Arch Gynecol Obstet (2011) 284:867873 DOI 10.

1007/s00404-010-1737-1

MATERNO-FETAL MEDICINE

Does cervical length and the lower placental edge thickness measurement correlates with clinical outcome in cases of complete placenta previa?
Moustafa M. Zaitoun Manal M. El Behery Azza A. Abd El Hameed Badeea S. Soliman

Received: 16 August 2010 / Accepted: 18 October 2010 / Published online: 27 November 2010 Springer-Verlag 2010

Abstract Objectives To evaluate the effectiveness of cervical length and the lower placental edge thickness measurement in predicting the risk of antepartum hemorrhage (APH) and emergency preterm cesarean delivery in women with complete placenta previa. Methods Fifty-four cases with conrmed diagnosis of complete placenta previa in third-trimester were subjected to transvaginal sonographic measurement of cervical length and lower placental edge thickness and correlated this to clinical outcome with regards to gestational age at delivery, ante partum hemorrhage, emergency cesarean section before 36 weeks due to massive hemorrhage and neonatal birth weight. Results Antepartum bleeding and emergency cesarean section rate before 36 weeks due to massive bleeding were signicantly higher in cases with thick lower placental edge or central placenta than cases with thin lower placental edge [16 cases (53.3%) vs. 5 cases (20.8%)] for the former and [14 cases (46.6%) vs. 4 cases (16.6%) for the later]. Antepartum bleeding was observed in 18 cases (51.4%) when cervical length measurements B30 mm of whom 16 cases (88.9%) had showed severe attack necessitated emergency cesarean delivery before 36 weeks versus 4 cases (21.1%) with cervical length C30 mm. By combining cervical length with lower placental edge thickness measurement sensitivity, specicity, positive predictive value (PPV) negative predictive value (NPV) and accuracy increased to 83.3, 78.4, 53.4, 79.8 and 89.7%, respectively for the prediction of antepartum bleeding and
M. M. Zaitoun M. M. El Behery (&) A. A. Abd El Hameed B. S. Soliman Zagazig University, Zagazig, Egypt e-mail: mbhry@hotmail.com

emergency cesarean section \36 weeks using receiver-operating characteristics curve with area under the curve 0.882. Conclusion Short cervical length at cut-off value B30 mm and increased lower placental edge thickness measurements may predict with high accuracy the risk of APH and emergency preterm cesarean delivery in patients with complete placenta previa. Keywords Cervical length Placental edge Placenta previa Preterm labor

Introduction Due to the widespread use of ultrasound and the rising cesarean section rates, the estimated prevalence of placenta previa rises between 0.2 and 0.3% of third-trimester pregnancies [1]. The clinical outcomes of placenta previa are highly variable with increased risk of antepartum hemorrhage (APH) and emergency preterm cesarean delivery, both are unpredictable. Although complete previa tends to be associated with earlier and more severe bleeding, a lesser degree of placenta previa may cause life-threatening hemorrhage necessitating blood transfusion and cesarean section [2, 3]. Thus, the degree of placenta previa is only one factor in prognosis and management. Transvaginal sonography (TVS) can provide detailed and accurate information about implantation site, placental migration, the lower placental edge and may help to predict delivery outcome [411]. Prediction of mode of delivery and the likelihood of APH is important in planning the management of placenta previa, including the method and time of intervention and the controversial area of outpatient management [12].

123

868

Arch Gynecol Obstet (2011) 284:867873

In a previous study on 104 women with conrmed diagnosis of placenta previa by TVS, women with placenta previa were at a relatively higher risk of developing complications if the lower placental edge was thick, and the author recommend integration of the shape of the lower placental edge into transvaginal sonographic assessment of placenta previa to improve the prediction of mode of delivery and clinical outcome [12]. Another prospective study conducted on 59 women with complete placenta previa in third-trimester of pregnancy suggested that shorter the cervix at the time of sonographic diagnosis, higher the risk of severe prematurity and emergency cesarean section due to massive maternal hemorrhage [13]. In a more recent prospective study on 68 women with conrmed diagnosis of placenta previa at delivery, a thirdtrimester cervical length of 30 mm or less was found to be associated with increased risk for hemorrhage, uterine activity, and preterm birth [14]. The objective of this study was to evaluate the effectiveness of cervical length and the lower placental edge thickness measurement in predicting the risk of APH and emergency preterm cesarean delivery in women with complete placenta previa.

Patients and methods This study was conducted in the High Risk Pregnancy Unit of Obstetrics and Gynecology Department of Zagazig University hospital, Egypt between January 2008 and June 2010. All asymptomatic pregnant women that have been referred for sonographic suspicion of placenta previa were invited to participate in this study, if they fulll the following inclusion criteria: (1) singleton pregnancy, (2) gestational age [28 weeks and less than 36 weeks as determined by a reliable date of last normal menstrual period and conrmed by early ultrasound exam before 20 weeks gestation, (3) diagnosis of complete placenta previa that was conrmed when the lower placental edge completely overlay the internal cervical os on TVS. Patients were excluded if they had (1) present or past history of bleeding in the current pregnancy, (2) multiple gestation, (3) previous or current history of threatened preterm labor, (4) history of cervical cerclage, ruptured membranes or evidence of polyhydramnios, (5) fetal growth restriction or fetal anomalies, (6) history of medical disorders complicating the pregnancy. Informed consent was taken from all participants and the study was approved by the local hospital ethics and research committee. All ultrasound examinations were performed using (GE healthcare, Voluson 730 pro V, Austria) medical system

equipped with a multifrequency 57.5 MHz transvaginal probe. Cervical length was measured at the time of diagnosis using TVS after bladder evacuation by the following technique. A sagittal plane was obtained to visualize the full length of the cervical canal and cervical length was measured by placing the callipers on the internal and external os. Three measurements were obtained and the shortest measurement was then considered. Visualization of the lower placental edge was done, while in the sagittal plane with the full length of the cervical canal and lower part of the uterus was in view. If the lower placental edge was not visualized, the transducer could be rotated 90 with the internal cervical os kept in view to detect the presence of placental tissue in the lower uterine cavity. The diagnosis of complete placenta previa was made when the internal cervical os was covered by placental tissue; it was dened as central if the placental edge was not visualized by TVS examination; it was dened as low-lying, when the lower edge of the placenta was within 3 cm from the internal cervical os as described previously [12]. The thickness of the lower placental edge was measured as the maximum thickness within a centimeter of the meeting point of the basal and chorionic plates and the estimated angle between these plates was measured. Thin placental edge was considered when the thickness was B1 cm and/or the angle was \45 (Fig. 1), otherwise it was referred to as thick edge (Fig. 2). Follow-up of all cases and a scheduled elective cesarean section was planned for all cases at completed 36 to 37 weeks of gestation following sonographic diagnosis of complete placental previa. Induction of fetal lung maturity was carried out by giving 12 mg dexamethasone IM daily for two doses. If vaginal bleeding occurred prior to the scheduled cesarean section, patients were admitted to the

Fig. 1 Thin lower placental edge

123

Arch Gynecol Obstet (2011) 284:867873

869 Table 1 Patients characteristics and pregnancy outcome in the study group Patient characteristics Maternal age (years) Parity N (%) Nulliparous Para1 or more Gestational age at diagnosis (weeks) Gestational age at delivery (weeks) Neonatal birth weight (g) Cervical length (mm) Nulliparous Para1 or more Thickness of lower placental edge (mm) 32 (59.3) 22 (40.7) 31.1 2.3 34. 2 2.4 2390.7 580.6 34.1 7.8 32.2 7.9 15.1 9.2 24 (44.4) 18 (40) 12 (22.6) 21 (38) 24 12.4 Mean SD 31.3 4.4

Fig. 2 Thick lower placental edge

N (%) of cases with thin placental edge N (%) of cases with central placental edge N (%) of cases with thick placental edge N (%) of cases with previous cesarean delivery Scan to delivery interval (days)

hospital and delivery was decided in accordance with the clinical condition of the patient. Delivery was performed at 36 completed weeks of gestation in cases of absent or mild vaginal bleeding. In cases of active ongoing bleeding that is clinically signicant, or in those with massive APH, an emergency cesarean section was done regardless of gestational age. The results of cervical length and lower placental edge thickness measurements were not disclosed to the obstetricians who perform the cesarean delivery. Previous uterine surgery, ultrasound scan to delivery interval, gestational age at delivery, neonatal weight, occurrence of antepartum bleeding, need for blood transfusion, cesarean hysterectomy and type of cesarean delivery (elective vs. emergency cesarean section due to massive hemorrhage) all were recorded. Mean cervical lengths and thickness of the lower placental edge were compared retrospectively in patients with and without antepartum bleeding, and in patients who had undergone elective cesarean section versus those who for whom an emergency cesarean section due to massive antepartum bleeding was performed.

section \36 weeks gestation. The areas under the ROC curves together with their standard error were evaluated. The best cut-off was evaluated in accordance with maximum likelihood ratio. The SPSS Version 13.0 (SPSS Inc., Chicago, IL, USA) statistical package was used to analyze data.

Results Sixty-seven cases with conrmed diagnosis of complete placenta previa on TVS were initially included in this study. Seven cases were excluded due to history of bleeding in the current pregnancy, two cases had fetal anomalies, four cases had threatened preterm labor. Thus, fty-four patients formed the nal study group. Patient characteristics and pregnancy outcome is shown in Table 1. 32 (59.3%) cases were primigravida, while 22 (40.7%) cases were para 1 or more. The mean SD gestational age at diagnosis was 31.1 2.3 weeks and the mean SD cervical length was 33.6 7.8 mm. No statistically signicant difference was noted between nulliparous and parous women regarding cervical length measurements (34.1 7.8 vs. 32.2 7.9 mm, respectively; P = 0.66). The mean gestational age at cesarean section was 34.2 2.4 weeks. The mean neonatal birth weight was 2390.7 580.6 g. Regarding lower placental edge thickness measurements 24 (44.4%) cases had thin edge, 18 (40%) had central edge and 12 (22.6%) cases had a thick lower placental edge.

Statistical analysis Data are reported as means and SDs. Continuous variables (cervical lengths) were compared using Students t test. A two-tailed value of P \ 0.05 was considered statistically signicant. A chi-squared test and Fishers exact test were used to assess the statistical signicance when the incidences in thin-edge and thick or central placental edge groups were compared. Receiver-operating characteristics (ROCs) curves was constructed to determine the accuracy of cervical length and combined cervical length and lower placental edge thickness to identify women at high risk for antepartum vaginal bleeding and emergency cesarean

123

870

Arch Gynecol Obstet (2011) 284:867873

Clinical outcome in women with a thin versus thick or central lower placental edge was shown in Table 2. No signicant difference was noted between both groups regarding incidence of previous cesarean delivery, the mean gestational age at delivery and mean neonatal birth weight were signicantly lower in patients with central or thick lower placenta edge than patients with thin lower placenta edge (32.8 3.4 vs. 36.1 4.3 weeks) for the former and (1965.5 505.3 vs. 2835.4 860.4 g) for the later. Scan to delivery interval was signicantly shorter with thick or central versus thin placental edge (26 12.7 vs. 19.4 14.4 days, respectively; P = 0.02). Antepartum bleeding and emergency cesarean section rate before 36 weeks due to massive bleeding were also signicantly higher in cases with central or thick lower placental edge than cases with thin lower placental edge, 16 cases (53.3%) versus 5 cases (20.8%) for the former and 14 cases (46.6%) versus 4 cases (16.6%) for the later. The incidence of complications in cesarean hysterectomy was higher in thick or central lower placental edge group 2 cases (6.6%) compared to none (0%) in cases with thin placental edge but the difference was not signicant. Also, the need for peripartum blood transfusion was signicantly higher in cases with central or thick placental edge. Based on ROC curve at a cut-off point B30 mm for cervical length the sensitivity, specicity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were 71.3, 70.6, 51.2, 64.5, 77.4%, respectively for predicting cases at high risk for emergency cesarean section \36 weeks with area under the curve (AUC) 0.681. By combining cervical length with lower placental edge thickness measurement sensitivity, specicity, PPV, NPV, and accuracy increased to 83.3, 78.4, 53.4, 79.8, and 89.7%, respectively for the prediction of antepartum bleeding and emergency cesarean section \36 weeks with AUC 0.882 (Fig. 3).

Table 3 shows clinical outcome in accordance with cervical length measurements at cut-off value 30 mm in women with conrmed diagnosis of complete placenta previa. No signicant difference was noted between both groups regarding incidence of previous cesarean delivery. The mean gestational age at delivery and mean neonatal birth weight were signicantly lower in patients with cervical length measurements B30 mm than patients with cervical length measurements [30 mm (33 2.3 vs. 35.7 5.2 weeks) for the former and (1.9 0.50 vs. 2.8 0.86 kg) for the later. Scan to delivery interval was signicantly shorter when cervical length was \30 mm than when it exceeds 30 mm in length (28 12.7 vs. 15.4 14.4 days; P = 0.008).

Fig. 3 Receiver operating characteristic curve showing the predictive performance of cervical length and lower placental edge thickness measurements for the occurrence of antepartum bleeding and emergency CS before 36 weeks gestation in women with complete placenta previa

Table 2 Clinical outcome in women with thin versus thick lower placental edge Parameter Previous cesarean delivery, N (%) Gestational age at delivery (weeks) APH, N (%) Scan to delivery interval (days), mean SD Emergency CS before 36 weeks, N (%) Elective CS C36 weeks, N (%) Cesarean hysterectomy, N (%) Peripartum blood transfusion (units), mean SD Mean birth weight (g), mean SD * Signicant P \ 0.05, ** highly signicant P \ 0.001 Thin-edge placenta (N = 24) 10 (47.6) 36.1 4.3 6 (25) 26 12.7 4 (16.6) 20 (83.4) 0 (0) 1.03 0.8 2835.4 860.4 Central or thick-edge placenta (N = 30) 11 (52.4) 32.8 3.4 16 (53.3) 719.4 14.4 14 (46.6) 16 (53.4) 2 (6.6) 3.1 1.4 1965.5 505.3 P (two-tailed test) 0.41 0.33 0.07 0.02* 0.06 0.11 0.07 0.08 0.003**

123

Arch Gynecol Obstet (2011) 284:867873 Table 3 Clinical outcome according to cervical length measurements in women with conrmed diagnosis of complete placenta previa Outcome Cervical length [30 mm (N = 19) Previous cesarean delivery N (%) Gestational age at delivery (weeks), mean SD APH, N (%) Scan to delivery interval (days) Emergency CS before 36 weeks, N (%) Elective CS C36 weeks Cesarean hysterectomy, N (%) Peripartum blood transfusion (units), mean SD Mean birth weight (kg), mean SD * Signicant P \ 0.05, ** highly signicant P \ 0.001 12 (57.2) 35.7 5.2 4 (21.1) 28 12.7 2 (10.5) 17 (89.5) 0 (0.0) 1.1 1.03 2.8 0.86 \30 mm (N = 35) 9 (42.8) 33 3.3 18 (51.4) 15.4 14.4 16 (45.7) 19 (54.3) 1 (2.8) 1.6 1.02 1.9 0.50 0.14 0.007** 0.004* 0.008** 0.002** 0.003** 0.03 0.14 0.003**

871

P (two-tailed test)

Antepartum bleeding was observed in 18 cases (51.4%) when cervical length measurements B30 mm of whom 16 cases (88.9%) had showed severe attack necessitated emergency cesarean delivery before 36 weeks and these gures were signicantly higher as compared to cases with cervical length measurements [30 mm in 4 cases (21.4%) P = 0.004. The need for peripartum blood transfusion was not signicantly different between patients with cervical length measurements B30 mm and patients with cervical length measurements [30 mm (P = 0.14), also Cesarean hysterectomy was done in one case with cervical length measurements B30 mm compared to none in cases with cervical length measurements [30 mm.

Discussion Women with placenta previa are at increased risk of prematurity and severe antepartum bleeding requiring an emergency cesarean delivery that are difcult to be predicted in the preclinical stage. Transvaginal sonography is the gold standard in diagnosis of placenta previa. Many authors had studied its role in assessment of distance of lower placental edge from the internal os and correlated this to the risk of APH, mode of delivery, emergency cesarean delivery and risk of prematurity [811, 15], but few studies added the value of lower placental edge thickness [12], and cervical length measurement and correlated this to clinical outcome in patients with placenta previa [13, 14]. In this study, the risk of antepartum bleeding among cases of complete placenta previa diagnosed by TVS in third-trimester was 45.7% and risk of emergency cesarean section before 36 weeks was 37.3%, there was no statistically signicant difference between nulliparous and parous

women in this aspect and also as regard to cervical length measurements. In a previous prospective study [12] on 104 patients diagnosed with placenta previa during third-trimester, the combined use of the shape of the lower placental edge with other sonographic ndings on TVS could improve the accuracy in predicting clinical outcome in placenta previa, and the author found that a thick lower placental edge could predict a particularly high risk group irrespective of degree and location of placenta. This agree with our results as we found that the risk of APH, emergency cesarean delivery and low neonatal birth weight is signicantly higher in cases with increased lower placental edge thickness, this might be explained by the abundant vasculature of the lower placental edge and the subplacental zone as suggested by Saitoh et al. [16] who prospectively examined 35 women with placenta previa using TVS at or beyond 28 weeks gestation with follow-up scans at 5- to 7-day intervals until cesarean section. The patients were classied into three groups based on placental edge sonographic features in relation to the internal cervical os. The authors concluded that echo free space in the lower edge of the placenta overlying the cervix is considered as a risk for occurence of sudden massive antepartum bleeding. There is a clear relationship between ultrasonographic cervical length and preterm birth, particularly among women with a prior preterm birth. A cut-off of 30 mm has been associated with a relative risk for preterm birth of 3.8 [17]. However, data are limited regarding cervical length in the setting of placenta previa. Stafford et al. [14] in their prospective study on 68 women with conrmed placenta previa at delivery found that women with a short cervix were more likely to require delivery for hemorrhage, and to deliver preterm than those with a normal cervical length.

123

872

Arch Gynecol Obstet (2011) 284:867873

Furthermore, evidence of regular uterine contractions was more common with a short cervix than with a longer cervix. Thus, they conclude that a third-trimester sonographic cervical length of 30 mm or less is associated with increased risk for hemorrhage, uterine activity, and preterm birth in pregnancies with placenta previa. In our study using ROC curve if cervical length B30 mm was taken as a cut-off, the risk of massive maternal hemorrhage and emergency cesarean section before 36 weeks was signicantly higher as compared to cases with cervical length above this threshold running in agreement with Stafford et al. [14]. When we combined cervical length with lower placental edge thickness measurements the sensitivity, specicity, PPV and NPV increased to 83.3, 78.4, 53.4, 79.8%, respectively for the prediction of antepartum bleeding and emergency cesarean section \36 weeks with AUC 0.882. In the study of Ghi et al. [13] they use cut-off value for cervical length measurement B31 mm and stated that below this value the risk of preterm cesarean section was almost one in two, whereas due to excellent NPV of the ultrasound examination whenever the cervical length was above this threshold, the need for preterm cesarean section was rare, also they showed that cervical shortening even to a minor degree might predict the risk of emergency preterm cesarean section due to earlier placental detachment with massive hemorrhage. The authors explained this increased risk by the increased chance of spontaneous preterm labor in women with a sonographically short cervix in thirdtrimester as reported previously [18, 19]. Placental detachment could occur with the premature onset of labor because of the inelastic nature of the placenta and its failure to adapt with the progressive cervical effacement [7]. In agreement with Ghi et al. [13] our nding showed that a short cervical length at a cut-off 30 mm there is a denite risk of emergency preterm cesarean delivery for ongoing active hemorrhage that is clinically relevant. However, in their study no signicant differences were noted between cases with and those without APH regarding cervical length measurements, in contrast our study showed a signicantly higher incidence of antepartum bleeding in cases with cervical length B30 versus those with cervical length above 30 mm [18 cases (51.4%) vs. 4 (21.4%); P = 0.004, respectively]. Among those presenting with antepartum bleeding 16 cases (88.9%), the attack was severe enough to require an emergency cesarean delivery before 36 weeks. This disagreement could be explained by a better predictive accuracy when we combined both cervical length with placental edge thickness using ROC curve for predicting the risk of APH in cases of complete placenta previa diagnosed at third-trimester.

Conclusion Short cervical length at cut-off value B30 mm and increased lower placental edge thickness measurements may predict with high accuracy the risk of APH and emergency preterm cesarean delivery in patients with complete placenta previa. We recommend transvaginal measurement of cervical length and lower placental edge thickness to be a part of the routine third-trimester scan in women with complete placenta previa to predict the risk of severe complications and to individualize patient management.
Conict of interest None.

References
1. Rosenberg T, Pariente G, Sergienko R, Wiznitzer A, Sheiner E (2010) Critical analysis of risk factors and outcome of placenta previa. Arch Gynecol Obstet [epub ahead of print] 2. Tuzovic L (2006) Complete versus incomplete placenta previa and obstetric outcome. Int J Gynaecol Obstet 93:110117 3. Green JR (1989) Placenta previa and abruptio placenta. In: Creasy RK, Resnik R (eds) Maternal fetal medicine: principles and practice. WB Saunders, Philadelphia, pp 592 4. Leerentveld RA, Gilber EC, Arnold MJ (1990) Accuracy and safety of transvaginal sonographic placental localization. Obstet Gynecol 76 (5 Pt 1):759762 5. Lodhi SK, Khanum Z, Watoo TH (2004) Placenta previa: the role of ultrasound in assessment during third trimester. J Pak Med Assoc 54:8183 6. Oppenheimer LW, Farine D (2009) A new classication of placenta previa: measuring progress in obstetrics. Am J Obstet Gynecol 201:227229 7. Oppenheimer L, Holmes P, Simpson N, Dabrowski A (2001) Diagnosis of low-lying placenta: can migration in the third trimester predict outcome? Ultrasound Obstet Gynecol 18:100102 8. Ghourab S, Al-Jabari A (2000) Placental migration and mode of delivery in placenta previa: transvaginal sonographic assessment during the third trimester. Ann Saudi Med 20(56):382385 9. Bhide A, Prefumo F, Moore J, Hollis B, Thilaganathan B (2003) Placenta edge to internal os distance in the late third trimester and mode of delivery in placenta previa. BJOG 110:860864 10. Vergani P, Ornaghi S, Pozzi I, Beretta P, Russo FM, Follesa I, Ghidini A (2009) Placenta previa: distance to internal os and mode of delivery. Am J Obstet Gynecol 201:266 11. Bronsteen R, Valice R, Lee W, Blackwell S, Balasubramaniam M, Comstock C (2009) Effect of a low-lying placenta on delivery outcome. Ultrasound Obstet Gynecol 33:204208 12. Ghourab S (2001) Third-trimester transvaginal ultrasonography in placenta previa: does the shape of the lower placental edge predict clinical outcome? Ultrasound Obstet Gynecol 18:103108 13. Ghi T, Contro E, Martina T, Piva M, Morandi R, Orsini LF et al (2009) Cervical length and risk of antepartum bleeding in women with complete placenta previa. Ultrasound Obstet Gynecol 33(2):209212 14. Stafford IA, Dashe JS, Shivvers SA, Alexander JM, McIntire DD, Leveno KJ (2010) Ultrasonographic cervical length and risk of hemorrhage in pregnancies with placenta previa. Obstet Gynecol 116:595600

123

Arch Gynecol Obstet (2011) 284:867873 15. Oyelese Y, Smulian JC (2006) Placenta previa, placenta accrete and vasa previa. Obstet Gynecol 107(4):927941 16. Saitoh M, Ishihara K, Sekiya T, Araki T (2002) Anticipation of uterine bleeding in placenta previa based on vaginal sonographic evaluation. Gynecol Obstet Invest 54:3742 17. Hibbard JU, Tart M, Moawad AH (2000) Cervical length at 1622 weeks gestation and risk for preterm delivery. Obstet Gynecol 96:972978

873 18. Rozenberg P, Gillet A, Ville Y (2002) Transvaginal sonographic examination of the cervix in asymptomatic pregnant women: review of the literature. Ultrasound Obstet Gynecol 19:302311 nchez M, Crispi F, Go mez O, 19. Palacio M, Sanin-Blair J, Sa s E (2007) The use of a Carreras E, Coll O, Cararach V, Grataco variable cut-off value of cervical length in women admitted for preterm labor before and after 32 weeks. Ultrasound Obstet Gynecol 29:421426

123

Copyright of Archives of Gynecology & Obstetrics is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.