Antenatal and prepregnancy care – prevention of perinatal morbidity and mortality

Pregnancy and childbirth remain the most important events for the survival of mankind. With advances in technology and improvements in lifestyle, mortality from pregnancy has fallen in the more resource rich countries although this unfortunately cannot be said for some of the low income countries with some having as many as one in six women dying from pregnancy and childbirth. It is therefore an inevitable progression that greater emphasis should move from dealing with complications to prevention, early identification and timely interventions. This philosophy has driven most advances in antenatal care. Indeed such significant progress has been made in screening that in some of the more advanced societies, those classified as low risk receive most of their antenatal care in the community. Despite this progress, significant challenges remain with the reliability of various predictive models of pregnancies that require specialized care. Additionally, survival of the at-risk mother and fetus once identified continues to be challenging.

In this themed issue of the journal dedicated to antenatal care, experts address various aspects of antenatal care starting with arguments about shifting the foundation of prevention to the pre-pregnancy period through providing information that could be used to counsel couples with recurrent pregnancy loss on prevention and outcome to prediction of fetuses at risk of significant perinatal morbidity and mortality using various modeling approaches.

A few years ago, Nicolaides [1] proposed the inverted pyramid of antenatal care. Ljubic [2] makes a very sound case for an “extended inverted pyramid” based on the premise that while significant progress has been made in the early detection of aneuploidy and some in the early prediction, prevention and treatment of some complications of pregnancies such as pre-eclampsia, we are a long way from not only predicting but being able to institute appropriate evidence-based interventions that will significantly reduce maternal and perinatal morbidity and mortality from these complications. He makes a very strong case for extending the pyramid to the subcellular level although I would rather refer to this as targeting the pre-implantation window and focusing on the signaling not only between gametes but also between the embryo and pre-implantation endometrium. An understanding of the molecular control of gametogenesis and how we can use modern technology to better select gametes for fertilization and interventions that will improve the quality of the embryo and the receptive endometrium could go a long way to reducing precisely what modern obstetrics has so far failed to achieve. These plausible arguments pose a challenge to the translational scientist and suggests an approach to pre-pregnancy care that is likely to revolutionize obstetric care just as the introduction of ultrasound did in the second half of the 20^th century.

That procreation is complex and fraught with significant risks is evidenced by the fact that about 15–25% of recognized pregnancies end in miscarriages with more than 80% of these occurring in the first trimester. A sub-group of couples suffer recurrent pregnancy loss (RPL) with estimates putting this at 1% [3]. Investigating and managing these couples has remained a major challenge especially in the approximately 50% where no obvious cause is found. Amongst the known causes are parental karyotypic abnormalities (prevalent in 3–4% of couples with RPL) and structural uterine abnormalities. For the former, two options are available to the couple – continuing to try in the hope that a healthy pregnancy will ensue or go for pre-implantation genetic diagnosis (PGD). PGD is not readily available and therefore, providing couples with information on outcomes if they continue to try is paramount to their decision making. Kabessa et al. [4] help in this regard. In a retrospective study of 52 patients with RPL and chromosome abnormalities (CA) and 297 controls they found a successful pregnancy rate following spontaneous conception of 53.8% in the RPL group with CA compared to 68% in the controls. In a similar vein, another study by Gabbai et al. [5] compared the outcome of pregnancies in women with RPL with uterine abnormalities and controls and found an increased pregnancy loss rate of 40% compared to 30.9%. These findings of Gabbai et al. [5] are not surprising except that the pregnancy loss rates are much higher and perhaps a classification based on type of abnormality (minor vs. major) would provide a better index of the implications of these abnormalities on outcome and by inference those who may benefit from interventions.

This year was the 40^th anniversary of the first baby delivered through IVF. Several advances have taken place in the field of assisted reproduction since the birth of Louise Brown in 1978. With the replacement of multiple embryos or single blastocyst is an increased risk of multiple pregnancies. While multiple pregnancies conceived naturally are associated with a higher miscarriage rate, there have been paradoxical observations of lower miscarriage rates in assisted reproductive technology (ART) twin pregnancies compared to ART singleton pregnancies. Several explanations have been suggested. Povoa et al. [6] provide another possible explanation. In a prospective study of 97 (50 singleton and 47 twin) pregnancies they performed ultrasound scans at 8–9 weeks and measured uteroplacental circulation and placental volume. They found the placental volume to be 1.6-fold larger in twin pregnancies who also had lower vascular density and blood perfusion in the intervillous spaces compared to singleton pregnancies. They conclude that these may paradoxically confer reproductive advantage to twin over singleton pregnancies following ART and therefore accounting for the lower miscarriage rate.

For pregnancies that progress beyond the first trimester, it is almost universally accepted that these should be offered morphometry assessment by ultrasound. The benefit of this is the identification of congenital malformations that are lethal, correctable in utero or ex utero. Traditionally in utero therapy/interventions have been very limited. Advances in prenatal imaging have accelerated at an amazing speed with significant improvements in the technology of imaging modalities and a better understanding of the embryological ontogeny. Over the last few years, emphasis has started moving away from diagnosis to in utero interventions and treatments that would improve outcome. These in utero interventions vary from simple drainage procedures to surgery to correct malformations that would otherwise be fatal ex utero, primarily to allow for further in utero development and maturation to improve survival. Nassr and colleagues [7] provide a comprehensive review of the fetal interventional procedures and surgeries that are commonly performed today with some historical perspectives and practical approaches for each of the interventions discussed using their personal experience as a guide. They also discuss the ethical considerations of fetal surgery and argue for the need to balance the benefits of any procedure with the risk to the mothers’ health (morbidity and mortality) and more importantly caution that procedures should be undertaken where there is demonstrable benefit to the fetus. The three criteria that ought to be fulfilled prior to any intervention must include (i) the high probability that the intervention will be life-saving or prevent disability for the fetus and child to be, (ii) the procedure poses low mortality risk and low or manageable risk of serious disease or injury or disability to the fetus and or child to be and (iii) that the risk of any ensuing maternal morbidity is low and that if this occurred it would be manageable. Undoubtedly one of the successes of the last decade is that of laser photocoagulation for twin-to-twin transfusion syndrome. While survival rates from this procedure are higher than those of other interventions which should be obsolete in most high resource countries, it is important to appreciate that survival is not synonymous with absence of handicap. Other interventions reviewed and discussed include various fetal-amniotic shunts for lower urinary tract obstructions and significant pleural effusions. Although fetoscopic endoluminal tracheal occlusion (FETO), the intervention to improve prognosis in congenital diaphragmatic hernia (CDH) has been shown to benefit a significant number of fetuses (70% survival at 12 months in a small sample and in very expert hands), there remains several problems with this technique and it is by and large considered an experimental procedure until more evidence is available. In my view, these unusual procedures should be centralized in supranational centers to allow for upskilling, quality assurance and therefore the proper assessment of its value in CDH.

As our understanding of the life course of some malformations advances, so too will be our approach to interventions in utero. An obvious malformation in this category is congenital pulmonary airway malformation (CPAM) previously referred to as congenital cystic adenomatoid malformation (CCAM). In the early 1990s pregnancies were terminated for the severe forms but it is now obvious that even the severe forms may regress, however, where there are complications such as hydrops, Nassr et al. [7] make the case for drainage to improve survival. A similar argument is made for lower urinary tract obstruction especially for cases of posterior urethral valve. The authors emphasize that while these interventions may apparently in the short term, be beneficial, long-term data may prove otherwise as was shown in the PLUTO trial in the UK [8].

Fetal cardiac intervention in hypoplastic left heart syndrome (HLHS) by way of using aortic balloon valvuloplasty allows the left ventricle to develop with a reported biventricular physiology achieved in about 43% of live born neonates. This is most certainly an exciting development and one whose introduction has been long awaited. It is likely that this will open the way to more in utero interventional cardiology. Although the first open fetal surgery was performed in 1981 by the team of Harrison in Texas, there is now a drive for more endoscopic surgical procedures for fetal malformations that lend themselves to corrective surgery. Irrespective of the various advances, Nassr and colleagues [7] rightly advise that prior to undertaking any interventional procedure, the fetus must have further screening including aneuploidy screening as the outcome of any procedure would be significantly influenced by the karyotype of the fetus.

The final piece of the antenatal jig-saw is monitoring and identification of those babies at risk of intrauterine complications or neonatal morbidity. Saviron-Cornudella et al. [9] from a cohort of 5234 Spanish singleton pregnancies generated customized models using conventional variables such as gestational age, maternal age, maternal body mass index, maternal height, parental height, PAPP-A, free BHCG, single umbilical artery, region of origin, pre-gestational diabetes and showed that using new predictor variables (especially paternal height) predicted ultrasound and birth weight better than the simple model of Gardosi et al. [10]. In fact, with their model (referred to as advanced) they demonstrated an improved prediction of small for gestational age (SGA) and large for gestational age (LGA). Such an improved prediction will result in better monitoring and timing of delivery to reduce morbidity and mortality. An assumption made in their modeling was that an estimate of SGA close to 10% is more precise (i.e. assuming that the prevalence of SGA ought to be 10%).

Shmueli et al. [11] compared sonographic prediction of SGA in fetuses presenting as breech compared to those in vertex presentation using various formulas. They found that irrespective of the formula applied, prediction of SGA and LGA was more accurate in vertex than in breech presentation, whichever formula was used. The difference in predicted weights was not statistically significant. These findings provide some confidence in the clinical application of these formulas, however, caution must be exercised in measuring head circumference in breech presentation as this could be problematic with a dolichocephalic head.

Kipfmueller et al. [12] interestingly sought to improve the prediction of outcome in babies with congenital diaphragmatic hernia by quantifying the pattern of blood flow in the pulmonary artery through measuring the ratio of the time to peak velocity (TPV) to right ventricular ejection time (RVET) and correlating it with the decision for or against extracorporeal membrane oxygenation (ECMO); early mortality, total duration of mechanical ventilation and total duration of oxygen supplementation. In 40 cases of CDH, they showed firstly, feasibility and secondly that these variables correlated well with outcomes. Cut-offs were identified below which the risk of ECMO increased 5.9-fold or there was early mortality, longer duration of mechanical ventilation, oxygen supplementation and a lower survival rate. These data are interesting but there is a need for larger studies to firstly confirm them and then assess their ease of use and applicability in large clinical trials.

One of the most difficult management decisions in obstetrics is that involving twin-pregnancies in which one has a severe congenital malformation. This is a bigger dilemma in monochorionic than in dichorionic twins. Reassuring are the data from a retrospective study of dichorionic twins (one with a malformation) by Algeri et al. [13] of 642 cases (56 with major/minor – 36/20 malformations vs. 586 controls) managed expectantly that showed no difference in outcome. While this is reassuring it would have been useful to investigate the psychological impact of the co-existing malformed fetus on the mother as this is more likely to be a factor considering that often dichorionic twins behave as two separate pregnancies.