Placenta accreta is diagnosed when the placenta adheres abnormally to the myometrium. Defined pathologically as the placental basal plate adhering to myometrial fibers without intervening decidua, this phenomenon is clinically identified in approximately one to two of 1,000 births in academic centers.1,2 Placenta accreta is associated with significant morbidity and is often managed by hysterectomy. However, in women desiring future fertility, uterine conservation can be considered and is sometimes feasible.3–5 Furthermore, accreta is sometimes diagnosed only pathologically in the absence of clinical identification or hemorrhage.6,7
Population-based and histologic studies have shown that a history of placenta accreta increases the risk for hemorrhage in a subsequent delivery.6,8 Nevertheless, the majority of patients with prior accreta will not experience a recurrence with the next pregnancy.9 Given the severe morbidity associated with accreta, patients and health care providers would benefit from knowing an individual's risk for recurrent abnormal placentation and postpartum hemorrhage. We therefore sought to study postpartum morbidity outcomes in patients with a history of a pathologically diagnosed placenta accreta.
MATERIALS AND METHODS
This retrospective cohort study examined women who delivered at Brigham and Women's Hospital and who had a pathologically diagnosed placenta accreta between January 1, 2007, and May 31, 2015. This study was approved by our institutional review board and deemed exempt from requiring individual patient consent.
To identify patients for inclusion, we used a word search of all Brigham and Women's Hospital pathology reports in the Partner's Healthcare electronic medical record database for delivering patients with the term “creta,” defined as placenta basalis abutting myometrium with no intervening decidua. The initial delivery with pathologic placenta accreta was defined as the index pregnancy. Once these patients were identified, we included those who returned to Brigham and Women's Hospital for a subsequent delivery. Patient demographics and clinical data were obtained by searching operative notes, delivery notes, pathology reports, and inpatient and outpatient clinical records. Clinically adherent placenta was described by the health care provider in the delivery or operative note as abnormally adherent or unable to be removed manually or with suction curettage.
Placentas are not routinely sent to the pathology department at our institution. Ultimately, placental examination is at the discretion of the health care provider, although it is advised in the following conditions: diabetes, hypertension, substance abuse, prematurity, intrapartum fever or infection, suspected abnormal placentation, significant vaginal bleeding, oligo- or polyhydramnios, multiple gestation, and fetal or neonatal abnormalities. The indication for initial placental analysis, as listed on the pathology report or stated in the patient record, was noted for all patients.
We used hemorrhagic and surgical morbidity in the index pregnancy as our major predictor variable, and similar morbidity in the subsequent pregnancy was used as our primary outcome. Morbidity at either delivery was categorized as minor or major. Minor morbidity included estimated blood loss of greater than 500–1,500 cc at vaginal delivery or greater than 1,000–1,500 cc at cesarean delivery; red blood cell transfusion of one to three units; sharp resection of a focally adherent placental portion at cesarean delivery or through a delayed curettage or hysteroscopic procedure; arterial ligations or oversewing of the placental bed at cesarean delivery; or a minor surgical procedure to control bleeding such as balloon tamponade or sharp or suction uterine curettage. Major morbidity included estimated blood loss greater than 1,500 cc, red blood cell transfusion greater than or equal to four units, uterine artery embolization, unplanned laparotomy, or hysterectomy.10,11 Although not all patients had a reported postoperative hematocrit, those with a drop of 10 or more points, but no other criteria for major morbidity, were classified as “minor” morbidity. Patients who were not associated with any of the aforementioned morbidities were classified as nonmorbid or incidental.
Our obstetric protocol indicates transfusion for symptomatic hypovolemia, ongoing active hemorrhage, or to maintain a hematocrit of at least 21%. The ultimate decision to administer blood products in this study was left to the attending obstetric and anesthesia providers. Estimated blood loss was assessed by the delivering providers and transcribed from delivery notes.
Statistical analysis was performed using SAS 9.3. Univariate comparison of morbidity outcomes was performed utilizing Fisher exact test for categorical variables and Wilcoxon analysis for continuous variables. A P value <.05 with two-tailed tests was considered statistically significant.
Figure 1 displays the acquisition of our comparison groups. There were 339 patients with pathologically diagnosed placenta accreta who did not undergo hysterectomy during the study period. Of these, 39 women (11.5%) returned for a subsequent delivery at our institution and were included in the study. None of the returning patients was diagnosed with increta or percreta in the index pregnancy. Three patients had more than one subsequent pregnancy, with only the first considered for this analysis. Just under half of the patients had no morbidity in the index pregnancy (n=18 [46%]) and thus had an incidentally identified placenta accreta. The other 54% (n=21) experienced any index morbidity, 76% minor and 24% major.
The indications for pathologic placental evaluation in the index pregnancy stratified by type of morbidity are listed in Table 1. Significantly more patients with a morbid index pregnancy outcome (18/21) had the placenta analyzed as a result of hemorrhage or a placental abnormality compared with 6 of 18 with an incidental accreta (86% compared with 33%, respectively, P<.01). Although by definition the incidental accreta group included no postpartum hemorrhages, these six patients had their placentas evaluated for resolved previa, manual removal, or adherent placenta without additional sequelae.
Index pregnancy characteristics are listed in Table 2. Demographic characteristics at the time of index delivery including maternal age, parity, prior cesarean delivery, mode of delivery, in vitro fertilization, prior uterine surgery, prior dilation and curettage, and multiples did not differ by morbidity outcome. Overall, 14 of the 39 patients (36%) had a clinically adherent placenta. When compared with those with no morbidity, patients with any morbidity in the index pregnancy did not differ with regard to rates of clinical adherence (22% compared with 48%, respectively, P=.18) but had significantly higher rates of manual placental extraction (42% compared with 87%, P=.04). No patient had a persistent placenta previa at either the index or subsequent delivery.
To evaluate subsequent pregnancy morbidity with a prior diagnosis of histologic placenta accreta, we compared patients according to morbidity of the index pregnancy (Table 3). Cesarean delivery rates were identical for those with and without index morbidity. No patient with an incidental placenta accreta in the index pregnancy experienced major morbidity in the subsequent pregnancy. Of the 21 patients with any morbidity in the index pregnancy, six (29%) had major morbidity in the subsequent pregnancy compared with zero patients who had no index morbidity, and this was statistically significant (P=.02) (Table 3). There were no placenta accretas, pathologic or clinical, diagnosed in the subsequent pregnancy for those with an incidental index accreta compared with a 24% (n=5) accreta rate in those with a morbid index delivery (P=.05). Only two of the five recurrent accretas were detected ultrasonographically.
Of note, our cohort included 25 patients who obtained an index pathologic accreta diagnosis but did not have a clinical diagnosis of an adherent placenta. Of these pathology-only accretas, 11 experienced morbidity in the index pregnancy, nine minor and two major. Three of the 25 (12%) experienced major morbidity in the subsequent pregnancy, all of whom had morbidity in the index pregnancy (two minor and one major). Viewed retrospectively, of the six patients with subsequent major morbidity, half had a pathology-only diagnosis with no adherence recognized in the index pregnancy.
The incidence of major postpartum morbidity after a prior pathologically diagnosed placenta accreta was significantly associated with the index pregnancy's clinical outcome. We focused on placental and hemorrhagic morbidity that we categorized as mild (minimally invasive procedure or mild hemorrhage) or major (need for invasive procedure or major hemorrhage). Our study implies that patients with a nonmorbid or incidental accreta are at low risk for major morbidity in a subsequent pregnancy. Alternatively, those with any clinical morbidity should be prepared for a possible major hemorrhage in the next pregnancy, which occurred in almost 30% in our series. Index morbidity appears to be more important than clinical accreta recognition, because half of those with major morbidity in the subsequent pregnancy had an index accreta that was not recognized as adherent.
A history of placenta accreta in a prior pregnancy was previously shown to be an independent risk factor for postpartum hemorrhage in a future pregnancy (2.1% compared with 0.6%) in a population-based study performed by Vinograd et al.12 This study used a broader group of index patients with placenta accreta, including both clinical and pathologic diagnoses. The current study has the additional advantages of stratifying patients with accreta according to prior clinical outcome and elucidating the role of incidentally identified accretas with no clinical morbidity.
A subset of placentas contains myometrial fibers adherent to the placental basal plate, often without clinical evidence of adherence or hemorrhage. These placentas have been previously termed “minimal lesions”13 or basal plate myometrial fibers.7 The incidence may range from 10% to 40% among placentas sent for pathologic analysis,6–8,13 calling into question this finding's clinical significance. Prior case–control studies have found that histologically adherent myometrial fibers were associated with accreta in a subsequent pregnancy, although this was limited to the subset of index patients with minimal or no decidua separating the myometrium from the basal plate.6,7
Although both of these studies associated a prior histologic accreta with subsequent abnormal placentation, they have more limited utility in counseling an individual patient for a subsequent delivery. The case–control design cannot provide subsequent morbidity rates nor did they identify which patients will be at highest risk. As a cohort study, the current analysis provides data that can be directly applied to individual patients based on the clinical context of the index accreta. Furthermore, we restricted our cohort to true accretas with absent intervening decidua to minimize the effect of false-positive diagnoses.
Limitations of our study include the small sample size. We had 339 diagnosed placenta accretas during the study period, and only 11.5% returned for a subsequent pregnancy. We are unable to account for patients who moved away or transferred care between deliveries. We do not believe that this would systematically vary across comparison groups, because our hospital delivers a large proportion of low-risk as well as high-risk patients. This low return rate may reflect selection patterns with patients avoiding a subsequent pregnancy after prior major morbidity. Although it should not invalidate our study's observations, it may affect its generalizability to those with highly morbid index deliveries. Furthermore, only placentas deemed by an individual health care provider to necessitate pathologic review were available, limiting the ability to generalize these results to a broader population. Another drawback is the retrospective nature of the study and reliance on past health care provider documentation. Despite these limitations, our study provides a useful and needed framework for patient counseling in this clinical situation, supporting the need for prospective research and systematic data collection to validate these findings.
Placenta accreta is a leading cause of severe maternal morbidity and mortality.14 Accordingly, patients and practitioners alike will be concerned when this diagnosis appears on a pathology report, particularly when unexpected. Our findings suggest that risk for major morbidity after a prior pathologically diagnosed placenta accreta depends on the clinical context. As shown in prior studies, a history of a postpartum hemorrhage is an important risk factor for hemorrhagic morbidity in a subsequent pregnancy.15,16 Additionally, we have observed that surgical management of a prior focal accreta warrants future caution.
With a 29% rate of subsequent major morbidity, preparation for hemorrhagic morbidity is indicated when a pathologically proven accreta was accompanied by postpartum hemorrhage or a surgical procedure. Alternatively, patients with nonmorbid accreta may be at lower risk for major hemorrhagic morbidity in a subsequent pregnancy, although careful assessment and management of these patients are warranted. Further trials are needed to study the outcomes of subsequent pregnancies to counsel patients who had a conservatively managed accreta.
1. Bailit JL, Grobman W, Rice MM, Reddy UM, Wapner RJ, Varner MW, et al. Morbidly adherent placenta treatments and outcomes. Obstet Gynecol 2015;125:683–9.
2. Mehrabadi A, Hutcheon JA, Liu S, Bartholomew S, Kramer MS, Liston RM, et al. Contribution of placenta accrete to the incidence of postpartum hemorrhage and severe postpartum hemorrhage. Obstet Gynecol 2015;125:814–21.
3. Fitzpatrick KE, Sellers S, Spark P, Kurinczuk JJ, Brocklehurst P, Knight M. The management and outcomes of placenta accreta, increta, and percreta in the UK: a population-based descriptive study. BJOG 2014;121:62–70.
4. Legendre G, Zoulovits FJ, Kinn J, Senthiles L, Fernandez H. Conservative management of placenta accreta: hysteroscopic resection of retained tissues. J Minim Invasive Gynecol 2014;21:910–3.
5. Perez-Delboy A, Wright JD. Surgical management of placenta accreta: to leave or remove the placenta? BJOG 2014;121:163–9.
6. Miller ES, Linn RL, Ernst LM. Does the presence of placental basal plate myometrial fibres increase the risk of subsequent morbidly adherent placenta: a case-control study. BJOG 2015 Aug 11 [Epub ahead of print].
7. Linn RL, Miller ES, Lim G, Ernst LM. Adherent basal plate myometrial fibers in the delivered placenta as a risk factor for development of subsequent placenta accreta. Placenta 2015;36:1419–24.
8. Khong TY, Werger AC. Myometrial fibers in the placental basal plate can confirm but do not necessarily indicate clinical placenta accreta. Am J Clin Pathol 2001;116:703–8.
9. Sentilhes L, Kayem G, Ambroselli C, Provansal M, Fernandez H, Perrotin F, et al. Fertility and pregnancy outcomes following conservative treatment for placenta accreta. Hum Reprod 2010;25:2803–10.
10. Shields LE, Wiesner S, Fulton J, Pelletreau B. Comprehensive maternal hemorrhage protocols reduce the use of blood products and improve patient safety. Am J Obstet Gynecol 2015;212:272–80.
11. Severe maternal morbidity: screening and review. Obstetric Care Consensus No. 5. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;128:e54–60.
12. Vinograd A, Wainstock T, Mazor M, Mastrolia SA, Beer-Weisel R, Klaitman V, et al. A prior placenta accreta is an independent risk factor for post-partum hemorrhage in subsequent gestations. Eur J Obstet Gynecol Reprod Biol 2015;187:20–4.
13. Sherer DM, Salafia CM, Minior VK, Sanders M, Ernst L, Vintzileos AM. Placental basal plate myometrial fibers: clinical correlations of abnormally deep trophoblast invasion. Obstet Gynecol 1996;87:444–9.
14. Publications Committee, Society for Maternal-Fetal Medicine, Belfort MA. Placenta accreta. Am J Obstet Gynecol 2010;203:430–9.
15. Combs CA, Murphy EL, Laros RK Jr. Factors associated with postpartum hemorrhage with vaginal birth. Obstet Gynecol 1991;77:69–76.
16. Kominiarek MA, Kilpatrick SJ. Postpartum hemorrhage: a recurring pregnancy complication. Semin Perinatol 2007;31:159–66.