- Email: [email protected]
Hypertensive disorders of pregnancy are an evolutionary adaptation to mitigate the reproductive consequences of the human physique
Medical Hypotheses (2006) 67, 796–801
http://intl.elsevierhealth.com/journals/mehy
Hypertensive disorders of pregnancy are an evolutionary adaptation to mitigate the reproductive consequences of the human physique Paul T-Y. Ayuk
*
41 Idbury close, Witney, Oxfordshire OX28 5FE, UK Received 25 March 2006; accepted 3 April 2006
Summary The aetiology of hypertensive disorders of pregnancy remains unknown, despite over 30 years of research. The prevalence and natural history of these disorders and the lack of progress in identifying a cause calls for a radical new approach. It is hypothesised that these disorders arise as a consequence of abnormal maternal regulatory mechanisms. The evolution of the physical characteristics unique to humans (bi-pedal gait and a large brain) resulted in a narrow pelvis and a large head. Such a physique is not conducive to viviparity and caused difficult, prolonged and obstructed labour with post-partum haemorrhage – the commonest causes of maternal mortality in the absence of modern medical care. In such circumstances, up to 6.5% of pregnant women will die as a direct consequence of pregnancy, mainly as a result of obstructed labour and haemorrhage. The death toll would have been much higher over millions of years of evolution. These conditions exerted significant adaptive and evolutionary pressure on our species. The adaptations necessary to mitigate the reproductive consequences of the human physique include activation of the coagulation system to reduce post-partum haemorrhage, increased blood pressure to peak after delivery and maintain cerebral perfusion in the face of post-partum blood loss and restriction of fetal growth to prevent obstructed labour. These adaptations must be regulated to guarantee their occurrence but limit their extent to prevent disease. Evidence for blood pressure regulation during pregnancy and a proposed mechanism to achieve this are presented. Regulation requires a redundant feto-placental signal and a single tightly controlled regulator. To guarantee that blood pressure rises, the feto-placental signal is predicted to be conveyed by several different molecules and to be produced in excess in all pregnancies. Normality is then maintained by a single tightly controlled regulator. This model predicts that the feto-placental factors that cause a rise in maternal blood pressure are multiple and produced in disease-causing concentrations in all pregnancies. Disease arises as a consequence of abnormalities in the maternal regulatory mechanism as occurs in say gestational diabetes mellitus. The search for a placental cause for pre-eclampsia is therefore futile. Research should focus on normal pregnancy and the identification of the factor that regulates maternal blood pressure in the second half of pregnancy. This factor will cause hypotension and prevent endothelial activation and have a role analogous to insulin in the regulation of glycaemia and the development of gestational diabetes mellitus. c 2006 Elsevier Ltd. All rights reserved.
* Tel.: +44 1993 778 509. E-mail address: [email protected].
0306-9877/$ - see front matter c 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2006.04.027
Hypertensive disorders of pregnancy are an evolutionary adaptation
Introduction The aetiology of hypertensive disorders of pregnancy remains unknown, despite decades of extensive research principally focused on the placenta. In the UK, and most other industrialised countries, hypertensive disorders of pregnancy are the second commonest cause of maternal mortality. Given the apparent adverse maternal consequences of hypertension, several survival advantages have been proposed, the commonest being a fetal survival advantage [1]. Maternal hypertension is proposed as an attempt by a poorly nourished fetus to increase its supply of nutrients by increasing the resistance of its mother’s peripheral circulation [1]. It is argued that the evolutionary price in terms of maternal mortality and morbidity seems to be worth paying to improve fetal survival [2]. A major flaw in this argument, however, is that fetal nutrition cannot be improved by increasing maternal peripheral resistance (including resistance in the utero-placental circulation) as this is invariably associated with decreased rather than increased perfusion. Furthermore, any adaptation that sacrifices the mother, or even causes disease in the mother in order to save the fetus is a suicidal adaptation as the fetus or neonate cannot survive maternal death. The causes of maternal mortality and morbidity in industrialised countries are an artefact of modern medical care and cannot be used to make accurate predictions of the evolutionary pressures and survival advantages engendered by specific adaptations. There are, however, parts of the world where there is little or no recognisable modern medical care but from where accurate data are obtainable. These data give an indication of the causes of maternal mortality and the evolutionary pressure faced by our species hundreds of thousands, or millions of years ago. This is the appropriate context within which predictions of survival advantage should be made. In Afghanistan, for instance, UNICEF and the US Centre for Disease Control and Prevention estimated maternal mortality rates of 1600 (1100–2000) per 100,000 live births [3] rising to 6500 (5000–8000) per 100,000 live births in remote areas [3]. The commonest causes of death were obstetric haemorrhage and obstructed labour (as opposed to venous thrombo-embolism and hypertensive disorders in industrialised countries), in keeping with the major causes of maternal mortality in the absence of modern medical care in other parts of the world. This study found that if a neonate’s mother died, the child only had a 25% chance of surviving till
797
the age of 1 year [3] and will almost certainly not have survived at all a million years ago. To put this into context, if 6.5% of pregnant women today died as a direct result of pregnancy, many, if not most competent women would regard pregnancy as too dangerous and opt not to reproduce with dire consequences for the species. A review of the veterinary literature indicates that obstructed labour and obstetric haemorrhage do not exact such a heavy toll on other species and the toll on the human species is likely to have been much higher thousands of years ago. These data reveal the scale of human reproductive wastage in the absence of modern medical care and give us an idea of the evolutionary pressures faced by our ancestors. For a species, loss of a fetus is a minor set-back, while loss of a neonate or an infant is a substantial loss. Loss of a mature individual who has survived the perils of intra-uterine life, labour, delivery and childhood; grown into adulthood to find a mate and conceive, if widespread and persistent would threaten the very survival of the species. In the context of high maternal, neonatal and infant mortality rates, it is inconceivable that an adaptation to improve fetal survival that causes more maternal mortality and morbidity (as currently proposed for pre-eclampsia) could have been selected. The proposition that hypertension in pregnancy is an adaptation to increase utero-placental perfusion and improve fetal survival is therefore illogical and not supported by the evidence that utero-placental perfusion is actually reduced in hypertensive pregnancies. The most efficient, and probably the only mechanism of ensuring fetal, neonatal and infant survival and therefore reproductive success is to ensure maternal survival.
Hypothesis Hypertensive disorders of pregnancy are an evolutionary adaptation to protect the human female, and therefore, the species from the risk of reproductive failure wrought by the human physique, the main dangers being obstetric haemorrhage and obstructed labour.
Evolution of the human physique Although estimates vary, the adaptations that set humans apart from other species (bi-pedal gait and a big head relative to body size) evolved 1–5 million years ago [4–6]. For an efficient transfer of forces necessary for bi-pedal gait, our species
798
Ayuk
evolved a narrow pelvis relative to height, bringing the legs close enough to prevent waddling. The evolution of a big head and a narrow pelvis is not an obvious recipe for viviparous success. The consequences are prolonged and obstructed labour with post-partum haemorrhage – the major causes of maternal mortality in the absence of medical care and therefore, the most likely reproductive challenges faced by our species over millions of years. The mechanism of parturition in humans, for instance, is strikingly different from that of our ape relatives. In the apes, the fetus emerges in the occipito-posterior position. This allows the mother to assist herself by pulling the fetus out of the birth canal, flexing the neck onto her torso while the neonate assists by climbing onto the mother’s abdomen. Delivery of a big head through a relatively narrow pelvis required a range of adaptations including changes in the mechanism of labour and expulsion of the fetus in the occipito-anterior position. This removed any possibility of the woman assisting herself as the most efficient line of traction would have been away from her trunk and impossible to achieve.
Adaptations and their implications The necessary adaptations to mitigate the reproductive consequences of the human physique become obvious. Firstly, activation of the coagulation system to prevent excessive blood loss after delivery. Secondly, a rise in blood pressure to peak after delivery and reduce the risk of post-partum hypotension and collapse. Blood pressure is more important than blood volume in maintaining cerebral perfusion and consciousness and the relative reduction in intra-vascular volume in hypertensive pregnancies would not pose a significant disadvantage. The limited value of blood volume can be noted in the observation that blood pressure falls in the first half of pregnancy (Fig. 1) despite a substantial rise in blood volume. Thirdly, restriction of fetal size by restricting utero-placental perfusion and fetal growth and/or by relatively early delivery thus reducing the risk of prolonged and obstructed labour. Finally, the almost universal practice of midwifery or assisted birth [7] would have replaced self-assistance found in apes. The first three adaptations would have been so critical to maternal survival that any individual that did not manifest these would have very little chance of surviving child birth. A signal from the pregnancy is needed to initiate any such adaptation and the maternal response has to be regulated if the adaptation is to be sufficient
Figure 1 (A) Changes in maternal blood pressure with gestation age in women at term with normal pregnancies. Data were collected from the woman’s antenatal records. (B) Rate of change of blood pressure with gestation age in the second half of pregnancy. There was a significant increase in systolic (j 0.51 ± 0.18 mm Hg/week; p = 0.005), mean arterial (. MAP: 0.63 ± 0.13 mm Hg/week; p < 0.0001) and diastolic (m 0.69 ± 0.13 mm Hg/week;
to provide benefit without being excessive and causing disease or death. For this, nature is likely to have selected the simplest and most efficient regulatory mechanism.
The signal The signal for maternal adaptation (a rise in blood pressure in this case) will come from the pregnancy (feto-placental unit). To guarantee that the maternal response occurs, there will have to be substantial redundancy in the signal. This will be achieved by ensuring that the signal is conveyed by a variety
Hypertensive disorders of pregnancy are an evolutionary adaptation of molecules (rather than by a single molecule) and that at any point in pregnancy, the signal is produced in excess of what is needed, removing the need for up/down regulation. An example of this is the regulation of maternal glucose concentration during pregnancy. Glucose is vital for fetal survival and growth and the feto-placental unit produces a range of molecules that increase maternal glucose concentration. These signals are produced in excess of what is needed and in sufficient concentration to cause maternal disease in the absence of a regulated maternal response. As a consequence, most, if not all women will develop gestational diabetes mellitus by term if they do not increase insulin production during pregnancy.
The regulator With a redundant signal, all that is required for efficient control is a single highly efficient and tightly controlled regulator, insulin in the case of glucose regulation. Abnormalities can arise in four ways: overproduction of or hypersensitivity to the regulator (causing hypoglycaemia or hypotension for glucose and blood pressure regulation respectively, the opposite of the desired effect with substantial survival disadvantages and would have been de-selected). Underproduction of or resistance to the regulator on the other hand, would have resulted in hyperglycaemia or hypertension for glucose and blood pressure regulation, respectively. Given that the initial intention of the adaptation was to raise blood glucose concentration or blood pressure, these latter abnormalities are likely to persist in the species. Thus hypertensive disorders, like hyperglycaemic disorders of pregnancy, will arise as a consequence of failure of the regulatory process rather than an excess of, or an abnormal factor produced by the feto-placental unit.
Evidence for blood pressure regulation in pregnancy Fig. 1 shows the antenatal blood pressure profile of 11 women with an uncomplicated first pregnancy at term. These data demonstrate what is already well known but poorly understood. Blood pressure falls in the first half of pregnancy despite an increase in maternal blood volume and cardiac out-put. This is a consequence of the development of the low resistance utero-placental circulation and the production of a variety of vasodilators (such as oestrogens and progesterone) from the feto-placental
799
unit and the effect of these molecules on vascular smooth muscle and endothelium. The fall in blood pressure occurs despite the activation of the rennin–angiotensin–aldosterone system in early pregnancy. This trend, if continued into the second half of pregnancy, would have profound survival disadvantages. The pregnant female has to stand up in order to collect food, escape from predators or defend herself and postural hypotension with decreased mobility in late pregnancy would have conferred a significant disadvantage. In the second half of pregnancy, there is a gradual but steady rise in blood pressure, despite the persistence of the utero-placental circulation and increasing concentrations of oestrogens, progesterone and other potential vasodilators. Diastolic blood pressure rose by 0.69 ± 0.13 mm Hg/week (p < 0.0001, n = 11; Fig. 1B). The factors responsible for this rise in blood pressure are very poorly understood. The rise is more marked in diastolic pressure, indicating that it is achieved by an increase in peripheral resistance while the gradual increase indicates that this is a regulated rather than a random process. The development of hypertension becomes almost inevitable in such a system, while the survival disadvantages associated with hypotension would have resulted in the elimination of this trait. It is unlikely that the development of hypertensive disorders in pregnancy will be understood without an understanding of the regulation of blood pressure during the second half of normal pregnancy.
Implications for pre-eclampsia With respect to blood pressure regulation, therefore, the signal that raises blood pressure during the second half of pregnancy (Fig. 1B) is feto-placental in origin and has considerable redundancy in that it is conveyed by several molecules and is produced in excess at all points in all pregnancies. In the absence of maternal regulatory mechanisms therefore, most, if not all pregnant women will develop hypertension by term. Hypertensive disorders arise not as a result of abnormal factors produced by the placenta but because of abnormalities in the regulatory mechanism in the mother. The implication is that a placental cause for pre-eclampsia will not be identifiable. A placental causative factor for pre-eclampsia is as likely to exist as a placental causative factor for gestational diabetes mellitus. If such a factor exists, it is produced in disease-causing concentrations in both normal pregnancies and pregnancies complicated by pre-eclampsia. Consequently, the
800 factor is unlikely to be identified by the current approach of case-control studies which relies on statistically significant differences between groups. On the other hand, blood pressure regulation, like glucose regulation, is likely to be achieved during pregnancy predominantly by a single molecule and abnormalities in these regulatory mechanisms cause disease. Gestational diabetes mellitus, for instance, is not caused by any placental factor but by maternal insulin resistance which may be detectable even before pregnancy. The non-pregnant woman encounters glucose challenges on a regular basis and it is therefore efficient to employ an identical mechanism for the regulation of glucose concentration during pregnancy. Pregnant women, however, face a unique set of circumstances with respect to blood pressure regulation and the risk of heavy blood loss following delivery. The regulatory response is therefore likely to be mediated by a factor unique to pregnancy. The focus of pre-eclampsia research has thus far been on the identification of the placental factors that cause endothelial activation. The hypothesis here is that there are several such factors and indeed several candidates have been identified. The discovery of a placental cause of endothelial activation has as much implication on the understanding of hypertensive disorders as the discovery of say human placental lactgen would have had on the understanding of the cause of gestational diabetes mellitus. The second hypothesis is that there is a single main regulator of blood pressure in the second half of pregnancy
Ayuk and abnormalities in this regulatory mechanism result in hypertension. It is widely acknowledged that hypertensive disorders are at the extreme of a normal adaptation in pregnancy and therefore very similar to gestational diabetes mellitus in this respect. It is unlikely that the cause of such disorders will be discovered without understanding how normality is achieved and maintained in the first place. The search for the placental cause of the disorder, on the other hand, is futile. The key to our understanding of pre-eclampsia is proposed to be a molecule that prevents endothelial injury and decreases blood pressure, a molecule with an analogous role to that of insulin in the regulation of maternal glycaemia and the development of gestational diabetes mellitus. Fig. 2 shows a schematic representation of the proposed evolution of hypertensive disorders of pregnancy. It is interesting that the major challenges of modern obstetrics – venous thrombo-embolism, pre-eclampsia and fetal growth restriction are the extremes of adaptations that saved our species from reproductive failure and extinction. The development of operative delivery, blood transfusion and use of utero-tonic agents has now made such adaptations disadvantageous but are too recent to reverse millions of years of evolution. In the absence of modern medical care, the risk of maternal mortality and morbidity from hypertensive disorders is small when considered in the context of the risks posed by obstetric haemorrhage and obstructed labour. In this context therefore, maternal mortality and morbidity from
Figure 2 Schematic representation of the evolution of hypertensive disorders of pregnancy. IUGR (intra-uterine growth restriction).
Hypertensive disorders of pregnancy are an evolutionary adaptation hypertensive disorders of pregnancy, like morbidity and mortality from venous thrombo-embolism, would be a price worth paying.
References [1] Haig D. Genetic conflicts in human pregnancy. Q Rev Biol 1993;68:495–532. [2] Redman CW, Sargent IL. Latest advances in understanding preeclampsia. Science 2005;308:1592–4.
801
[3] Bartlett Linda A, Mawji Shairose. Maternal mortality in Afghanistan: magnitude, causes, risk factors and preventability; 2002. Available from: http://www.afghana.com/ Articles/maternalmortalityafghanistan.doc. [4] Rosenberg K, Trevathan W. Bipedalism and human birth: the obstetrical dilemma revisited. Evol Anthropol 1996;4: 161–8. [5] Rak Y. Lucy’s pelvic anatomy: its role in bipedal gait. J Hum Evol 1991;20:283–90. [6] Tague R, Lovejoy C. The obstetric pelvis of A.L. 288-1 (Lucy). J Hum Evol 1986;15:237–55. [7] Trevathan Wenda R. Human birth: an evolutionary perspective. New York: Aldine de Gruyter; 1987.
http://intl.elsevierhealth.com/journals/mehy
Hypertensive disorders of pregnancy are an evolutionary adaptation to mitigate the reproductive consequences of the human physique Paul T-Y. Ayuk
*
41 Idbury close, Witney, Oxfordshire OX28 5FE, UK Received 25 March 2006; accepted 3 April 2006
Summary The aetiology of hypertensive disorders of pregnancy remains unknown, despite over 30 years of research. The prevalence and natural history of these disorders and the lack of progress in identifying a cause calls for a radical new approach. It is hypothesised that these disorders arise as a consequence of abnormal maternal regulatory mechanisms. The evolution of the physical characteristics unique to humans (bi-pedal gait and a large brain) resulted in a narrow pelvis and a large head. Such a physique is not conducive to viviparity and caused difficult, prolonged and obstructed labour with post-partum haemorrhage – the commonest causes of maternal mortality in the absence of modern medical care. In such circumstances, up to 6.5% of pregnant women will die as a direct consequence of pregnancy, mainly as a result of obstructed labour and haemorrhage. The death toll would have been much higher over millions of years of evolution. These conditions exerted significant adaptive and evolutionary pressure on our species. The adaptations necessary to mitigate the reproductive consequences of the human physique include activation of the coagulation system to reduce post-partum haemorrhage, increased blood pressure to peak after delivery and maintain cerebral perfusion in the face of post-partum blood loss and restriction of fetal growth to prevent obstructed labour. These adaptations must be regulated to guarantee their occurrence but limit their extent to prevent disease. Evidence for blood pressure regulation during pregnancy and a proposed mechanism to achieve this are presented. Regulation requires a redundant feto-placental signal and a single tightly controlled regulator. To guarantee that blood pressure rises, the feto-placental signal is predicted to be conveyed by several different molecules and to be produced in excess in all pregnancies. Normality is then maintained by a single tightly controlled regulator. This model predicts that the feto-placental factors that cause a rise in maternal blood pressure are multiple and produced in disease-causing concentrations in all pregnancies. Disease arises as a consequence of abnormalities in the maternal regulatory mechanism as occurs in say gestational diabetes mellitus. The search for a placental cause for pre-eclampsia is therefore futile. Research should focus on normal pregnancy and the identification of the factor that regulates maternal blood pressure in the second half of pregnancy. This factor will cause hypotension and prevent endothelial activation and have a role analogous to insulin in the regulation of glycaemia and the development of gestational diabetes mellitus. c 2006 Elsevier Ltd. All rights reserved.
* Tel.: +44 1993 778 509. E-mail address: [email protected].
0306-9877/$ - see front matter c 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2006.04.027
Hypertensive disorders of pregnancy are an evolutionary adaptation
Introduction The aetiology of hypertensive disorders of pregnancy remains unknown, despite decades of extensive research principally focused on the placenta. In the UK, and most other industrialised countries, hypertensive disorders of pregnancy are the second commonest cause of maternal mortality. Given the apparent adverse maternal consequences of hypertension, several survival advantages have been proposed, the commonest being a fetal survival advantage [1]. Maternal hypertension is proposed as an attempt by a poorly nourished fetus to increase its supply of nutrients by increasing the resistance of its mother’s peripheral circulation [1]. It is argued that the evolutionary price in terms of maternal mortality and morbidity seems to be worth paying to improve fetal survival [2]. A major flaw in this argument, however, is that fetal nutrition cannot be improved by increasing maternal peripheral resistance (including resistance in the utero-placental circulation) as this is invariably associated with decreased rather than increased perfusion. Furthermore, any adaptation that sacrifices the mother, or even causes disease in the mother in order to save the fetus is a suicidal adaptation as the fetus or neonate cannot survive maternal death. The causes of maternal mortality and morbidity in industrialised countries are an artefact of modern medical care and cannot be used to make accurate predictions of the evolutionary pressures and survival advantages engendered by specific adaptations. There are, however, parts of the world where there is little or no recognisable modern medical care but from where accurate data are obtainable. These data give an indication of the causes of maternal mortality and the evolutionary pressure faced by our species hundreds of thousands, or millions of years ago. This is the appropriate context within which predictions of survival advantage should be made. In Afghanistan, for instance, UNICEF and the US Centre for Disease Control and Prevention estimated maternal mortality rates of 1600 (1100–2000) per 100,000 live births [3] rising to 6500 (5000–8000) per 100,000 live births in remote areas [3]. The commonest causes of death were obstetric haemorrhage and obstructed labour (as opposed to venous thrombo-embolism and hypertensive disorders in industrialised countries), in keeping with the major causes of maternal mortality in the absence of modern medical care in other parts of the world. This study found that if a neonate’s mother died, the child only had a 25% chance of surviving till
797
the age of 1 year [3] and will almost certainly not have survived at all a million years ago. To put this into context, if 6.5% of pregnant women today died as a direct result of pregnancy, many, if not most competent women would regard pregnancy as too dangerous and opt not to reproduce with dire consequences for the species. A review of the veterinary literature indicates that obstructed labour and obstetric haemorrhage do not exact such a heavy toll on other species and the toll on the human species is likely to have been much higher thousands of years ago. These data reveal the scale of human reproductive wastage in the absence of modern medical care and give us an idea of the evolutionary pressures faced by our ancestors. For a species, loss of a fetus is a minor set-back, while loss of a neonate or an infant is a substantial loss. Loss of a mature individual who has survived the perils of intra-uterine life, labour, delivery and childhood; grown into adulthood to find a mate and conceive, if widespread and persistent would threaten the very survival of the species. In the context of high maternal, neonatal and infant mortality rates, it is inconceivable that an adaptation to improve fetal survival that causes more maternal mortality and morbidity (as currently proposed for pre-eclampsia) could have been selected. The proposition that hypertension in pregnancy is an adaptation to increase utero-placental perfusion and improve fetal survival is therefore illogical and not supported by the evidence that utero-placental perfusion is actually reduced in hypertensive pregnancies. The most efficient, and probably the only mechanism of ensuring fetal, neonatal and infant survival and therefore reproductive success is to ensure maternal survival.
Hypothesis Hypertensive disorders of pregnancy are an evolutionary adaptation to protect the human female, and therefore, the species from the risk of reproductive failure wrought by the human physique, the main dangers being obstetric haemorrhage and obstructed labour.
Evolution of the human physique Although estimates vary, the adaptations that set humans apart from other species (bi-pedal gait and a big head relative to body size) evolved 1–5 million years ago [4–6]. For an efficient transfer of forces necessary for bi-pedal gait, our species
798
Ayuk
evolved a narrow pelvis relative to height, bringing the legs close enough to prevent waddling. The evolution of a big head and a narrow pelvis is not an obvious recipe for viviparous success. The consequences are prolonged and obstructed labour with post-partum haemorrhage – the major causes of maternal mortality in the absence of medical care and therefore, the most likely reproductive challenges faced by our species over millions of years. The mechanism of parturition in humans, for instance, is strikingly different from that of our ape relatives. In the apes, the fetus emerges in the occipito-posterior position. This allows the mother to assist herself by pulling the fetus out of the birth canal, flexing the neck onto her torso while the neonate assists by climbing onto the mother’s abdomen. Delivery of a big head through a relatively narrow pelvis required a range of adaptations including changes in the mechanism of labour and expulsion of the fetus in the occipito-anterior position. This removed any possibility of the woman assisting herself as the most efficient line of traction would have been away from her trunk and impossible to achieve.
Adaptations and their implications The necessary adaptations to mitigate the reproductive consequences of the human physique become obvious. Firstly, activation of the coagulation system to prevent excessive blood loss after delivery. Secondly, a rise in blood pressure to peak after delivery and reduce the risk of post-partum hypotension and collapse. Blood pressure is more important than blood volume in maintaining cerebral perfusion and consciousness and the relative reduction in intra-vascular volume in hypertensive pregnancies would not pose a significant disadvantage. The limited value of blood volume can be noted in the observation that blood pressure falls in the first half of pregnancy (Fig. 1) despite a substantial rise in blood volume. Thirdly, restriction of fetal size by restricting utero-placental perfusion and fetal growth and/or by relatively early delivery thus reducing the risk of prolonged and obstructed labour. Finally, the almost universal practice of midwifery or assisted birth [7] would have replaced self-assistance found in apes. The first three adaptations would have been so critical to maternal survival that any individual that did not manifest these would have very little chance of surviving child birth. A signal from the pregnancy is needed to initiate any such adaptation and the maternal response has to be regulated if the adaptation is to be sufficient
Figure 1 (A) Changes in maternal blood pressure with gestation age in women at term with normal pregnancies. Data were collected from the woman’s antenatal records. (B) Rate of change of blood pressure with gestation age in the second half of pregnancy. There was a significant increase in systolic (j 0.51 ± 0.18 mm Hg/week; p = 0.005), mean arterial (. MAP: 0.63 ± 0.13 mm Hg/week; p < 0.0001) and diastolic (m 0.69 ± 0.13 mm Hg/week;
to provide benefit without being excessive and causing disease or death. For this, nature is likely to have selected the simplest and most efficient regulatory mechanism.
The signal The signal for maternal adaptation (a rise in blood pressure in this case) will come from the pregnancy (feto-placental unit). To guarantee that the maternal response occurs, there will have to be substantial redundancy in the signal. This will be achieved by ensuring that the signal is conveyed by a variety
Hypertensive disorders of pregnancy are an evolutionary adaptation of molecules (rather than by a single molecule) and that at any point in pregnancy, the signal is produced in excess of what is needed, removing the need for up/down regulation. An example of this is the regulation of maternal glucose concentration during pregnancy. Glucose is vital for fetal survival and growth and the feto-placental unit produces a range of molecules that increase maternal glucose concentration. These signals are produced in excess of what is needed and in sufficient concentration to cause maternal disease in the absence of a regulated maternal response. As a consequence, most, if not all women will develop gestational diabetes mellitus by term if they do not increase insulin production during pregnancy.
The regulator With a redundant signal, all that is required for efficient control is a single highly efficient and tightly controlled regulator, insulin in the case of glucose regulation. Abnormalities can arise in four ways: overproduction of or hypersensitivity to the regulator (causing hypoglycaemia or hypotension for glucose and blood pressure regulation respectively, the opposite of the desired effect with substantial survival disadvantages and would have been de-selected). Underproduction of or resistance to the regulator on the other hand, would have resulted in hyperglycaemia or hypertension for glucose and blood pressure regulation, respectively. Given that the initial intention of the adaptation was to raise blood glucose concentration or blood pressure, these latter abnormalities are likely to persist in the species. Thus hypertensive disorders, like hyperglycaemic disorders of pregnancy, will arise as a consequence of failure of the regulatory process rather than an excess of, or an abnormal factor produced by the feto-placental unit.
Evidence for blood pressure regulation in pregnancy Fig. 1 shows the antenatal blood pressure profile of 11 women with an uncomplicated first pregnancy at term. These data demonstrate what is already well known but poorly understood. Blood pressure falls in the first half of pregnancy despite an increase in maternal blood volume and cardiac out-put. This is a consequence of the development of the low resistance utero-placental circulation and the production of a variety of vasodilators (such as oestrogens and progesterone) from the feto-placental
799
unit and the effect of these molecules on vascular smooth muscle and endothelium. The fall in blood pressure occurs despite the activation of the rennin–angiotensin–aldosterone system in early pregnancy. This trend, if continued into the second half of pregnancy, would have profound survival disadvantages. The pregnant female has to stand up in order to collect food, escape from predators or defend herself and postural hypotension with decreased mobility in late pregnancy would have conferred a significant disadvantage. In the second half of pregnancy, there is a gradual but steady rise in blood pressure, despite the persistence of the utero-placental circulation and increasing concentrations of oestrogens, progesterone and other potential vasodilators. Diastolic blood pressure rose by 0.69 ± 0.13 mm Hg/week (p < 0.0001, n = 11; Fig. 1B). The factors responsible for this rise in blood pressure are very poorly understood. The rise is more marked in diastolic pressure, indicating that it is achieved by an increase in peripheral resistance while the gradual increase indicates that this is a regulated rather than a random process. The development of hypertension becomes almost inevitable in such a system, while the survival disadvantages associated with hypotension would have resulted in the elimination of this trait. It is unlikely that the development of hypertensive disorders in pregnancy will be understood without an understanding of the regulation of blood pressure during the second half of normal pregnancy.
Implications for pre-eclampsia With respect to blood pressure regulation, therefore, the signal that raises blood pressure during the second half of pregnancy (Fig. 1B) is feto-placental in origin and has considerable redundancy in that it is conveyed by several molecules and is produced in excess at all points in all pregnancies. In the absence of maternal regulatory mechanisms therefore, most, if not all pregnant women will develop hypertension by term. Hypertensive disorders arise not as a result of abnormal factors produced by the placenta but because of abnormalities in the regulatory mechanism in the mother. The implication is that a placental cause for pre-eclampsia will not be identifiable. A placental causative factor for pre-eclampsia is as likely to exist as a placental causative factor for gestational diabetes mellitus. If such a factor exists, it is produced in disease-causing concentrations in both normal pregnancies and pregnancies complicated by pre-eclampsia. Consequently, the
800 factor is unlikely to be identified by the current approach of case-control studies which relies on statistically significant differences between groups. On the other hand, blood pressure regulation, like glucose regulation, is likely to be achieved during pregnancy predominantly by a single molecule and abnormalities in these regulatory mechanisms cause disease. Gestational diabetes mellitus, for instance, is not caused by any placental factor but by maternal insulin resistance which may be detectable even before pregnancy. The non-pregnant woman encounters glucose challenges on a regular basis and it is therefore efficient to employ an identical mechanism for the regulation of glucose concentration during pregnancy. Pregnant women, however, face a unique set of circumstances with respect to blood pressure regulation and the risk of heavy blood loss following delivery. The regulatory response is therefore likely to be mediated by a factor unique to pregnancy. The focus of pre-eclampsia research has thus far been on the identification of the placental factors that cause endothelial activation. The hypothesis here is that there are several such factors and indeed several candidates have been identified. The discovery of a placental cause of endothelial activation has as much implication on the understanding of hypertensive disorders as the discovery of say human placental lactgen would have had on the understanding of the cause of gestational diabetes mellitus. The second hypothesis is that there is a single main regulator of blood pressure in the second half of pregnancy
Ayuk and abnormalities in this regulatory mechanism result in hypertension. It is widely acknowledged that hypertensive disorders are at the extreme of a normal adaptation in pregnancy and therefore very similar to gestational diabetes mellitus in this respect. It is unlikely that the cause of such disorders will be discovered without understanding how normality is achieved and maintained in the first place. The search for the placental cause of the disorder, on the other hand, is futile. The key to our understanding of pre-eclampsia is proposed to be a molecule that prevents endothelial injury and decreases blood pressure, a molecule with an analogous role to that of insulin in the regulation of maternal glycaemia and the development of gestational diabetes mellitus. Fig. 2 shows a schematic representation of the proposed evolution of hypertensive disorders of pregnancy. It is interesting that the major challenges of modern obstetrics – venous thrombo-embolism, pre-eclampsia and fetal growth restriction are the extremes of adaptations that saved our species from reproductive failure and extinction. The development of operative delivery, blood transfusion and use of utero-tonic agents has now made such adaptations disadvantageous but are too recent to reverse millions of years of evolution. In the absence of modern medical care, the risk of maternal mortality and morbidity from hypertensive disorders is small when considered in the context of the risks posed by obstetric haemorrhage and obstructed labour. In this context therefore, maternal mortality and morbidity from
Figure 2 Schematic representation of the evolution of hypertensive disorders of pregnancy. IUGR (intra-uterine growth restriction).
Hypertensive disorders of pregnancy are an evolutionary adaptation hypertensive disorders of pregnancy, like morbidity and mortality from venous thrombo-embolism, would be a price worth paying.
References [1] Haig D. Genetic conflicts in human pregnancy. Q Rev Biol 1993;68:495–532. [2] Redman CW, Sargent IL. Latest advances in understanding preeclampsia. Science 2005;308:1592–4.
801
[3] Bartlett Linda A, Mawji Shairose. Maternal mortality in Afghanistan: magnitude, causes, risk factors and preventability; 2002. Available from: http://www.afghana.com/ Articles/maternalmortalityafghanistan.doc. [4] Rosenberg K, Trevathan W. Bipedalism and human birth: the obstetrical dilemma revisited. Evol Anthropol 1996;4: 161–8. [5] Rak Y. Lucy’s pelvic anatomy: its role in bipedal gait. J Hum Evol 1991;20:283–90. [6] Tague R, Lovejoy C. The obstetric pelvis of A.L. 288-1 (Lucy). J Hum Evol 1986;15:237–55. [7] Trevathan Wenda R. Human birth: an evolutionary perspective. New York: Aldine de Gruyter; 1987.