An anthropological perspective on the evolutionary context of preeclampsia in humans
Article Outline
Abstract
Preeclampsia/eclampsia is a dangerous condition unique to humans that is associated with an energetically expensive developing fetal brain and extremely invasive implantation of the trophoblast. We review here the evolutionary history of human pregnancy and childbirth to set a context for evolutionary hypotheses about the origin of preeclampsia. Humans are characterized by having large brains, bipedal locomotion and helpless newborns. These distinctive aspects of our biology arose independently but together constrain pregnancy and childbirth leading to an unusual mechanism of birth, cephalopelvic disproportion, shoulder dystocia, difficult labors, and neonates requiring high levels of parental care. Our cultural adaptation in the form of assistance during childbirth and intensive parental investment make it possible to balance those constraints. Preeclampsia probably arose only after the increase in human brain size and modern human mechanism of birth. Like the other risks of childbirth, preeclampsia also represents a risk associated with these distinctive aspects of human pregnancy and childbirth and is mitigated today by medical intervention. We speculate that, like assistance during childbirth, cultural intervention during pregnancy may extend into the past.
Keywords: Preeclampsia, Human evolution, Paleoanthropology, Childbirth
Preeclampsia, a major complication of human pregnancy worldwide, occurs in 10% of pregnancies, is a common cause of premature birth and is the leading cause of maternal mortality in developed countries (Sibai et al., 2005). In addition to dangers to the mother and child during pregnancy, preeclampsia is associated with lifelong health effects for mothers. Preeclampsia/eclampsia is found only in humans and is related to the evolution of increased cranial capacity in the genus Homo that requires a very deep invasion of the placenta into maternal tissue for adequate oxygen exchange for the fetal brain (Robillard et al., 2003a, Chaline, 2003).
Martin (2003) showed that many other features of human reproduction (including the hemochorial placenta and loss of oestrus) that are often thought to be unique to our species, actually occur in other primates as well. He provided a context for examining the human experience of preeclampsia in light of the general trends of life history parameters and reproductive and sexual strategies of primates. Martin's ‘maternal energy hypothesis’ proposed that intrauterine development of the human brain is constrained by maternal energy resources. He argued that the human reproductive pattern carries with it a number of risks, suggesting that there must be significant reproductive benefits to balance those risks and that humans have evolved compensatory mechanisms to deal with some of those risks. Trevathan, 1987, Trevathan, 1988 has argued that the human universal characteristic of assistance in childbirth (or ‘obligate midwifery’) is just such a compensatory mechanism to deal with some of the risks of parturition. Martin (2003) suggested that the deeply invasive placenta in humans was also a compensatory mechanism (to deal with the energetically expensive fetal brain) and that preeclampsia is a failure of that mechanism. Robillard et al., 2002, Robillard et al., 2003b have also argued that hypertensive disorders of pregnancy are a result of a defect of the normal characteristically human deep invasion of the trophoblast, required to provide nutrients for the growing fetal brain. They proposed that this phenomenon can explain what they called the ‘extravagant’ features of human sexuality including low fertility, concealed ovulation, loss of oestrus, high rate of spontaneous miscarriage and even perhaps long term pair bonding.
Understanding the evolutionary explanation for vulnerability to disease (‘Darwinian Medicine’) often has practical benefits in medical treatment (Williams and Nesse, 1991). Evolution involves tinkering with a preexisting biological system and often results in compromises that carry significant costs in the form of disease susceptibility or other apparent flaws in design.
Here, we examine the fossil record and comparative anatomy of humans and our closest living relatives to understand how humans evolved to give birth to large-brained, helpless young who require great pre-natal resources and who suffer among other risks from preeclampsia. We focus on changes in human anatomy that we can see directly in the fossil record—obviously this does not include the propensity for preeclampsia, but we may make some inferences about that from our reconstruction of other changes in human reproduction.
Humans differ morphologically from our closest relatives by being bipedal, having large brains and giving birth to helpless newborns. These features evolved at different times in our evolutionary history and for different reasons, though each is affected by the others and each has a significant influence on the birth process. Human reproduction places important constraints on the size of the newborn brain. These include cephalopelvic disproportion, difficult and risky labor, and newborns requiring a great deal of cultural buffering or care. Preeclampsia apparently represents another one of those constraints. Regardless of whether the proximate cause of preeclampsia is immunological, vascular, a combination of the two or something quite different, its ultimate cause seems to be related to our deeply invasive placenta which, in turn, is related to our expensive developing fetal brain. Understanding the evolution of those human characteristics that we can examine in the fossil record may suggest some hypotheses about the origin of preeclampsia.
In addition to our anatomical differences from the great apes, humans are unique in that we depend on material culture or tools, and learned social rules for our survival. As part of our cultural adaptation, we communicate through systems of abstract symbols (language), practice rituals and have socially derived rules governing our behavior. We have previously argued that an important part of the human adaptation is assistance during childbirth (Trevathan, 1987, Trevathan, 1988, Rosenberg and Trevathan, 2002). Here, we speculate that this intervention may even extend into pregnancy—for example, to mitigate the risks of preeclampsia.
We know today from fossil evidence that human ancestors were bipedal by at least about 5 million years ago in Africa. These early members of our lineage had a modern pattern of locomotion but had brains that were much smaller than modern humans—in proportion to their body sizes, very similar to those of living chimpanzees. Significant brain expansion began only with the origin of our genus, Homo, about 2.5 million years ago. The timing of the origin of language is controversial but it and other complex symbolic behaviors first occurred much later in our evolution. It is important to emphasize that these human distinctions arose at different times over the course of human evolution, rather than all at once (Wolpoff, 1999). Several of these distinctions impact the way in which humans give birth. Modern human pelvic morphology, especially in females, is a result of a compromise between conflicting constraints imposed by bipedalism and large newborn brains (Lovejoy, 1988). On the one hand, the large newborn crania would pass most easily through a wide, spacious birth canal. On the other hand, a broad pelvis hinders efficient bipedal locomotion making for a waddling gait that swings body weight back and forth across the pelvic breadth at each step and is energetically costly. This compromise creates a situation of intense natural selection and childbirth presents significant risks of injury and mortality to both infants and mothers.
Although humans are not the only primates that have difficulty during childbirth (monkeys are also characterized by a tight fit between the neonatal head and the maternal pelvis) (Fig. 1) (Leutenegger, 1982, Stoller, 1995), three significant differences between the mechanism of birth in humans and other primates make childbirth a unique experience in our species. (1) In humans, the infant head and body generally pass through a series of rotations during birth in response to the close correspondence between infant head and shoulder dimensions and maternal pelvic dimensions. Although there is variation in this process, typically the head enters the birth canal facing sideways and then (2) turns so that the infant usually exits the birth canal in an occiput anterior presentation, (3) like the head, human shoulders fit tightly within the birth canal and present an additional challenge to the birth process (Trevathan, 1988, Trevathan and Rosenberg, 2000).
Fig. 1.
The relationship between the size of the maternal pelvic inlet and the size of the neonatal head in a range of primate species. Transverse diameters of the maternal inlets are held constant and other dimensions scaled relative to that. For each species, the outer oval represents the average maternal pelvic inlet; the black oval represents the average neonatal cranium. Note that in the monkeys and gibbon (Ateles, Nasalis, Macaca and Hylobates) the dimensions of the neonatal cranium are only slightly smaller than the dimensions of the mother's pelvic inlet. In great apes (Pongo, Pan and Gorilla) the pelvic inlet is relatively spacious. In humans, the neonatal cranium is actually longer than the anterior–posterior dimension of the pelvic inlet, requiring the head to enter the inlet facing transversely. Modified from Schultz (1969).
The mechanism of birth and the fetal emergence pattern account for many unique aspects of human maternal behavior (Trevathan, 1987, Trevathan, 1988). Because the human infant typically emerges from the birth canal facing in the opposite direction from its mother, it is difficult for her to reach down as non-human primate mothers (Fig. 2), to clear a breathing passage for the infant or to remove the umbilical cord from around its neck. If a human mother does try to assist in the delivery of her infant by guiding the infant from the birth canal, she risks pulling it against the body's natural curve of flexion, possibly damaging the infant's spinal cord, brachial nerves and muscles.
Fig. 2.
A monkey mother in three stages of childbirth. Note that the infant is delivered facing the same direction as the mother and the mother is able to reach down and guide the baby to her breast as it emerges from the birth canal (Courtesy of Ryoichi Saito).
The human cultural adaptation to this challenge is to seek assistance during birth (Trevathan, 1987, Trevathan, 1988). Today, virtually all women in all societies seek assistance during delivery from relatives, midwives or obstetricians. Although individual women in special circumstances may give birth alone, it is rarely, if ever, a cultural norm. Although the presence of an assistant may put the mother and infant at risk by attracting predators or increasing the chance for infection, at some point in our evolutionary past, the advantages of assistance to mothers and infants outweighed the disadvantages and the species typical pattern of assisted birth emerged. For humans, in contrast to other animals, birth is a social rather than a solitary event.
Another consequence of human brain expansion is a constraint on brain growth during pregnancy. The brain is an expensive organ to maintain, both in utero and after birth. Across mammals, large brains are associated with long gestations (Martin, 1983). In turn, these factors are usually associated with giving birth to relatively precocial young. Most primates fit the model of having long gestation periods, large and developed brains at birth and relatively precocial young. Their newborns are fully furred at birth, with eyes and ears open, capable of some degree of independent movement and generally born as singletons rather than in larger litters.
Human neonates are exceptional because they do not fit the pattern typical of precocial animals. They are like other primates with respect to most of these characteristics, but they are born with a smaller fraction of adult brain size than would be predicted from other life history parameters. Associated with this is a degree of helplessness and a need for parental care that is greater than that seen in most anthropoids. This unusual helplessness has sometimes been described as secondary altriciality (Portmann, 1941). But human infants are not truly altricial, like newborn mice. They are precocial babies born at a relatively early stage in their development. Montagu (1989) referred to the human period immediately after birth as ‘exterogestation’, suggesting that the human newborn continues to function more as a fetus than an infant, but protected by a sort of cultural womb of parents and other caretakers.
The average human gestation length of 38 weeks is not very different from the gestation periods of the great apes—32 weeks for chimpanzees and 38 weeks for gorillas and orangutans, making it unlikely (Martin and MacLarnon, 1990), as some scholars have suggested (Trinkaus, 1984, Gould, 1975), that the gestation length of our smaller bodied ancestors was significantly greater (in absolute or relative terms) than in humans today (Rosenberg, 1992). What is really different about human newborns is the relative size of their brains compared to adult brain size. This is achieved by a shift in the position of birth relative to the rapid period of brain growth. In most primates, birth takes place at about the point when fetal brain growth slows down. In humans, brain growth continues at fetal rates well past birth.
A number of factors constrain human gestation length. Too short a gestation typically results in high-risk low-birth weight infant. Natural selection in our ancestors would have been very strong against small birth weight. A longer gestation results in depleted fat stores and a head too big to pass through the birth canal. But the trade-off is that a full-term human baby is born with only 26% adult brain size (Schultz, 1969) and requires intensification of parenting effort. There may be several advantages to being born when the brain is still undergoing rapid growth, most notably the advantage of exposure to environmental stimuli (such as language) in the extrauterine environment. But newborn human infants would not survive birth for the first portion of postnatal life without assistance from highly motivated caretakers. The more precocial newborn rhesus macaque with 57% of adult brain size at birth can even assist in its own delivery by pulling itself from the birth canal once its hands are free. Even if the mother ignores it, the infant can cling to her ventrum, find the nipple on its own and begin nursing without active assistance from her. Although she may interact with her infant in varying ways, nursing and staying in contact with her are largely up to the infant. This sort of laissez-faire behavior on the part of the mother would greatly reduce survival of the helpless human infant, so a different way of caring for infants has emerged over the past 2 million years since the onset of delayed brain development in the infant (Trevathan, 1990). In fact, this greater reliance on maternal action may go back as far as the evolution of bipedalism, which involved the loss of the ability of a newborn to grasp with all fours to a hairy mother. The result is that although human infants have a wide array of ‘adorable’ behaviors that serve to attract the caretaker and invite interaction, most of the responsibility for maintaining interaction is up to the caretakers. Of course, the more intensified parenting behaviors must continue for several years for human offspring, far exceeding the time and energy investments of non-human primate and other mammalian mothers.
Lovejoy (1981) and Fisher (1989) have suggested that the intensification of parenting effort was one of the factors contributing to the origin of the pair-bond in humans. It is reasonable to argue that human mothers, encumbered with great energetic needs in the last trimester of pregnancy and the first year or more of an infant's life, had more surviving offspring if they were provisioned by adult males and other relatives (e.g., grandmothers). Thus, the demands of full-term helpless infants may mean that in addition to assistance at birth, parenting practices are at the center of hominid survival strategies.
The evolution of the birth process can be inferred from the fossil record that includes pelvic morphology and from our knowledge of the probable size of the infant brain case at birth. The earliest members of our lineage are well represented by fossils from East and South Africa. The australopithecines were shorter than modern humans (about 3.5
ft or 1.07m in stature) and combined a modern bipedal form of locomotion with a relatively small ape-sized brain (Leakey and Walker, 1997, Lovejoy, 1988). Paleoanthropologists have argued that because the australopithecine birth canal had a constant platypelloid shape throughout its length there would have been ‘no bony resistance to fetal descent’ and the neonate would have moved through the birth canal with its head in a transverse orientation (Tague and Lovejoy, 1986). Rotation of the head within the birth canal would have been not only unnecessary but actually impossible, given the short anterior–posterior dimension of each pelvic plane. This mechanism of birth is unlike that known for any living animal and is an example of the mosaic rather than linear (or progressive) nature of the evolutionary process. Australopithecine birth is not halfway between the ancestral pattern for humans and apes and the modern pattern, it is something quite different (Fig. 3). This suggests that neonatal head rotation evolved later in humans as brain size increased, especially in our genus, Homo. The relatively small newborn brain size of these early humans means that selection for a deeply invasive placenta had not probably taken place (and hence it is unlikely that these early humans suffered from preeclampsia).
Fig. 3.
Comparison of the birth mechanism in chimpanzee (Pan), an Australopithecine (A.L. 288-1) and a modern human (Homo). The diagram shows the ‘midwife or obstetrician's’ eye view of a neonatal head passing through the birth canal. In each drawing, the maternal pelvis and neonatal head are shown from below, with the sacrum at the bottom of the picture and the pubic symphysis at the top. Modified from Tague and Lovejoy (1986).
With an increased adult brain size approximately 1.5 million years ago, at the time of Homo erectus, the size of the pelvis and non-rotational mechanism of birth placed a limit on prenatal brain growth. Martin (1983) argued that, like modern humans, these ancestors had babies who were relatively helpless at birth with much brain growth occurring postnatally. Thus, Homo erectus babies were probably quite helpless at birth, and probably would have needed increasingly complex parental care passed on by cultural traditions.
Walker and Ruff (1993) and Ruff (1995) have argued that the evolution of rotational birth occurred sometime after 1.5 million years ago when the birth canal expanded anteroposteriorly. This would have allowed infants to be born with heads that were larger in absolute terms than apes though of course smaller relative to adult head size because of shifts in the timing of birth.
By 300,000 years ago, our ancestors had evolved modern brain size and essentially modern pelvic morphology. It can thus be concluded that these archaic humans probably gave birth much as modern humans do, not only with respect to the mechanism of birth but also in associated behavior, such as the presence of attendants and the state of development of the newborn (Rosenberg and Trevathan, 2002). Whether preeclampsia was present in Homo erectus is uncertain, but it certainly likely that by the time of archaic humans like Neandertals (100,000 years ago) (when brain/body relationships were similar to those of modern humans), our ancestors also experienced some risk of preeclampsia as part of pregnancy and childbirth.
One result of the anatomical changes in the pelvis through human evolution and the associated changes in the ways in which human infants are born is that birth has been transformed from the solitary event that it is for most mammals into a social and cultural event often marked by culturally specific rules and ritual behavior. As bipedalism evolved and brain size increased, natural selection favored the behavior of seeking assistance during birth. As infant helplessness evolved, also as a response to the conflicting selective demands on the pelvis, natural selection would have favored the presence of an assistant. In the long run, seeking another person as an assistant during labor would have reduced mortality and therefore would have been favored by natural selection. Humans have evolved to give birth in a social and cultural context. This suggests that the desire for supportive, familiar and knowledgeable people at birth is deeply rooted in our evolutionary history and that our birth process may have encouraged the formation of intense social bonds between women.
It is the nature of the human adaptation over the course of the last two or 3 million years to find cultural solutions to biological problems. Tools, language, rituals, and complex social systems such as kinship and medical science are all examples of a web of cultural responses to the needs of survival and reproduction. The distinctive way in which human women give birth is also a part of this cultural and biological adaptation and one of the defining features of our humanity.
Perhaps we could extend the argument made by Trevathan, 1987, Trevathan, 1988 about assistance during childbirth to pregnancy. Might it be possible that alert, experienced midwives or other attendants could spot warning signs of preeclampsia and act to minimize the risk of seizure by encouraging women to rest (as they do now) or to induce labor in extreme situations? If it were the case that attendants could recognize warning signs and intervene to reduce maternal mortality in even some cases, that would be enough to improve the reproductive success of women at risk for this condition. As in childbirth, some experience and wisdom on the part of attendants might reduce the mortality of mothers and infants at risk. Of course, this is also true after birth when extremely helpless infants need very intensive parental care.
If preeclampsia is such a significant risk to modern humans, why did natural selection not act to eliminate, or at least minimize, its presence? The risk of eclampsia is one of the constraints on the timing of human birth along with cephalopelvic disproportion and infant helplessness. The balance is obtained at the point where overall reproductive success for the mother and infant are highest (keeping in mind that their interests are sometimes in conflict). The fact that humans are able to tolerate such a high risk of preeclampsia and the other risks associated with pregnancy and childbirth speaks to the benefits that accrue as a result of having a large brain. It is because our adaptation is a cultural one that we benefit so much from being such encephalized creatures, and it is because we are encephalized creatures giving birth to helpless babies that we benefit so much from cultural assistance in the reproductive process.
Natural selection in the last 2 million years has tripled human brain size but at great cost. A larger developing brain is energetically expensive and requires a deeply invasive placenta as well as presenting obstetrical challenges during delivery. On the other hand, babies born at too early a developmental stage are extremely vulnerable, as we know today from experience with extremely small premature infants. Increasing the width of the pelvis to make the birth canal more spacious produces inefficient gait. Evolution is not a global optimization strategy that produces a perfect system and the way humans have babies is surely a superb example of this. Natural selection works as a tinkerer rather than a designer working with the constraints presented by the particular evolutionary context of a species. In this case, those constraints include bipedalism, large brain size and helpless newborns. We are able to balance those constraints in large part because we are cultural animals.
References
- . Increased cranial capacity in hominid evolution and preeclampsia. J. Reprod. Immunol. 2003;59:137–152
- . Evolution of human serial pairbonding. Amer. J. Phys. Anthropol. 1989;78:331–354
- Gould, S.J. 1975. Allometry in primates, with emphasis on scaling and the evolution of the brain. In: Szalay, F. (Ed.), Approaches to Primate Paleobiology, Contrib. Primatol. 5, 244–292.
- . Early hominids fossils from Africa. Sci. Am. 1997;276:74–79
- . Encephalization and obstetrics in primates with particular reference to human voluiton. In: Armstrong E, Falk D editor. Primate Brain Evolution: Methods and Concepts. New York: Plenum Press; 1982;p. 207–222
- . The origin of man. Science. 1981;211:341–350
- . Evolution of human walking. Sci. Am. 1988;259:118–125
- Martin, R.D., Human brain evolution in an ecological context, Fifty-second James Arthur lecture on the evolution of the human brain. American Museum of Natural History, 1983.
- . Human reproduction: a comparative background for medical hypotheses. J. Reprod. Immunol. 2003;59:111–135
- . Reproductive patterns in primates and other mammals: the dichotomy between altricial and precocial offspring. In: DeRousseau CJ editors. Primate Life History and Evolution. New York: Wiley-Liss; 1990;p. 47–79
- . Growing Young. second ed.. Granby, Massachusetts: Bergin and Garvey Publishers; 1989;
- . Die Tragzeiten der Primaten und die Dauer des Schwangerschaft beim Menschen: Ein Problem der vergleichenden Biologie. Revue Suisse de Zoologie. 1941;48:511–518
- . Evolutionary adaptations to pre-eclampsia/eclampsia in humans: low fecundability rate, loss of oestrus, prohibitions of incest and systematic polyandry. Am. J. Reprod. Immunol. 2002;47:104–111
- . Pre-eclampsia and human reproduction. An essay of long term reflection. J. Reprod. Immunol. 2003;59:93–100
- . Preeclampsia/eclampsia and the evolution of the human brain. Curr. Anthropol. 2003;44:130–134
- . The evolution of modern human childbirth. Yearbook Phys. Anthropol. 1992;35:89–124
- . Birth, obstetrics and human evolution. Br. J. Obstet. Gynaecol. 2002;109:1199–1206
- . Biomechanics of the hip and birth in early Homo. Am. J. Phys. Anthropol. 1995;98:527–574
- . The Life of Primates. London: Weidenfeld and Nicolson; 1969;
- . Pre-eclampsia. Lancet. 2005;365:785–799
- Stoller, M., 1995. The obstetric pelvis and mechanism of labor in nonhuman primates. Ph.D. Dissertation. University of Chicago.
- . The obstetric pelvis of A. L. 288-1 (Lucy). J. Hum. Evol. 1986;15:237–255
- . Human Birth: An Evolutionary Perspective. New York: Aldine de Gruyter; 1987;
- . Fetal emergence patterns in evolutionary perspective. Amer. Anthropol. 1988;90:674–681
- . Evolutionary obstetrics. In: Trevathan WR, Smith EO, McKenna JJ editor. Evolutionary Medicine. Oxford: Oxford University Press; 1990;p. 183–207
- . The shoulders follow the head: postcranial constraints on human childbirth. J. Hum. Evol. 2000;39:583–585
- . Neandertal pubis morphology and gestation length. Curr. Anthropol. 1984;25:509–514
- . The reconstruction of the pelvis. In: Walker A, Leakey REF editor. The Nariokotome Homo erectus skeleton. Cambridge: Harvard University Press; 1993;p. 221–233
- . The dawn of Darwinian medicine. Quart. Rev. Biol. 1991;66:1–22
- . Paleoanthropology. second ed.. Boston: McGraw Hill; 1999;
PII: S0165-0378(07)00076-9
doi:10.1016/j.jri.2007.03.011
© 2007 Elsevier Ireland Ltd. All rights reserved.
