Improving the detection of risk factors for stillbirth

Abstract

Stillbirth is a global tragedy, claiming over 2.6 million lives annually. In Australia, seven in 1000 babies are stillborn - a figure that has remained static over the last three decades. The majority of stillbirths after 24 weeks occur late in pregnancy, after 34 weeks’ gestation, with the risk increasing exponentially at term. This is a time where, if delivery were expedited, a healthy newborn is almost invariable.

Fetal growth restriction due to uteroplacental insufficiency has consistently been identified as a major risk factor for stillbirth. Small-for-gestational-age (<10th centile) fetuses are known to have a three-four fold increased stillbirth risk, yet current antenatal care detects only a minority of these fetuses. Improved antenatal detection of fetuses destined to be born small is recognised internationally as a leading priority to reduce stillbirth risk. Where the fetus is suspected to be small, stillbirth risk is halved, through a combination of increased surveillance and timely delivery.

Fetuses may also be growth restricted yet not small-for-gestational-age. While being small is a leading risk factor for stillbirth, the relationship between birthweight centile and perinatal death is continuous. Fetuses between the 75th and 97th centiles are at the lowest risk for stillbirth, with a stepwise increase in risk at every centile below the 50th. This body of work investigates novel blood-based biomarkers, and applications of existing ultrasound techniques, to better identify the growth restricted fetus.

The first aim of our study was to identify novel biomarkers of term fetal growth restriction. To do so, we collected over 2000 blood samples from pregnant women at 28, and at 36 weeks’ gestation. We systematically screened the 36 week blood samples to identify potential biomarkers that could predict infants destined to be born small-for-gestational-age using a nested case-control approach. We focussed on proteins originating from genes highly expressed in the placenta relative to other tissues. We found four proteins that were significantly altered in the circulation of women destined to birth a small-for-gestational-age infant.

In particular, we identified serine peptidase inhibitor Kunitz type 1 (SPINT1) as a novel biomarker of uteroplacental function. SPINT1 demonstrated better predictive performance (measured by area under the receiver operator characteristic curve) than placental growth factor, the current best biomarker of late pregnancy fetal growth restriction. Importantly, the predictive accuracy of low SPINT1 was unaltered when measured in the entire 2000 samples. We also found SPINT1 levels to be lower in the maternal blood at 28 weeks’ gestation among pregnancies destined to deliver a small-for-gestational-age infant. SPINT1 levels showed a biological gradient across all birthweight centiles, mirroring the observed continuum of perinatal risk. SPINT1 showed significant correlations with birthweight centile, neonatal lean mass, placental weight, and ultrasound measurement of uterine perfusion. Together these data suggest that SPINT1 has potential as a novel biomarker of uteroplacental insufficiency and fetal growth restriction. Our findings suggest that SPINT1 shows stronger correlations with indicators of uteroplacental function than any other biomarker reported to date.

Our second aim was to determine whether we could add to existing ultrasound parameters to improve detection of the growth restricted fetus in late pregnancy. We investigated Doppler ultrasound measurement of maternal, placental and fetal vessels which demonstrate altered resistance in the face of uteroplacental insufficiency. In early-onset fetal growth restriction (<32-34 weeks), ultrasound Doppler parameters tend to deteriorate according to a well-established pattern, beginning with increasing umbilical artery resistance. This is followed by decreasing cerebral vascular resistance and finally abnormal waveforms of precordial veins such as the ductus venosus. Umbilical artery Doppler measurement has been found to reduce the risk of perinatal death in high risk pregnancies. Furthermore, decisions regarding timing of delivery in early fetal growth restriction are well informed by Doppler parameters. In late pregnancy fetal growth restriction (>32-34 weeks) optimal ultrasound assessment of uteroplacental function has not been determined. Routine Doppler ultrasound in low-risk populations has not been found to confer benefit. Moreover, most adverse outcomes in late pregnancy occur in the presence of a normal umbilical artery pulsatility index.

In this study, we interrogated all fetoplacental parameters individually, and in combination, for their ability to identify fetal growth restriction in late pregnancy. We recorded the resistance of the maternal uterine arteries, the umbilical artery, and the fetal middle cerebral artery, aortic isthmus, ductus venosus, and renal arteries. We calculated the well-described cerebroplacental ratio (middle cerebral artery pulsatility index divided by that of the umbilical artery). We also examined the reproducibility of the fetal left myocardial performance index. We first screened each of the vessels’ Doppler measurements for their ability to predict a small-for-gestational-age infant. Mean uterine artery pulsatility index and the cerebroplacental ratio demonstrated the highest areas under receiver operator characteristic curves. This led us to describe for the first time, a novel Doppler combination, the cerebral-placental-uterine ratio (CPUR). This new combination, the cerebroplacental ratio divided by the uterine artery pulsatility index, proved to be advantageous. The CPUR predicted birthweights <10th centile, <5th centile or <3rd centile with greater sensitivity than either of its constituent parameters. It also significantly correlated with all other antenatal and neonatal indicators of uteroplacental function measured – third trimester fetal growth velocity, neonatal body fat percentage and ponderal index. Importantly, CPUR also demonstrated a biological gradient mirroring the continuum of adverse perinatal outcome associated with uteroplacental insufficiency. Together these data suggest the CPUR is a superior ultrasound measure of uteroplacental function, with higher sensitivity for fetal growth restriction in late pregnancy than previously described parameters.

Our third aim was to determine whether reduced third trimester growth velocity is a measurable indicator of uteroplacental insufficiency among appropriate-for-gestational-age fetuses. Although being small-for-gestational-age is a major risk factor for stillbirth, 50% of stillbirths (and 70% of term stillbirths), occur among ≥10th centile fetuses. Low antenatal growth velocity has previously been found to identify fetuses at high risk of adverse outcome among the small-for-gestational-age. The significance of growth velocity among fetuses born appropriate-for-gestational-age was previously unknown.

We measured the change in estimated fetal weight centile, and the change in abdominal circumference centile, over eight weeks in fetuses born appropriate-for-gestational-age. We correlated the change in third trimester fetal growth centiles with antenatal, intrapartum and postnatal indicators of uteroplacental insufficiency. Fetuses that exhibited a fall in estimated fetal weight or abdominal circumference of >30-35 centiles were at increased risk of all indicators of uteroplacental insufficiency. These were: (i) adaptive brain sparing at 36 weeks, as evidenced by an abnormal cerebroplacental ratio; (ii) neonatal acidosis on cord blood samples from fetuses who underwent the hypoxic challenge of labour; and (iii) low neonatal body fat percentage. We suggest that reduced antenatal fetal growth velocity is associated with uteroplacental insufficiency regardless of fetal size. This may allow clinicians to detect a previously unrecognised cohort potentially at increased risk of stillbirth.

In conclusion, this thesis has identified novel blood-based biomarkers and obstetric ultrasound parameters of uteroplacental insufficiency and fetal growth restriction. These are: (i) protein biomarker SPINT1; (ii) the CPUR Doppler combination; and (iii) reduced third trimester fetal growth velocity in appropriate-for-gestational-age fetuses. These measurable indicators each have the potential to be validated and clinically translated for more effective identification of fetuses at risk of late pregnancy stillbirth.