Impact of cord entanglement on perinatal outcome of monoamniotic twins: a systematic review of the literature

Authors

  • A. C. Rossi,

    1. Department of Obstetrics and Gynaecology, University of Bari, Bari, Italy
    Search for more papers by this author
  • F. Prefumo

    Corresponding author
    1. Maternal-Fetal Medicine Unit, Department of Obstetrics and Gynaecology, University of Brescia, Brescia, Italy
    • Department of Obstetrics and Gynaecology, University of Bari, Bari, Italy
    Search for more papers by this author

Correspondence: Dr F. Prefumo, Maternal-Fetal Medicine Unit, Department of Obstetrics and Gynaecology, University of Brescia, Piazzale Spedali Civili, 1, 25123 Brescia, Italy (e-mail: federico.prefumo@gmail.com)

ABSTRACT

Objective

To review the current literature concerning perinatal outcome of monoamniotic pregnancies with cord entanglement.

Methods

A search in PubMed, EMBASE and MEDLINE for articles published between January 2000 and December 2011 was performed, using the following keywords: monoamniotic/monochorionic pregnancy(ies); twins; cord entanglement; fetal ultrasound/surveillance. Inclusion criteria for the study were: monoamniotic twins with documented cord entanglement at delivery, and perinatal outcome reported as proportional rates. Exclusion criteria were: higher-order multiple pregnancy; selective feticide; presence of twin reversed arterial perfusion sequence; conjoined twins; fewer than four cases in the series; and non-English language publication. Survival rates were stratified for method of prenatal management, sonographic diagnosis of cord entanglement and delivery mode. A meta-analysis was also performed using data from articles that stratified outcome according to the presence or absence of cord entanglement at birth. Comparison between neonates with cord entanglement and controls was deemed significant if the 95% CI of the pooled odds ratios did not encompass 1. MOOSE (Meta-analysis Of Observational Studies in Epidemiology) guidelines were followed.

Results

Nine articles met the inclusion criteria for this review, including a total of 114 monoamniotic twin sets (228 fetuses) with cord entanglement. The overall survival rate was 202/228 (88.6%). Perinatal mortality occurred in 26 (11.4%) fetuses; of these, 17 (65%) died in utero and nine (35%) died at birth. Five neonatal deaths occurred as a result of prematurity, two were related to structural abnormalities and two were caused by cord entanglement. Sonographic visualization of cord entanglement did not improve outcome. Four articles were included in the meta-analysis, with no significant difference in mortality between controls (n = 66 fetuses) and twins with cord entanglement (n = 82 fetuses), and higher morbidity in controls.

Conclusions

Cord entanglement does not contribute to prenatal morbidity and mortality in monoamniotic twin pregnancies.

INTRODUCTION

Monoamniotic pregnancies occur rarely, representing about 1% of cases of monozygotic twins. It is generally believed that monoamniotic twins are at high risk of fetal death and neonatal morbidity, secondary to umbilical cord entanglement[1, 2], prematurity and congenital anomalies. A significant proportion of cases of intrauterine fetal demise can be attributed to cord-related complications, in which the fetal circulation is compromised by occlusion of the umbilical vessels.

A relatively recent review on cases of monoamniotic pregnancies showed that cord entanglement was documented sonographically in approximately 22% of cases[3]. This is in contrast to the common finding of cord entanglement at the time of delivery in the vast majority of monoamniotic pregnancies[4]. It is noteworthy that the authors included literature from the early 1990s. Because significant advances in obstetric ultrasound resolution have allowed better assessment of monoamniotic pregnancies over the past decade, it can be speculated that prenatal diagnosis of cord entanglement has improved over time. Color flow mapping and Doppler velocimetry are associated with a positive predictive value of 89% for diagnosis of cord entanglement[5]. The use of three-dimensional (3D) ultrasound may also enhance the accuracy of cord entanglement assessment, since 3D volume-rendered images provide additional details[6]. Nonetheless, prenatal management remains controversial. Some authors report that cord entanglement appears in early pregnancy[4, 7] and its complications are unpredictable and may occur acutely. Therefore, elective Cesarean section may prevent umbilical cord injury, although the optimal timing for delivery has not been determined. Other authors recommend expectant management and hospitalization of the mother for continuous fetal monitoring, with prompt delivery if fetal distress becomes apparent[1].

Because of the low prevalence of monoamniotic pregnancies, most of the literature consists of case reports and small series. We therefore performed a review of literature published in the last decade concerning perinatal outcome of monoamniotic pregnancies with documentation of cord entanglement at birth.

Methods

A search in PubMed, EMBASE, MEDLINE and reference lists of selected articles was performed in order to find relevant articles from the period between January 2000 and December 2011 that described perinatal outcome of monoamniotic twin pregnancies complicated by cord entanglement. The search was limited to articles published from 2000 onwards because improvements in ultrasound techniques over the past decade have allowed more accurate prenatal diagnosis than in previous years; missed diagnoses of cord entanglement by ultrasound would have been more likely in previous years and would have biased our results. Keywords for the search were: monoamniotic/monochorionic pregnancy(ies); twins; cord entanglement; fetal ultrasound; fetal surveillance. Studies were included according to the following criteria: monoamniotic twins with documented cord entanglement at the time of delivery and perinatal outcome reported as proportional rates rather than graphs or percentages. Cord entanglement at birth was defined by the macroscopic appearance of a twisted umbilical cord in several coils. Exclusion criteria for study selection were: higher-order multiple pregnancy; selective feticide; presence of twin reversed arterial perfusion sequence; conjoined twins; and non-English language publication. In order to reduce publication bias, case reports and case series with fewer than four cases were also excluded.

Perinatal outcome consisted of overall survival rates, number of surviving twins for each pregnancy, intrauterine demise and neonatal mortality and morbidity. Neonatal morbidity was defined as at least one of the following: Apgar score < 7 at 5 min; respiratory distress syndrome; admission to a neonatal intensive care unit; congenital heart block; necrotizing enterocolitis; and neurological impairment. Neonatal death included death at or within 28 days following birth.

Data collected from each study were: sample size; method of prenatal management; and whether diagnosis of cord entanglement was by ultrasound. Study selection and data extraction were performed independently by each of the authors, following MOOSE (Meta-analysis Of Observational Studies and Epidemiology) guidelines[8]. Discordance was resolved by consensus. When required, efforts to contact the corresponding author from the original publication were made in an attempt to obtain unpublished data. Risk of bias within and across studies was assessed according to the Cochrane Collaboration's tool for assessing risk of bias.

Articles that stratified outcome according to the presence or absence of cord entanglement at birth were pooled for meta-analysis. This was performed to compare overall survival rate and perinatal mortality (both intrauterine demise and neonatal mortality rate) between twins with cord entanglement and those without. Controls were defined as twins without evidence of cord entanglement at birth. For this purpose, interstudy heterogeneity was defined, according to Higgins et al.[9], as the percentage of total variation across studies due to heterogeneity rather than chance (I2) and was tested with the chi-square test for heterogeneity at a significance level of P = 0.10. A random effects model was generated whenever I2 was ≥ 25%. Pooled odds ratio and 95% CIs were calculated. Intergroup comparisons were deemed statistically significant if the 95% CI did not encompass 1. Meta-analysis was performed with RevMan (Revision Manager, Version 5.1 for Macintosh, Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration 2011).

Results

Figure 1 shows the steps involved in study selection. Nine articles met the inclusion criteria, with a total of 114 monoamniotic twin sets with cord entanglement (228 fetuses)[2, 4, 7, 10-15]. The characteristics of each study are summarized in Table 1. The overall survival rate was 202/228 (88.6%). There were no survivors in 8/114 (7.0%) twin sets, single survival in 10/114 (8.8%) and double survival in 96/114 (84.2%). At least one survivor was present in 106/114 (93.0%) pregnancies. Neonatal morbidity affected 48/228 (21.1%) twins. Perinatal mortality occurred in 26 (11.4%) fetuses; of these, 17 (65%) died in utero and nine (35%) died at birth. Gestational age at time of intrauterine fetal demise ranged from 14 to 33 weeks. Of the nine neonatal deaths, two were related to structural abnormalities (one case each of intra-abdominal tumor and transposition of great vessels). A further five deaths were a result of prematurity, and two neonatal deaths were attributed to cord entanglement.

Figure 1.

Flow chart of studies included in the systematic review.

Table 1. Characteristics of studies included in the systematic review
First authorYear of publicationCountryTwin sets with CE (n)Prenatal managementUltrasoundvisualizationof CEOverall survivors (n)Neonatal morbidity (n)IUFD (n)NND (n)
  1. Outcomes for individual pregnancies have been given where information was available. *Ultrasound performed according to maternal and fetal condition. CE, cord entanglement; FHR, fetal heart rate monitoring; IUFD, intrauterine fetal demise; NA, not available; NND, neonatal death; US, serial ultrasound examinations.

Pasquini[7]2006UK20USYes402200
Dias[4]2010UK18NAYes32022
Arabin[10]2009Netherlands18NAYes27054
Hack[12]2009Netherlands2Maternal/fetal condition*NA0031
Cordero[2]2006USA15Maternal/fetal condition*NA30200
Sau[13]2003UK5USYes2100
    USYes 110
    USYes2200
    USNo0020
    USYes1001
Suzuki[14]2001Japan10NANA2000
      2100
      2000
      0020
      2100
      2000
      2000
      2200
      1110
      2100
DeFalco[11]2006USA11USNA2100
    USNA2200
    USNA2100
    USNA2200
    USNA2200
    USNA1110
    USNA2000
    FHRNA2000
    USNA2200
    USNA2200
    USNA2000
Baxi[15]2010USA15US/FHRNA29101

Four articles were pooled to compare perinatal outcome of pregnancies complicated by cord entanglement and controls[2, 11, 13, 14]. No significant differences in outcome were observed between the two groups with regard to overall survival rate (Figure 2), and neonatal morbidity was higher in the controls (Figure 3).

Figure 2.

Forest plot of odds ratio for overall mortality according to presence or absence of cord entanglement in monoamniotic twins. Numbers refer to number of fetuses. M–H, Mantel–Haenszel test.

Figure 3.

Forest plot of odds ratio for neonatal morbidity according to presence or absence of cord entanglement in monoamniotic twins. Numbers refer to number of fetuses. M–H, Mantel–Haenszel test.

Prenatal management was described in all except three articles[4, 10, 14], giving a total of 68 twin sets with cord entanglement for which information on prenatal management was available. Prenatal management consisted of serial ultrasound examinations of fetal growth and Doppler flow in 35/68 (51.5%) twin sets and serial biophysical assessment with daily fetal heart monitoring after 26–28 weeks in 15/68 (22.1%) twin sets, whereas in 17/68 (25.0%) pregnancies ultrasound was performed according to maternal and fetal conditions and in one (1.5%) pregnancy fetal heart monitoring was performed. In the remaining 46 (40.4%) pregnancies, prenatal management was not described. The overall survival rate of twins that were serially assessed was 96/102 (94.1%), with one case of dual demise (1/51, 2%), 4/51 (8%) pregnancies with a single survival and 46/51 (90%) with double survival. Neonatal morbidity was diagnosed in 40/102 (39%) neonates. Perinatal mortality, which occurred in six twins, consisted of four intrauterine deaths and two neonatal deaths due to prematurity. Of the 17 monoamniotic pregnancies that were managed according to maternal and fetal condition, 30/34 (88%) fetuses survived overall. In 2/17 (12%) pregnancies neither twin survived and there was double survival in the remaining 15/17 (88%) twin sets. Neonatal morbidity occurred in 2/34 (6%) fetuses. Perinatal mortality was observed in four fetuses, with intrauterine demise in 3/4 (75%) and neonatal death in 1/4 (25%).

Four articles[4, 7, 10, 13] reported on whether cord entanglement was searched for systematically and found on ultrasound examination, giving information on 61 monoamniotic pregnancies in total: cord entanglement was detected by ultrasound in 60/61 (98%) pregnancies and missed in 1/61 (2%). Gestational age at diagnosis of cord entanglement ranged from 11[4, 10] to 30[7] weeks. When cord entanglement was observed by ultrasound, the overall survival rate was 105/120 (88%). There were no survivors in 4/60 (7%) twin sets, single survival in 7/60 (12%) and double survival in 49/60 (81%). Neonatal morbidity affected 26/120 (21.7%) fetuses. A total of 15 fetuses died; of these, eight (53%) died in utero and seven (47%) at birth because of prematurity or major abnormalities. In the only pregnancy in which cord entanglement was not observed sonographically, a double intrauterine fetal death occurred before 19 weeks of gestation[13].

Discussion

This review has shown that cord entanglement is a minor complication of monoamniotic twin pregnancies. In approximately 90% of cases, at least one twin survived, and in 84% of cases, both twins survived. Although perinatal mortality occurred in approximately 10% of cases, only two of the neonatal deaths were associated with cord entanglement, whereas the others were due to prematurity or major congenital abnormalities. This could explain our finding that adverse outcome was more frequent in the control group compared with twins affected by cord entanglement in studies that reported outcomes according to the presence or absence of cord entanglement at birth.

We were unable to demonstrate that sonographic detection of cord entanglement improves perinatal outcome, since we observed that survival rate was around 88% when cord entanglement was detected by ultrasound. This confirms that both neonatal morbidity and mortality of monoamniotic pregnancies do not depend on the presence of cord entanglement, but that other factors, such as twin–twin transfusion syndrome, prematurity and congenital anomalies, contribute to poor outcome. Therefore, prenatal management of monoamniotic pregnancies should focus on the detection of sonographic signs of twin–twin transfusion syndrome, structural anomalies or risk of prematurity, rather than the presence of cord entanglement.

It has been suggested that prenatal management of monochorionic twin pregnancies, both monoamniotic and diamniotic, should consist of ultrasound assessment of fetal growth, blood flow and amniotic fluid every 2 weeks. This is in agreement with our finding that the overall survival rate in twins who received serial surveillance by ultrasound was higher (94%) compared with twins managed according to maternal and fetal conditions (88%). However, this study did not aim or have sufficient power to demonstrate the effect of any particular antenatal management protocol on pregnancy outcome.

Other limitations of this review should be considered. An abnormally high systolic/diastolic ratio and the presence of a notch in the umbilical artery velocity waveform might indicate narrowing of the umbilical vessels involved in the cord entanglement, which could lead to circulatory compromise[13]. However, most of the reviewed articles did not specify whether Doppler anomalies were present and therefore we could not comment on whether they were associated with neonatal morbidity and mortality. Furthermore, some of the included articles were retrospective and could not have assessed systematically the presence of cord entanglement during ultrasound examination.

The type of prenatal management, local ultrasound protocols, as well the policy for the timing of delivery in monoamniotic pregnancies used in different studies may have affected the association between cord entanglement and perinatal outcome. When emergency delivery was performed for non-reassuring fetal heart traces, it was not possible to establish whether fetal distress was directly attributable to cord entanglement or other underlying conditions. In addition, cord entanglement complications may occur progressively or may occur suddenly in response to fetal movements. Further studies are needed to investigate whether elective delivery at a prespecified gestational age is effective in reducing perinatal mortality and morbidity in monoamniotic twins with a prenatal diagnosis of cord entanglement. Ideally, only monoamniotic pregnancies with cord entanglement and without any other conditions that may influence perinatal outcome should be included. Randomized clinical trials would probably define the optimal mode of delivery, but due to the rarity of monoamniotic pregnancies, such trials would require a very long period of time.

Finally, in the pooled studies we did not find sufficient information to investigate whether perinatal outcome is associated with inpatient vs outpatient management of monoamniotic twins with prenatal diagnosis of cord entanglement.

In conclusion, this review has shown that umbilical cord entanglement does not contribute to perinatal morbidity and mortality in monoamniotic twin pregnancies, probably as a result of close fetal monitoring.

Ancillary