Volume 201, Issue 3 , Pages 266.e1-266.e5, September 2009
Placenta previa: distance to internal os and mode of delivery
Article Outline
Objective
The purpose of this study was to relate the mode of delivery and outcomes in a cohort of cases of placenta previa that had the last transvaginal ultrasonographic scan <28 days before delivery.
Study Design
Cases in which the placental edge overlapped the internal cervical (n = 42) underwent cesarean section delivery. Labor was allowed in those with placental edge to internal os distance of 1-10 mm (group 1, 24 women) and those with a distance of 11-20 mm (group 2, 29 women).
Results
Rates of cesarean section delivery (75% vs 31%; odds ratio, 6.7; 95% confidence interval [CI], 2–22) and of bleeding before labor (29% vs 3%; odds ratio, 11.5; 95% CI, 1.6–76.7) were higher in group 1 than in group 2. Blood loss at delivery (662 ± 466 mL vs 510 ± 547 mL) and rate of severe postpartum hemorrhage (21% vs 10%; odds ratio, 2.3; 95% CI, 0.5–9.7) were similar in the 2 groups.
Conclusion
More than two-thirds of women with a placental edge to cervical os distance of >10 mm deliver vaginally without increased risk of hemorrhage.
Key words: cesarean section delivery, placenta previa, transvaginal sonography
Transvaginal ultrasonography has become the technique of choice to evaluate cases with suspected placenta previa. During the third trimester, measurement of the distance from the placental edge to the internal cervical os is used commonly to gauge the likelihood of the need for cesarean delivery. There is consensus that a placenta previa that totally or partially overlies the internal os requires delivery by cesarean section. More controversial is the optimal mode of delivery when the placenta lies in proximity of the internal os.
For Editors' Commentary, see Table of Contents
See related editorial, page 227
Three studies concluded that a placental edge to cervical os distance of >20 mm permits a safe vaginal delivery.1, 2, 3 However, the optimal mode of delivery for cases with a distance from placental edge to internal os between 1 and 20 mm is unclear presently because of the paucity of data. Despite this uncertainty, the Royal College of Obstetricians and Gynaecologists issued a category B recommendation that “a placental edge less than 20 mm from the internal os is likely to need delivery by caesarean section,”4 which has been echoed by the Society of Obstetricians and Gynecologists of Canada: “A distance of 20 to 0 mm away from the os is associated with a higher cesarean section delivery rate although vaginal delivery is still possible depending on clinical circumstances.”5 Although the American College of Obstetricians and Gynecologists has not pronounced itself on the topic, a recent Clinical Expert Series report concluded that “(women with placenta previa) with a placental edge to internal os distance of less than 2 cm should be delivered by cesarean.”6
However, the rate of cesarean section delivery among cases with a distance between 1 and 20 mm ranges from 40-90%.1, 2, 3, 7 A recent review on the subject recommended further studies on the topic.5 The aims of our study were to relate transvaginal ultrasound findings with the mode of delivery and outcomes in a large series of cases with placenta previa distance between 1 and 20 mm from the cervical os and to examine the rates of cesarean section delivery at different cutoff distances within this group.
Materials and Methods
This is a retrospective analysis of a cohort of singleton pregnancies with known placenta previa who delivered at our hospital between January 2003 and August 2008. All women who were registered for antenatal care at our clinic underwent an ultrasound scan in the early second trimester for anatomy survey. Those women with a low-lying placenta at transabdominal scan underwent a transvaginal ultrasound examination; cases with a diagnosis of placenta previa were followed with serial transvaginal examinations until delivery. In addition, all women with antepartum hemorrhage had a transvaginal scan, and women with placenta previa were followed in a similar way. All cases that were followed because of placenta previa were recorded in a dedicated log book. The diagnosis of placenta previa was made when the lower edge of placental tissue was within 20 mm of the internal cervical os or it overlapped it at transvaginal ultrasound examination. A placenta-to-cervical os distance of >20 mm was classified as low-lying. The exact distance between the center of the internal cervical os and the leading edge of the placenta was measured by transvaginal sonography after voiding. Placenta accreta was diagnosed at ultrasonography in the presence of loss of interface between myometrium and chorion and intraplacental lacunae with blood flow at color flow mapping. Such cases underwent confirmatory studies with magnetic resonance imaging and were scheduled for cesarean hysterectomy at 34-35 weeks of gestation.
A consistent protocol was implemented on all women with a diagnosis of placenta previa. Scans were repeated monthly during pregnancy to assess the extent of placental migration. In the presence of bleeding, patients were hospitalized, and corticosteroids were administered for fetal lung maturity enhancement if the pregnancy was <34 weeks of gestation. Cases in which the placental edge reached or overlapped the internal cervical os underwent prelabor cesarean section delivery at ≤37-38 weeks of gestation in cases of intractable hemorrhage. Cases in which the placental edge was between 1 and 20 mm of the internal cervical, the fetus was in cephalic presentation position, and there were no contraindications for vaginal delivery were allowed to labor after evaluation of their clinical situation. In this group, nulliparous women with a history of cesarean section delivery were not admitted to the trial of labor. Excluded from the present analysis were cases in which the distance between placental edge and cervical os became >20 mm during follow-up monitoring and women who underwent the last sonographic scan >28 days before delivery.
Information about the woman, course of pregnancy, mode of delivery with indications for all cesarean section, and blood loss at delivery was registered prospectively in a dedicated log book, which was audited periodically, in the labor and delivery theater. Blood loss at delivery was quantified with the use of graduated collection bags at vaginal delivery and of suction bottles and weighted surgical pads at cesarean delivery. Postpartum hemorrhage was defined as blood loss of >1000 mL for cesarean deliveries and >500 mL for vaginal deliveries. The study was approved on April 13, 2006 (protocol no. 236) by the institutional review board.
Statistical analysis
The likelihood of vaginal delivery and other obstetric variables was compared between women with a distance of 1-10 mm vs 11-20 mm with the use of χ2 test and Fisher exact test for categoric variables and 1-way analysis of variance for continuous variables (SPSS software, version 15; SPSS Inc, Chicago, IL).
Results
A total of 14,973 women delivered during the study period, 120 of whom had a diagnosis of placenta previa during pregnancy (8‰). There were no cases with prenatal diagnosis of placenta previa who were lost to follow-up evaluation. Sixteen cases were excluded from analysis because the placental edge to cervical os distance became >20 mm during follow-up scans; cesarean delivery occurred in 19% of them (3/16 women). Nine cases were excluded, because the last transvaginal sonographic examination was performed >28 days before delivery. The mean scan-to-delivery interval among the remaining patients was 10.0 ± 7.1 days.
Table 1 shows a comparison of the cases that is based on whether the placenta crossed the internal cervical os or not. A cesarean section delivery was performed in all 42 cases with the placental edge crossing the internal cervical os. In this group, there were 6 cases of cesarean hysterectomy: 5 cases because of placenta accreta and 1 case because of uterine atony. All 5 cases of placenta accreta were diagnosed during pregnancy by ultrasonography and confirmed by magnetic resonance imaging; 4 cases had an anterior placenta and a history of cesarean section delivery, whereas the remaining case had 2 previous vaginal deliveries and a curettage as only risk factor. The case of cesarean hysterectomy because of postpartum atony occurred in a 38-year-old woman (4 pregnancies, 3 viable offspring) with a posterior placenta previa that was diagnosed at 19.5 weeks of gestation that did not migrate at follow-up scans. She never experienced bleeding episodes during pregnancy. A prelabor cesarean section delivery was performed at 37.0 weeks of gestation for fetal growth restriction; she subsequently experienced uterine atony with a blood loss of 2000 mL, which required blood transfusion and hysterectomy after failure to respond to conservative measures.
TABLE 1. Population characteristics among women with placenta previa
| Characteristic | Placenta overlapping cervical os (n = 42) | Cervix-to-placenta distance of 1-20 mm (n = 53) | P value or odds ratio (95% CI) |
|---|---|---|---|
| Maternal age (y)a | 34.4 ±4.9 | 34.0±3.4 | .63 |
| Nulliparity (n) | 27(64%) | 32(60%) | 1.2(0.5–2.7) |
| Days from last scan to deliverya | 9.0±6.1 | 10.0±8.1 | .51 |
| Anterior placenta (n) | 25(59%) | 21(40%) | 2.2(0.9–5.1) |
| Antepartum hemorrhage (n) | 11(26%) | 8(15%) | 2.0(0.7–5.4) |
| Weeks of gestation at deliverya | 35.5±3.6 | 37.8±3.2 | .001 |
| Delivery at <37 weeks of gestation (n) | 21(50%) | 12(23%) | 3.4(1.4–8.2) |
| Blood loss at delivery (mL)a | 1195±1588 | 579±513 | .009 |
| Postpartum hemorrhage (n) | 12(28%) | 8(15%) | 2.3(0.8–6.0) |
| Postpartum hemorrhage >1000 mL (n) | 12(29%) | 3(6%) | 6.7(1.8–23.7) |
| Cesarean delivery (n) | 42(100%) | 27(51%) | <.001 |
| Prelabor cesarean delivery (n) | 39(93%) | 25(47%) | 14.6(3.7–67.5) |
| Birthweight (g)a | 2578±717 | 3038±690 | .002 |
aData are given as mean ± SD. |
The remaining 53 women had a placental edge-to-internal os distance of 1-20 mm at the last ultrasound scan that was performed <28 days before delivery; they were divided into 2 groups according to placental edge-to-os distance: group 1 (n = 24), the distance was 1-10 mm; group 2 (n = 29), the distance was 11-20 mm (Table 2). Rates of antepartum hemorrhage and cesarean delivery were higher in group 1 than in group 2; in particular, the only 3 cases of prelabor cesarean section delivery because of bleeding occurred in group 1, as did the only 2 cases of cesarean section delivery during labor. Rates of postpartum hemorrhage were similar in the 2 groups. In group 2, none of the women underwent cesarean delivery during labor, and blood loss was similar in those who underwent cesarean delivery (n = 9) vs vaginal delivery (n = 20; 550 ± 306 mL vs 492 ± 633 mL, respectively; P = .79). Only 1 of 53 cases required postpartum hysterectomy for uterine atony. The woman was 39 years old (3 pregnancies, 2 viable offspring) and was diagnosed with placenta previa at 18 weeks of gestation. The diagnosis was confirmed at 28.3 and 39.1 weeks of gestation, with a placental edge to internal os distance of 12 mm. She never experienced bleeding episodes during pregnancy. She went into spontaneous labor at 41.4 weeks of gestation and delivered vaginally of a male infant who weighed 4050 g, with an Apgar score of 10 at 5 minutes. She subsequently experienced uterine atony that was unresponsive to conservative measures, which lead to a blood loss of 3000 mL that required blood transfusion and hysterectomy.
TABLE 2. Comparison of women with placenta previa according to the placental edge to internal os distance
| Variable | Cervix-to-placenta distance of 1-10 mm (n = 24) | Cervix-to-placenta distance of 11-20 mm (n = 29) | P value or odds ratio (95% CI) |
|---|---|---|---|
| Maternal age (y)a | 35.2±3.8 | 33.1±2.8 | .02 |
| Nulliparity (n) | 14(58%) | 18(62%) | .9(0.3–2.5) |
| Weeks of gestation at last scana | 36.2±2.7 | 36.4±2.8 | .81 |
| Days from last scan to deliverya | 10.3±9.5 | 10.2±6.9 | .96 |
| Anterior placenta (n) | 7(29%) | 14(48%) | .4(0.1–1.4) |
| Antepartum hemorrhage (n) | 7(29%) | 1(3%) | 11.5(1.6–76.7) |
| Weeks of gestation at deliverya | 37.7±3.3 | 37.9±3.1 | .83 |
| Delivery at <37 weeks of gestation (n) | 6(25%) | 6(21%) | 1.3(0.4–4.5) |
| Blood loss at delivery (mL)a | 662±466 | 510±547 | .29 |
| Postpartum hemorrhage (n) | 5(21%) | 3(10%) | 2.3(0.5–9.7) |
| Postpartum hemorrhage >1000 mL (n) | 2(8%) | 3(10%) | .8(0.1–4.4) |
| Cesarean delivery (n) | 18(75%) | 9(31%) | 6.7(2.0–22.0) |
| Prelabor cesarean delivery (n) | 16(67%) | 9(31%) | 4.4(1.2–16.9) |
| Birthweight (g)a | 3077±652 | 3006±729 | .71 |
aData are given as mean ± SD. |
Among the 39 prelabor cesarean section deliveries of women with placenta overlapping the internal os, 10 deliveries (26%) were performed because of bleeding, compared with 3 of 25 deliveries (12%) among cases with a placental distance of 1-20 mm (odds ratio, 2.5; 95% confidence interval, 0.5–13.3). The remaining indication for prelabor cesarean section delivery was the presence of placenta previa at term in the former group; the indications in the latter group are shown in Table 3. Cesarean section delivery during labor was performed in 3 of 42 women (7%) with complete placenta previa because of preterm labor and in 2 of 53 women (4%) with placental edge to os distance of 1-20 mm, because of fetal intolerance to labor in 1 case and hemorrhage in the other case.
TABLE 3. Indications for prelabor cesarean section delivery in women with placenta previa according to the placental edge to internal os distance
| Indication | Cervix-to-placenta distance of 1-10 mm (n = 16) | Cervix-to-placenta distance of 11-20 mm (n = 9) | P value or odds ratio (95% CI) |
|---|---|---|---|
| Uterine scara | 4(25%) | 1(11%) | 2.7(0.2–74.5) |
| Antepartum hemorrhage | 3(19%) | 0 | .46 |
| Malpresentation | 3(19%) | 4(44%) | .3(0–2.4) |
| Myomas | 3(19%) | 0 | .46 |
| Maternal indicationsb | 3(19%) | 2(22%) | .8(0.1–9.2) |
| Fetal growth restrictionc | 2(13%) | 3(33%) | .3(0–3.0) |
| Fetal macrosomiac | 1(17%) | 0 | 1.00 |
aAll of these women were nulliparous. The case with a uterine scar that was allowed to labor had a previous successful vaginal birth after cesarean section delivery; |
bmaternal indications included: micromelia, fever and acute pulmonary edema, chronic myeloid leukemia, preeclampsia, and open perineal fistula; |
cfetal growth restriction was defined as an ultrasonographic abdominal circumference at <10th percentile; fetal macrosomia was defined as an ultrasonographic estimated fetal weight >4000 g. |
Comment
Although official organizations and expert opinions discourage obstetricians from allowing labor and attempting vaginal delivery in presence of a placenta previa with a placental edge within 20 mm from the internal os, our results show that such cases are quite heterogeneous. Whereas a distance <10 mm from the cervix to the placental edge is associated with a need for cesarean section delivery in 75% of cases, most cases (69%) with a distance between 11 and 20 mm can deliver vaginally if allowed to labor. Similar findings were reported by Dawson et al,2 who used translabial evaluation of placenta previa in 18 cases and reported a cesarean section delivery rate of 90% (10/11 deliveries) for a distance of 0-10 mm and of 29% (2/7 deliveries) for a distance of 10-20 mm. The few other series on the subject lumped together cases with distances <20 mm and thus failed to appreciate the presence of a relevant cutoff within such a distance.1, 3 Moreover, the small number of cases with a distance between 1 and 20 mm in the other series on the subject (<40 cases) most likely prevented them from identifying a different and more appropriate cutoff within such distance to allow labor.1, 3, 8 A corollary conclusion of our findings is that a modern classification of placenta previa should abandon the time-honored definitions of “marginal previa” vs “low-lying placenta.” A direct measurement of the placental edge to cervical os distance with transvaginal sonography optimally permits the identification of women with different risk of cesarean section delivery. A second corollary is that a threshold distance of ≥11 mm not only identifies cases with a high likelihood of successful vaginal delivery but also identifies cases that are at low risk of bleeding either before or during labor. No woman with a distance between 11 and 20 mm who was allowed to labor had significant bleeding before or during labor, compared with 3 women with a distance between 1 and 10 mm.
The higher rate of vaginal delivery in the presence of a placental edge between 1 and 20 mm from the cervical os in our series, compared with other series in the literature, may reflect an overall attitude at our institution to encourage vaginal birth. The liberal policy of admitting women to labor in our institution is witnessed by the low rate of cesarean delivery with a placenta-to-os distance >20 mm (19%; 3/16 deliveries), which in other series was 56% (22/39 deliveries) and 51% (27/53 deliveries).3, 7 A placenta considered “too low” may in fact lower the threshold for cesarean section delivery in the presence of other factors, such as even minimal bleeding.
It should be noted that a placental edge within 20 mm of the internal os represents an important risk factor for prelabor cesarean section delivery and for indications such as malpresentation or presence of uterine scar, which in our series accounted for 47% of cases. We have noted that the shorter the placenta-to-os distance, the higher the risk of cesarean section delivery before labor.
We hope that our data provide reassurance to obstetricians about the safety and success of allowing labor with a placental distance >10 mm from the os. Such reassurance may eventually prompt other institutions to confirm our findings in larger series.
References
- . What is a low-lying placenta?. Am J Obstet Gynecol. 1991;165:1036–1038
- . Translabial ultrasonography and placenta previa: does measurement of the os-placenta distance predict outcome?. J Ultrasound Med. 1996;15:441–446
- . Placental edge to internal os distance in the late third trimester and mode of delivery in placenta praevia. BJOG. 2003;110:860–864
- Royal College of Obstetricians and Gynaecologists. Placenta praevia and placenta praevia accreta: diagnosis and management. Medical Specialty Society. 2001 January (revised 2005 October). Guideline no. 27. NGC: 004763. National Guideline Clearinghouse. Available at: www.guideline.gov. Accessed June 1, 2009.
- . Diagnosis and management of placenta previa. J Obstet Gynaecol Can. 2007;29:261–273
- . Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol. 2006;107:927–941
- . A sonographic assessment of different patterns of placenta previa “migration” in the third trimester of pregnancy. J Ultrasound Med. 2005;24:773–780
- . Third trimester transvaginal ultrasonography in placental previa: does the shape of the lower placental edge predict clinical outcome?. Ultrasound Obstet Gynecol. 2001;18:103–108
Cite this article as: Vergani P, Ornaghi S, Pozzi I, et al. Placenta previa: distance to internal os and mode of delivery. Am J Obstet Gynecol 2009;201:266.e1-5.
Authorship and contribution to the article is limited to the 7 authors indicated. There was no outside funding or technical assistance with the production of this article.
PII: S0002-9378(09)00629-2
doi:10.1016/j.ajog.2009.06.009
© 2009 Mosby, Inc. All rights reserved.
Refers to article:
-
A new classification of placenta previa: Measuring progress in obstetrics
Volume 201, Issue 3 , Pages 266.e1-266.e5, September 2009
