Objective
To investigate the use of the fetal cerebroumbilical ratio to predict intrapartum compromise in appropriately grown fetuses.
Study Design
A prospective observational study set at Queen Charlotte’s and Chelsea hospital, London, UK. Fetal biometry and Doppler resistance indices were measured in 400 women immediately before established labor. Labor was then managed according to local protocols and guidelines, and intrapartum and neonatal outcome details recorded.
Results
Infants delivered by cesarean section for fetal compromise had significantly lower cerebroumbilical ratios than those born by spontaneous vaginal delivery (1.52 vs 1.82, P ≤ .001). Infants with a cerebroumbilical ratio <10th percentile were 6 times more likely to be delivered by cesarean section for fetal compromise than those with a cerebroumbilical ratio ≥10th percentile (odds ratio, 6.1; 95% confidence interval, 3.03−12.75). A cerebroumbilical ratio >90th percentile appears protective of cesarean section for fetal compromise (negative predictive value 100%).
Conclusion
The fetal cerebroumbilical ratio can identify fetuses at high and low risk of a subsequent diagnosis of intrapartum compromise, and may be used to risk stratify pregnancies before labor.
The intrapartum period represents the time during pregnancy when the fetoplacental relationship is challenged to the highest degree. Uterine contractions are associated with up to 60% decline in uterine artery flow velocities, and the associated reduction in placental perfusion may precipitate fetal compromise. Fetal hypoxia can result in significant neonatal sequelae including neurologic injury, seizures (neonatal encephalopathy), and death. Although in the majority of infants, the etiology of cerebral palsy or neonatal encephalopathy are in the antenatal period, a significant proportion (14.5%) are associated with intrapartum hypoxia. In term infants, hypoxic-ischemic encephalopathy occurs in 2-3 cases per 1000 live births in developed countries, whereas this incidence is almost 10-fold greater in developing nations. Up to 63% of cases of intrapartum hypoxia occurs in pregnancies with no antenatal risk factors making identification of the fetus at risk of intrapartum complications difficult.
Although in individual cases electronic fetal heart rate monitoring may detect fetal hypoxia, its widespread use has not resulted in a decrease in the incidence of cerebral palsy in developed countries. This may be because electronic fetal heart rate monitoring has a poor positive predictive value and is not being appropriately targeted to an at risk population. Furthermore, the use of continuous electronic fetal heart rate monitoring has resulted in increased rates of cesarean section and instrumental delivery for presumed fetal compromise. Individual risk factor assessment based on maternal history, admission cardiotocography (CTG), or measurement of amniotic fluid index (AFI) may be used to identify a population to whom electronic fetal heart rate monitoring should be applied. However, a recent Cochrane review assessed the use of admission CTG as a screening test for intrapartum fetal compromise and concluded that there was no evidence to support this practice. Similarly, assessment of admission AFI has been demonstrated to be of limited value in identifying fetuses at risk of compromise.
During periods of either acute or chronic hypoxia the fetus preferentially redistributes its cardiac output to perfuse vital organs such as the brain. A decrease in the fetal middle cerebral artery pulsatility index (MCA PI) is suggestive of this cerebral distribution, which is a physiologic response seen in fetuses exposed to hypoxic conditions. These changes can be measured using pulse wave Doppler, with a low middle cerebral artery and a high umbilical artery pulsatility index (low cerebroumbilical [C/U] ratio) indicative of cerebral redistribution.
We aimed to investigate the predictive value of the fetal C/U ratio, measured just before established labor, in identification of fetuses at risk of intrapartum fetal compromise and subsequent obstetric intervention. We hypothesized that those fetuses with a low C/U ratio would be at increased risk of compromise during labor, leading to increased rates of obstetric intervention and emergency delivery.
Materials and Methods
Four hundred women at term (37-42 weeks) were recruited at Queen Charlotte’s and Chelsea Hospital, Imperial College Healthcare NHS Trust, London, UK. All recruited patients were approached before active labor (cervical dilatation ≤4 cm), had singleton pregnancies, were identified as low risk on the basis individual chart review with no fetal concerns identified antenatally. Exclusion criteria included cervical dilatation of >4 cm, multiple pregnancy, preeclampsia, previously identified fetal growth restriction, known fetal anomaly, evidence of intrauterine infection, or maternal age <16 years. We aimed to recruit women likely to deliver within 72 hours. Ethical approval for this study was granted by the London Research Ethics Committee (Ref no: REC 10/H0718/26).
All women had an ultrasound examination performed in a supine position with the head of the bed elevated at 45 degrees. Fetal biometry and AFI were recorded. In addition, the pulsatility index of the umbilical artery and middle cerebral artery was recorded using an automated trace of at least 3 consecutive waveforms. All Doppler waveforms were recorded in the absence of fetal breathing movements or uterine contractions. The angle of insonation was kept <30 degrees. Each parameter was recorded 3 times and a mean of these values used for data analysis. In a subcohort of 50 women, recording of Doppler indices were repeated by the same operator after a 30-minute interval to ascertain intraobserver variability and by a different trained operator to establish the interobserver variability of these parameters. Information about maternal age, ethnicity, parity, booking blood pressure, gestation at onset of labor, body mass index (BMI), history of smoking, preexisting maternal medical disorders, history of previous fetal growth restriction, stillbirth, or neonatal death was documented. All ultrasound examinations were undertaken by a single trained practitioner using a portable ultrasound machine (GE Volusune, 4-8 MHz transabdominal curvilinear transducer; GE, Chalfont St. Giles, UK). Information obtained from the ultrasound examination was not made available to either the clinicians responsible for the patient’s intrapartum care, or the patients themselves, to ensure that management of labor was not influenced by the ultrasound findings.
Labor was then managed as per local protocols and guidelines. After delivery, intrapartum and neonatal outcome data were collected from patient case notes.
Outcome measures for this study included mode of delivery, diagnosis of fetal compromise (based on CTG abnormalities, abnormal fasting blood sugar [FBS] [pH <7.20], or both), presence of meconium stained liquor or CTG abnormalities (classified according to National Institute for Health and Clinical excellence [NICE] guidelines ) and duration of labor before a diagnosis of fetal compromise. Neonatal outcome was assessed by examining birthweight, as well as a composite neonatal outcome score that included cord arterial pH and base excess at delivery, Apgar score at 1 and 5 minutes, and neonatal unit admission.
Data analysis was performed using SPSS version 19 (SPSS, Inc, Cary, NC). Infants were categorized according to mode of delivery and C/U ratio (<10th percentile, 10th-90th percentile, >90th percentile). Statistical analysis included 1-way analysis of variance (ANOVA), χ 2 test, independent sample 2-tailed t tests, and logistic regression.
Results
Four hundred women were recruited to the study over a 1-year period. Patient demographics are reported in Table 1 and mode of delivery details for all patients in Table 2 . No difference in maternal age, BMI, or ethnicity, was observed between the different modes of delivery groups.
| Demographic | Overall | Emergency cesarean fetal compromise | SVD | Instrumental fetal compromise | Instrumental prolonged second stage | Emergency cesarean other | ANOVA/χ 2 P value |
|---|---|---|---|---|---|---|---|
| Number of patients | 400 | 46 (11.5%) | 175 (43.8%) | 71 (17.8%) | 45 (11.3%) | 63 (15.8%) | |
| % primiparous | 65.5% (262/400) | 89.1% (41/46) | 37.1% (65/175) | 87% (62/71) | 80.0% (36/45) | 92.1% (58/63) | < .001 |
| Mean maternal age | 32 (18-47) | 32 | 32 | 33 | 31 | 33 | .38 |
| Mean BMI | 25 (17-42) | 26 | 25 | 24 | 24 | 25 | .05 |
| Ethnicity, % | |||||||
| White | 65.5% (262/400) | 56.5% (26/46) | 66.3% (116/175) | 77.5% (55/71) | 62.2% (28/45) | 58.7% (37/63) | .62 |
| Asian | 16% (64/400) | 19.6% (9/46) | 12.6% (22/175) | 14.1% (10/71) | 20% (9/45) | 22.2% (14/63) | .44 |
| Afro-Caribbean | 13% (52/400) | 15.2% (7/46) | 15.4% (27/175) | 7.0% (5/71) | 6.7% (3/45) | 15.9% (10/63) | .32 |
| Other | 5.5% (22/400) | 8.7% (4/46) | 5.7% (10/175) | 1.4% (1/71) | 11.1% (5/45) | 3.2% (2/63) | .18 |
| Mean gestation of ultrasound scan | 40 wks + 2 d (37 – 42 + 1) | 40 wks + 6 d | 40 wks + 1 d | 40 wks + 3 d | 40 wks + 3 d | 40 wks + 3 d | .04 |
| Neonatal outcomes | |||||||
| Birthweight | 3517 g (1780–4940 g) | 3475 g | 3523 g | 3399 g | 3552 g | 3641 g | .047 |
| Birthweight percentile | 53 (1-99) | 48 | 55 | 45 | 54 | 60 | .01 |
| Apgar <7 at 5 min | 4/400 (1%) | 1/46 (2.2%) | 0/175 (0%) | 1/71 (1.4%) | 1/45 (2.2%) | 1/63 (1.6%) | .49 |
| Cord arterial pH <7.20 | 114/400 (28.5%) | 10/46 (21.7%) | 51/175 (29.1%) | 35/71 (49.3%) | 9/45 (20%) | 9/63 (14.3%) | .002 |
| Neonatal unit admission | 1.5% (6/400) | 2.2% (1/46) | 0.6% (1/175) | 4.2% (3/71) | 0% (0/45) | 1.6% (1/63) | .21 |
| Neonatal encephalopathy | 0% (0/400) | 0% (0/46) | 0% (0/175) | 0% (0/71) | 0% (0/45) | 0% (0/63) | n/a |
| Composite neonatal outcome score | 0.71 | 0.65 | 0.66 | 1.30 | 0.55 | 0.33 | < .001 |
- 1
Apgar = >7 @ 1 min = 0, <7 @ 1 min = 1, <7 @ 5 min = 2
- 2
Cord arterial pH = >7.20 = 0, <7.20 = 1, <7.10 = 2, <7.00 = 3
- 3
Base excess = <−8 = 0, >−8 and <−12 = 1, >−12 = 2
- 4
NNU admission = No = 0, Yes = 1
| 109 cesarean section (27%) | 175 SVD (44%) | 116 instrumental deliveries (29%) | ||
|---|---|---|---|---|
| 46 for presumed fetal compromise (42%) | 63 other indication (58%) | 71 for presumed fetal compromise (61%) | 45 for prolonged second stage (39%) | |
The mean umbilical artery pulsatility index (UA PI) for the entire cohort was 0.80 (range, 0.46−1.53). The UA PI had good intra- and interobserver variability (correlation coefficient of 0.98 and 0.97, respectively).Infants delivered by cesarean section for presumed fetal compromise had the highest mean UA PI, whereas those delivered by instrumental delivery for a prolonged second stage had the lowest mean UA PI (0.86 vs 0.76, P = .002). Infants born by spontaneous vaginal delivery (SVD) had a mean UA PI significantly lower than the cesarean section cohort (0.79 vs 0.86, P = .005). Infants born by instrumental delivery (ventouse or forceps) for presumed fetal compromise had a mean UA PI of 0.82 ( P = .16). The difference in UA PI between the mode of delivery groups was statistically significant when compared using 1-way ANOVA ( P = .009). Receiver-operator curves (ROC) were constructed to evaluate the use of UA PI to predict cesarean delivery for presumed fetal compromise and resulted in an area-under-curve (AUC) of 0.63.
The mean MCA PI for the entire cohort was 1.37 (range, 0.68−2.29). Similarly, the MCA PI had good intra- and interobserver variability (correlation coefficient of 0.97 and 0.95, respectively). Infants delivered by cesarean section for presumed fetal compromise had the lowest mean MCA PI when compared with those delivered by instrumental delivery for a prolonged second stage (1.26 vs 1.41, P = .004) or those born by SVD (1.26 vs 1.41, P = .001). Infants born by instrumental delivery for presumed fetal compromise had a mean MCA PI of 1.32 ( P = .17). When all mode of delivery groups were compared using 1-way ANOVA, the difference in mean MCA PI was statistically significant ( P = .004). ROC curve analysis for the MCA PI as a predictor of cesarean section for presumed fetal compromise resulted in an AUC of 0.64.
MCA PI and UA PI were then used to calculate the C/U ratio. The mean C/U ratio for the entire cohort was 1.77 (range, 0.59−3.15). The C/U ratio, too, had good intra- and interobserver variability (correlation coefficient of 0.97 and 0.94, respectively). Infants delivered by cesarean section for presumed fetal compromise had the lowest mean C/U ratio, whereas those delivered by instrumental delivery for a prolonged second stage had the highest mean C/U ratio (1.52 vs 1.90, P ≤ .001). Infants born by SVD had a mean C/U ratio of 1.83 and those born by instrumental delivery for presumed fetal compromise had a mean C/U ratio of 1.67, both significantly higher than the cesarean section cohort ( P ≤ .001 and P = .04, respectively). When all modes of delivery were compared using 1-way ANOVA, the difference in mean C/U ratio remained statistically significant ( P ≤ .001). ROC curve analysis for the C/U ratio as a predictor of cesarean section for presumed fetal compromise resulted in an AUC of 0.69. The optimal positive discriminatory C/U ratio was 1.24 resulting in a sensitivity of 32.5%, specificity of 93.2%, and a positive predictive value of 36.4% for presumed fetal compromise. Conversely, the optimal negative discriminatory C/U ratio was 2.34 and resulted in a 100% negative predictive value (NPV).
The C/U ratio was normally distributed throughout the study population (Kolmogorov-Smirnov test P = .13). This distribution resulted in a 10th percentile C/U ratio value of 1.24 and a 90th percentile value of 2.35, with a median of 1.76. Cases were subclassified according to C/U ratio percentile group as <10th percentile, 10th-90th percentile, and >90th percentile on the basis of our ROC curve analyses. No significant differences in maternal age, BMI, ethnicity, or parity were observed between the different C/U ratio percentile groups. These results are summarized in Table 3 .
| Demographic | Overall | C/U ratio <10th percentile | C/U ratio 10th-90th percentile | C/U ratio >90th percentile | ANOVA/χ 2 P value |
|---|---|---|---|---|---|
| % primiparous | 65.5% (262/400) | 84% (37/44) | 64.9% (205/316) | 50% (20/40) | .15 |
| Mean maternal age | 32 (18-47) | 32 | 32 | 34 | .16 |
| Mean BMI | 25 (17-42) | 25 | 25 | 24 | .19 |
| Ethnicity, % | |||||
| White | 65.5 (262/400) | 61.4 (27/44) | 66.1 (209/316) | 65.0 (26/40) | .94 |
| Asian | 16.0 (64/400) | 15.9 (7/44) | 16.5 (52/316) | 12.5 (5/40) | .84 |
| Afro-Caribbean | 13.0 (52/400) | 15.9 (7/44) | 12.7 (40/316) | 12.5 (5/40) | .85 |
| Other | 5.5 (22/400) | 6.8 (3/44) | 4.7 (15/316) | 10.0 (4/40) | .38 |
| Mean gestation at time of U/S | 40 wks + 2 d | 40 wks + 5 d | 40 wks + 3 d | 39 wks + 5 d | .004 |
| Evidence of fetal compromise | |||||
| Rate of meconium stained liquor | 10.8% (43/400) | 22.7% (10/44) | 10.1% (32/316) | 2.5% (1/40) | .02 |
| Rate of CTG abnormalities | 35.3% (141/400) | 86% (38/44) | 31.0% (98/316) | 12.5% (5/40) | < .001 |
| Delivery category | |||||
| Cesarean section fetal compromise | 11.5% (46/400) | 36.4% (16/44) | 9.5% (30/316) | 0% (0/40) | < .001 |
| Fetal compromise diagnosed at any time during labor | 29.3% (117/400) | 63.6% (28/44) | 26.6% (84/316) | 12.5% (5/40) | < .001 |
| SVD | 43.8% (175/400) | 22.7% (10/44) | 44.9% (142/316) | 57.5% (23/40) | .04 |
| Vaginal delivery of any kind | 72.8% (291/400) | 52.3% (23/44) | 74.1% (234/316) | 85% (34/40) | .18 |
| Cesarean section other indication | 15.8% (63/400) | 11.4% (5/44) | 16.5% (52/316) | 15% (6/40) | .71 |
| Length of labor before fetal compromise diagnosed | 510 min (0-1817) | 576 min | 493 min | 422 min | .40 |
| Neonatal outcome | |||||
| Birthweight | 3518 g (1780-4940 g) | 3448 g | 3514 g | 3622 g | .23 |
| Birthweight percentile | 53 (1-99) | 47 | 53 | 63 | .04 |
| Apgar <7 at 5 min | 4/400 (1%) | 1/44 (2.3%) | 3/316 (0.9%) | 0/40 (0%) | .57 |
| Cord arterial pH <7.20 | 114/400 (28.5%) | 13/44 (29.5%) | 90/316 (28.5%) | 11/40 (27.55%) | .98 |
| Neonatal unit admission | 1.5% (6/400) | 4.5% (2/44) | 1% (3/316) | 2.5% (1/40) | .19 |
| Neonatal encephalopathy | 0% (0/400) | 0% (0/44) | 0% (0/316) | 0% (0/40) | n/a |
| Composite neonatal outcome score | 0.71 | 0.75 | 0.71 | 0.70 | .96 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
