiple anastomotic blood placenta accreta; placenta
vessels). Type 1: upper posterior bladder; type 2: parametrial; type 3: low posterior bladder; and accreta spectrum;
type 4: low posterior bladder and fibrosis. After the ligature of newly formed vessels between placenta percreta
the uterus and pelvic organs, the fetus was delivered through an upper segmental hysterotomy.
Hemostasis was achieved by selective ligature of vesical-uterine and colpo-uterine vessels. Then,
the invaded myometrium and the entire placenta were removed totally in bloc and until
detected healthy tissue in both edges, to guarantee the most physiological hysterotomy in the
uterine segment. The uterus was closed with a polyglactin suture, double-layer technique. The
main outcome measurements were the uterine conservation, the blood loss and other complica-
tions classified according to intrasurgical classification.
Results: From 452 accepted patients, 326 patients had a confirmed diagnosis of placenta
accreta spectrum by histology analysis and surgical-clinical findings. In 126 cases, placenta
accreta spectrum was excluded used the same diagnostic criteria (Type 0 or false positive PAS).
They were identified 248 cases as type 1, 44 as type 2, 23 as type 3 and 11 as type 4. Uterine
conservation was possible in the 81% of type 1 invasion with 500 mL of blood loss (interquartile
range, IQR ¼ Q3  Q1). The modified Pfannenstiel was the most commonly used incision, while
midline incision was chosen in all emergencies or in patients with a previous midline incision.
Hysterotomy made in the upper part of the uterine segment presented normally attached pla-
centas and not accreta. Selective vessel ligature, also named custom-made hemostasis method
(CMHM) was effective at stopping or preventing bleeding associated with PAS. The entire pla-
centa and the invaded area are removed in block, to guarantee to perform the uterine repair
with healthy tissue and to avoid a recurrence in the subsequent cesarean. The uterine-ovary
artery axis is never occluded or obliterate to guarantee the uterine-endometrial and ovary blood
supply as before surgery. No significant differences existed according to the population; how-
ever, the presence of total occlusive placenta previa was more frequent in types 3 and 4, which
were also associated with older mothers and age-related collagen changes. Lateral and lower
segment invasions (types 2 and 3) were most commonly associated with previous terminations
of pregnancy, curettage, and manual removal of the placenta. Blood loss and technical difficulty
were clearly associated to the invaded area, while invasion degree was a poor marker to predict-
ing bleeding or complications in all locations Uterine conservation was possible in 202/248
(81.5%) of type 1, 21/44 (47.7%) of type 2, 5/23 (21.8%) of type 3 and 0/11 (0%) of type four
cases. Type 0 (false positive) were excluded of statistical analysis, and the uterus was preserved
in 100% of cases. In a separate report, we will describe the maternal and fetal outcomes as well
as 204 subsequent pregnancies after the use of one-step resective reconstructive technique.

CONTACT Jose Miguel Palacios-Jaraquemada [email protected] Department of Obstetrics and Gynaecology, Cemic University Hospital, Galvan
4102, C1431 Ciudad Aut
onoma de Buenos Aires, Argentina
Supplemental data for this article is available online at https://doi.org/10.1080/14767058.2020.1716715
ß 2020 Informa UK Limited, trading as Taylor & Francis Group
276 J. M. PALACIOS-JARAQUEMADA ET AL.

Conclusions: Using the resective-reconstructive approach (one-step conservative surgery) to the

management of invasive placenta, the uterus can be preserved with minimal morbidity and
reduced blood loss in almost 80% of cases. Precis preventing hysterectomy in 80% of placenta
accreta spectrum.

Introduction uterine wall, minimizing the possibility of recurrence.

In this paper, we describe the operative results using
The incidence of placenta accreta spectrum (PAS) is
a resective reconstructive approach according to PAS
rising worldwide and mainly result of rising cesarean
intrasurgical staging.
section rates. Prior to the twentieth century, almost all
cases of PAS were lethal due to massive bleeding. The
first successful case of conservative treatment of PAS Materials and methods
was published in 1933 [1], but complications such as
Four hundred nine patients had a previous image
infection and unexpected hemorrhage remain possible
diagnosis of PAS made by specialized ultrasound,
[2]. Conservative treatment requires such a high level
Doppler and MRI and 43 were diagnosed only at the
of surgical expertise [3] that in up to 75% of cases,
time of cesarean section. Four hundred eleven
cesarean hysterectomy is performed [4,5]. However, patients had elective scheduled surgery and 41 had
uterine reconstructive techniques can be used to emergency surgery. In 126 cases, the placenta was
reduce the frequency of morbidity [6] and to preserve adherent to the dehiscent uterine scar, but without
the uterus for a subsequent pregnancy [7]. Currently, invading it (Type 1, false positive) hence, they were
there is not a general recommended approach to excluded from this study. Thus, 326 patients with a
management, partly because most studies fail to true diagnosis of invasive placenta (histologically con-
accurately categorize PAS. Preoperative ultrasound firmed) were included for surgical staging and statis-
evaluation is an excellent tool for diagnosis, but there tical analysis. Clinical diagnosis of PAS included:
is a poor correlation between images and technical increased, tortuous vascularity along the uterine sur-
complications that arise at surgery [8]. In particular, face, and a markedly distended lower uterine segment
the usefulness of ultrasound relies upon the operator that bulges toward the pelvic sidewalls, an evident
skill and experience [9]. A widely applicable prenatal placental invasion thought uterine surface or pelvic
screening protocol has the potential to improve out- organs. When PAS was diagnosed at cesarean section
come and avoid the most common obstetric complica- and the fetal-maternal condition was good, the sur-
tions [10]. Identifying new vascularization with gery was delayed until skilled assistance was available.
determination of the involved organ and, the identifi- While fetal or maternal conditions needed immediate
cation of adhesions and the topographical delineation delivery, a fundal hysterotomy was performed to
of placental invasion, are all important features to be deliver the baby and definitive surgery was delayed.
considered before surgery [11], like necessary skills The umbilical cord was then cut, and the uterus and
[12,13]. In 1995, we introduced the one-step conserva- laparotomy closed until expert assistance was avail-
tive surgery, including selective devascularization, full able. Objective topographic invasion parameters and
resection of the invaded tissues and repair in a single local features at surgery determined the following PAS
operation. In this technique, selective ligature of newly classification into four groups (Figures S1–S2): type 0:
formed vessels performed uterine hemostasis, thereby the placenta reached the serosa through a uterine
occluding the vesical-uterine and the colpo-uterine wall defect, adhesive but without the true placental
blood supply [14]. After resection of placenta and invasion or presence of newly-vascularization or; this
invaded area, the uterus is then sutured in two layers, group was considered a “false PAS”; type 1: the pla-
and the posterior bladder muscular layer is reinforced centa reached to or beyond the serosa, with the pres-
by a double layer suture of 000 polyglactin acid. This ence of newly-formed vessels between the uterus,
procedure restores uterine anatomy and preserves its placenta, and bladder and in relation with the upper
reproductive function for subsequent pregnancies. posterior bladder; type 2: the placenta reached to or
Abnormal decidualization over scar tissue is strongly beyond the serosa with placental parametrial invasion;
associated with PAS [8] therefore, the complete type 3: the placenta invaded the posteroinferior area
removal of the placenta and involved uterine tissue of the bladder (trigone-cervical invasion) with shared
optimize the chances of reestablishing a normal blood supply between the bladder, cervix and the
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 277

Table 1. Population features.

Invasion type Total (326) T1 (248) T2 (44) T3 (23) T4 (11) P
Age (mean SD) 31.85 (5.55) 31.33 (5.80) 32.32 (4.69) 34.43 (4.70) 36.81 (3.76) F ¼ 5.41
0.0012 (1)
Parity (median P10–P90) 2 (0–4) 2 (0–4) 1 (0–3) 1 (1–4) 1 (1–3) Chi-squared ¼ 9.761 with 3 df
p ¼ .0207 (2)
BMI (mean SD) 23.62 (2.10) 23.42 (2.09) 24.30 (1.96) 24.00 (2.09) 24.64 (2.35) F ¼ 3.41
p ¼ .0179 (1)
Number of previous cesarean 1 (0–3) 1 (0–4) 1 (0–2) 1 (0–3) 1 (0–3) Chi-squared ¼ 5.385 with 3 df
(median P10–P90) p ¼ .1457 (2)
Total occlusive placenta previa (%) 34.66 34.68 0.00 78.26 81.82 Pearson chi2 (3) ¼ 53.446 p < .001
D&C (%) 46.32 39.52 72.73 69.57 45.45 Pearson chi2 (3) ¼ 21.958 p < .001
Analysis of Variance; Kruskal–Wallis equality-of-populations rank test; Pearson chi2.
df: Degrees of freedom; T1: upper bladder; T2: parametrium; T3: low bladder; T4: low bladder þ fibrosis.

vagina; type 4: has an identical features that type 3, then a continuous suture with a suture of 1 polyglac-
but with the addition of intensive fibrosis between the tin acid. In case of dense adhesions between the pla-
bladder and uterus. Table S2 shows a detailed justifi- cental invasion and the posterosuperior bladder wall
cation of topography classification and its implication we used retrovesical finger bypass [20]. When the
for clinical practice. The laparotomy approach used bladder was opened, damaged borders were resected
was mostly a modified Pfannenstiel or midline incision. with sharp scissors and it was repaired in two layers
The modified Pfannenstiel involved cutting skin in the with 3-0 polyglactin acid (Figure S10). After suture,
standard way and then the cephalic skin and subcuta- urine leakage was tested with an infusion of methy-
neous fat were dissected as a flap over the anterior lene blue solution. The bladder muscular layer was
abdominal fascia as far up as the umbilicus. Then the usually reinforced with 3-0 polyglactin acid sutures. In
anterior aponeurosis of the rectus abdominis was all cases we placed a sheet of 10  15 cm of regener-
opened in the midline. The uterus was dissected free ated cellulose (SurgicelTM, Ethicon) between the uterus
of the posterior wall of the bladder as far down as the and the bladder as antiadherent barrier (Figure S11).
cervix while the vesicouterine vessels were ligated After resection of the invaded area or the hysterec-
with Vicryl 1 (Figures S3, S4 and S5). The peritoneum tomy, fibrinogen was measured using the Clauss
medial to the round ligaments was cut to separate the method [21]. In a level below 200 mg/dL we adminis-
TM
pelvic subperitoneal fat from the uterus in the ceph- tered 2–3 vials of fibrinogen (Hemocomplettant , CSL
alic-caudal direction, thus allowing the visualization of Behring, Germany) or 2 U of cryoprecipitate/kg/body
the parametrial space. Hysterotomy was performed in weight. All stable patients received low molecular
the upper segment, immediately above of myometrial weight heparin, i.e. enoxaparine 40 mg/day sc.
TM
invasion area (Figure S6). Lower uterine hemostasis (Clexane , Sanofi-Aventis, Argentina) 8 h after surgery
was performed using square compression sutures [15] provided the platelet count was 100,000 and contin-
or by ligature of the colpouterine vessels (Figure S7) ued until patients were fully mobile [13]. The rate of
at 3, 6 and 12 o’clock [16]. In diagnosed severe type uterine conservation and blood loss was analyzed
3–4 invasion, an elastomeric balloon was placed at the according to the PAS intrasurgical classification. A pla-
level of the infrarenal aorta. In cases of unexpectedly cental pathologist specialist analyzed all specimens
heavy bleeding, blood loss was controlled by manual (placentas and invaded areas) according to 3 marked
compression of the aorta over the sacral promontory areas on the surgical specimen made before formalde-
or an emergency inflating of an aortic balloon [17]. To hyde fixation (Figure S12). In multiple invasion degree
estimate blood loss, we resorted to visual assessment areas, diagnosis was made by the more extensive one.
complemented, in cases of major loss, by measure- The main population features are described in Table 1.
ment of hemodynamic parameters including blood In a separate report, we will describe the maternal
lactate, arterial pH, base excess, bicarbonate, and urin- and fetal outcomes as well as 204 subsequent preg-
ary output [18,19]. A conservative uterine procedure nancies after uterine repair.
was only attempted with stable hemodynamic param-
eters and a presence of 2 cm of healthy myometrium
Statistical analysis
visually identified above the cervix. In all resective-
conservative cases, all invaded myometrium and the The main surgical variables were analyzed in 326 true
entire placenta was removed (Figures S8 and S9). The cases of PAS; clinical features at diagnosis is described
uterus was repaired in two layers, first mattress and in Table 2. In order to investigate variations among
278 J. M. PALACIOS-JARAQUEMADA ET AL.

Table 2. Clinical features.

Total (326) T1 (248) T2 (44) T3 (23) T4 (11) p
Gestational age at diagnosis in days (Mean SD) 184.84 (38.64) 188.91 (37.19) 191.18 (35.31) 148.21 (32.46) 144.45 (34.75) F ¼ 13.63 p < .001 (1)
Gestational age at birth in days (Mean SD) 249.47 (8.36) 250.03 (7.55) 252.15 (7.58) 237.95 (10.7) 250.36 (4.17) F ¼ 19.23 p < .001 (1)
Time from diagnosis to birth in days (Mean SD) 64.63 (35.26) 61.12 (33.91) 60.97 (30.59) 89.73 (36.74) 105.90 (36.33) F ¼ 10.78 p < .001 (1)
Analysis of variance; T1: upper bladder; T2: parametrium; T3: low bladder; T4: low bladder þ fibrosis.

Table 3. Type of surgery.

Total (326) T1 (248) T2 (44) T3 (23) T4 (11) p
Emergency surgery (%) 13.80 11.29 11.36 52.17 0.00 Pearson chi2 (3) ¼ 31.758 p < .001
Hysterectomy (%) 29.45 18.15 50.00 78.26 100.00 Pearson chi2 (3) ¼ 76.926 p < .001
Cause of hysterectomy
Hemorrhage (%) 3.99 2.82 4.55 17.39 0.00 Pearson chi2 (9) ¼ 99.889 p < .001
Lack of tissue to repair (%) 23.31 12.50 45.45 60.87 100.00
Elective (%) 2.15 2.82 0.00 0.00 0.00
T1: upper bladder; T2: parametrium; T3: low bladder; T4: low bladder þ fibrosis.

and between groups, we used analysis of variance complications. Manual aortic compression was used in
(ANOVA). Fisher’s exact test was used to assess statis- 26 cases of massive unexpected bleeding during dis-
tical significance when sample sizes were small while section of the lower uterine segment (below the peri-
the Kruskal–Wallis test was used, as a nonparametric toneal reflection). Hysterectomy was performed in 98
method, for testing whether samples originated from cases, 93 of them because of extensive lower uterine
the same distribution and for comparing two or more tissue damage and the remaining 6, owing to untreat-
independent samples of equal or different sample able bleeding. Uterine conservation was possible in
sizes. All statistical analyses were performed using 202/248 (81.5%) of type 1, 21/44 (47.7%) of type 2, 5/
StataCorp. 2015. Stata Statistical Software: Release 14. 23 (21.8%) of type 3 and 0/11 (0%) of type four cases.
College Station, TX: StataCorp LP. The interquartile Lower parametrial invasion (Figure S13) and trigonal
range was used to determine the distribution of a bladder invasion (Figure S14, 15 and 16) were associ-
numerical variable that contains the middle 50% ated with increased bleeding and technical difficulty.
(approximately) of the values in the distribution. The bladder was dissected without the requirement
Microsoft Excel (2010). The Institutional Review Board for any bladder tissue resection, but it was opened in
of CEMIC University Hospital approved the study (CE 46 patients due to dense adhesions or detrusor thin-
1045). 25 July 2016 ning. We used retrovesical finger bypass in 16 cases of
dense bladder fibrosis. Complications can be catego-
rized as grade 1: included minor risk events not requir-
Results
ing invading therapy (36 cases-11.04%); grade 2:
We included 326 patients with a true diagnosis of included potentially life-threatening complications
PAS: 248 patients were identified as type 1, 44 as type with the need of intervention (25 cases-7.6%); grade 3:
2, 23 as type 3 and 11 as type 4. In total, 302 patients complications were defined as complications leading
were operated on using a modified Pfannenstiel inci- to lasting disability or organ resection, except uterus
sion; 50 were operated on using a midline incision, 41 (1 case-0.003%), and grade 4: complication indicated
as emergency surgery and 9 through a previous mid- death of a patient due to a complication (0 case-0%).
line incision. Planned or emergency surgery and failure Complications are summarized in Table 4. Bleeding
cause of conservative procedure are described in was closely related to the invasion topography
Table 3. Local hemostasis performed at the cervical- (Table 5) and types were associated with the specific
vaginal junction before resecting the invaded area blood supply of each of them. Lower vesical and sub-
resulted in a manifest visual reduction of blood loss peritoneal parametrial invasions were the bloodiest
after removal of the placenta. In six cases of diag- invasion areas. Estimated blood loss in type 1 was
nosed severe trigonal and cervical vascular invasion, 500–1000 ml, 500–1500 ml in type 2; 500–2500 ml in
we placed elastomeric aortic balloons; nevertheless, type 3 and 1000–2500 ml in type 4. Interquartile range
they were only inflated in two cases for massive for groups is shown in Table 6. Overall, in 107 cases
bleeding for 8 and 15 min respectively. In one case, a occurred urologic complications, but, in 97, 19% of
femoral arterial thrombosis was detected during fol- them consisting of bladder opening or muscular wall
low-up and treated by open thrombectomy without reinforcement. In four cases the ureters were ligated,
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 279

Table 4. Complications.
Total (326) (%) T1 (248) (%) T2 (44) (%) T3 (23) (%) T4 (11) (%) p
Total of complications 40.49 31.05 50.00 95.65 100.00 Pearson chi2 (3) ¼ 56.0384 p<.001
Hemorrhage > 1500 ml (IO) 28.83 22.58 40.91 56.52 63.64 Pearson chi2 (3) ¼ 22.937 p<.001
Global urinary complications 32.82 22.18 45.45 91.30 100.00 Pearson chi2 (3) ¼ 74.119 p<.001
Bladder opening (IO) 14.42 8.87 15.91 30.43 100.00 Pearson chi2 (3) ¼ 76.342 p<.001
Bleeding from posterior bladder (PO) 3.07 2.02 4.55 13.04 0.00 Fisher’s exact ¼ 0.04
Bladder muscular reinforce (IO) 17.79 11.69 18.18 52.17 81.82 Pearson chi2 (3) ¼ 55.730 p<.001
Urinary retention (PO) 1.23 1.61 0.00 0.00 0.00 Fisher’s exact ¼ 1
Ureteral damage (IO) 1.23 0.00 9.09 0.00 0.00 Fisher’s exact ¼ 1
Disseminated intravascular coagulation (IO) 3.68 2.82 2.27 17.39 0.00 Fisher’s exact ¼ 0.026
Hypovolemic shock (IO) 0.92 0.40 0.00 8.70 0.00 Fisher’s exact ¼ 0.038
Reoperations 2.45 1.21 6.82 8.70 0.00 Fisher’s exact ¼ 0.027
Femoral thrombosis (PO) 0.31 0.00 0.00 4.35 0.00 Fisher’s exact ¼ 0.104
Prolonged ileous (PO) 3.99 3.23 6.82 8.70 0.00 Fisher’s exact ¼ 0.276
Metabolic acidosis (IO) 1.84 1.61 2.27 4.35 0.00 Fisher’s exact ¼ 0.493
Suprafascial hematoma (PO) 0.31 0.40 0.00 0.00 0.00 Fisher’s exact ¼ 0.104
Secondary uterine bleeding (PO) 0.31 0.40 0.00 0.00 0.00 Fisher’s exact ¼ 1
T1: upper bladder; T2: parametrium; T3: low blader; T4: low bladder þ fibrosis; IO: intraoperative; PO: postoperative.

Table 5. Bleeding. Eight cases showed continuous vaginal bleeding

Type which appeared within 3 h of the procedure; these
Hemorrhage T1 n ¼ 248 T2 n ¼ 44 T3 n ¼ 23 T4 n ¼ 11 Total cases were associated with a low level of fibrinogen
Grade 1 5 0 0 0 5 (less than 100 mg/dL). They were reoperated on after
(<750 ml) (%) 2.02 0.00 0.00 0.00 1.53 a cryoprecipitate infusion of lyophilized fibrinogen
Grade 2 187 26 10 4 227 TM
(750–1500 ml) (%) 75.40 59.09 43.48 36.36 69.63 (Hemocomplettant , CSL Behring). In five of these
Grade 3 30 6 5 5 46 cases, the uterus was preserved after the correction of
(1500–2000 ml) (%) 12.10 13.64 21.74 45.45 14.11
Grade 4 26 10 7 2 45 coagulation and the additional use of vascular control
2000–3000 ml (%) 10.48 22.73 30.43 18.18 13.80 stitches. In the remaining three cases, the hemo-
Grade 5 0 2 1 0 3
>3000 ml (%) 0.00 4.55 4.35 0.00 0.92 dynamic parameters remained unstable and hysterec-
Total 248 44 23 11 326 tomy was performed. In one patient, open surgery
100.00 100.00 100.00 100.00 100.00
removed retained laparotomy gauze.
Pearson chi2 (12) ¼37.6791; Pr ¼ 0.000.
Forty-three previously undiagnosed cases of PAS
discovered at cesarean section were managed accord-
Table 6. Bleeding.
ing to the maternal and/or fetal condition. In 12/43
Type T1 n ¼ 248 T2 n ¼ 44 T3 n ¼ 23 T4 n ¼ 11
(27.9%) cases, no intervention was needed until after
Q1 min (25%) 1000 ml 1500 1000 ml 1500 ml
Q2 median (50%) 1500 ml 1500 ml 2000 ml 2000 ml specialized surgical exploration was available
Q3 (75%) 1500 ml 2000 ml 2000 ml 2000 ml (30–45 min). In 26/43 (60.46%) cases, the baby was
IQR 500 ml 500 ml 1000 ml 500 ml
delivered through a uterine incision away from the
IQR: interquartile range; T1: upper bladder; T2: parametrium; T3: low
bladder; T4: low bladder þ fibrosis. placental attachment without touching the placenta
IQR ¼ Q3  Q1. and definitive surgery was delayed by 48–72 h. In 5/43
(11.62%) cases, the baby was delivered by fundal
all cases of extensive parametrial invasion in which access, but the obstetrician insisted on attempting to
ureteral catheterization was not possible or available. detach the placenta, which resulted in massive bleed-
Detection of all ureteral injuries occurred within the ing. In this group, prompt abdominal internal aortic
first postoperative week and they were treated with compression controlled the bleeding. Hysterotomy
early ureter reimplantation except for one case in made in the upper part of the uterine segment [22]
which intensive ureteropelvic junction fibrosis made it presented normally attached placentas and not
impossible to perform the anastomosis and nephrec- accreta. Custom-made hemostasis method (CMHM)
tomy was necessary. One case was reoperated on indicates a selective vessel ligature through a previous
twice, firstly due to an infection of the remaining opening of pelvic fascia’s, the procedure was effective
lower segment, and secondly due to an abscess in the at stopping or preventing bleeding associated with
vaginal vault (drained by open surgery). Seven PAS [7,16]. No significant differences existed according
patients (5 from hysterectomy and 2 from conservative to the population; however, the presence of total
group) were reoperated on due to abdominal bleed- occlusive placenta previa was more frequent in types
ing originating from cauterized retrovesical vessels; 3 and 4, which were also associated with older moth-
hemostasis was achieved using 3-0 polyglactin sutures. ers and age-related collagen changes [8,23]. Lateral
280 J. M. PALACIOS-JARAQUEMADA ET AL.

and lower segment invasions (types 2 and 3) were histopathology confirmation. Patients without intrasur-
most commonly associated with previous terminations gical or histopathological diagnosis of PAS were
of pregnancy, curettage, and manual removal of the excluded from the statistical analysis. Skill in the surgi-
placenta [24]. cal technique, including special hemostasis and dissec-
Use of hemocomponents was variable, although tion maneuvers was acquired by senior obstetricians
superior in hysterectomy than in conservative cases. assisting at 5–7 surgeries before taking charge of
Its use was based in individual preferences and did cases themselves, to initially ensure the reproducibility
not follow a preestablished protocol (Table S1). of the methodology.
Fibrinogen dropped below 200 mg/dL after removal
of the invaded area or hysterectomy in 29/248 (11.7%)
Interpretation
patients. Histopathologist analyses not always
matched with the invasion degree seen by objective Placental invasion is customary classified by using a
invasion features, fact that is coincident with current histological degree, but this is not useful for surgical
pathological review [25]. This may occur because in purposes, especially because the diagnosis of invasion
one piece, many invasion degrees could coexist. Blood degree is retrospective. It is recognized, that prenatal
loss according to invasion degree by type is shown in diagnosis of PAS is not 100% accurate, and although
Table S3. Cases undiagnosed before the cesarean sec- relation between images and complications was investi-
tion were not associated with more complications gated, this has proved not possible up till now
except when the placenta was removed without previ- [8,23–27]. In contrast, the invasion topography is a good
ous surgical intervention to ensure hemostasis. A com- predictor of complications and blood loss [28,29]
plete and anonymous data of the 326 patients was although its potential is not thus far fully understanding.
included in the Table S1. PAS type 1 was the most frequent PAS type (80% of
cases) and that it is associated with less bleeding and a
higher rate of conservative reconstructive treatment
Discussion
than the other groups. Although a planned delivery is
Main findings the goal, a contingency plan or a strategy should be dis-
Preoperative topography of PAS enabled us to predict cussed for each patient [30]. This is particularly relevant
the likelihood of bleeding, postoperative complica- in view of the high percentage of false negative cases
tions, and the possibility of uterine repair. Knowledge [4]. History of “unsafe abortion” must be sought in all
of the specific arterial pedicles by areas allow the use patients with a PAS [31,32] so as to any unusual and
of specific vascular control methods. Surgical classifica- dangerous locations can be anticipated. This is particu-
tion could be performed even in undiagnosed cases larly important in low parametrial invasions [33] which
of PAS, using simple and reversible measures during require the use of ureteral catheterization or surgical
cesarean, such as proper opening of the pelvic fascia’s identification to avoid unintentional injuries [34],
and wide retrovesical dissection. PAS type 1 was the although and even using preventive measures, damage
most frequent category (80% of cases) and it was is not always evitable [35,36] In this group (T2) blood
associated with less bleeding and a higher rate of loss could be extremely difficult to control. Areas with
uterine conservation than the other groups. Surgical an exuberantly interconnected blood supply or aberrant
management of PAS may be individualized according vessels could be cause of unintentional damage during
surgical findings and to reduce the possibility of false embolization [37,38]. Training in obstetric vascular anat-
positive. Custom-made hemostasis demonstrated a omy allowed the use of low-cost selective hemostasis,
high accuracy, reducing surgery time, as the use of especially by compression hemostasis [39] that there
expensive equipment and some complications associ- are low-priced than endovascular methods [40]. In the
ated to embolization techniques. recent years, the International Federation of
Gynecology and Obstetrics (FIGO) proposed a new clas-
sification of the placenta accreta spectrum cases [41]
Strengths and limitations
according clinical and histological criteria. Although this
All the patients included in the study were diagnosed, proposal was never validated by any surgical series, we
operated on, and managed by the authors according analyzed our cases according this suggestion in the
to a described technique and methods. PAS was con- Table S3. The evaluation proved that, by our series,
firmed in all involved patients according clinical crite- there is no correlation among the FIGO classification
ria, image analysis (US, Doppler and MRI) and with respect to blood loss and others significant
 THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE 281

problems during surgery. In addition, a wide variation of normales enza alcuna complicanza (reasorbimiento
the degree of placental invasion for the same specimen autodigestione uterina della placenta?). Policlin. 1933;
40(9):347–349 [in Italian].
was recently published [42]. This fact will make it diffi-
[2] Teo SB, Kanagalingam D, Tan HK, et al. Massive post-
cult to use the classification of FIGO, since it is not speci- partum haemorrhage after uterus-conserving surgery
fied how to classify a case that has multiple degrees of in placenta percreta: the danger of the partial pla-
invasion. This concept is not new and was originally centa percreta. BJOG. 2008;115(6):789–792.
published in 2016 by a group of pathologists at the [3] Fox KA, Shamshirsaz AA, Carusi D, et al. Conservative
Beth Israel Deaconess Medical Center, Boston, USA [26]. management of morbidly adherent placenta: expert
review. Am J Obstet Gynecol. 2015;213(6):755–760.
Other scores or PAS classification were created and vali-
[4] Fitzpatrick KE, Sellers S, Spark P, et al. The manage-
dated by case series with promissory results [43]. A pro- ment and outcomes of placenta accreta, increta, and
spective, randomized and comparative study is percreta in the UK: a population-based descriptive
especially difficult to perform in PAS. The inherent vari- study. BJOG. 2014;121(1):62–71.
ability of surgery requires the precise definition of inter- [5] Jauniaux E, Collins SL, Jurkovic D, et al. Accreta pla-
ventions and the close monitoring of quality, which centation: a systematic review of prenatal ultrasound
imaging and grading of villous invasiveness. Am J
depend on a learning curve. Life threatening situations Obstet Gynecol. 2016;215(6):712–721.
may cause of difficulties with recruitment, consent, and [6] Chandraharan E, Rao S, Belli AM, et al. The triple-P
randomization [44]. In some communities with high procedure as a conservative surgical alternative to
incidence of PAS, habitually women refuse to accept a peripartum hysterectomy for placenta percreta. Int J
hysterectomy as the primary plan since they perceived Gynecol Obstet. 2012;117(2):191–194.
[7] Palacios Jaraquemada JM, Pesaresi M, Nassif JC, et al.
hysterectomy as a loss of female identity [45,46]. We
Anterior placenta percreta: surgical approach, hemo-
suggest using evidence derived from large case series stasis and uterine repair. Acta Obstet Gynecol Scand.
to inform decision-making when no other higher level 2004;83(8):738–744.
of evidence is available [47]. [8] Jauniaux E, Collins S, Burton GJ. Placenta accreta
spectrum: pathophysiology and evidence-based anat-
omy for prenatal ultrasound imaging. Am J Obstet
Conclusions Gynecol. 2018;218(1):75–87.
[9] Rac MW, Dashe JS, Wells CE, et al. Ultrasound predic-
Use of one-step resective-reconstructive technique in tors of placental invasion: the Placenta Accreta Index.
PAS allows the preservation of the uterus in an increased Am J Obstet Gynecol. 2015;212(3):343.e1–343.e7.
number of cases. Selective hemostasis reduces blood [10] Jauniaux E, Bhide A. Prenatal ultrasound diagnosis
loss, surgical time and costs. Intrasurgical classification of and outcome of placenta previa accreta after cesar-
ean delivery: a systematic review and meta-analysis.
placental invasion according to topographic areas has a
Am J Obstet Gynecol. 2017;217(1):27–36.
close correlation with bleeding and complications, which [11] Palacios-Jaraquemada JM, Bruno C, Clavelli W. Morbid
allows the selection of a surgical strategy designed to adherent placenta: prediction, diagnosis and manage-
reduce maternal morbidity and mortality. ment. Fet Matern Med Rev. 2007;18(4):357–381.
[12] Shamshirsaz AA, Fox KA, Salmanian B, et al. Maternal
morbidity in patients with morbidly adherent pla-
Acknowledgements centa treated with and without a standardized multi-
disciplinary approach. Am J Obstet Gynecol. 2015;
We thank Professor P. J. Steer and P. Prefumo for reviewing
212(2):218.e1–218.e9.
the manuscript and editing it for English. No financial sup-
[13] Silver RM, Fox KA, Barton JR, et al. Center of excel-
port was received for this study.
lence for placenta accreta. Am J Obstet Gynecol.
Review. 2015;212(5):561–568.
[14] Palacios-Jaraquemada JM, et al. One-Step conserva-
Disclosure statement
tive surgery for abnormal invasive placenta (placenta
No potential conflict of interest was reported by the authors. accreta, increta, percreta). In. Arulkumaran S, Karoshi
M, Keith LG, editors. A comprehensive textbook of
postpartum hemorrhage: an essential clinical refer-
ORCID ence for effective management. 2nd ed. London:
Sapiens Publishing Ltd; 2012. p. 263–271.
Jose Miguel Palacios-Jaraquemada http://orcid.org/0000-
[15] Cho JH, Jun HS, Lee CN. Hemostatic suturing tech-
0002-5240-5320
nique for uterine bleeding during cesarean delivery.
Obstet Gynecol. 2000;96(1):129–131.
References [16] Palacios-Jaraquemada JM. Efficacy of surgical techni-
ques to control obstetric hemorrhage: analysis of 539
[1] Capechi E. Placenta accreta abandonata in utero cases. Acta Obstet Gynecol Scand. 2011;90(9):
cesarizzato. Ritorno progressivo di questo allo stato 1036–1042.
282 J. M. PALACIOS-JARAQUEMADA ET AL.

[17] Wu Q, Liu Z, Zhao X, et al. Outcome of pregnancies [33] Palacios-Jaraquemada JM, Bruno CH, Martın E. MRI in
after balloon occlusion of the infrarenal abdominal the diagnosis and surgical management of abnormal
aorta during caesarean in 230 patients with placenta placentation. Acta Obstet Gynecol Scand. 2013;92(4):
praevia accreta. Cardiovasc Intervent Radiol. 2016; 392–397.
39(11):1573–1579. [34] Baggish MS, Karram MM. Atlas of pelvic anatomy and
[18] Zuckerwise LC, Pettker CM, Illuzzi J, et al. Use of a gynecologic surgery. Philadelphia: WB Saunders; 2001.
novel visual aid to improve estimation of obstetric p. 204–213.
blood loss. Obstet Gynecol. 2014;123(5):982–986. [35] Crocetto F, Esposito R, Saccone G, et al. Use of routine
[19] Schorn MN. Measurement of blood loss: review of the ureteral stents in cesarean hysterectomy for placenta
literature. J Midwif Womens Health. 2010;55(1):20–27. accreta. J Matern Fetal Neonatal Med. 2019;1–4.
[20] Pelosi MA, 3rd, Pelosi MA. Modified cesarean hyster- [36] Celik S, Celik H, Soyer Calıskan C, et al. Bladder filling
ectomy for placenta previa percreta with bladder before accreta surgery is a very effective method for
invasion: retrovesical lower uterine segment bypass. preventing bladder injury: a retrospective cohort
Obstet Gynecol. 1999;93(5 Pt 2):830–833. study. J Matern Fetal Neonatal Med. 2019;1–6. DOI:10.
[21] Mackie IJ, Kitchen S, Machin SJ, on behalf of the 1080/14767058.2019.1670809
Haemostasis and Thrombosis Task Force of the British [37] Qiu J, Fu Y, Huang X, et al. Acute pulmonary embol-
Committee for Standards in Haematology, et al. ism in a patient with cesarean scar pregnancy after
Guidelines on fibrinogen assays. Br J Haematol. 2003; receiving uterine artery embolization: a case report.
121(3):396–404. TCRM. 2018;14:117–120.
[22] Ward CR. Avoiding an incision through the anterior [38] Baker CM, Winkel CA, Subramanian S, et al. Estimated
previa at cesarean delivery. Obstet Gynecol. 2003; costs for uterine artery embolization and abdominal
102(3):552–554. myomectomy for uterine leiomyomata: a comparative
[23] Nelson SM, Telfer EE, Anderson RA. The ageing ovary study at a single institution. J Vasc Interv Radiol.
and uterus: new biological insights. Hum Reprod 2002;13(12):1207–1210.
Update. 2013;19(1):67–83. [39] Shmakov RG, Vinitskiy AA, Chuprinin VD, et al.
[24] Urner F, Zimmermann R, Krafft A. Manual removal of Alternative approaches to surgical hemostasis in
the placenta after vaginal delivery: an unsolved prob- patients with morbidly adherent placenta undergoing
lem in obstetrics. J Preg. 2014;2014:274651. fertility-sparing surgery. J Matern Fetal Neonatal Med.
[25] Dannheim K, Shainker SA, Hecht JL. Hysterectomy for 2019;32(12):2042–2048.
placenta accreta; methods for gross and microscopic [40] Neumann PJ, Weinstein MC. The changing economics
pathology examination. Arch Gynecol Obstet. 2016; of Medical Technology Institute of Medicine.
293(5):951–958. Washington, DC: The National Academies Press,
[26] Panwar P, Lamour G, Mackenzie NC, et al. Changes in Institute of Medicine; 1991.
structural-mechanical properties and degradability of [41] Jauniaux E, Ayres-de-Campos D, Langhoff-Roos J,
collagen during aging-associated modifications. J Biol FIGO Placenta Accreta Diagnosis and Management
Chem. 2015;290(38):23291–23306. Expert Consensus Panel, et al. FIGO classification for
[27] Bowman ZS, Eller AG, Kennedy AM, et al. Accuracy of the clinical diagnosis of placenta accreta spectrum
ultrasound for the prediction of placenta accreta. Am disorders. Int J Gynecol Obstet. 2019;146(1):20–24.
J Obstet Gynecol. 2014;211(2):177.e1–177.e7. [42] Jauniaux E, Hussein AM, Zosmer N, et al. A new meth-
[28] Lim G, Horowitz JM, Berggruen S, et al. Correlation of odologic approach for clinico-pathologic correlations
probability scores of placenta accreta on magnetic in invasive placenta previa accreta. Am J Obstet
resonance imaging with hemorrhagic morbidity. Gynecol. 2019. DOI:10.1016/j.ajog.2019.11.1246
J Clin Anesth. 2016;34:261–269. [43] Marsoosi V, Ghotbizadeh F, Hashemi N, et al.
[29] Palacios-Jaraquemada JM, Karoshi M, Keith LG, et al. Development of a scoring system for prediction of
Uterovaginal blood supply: the S1 and S2 segmental placenta accreta and determine the accuracy of its
concepts and their clinical relevance. In: Sir results. J Matern Fetal Neonatal Med. 2018;17. DOI:10.
Sabaratnam Arulkumaran, Mahantesh Karoshi, Louis 1080/14767058.2018.1531119
G. Keith, Andre B. Lalonde and Christopher B-Lynch, [44] McCulloch P, Taylor I, Sasako M, et al. Randomised tri-
editors. A comprehensive textbook of postpartum als in surgery: problems and possible solutions. BMJ.
hemorrhage: an essential clinical reference for effect- 2002;324(7351):1448–1451.
ive management. 2nd ed. London: Sapiens Publishing [45] Martins C, Pinto B, Soares M, et al. Feminine identity:
Ltd; 2012. p. 19–23. the representation of the uterus for women under-
[30] Committee on Obstetric Practice. Committee Opinion going hysterectomy. R Pesq: Cuid Fundam Online.
no. 529: placenta accreta. Obstet Gynecol. 2012; 2013;5(4):574–582.
120(1):207–211. [46] Solbraekke KN, Bondevik H. Absent organs – present
[31] Vasquez DN, Das Neves AV, Golubicki JL, et al. Critical selves: exploring embodiment and gender identity in
care providers’ opinion on unsafe abortion in young Norwegian women’s accounts of hysterectomy.
Argentina. Int J Gynecol Obstet. 2012;116(3):249–252. Int J Qual Stud Health Well Being. 2015;10(1):26720.
[32] Rasch V. Unsafe abortion and postabortion care – an [47] Murad MH, Sultan S, Haffar S, et al. Methodological
overview. Acta Obstet Gynecol Scand. 2011;90(7): quality and synthesis of case series and case reports.
692–700. BMJ Evid Based Med. 2018;23(2):60–63.