THE TWO FACES OF PERINATAL EPIDEMIOLOGY

1
THE TWO FACES OF PERINATAL EPIDEMIOLOGY

• Nigel Paneth
• NIH-CHIR course in
• Reproductive and Perinatal Epidemiology
• Woods Hole, MA
• July 14, 2005
• http//www.epi.msu.edu/faculty/paneth.htm

2
WHAT DOES PERINATAL REFER TO?

• Classical definition
• The period from 28 weeks of gestation to age one
  week
• Expanded definition
• The period from 20 weeks of gestation to age one
  month
• My more conceptual definition
• Phenomena that take place from the time the
  organs are formed (fetal life) until the baby has
  adapted to post-natal life (the first few months)

3
THREE KEY CONCEPTS PERTAINING TO THE PERINATAL
PERIOD

• Competition
• In some ways, the fetus/baby and the mother
  compete for resources and for survival. At times
  two or more fetuses may compete with each other
  for resources.
• Adaptation
• The fetus must adapt to labor and to
  extra-uterine life. The mother must adapt to the
  stresses of labor.
• Magnification
• Events taking place in the perinatal period may
  have magnified impacts on later development

4
COMPETITION

• Mother and fetus/baby should ideally both
  survive, but the fetus is in competition with
  mother for nutrients and at times may have
  different metabolic or physiologic priorities.
  Under nearly all circumstances, mothers survival
  will take precedence over the baby. Some fetal
  conditions may reflect the maternal need to
  survive.
• EXAMPLE Below a threshold of 1,500 calories per
  day, preservation of maternal weight takes
  precedence over fetal growth

5
ADAPTATION

• Both mother and fetus must pass through some key
  passages during the perinatal period. Failure to
  adapt to the requirements of these passages will
  lead to death or damage to mother or fetus
• EXAMPLES
• Maternal passage through labor
• Fetal passage through labor
• Fetal adaptation to post-natal breathing

6
MAGNIFICATION

• The fetus grows and develops rapidly, and it is
  possible that exposures occurring at certain
  times will be much more important than those that
  occur at less sensitive periods of development.
• EXAMPLE lack of thyroid hormone during the
  second trimester may permanently damage white
  matter

7
EPIDEMIOLOGICAL IMPLICATIONS OF THESE CONCEPTS

• Stress - The normal fetus undergoes a period of
  extreme asphyxial stress in labor, while the
  delivering mother experiences the most extreme
  hemorrhagic stress of ordinary human experience.
• Risk For both mothers and babies, labor is the
  period of time in life when the risk of dying is
  highest.

8
STRESS

• ASPHYXIA Stanley James and Virginia Apgar at
  Columbia showed in the 1950s that normal labors
  are asphyxiating. Even a very healthy newborn can
  have blood gas values at birth that if found in a
  child or adult would mandate immediate admission
  to intensive care.
• HEMORRHAGE In underdeveloped countries,
  bleeding is the leading cause of maternal death.
  Even in developed countries, 5 or so of mothers
  lose more than a liter of blood in delivery.
• Magann EF et al Postpartum hemorrhage after
  vaginal birth an analysis of risk factors. South
  Med J. 2005 98419-22

9
RISK OF FETAL DEATH IN LABOR

• Intrapartum fetal death rates (death between the
  onset of labor and birth) tend to be around 1 per
  thousand babies entering labor in developed
  countries.
• If the average duration of labor is 12 hours, an
  intrapartum fetal mortality rate of 5/1,000
  fetuses at risk (a rate recently found in rural
  China1), means that the death rate of labor can
  be as high as 1 per day.
• Even at a Canadian tertiary care center, a recent
  intrapartum fetal death rate was found to be
  0.65/1,000.2
• This daily rate of death is not even found at age
  100.
• Wen SW et al J Perinatol. 2004 Feb24(2)77-81.
• Mattatall FM et al Am J Obstet Gynecol. 2005
  May192(5)1475-7

10
ORDINARY LABOR IS AS RISKY TO FETAL LIFE AS A
SEVERE EPIDEMIC/DISASTER SITUATION

• Only the worst epidemic situations come close to
  labor in terms of risk of death.
• In the highest mortality South Darfur town,
  mortality under age five in late 2004 was
  0.59/1,000/day, half the rate of death for
  fetuses in Canadian tertiary care
• Grandesso F et al JAMA. 20052931490-4.

11
MATERNAL MORTALITY

• Maternal mortality rates (most of which are
  related to labor and delivery in undeveloped
  countries) differ more between developed and
  underdeveloped countries than any other mortality
  measure.
• Maternal mortality rates declined more than any
  other population mortality parameter (including
  infant mortality) in Western countries in the
  20th century.
• Maternal mortality rates of 1 per 100 pregnancies
  (common in the US in 1900) are not rare even now
  in some parts of the world.

12
CONCLUSION 1

• The perinatal period is a time of such high risk
  to mothers and babies that it is likely that it
  has had great influence on the human genotype

13
A SPECIAL DIFFICULTY OF PERINATAL EPIDEMIOLOGY
SO MUCH IS HIDDEN FROM VIEW
14
HIDDEN DURING PREGNANCY

• Nearly all of the processes critical to
  competition, adaptation and magnification in
  pregnancy take place where we cannot see them.
  Thus we can only indirectly diagnose fetal
  disease. Often we mistake maternal disease
  manifestations (which we can see) for fetal
  disease (which we cannot see), or measurable
  fetal effects, such as birthweight, for subtler
  processes in the mother/placenta that influence
  birthweight.

15
HIDDEN IN INFANCY

• Even after birth, the infant often cannot
  provide direct information about disease
  manifestations, and the rapid rate of change in
  the fetus/infant further impairs ability to
  diagnose.
• For example, we cannot directly diagnose
  neurodevelopmental disorders until the function
  of interest is capable of manifestation. This is
  why cerebral palsy and mental retardation cannot
  be diagnosed in infancy.

16
CONCLUSION 2

• Because so much is hidden from view in perinatal
  epidemiology, confounding is much more difficult
  to detect than in other epidemiologic fields.

17
WHAT DO I MEAN BY THE TWO FACES OF PERINATAL
EPIDEMIOLOGY?

• The main thing I mean by this term is that
  unlike most of epidemiology which has two poles
  exposure and outcome, the perinatal period is
  both an exposure and an outcome, at times facing
  backwards towards the mother and her history, at
  times facing forward to the child and its future
  development.

18
BUT THERE ARE OTHER DICHOTOMIES NOT SEEN IN OTHER
TYPES OF EPIDEMIOLOGY

• Fetus and the newborn In a few seconds, the
  newborn achieves independent existence after
  spending nine months tied to his or her mothers
  blood supply for food and oxygen
• Placenta and fetus The embryo divides into two
  parts and the placenta then becomes an essential
  support system for the baby. But sometimes the
  support system interferes with fetal development
• A baby and its twin Twins and other multiples
  have their own unique and complicated
  interactions (e.g. t-t transfusion, freemartins)

19
THE MOST IMPORTANT DICHOTOMY

• COMPETITION BETWEEN MOTHER AND CHILD

20
CONCLUSION 3

• For each set of genes that have evolved because
  of perinatal pressures, one must consider whether
  they are designed for optimal fetal or for
  optimal maternal survival, which may be in
  competition.

21
THESE IDEAS LEAD ME TO RAISE A FUNDAMENTAL
QUESTION

• IS THE FETAL ORIGINS HYPOTHESIS POSSIBLY A
  MATERNAL ORIGINS HYPOTHESIS?

22
THRIFTY GENOTYPE

• The concept that we have genes that have evolved
  to make the maximum use of foodstuffs. We thus
  are genetically programmed to gain weight, retain
  fats, and thus be at risk for diabetes and
  cardiovascular disease under conditions of food
  abundance. Concept first proposed in 1962 by
  James Neel of the University of Michigan.
• Neel JV Diabetes mellitus a "thrifty"
  genotype rendered detrimental by "progress"?Am J
  Hum Genet. 1962 Dec14353-62

23
THRIFTY PHENOTYPE

• The concept that fetal malnutrition sets in
  motion processes that create fetal nutritional
  thrift which in turn lead to higher rates of CVD
  in later life. Concept proposed by Hales and
  Barker in 1992. It is another name for the
  fetal origins hypothesis
• Hales CN, Barker DJ Diabetologia.
  199235595-601

24
THRIFTY GENOTYPE? THRIFTY PHENOTYPE?

• OR
• THE NIFTY, SHIFTY MATERNOTYPE?

25
NIFTY SHIFTY MATERNOTYPE

• To maximize a successful maternal and fetal
  outcome of pregnancy, the mother develops several
  components of the cardiovascular risk state.
• I hypothesize that genes that predispose to
  cardiovascular disease originated as genes
  designed either to facilitate the shift of
  resources from mother to fetus/infant during
  pregnancy and lactation or to facilitate optimum
  survival of the mother.

26
WHY SHIFTY AND NIFTY?

• SHIFTY MATERNOTYPE a genotype rich in genes
  that promote the shift of nutrients from mother
  to baby
• NIFTY MATERNOTYPE a genotype rich in genes that
  promote maternal survival
• Both genotypes can predispose to cardiovascular
  disease, but by different mechanisms

27
FETAL ORIGINS

• A corollary of this hypothesis is that some of
  the fetal phenotypes that have been linked to
  adult cardiovascular risk may in fact reflect
  maternal genotypes, seen in the fetus either
  because of shared genetic heritage with their
  mothers, or because of the consequences of the
  maternal genotype on fetal development.

28
TYPICAL CARDIOVASCULAR RISK FACTORS

• Increased body mass index
• Increased fat deposition
• Hyperlipemia, hypertriglyceridemia
• Hyperinsulinism and insulin resistance
• Tendency to thrombophilia

29
THE PREGNANT STATE (ESPECIALLY THIRD TRIMESTER)

• WEIGHT Increase in BMI and fat deposition
• LIPIDS increases in all lipid fractions,
  especially triglycerides
• INSULIN Massive increase in insulin secretion
  and corresponding increase in insulin resistance
  and propensity to diabetes
• COAGULATION Dominance of pro-coagulant state (?
  thrombin, PAI, thromboxane, factors VII, VIII, X,
  and greatly enhanced risk of thrombotic disorders)

30
FAR OR PSEUDOPREGNANT?

• The metabolic syndrome or syndrome X the
  clustering of hypertension, hyptriglyceridemia,
  hyperinsulinemia, fat deposition and hyperlipemia
  is in some sense a partial replication of the
  pregnant state.

31
EXAMPLE OF A SHIFTY MATERNOTYPE AFFECTING INFANT
CVD RISK

• Infants of diabetic mothers are
  characteristically large at birth. Maternal
  processes have been very successful in
  transferring calories to the fetus. Such children
  appear to be at higher risk of obesity and Type
  II diabetes. Some populations, such as Pima
  Indians, are notable for big babies and high risk
  of type 2 diabetes.

32
EXAMPLE OF A NIFTY MATERNOTYPE AFFECTING CVD RISK

• The intense pro-coagulant maternal state in
  pregnancy might promote placental clots that
  interfere with fetal perfusion and produce
  impaired fetal growth. The maternal genes
  promoting coagulation are passed on to the fetus
  for whom they promote CVD risk (as they do in the
  mother). This gives the impression that impaired
  fetal growth leads to CVD.

33
PRE-ECLAMPSIA

• Pre-eclampsia may be a maternal response to the
  fetal drive to obtain more nutrients the
  originary lesion appears to be failure of
  placental trophoblast to convert spiral arteries
  into vasoconstrictor-resistant low pressure high
  flow conduits. This may be an extreme
  illustration of a maternal protective (nifty)
  gene, and it is likely to be found in populations
  with low birthweight and hypertension, the latter
  from an excess of genes promoting
  vasoconstriction.

34
BIRTHWEIGHT

• High birthweights in a population (except perhaps
  in cold climates where infant fat is critical to
  survival) likely represents the shifty
  maternotype, while low birthweight populations
  might represent nifty maternotypes, sacrificing
  fetal growth for maternal survival.
• Populations with high birthweight might suffer
  especially from diabetes with low birthweight,
  especially from hypertension.

35
FINAL THOUGHTS

• Perinatal epidemiology is a very rich and
  exciting field of research with profound
  implications for human health
• It is a relatively small field, and greatly in
  need of young investigators
• New ways of looking at the interactions of
  mothers, babies and placentas open up exciting
  avenues of research