The Risk Factors and Neonatal outcomes of Isolated Single Umbilical Artery in Singleton Pregnancy: A Meta-analysis

The current meta-analysis aims to evaluate the risk factors and neonatal outcomes of isolated Single Umbilical Artery (iSUA) in singleton pregnancy. Standard Mean Difference (SMD) or Weighted Mean Difference (WMD) was pooled for the maternal age, gravidity and parity, neonate birth weight and Apgar score one and five minutes after birth. We also pooled the odds ratios (ORs) at 95% confidence intervals (CIs) for maternal smoking status, the rate of neonate delivery before 37 or 34 weeks, Cesarean section (CS), the rate of being admitted to neonatal intensive care unit (NICU) and the serious adverse neonate outcome. Results show that maternal primigravidity [OR: −0.082, CI (−0.152, −0.011), p = 0.023] and female sex of the neonate [OR: 0.805, CI (0.673, 0.963), p = 0.017] were associated with higher risks of iSUA. As compared to normal neonates, the neonates with iSUA had lower birth weight, worse Apgar score, increased risk of delivery before the normal gestational age, increased rate of CS due to fetal distress, increased rate of admission to NICU and prolonged NICU stay. However, no difference in neonatal mortality was observed. Maternal primigravidity and female neonate might associate with increased risk of iSUA. Identification of iSUA is of great importance for prenatal diagnosis and may improve neonatal outcomes.

The risk factors of iSUA. The results from different analysis methods for evaluating the risk factors of isolated SUA are summarized in Table 2. For iSUA, the overall OR was −0.082 (95% CI, −0.152-0.011, P = 0.023, Fig. 2a) in maternal risk factor analysis, which indicated that pregnant women who had low gravidity were associated with an increased risk for iSUA. No heterogeneity was observed (I 2 = 0%). As to fetal risk factors, female fetuses were associated with an increased risk for SUA with the overall OR of 0.805 (95% CI, 0.673-0.963,   The correlation between iSUA and adverse neonatal outcomes. We further analyzed the differential effects of iSUA and normal umbilical artery in adverse neonatal outcomes. Results showed that the iSUA   Table 3. Meta-analysis for general neonate outcome between isolated SUA and normal umbilical artery. *Indicates WMD.   Table 4. Meta-analysis for adverse neonate outcome between isolated SUA and normal umbilical artery. *Indicates SMD. Delivery < 37 wk: neonates born when the gestational age was shorter than 37 weeks. NICU: neonatal intensive care unit. CS: Cesarean section. ( Table 4). On the contrary, iSUA was not a risk factor for neonatal mortality [OR: 1.749 (95% CI, 0.780-3.925, P = 0.175)] (Supplementary Figure 2).
Assessment of publication bias. The funnel plot, Begg's test and Egger's test were utilized to assess the potential publication bias of included studies. The results indicated no evidence of publication bias for all the subgroup analyses (Tables 2, 3 and 4).

Discussion
SUA is an abnormal condition of the umbilical cord in which one artery is missing. The prevalence of SUA ranges from 0.2% to 11% 1,9,15 . Over the past 30 years, numerous maternal and fetal risk factors were reported to be associated with SUA 1, 4-8 . However, due to small sample size and difference in research subjects, previous studies provided limited information and arrived at controversial conclusions. Therefore, a systematic meta-analysis of all available qualified studies may provide us with definitive answers. Meta-analysis is a powerful approach for investigating the risk factors and neonatal outcomes of iSUA in singleton pregnancy. To date, only 1 meta-analysis publication has focused on the association among iSUA, fetal growth, aneuploidy and perinatal mortality 16 . The results suggested that there was no significant association of iSUA with fetal growth, perinatal mortality or aneuploidy. A large-scale prospective cohort study is needed to reach definitive conclusions on the appropriate work-up in iSUA pregnancies. Thus, we collected a large number of qualified studies for a meta-analysis, aiming to evaluate the risk factors and neonates outcomes of iSUA in singleton pregnancy.
In the current study, we revealed that both maternal status and neonatal sex might be risk factors for iSUA. iSUA itself is a risk factor for Cesarean section and iSUA neonates might have prolonged NICU stay. This result is consistent with the findings of previous reports 1, 6, 14 that suggested that iSUA was associated not only with anomalies at birth and but also with increased risk of adverse pregnancy outcomes. In our meta-analysis, we found that iSUA neonates had a higher rate of preterm birth < 37 weeks, which suggested that iSUA might be a risk factor for premature birth. Similar results have been published in a previous report 6 . However, our results did not suggest that iSUA is a risk factor for neonatal mortality, which is different from the finding in a previous study in which iSUA was shown to be associated with perinatal mortality. This might be due to a much larger sample size in the current meta-analysis. It should be noted that a meta-analysis 17 study on a similar issue was recently published after the submission of our manuscript. This study was focused on evaluating the association of iSUA with pregnancy outcomes and perinatal outcomes. No association between iSUA and pregnancy outcomes was identified. iSUA was found to be correlated only with some perinatal complications in this study 17 .
Despite the clinical significance of our study, there are still some limitations. First, selection bias might exist as all studies included in this meta-analysis were published in English. Studies in languages other than English that may have impact on the evaluation were excluded. Second, although all cases and controls in each study were well-defined following the inclusion criteria, there might be factors that were not taken into account but might influence our results if included. Future analyses including more studies on these risk factors are needed to further confirm our findings.
In this article, we presented a meta-analysis to evaluate the neonatal outcomes and possible risk factors associated with iSUA. Our results suggest that maternal primivalidity and the female sex of neonates might be risk factors for iSUA. Fetuses and neonates with SUA and/or iSUA have increased risk of adverse outcomes. Therefore, the diagnosis of iSUA is necessary during pregnancy and attention should be paid to adverse outcomes associated with iSUA for neonates. As such, surveillance with iSUA would improve neonatal outcomes.

Methods
Data collection. For the first-round exclusion, articles were searched in the NCBI Global Cross-database, including PUBMED, EMBASE and MEDLINE, using "single umbilical artery" or "two umbilical vessels" or "SUA" and ("fetal" or "prenatal") and ("Three-vessel cord" or "Normal umbilical cord" or "control" or "two Umbilical arteries" or "3VD") as key words. Where appropriate, Standard Mean Difference (SMD) or Weighted Mean Difference (WMD) were pooled for the maternal age, gravidity and parity, neonatal birth weight and Apgar score at both one and five minutes postpartum, and odds ratios (ORs) for maternal smoking status, the rate of neonatal delivery before 37 weeks or 34 weeks, Cesarean section (CS) because of fetal distress, the rate of admission to NICU, and the serious adverse neonatal outcome (neonatal mortality to evaluate fetal development as the outcome).
Inclusion and exclusion criteria. Retrospective cohort studies and retrospective case-control studies were considered eligible if the iSUA described in the study was identified by ultrasound in single pregnancy. In other words, our meta-analysis included all studies on singletons with at least 24 weeks' gestation and SUA at birth but without identifiable congenital anomalies 4 . SUA was considered to be isolated if there were no additional structural anomalies and markers for an euploidy or small for gestational age (SGA) at the time of the ultrasound scan 15,16 . Studies were excluded if iSUA was not diagnosed at birth or if the study was only on twin pregnancies. Other exclusion criteria were: presence of any fetal malformation, marker of euploidy by ultrasound examination, chromosomal abnormality established by fetal karyotyping or multi-fetal pregnancy, or absence of a SUA umbilical cord at delivery or by pathological examination 9 .
Possible risk factors included maternal age, gravidity, parity, smoking status, the BMI and neonatal sex. Neonatal data included birth weight, Apgar scores, mode of delivery, rate of preterm delivery (neonatal delivery before 37 weeks or 34 weeks), NICU admissions, and neonatal admission days (length of stay) 11 . Composite adverse outcomes included the following: Cesarean section (CS) because of fetal distress, prolonged neonatal admission and low Apgar score 8 . Neonatal mortality was considered as the serious adverse event.