Longitudinal change in cerebro-placental ratio (CPR) between 37 and 40 weeks of pregnancy is associated with non-reassuring fetal status and increased risk of cesarean section

Abstract Objective To evaluate in low-risk pregnancies if longitudinal change in cerebro-placental ratio (CPR) between 37 and 40 weeks of pregnancy is associated with cesarean section (CS) for non-reassuring fetal status (NRFS) during labor. Methods This is a prospective observational study of women with singleton low-risk pregnancies who underwent an ultrasound scan at 36 + 0 to 37 + 6 and 39 + 0 to 41 + 6 weeks of pregnancy, when the CPR was calculated from the middle cerebral artery (MCA) and umbilical artery (UA) pulsatility indices. Managing professionals were kept blinded to the Doppler results. The association of the longitudinal change between both CPR (z-velocity) to CS for NRFS was evaluated by logistic regression. Results A total of 401 pregnancies were included. The mean time interval between both CPR evaluations was 21 days (SD 7). A CS for fetal distress was performed in 7% of pregnancies. Independent of the CPR at 37 weeks, the likelihood of CS for fetal distress was significantly decreased by the longitudinal changes from 37 to 40 weeks (OR 0.61, 95%CI 0.4–0.92; p=.018). This association remained significant after further adjustment for potential confounders (nulliparity, maternal weight at booking and estimated fetal weight at 37): (OR 0.64, 95%CI 0.41–0.98; p=.044). Conclusions The longitudinal change of CPR between 37 and 40 weeks is associated with the need for CS for NRFS during labor.


Introduction
Despite the advances in fetal and maternal care, there is still a lack of tests to identify and predict the fetus at risk for adverse perinatal outcome. The role of Doppler is well established in fetal medicine to detect pathological conditions attributable to placental insufficiency. In small-for-gestational age (SGA) fetuses with an early onset and detection, the umbilical artery (UA) has shown to reduce perinatal mortality and morbidity [1] and it is a standard of care in fetal monitoring. In late-onset SGA (growth defect that occurs after 32 weeks of gestation), however, the UA Doppler is not useful for giving us information about placental insufficiency [2]. Under hypoxic conditions, the fetus redistributes its flow toward the brain, the so-called brain sparing. This is evident in the middle cerebral artery (MCA) flow that shows decreased pulsatility at Doppler evaluation. The cerebro-placental ratio (CPR) is the MCA pulsatility index (PI) and the UA PI, and gestation-specific reference centiles have already been described [3][4][5]. The CPR is a predictor of adverse perinatal outcome and it is primarily used for the assessment of fetal wellbeing [6,7]. Late-onset SGA babies with pathological CPR have a 10-fold increased risk of poor perinatal outcome [8], suggesting that the CPR could be used as a useful marker to differentiate constitutional smallness from late fetal growth restriction (FGR). Our group has also reported that fetal Doppler assessment allows predicting those SGA pregnancies at higher risk of cesarean delivery for non-reassuring fetal status (NRFS) [9].
It is implausible that placental insufficiency is only present in babies below the 10th centile at birth [10], especially if we take into consideration the fact that between the 10th and the 50th centile of birth weight perinatal mortality remains increased compared to the babies with an appropriate weight for the gestational age (AGA) >50th centile, suggesting that placental insufficiency could explain part of this mortality. The CPR has been proposed as a marker of adverse perinatal outcome not only in SGA but also in appropriately grown for gestational age (AGA) fetuses [11][12][13][14][15]. There is also evidence that an abnormal CPR at admission for labor is associated with fetal distress [11], supporting the hypothesis that CPR is a latent mirror of placental insufficiency. However, there is no evidence regarding the clinical meaning of CPR changes during the last weeks of pregnancy.
In this study, we sought to investigate to which extent the longitudinal change in CPR between 37 and 40 weeks of pregnancy is associated to cesarean section (CS) for NRFS during labor.

Materials and methods
This is a prospective study of low-risk pregnant women who attended a referral university hospital (Barcelona Hospital Clinic) from 2017 to 2020. These women were primarily included in a randomized controlled trial, the RATIO37 study [16]. Inclusion criteria were: live singleton non-malformed fetus, available first-trimester US dating, maternal age at recruitment !18 years and capacity to give informed consent. Pregnancies were excluded if the following criteria were met after recruitment: abnormal karyotype, structural abnormalities, congenital infections, pregnancy complication, and birth weight below the 10th centile (according to local standards) [17]. All pregnancies were dated according to the 1st-trimester crown-rump length measurement [18].
At 36 þ 0 to 37 þ 6 weeks, all women had Doppler scan evaluation, and those with an estimated fetal weight (EFW) >10th centile according to local standard [17] were invited to have a second scan at 39 þ 0 to 40 þ 6 weeks. The local ethics committee approved the study and all women signed informed consent to participate.

Measurements
All Doppler ultrasound examinations were performed by certified sonographers using either a General Electric Voluson E8 (Boston, MA), Siemens Sonoline Antares (Mountain View, CA), or Aloka alpha 7 (Tokyo, Japan). The CPR was calculated as the simple ratio between the MCA PI and the UA PI [19]. The Doppler measurements were performed adhering to the recommendation by the International Society of Obstetrics and Gynaecology (ISUOG) [20].

Management and outcome definition
Labor management was performed according to local protocols by professionals blinded to the Doppler results. The study outcome was a need for a CS for NRFS. Indication for cesarean delivery for NRFS was based on abnormal fetal heart rate monitoring and abnormal fetal scalp blood pH during intrapartum monitoring. Continuous fetal heart monitoring was carried out, and tracings were classified according to the following three-tiered system: (i) normal, baseline 110-160 beats per minute (bpm), variability >5 bpm and absence of decelerations; (ii) suspicious (one nonreassuring criterion present), baseline 100-109 or 161-180 bpm, variability <5 bpm for less than 90 min, recurrent (>50% contractions) typical variable decelerations for more than 90 min, and a single prolonged deceleration for up to 3 min; or (iii) pathological, more than one non-reassuring criterion or the presence of any abnormal feature, including baseline <100 or >180 bpm or sinusoidal patterns for more than 10 min, variability <5 bpm for more than 90 min, recurrent atypical variable decelerations for more than 30 min, late decelerations for more than 30 min, and a single prolonged deceleration for more than 3 min.
In cases with pathological fetal heart rate or a suspicious pattern not presenting fetal heart rate acceleration after digital fetal scalp stimulation, fetal scalp blood sampling was performed. Levels were considered abnormal with values less than 7.15 or 7.20 on two occasions within 30 min. If cervical conditions did not allow fetal scalp sampling, cesarean delivery was considered for NRFS based on the persistence of abnormal tracings after pessary withdrawal, oxytocin suspension, and 10 min of intravenous infusion of 200 mg/min ritodrine.

Statistical analysis
The CPR measurements were transformed to z-values according to Baschat and Gembruch [19]. Longitudinal CPR changes were quantified by calculating the z-velocity as: (last CPR measurementfirst CPR value)/interval between scans in days. The association between CPR measurements with the occurrence of CS for NRFS was evaluated by logistic regression, adjusted for nulliparity, maternal weight at booking and EFW at the 37-week scan. All statistical analyses and graph constructions were performed using IBM SPSS Statistics (Armonk, NY). p Values <.05 were considered statistically significant.

Results
A total of 407 were invited to participate, of which six failed at the 40-week scan, leaving 401 women for the analysis.
The mean gestational age at first scan was 37.4 weeks (SD 1.2) of gestation. The second scan was performed at 39.6 weeks of gestation (SD 0.9). The mean time interval between both CPR measurements was 3.1 weeks (SD 1).
The characteristics of the study population are shown in Table 1.
The mean gestational age at birth was 39.9 weeks (SD 4). A CS for NRFS was performed in 27 (7%) of the pregnancies. Figure 1 shows the individual trends of CPR z-velocity between both examinations by the occurrence of CS for NRFS.
The proportion of women that needed a CS for NRFS was significantly higher in those with a CPR z-  velocity in the lower quartile (z-velocity less than À0.7): 13/100 (13%) vs. 14/301 (4.7%); p¼.004. Table 2 shows the perinatal outcomes by the CPR z-velocity quartile and the composite adverse outcome (CAPO) is defined as NRFS CS or admission to NICU for more than three days or neonatal acidosis. Table 3 shows logistic regression model for the association with CS for NRFS. For each unit increase of CPR z-velocity from 37 to 40 weeks, there was a 39% independent reduction in the likelihood of CS for NRFS (OR 0.61, 95%CI 0.4-0.92; p¼.018). This association remained significant after further adjustment for potential confounders (nulliparity, maternal weight at booking and EFW at 37): OR 0.64, 95%CI 0.41-0.98; p¼.044).

Discussion
The detection of fetuses at risk for adverse perinatal outcomes remains a challenge. The role of fetal Doppler to detect pathological conditions secondary to placental insufficiency in growth-restricted fetuses has been extensively studied [1,6,8]. However, the same role in normally grown fetuses to capture latent placental insufficiency is controversial [21,22]. Recent studies have demonstrated that low CPR in AGA fetuses is significantly associated with adverse neonatal outcomes [12,23,24], including operative delivery for fetal compromise, NICU admission, stillbirth, and neonatal morbidity. In this study, we have investigated how longitudinal changes in CPR between 37 and 40 weeks of pregnancy are associated with a CS for NRFS during labor.
There is some evidence of the cross-sectional value of CPR measurements in the prediction of adverse outcomes in the general population of pregnancies. In 2013, Prior et al. showed that the CPR could identify fetuses at high risk of a subsequent intrapartum compromise, and suggested that CPR could be used to stratify risk pregnancies before labor [11]. In 2016, Khalil et al. published a retrospective study suggesting that CPR measured at the third trimester could be a predictor of stillbirth and perinatal loss [25]. Later, the same group reported that the combination of EFW, CPR, and uterine Doppler in the third trimester could identify the majority of fetuses at risk of stillbirth [26]. Similar to our study, Sabdia et al. found the CPR zscore at 35-37 weeks of gestation was a good  predictor of adverse perinatal outcome [27] and Flatley et al. suggest that both the individual CPR zscore and the magnitude and direction of change in CPR z-score can identify pregnancies at risk of various adverse perinatal outcomes [28]. In 2017, we reported that in low-risk pregnancies Doppler evaluation at 37 weeks of pregnancy improved the prediction of adverse perinatal outcome although it did not improve the prediction of SGA and FGR [29]. In a recent meta-analysis by Nassr et al. revealed that low CPR was associated with an increased risk of CS for fetal distress (OR 4.49; 95%CI 1.63-12.42), low APGAR score (OR 4.01; 95%CI 2.65-6.08), neonatal unit admission (OR 9.65; 95%CI 3.02-30.85), and neonatal complications (OR 11.00; 95%CI 3.64-15.37) [8]. There is no good-quality evidence, however, about the clinical value of Doppler evaluation in the general population. Our group conducted a multicenter randomized controlled trial (RATIO37 study) evaluating a strategy of labor induction at term based on the CPR [16].
It would be plausible that the change of CPR measurements over the lasts weeks of pregnancy adds predictive value to the cross-sectional evaluation. However, this has been scarcely evaluated. Kalafat et al. reported that longitudinal assessment of Doppler parameters was not useful in improving the detection of stillbirth in SGA pregnancies, as compared with a single-point assessment [30]. In 2011, Oros et al. investigated the role of longitudinal changes in uterine, umbilical, and fetal cerebral Doppler indices in lateonset SGA fetuses. They showed that late-onset SGA fetuses with normal Doppler velocimetry upon diagnosis show progression from 37 weeks' gestation with worsening CPR followed by a decrease in MCA-PI [31]. Khalil et al. described differences in the temporal changes of uterine Doppler between normal pregnancies women who develop hypertensive disease [32]. Indeed, our findings suggest that the longitudinal changes of CPR between 37 and 40 weeks are associated with the need for CS for NRFS during labor.
The strengths of our study are its prospective design in a low-risk cohort of patients attending for routine assessment of fetal wellness, and that both patients and managing professionals were blinded to the results. A strong point of our study favoring its external validity is that the Doppler studies described here can be reliably and non-invasively performed by doctors and midwives using portable ultrasound machines in a bedside way. However, we acknowledge some weak points. First, additional Doppler evaluations could have improved the performance of the longitudinal evaluation by allowing to explore other non-linear trends. Second, our sample size prevented us from exploring the predictive capacity (rather than the association) of the Doppler changes. Also, our sample size did not allow us to evaluate as an outcome neonatal morbidity. We also admit barriers to implementation. It can be argued that not every hospital has an available ultrasound machine to perform a Doppler measurement before delivery. The cost of increasing the number of ultrasounds performed in the third trimester is not negligible.
Further research will be needed to investigate more specifically the extent to which these parameters could be used to predict perinatal morbidity and guide the mode of delivery. These studies should elucidate the best gestational age for assessment and the predictive capacity for short-and long-term morbidity.