Prenatal medicine: The child is the father of the man*

To see the world in a grain of sand

And eternity in a wild flower,

To hold the world in the palm of your hand

And infinity in an hour.

William Blake

No period in an individual's life is as critical in determining his or her future health and propensity to disease as the perinatal period. Several epidemiological, animal and clinical studies indicate that insults during the earliest stages of life can profoundly affect our health as adults and cause disease [1]. This realization goes beyond the obvious examples of genetic or congenital diseases. Intrauterine malnutrition, for example, is a significant risk factor for the development of chronic hypertension [2-8], diabetes [9-11], stroke [1],[3] and death from coronary artery disease in adults [1],[3]. Undernutrition during fetal or early postnatal life may produce permanent changes in lipid metabolism 12,13 and in the hemostatic factors [14] that increase the risk for cardiovascular disease. Raised total and low density lipoprotein (LDL) cholesterol have been demonstrated in adults born with a small abdominal circumference [12], and elevated concentrations of fibrinogen and factor VII, both risk factors for cardiovascular disease, are found in men born with short stature and a small abdominal circumference who fail to gain weight during infancy [12]. Some cases of male infertility have their origins in inadvertent intrauterine exposure to environmental estrogens. Moreover, the possibility must be considered that the differences in essential hypertension, diabetes and other disease observed in certain ethnic groups are not due in any meaningful way to genetic predisposition, but are rather the result of differences in the quality of the intrauterine and postnatal environment.

Clearly, the prospect that the quality of life of future generations may be determined to a large extent by the health care delivered during fetal and neonatal life is a daunting prospect with staggering implications for obstetricians, neonatologists and society.

Yet, perinatologists should not be surprised by these findings. Implicit in much obstetric and neonatal practice is the belief that the early identification and correction of fetal or neonatal abnormalities may prevent not only early death but also chronic disease. To be sure, this fundamental tenet is questioned whenever our expectations are not met, as happened recently when the New England Journal of Medicine reported that electronic fetal heart rate monitoring failed to predict cerebral palsy [15-18], a finding widely discussed in the New York Times and other major news agencies. But, these frustrations serve only to underscore the need to rethink our concepts of perinatal disease rather than dismissing the hope of improving adult health perinatal intervention – and ironically, nothing demonstrates this better than cerebral palsy.

Cerebral palsy is a protean motor disorder that is often associated with mental retardation. The simplistic and erroneous view that cerebral palsy had a single cause, intrapartum hypoxia, dates back to 1862 when William John Little first noted an association between ‘spastic rigidity’ and abnormalities of the birthing process [19], which led him to assume that nothing besides abnormalities of birth could cause the clinical picture we now call cerebral palsy [20]. This view was subsequently reinforced by studies in animals [21]. The observed association between asphyxia neonatorum and cerebral palsy, which provided the impetus for the introduction of electronic fetal heart rate monitoring during labor 16,17,[20] and engendered the hope that it would reduce the incidence of cerebral palsy. When these high hopes were repeatedly dashed by randomized clinical trials and epidemiological studies [15-18],22,23, it was only natural to question the adequacy of our technology and the validity of the doctrine that intrauterine conditions may determine health in adulthood rather than admit our hopes were ill-founded to begin with.

One hundred and fifty years later, it is obvious that William John Little overstated his case but less obvious that the right lessons have been learned from it. Cerebral palsy is not, of course, a single condition, but – like so many other obstetric diagnoses – a clinical syndrome that can be caused by many factors during fetal life [24-27], the intrapartum period or even after birth [27]. lntrapartum hypoxia accounts for only a small proportion of cases, probably less than 10%, so it is hardly surprising that electronic fetal heart rate monitoring has not had a detectable impact on the incidence of this syndrome.

Wrong-headed ideas beget wrong technologies, not defective ones. Therefore, as we look forward with hopes that other disorders of pregnancy and labor will be better managed in the future, we must be careful not to repeat the mistakes of the past but learn the lessons that cerebral palsy and our attempts to conquer it with electronic fetal heart rate monitoring had to teach us.

Perinatal medicine is at a stage when we can only recognize clinical syndromes rather than distinct disease entities caused by specific pathological mechanisms. This is true of each of the five conditions that cause most cases of perinatal mortality and morbidity, and is reflected in the terms we have given to them: premature labor/delivery, premature rupture of membranes, small for gestational age, congenital anomalies and pregnancy-induced hypertension. The term ‘preterm labor’, for example, does not tell us whether the condition is caused by an infection, a vascular insult, uterine overdistension, or some other pathological process [28], nor does ‘small for gestational age’ tell us whether a viral infection, a cytogenetic disorder or retarded intrauterine growth has caused a baby to be smaller than it should have been 29,30. If we are to dispel the unreasonable but prevailing hope that a single diagnostic test and treatment will detect and cure each of these conditions, it is essential to recognize that these and other obstetric diagnoses are, like cerebral palsy, essentially ‘syndromic’ in nature.

A second and perhaps even more important concept is that many obstetric syndromes result from adaptive responses of the maternal–fetal unit to pathological insults. Pre-eclampsia, for example, is a response to early alterations in uteroplacental vascular resistance, and the raised blood pressure is a manifestation and not the cause of the ‘disease’[31]. Similarly, premature labor is a mechanism of host defense against intrauterine infection that allows the fetus to exit a hostile environment [32], just as fetal growth retardation is an adaptation to decreased intrauterine availability of nutrients. Once these syndromes are seen in this light, it is easy to understand why antihypertensive therapy has not improved perinatal outcome in pre-eclampsia or why tocolysis has been ineffective in treating preterm labor associated with infection. These therapies are directed at the maternal–fetal responses to disease and not at the disease process itself, and, inasmuch as the responses may be adaptive, intervention may do more harm than good.

A third general feature of most obstetric syndromes is that they are often the late clinical manifestations of chronic pathophysiological processes that have a long subclinical phase. For example, abnormal Doppler waveforms in the uterine arteries [33-37] or abnormal blood pressure responses to angiotensin II [38] can be detected weeks before pre-eclampsia becomes clinically manifest, and ascending uterine infections, rather than being an acute process, may precede the clinical onset of preterm labor or preterm premature rupture of membranes by days or weeks [39-41]. Although this chronicity offers obvious opportunities for early disease detection, we must resist the temptation of rushing headlong into simplistic screening programs in the hope that a single test will pick up most abnormalities early in pregnancy if only we can find the right test.

The multiple etiologies underlying each of these syndromes make it unlikely that any one test will identify all or even most cases. The low prevalence of these disorders will mean that large number of normal women will have to be screened and the false-positive rate of the screening tests will have to be extremely low. However, even with low false-positive rates and high sensitivities, the true positive predictive values of diagnostic tests remain very low if the disease being screened is uncommon enough. The good news is that, if ‘obstetric obstetrical syndromes’ have a long subclinical phase and they are adaptive in nature, then the possibility that their deleterious effects on the fetus/newborn can be prevented is very real [42-45].

At the turn of the century, Ballantyne said that the goals of prenatal care should be to prevent fits, treat maternal diseases, and to identify monsters antenatally 46,47. Obstetricians can look back with satisfaction on the progress they have made in each of these areas, but new challenges lie ahead. First, the time has come to apply what we have learned about the basic mechanisms of development in the fruit fly and rodents to try to understand the make-up of the human embryo and the genesis of congenital anomalies in humans. It is now our task to apply the techniques of developmental biology to perinatology, for this is the only effective way to prevent and treat these disorders.

The second challenge is to define the pathophysiological mechanisms underlying our great obstetric syndromes at the molecular and cellular levels. Only in this way can we hope to develop effective screening programs for the chronic intrauterine diseases that usually manifest themselves clinically in the third trimester, much too late in their natural history for us to alter perinatal outcome. Comprehensive screening programs, eventually based on non-invasive alternatives to the invasive tests we have today, are required to identify mothers and fetuses destined to develop pre-eclampsia, premature labor, placental abruption, premature rupture of membranes and fetal growth retardation. But, they must be identified early enough to allow intervention to prevent not only the clinical manifestation of disease but the long-term handicap it causes.

We are all indebted to Professor Gian Carlo Di Renzo, who was quick to see in these new developments the need to create a forum for the rapid exchange of information in Perinatal Medicine. I am certain that time will tell that Prenatal and Neonatal Medicine, which is the product of his foresight, will have greatly helped us to meet the huge challenge that lies ahead: to implement the practice of Developmental Medicine. I would like, therefore, to acknowledge, on behalf of the Editors and the Editorial Board. Professor Di Renzo's leadership, and to thank him for agreeing to serve as Editor-in-Chief of The Journal of Maternal-Fetal and Neonatal Medicine.