5
Preeelampsia: What We Know and What We Do Not Know James M. Roberts Preeclampsia remains a major health problem for mothers and infants. Studying the entire patho- physiology of preeclampsia rather than "pregnancy-induced hypertension" has greatly expanded our knowledge of the disorder. Current thinking approaches preeclampsia as a 2 stage disorder: reduced placental perfusion usually secondary to abnormal implantation and a consequent maternal disorder characterized by endothelial dysfunction and subsequent pathophysiological changes. We know much about the 2 stages and less about their linkage. It is evident that reduced placental perfusion is not sufficient to account for the pathophysiology. Reduced perfusion and abnormal implantation occur in other conditions (intrauterine growth restriction and preterm labor) without the maternal syn- drome. This leads to the hypothesis that reduced placental perfusion must interact with maternal constitutional factors to generate the systemic pathophysiology of preeclampsia. The similarities of these risk factors and metabolic alterations between preeclampsia and atherosclerosis suggest a common pathophysiology. Oxidative stress is postulated as the genesis of endothelial dysfunction in atherosclerosis. The author proposes that oxidative stress secondary to reduced placental perfusion leads to endothelial dysfunction, linking the 2 stages of the syndrome. Copyright 2000 by W.B. Saunders Company E clampsia was first described 2,000 years ago as seizures associated with pregnancy and resolving with delivery. 1 It was not until the end of the 19th century that the associa- tion with proteinuria and increased blood pressure ted to the conclusion that this preg- nancy abnormality was more than a seizure disorder. Rather, the occurrence of new onset increased blood pressure, especially in associ- ation with proteinuria, identified a clinical state in which the maternal condition could rapidly deteriorate and in which the fetus was at increased risk independent of seizures. Since that time, the 2 markers have been used in several classification schemes to define this syndrome. There are several variations cur- rently in use likely reflecting the fact that the findings by which the syndrome is defined were selected for historical reasons rather From the Department of Obstetrics, Gynecology and Reproductive Sciences, Magee-Womens Research Institute, University of Pitts- burgh, Pittsburgh, PA. Supported by National Institutes of Health (NIH) Grant: PO1 HD 30367. Address reprint requests toJames M. Roberts, MD, Director, Magee- Womens Research Institute, 204 Craft Ave, Suite 610, Pittsburgh, PA 15213; e-mail: [email protected] Copyright 2000 by W.B. Saunders Company O146-0005/00/2401-0007510. 00/0 than because of a close association with patho- physiology or adverse outcome. It is abun- dantly clear that preeclampsia is not simply pregnancy-induced hypertension and protein- uria. 2 Appreciation of this fact has led to an enormous increase in our understanding of the disorder in the last 15 years. There are many things we now understand about the disorder but likewise many we do not. Preeclampsia: What We Know Impact Preeclampsia is a leading cause of maternal mor- tality and increases perinatal mortality 5-fold. The increase in perinatal mortality is due, at least, in part to a management strategy which is now, as it was 100 years ago, delivery to termi- nate the pregnancy specific disease. While deliv- ery is always beneficial to the mother, it may result in the transition of a sick fetus in utero to a sick premature neonate in the nursery. Nearly 15% of preterm births are indicated preterm births for preeclampsia. Advances in therapy must target both maternal and neonatal well being. 24 Seminars in Perinatology, Vol 24, No 1 (February), 2000: pp 24-28

Preeclampsia: What we know and what we do not know

Embed Size (px)

Citation preview

Preeelampsia: What We Know and What We Do Not Know

James M. Roberts

Preeclampsia remains a major health problem for mothers and infants. Studying the entire patho- physiology of preeclampsia rather than "pregnancy-induced hypertension" has greatly expanded our knowledge of the disorder. Current thinking approaches preeclampsia as a 2 stage disorder: reduced placental perfusion usually secondary to abnormal implantation and a consequent maternal disorder characterized by endothelial dysfunction and subsequent pathophysiological changes. We know much about the 2 stages and less about their linkage. It is evident that reduced placental perfusion is not sufficient to account for the pathophysiology. Reduced perfusion and abnormal implantation occur in other conditions (intrauterine growth restriction and preterm labor) without the maternal syn- drome. This leads to the hypothesis that reduced placental perfusion must interact with maternal constitutional factors to generate the systemic pathophysiology of preeclampsia. The similarities of these risk factors and metabolic alterations between preeclampsia and atherosclerosis suggest a common pathophysiology. Oxidative stress is postulated as the genesis of endothelial dysfunction in atherosclerosis. The author proposes that oxidative stress secondary to reduced placental perfusion leads to endothelial dysfunction, linking the 2 stages of the syndrome. Copyright �9 2000 by W.B. Saunders Company

E clampsia was first described 2,000 years ago as seizures associated with pregnancy

and resolving with delivery. 1 It was not until the end of the 19th century that the associa- tion with pro te inur ia and increased b lood pressure ted to the conclusion that this preg- nancy abnormal i ty was more than a seizure disorder. Rather, the occur rence of new onset increased b lood pressure, especially in associ- ation with proteinuria , identified a clinical state in which the maternal condi t ion could rapidly deter iorate and in which the fetus was at increased risk i n d e p e n d e n t of seizures. Since that time, the 2 markers have been used in several classification schemes to define this syndrome. There are several variations cur- rently in use likely reflecting the fact that the findings by which the syndrome is defined were selected for historical reasons ra ther

From the Department of Obstetrics, Gynecology and Reproductive Sciences, Magee-Womens Research Institute, University of Pitts- burgh, Pittsburgh, PA. Supported by National Institutes of Health (NIH) Grant: PO1 HD 30367. Address reprint requests to James M. Roberts, MD, Director, Magee- Womens Research Institute, 204 Craft Ave, Suite 610, Pittsburgh, PA 15213; e-mail: [email protected] Copyright �9 2000 by W.B. Saunders Company O146-0005/00/2401-0007510. 00/0

than because of a close association with patho- physiology or adverse outcome. It is abun- dantly clear that preeclampsia is no t simply pregnancy- induced hyper tension and protein- uria. 2 Apprecia t ion of this fact has led to an enormous increase in our unde r s t and ing of the disorder in the last 15 years. There are many things we now unders tand about the disorder but likewise many we do not.

Preeclampsia: What We Know

Impact

Preeclampsia is a leading cause of maternal mor- tality and increases perinatal mortality 5-fold. The increase in perinatal mortality is due, at least, in part to a management strategy which is now, as it was 100 years ago, delivery to termi- nate the pregnancy specific disease. While deliv- ery is always beneficial to the mother, it may result in the transition of a sick fetus in utero to a sick premature neonate in the nursery. Nearly 15% of preterm births are indicated preterm births for preeclampsia. Advances in therapy must target both maternal and neonatal well being.

24 Seminars in Perinatology, Vol 24, No 1 (February), 2000: pp 24-28

Preeclampsia 2 5

Pathology and Pathophysiology

The pathological changes of preeclampsia are most consistent with reduced perfusion ra ther than mechanical disruption of blood vessels. 3 In the liver, for example, the characteristic hemor- rhage and necrosis are most consistent with pro- found vasoconstriction leading to infarction with bleeding into the necrotic area. In addition, subendocardial necrosis, a characteristic change in hypovolemic shock, is also present in women dying with eclampsia. These pathological evi- dences of reduced organ blood flow are sup- por ted by the pathophysiological findings of re- duced perfusion to virtually any organ in the preeclamptic woman. 3

Certain pathological and pathophysiological changes are especially revealing. The kidney of the woman with preeclampsia manifests a char- acteristic morphological change present in no other form of hypertension. This change, termed glomerular endotheliosis, consists of oc- clusion of the glomerular capillary lumen by swollen endothelial and mesangial cells with in- clusions in the basement membrane. Interest- ingly, despite the proteinuria, the renal podo- cytes are completely normal. ~ In addition to showing that preeclampsia is not merely a vari- ant of essential hypertension, the p ro found changes in the glomerular endothelial cells sug- gest that vascular endothel ium may be an impor- tant target in this disorder. Another finding pro- viding special insight is the abnormality of the spiral arteries that perfuse the intervillous space. Ordinarily these vessels undergo striking modi- fication in pregnancy. 4 Their diameter increases dramatically and the smooth muscle and inner elastic lamina of the vessel wall are lost. Endo- thelium no longer lines these cells but ra ther trophoblast cells that express endothel ia l anti- gens. 5 These changes do not take place in the vessels that supply the placenta of the pre- eclamptic woman. 6 They may occur in some su- perficial decidual vessels but never extend to the inner third of the myometr ium as is characteris- tic of normal pregnancy. These findings illus- trate 2 important points. The first is that a char- acteristic finding in preeclampsia and one which is likely pathophysiologically important, is re- duced placental perfusion. Second, the physio- logical modification that occurs, at least, in part secondary to trophoblastic invasion, is ordinarily

completed by 20 to 22 weeks gestation. Thus, al though preeclampsia characteristically pre- sents in late gestation, the root causes are present much earlier.

The pathophysiological changes of pre- eclampsia fur ther support the presence of the disorder from earlier gestation than its clinical presentation. 3 The reduced organ perfusion of preeclampsia is secondary to profound vasocon- striction. Rather than an increase in ei ther usual or unique pressors, vasoconstriction is secondary to an increased sensitivity to all pressors. Quite importantly this can be shown in groups of women destined to develop preeclampsia weeks to months before clinically evident disease. Ac- tivation of the coagulation cascade, especially platelets, likely fur ther reduces organ perfusion by the formation of microthrombi. As with vas- cular sensitivity, increased platelet turnover is present months before clinical preeclampsia. Fi- nally, fluid is lost f rom the vascular compart- ment secondary to vasoconstriction and and to an endothelial leak. These changes fur ther re- duce organ perfusion. Again the consequent re- duct ion in plasma volume antedates the diagnos- tic findings of preeclampsia. ~

Endothelial Dysfunction in Preeelampsia

Normal endothel ium buffers response to circu- lating pressors, prevents activation of platelets, activates circulating anticoagulants, and main- tains fluids in the intravascular compartment . 7 Thus, the earliest pathophysiological changes of preeclampsia are all consistent with abnormali- ties of endothelial function. This is fur ther sup- ported hy the morphological changes in the glo- merular capillary endothel ium. In addition, biochemical markers of endothelial activation, increased cellular f ibronecton, von Willebrand Factor, vascular cell adhesion molecule, and th rombomodul in amongst others, are increased in the circulation of women with preeclampsia. Many of these markers, as well as other indica- tors of endothelial dysfunction, are present weeks to months prior to clinical preeclampsia. Endothelial dysfunction is posited to be a central feature leading to the other pathophysiological changes of preeclampsia explaining the multior- gan involvement and protean manifestations.

26 James M. Robe~r

The Stages of Preeclampsia

Preeclampsia is a 2 stage disorder, s Preeclampsia is a pregnancy specific disease that begins to abate with delivery of the placenta. Sixty years ago Page 9 proposed that the placenta was tile impor tan t pregnancy c o m p o n e n t and that the feature that resulted in preeclampsia was re- duced placental perfusion. The abnormal im- plantation, typical of preeclampsia, the associa- tion of preeclampsia with medical disorders characterized by microvascular disease and with obstetrical conditions with large placentas sup- por ted this hypothesis. Subsequently, fur ther suppor t was provided by direct measurements of intervillous blood flow and animal experiments. The reduced pertiasion of the intervillous space is the first stage that culminates in stage 2: en- dothelial dysfunction and systemic maternal dis- ease.

Preeclampsia: What We Do Not Know

What is the l inkage be tween the 2 stages of preec lampsia : ma te rna l f e t a l / p l acen ta l inter- actions. As we begin to explore the l inkage be tween r edu ced placenta l pe r fus ion and the ma te rna l systemic mani fes ta t ions of pre- eclampsia, it is evident that a l though r educed per fus ion seems necessary for p reec lamps ia it is clearly not sufficient. Not all mo the r s with growth res t r ic ted infants have preeclampsia . Some condi t ions that p red i spose to pre- ec lampsia (obesity and insulin resistance) are associated with large infants. Perhaps, most impor tant ly , the a b n o r m a l implan ta t ion char- acteristic o f p reec l amps ia is p r e sen t in women with growth rest r ic ted infants and in one third of women del iver ing p re t e rm. 6,1~ This suggests that the r educed per fus ion mus t in terac t with ma te rna l factors to result in preec lampsia . It is quite likely that these risk factors may be dif- f e r en t for d i f ferent women . For example , sev- eral genet ic po l ym orph i s m s are associated with p r e e c l a m p s i a - - b u t not in all popula t ions . An ang io t ens inogen var iant is associated with p reec l amps ia in Utah and J a p a n II but no t En- g land ~2 and Pi t t sburgh and a var iant o f the homocys te ine re la ted gene me thy lene tetrahy- drofola te reductase is p r e sen t in p reec lampt ic w o m e n in J a p a n and Italy but not in 2 US p o p u l a t i o n s ) ~ The exis tence of d i f ferent risk factors for d i f ferent w o m e n indicates that

there will no t likely be any 1 p reec l amps ia gene and also suggests that prevent ive therapy may need to be ta i lored to individual popula - tions.

Preeclampsia and Atherosclerosis

As we begin to look at the risk factors that could predispose to preeclampsia, the similarity of these risk factors to those for atherosclerosis is striking. Similar risks include obesity, diabetes, hypertension, black race, and hyperhomocys- teinemia. Fur ther suppor t for c o m m o n risk fac- tors comes f rom the relationship of preeclamp- sia to cardiovascular disease in later life. The classic work of Chesley et a114 indicates that women with eclampsia in their first p regnancy have no excess of cardiovascular disease in later life compared with appropr ia te ly matched women with unknown pregnancy history. How- ever, if pregnancy history is known, women who have been pregnan t and never developed pre- eclampsia have a lower risk of cardiovascular disease than the general female populat ion. 15 In addit ion to the commonal i ty of risk factors indi- cated by this relationship, there are o ther inter- esting similarities between the disorders. The endothe l ium is target in both and the character- istic dyslipidemia predisposing to atherosclerosis in women, elevated triglycerides, reduced high density l ipoprotein, and increased low density l ipoprotein (LDL) cholesterol in association with small dense LDL, are also present in women with preeclampsia./6,17

Oxidative Stress: The Linkage Between Reduced Placental Perfusion and Preeclampsia

Information on the pathogenesis of atherosclero- sis may provide clues as to the linkage of reduced placental perfusion to endothelial dysfunction in preeclampsia. Oxidative stress, with the formation of oxidized LDL from small dense LDL in the subendothelial space, has been postulated as the initiator of endothelial damage in atherosclero- sis. is There is also abundant evidence of oxidative stress in preeclampsia.19 Markers of oxidative stress are increased in blood and tissues of women with the disorder and there are reports of increased concentration of antibodies directed against oxi- dized LDL. Perhaps alteration of endothelial func- tion by reactive oxygen species generated in the

Preeclampsia 2 7

intervillous space can link reduced placental per- fusion to systemic maternal disease in preeclamp- sia. How might oxidative stress arise in preeclamp- sia? One of the primary sources of oxidative stress is reduced organ perfusion with subsequent return of normal oxygenation. A secondary wave of injury due to the formation of reactive oxygen species with reperfusion accounts for much of the tis- sue damage with coronary occlusion or stroke. With reduced peffusion, adenosine triphosphate cannot be fully oxidized and breaks down to aden- osine providing increased substrate for the bi- functional enzyme, xanthine oxidase/dehydroge- naseY ~ The enzyme is increased in response to hypoxia and preferentially functions as its oxidase form to produce uric acid and superoxide radical. With reinstitution of oxygen delivery, there is an abundant production of this radical. Uterine blood flow decreases with many normal activities. Changes in posture strikingly alter uterine blood flow. Increased requirements for nutrient and ox- ygen delivery to other organs (eg with exercise) reduces uterine blood flow. In labor, the blood flow to the placenta is transiently reduced with every uterine contraction. Adaptive mechanisms appear to prevent the formation of reactive oxygen species in the normal pregnant woman. However, in a setting of reduced placental perfusion, as is present in preeclampsia, might these normal re- ductions of uterine blood flow result in hypoxia sufficient to generate oxidative stress? In support of this hypothesis we have demonstrated increased xanthine oxidase/dehydrogenase and increased oxidase activity in the invasive cytotrophobtasts of preeclamptic women, z: Reactive oxygen species in general are quite labile; nonetheless, there are several possibilities for transfer of their effects to endothelium. The simplest would be oxidative in- jury by stable products of lipid oxidation such as malondialdehyde, known to be increased in the blood of preeclamptic women. Alternatively, nen- trophils and monocytes passing through the inter- villous space could be activated by exposure to reactive oxygen species and interact with endothe- lial cells to generate oxidative stress. This is also supported by the observation that these cells are activated in preeclamptic women. 22 Knight et a123 have shown increased deportation of syncitial flag- ments in preeclamptic women. If these portions of trophoblast were modified by oxidative stress, they could also provide a source of reactive oxygen species interacting with endothelium. Whatever

the transport mechanism, the consequences of this challenge would be influenced by maternal factors. Women with lipids with increased sensitiv- ity to oxidation (small dense LDL) would be espe- cially effected by this as would women with other potential sources of endothelial injury or oxidative stress (eg, hyperhomocysteinemia). Similarly, women with dietary deficiency of anti oxidants or genetically determined differences in resistance to oxidative stress would be at increased risk.

Summary

Our understanding of preeclampsia has in- creased as we consider the entire spectrum of the disease and not merely hypertension and renal dysfunction. Evolving data suggest the fol- lowing hypothesis. Preeclampsia results f rom pro-oxidants generated, at least, in part second- ary to reduced placental perfusion. These active species interact with maternal constitutional fac- tors, fur ther modified by pregnancy specific changes to generate oxidative stress that alters systemic endothelial function.

References

1. Chesley LC: Hypertensive disorders of pregnancy. New York, NY, Appleton-Centmy-Crofts, 1978

2. Roberts JM, Redman CWG: Pre-eclampsia: More than pregnancy-induced hypertension. Lancet 341:1447- 1451, 1993

3. Roberts JM: Pregnancy related hypertension, in Creasy RK, Resnik R (eds): Maternal Fetal Medicine. Philadel- phia, PA, Saunders, 1998

4. Pijnenborg R, Robertson WB, Brosens I, et al: Tropho- blast invasion and the establishment of haemochorial placentation in man and laboratory animals. Placenta 2:71-92, 1981

5. Zhou Y, Fisher SJ, Janatpour M, et al: Human cytotro- phoblasts adopt a vascular phenotype as they differenti- ate. A strategy for successful endovascular invasion? J Clin Invest 99:2139-2151, 1997

6. Khong TY, De Wolf F, Robertson WB, et al: Inadequate maternal vascular response to placentation in pregnan- cies complicated by pre-eclampsia a n d b y small-tor-ges- rational age infants. BrJ Obstet Gynaecol 93:1049-1059, 1986

7, Roberts JM: Endothelial dysfunction in preeclampsia [review]. Semin Reprod Endocrinol 16:5-15, 1998

8. Redman CWG: Current topic: Pre-eclampsia and the placenta. Placenta 12:301-308, 1991

9. Page EW: The relation between hydatid moles, relative ischemia of the gravid uterus, and the placental origin of edampsia. Am J Obstet Gynecol 37:291-293, 1939

2 8 James M. Roberts

10. Arias F, Rodfiquez L, Rayne SC, et al: Maternal placental vasculopathy and infection: Two distinct subgroups among patients with preterm labor and preterm ruptured mem- branes. AmJ Obstet Gynecol 168:585-591, 1993

11. Ward K, Hata A, Jeunemaitre X, et al: A molecular variant of angiotensinogen associated with preeclamp- sia. Nat Genet 59-61, 1993

12. Morgan L, Baker P, Pipkin FB, et al: Pre-eclampsia and the angiotensinogen gene. Br J Obstet Gynaecol 102: 489-490, 1995

13. Powers R, Minich L, Lykins D, et al: Methylenetetrahydro- folate reductase polymorphism, folate and susceptibility to preeclampsia. J Soc Gynecol Investig 6:74-79, 1999

14. Chesley LC, AnnittoJE, Cosgrove RA: The remote prog- nosis of eclamptic women: Sixth periodic report. Am J Obstet Gynecol 124:446-459, 1976

15. Fisher KA, Luger A, Spargo BH, et ah Hypertension in pregnancy: Clinical-pathologicaI correlations and re- mote prognosis. Medicine 60:267-276, 1981

16. Sattar N, Bendomir A, Berry C, et ah Lipoprotein subfrac- tion concentrations in preeclampsia: pathogenic parallels to atherosclerosis. Obstet Gynecol 89:403408, 1997

17. Hubel CA, McLaughlin MK, Evans RW, et al: Fasting

serum triglycerides, free fatty acids, and malondialde- hyde are increased in preeclampsia, are positively corre- lated, and decrease within 48 hours post partum. Am J Obstet Gynecol 174:975-982, 1996

18. Witztum J: The oxidation hypothesis of atherosclerosis. Lancet 344:793-795, 1994

19. Hubel CA: Oxidative stress and preeclampsia. Fetal Ma- ternal Med Rev 9:73-101, 1997

20. Hamer I, Wattiaux R, Wattiaux-DeConinck S: Deleteri- ous effects of xanthine oxidase on rat liver endothelial cells after ischemia/reperfusion. Biochim Biophys Acta 1269:145-152, 1995

21. Many A, Hubel C, Fisher SJ, et ah Invasive cytotropho- blast manifest evidence of oxidative stress in preeclamp- sia. AmJ Pathol 1999 (in press)

22. Clark P, Path MRC, Boswell F, Greer IA: The neutrophil and preeclampsia. Semin Reprod Endocrinot 16:57-64, 1998

23. Knight M, Redman CWG, Linton EA, et al: Shedding of syncytiotrophoblast microvilli into the maternal circula- tion in pre-eclamptic pregnancies. BrJ Obstet Gyamecol 105:632-640, 1998