Peripheral cord insertion is associated with adverse pregnancy outcome only when accompanied by clinically significant placental pathology
ABSTRACT
Objective
To examine the relationship between umbilical cord insertion site, placental pathology and adverse pregnancy outcome in a cohort of normal and complicated pregnancies.
Methods
Sonographic measurement of the cord insertion and detailed placental pathology were performed in 309 participants. Associations between cord insertion site, placental pathology and adverse pregnancy outcome (pre-eclampsia, preterm birth, small-for-gestational age) were examined.
Results
A total of 93 (30%) participants were identified by pathological examination to have a peripheral cord insertion site. Only 41 of the 93 (44%) peripheral cords were detected by prenatal ultrasound. Peripherally inserted cords were associated significantly (P < 0.0001) with diagnostic placental pathology (most commonly with maternal vascular malperfusion (MVM)); of which 85% had an adverse pregnancy outcome. In cases of isolated peripheral cords, without placental pathology, the incidence of adverse outcome was not statistically different when compared to those with central cord insertion and no placental pathology (31% vs 18%; P = 0.3). A peripheral cord with an abnormal umbilical artery (UA) pulsatility index (PI) corresponded to an adverse outcome in 96% of cases compared to 29% when the UA-PI was normal.
Conclusions
This study demonstrates that peripheral cord insertion is often part of the spectrum of findings of MVM disease and is associated with adverse pregnancy outcome. However, adverse outcome was uncommon when there was an isolated peripheral cord insertion and no placental pathology. Therefore, additional sonographic and biochemical features of MVM should be sought when a peripheral cord is observed. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.
CONTRIBUTION
What are the novel findings of this work?
There is conflicting evidence regarding the relationship between peripheral cord insertion and pregnancy outcome. This study used detailed placental pathology and found peripheral cords are most common with maternal vascular malperfusion; however, adverse pregnancy outcome is associated with pathology within the placenta, not with the cord insertion.
What are the clinical implications of this work?
These findings suggest that patients with peripheral cord insertion, normal Doppler ultrasound and normal circulating placental growth factor levels are at low risk of adverse outcome, and additional fetal surveillance is not justified.
INTRODUCTION
The placenta is a vital organ for normal fetal development, as it allows the hemochorial exchange between the maternal and fetal circulatory systems that supply the intervillous and villous vascular tree, respectively. Typically, the villous tree is symmetrically formed when the umbilical cord inserts paracentrally into the placenta. However, in 6–11% of pregnancies the cord inserts peripherally, potentially affecting efficiency of placental function1, 2. Studies of pregnancies with peripheral cord insertion have yielded variable results. While several studies report that peripheral cord insertion has no relationship with adverse pregnancy outcome3-5, others found associations with a variety of complications including pre-eclampsia6, 7, preterm birth2, 6, 8, 9, small-for-gestational-age (SGA) neonates6, 7, 9-11, emergency Cesarean delivery6, 9, admission to the neonatal intensive care unit6, 7, low Apgar score6 and a variety of fetal malformations6, 12. A 2017 systematic review13 of 17 clinical studies concluded that peripheral cord insertion was associated significantly with emergency Cesarean delivery, but the number of studies assessing other adverse pregnancy outcomes was too small to make conclusive associations. A limitation common to these studies was that they did not systematically consider placental pathology, although it may contribute substantially to the development of pre-eclampsia, impaired fetal growth and adverse perinatal outcome. Peripheral cord insertion is more frequent in placentas with features of either maternal vascular malperfusion (MVM)14 or fetal vascular malperfusion (FVM)15, illustrating that abnormal cord insertion and some types of placental pathology may be linked mechanistically. The aim of the present study was to examine the relationship between umbilical cord insertion site, placental pathology and adverse pregnancy outcome in a cohort of normal and complicated pregnancies.
METHODS
Patient cohort
A total of 427 pregnant individuals were recruited for fetal ultrasound examination in a study that evaluated a new methodology for measuring umbilical artery (UA) hemodynamics16. Inclusion criteria were: singleton pregnancy; maternal age between 18 and 45 years; and a body mass index (BMI) < 45 kg/m2. Written informed consent was obtained from each participant. Datasets were excluded when a major fetal abnormality was detected or the patient withdrew at any point during the study. Obstetric data were obtained from the participants' medical records. Adverse pregnancy outcome was defined as one or more of: pre-eclampsia (diagnosed according to the American College of Obstetricians and Gynecologists (ACOG) guidelines17); preterm birth (delivery < 37 weeks' gestation); or SGA neonate (birth weight < 10th centile according to Canadian population growth charts18). Ethical approval was obtained from the ethics boards of The Hospital for Sick Children, Toronto, ON, Canada (REB number 1000051548), Mount Sinai Hospital, Toronto, ON, Canada (REB number: 15-0279-A) and Johns Hopkins Medicine, Baltimore, MD, USA (IRB number: 00082717).
Doppler ultrasound assessment
Ultrasound examinations were conducted between 26 and 32 weeks of gestation by dedicated certified research sonographers, using either a Philips iU22 (Philips Healthcare, Andover, MA, USA) or a General Electric Voluson e10 (GE Healthcare, Zipf, Austria) ultrasound machine. Placental morphological measurements were obtained using the curve–linear method at the maternal interface19. Color Doppler was used to confirm the placental cord insertion site and the transducer was rotated to identify the imaging plane with the shortest distance from the cord insertion to the edge of the placenta. For the purpose of this study, from the ultrasound examination we defined a peripheral cord as a distance of less than 3.0 cm from the cord insertion site to the placental edge. Pulsed-Doppler spectra were collected for the two UAs in a free loop of the umbilical cord away from the cord insertion site20 and the pulsatility index (PI) was computed from the traced average Doppler waveforms as the difference between the peak systolic and end-diastolic velocities, divided by the mean velocity over the cardiac cycle. UA Doppler waveforms were not recorded for 20 of the participants.
Placental pathology
After delivery, placentas were fixed in 10% formaldehyde for 48 h and then evaluated by experienced perinatal pathologists using the Amsterdam working-group definitions for placental lesions21. Gross findings were recorded, including placental weight and dimensions (longest axis, orthogonal axis and thickness (cm)), number of cord vessels, cord insertion and cord coiling. For the pathological analysis, a peripheral cord was defined as a cord inserted < 3.0 cm from the placental edge and a marginal cord was defined more strictly as < 2.0 cm from the placental edge5, 22, 23. However, there is no consensus on the definition of marginal cord and therefore we also investigated cords inserted < 1.0 cm from the placental edge21, 24. The fixed placental disc was then cut into a series of 2-cm thick slices to assess gross pathology; gross lesions were sampled, together with normal tissue, a segment of umbilical cord and a fetal membrane roll. The sampled tissues were then paraffin embedded, cut into 4-µm thick sections and stained with hematoxylin and eosin to assess microscopic structures. Placental pathology was classified into five diagnostic categories14: (1) no pathology, minor non-diagnostic findings (e.g. small placenta without any additional pathology or features of acute chorioamnionitis); (2) MVM; (3) FVM; (4) non-infectious chronic maternal inflammation (chronic villitis of unknown etiology, chronic intervillositis); and (5) massive perivillous fibrin deposition. MVM was diagnosed using the criteria outlined by Zur et al.25. In addition to a small placental size for gestational age (< 10th centile), the placenta must show evidence of one or more macroscopic (infarction, hemorrhage) and histological features (distal villous hypoplasia, accelerated villous maturation, syncytial knots, decidual vasculopathy).
Statistical analysis
All statistical tests were performed using the R statistical software package (www.r-project.org). Data are reported as mean ± SD. The UA-PIs were transformed into Z-scores using the INTERGROWTH-21st equations26. For the analysis of continuous variables (maternal age, pregestational BMI, gestational age at delivery, birth weight, UA-PI Z-score), a one-way ANOVA was used to evaluate the effect of group (central cord and no placental pathology; central cord and significant placental pathology; peripheral cord and no other placental pathology; peripheral cord and significant placental pathology). Cord insertion site was treated as a dichotomous variable to be consistent with previous studies13. If ANOVA was significant, Tukey Honestly Significant Difference (HSD) post-hoc tests were performed. For the analysis of the remaining clinical characteristics, pregnancy outcome (normal, adverse), placental pathology (none, significant) and UA-PI (normal, abnormal), the variables were treated as categorical and analyzed using Fisher's exact test to evaluate the effect of the group. If the Fisher's test was significant, a post-hoc pairwise comparison using Fisher's exact test was performed with a Bonferroni correction for multiple comparisons. A P-value of < 0.05 was considered significant.
RESULTS
Patient characteristics
Of 427 patients who consented to participate in the study, 309 had an ultrasound examination in which their placental umbilical cord insertion could be measured confidently and had their placenta assessed for pathology (25 withdrew and 93 had placentas that were not sent for pathological examination). During the prenatal ultrasound examination, 54 of the 309 (17%) participants were identified as having peripheral cord insertion. After delivery, 93 (30%) participants were identified by pathological examination to have a peripheral cord insertion, with 41 of the 93 (44%) having previously been identified prenatally (Figure 1). There were 69 (22%) participants with diagnostic placental pathology. Placental pathology was present in 35/216 (16%) of the participants with a central cord insertion and 34/93 (37%) of the participants with a peripheral cord insertion (P < 0.0001). We then stratified participants into four groups based on their cord insertion site from the pathological examination and the presence of significant placental pathology. There were 181 of the 309 (59%) participants with a central cord insertion and no clinically significant placental pathology. Thirty-five (11%) had a central cord insertion and a significant placental pathology diagnosis. Fifty-nine (19%) had a peripheral cord insertion with no significant pathology diagnosis, and 34 (11%) had a peripheral cord insertion with a significant placental pathology diagnosis. The clinical characteristics of the four groups are summarized in Table 1.
| Central cord insertion | Peripheral cord insertion | |||
|---|---|---|---|---|
| Characteristic | No placental pathology (n = 181) | Significant placental pathology (n = 35) | No placental pathology (n = 59) | Significant placental pathology (n = 34) |
| MA at delivery (years) | 33.6 ± 4.2 | 34.5 ± 3.8 | 33.0 ± 4.3 | 32.9 ± 4.7 |
| BMI (kg/m2) | 25.0 ± 5.1 | 25.2 ± 5.5 | 25.2 ± 5.0 | 24.4 ± 5.5 |
| GA at delivery (weeks)* | 38.1 ± 1.9 | 35.4 ± 3.8§ | 37.9 ± 2.0¶ | 33.4 ± 4.7§¶** |
| Cesarean delivery | 76/181 (42) | 22/35 (63) | 29/59 (49) | 24/34 (71)§ |
| Birth weight (g)† | 3240 ± 630 | 2260 ± 1070§ | 2960 ± 700¶ | 1720 ± 1000§¶** |
| Infant female sex† | 89/181 (49) | 17/35 (49) | 32/58 (55) | 16/33 (48) |
| NICU admission | 27/181 (15) | 15/35 (43)§ | 10/59 (17) | 21/34 (62)§** |
| Pre-eclampsia | 6/181 (3) | 8/35 (23)§ | 1/59 (2)¶ | 4/34 (12) |
| SGA† | 12/181 (7) | 17/35 (49)§ | 13/58 (22)¶ | 23/33 (70)§¶** |
| MVM pathology‡ | — | 15/35 (43) | — | 21/34 (62) |
| FVM pathology | — | 14/35 (40) | — | 5/34 (15) |
| CHIV pathology | — | 5/35 (14) | — | 7/34 (20) |
| MPFD pathology | — | 1/35 (3) | — | 1/34 (3) |
- Data are given as mean ± SD or n/N (%).
- * Data missing for one participant (peripheral cord, no placental pathology).
- † Data missing for two participants (peripheral cord, no pathology and peripheral cord, significant pathology).
- ‡ Nine participants had primary diagnosis of maternal vascular malperfusion (MVM), but also features of fetal vascular malperfusion (FVM).
- § P < 0.05 when compared with central cord, no pathology;
- ¶ P < 0.05 when compared with central cord, significant pathology;
- ** P < 0.05 when compared with peripheral cord, no pathology.
- BMI, body mass index; CHIV, chronic histiocytic intervillositis; GA, gestational age; MA, maternal age; MPFD, massive perivillous fibrin deposition; NICU, neonatal intensive care unit; SGA, small-for-gestational age.
Association with adverse pregnancy outcome
Obstetric outcome was reviewed to explore the relationship between cord insertion at delivery (measured on pathology) and adverse pregnancy outcome (pre-eclampsia, preterm birth, SGA). Figure 2 illustrates how the incidence of adverse pregnancy outcome varied between pregnancies with central and peripheral cord insertions, stratified by the presence or absence of a significant placental pathology diagnosis. Pregnancies in which the placenta had a pathological finding had significantly higher rates of adverse pregnancy outcome (P < 0.0001). Of 36 participants with MVM, 33 (92%) had an adverse pregnancy outcome. Peripheral cord insertion was associated significantly with adverse pregnancy outcome: 51% of placentas with a peripheral cord insertion had an adverse outcome, compared to 25% for central cord insertions (P < 0.0001). However, in the absence of any diagnostic placental pathology, there was no difference in the incidence of adverse pregnancy outcome between placentas with a central cord or a peripheral cord insertion (18% vs 31%; P = 0.3) (Figure 2a). While not statistically significant, the risk of adverse pregnancy outcome in the group with a peripheral cord insertion was 2-fold higher (odds ratio, 2.0 (95% CI, 1.1–4.0)) than with a central cord insertion, and this finding may be clinically relevant. When the placenta had diagnostic placental pathology, there was no difference in the incidence of adverse pregnancy outcome based on cord insertion site (63% and 85% for central and peripheral cord insertions, respectively; P = 0.3). These results were replicated with a more stringent definition of an abnormal cord insertion site (i.e. < 2.0 cm from the placental edge). There was no difference in adverse outcome between pregnancies with central and marginal cord insertions (< 2.0 cm from the placental edge) in placentas without clinically significant pathology (19% vs 28%; P = 1.0), and the incidence of adverse pregnancy outcome with marginal cord insertion < 2.0 cm varied depending on the presence of clinically significant placental pathology (28% vs 96%; P < 0.0001) (Figure 2b). Unlike with peripheral cord insertion, there was a significant increase in the incidence of adverse pregnancy outcome in placentas with clinically significant pathology with marginal cord insertions when compared to central cord insertions (63% vs 96%; P = 0.02). While the sample size of placentas with a cord insertion less than 1.0 cm was small (n = 29), the same trends were observed (Figure 2c).
) or normal outcome (
), according to: (a) peripheral cord insertion (< 3.0 cm from placental edge), (b) marginal cord insertion < 2.0 cm and (c) marginal cord insertion < 1.0 cm. *P < 0.05. **P < 0.0001.Association with placental pathology
To explore further the relationship between peripheral cord insertion and placental pathology, we stratified the distance from the placental edge into four groups (< 1.0 cm, < 2.0 cm, < 3.0 cm and ≥ 3.0 cm). The distance of the cord insertion from the placental edge was associated inversely with placental pathology (P < 0.0001) (Figure 3). Post-hoc analysis revealed that the marginal cord insertions (< 2.0 cm from the placental edge) and peripheral cord insertions (< 3.0 cm) each had a significantly higher proportion of placentas with clinically significant pathology (37% for both) compared to central cord insertions (≥ 3.0 cm) (16%) (P = 0.004 and P = 0.001, respectively). There were no significant differences between the three stratified groups of abnormal cord insertions.
, placental pathology; 
, no pathology.Association with UA-PI
In addition to evaluating the cord insertion site, standard-of-care prenatal assessment would involve measurement of UA-PI using Doppler ultrasound. There was no relationship between the UA-PI Z-score and distance of cord insertion from the placental edge (P = 0.1, one-way ANOVA; Figure 4a). We then investigated whether cord insertion site, combined with the UA-PI, would better reflect adverse pregnancy outcome than the cord insertion site alone. In participants with a peripheral cord insertion and abnormal UA-PI (> 90th percentile for gestational age)20 the incidence of adverse pregnancy outcome was 96% (27/28) compared to 29% (18/62) when the UA-PI was within the normal range for gestational age (P < 0.0001; Figure 4b). Of 18 participants with a peripheral cord insertion, normal UA-PI and adverse pregnancy outcome, 10 had additional placental pathology on pathological examination. Of 36 participants with MVM, 28 (78%) had abnormal UA-PI. When UA-PI was abnormal, there was no significant difference in the incidence of adverse pregnancy outcome between peripheral and central cord insertions (96% vs 82%; P = 0.2).
). *P < 0.0001. 
, Normal pregnancy outcome.DISCUSSION
In this study, we demonstrate that adverse pregnancy outcome in participants with a peripheral cord insertion is dependent upon the presence or absence of clinically significant pathology inside the placenta, rather than from the abnormal cord insertion. An isolated peripheral cord was not associated with an increased incidence of adverse outcome, and the distance of the peripheral cord from the placental edge (< 3.0 cm, < 2.0 cm or < 1.0 cm) also did not affect pregnancy outcome. When peripheral cord insertion and placental pathology coexist, the clinical effects are most notable with significantly lower birth weight, earlier delivery and subsequently higher neonatal intensive care unit admission numbers. A peripheral cord insertion was more likely to occur with specific placental pathology, especially with the most common of these, MVM disease27, 28. Therefore, when a peripheral placental cord insertion is detected antenatally, additional prenatal diagnostic assessment, including UA and uterine artery Doppler ultrasonography, is indicated to identify pregnancies with coexistent MVM disease29. As discussed by Kingdom et al.27, if UA and uterine artery Doppler and placental growth factor levels are normal, the definitive placenta is well-perfused, regardless of cord insertion site and the pregnancy is therefore at low-risk for developing adverse maternal or perinatal complications. The present study affirms these recommendations, with 71% of participants having a normal pregnancy outcome when the UA-PI was normal despite evidence of a peripheral cord insertion.
In a recent prospective study of 856 low-risk women14, 53% of women meeting the minimum diagnostic criteria for MVM had a normal pregnancy outcome. As discussed in Zur et al.25, this finding suggests that the current diagnostic criteria for MVM are not sufficiently strict, in part because they were developed prior to the establishment of maternal circulating placental growth factor as a robust biomarker of global placental function27, 28, 30. For the MVM disease spectrum of findings to cause adverse pregnancy outcome, they must be multidimensional with an abnormal cord insertion combined with macroscopic and histological features of damaged areas of placental villi (infarction and abnormal villous differentiation). This conclusion is consistent with previous work that describes marginal cords as only being clinically meaningful when accompanied by abnormal placental histological findings31. While most studies investigating the relationship between abnormal cord insertion and pregnancy outcome have not included detailed placental pathology, a study by Tantbirojn et al.32 reported that gross cord abnormalities (true knot, nuchal cord, long cord, hypercoiled cord, abnormal cord insertion) were associated significantly with placental histological features of FVM, and that cases with both abnormal cords and features of FVM had a higher incidence of adverse neonatal outcome (reviewed by Kingdom et al.27). In addition to FVM, this study also found peripheral cord insertion to be associated with both MVM and chronic histiocytic intervillositis placental pathologies. As demonstrated by modeling, it may be that placentas with a non-central cord insertion have a sparser chorionic vascular distribution33 and a more asymmetric chorionic vessel structure that results in more heterogeneous fetal blood flow within the placenta34.
Two recent studies have investigated the relationship between adverse pregnancy outcome and marginal cord insertion, as measured by prenatal ultrasound. The results are conflicting, with one study reporting a significant association between marginal cords < 1.0 cm from the placental edge and adverse outcome35 and the second finding no association with adverse outcome, regardless of the distance of the cord insertion from the placental edge5. One interesting finding in this study is that only 44% of the placentas found to have a peripheral cord insertion at pathological examination had a prenatal observation of this finding at their ultrasound examination. This low detection rate illustrates one of the challenges with ultrasound accuracy, especially with a posterior placenta, in which the larger fetus lies on top of the placenta, thus obscuring the placental surface anatomy. While previous work has reported that cord insertion is visualized reliably in 91–100% of cases23, 24, 36, one simple explanation may be the more advanced gestational age in the present study (26–32 weeks). Placental cord insertion is far easier to ascertain by ultrasound in earlier gestation, up to the 18–20-week fetal anatomical ultrasound examination period. Therefore, cord insertion and placental assessment are best performed at this stage14.
A potential limitation of this study is that not all participants' placentas had a pathological examination. Most of the missing placental pathology was from low-risk pregnancies with normal term outcome; this selection bias may therefore account for the higher rate of pathological findings in comparison to other large cohort studies (22% in the present study compared to 14%14). Another limitation is that the sample size for placentas with a cord insertion < 1.0 cm from the placental edge was small; therefore, comprehensive placental health assessment studies should target this population and include pregnancies with two-vessel cords and those with velamentous cord insertions into the fetal membranes, including when vasa previa is diagnosed. Finally, while a routine ultrasound examination is typically recommended at 18–20 weeks of gestation, the ultrasound data from this study were collected later in pregnancy (26–32 weeks). The association between peripheral cord insertion, placental pathology, UA-PI and pregnancy outcome may have been modified by gestational age at examination. Future investigation of these associations at 18–20 weeks of gestation will determine if a third-trimester scan should be recommended when a peripheral cord insertion is observed.
In conclusion, the principal findings of this work show that there is a significant and clinically relevant relationship between peripheral cord insertion and placental pathology, and that adverse pregnancy outcome is related to the pathology within the placenta rather than the location of the cord insertion itself. For patients with an isolated peripheral cord insertion and, when available, normal additional placental findings (e.g. uterine and UA Doppler), additional fetal surveillance is not justified.
ACKNOWLEDGMENTS
We thank the pregnant people who participated in the study. This work was supported by the Eunice Kennedy Shriver National Institutes of Child Health and Human Development of Human Health Grant 1U01HD087177-01 and a Janeway Foundation Research Grant.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
