Chesley's Hypertensive Disorders in Pregnancy || Cardiovascular Alterations in Normal and Preeclamptic Pregnancy

CHAPTER 14

Cardiovascular Alterations in Normal andPreeclamptic Pregnancy

JUDITH U. HIBBARD, SANJEEV G. SHROFF AND MARSHALL D. LINDHEIMER

Editors’ comment: The heart was a brief bystander inChesley’s original text, while exploration of cardiac perfor-mance during pregnancy as well as a new school hypothe-sizing an important role for aberrant cardiac output in thegenesis of the preeclampsia syndrome were just emergingwhen the second edition went to press. The 1999 edition alsodiscussed normal cardiovascular changes in pregnancy, thepathophysiological alterations, especially preeclampsia, be-ing assigned to different chapters. In this edition the normal(previously Chapter 3 in the second edition by McLaughlinand Roberts) and pathological alterations (previously Chap-ter 8 by Conrad and Lindheimer) during pregnancy havebeen combined in a single chapter, with the expectation thatthis will be more logical and easier for the reader. A numberof critical figures and tables from those second edition chap-ters are reproduced here.

INTRODUCTION

There are striking physiologic cardiovascular changes dur-ing pregnancy that assure adequate uterine blood flow, aswell as appropriate oxygenation and nutrient delivery to thefetus. Further, these compensatory mechanisms allow themother to function normally during the altered physiologicstate of pregnancy. Both knowledge of and understanding theroles of these changes are particularly critical if we are todevelop clinical strategies to manage pregnant women withchronic or new hypertension and especially preeclampsia,the systemic disorder which in its severest forms causesserious morbidity, and even death. Thus, how the latter dis-ease impacts the cardiovascular system may be integral toappropriate therapy, but as we shall see, studies designed todocument effects of preeclampsia on the cardiovascular sys-tem have not always produced uniformity in findings. Thischapter begins with a review of normal cardiac and hemo-dynamic function during pregnancy followed by a survey ofknowledge regarding cardiac performance and vascularchanges in preeclampsia, focusing on recent progress madepossible by advances in noninvasive technology.

Chesley’s Hypertensive Disorders in Pregnancy.ISBN: 978-0-12-374213-1 249

HEMODYNAMICS AND CARDIAC FUNCTIONIN NORMAL PREGNANCY

Nearly 100 years ago, Lindhard,1 reporting on normal car-diovascular adaptations in pregnancy, described a 50% in-crease in cardiac output as measured by a dye-dilutiontechnique. Since then multiple methodologies have beenemployed to assess cardiovascular function in the gravidaresulting in a myriad of findings. The “gold standard” forsuch evaluation remains flow directed pulmonary arterycatheters using thermodilution methodology. Given the in-vasive nature of these techniques, only cross-sectional inves-tigations are feasible. Fortunately, noninvasive M-modeechocardiography and continuous and pulsed-wave Dopplertechniques, validated against an invasive technique, permitserial determinations throughout both normal and abnormalpregnancy.

One must be cautious in regard to several methodologicissues that impact cardiovascular parameters in pregnancy.These are maternal posture during data collection, whetheror not the patient received fluids or vasoactive medicationsprior to data acquisition, and whether she is in active labor.2

In addition, maternal body habitus can also affect cardiacmeasurements. Some, but not all investigations, have con-trolled for these potential confounding issues.

Systemic Arterial Hemodynamics in NormalPregnancy

There are significant decreases in both systolic and dias-tolic blood pressure, noted as early as 5 weeks gestation3

(Fig. 14.1). Interestingly, Chapman et al.4 noted a significantdecrease in blood pressure during the luteal compared to thefollicular phase of the menstrual cycle, data suggesting ahormonal origin of the fall in blood pressure that persistsand increases further when conception occurs. The decreasein blood pressure during pregnancy is characterized as fol-lows: Decrements in diastolic levels exceed those in systoliclevels, the former averaging 10 mm Hg below the baselinevalue. Mean blood pressure nadirs at 16–20 weeks, these

Copyright 2009, Elsevier, Inc.All rights of reproduction in any form reserved.

FIGURE 14.1 Mean blood pressure by gestational age in 6000white women 25 to 34 years of age who delivered single-terminfants. (Reprinted with permission.3)

FIGURE 14.2 Averaged mean arterial blood pressures (W 1 SE)in women who remained normotensive throughout pregnancy(o–o), in women who developed preeclampsia (*–*), and inwomen who developed hypertension (&–&) by periods of 4 weeks.(Reprinted with permission.5)

250 Chesley's Hypertensive Disorders in Pregnancy

changes persisting to the third trimester3,5 (Figs. 14.1 and14.2). In the mid-third trimester blood pressure graduallyrises, often approaching the pre-pregnancy values.5–17 Thereare diurnal fluctuations in normal pregnancy, similar topatterns in the nonpregnant state, the nadir occurring atnight.18–20 Finally, all these observations have also beenverified using 24-hour ambulatory blood pressure monitor-ing protocols21 (Fig. 14.3A, B).

Cardiac output increases 35–50% during gestation,half or more of this increase established by 8 weeks ofpregnancy.14 The earliest evidence of this change (in parallel

FIGURE 14.3 Differences in systolic (A) and diastolic (B) blood press48-hour ambulatory blood pressure monitoring throughout gestation. (Re

with decreases in mean arterial pressure) can be detected inthe luteal phase of the menstrual cycle.3 If no conceptionoccurs there is a significant reversal of all these changes,detectable in the next follicular phase (see Fig. 15.1). Thesedata implicate the corpus luteum in the observed changes,and suggest a hormonal role in the early cardiovascularchanges of pregnancy.

The significant increase in cardiac output is well estab-lished by gestational week 5, rising further to 50% abovepre-pregnancy values by gestational weeks 16–20, then in-creasing slowly or plateauing until term8–10,12,13,16,17,22,23

(Fig. 14.4). These data were acquired with the women posi-tioned in left lateral recumbency. Several investigators havenoted different patterns including a decrease in cardiacoutput from the peak pregnancy value when close to

ure between groups of pregnant women systematically measured byprinted with permission.21)

FIGURE 14.4 Longitudinal studies of cardiac output, strokevolume, and heart rate started before conception and continuedthrough to the postnatal period. (Reprinted with permission.16)

CHAPTER 14 � Cardiovascular Alterations in Normal and Preeclamptic Pregnancy 251

term6,11,24,25; these discrepancies might relate to failingto control posture or to correct for body surface area.6,11

However, two investigators whose descriptions also deviatedfrom our summary composite24,25 did control for both pos-ture and body surface area.

Stroke volume and heart rate, the two determinants ofcardiac output, appear to rise sequentially, with the increasesapparent by 5–8 weeks gestation.6,9,14 Stroke volume con-tinues to increase until gestational week 16, plateauing there-after, while heart rate continues to increase slowly into thirdtrimester9,10,14 (Fig. 14.4).

Capeless and Clapp14 noted that 50% of the increase incardiac output had occurred by gestational week 8, this earlychange primarily due to increased strokevolume (not to heartrate changes). These investigators also noted that parousgravidas had a greater rise and rate of change in stroke

volume compared to nulliparas, as well as a greater dropin systemic vascular resistance, but there was no effect onheart rate.8

Postural changes can impact heart rate, blood pressure,and cardiac output. Both heart rate and blood pressure aresignificantly lower in lateral recumbency, while cardiacoutput is increased in this position. There is a reduction incardiac output upon standing, noted in first trimester,which becomes significantly attenuated in the second tri-mester and absent by the mid third trimester.26 Intravas-cular volume is progressively amplified up to 40%,perhaps contributing to the aforementioned changes (seeChapter 15).

To summarize, most evidence supports the followingscenario of physiologic cardiovascular changes duringpregnancy: There is an early decrease in systolic, diastolic,and mean blood pressures (diastolic more than systolic), anearly increase in cardiac output that continues to rise orplateau into the third trimester, and increases in both strokevolume and heart rate contributing to this increase in car-diac output.

Venous System in Normal Pregnancy

MEAN CIRCULATORY FILLING PRESSURE

The venous return to the right heart maintains filling pres-sure, permitting adaptation to changing cardiac outputrequirements. A prerequisite for such regulation is that thevascular bed with appropriate tone should be adequatelyfilled with blood. The mean circulatory filling pressure(MCFP) characterizes this steady state veno-cardiac fill-ing.27 The MCFP is the pressure recorded in the vasculartree at equilibrium and in the absence of any blood flow,which is the pressure in the circulation after the heart hasbeen arrested and the system has come into equilibrium.28

The MCFP thus provides an indication of the relationshipbetween changes in blood volume compared to the size of thecirculatory compartment and, as such, indicates to whatextent the vascular compartment accommodates the largegestational increases in total blood volume.29 Venous smoothmuscle activation and changes in blood volume are mechan-isms for changing the MCFP.30

Data assessing MCFP in pregnancy derive from the an-imal literature as, of course, one cannot and should notdesign experiments that require stopping a human heart!Thus one must be circumspect about their relevance to hu-man pregnancy, as not only may we be dealing with speciesdifferences, but all these experiments concerning MCFPrequire studies in anesthetized animals.

The MCFP is slightly, but significantly, elevated in preg-nant dogs, rabbits, sheep, and guinea pigs,29,31–34 though onegroup studying guinea pigs noted no differences in compar-ison to the nongravid state.35 Furthermore, pregnant dogsrespond to epinephrine infusions with a rise in MCFP in a


manner similar to nonpregnant controls, suggesting that theyare able to increase their vascular tone normally. Also, theslope relating MCFP to changes in blood volume in moststudies appears not to be altered by gestation,31,32,34,35 ratherthe blood volume (BV)–MCFP relationship is merely shiftedto the right (i.e., increased unstressed volume). One group,29

however, did suggest a decrease in the slope of theBV–MCFP relationship, interpreted as increased compli-ance. Both increased venous unstressed volume and compli-ance permit the large increases in intravascular volume,characteristic of gestation, to occur with very little rise inMCFP. As a whole, the increased MCFP with the aforemen-tioned changes in venous unstressed volume and compliancecan be interpreted as showing that the “stressed” (i.e., dis-tending) component of the intravascular volume is elevatedin pregnancy.

The higher MCFP has led some to suggest the increasedcardiac output may be secondary to relative overfilling of thecirculatory system during pregnancy31,32 (see Chapter 15,Normal and Abnormal Volume Homeostasis, for discussionswhether the gestational increase in blood volume should beconsidered “under-fill,” “over-fill,” or “normal-fill”). Sinceblood volume–MCFP relationships represent a measure oftotal circulatory compliance36 and for practical purposes totalbody venous compliance,37 such data would suggest venoustone was unaltered in pregnancy, a finding that supports thosewho see pregnancy as vascular overfill. As noted, the con-cepts of under-fill, over-fill, and normal-fill in interpreting thevolume changes in pregnancy are discussed in greater detailin Chapter 15. Concerning the meaning of MCFP in thesediscussions, one must recall the circumstances of how thisindex is measured, and the many interpretative problems withthese anesthetized animal preparations.

In terms of cardiovascular homeostasis, the “active” reg-ulation of capacitance remains a key question during preg-nancy as this is one of the important mechanisms thatinfluence venous return of blood to the heart, the systemicreflex capacitance in humans estimated at �5 mL/kg.27 Thedensely innervated mesenteric venous microcirculationappears to have a key function in controlling changes invascular capacity. For example, most of the blood volumeshift in the intestinal vascular bed during activation of thebaroreceptor reflex occurs in the intestinal venules.27

VENOUS TONE REGULATION

The venous system in pregnancy appears to be a neglectedarea of research. Hohmann et al.38 studied adrenergic regu-lation of venous capacitance in pregnant and pseudopregnantrats, noting a progressive decline in the sensitivity to adren-ergic nerve stimulation from cycling to late gestation. Thereduced sympathetic nerve response was associated withmarked increases in sensitivity to exogenously applied epi-nephrine during pregnancy suggesting denervation super-sensitivity.

It should be noted that venous pressure–volume or wallstress–strain relationships cannot accurately be character-ized (either in vivo or in vitro) without defining thecontractile state of the vascular smooth muscle.27 Also,measurements made in vivo cannot distinguish betweenwall structural changes and those caused by differencesin venous tone. In one study, where vascular smooth mus-cle was inactivated prior to assessing stress–strain relation-ships in pregnant rodents, the compliance of the mesentericcapacitance veins decreased by 40%.39 However, the un-stressed volume doubled in comparison to the nonpregnantfemales.

In human studies, noninvasive measures suggest thatvenous distensibility increases with pregnancy in someinvestigations,40–44 but not in others.45–47 We could locateno longitudinal measures of humanvenous distensibility thatincluded pre-conception values.

In summary, despite the importance of venous function tocardiovascular volume homeostasis, knowledge of either thenormal or pathophysiologic status of this system is quitelimited. It appears that venous distensibility (compliance)increases during gestation. Animal data, obtained under lessthan ideal experimental conditions, suggest that the in-creased vascular capacitance does not quite accommodatethe increase in blood volume (“over-fill”). Whether MCFPchanges in human gestation is unknown. Information onsympathetic regulation of venous function both in terms ofregulating venous return and fluid exchange at the capillarybed would be of interest, but again information is spotty. Theveins andwhat influences their function in pregnancy requirea great deal of attention, and limited information is availableregarding what happens in preeclampsia.

Systemic Arterial Properties in Normal Pregnancy

The gestational increase in cardiac output and decrement inblood pressure have traditionally been ascribed to themarked decrease in total systemic vascular resistance thatoccurs early in gestation.9,14,24,48 It should be recognized,however, that other changes may be involved. For example,both left ventricular and systemic arterial mechanical prop-erties (ventricular afterload) have a potential to alter system-ic hemodynamics. Afterload, or the arterial system load theheart experiences, is the mechanical opposition experiencedby the blood ejected from the left ventricle. This oppositioncan be considered to have two components: one steady, theother pulsatile. The steady component, quantified in terms oftotal systemic vascular resistance, is determined by the prop-erties of the small caliber resistance vessels (e.g., effectivecross-sectional area) and blood rheological properties(e.g., viscosity).

There are other considerations. Due to the pulsatile na-ture of cardiac ejection, oscillations in pressure and flowexist throughout the arterial tree and thus, the pulsatile com-ponent of the arterial load needs to be considered. Physically,


the pulsatile arterial load is determined by the (visco)elasticproperties of the arterial vessel wall, architectural features ofthe arterial circulation (i.e., network of branching tubes), andblood rheological properties.49 Quantitative indices of pul-satile load include the global arterial compliance, aorticcharacteristic impedance, and measures of wave propagationand reflection.49 Global arterial compliance is a measure ofthe reservoir properties of both the conduit and peripheralarterial tree. In contrast to arterial compliance, which is aglobal property (i.e., belonging to the entire circulation),characteristic impedance quantifies a local property(i.e., belonging to the site of pressure/flow measurement),being determined by local vascular wall stiffness and geo-metric properties. Pulse wave velocity and global reflectioncoefficient are indices often used to describe wave propaga-tion and reflection within the arterial tree. Understanding theinterplay of both steady and pulsatile components shouldlead to a better grasp of cardiovascular performance in preg-nancy with its marked changes in both components.

To this end Poppas et al.10 serially studied 14 normal,normotensive gravidas throughout pregnancy and 8 weekspostpartum using noninvasive measures of instantaneousaortic pressure and flow velocities to assess both conduitand peripheral vessels. These investigators verified thatsystemic vascular resistance, the steady component ofthe arterial load, decreases very early in pregnancy, andcontinues to decrease significantly through the remainderof pregnancy, though less so in the latter weeks of gesta-tion10 (Fig. 14.5). Global arterial compliance increasedby 30% during the first trimester and was maintainedthereafter throughout pregnancy, temporally relating tothe decreased systemic vascular resistance. By 8 weekspostpartum global compliance returned to normal levels10

FIGURE 14.5 Temporal changes in global arterial compliance(ACA) and total vascular resistance (TVR) during normal pregnan-cy. Data are normalized to 8-week postpartum control values(meanW SEM; *p < .05, first, second, or third trimester vs. 8-weekpostpartum control; †p < .05, second or third vs. first trimester).(Reprinted with permission.10)

(Fig. 14.5). There was a tendency for aortic characteristicimpedance to fall, and the magnitude of arterial wavereflections was reduced during late pregnancy, along witha delay in the timing of reflected waves. Similar resultsdocumenting increased compliance have been noted inpregnant animal models29,35,50–53 and in other studies ofpregnant humans,44,54 as well as nonpregnant humans.55

Mone et al.24 also noted decreased characteristic imped-ance in a cohort of 33 normal gravidas.

It is clear that both steady and pulsatile arterial loaddecrease during normal pregnancy, indicating a state of pe-ripheral vasodilation and generalized vasorelaxation thatinvolves both the peripheral (resistance) vessels and conduitvessels. The magnitudes of the fall in systemic vascularresistance and the rise in cardiac output seem to be equiva-lent, which results in a very small change (fall) in meanarterial pressure. The decrement in pulsatile arterial load(i.e., increased global compliance, decreased characteristicimpedance, decreased reflection coefficient) appears to beprimarily due to a generalized increase in vascular disten-sibility, which, in turn, may be related to reduced smoothmuscle tone and vascular remodeling.10

Left Ventricular Properties in Normal Pregnancy

Left ventricular mass increases in normal pregnancy. Theincrease has been described by some6,10,23,24,56 as modest,averaging 10–20%, while others9,17,22 have reported incre-ments as great as 40%. An increase in ventricular massshould contribute to the heart’s ability to increase its totalpower (described below). Of note, in most studies theincrease in mass does not meet criteria for ventricular hy-pertrophy (>2 SD above the mean for normal population),as might occur, for example, in patients with chronic hyper-tension. Also, ventricular mass reverts to nonpregnant valuespostpartum.2 There is a mild increase in left ventricular end-diastolic chamber diameter noted by many,9,23,56,57 but notall10,12,24,25,58 investigators.

Normal pregnancy is associated with an increase in thecross-sectional area of the left ventricular outflow tract,measured at the aortic annulus.2,8–11,16,22,24,25,48,59 Thus, itis important to assess aortic diameter at each echocardiogra-phically obtained set of hemodynamic parameters in longi-tudinal studies.2 These findings further highlight the risks thenormal changes in pregnancy create for womenwith diseasesknown to be associated with compromised aortic roots(e.g., Marfan’s or Turner’s syndromes). That is, pregnancymay precipitate rupture or dissection.60,61

Evaluation of left ventricular myocardial contractilityin pregnancy has produced conflicting results. Use oftraditional ejection phase indices of left ventricular per-formance is problematic as these indices are unableto distinguish alterations in contractility from changes inventricular load.2,62,63 Thus, some of the variability in theresults related to the assessment of left ventricular

FIGURE 14.6 Average end-systolic wall stress (sES)-rate cor-rected velocity of fiber shortening (Vcfc) obtained in normotensivepregnant control subjects before delivery and 1 day and 4 weeksafter delivery. From visit 1 to visit 3, data points shifted rightwardand downward (arrow) but still fell on the mean contractility line,indicating increase in afterload without changes in contractility.(Reprinted with permission.2)


myocardial contractility may be attributable to the use ofload-dependent indices.

Lang et al.2 studied ten normal gravidas in early labor andone day and 4 weeks postpartum and quantified left ventric-ular myocardial contractility using themeasurements of end-systolic wall stress (sES) and rate-corrected velocity of fibershortening (Vcfc) (Fig. 14.6). Note, sES–Vcfc relationshipyields a preload independent and afterload adjusted charac-terization of left ventricular myocardial contractility. sES

and Vcfc data from an individual pregnant subject werecompared to a nomogram, i.e., sES–Vcfc relationship con-structed by studying a large group of normal, nonpregnantindividuals, both in their basal state and after pharmacologicmanipulation of afterload and preload. The sES–Vcfc datapoints for the normal gravidas were shifted rightward anddownward, remaining superimposed on the nomogram, in-dicating a decrease in afterload without any changes in leftventricular contractility2 (Fig. 14.6). These observationswere verified by Poppas et al.10 who reported an invariantleft ventricular myocardial contractility throughout preg-nancy in a cohort of normal gravidas. Simmons et al.23

similarly demonstrated unchanged contractility in 44 preg-nant women.

Some studies have claimed that left ventricular myocar-dial contractility changes during normal pregnancy. For ex-ample, Mone et al.24 have reported a reversible fall in leftventricular myocardial contractility. This conclusion wasbased on the observation that Vcfc progressively diminishedduring gestation (by 7% at term) even though sES wasdeclining over the same time period (by 15% at term). How-

ever, a comparison with the nomogram indicated that thegroup-averaged values of sES–Vcfc points during pregnancywere above the normal contractility line and were within thestatistical bounds of the normal (nonpregnant) population.Similarly, Gilson et al.12 have reported an enhanced leftventricular myocardial contractility. While there were nosignificant changes in Vcfc, sES decreased by 12% overthe observation period (early to late gestation). This obser-vation, if anything, would imply decreased myocardial con-tractility. Their conclusion of enhanced contractility wasbased upon the observed decrease in sES/Vcfc ratio, whichthey claim to be a load-independent index of myocardialcontractility as proposed by Colan et al.64 Interestingly,Colan et al.64 never proposed the sES/Vcfc ratio as an indexof myocardial contractility; they used the position of thesES–Vcfc point relative to the normal contractility line toquantify contractility in an individual subject. Furthermore,sES/Vcfc ratio is highly load-sensitive; it will change signif-icantly as one moves along a given sES–Vcfc relationship(i.e., fixed myocardial contractility by definition). Thus, theconclusion of enhanced myocardial contractility by Gilsonet al.12 appears to be erroneous.

To summarize, most of the evidence supports a conclu-sion that left ventricular myocardial contractility, as assessedby load-independent indices, is essentially unchanged innormal gestation. The data would be more secure, however,if the contractility nomogram used in future studies werederived exclusively from a female population of reproduc-tive age.

Coupling Between Left Ventricle and SystemicArterial Circulation in Normal Pregnancy

From the mean pressure-flow perspective, the coupled leftventricle–arterial circulation system produces significantlyhigher cardiac output during normal gestation, with littlechange in mean blood pressure (a small decrease is typicallyobserved). This coupled equilibrium of mean pressure andflow is achieved by a significant peripheral vasodilation(reduced systemic vascular resistance) and increases in heartrate, left ventricular preload (end-diastolic volume), andmuscle mass, without any significant changes in left ventric-ular myocardial contractility.

From the perspective of pulsatile hemodynamics, math-ematical simulation-based analysis indicates that if the fall inthe steady arterial load (i.e., decrease in systemic vascularresistance) were not accompanied by the fall in pulsatilearterial load (e.g., increase in global compliance) then arte-rial pulse pressure would have increased significantly, withthe decrement in diastolic pressure being significantly great-er than that in systolic pressure.10 Thus, the increase inglobal arterial compliance that accompanies the profoundperipheral vasodilation prevents the undesirable increase inpulse pressure and diastolic hypotension that can compro-mise myocardial perfusion.

FIGURE 14.7 Temporal changes in steady (WSTD) and oscillato-ry (WOSC) power during normal pregnancy. Data are normalized to8-week postpartum control values (mean W SEM; *p < .05, first,second or third trimester vs. 8-week postpartum control; †p < .05,second or third vs. first trimester). (Reprinted with permission.10)


Left ventricular hydraulic power, another functionalindex of the coupled left ventricle–arterial circulation sys-tem, has been evaluated during normal pregnancy. Twocomponents comprise total power: steady and oscillatorypower. The oscillatory component is considered to bewasted power as it does not result in net forward flow ofthe blood. Thus, the ratio of the oscillatory to total poweris often used as an index of inefficiency of the coupled leftventricle–arterial circulation system. Although both totaland oscillatory power increased significantly throughoutpregnancy, peaking in the third trimester (Fig. 14.7), theratio of the oscillatory to total power did not changesignificantly throughout gestation.10 The aforementionedmathematical simulation-based analysis revealed that thisratio would have doubled (i.e., efficiency decreasedby a factor of 2) had the increase in global arterial com-pliance not accompanied the fall in systemic vascularresistance.

In summary, the various systemic arterial and left ven-tricular mechanical properties undergo coordinated changesin normal pregnancy that result in significantly increasedcardiac output with little changes in mean and pulse pres-sures and ventriculo-arterial coupling efficiency. The fall inthe pulsatile arterial load (e.g., increase in global arterialcompliance) that accompanies the fall in the steady load(systemic vascular resistance) in normal pregnancy is con-sidered to be an adaptive response for several reasons: 1)markedly increased intravascular volume can be accommo-dated without a concomitant increase in mean arterial pres-sure, 2) increase in pulse pressure and diastolic hypotensionare prevented, and 3) the efficiency of the mechanical energytransfer from the left ventricle to the arterial circulation ismaintained.

HEMODYNAMICS AND CARDIAC FUNCTIONIN PREECLAMPSIA

Having underscored that assessing the cardiovascular systemin normal pregnancy is fraught with methodological and de-sign pitfalls, we note emphatically that such problems andchallenges are even greater when studying preeclampsia. Forexample, the true diagnosis of preeclampsia is never certain byclinical criteria alone.65 There may be other concomitant pa-thology simultaneously influencing the cardiovascular systemsuch as renal disease or diabetes. Finally, no matter how wellthe experiments are designed, certain confounders may bedifficult to eliminate or circumvent, as for example treatmentwith magnesium sulfate or other antiseizure agents, antihyper-tensive medications, or parenteral fluid prescription, as well aswhether or not the preeclamptic subject is in active labor.

Systemic Arterial Hemodynamics in Preeclampsia

While preeclampsia is characterized by the development ofhypertension late in gestation, patients destined to developpreeclampsia have been documented to have an elevatedmean blood pressure when compared to women who havenormotensive gestations, an observation present as early asgestational week 95,15,45,66 (Fig. 14.2). These differenceshave also been shown using ambulatory blood pressure tech-nology21 (Fig. 14.3). Clearly, elevated blood pressure in earlypregnancy is a risk factor for developing preeclampsia later;however, blood pressure values alone are a poor predictor foractually determining who will develop preeclampsia.21,67–70

A change in the normal diurnal pattern of blood pressurein women destined to develop preeclampsia has been noted,with either obliteration of the decrease in nocturnal pressureor a shift in the timing of the blood pressure nadir.20,71,72

Recently alterations in cardiovascular regulatory behaviorhave been suggested to predict preeclampsia by assessingsystolic as well as diastolic beat-to-beat pressures and heartrate by variability and coupling analysis. Discriminant func-tion analysis of the parameters predicted preeclampsia withan 88% sensitivity and specificity and if assessment of uter-ine artery resistance was added, a 70% positive predictivevalue at 18–26 weeks gestation was realized.73,74

Normal pregnancy, as noted above, is accompanied byincreased intravascular volume, high cardiac output, andvasodilation (a low-resistance systemic circulation). Withthe onset of overt preeclampsia there is a shift to a low-output, high-resistance state, and intravascular volume issignificantly lower than in the normal pregnant state (Chap-ter 15). This traditional characterization of preeclampsia as astate of decreased intravascular volume, lower cardiac out-put, and vasoconstriction is not observed by all investigators.A myriad of problems may contribute to the variable obser-vations, including manipulation of patients’ volume statusprior to evaluation that can alter the pathophysiological pic-ture present before the treatment.

TABLE 14.1 Hemodynamic profile in untreated and treated preeclamptic patients and normotensive pregnant women

Preeclamptics,Untreated (n = 87)

Pa NormotensiveControls (n = 10)

Pb PreeclampticsTreated (n = 47)

Mean intra-arterial pressure (mm Hg) 125 (92–156) <.001 83 (81–89) <.001 120 (80–154)c

Cardiac index (L � min�1 � m�2) 3.3 (2.0–5.3) <.001 4.2 (3.5–4.6) NS 4.3 (2.4–7.6)c

Systemic vascular resistance index(dyne � sec � cm�5 � m2)

3003 (1771–5225) <.001 1560 (1430–2019) <.005 2212 (1057–3688)c

Pulmonary capillary wedge pressure (mm Hg) 7 (1–20) NS 5 (1–8) <.05 7 (0–25)

Values given are median (range). (Reprinted with permission.83)NS, not significant.a Differences between untreated preeclamptic patients and normotensive controls.b Differences between pharmacologically treated preeclamptic patients and normotensive controls.c p< .05 vs. untreated nulliparous patients.


The dilemma alluded to above, for example, can beappreciated when one reviews a series of studies during the1980s performed using “gold standard” invasive monitoringwith Swan–Ganz pulmonary artery catheters that producedmarkedly contrasting results.75–82 Some investigators noteddecreased cardiac output, others increased or unchanged,and there was wide variation in the measured peripheralvascular resistance. However, a landmark investigation ofVisser and Wallenburg,83 we believe, pinpoints the reasonsfor these diverse findings.

Two unique features distinguish the work of Visser andWallenburg.83 First, they compared a group of untreated or“virgin” preeclamptics to a cohort of treated preeclamptics,and second, the number of subjects included, 87 and 47respectively, remains the largest investigation using invasivetechnology undertaken through 2009. It is unlikely that thisinvestigation will ever be repeated due to the fact that cen-tral monitoring with pulmonary artery catheters is rarelyused today in managing preeclamptic patients, the wideavailability of noninvasive techniques for assessment havingreplaced the invasive ones. These investigators carefullychose their subjects, using strict criteria and only selectingwomen with diastolic pressure of 100 mm Hg measuredtwice four hours apart, proteinuria �0.5 g/L, onset �20weeks of gestation, and complete recovery postpartum.Patients with medical disorders such as chronic hyperten-sion, cardiac or renal disease were excluded from enroll-ment. Patients receiving intravenous fluids, antihypertensivemedication, antiseizure therapy or any other type of medi-cation were considered “treated,” while those gravidas whohad not yet received any of the aforementioned therapieswere the “pure (or virgin) preeclamptic” group. However,the women were not randomized to treatment groups. Final-ly, a group of normotensive gravidas studied in anotherprotocol were used for further comparison.79 (Of note,studying consenting normal pregnant volunteers with cen-tral monitoring was defended in the 1980s, the risk believedlow, and the need to obtain normative data to improveintensive care monitoring and treatment of gravidas wasbelieved to be consistent with equipoise considerations.

The status of noninvasive technology, we believe, wouldcurrently preclude such studies.)

Table 14.1 summarizes the hemodynamic measurementsof Visser andWallenburg.83 As expected, mean arterial pres-sure was highest in the untreated preeclamptic group(125 mm Hg), underscoring disease severity in relation totreated preeclamptics and normotensive pregnancies. Notethe remarkably consistent hemodynamic parameters in the“virgin” preeclamptics, that is: a significantly decreased car-diac index, as well as the marked and significantly increasedsystemic vascular resistance. This contrasts with the resultsfrom treated preeclamptics. Note further that capillarywedge pressure remained normal. These findings stronglysupport the concept of preeclampsia being predominantlyassociated with low cardiac output, markedly increased sys-temic resistance, and an increased afterload, and suggestexplanations for the variable findings of others. This signalwork confirms the same group’s preliminary findings in asmaller group studied previously.79

Lang et al.,2 using noninvasive techniques to compare 10severe preeclamptics and 10 normotensive gravidas, eachevaluated in labor, and again one day and 4 weeks postpar-tum, obtained results similar to Visser and Wallenburg83:significantly decreased cardiac output and increased system-ic vascular resistance during preeclampsia, both groups hav-ing similar hemodynamic profiles at postpartum follow-up.

While the cardiovascular changes during overt preeclamp-sia seem clear, there are differences of opinions regarding thechanges that precede clinical presentation of the disease. Thegeneral impression is that there is evidence of a vasoconstric-tor state well before overt disease manifests. For instanceincreased sensitivity to pressor substances, increments in cir-culating antiangiogenic factors that affect the vasculature in amanner that opposes vasodilatation and lower intravascularvolume, precedes the disease by many weeks.84–88 Theseissues are explored further elsewhere in the text (Chapters 6and 15). However, there is an alternate view that womendestined to develop preeclampsia have an exaggeration ofthe normal increase in cardiac output, and a similarly exag-gerated low-resistance state in early pregnancy.89

FIGURE 14.8 Median values for cardiac output and total periph-eral resistance in the different patient populations plotted againstgestation. Normal: normotensive cohort; PET: preeclamptic cohort;GH: gestational hypertensive cohort. (Reprinted with permission.90)


Aview that preeclampsia is the end result of a pregnancyoriginally marked by an excessive increase in cardiac outputwith an exaggerated compensatory decrease in systemic pe-ripheral resistance has been championed by Easterlinget al.89 and Bosio et al.90 These authors liken the preclinicalphase of preeclampsia to that preceding overt essential hy-pertension. Messerli et al.91 described this prehypertensivephase as one of increased cardiac output accompanied by areflex decrease in systemic resistance, a protective mecha-nism against the appearance of elevated blood pressure.Eventually this autoregulation mechanism fails, afterloadincreases, cardiac output decreases, and hypertensionbecomes manifest. Thus in the combined views of Easter-ling89 and Bosio90 preeclampsia would occur in pregnantwomen with exaggerated increases in cardiac output, andnormal or slightly increased gestational falls in peripheralvascular resistance, but at the time the disease becomes overtthere is a “crossover” to a high-resistance low-output state. Ifthis theory were to prove correct, it would be logical to try toidentify women with exaggerated increases in cardiac outputearly in pregnancy and treat them with drugs that loweroutput such as beta-blockers, and indeed such a study hasappeared.92 However, as discussed below, we have a numberof problems with theories of Easterling89 and Bosio.90

Easterling et al.89 studied nulliparous gravidas seriallyusing noninvasive techniques to measure mean arterial pres-sure and cardiac output. Of 179 women, 89 had normalpregnancies, 9 developed preeclampsia, and 81gestationalhypertension. The women destined to develop preeclampsiahad higher mean arterial pressures and cardiac output, withlower systemic resistance (the latter not significant) beforedisease manifestations. In this study there were no signifi-cant changes detected in either the high cardiac output stateor in the reduced systemic resistance when overt signs andsymptoms of preeclampsia occurred. The higher initialblood pressures in the eventual preeclamptics is consistentwith findings of other investigators, but the early elevatedcardiac output with normal or reduced vascular resistancehad not been described previously. Of concern, there was ahigh drop-out rate, and only 9 women developed preeclamp-sia, mainly mild disease (averaging delivery at gestationalweek 39.4). Also, neonatal birth weight was similar to thecontrols. Other concerns were the use of a 1+ qualitativedetermination to define proteinuria.

Easterling et al.89 used a continuous wave Doppler ultra-sound system with no range-gating or imaging capabilities,and measurements of the aortic annulus were obtained byA-mode ultrasound, measurements that are more prone toangulation errors.58 The preeclamptic women had a muchhigher body mass than the control population, and there wasno correction for this factor in calculating cardiac output.The preeclamptics were on average 12 kg heavier thanthe nonpreeclamptics (BMI 29.8 kg/m2 vs. 24.9 kg/m2

calculated from the data provided) and thus “obesity,” afactor known to increase cardiac output,93 could explain

the increments recorded. Finally, the authors did not provideresults for the 89 women developing gestational hyperten-sion, a complication of pregnancy in many women whoeventually manifest essential hypertension. Given this re-markably high incidence of women with gestational hyper-tension in the study, (nearly half of the group), comparisonsof cardiac output and vascular resistance between the twohypertensive groups would have been instructive.

While the observations of Easterling et al.89 raised ques-tions, a second study by Bosio et al.90 appeared. It was a muchlarger (400 gravidas) longitudinal survey that utilized similarnoninvasive techniques. They too describe a high cardiacoutput state prior to evidence of clinical disease in the 20women who developed preeclampsia, but here systemic vas-cular resistance in the latent phase was definitely normal.After the “cross-over,” however, and similar to Visser andWallenberg83 and Lang et al.,2 but unlike Easterling et al.,89

they noted markedly decreased cardiac output and increasedsystemic vascular resistance. Bosio et al.90 chose not to correctcardiac output for maternal weight, but stated that even afteradjusting for BMI statistical significance was maintained; thedata however are not presented. Their group of 24 womendeveloping gestational hypertension remained with high car-diac output after disease manifestation90 (Fig. 14.8). These

FIGURE 14.9 Global arterial compliance index, a measure of theelasticity component of the entire system, both conduit and periph-eral vessels, is depicted for the 3 study groups: normal controls solidbar, preeclamptic patients striped bar, chronic hypertension withsuperimposed preeclampsia checked bar. *p < .05 compared withnormal control group. (Reprinted with permission.95)

FIGURE 14.10 For a matched pressure, the preeclampsia grouphad a lower compliance indicating that the disease process had anindependent effect on vascular properties: Average ACarea-NL–index-pressure relationships for the three groups (lines), where ACarea-NL–index is the global AC indexed to BSA and calculated using the areamethod and a pressure-dependent, nonlinear compliance model.ACarea-NL–index values (mean [SEM]) corresponding to the averagepressure during the diastolic period for the three groups are depictedby the solid symbols (control: 1.69 [0.12] mL mmHg�1 m�2; pre-eclampsia: 1.19 [0.12] mL mmHg�1 m�2; chronic hypertension(HTN) + preeclampsia: 0.94 [0.10] mL mmHg�1 m�2. These threeACarea-NL–index values are almost identical to the values estimatedusing the linear (standard) model. The open symbol representsACarea-NL–index for the preeclampsia group at a pressure equal tothat of the control group. (Reprinted with permission.94)


investigators have suggested that the hemodynamic changespromote endothelial injury and trigger the low-output vaso-constricted state of clinical disease.

The theory that preeclampsia may be the result of in-creased cardiac output early in gestation was discussed indetail because “minority” theories deserve scrutiny andsometimes surprise us all. Although some studies have sug-gested that women with exaggerated cardiac output in earlypregnancy be treated with beta receptor blocking drugs toprevent preeclampsia,92 this suggestion appears to be pre-mature. Finally, there would be a need to explain how thistheory is compatible with all the evidence that the preclinicalstate of preeclampsia is one marked by increased pressorresponses, lower intravascular volume, and increased levelsof circulating antiangiogenic proteins that impair thevessels’ ability to vasodilate.

Systemic Arterial Properties in Preeclampsia

CROSS-SECTIONAL STUDIESPreviously, effects of preeclampsia on systemic arterial loadhave been described mostly in terms of the steady compo-nent: peripheral vasoconstriction as indicated by an increasein systemic vascular resistance. More recent approaches in-clude assessment of the pulsatile component of arterial loadand the complex interactions of the components of thecardiovascular system in the face of a low-output, high-afterload disease state. Hibbard and colleagues94 performeda cross-sectional study comparing preeclamptics, chronichypertensives with superimposed preeclampsia, and normo-tensive women admitted in preterm labor. All were receivingmagnesium sulfate for either seizure prophylaxis or tocoly-sis, so two additional control groups, normal laboring wom-en with epidural anesthesia and a second normotensivelaboring group receiving neither epidural nor magnesium,were included in order to eliminate confounding factors.This study confirmed that total vascular resistance, thesteady component of the arterial load, was significantlyelevated in both hypertensive groups, but more so in thechronic hypertensives with superimposed preeclampsia.Global arterial compliance was significantly lower in thepure preeclamptic group and with an even greater decrementin chronic hypertensive women with superimposed pre-eclampsia95 (Fig. 14.9). The authors further showed that asubstantial component of the decreased global compliancewas not completely attributable to the rise in blood pres-sure, but independently related to preeclampsia per se94

(Fig. 14.10). These findings indicated the overall reservoirproperties of the systemic arterial circulation to be compro-mised. Lower compliance (or higher stiffness) was observedfor large conduit arteries as well. For example, the magni-tude of the first harmonic of aortic input impedance (Z1)was significantly elevated in both hypertensive groups95

(Fig. 14.11) and aortic characteristic impedance tended tobe greater in both hypertensive groups, though not signifi-

cantly so. Reflection index (RI), a measure of wave reflec-tions within the arterial system, was increased in thepreeclamptics and more so in those with chronic hyperten-sion with superimposed preeclampsia. Thus, during the latestages of preeclampsia, both steady and pulsatile compo-nents of arterial load are elevated as compared to the normalpregnancy.

Other studies have also reported similar observationsregarding lower arterial compliance and increased systemic

FIGURE 14.11 The magnitude of the first harmonic of the inputimpedance spectrum index, representing impedance properties ofboth large and small vessels, is depicted for the three study groups:normal controls solid bar, preeclamptic patients striped bar, chronichypertensionwith superimposed preeclampsia checked bar. *p < .05compared with normal control group. (Reprinted with permission.95)


vascular resistance during late gestation in preeclamptic sub-jects. For example, Elvan-Taspinar et al.96 and Tihtonenet al.97 have studied preeclamptic and chronic hypertensive(without superimposed preeclampsia) subjects during thethird trimester of pregnancy. The carotid-to-femoral pulsewave velocity (PWVcf) was used to quantify aortic stiffness(inverse of compliance) in the Elvan-Taspinar study96 andthe ratio of stroke index to pulse pressure (SI/PP) was used toquantify arterial compliance in the Tihtonen study.97 PWVcf

was significantly higher and SI/PP ratio was significantlylower in preeclamptic subjects compared to the values forthe control group (i.e., normal pregnancy), indicating re-duced arterial compliance in preeclampsia. Interestingly,stiffness (compliance) measures changed less in the chronichypertension group (without superimposed preeclampsia)compared to the preeclamptic group. In both studies, sys-temic vascular resistancewas greater in the preeclamptic andchronic hypertensive groups compared to the control group.

LONGITUDINAL STUDIESSerial measurements of systemic arterial properties havebeen reported. In nulliparous pregnant women studied lon-gitudinally, gravidas destined to develop preeclampsia werenoted to have elevated pulse pressures early in pregnancy,indicating higher arterial stiffness or lower compliance.98

Although pulse pressure is not as good a measure of pulsatilearterial load as global arterial compliance or aortic charac-teristic impedance, it is related to vascular compliance(stiffness) properties: high pulse pressure with similar strokevolume and heart rate corresponds to high stiffness or lowcompliance. Similarly, Oyama-Kato et al.99 have reportedincreased brachial-to-ankle pulse wave velocity throughoutpregnancy in subjects destined to develop preeclampsiacompared to normotensive controls, suggesting that arterialstiffness is significantly increased (or compliance is reduced)in these subjects. The findings of these two studies suggestthat the increase in arterial compliance (or decrease in arte-rial stiffness) seen in normal pregnancy is absent or signif-

icantly diminished in preeclampsia. Furthermore, thisaberrant compliance (stiffness) response is consistent withthe views that alterations in vascular reactivity leading to avasoconstricted state occur long before the development ofpreeclampsia.84,100

Differences in arterial properties noted in preeclampticsubjects during gestation may persist postpartum. Usingvenous occlusion plethysmography, studies have shown thatendothelial dysfunction persists in prior preeclamptics up toone year101 or 5–6 years102 postpartum. There are a numberof reports from the laboratory of Peeters and collea-gues,44,103,104 the results of which suggest that formerlypreeclamptic women have endothelial abnormalities, inde-pendent of whether or not they eventually develop chronichypertension after an index gestation complicated bypreeclampsia. Those who remained normotensive and hadendothelial dysfunction were termed “latent” hyperten-sives.104 Furthermore, when some of these women were rest-udied early in subsequent gestations (fifth to seventhgestational weeks), none had the normal rise in femoralvascular compliance.44 These results support observationsby Hibbard et al.94,95 in normal gestations, but also suggestthat those with a history of preeclampsia have impairedvascular compliance long-term. Haukkamaa et al.105 pre-sented pulse-wave velocity evidence for long-term vascularabnormalities in previously preeclamptic women, but not allinvestigators find support for this notion.106

Whether decreased arterial compliance (or increasedstiffness) persists after pregnancy in those women who suf-fered preeclampsia is not known. It would be instructive toapply our techniques described above to such women in alongitudinal study. Finally, a problem with most of theseinvestigations is one similar to long-term follow-up studiesof cardiovascular risk in preeclampsia: preeclamptic womenare compared to those with normotensive deliveries. Theformer population harbors more women destined to havechronic hypertension as well as cardiovascular and metabol-ic disorders later in life. In essence preeclampsia is a riskmarker for these disorders while normotensive deliveriesselect a particularly healthy population, making it difficultto ascribe observations made postpartum to the risk theyhave been marked for or to the preeclampsia per se!

Left Ventricular Properties in Preeclampsia

Left ventricular muscle mass and end-diastolic diameterduring late gestation were similar between preeclampticand control subjects,2 indicating that left ventricular struc-tural changes in preeclampsia are similar to those observedin normal pregnancy.

Although a number of studies have reported data regard-ing cardiovascular function in preeclampsia (e.g., cardiacoutput, left ventricular ejection phase indices such as frac-tional shortening),75–78,107,108 few studies have examinedleft ventricular myocardial contractility using techniques

ttpeTliocwTcssupiamiwoImecdn

V

Frwauiw


hat eliminate the confounding effects of loading condi-ions.2,62,63 Lang et al.2 carefully selected ten preeclampticrimigravid women, strictly defining the disease,85 withoutvidence of chronic hypertension, cardiac or renal disease.he subjects were studied at three time points: 1) prior toabor and prior to administering any antihypertensive med-cations, though magnesium sulfate had been initiated, 2)ne day postpartum when magnesium sulfate had been dis-ontinued and subjects were still hypertensive, and 3) foureeks after delivery when the subjects were normotensive.en normotensive gravidas who underwent identical proto-ols served as controls. As discussed before, end-systolictress (sES)-rate corrected velocity of circumferential fiberhortening (Vcfc) data were used to characterize left ventric-lar myocardial contractility. As depicted in Figure 14.12,reeclamptic subjects had sES–Vcfc values that were super-mposed upon the normal contractility line (the nomogram)t each study point, results similar to those obtained in nor-otensive controls2 (Fig. 14.6). Proceeding from study 1 to 3n preeclamptics, sES–Vcfc points shifted leftward and up-ard along the nomogram, indicating decreased afterloadver timewithout any change in left ventricular contractility.n other words there was decreased left ventricular perfor-ance with acute preeclampsia, but when afterload wasliminated as a confounding variable, it was obvious thatontractility remained normal. This was the appropriate car-iovascular response in the face of increased afterload, andot a pathologic response.In another investigation employing left ventricular sES–

cfc data to evaluate myocardial contractility in preeclamp-

IGURE 14.12 Average end-systolic wall stress (sES)-rate cor-ected velocity of fiber shortening (Vcfc) data obtained in patientsith preeclampsia before delivery, 1 day after delivery, and 4 weeksfter delivery. From visit 1 to visit 3, data points shifted leftward andpward (arrow) but still fell on the mean contractility line, indicat-ng decreased afterload without changes in contractility. (Reprintedith permission.2)

sia, Simmons et al.23 studied 15 preeclamptic gravidas andcompared them to 44 normal controls. They reported similarmyocardial contractility in the two groups, thereby confirm-ing the findings of Lang et al.2 that left ventricular contrac-tility is unchanged in preeclampsia.

FACTORS THAT MAY EXPLAIN VASCULARCHANGES IN PREGNANCY

Normal Pregnancy

AUTONOMIC NERVOUS SYSTEMAutonomic regulation of the cardiovascular system is aprime candidate for modification during gestation, but itsstudy is complicated by baseline differences in heart rate,blood pressure, and blood volume that accompany pregnan-cy.109 In this text’s first edition, Chesley reviewed strikingobservations made during the 1950s and 1960s, demonstrat-ing the importance of the autonomic nervous system inmaintaining blood pressure when changing posture. Whena pregnant woman lies supine, autonomic blockade withtetramethyl ammonium or spinal anesthesia resulted inmarked hypotension, alleviated by assuming a lateral recum-bent position, while similar treatment of nonpregnant sub-jects had but minimal effects. These studies suggested thatthe predominant effect of autonomic blockade was through aloss of vasomotor tone. As noted, a reduction in afferentoutflow to the resistance vasculature could explain the re-duction in systemic vascular resistance. However, very fewin-depth studies followed these observations.

Two studies in animal models are of interest. Pan et al.,110

in an extensive study of blood pressure regulation in preg-nant rats, observed similarly large decreases in blood pres-sure (40–50%) after either ganglionic or selective alphaadrenoreceptor blockade in gravid and nonpregnant animals,suggesting that the adrenergic nervous system was of quan-titatively similar importance in the pregnant and nonpreg-nant states of that species. Assessing nervous activitydirectly would be needed to confirm that this smooth muscleresponse to sympathetic neurotransmitters might be differentin the pregnant animal. Such studies were performed byO’Hagen and Casey.111 They evaluated renal sympatheticactivity in chronically instrumented rabbits, eliminatinganesthesia as a confounder, and observed no effect of preg-nancy on renal sympathetic nerve activity in late gestation.

At first glance it would seem that postural differencesshould make comparisons between rodents and humans im-prudent. Howeverwe have cited the above studies, in relationto the carefully conducted studies in humans by Schobel etal.112 that are detailed further below and in Chapter 15. Theselatter investigators, utilizing peroneal nerve microneurologytechniques, also noted no differences in basal sympatheticactivity when age-matched pregnant and nonpregnant wom-en were compared, their data, too, suggesting that pregnancy


does not affect basal regulation of vascular tone by the sym-pathetic nervous system.

There is a limited animal literature regarding autonomicnervous system function in pregnancy relating to pressure-mediated changes in heart rate and blood pressure, but bar-oreflex regulation of heart rate involves both vagal and sym-pathetic nerve affects.109,111 More recent studies controlledfor this by measuring the response to a variety of sti-muli.109,111 Also we could locate no studies relating to af-ferent inputs to the central nervous system in response topressure stimuli.

The most meaningful studies, of course, are those in theconscious state. However results in both animals and humansgive discordant results, as the baroreflex-mediated heart rateresponse to increasing blood pressure has been reported asenhanced, unchanged, or depressed in pregnancy.113–118

These discrepancies may relate to the period in pregnancystudied, as well as differences in control of heart rate be-tween the pregnant and nonpregnant states. Of interest hereis that in those experiments where the normal gestationaldifference in resting heart rate is present prior to imposedpressure steps, the reflex tachycardia is accentuated by ges-tation109,118,119 suggesting that pregnancy augments sympa-thetic activity to the heart in response to elevations in bloodpressure. (In these same studies, pregnancy had little effecton heart rate to decreases in blood pressure.)

The tachycardic response to hypotension appears unaf-fected by pregnancy, but this does not reflect the overallsympathetic response to hypotension. Pregnant rats showan attenuated ability to increase sympathetic nerve outputin response to a hypotensive challenge.115 Pregnancy doesnot seem to affect basal postganglionic sympathetic nerveactivity in skeletal blood vessels, nor is there increased ac-tivity in response to a cold pressor test.112

Finally, while alterations in the autonomic nervous sys-tem during pregnancy appear more related to pressure sti-muli than to changing basal tonemaintenance per se, we notethat vascular reactivity to neurotransmitters is reduced dur-ing normal gestation,120–122 and call attention to a landmarkstudy that addressed adrenergic vascular reactivity, per-formed in 1985 by Nisell and colleagues.122 They combinedmeasuring blood levels of administered agonists, calf bloodflow, and cardiac output in both pregnant women and non-pregnant controls, and were able to simultaneously verifythat both groups were receiving equivalent stimuli, and todetermine the different components of the pressor response.The absolute pressor response to norepinephrine was similarbetween groups, but the rise in pregnant women was duesolely to increased cardiac output, while the pressure rise inthe nonpregnant state was due to vasoconstriction. This ob-servation is an elegant example of why examining the changein blood pressure alone is not necessarily a measure of sys-temic vascular reactivity. Using the identical change in bloodpressure alone, one might erroneously conclude there wereno changes in pregnancy. However, the simultaneous record-

ing of cardiac output permitted the correct conclusion (nor-mal pregnancy had blunted the systemic response tonorepinephrine). Since as noted above basal sympatheticoutflow does not decrease during human gestation, at leastin the skeletal bed,112 then the vascular response to vasocon-striction may be reduced. Thus, a given level of nervousactivity may produce a different vascular response duringpregnancy, compared to the nonpregnant state. There mayalso be local mechanisms within the arterial tree itself thataccount for the relaxation observed during pregnancy, per-haps due to humeral signals deriving from ovary, placenta, orpossibly from the pituitary. In summary, we still have a lot tolearn about the autonomic nervous system in pregnancy.

VASCULAR WALL REMODELING AND SMOOTH MUSCLE

TONE

Both increased vascular distensibility and the presence of theuteroplacental circulation could contribute to the observedincrease in global arterial compliance during normal preg-nancy. The second possibility is unlikely to be a major factorbecause compliance changes occur very rapidly (early in thefirst trimester) and several studies have shown that most ofthe pregnancy-associated hemodynamic changes can bereproduced simply by sex steroid50,55 or other hormon-al123,124 administrations. Potential mechanisms for increasedarterial distensibility can be divided into three categories: 1)passive changes in vessel wall properties secondary to re-duced distending pressure, 2) vascular wall remodeling, and3) reduced smooth muscle tone. The first factor is unlikely toplay a major role because changes in distending pressureduring normal pregnancy are very small. Vascular wall remo-deling is comprised of geometric (e.g., increase in vascularwall area) and compositional (e.g., relative amounts of wallconstituents such as elastin and collagen) components. Osoland colleagues125–127 have reported that uterine radial arter-ies from late pregnant rats were characterized by a signifi-cant increase in vascular wall cross-sectional area whencompared to those from nonpregnant rats, and smooth mus-cle cell hypertrophy and hyperplasia were underlying factorsin the observed increase in vascular wall area. Griendlinget al.128 have reported similar smooth muscle hypertrophy ofuterine arteries during late pregnancy in sheep and this wasaccompanied by compositional remodeling (decreased col-lagen and unchanged elastin). Interestingly, carotid arteriesdid not show any geometric or compositional remodeling.Mackey et al.129 observed reduced collagen and elastin con-centrations in rat mesenteric resistance arteries during latepregnancy. Thus, geometric and/or compositional remodel-ing seems to be occurring during late pregnancy in an artery-specific manner. However, it not entirely clear whether thesechanges are present early in pregnancy and contribute to theobserved alterations in the vascular mechanical properties.Finally, ample evidence exists that a reduction in smoothmuscle tone contributes to the changes in arterial properties


in normal pregnancy. Factors responsible for this includehormonal signals, particularly estrogens130–132 or relaxin,123

or the release of endothelial relaxing factors such as NO orVEGF or PlGF or circulating angiogenic factors.88

HORMONAL SIGNALS

Estrogen decreases smooth muscle tone in animals andhumans,132–135 lowers blood pressure and carotid-femoralpulse wave velocity in postmenopausal women,135,136 andalso promotes endothelium-dependent vasodilation in wom-en with premature ovarian failure.137 Activation of estrogenreceptor alpha modulates nitric oxide production and theresponse of smoothmuscle relaxation in segments of isolatedmouse aorta.138

The peptide relaxin, produced primarily by the corpusluteum, has been investigated as a candidate hormone pro-moting the vascular changes in normal pregnancy. As de-tailed in Chapter 17, relaxin may be the hormone mostresponsible for the physiologic gestational alterations in re-nal hemodynamics.139,140 Similarly, chronic administrationof recombinant human relaxin to conscious, female, non-pregnant rats a) reduced the steady arterial load by decreas-ing systemic vascular resistance, b) increased cardiac output,and c) reduced the pulsatile arterial load as assessed byindices of systemic arterial compliance.124 Interestingly, ex-ogenous relaxin administration to male rats at doses thatwere efficacious in females increased both cardiac outputand systemic arterial compliance, and reduced systemic ar-terial resistance.141 Finally, neutralization of endogenouscirculating relaxin through administration of specific ratrelaxin antibodies during early gestation abolished thechanges in cardiac output, systemic vascular resistance andglobal arterial compliance observed in rats infused with anirrelevant antibody.123 Thus, relaxin appears to play an im-portant role in many of the systemic arterial (hemodynamicsand mechanical properties) and renal changes during normalpregnancy,142 especially during early gestation.Mechanismsof how relaxin induces decreased smooth muscle tone aredetailed in Chapter 17. Relaxin activates the endothelial ETbreceptor/nitric oxide vasodilatory pathway,142 stimulatesvascular endothelial growth factor (VEGF) in vitro,143 andhas angiogenic properties.142

Carbillon et al.144 reviewed the evidence that nitric oxideplays a role in the regulation of vascular tone in pregnancy,contributing to the low systemic resistance. That a nitricoxide-mediated pathway may contribute to the cardiovascu-lar changes described above was addressed by Williams etal.145 in a pregnant dog model. These investigators, studyingarterioles from the dogs, suggested that nitric oxide producedby endothelium enhanced release of nitrites, exerting greatercontrol over shear stress-induced vasodilation, promotingcoupling of oxygen delivery and efficiency of the heart.Angiogenic factors such as VEGF and TGF-b1 maintainvascular homeostasis and endothelial health in normal preg-

nancy88 and are discussed in Chapter 6. Their roles in ges-tational cardiovascular changes remain to be explored.

Preeclampsia

Imaging and other technologies used to probe the cardiovas-cular system were discussed above, the data reviewed sug-gesting that all components of the arterial system are affectedby preeclampsia. Factors responsible for the pathologicalcardiovascular changes are not completely understood.

An attempt to explain the vascular changes in pre-eclampsia would be mostly speculation, and we will focuson the meaning of a few experimental observations. Asnoted in the section devoted to the autonomic nervoussystem, Nisell et al.122 infused norepinephrine in normalpregnant and nonpregnant women and measured cardiacoutput during the infusion. Blood pressure in the normalpregnant group increased secondary to an increase in car-diac output alone while in the nonpregnant state pressureincrements were due to increased systemic vascular resis-tance and a small decrease in cardiac output. These inves-tigators also studied preeclamptic women and in thesesubjects the enhanced pressure response was secondaryto an exaggerated rise in systemic vascular resistance.These results are consistent with observations that vascularreactivity is attenuated in normal pregnancy, but augment-ed in preeclamptics.

In studies of Schobel,112 described above and in Chapter15, postganglionic sympathetic nerve activity was three-foldhigher in preeclamptic women, the values normalizing post-partum, coinciding with the return of their blood pressures tonormal nonpregnant levels. The observations of sympatheticoveractivity correlate with the increments in total vascularresistance, decreased arterial compliance and increased im-pedance that accompanied the increased blood pressure not-ed in our own studies.94 We note though that theseobservations are inconsistent with old studies where phar-macological blockade of the autonomic system had no effecton the elevated blood pressure of preeclamptic wom-en.146,147 They are also at variancewith data indicating lowerheart rates in preeclampsia.

A final hint from human studies permits speculation thatvascular remodeling may be involved in the cardiovascularfindings of preeclampsia. Omental resistance arteries frompreeclamptic women were analyzed demonstrating irregularthickness of the elastic lamina, incomplete basement mem-brane, and a changed location and arrangement of endothe-lial cells compared to normotensive controls as demonstratedby electron microscopy.148 These findings, however, couldbe secondary to the disease process rather than the incitingevent that stimulates cardiovascular changes.

Genetic factors addressed in Chapter 4 may also be inte-gral while a new chapter (Chapter 6) details the roleof circulating antiangiogenic factors and their effects onvasodilation.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2


Summary

A host of changes in the cardiovascular system occurduring the course of a normal pregnancy and includeremarkable increases in cardiac output, arterial compli-ance, and total blood volume while blood pressure andsystemic vascular resistance fall, all in the face of main-taining normal myocardial contractility. An impressivereversal of many of these parameters occurs prior toand during the acute onset of preeclampsia commencingwith a marked reversal of the vasculature’s resistance topressor hormones, culminating not only in a markedvascular sensitivity but hypertension characterized by alow cardiac output, high systemic vascular resistant state.There may be a hyperdynamic, low-resistance diseasestate preceding this but whether this represents a subsetof women who have or will eventually have chronichypertension remains to be determined. Such a statehowever is hard to reconcile with the early change invascular reactivity, the knowledge that volume decreasesbefore the advent of hypertension, and the exaggeratedrise in circulating antiangiogenic proteins that impedevasodilation.

Both steady and pulsatile arterial load fail to decrease inpreeclampsia, as occurs in normal pregnancy, involvingchanges in both conduit and small vessels. However, theability of the cardiovascular system to adapt, and for theheart to continue functioning with normal myocardial con-tractility during preeclampsia has been demonstrated by ourgroup. Abnormal adaptive mechanisms may be secondary tochanges in vascular tone or vascular wall elements, changesthat are just becoming understood, and may have implica-tions for a woman later in life.

References

1. Lindhard J. Uber das Minutenvolume des Herzens bei Ruheund bei Muskelarbeit. Pflugers Arch. 1915;161:223.

2. Lang RM, Pridjian G, Feldman T, Neumann A, LindheimerM, Borow KM. Left ventricular mechanics in preeclampsia.Am Heart J. 1991;121:1768–1775.

3. Christianson RE. Studies on blood pressure during pregnancy.I. Influence of parity and age. Am J Obstet Gynecol. 1976;125:509–513.

4. Chapman AB, Zamudio S, Woodmansee W, et al. Systemicand renal hemodynamic changes in the luteal phase of themenstrual cycle mimic early pregnancy. Am J Physiol.1997;273:F777–F782.

5. Moutquin JM, Rainville C, Giroux L, Raynauld P, Amyot G,Bilodeau R, Pelland N. A prospective study of blood pressurein pregnancy: Prediction of preeclampsia. Am J Obstet Gyne-col. 1985;151:191–196.

6. Duvekot JJ, Cheriex EC, Pieters F, Menheere P, Peeters L.Early pregnancy changes in hemodynamics and volume ho-meostasis are consecutive adjustments triggered by a primaryfall in systemic vascular tone. Am J Obstet Gynecol. 1993;169:1382–1392.

7. Duvekot JJ, Peeters LLH. Maternal cardiovascular hemody-namic adaptation to pregnancy.Obstet Gynecol Surv. 1994;49(Suppl):S1–S14.

8. Clapp JF, Capeless E. Cardiovascular function before, duringand after the first and subsequent pregnancies. Am J Cardiol.1997;80:1469–1473.

9. Robson SC, Hunter S, Boys RJ, Dunlop W. Serial study offactors influencing changes in cardiac output during humanpregnancy. Am J Physiol. 1989;256:H1060–H1065.

0. Poppas A, Shroff SG, Korcarz CE, Hibbard JU, Berger DS,Lindheimer MD, Lang RM. Serial assessment of the cardio-vascular system in normal pregnancy: Role of arterial com-pliance and pulsatile arterial load. Circulation. 1997;95:2407–2415.

1. van Oppen AC, Van der Tweel I, Alsbach GP, Heethaar RM,Bruinse HW. A longitudinal study of maternal hemodynamicsduring normal pregnancy. Obstet Gynecol. 1996;88:40–46.

2. Gilson GJ, Samaan S, Crawford MH, Qualls CR, Curet LB.Changes in hemodynamics, ventricular remodeling, and ven-tricular contractility during normal pregnancy: A longitudinalstudy. Obstet Gynecol. 1997;89:957–962.

3. Lees MM, Taylor SH, Scott DB, Kerr MG. A study of cardiacoutput at rest throughout pregnancy. J Obstet Gynaecol BrCommonw. 1967;74:319–328.

4. Capeless EL, Clapp JF. Cardiovascular changes in early phaseof pregnancy. Am J Obstet Gynecol. 1989;161:1449–1453.

5. Page EW, Christianson R. The impact of mean arterial pres-sure in the middle trimester upon the outcome of pregnancy.Am J Obstet Gynecol. 1976;125:740–746.

6. Hunter S, Robson SC. Adaptation of the maternal heart inpregnancy. Br Heart J. 1992;68:540–543.

7. Desai DK, Moodley J, Naidoo DP. Echocardiographic assess-ment of cardiovascular hemodynamics in normal pregnancy.Obstet Gynecol. 2004;104:20–29.

8. Beilin LJ, Deacon J, Michael CA, Vandongen R, Lalor CM,Barden AE, Davidson L. Circadian rhythms of blood pressureand pressor hormones in normal and hypertensive pregnancy.Clin Exp Pharmacol Physiol. 1982;9:321–326.

9. Seligman SA. Diurnal blood-pressure variation in pregnancy.J Obstet Gynaecol Br Commonw. 1971;78:417–422.

0. Benedetto C, Zonca M, Marozio L, Dolci C, Carandente F,Massobrio M. Blood pressure patterns in normal pregnancyand in pregnancy-induced hypertension, preeclampsia, andchronic hypertension. Obstet Gynecol. 1996;88:503–510.

1. Hermida RC, Ayola DE. Prognostic value of office and am-bulatory blood pressure measurements in pregnancy. Hyper-tension. 2002;40:298–303.

2. Mabie WC, DiSessa TG, Crocker LG, Sibai BM, Arheart KL.A longitudinal study of cardiac output in normal human preg-nancy. Am J Obstet Gynecol. 1994;170:849–856.

3. Simmons LA, Gillin AG, Jeremy RW. Structural and func-tional changes in left ventricle during normotensive and pre-eclamptic pregnancy. Am J Physiol Heart Circ Physiol.2002;283:H1627–H1633.

4. Mone SM, Sanders SP, Colan SD. Control mechanisms forphysiological hypertrophy of pregnancy. Circulation. 1996;94:667–672.

5. Geva T,MauerMB, Striker L, KirshonB, Pivarnik JM. Effectsof physiologic load of pregnancy on left ventricular contrac-tility and remodelling. Am Heart J. 1997;133:53–59.


26. Del Bene R, Barletta G, Mello G, et al. Cardiovascular func-tion in pregnancy: effects of posture. Br J Obstet Gynaecol.2001;108:344–352.

27. Monos E, Berczi V, Nadasy G. Local control of veins: Bio-mechanical, metabolic, and humoral aspects. Physiol Rev.1995;75:611–666.

28. Guyton AC, Polizo D, Armstrong GG. Mean circulatory fill-ing pressure measured immediately after cessation of heartpumping. Am J Physiol. 1954;179:261–267.

29. Humphreys PW, Joels N. Effect of pregnancy on pressure-volume relationships in circulation of rabbits. Am J Physiol.1994;267:R780–R785.

30. Rothe CF. Physiology of venous return. An unappreciatedboost to the heart. Arch Intern Med. 1986;146:977–982.

31. Cha SC, Aberdeen GW, Nuwayhid BS, Quillen EW Jr. Influ-ence of pregnancy on mean systemic filling pressure and thecardiac function curve in guinea pigs. Can J Physiol Pharma-col. 1992;70:669–674.

32. Douglas BH, Harlan JC, Langford HG, Richardson TQ. Effectof hypervolemia and elevated arterial pressure on circulatorydynamics of pregnant animals. Am J Obstet Gynecol.1967;98:889–894.

33. Goodlin RC, Niebauer MJ, Holmberg MJ, Zucker IM. Meancirculatory filling pressure in pregnant rabbits. Am J ObstetGynecol. 1984;148:224–225.

34. Tabsh K, Monson R, Nuwayhid B. Regulation of cardiacoutput in ovine pregnancy. Society for Gynecologic Investi-gation 1995; 32nd meeting, 53P.(abstr).

35. Davis LE, Hohimer AR, Giraud GD, Paul MS, Morton MJ.Vascular pressure-volume relationships in pregnant and estro-gen-treated guinea pigs. Am J Physiol. 1989;257:R1205–R1211.

36. Richardson TQ, Stallings JO, Guyton AC. Pressure-volumecurves in live, intact dogs. Am J Physiol. 1961;201:471–474.

37. Yamamoto J, Trippodo NC, Ishise S, Frohlich ED. Total vas-cular pressure-volume relationship in the conscious rat. Am JPhysiol. 1980;238:H823–H828.

38. Hohmann M, Keve TM, Osol G, McLaughlin MK. Norepi-nephrine sensitivity of mesenteric veins in pregnant rats. Am JPhysiol. 1990;259:R753–R759.

39. Hohmann M, Mackey K, Davidge S, McLaughlin MK. Ve-nous remodeling in the pregnant rat. Clin Exp Hypertens B.1991;10:307–321.

40. Fawer R, DettlingA,Weihs D,Welti H, Schelling JL. Effect ofthe menstrual cycle, oral contraception and pregnancy onforearm blood flow, venous distensibility and clotting factors.Eur J Clin Pharmacol. 1978;13:251–257.

41. Goodrich SM, Wood JE. Peripheral venous distensibility andvelocity of venous blood flow during pregnancy or during oralcontraceptive therapy. Am J Obstet Gynecol. 1964;90:740–744.

42. Pickles CJ, Brinkman CR, Stainer K, Cowley AJ. Changes inperipheral venous tone before the onset of hypertension inwomen with gestational hypertension. Am J Obstet Gynecol.1989;160:678–680.

43. Sakai K, Imaizumi T, Maeda H, Nagata H, Tsukimori K,Takeshita A, Nakano H. Venous distensibility during pregnan-cy: Comparisons between normal pregnancy and preeclamp-sia. Hypertension. 1994;24:461–466.

44. Spaanderman MEA, Willekes C, Hoeks APG, Ekhart THA,Peeters LLH. The effect of pregnancy on the compliance oflarge arteries and veins in healthy parous control subjects andwomen with a history of preeclampsia. Br J Obstet Gynaecol.2000;183:1278–1286.

45. Smith AJ, Walters WA, Buckley NA, Gallagher L, Mason A,McPherson J. Hypertensive and normal pregnancy: A longi-tudinal study of blood pressure, distensiblility of dorsal handveins and the ratio of the stable metabolites of thromboxaneA2 and prostacyclin in plasma. Br J Obstet Gynaecol.1995;102:900–906.

46. Duncan SL, Bernard AG. Venous tone in pregnancy. J ObstetGynaecol Br Commonw. 1968;75:142–150.

47. Edouard DA, Pannier BM, London GM, Cuche JL, Safar ME.Venous and arterial behavior during normal pregnancy. Am JPhysiol. 1998;274:H1605–H1612.

48. Robson SC, DunlopW,MooreM, Hunter S. Combined Dopp-ler and echocardiographic measurement of cardiac output:theory and application in pregnancy. Br J Obstet Gynaecol.1987;94:1014–1027.

49. Shroff SG. Pulsatile arterial load and cardiovascular function– facts, fiction and wishful thinking. Therapeutic Res.1998;19:59–66.

50. Hart MV, Hosenpud JD, Hohimer AR, Morton MJ. Hemody-namics during pregnancy and sex steroid administration inguinea pigs. Am J Physiol. 1985;249:R179–R185.

51. Slangen BF, van Ingen Schenau DS, van Gorp AW, De MeyJG, Peeters LL. Aortic distensibility and compliance in con-scious pregnant rats. Am J Physiol. 1997;272:H1260–H1265.

52. Danforth DN, Manalo-Estrella P, Buckingham JC. The effectof pregnancy and of enovid on the rabbit vasculature. Am JObstet Gynecol. 1964;88:952–962.

53. Mackey K, Meyer MC, Stirewalt WS, Starcher BC,McLaughlin MK. Composition and mechanics of mesentericresistance arteries from pregnant rats. Am J Physiol. 1992;263:R2–R8.

54. Hart MV, Morton MJ, Hosenpud JD, Metcalfe J. Aortic func-tion during normal human pregnancy. Am J Obstet Gynecol.1986;154:887–891.

55. Slater AJ, Gude N, Clarke IJ, Walters WAW. Haemodynamicchanges and left ventricular performance during high-doseoestrogen administration to male transsexuals. Br J ObstetGynaecol. 1986;93:532–538.

56. Katz R, Karliner JS, Resnick R. Effects of a natural volumeoverload state (pregnancy) on left ventricular performance innormal human subjects. Circulation. 1978;58:434–441.

57. Rubler S, Damani PM, Pinto ER. Cardiac size and perfor-mance during pregnancy estimated with echocardiography.Am J Cardiol. 1977;40:534–540.

58. Lee W, Rokey R, Cotton DB. Noninvasive maternal strokevolume and cardiac output determinations by pulsed Dopplerechocardiography. Am J Obstet Gynecol. 1988;158:505–510.

59. Easterling TR, Benedetti TJ, Schmucker BC, Carlson K,Mill-ard SP. Maternal hemodynamics and aortic diameter in nor-mal and hypertensive pregnancies. Obstet Gynecol.1991;78:1073–1077.

60. Immer FF, Bansi AG, Immer-Bansi AS, McDougall J, ZehrKJ, Schaff HV, Carrel TP. Aortic dissection in pregnancy:analysis of risk factors and outcome. Ann Thorac Surg.2003;76:309–314.


61. Practice Committee of the American Society for Reproduc-tive Medicine: Increased maternal cardiovascular mortalityassociated with pregnancy in womenwith Turner’s syndrome.Fertil Steril. 2006;86(5 Suppl):S127–S128.

62. Borghi C, Esposti DD, Immordio V, Cassani A, Boschi S,Bovicelli L, Ambrosioni E. Relationship of systemic hemody-namics, left ventricular structure and function, and plasmanatriuretic peptide concentrations during pregnancy complicat-ed by preeclampsia. Am J Obstet Gynecol. 2000;183:140–147.

63. Scardo J, Kiser R, Dillon A, Brost B, Newman R. Hemody-namic comparison of mild and severe preeclampsia: conceptof stroke systemic vascular resistance index. J Mater FetalMed. 1996;5:268–272.

64. Colan SD, Borow KM, Neumann A. The left ventricular end-systolic wall stress-velocity of fiber shortening relation: aload-independent index of myocardial contractility. JAmCollCardiol. 1984;4:715–724.

65. Fisher SJ, Roberts JM. Defects in placentation and placentalperfusion. In: Lindheimer MD, Roberts JM, Cunningham FG,eds. Chesley’s Hypertensive Disorders of Pregnancy. 2nd ed.Stamford: Appleton and Lang; 1999:377–394. (Revised asChapter 5 of this edition).

66. Reiss RE, O’Shaughnessy RW, Quilligan TJ, Zuspan FP. Ret-rospective comparison of blood pressure course during pre-eclamptic and matched control pregnancies. Am J ObstetGynecol. 1987;156:894–898.

67. Conde-AgudeloA, Belizan JM, Lede R, Bergel EF.What doesan elevated mean arterial pressure in the second half of preg-nancy predict-gestational hypertension or preeclampsia? AmJ Obstet Gynecol. 1993;169:509–514.

68. Sibai BM, Gordon T, Thom E, Zuspan FP. Risk factors forpreeclampsia in healthy nulliparous women: A prospectivemulticenter study. The NIH Maternal Fetal Medicine Units.Am J Obstet Gynecol. 1995;172:642–648.

69. Sibai BM, Ewell M, Levine RJ, et al. Risk factors associatedwith preeclampsia in healthy nulliparous women. The Calci-um for Preeclampsia Prevention (CPEP) Study Group. Am JObstet Gynecol. 1997;177:1003–1010.

70. Conde-Agudelo A, Villar J, Lindheimer M. World HealthOrganization systematic review of screening tests for pre-eclampsia. Obstet Gynecol. 2004;104:1367–1391.

71. Redman CWG, Beilin LJ, Bonnar J. Reversed diurnal bloodpressure rhythm in hypertensive pregnancies. Clin Sci MolMed Suppl. 1976;51:687s–689s.

72. Beilin LJ, Deacon J, Michael CA, Vandongen R, Lalor CM,Barden AE, Davidson L. Circadian rhythms of blood pressureand pressor hormones in normal and hypertensive pregnancy.Clin Exp Pharmacol Physiol. 1982;9:321–326.

73. Malberg H, Bauernshcmitt R, Voss A, Walther T, Faber R,Stepan H,Wessel N. Analysis of cardiovascular oscillations: anew approach to the early prediction of pre-eclampsia.Chaos.2007;17:015113.

74. Walther T,Wessel N,Malberg H, Voss A, Stepan H, Faber R. Acombined technique for predicting pre-eclampsia: concurrentmeasurement of uterine perfusion and analysis of heart rateand blood pressure variability. J Hypertens. 2006;24:747–750.

75. Benedetti TJ, Cotton DB, Read JC, Miller FC. Hemodynamicobservations in severe pre-eclampsia with a flow-directedpulmonary artery catheter. Am J Obstet Gynecol. 1980;136:465–470.

76. Strauss RG, Keefer JR, Burke T, Civetta JM. Hemodynamicmonitoring of cardiogenic pulmonary edema complicatingtoxemia of pregnancy. Obstet Gynecol. 1980;55:170–174.

77. Rafferty TD, Berkowitz RL. Hemodynamics in patients withsevere toxemia during labor and delivery. Am J Obstet Gyne-col. 1980;138:263–270.

78. Phelan JP, Yurth DA. Severe preeclampsia: I. Peripartum he-modynamic observations. Am J Obstet Gynecol. 1982;144:17–22.

79. Groenendijk R, Trimbos JBMJ, Wallenburg HCS. Hemody-namic measurements in preeclampsia: preliminary observa-tions. Am J Obstet Gynecol. 1984;150:232–236.

80. Clark SL, Greenspoon JSA, Aldahl D, Phelan JP. Severepreeclampsia with persistent oliguria: Management of hemo-dynamic subsets. Am J Obstet Gynecol. 1986;154:490–494.

81. Cotton DB, Lee W, Huhta JC, Dorman KF. Hemodynamicprofile of severe preeclampsia-induced hypertension. Am JObstet Gynecol. 1988;158:523–529.

82. Mabie WC, Ratts TE, Sibai BM, et al. The central hemody-namics of severe preeclampsia. Am J Obstet Gynecol. 1989;161:1443–1448.

83. Visser W, Wallenburg HCS. Central hemodynamic observa-tions in untreated preeclamptic patients. Hypertension.1991;17:1072–1077.

84. Gant NF, Daley GL, Chand S, Whalley PJ, MacDonald PC. Astudy of angiotensin II pressor response throughout primigra-vid pregnancy. J Clin Invest. 1973;52:2682–2689.

85. Lindheimer MD, Katz AI. Preeclampsia: pathophysiology,diagnosis, and management. Annu Rev Med. 1989;40:233–250.

86. Maynard SE, Min JY, Merchan J, et al. Excess placentalsoluble fms-like tyrosine kinase 1 (sFlt1) may contribute toendothelial dysfunction, hypertension, and proteinuria in pre-eclampsia. J Clin Invest. 2003;111:649–658.

87. Levine RJ, Maynard SE, Qian C, et al. Circulating angiogenicfactors and the risk of preeclampsia. N Eng J Med. 2004;350:672–683.

88. Maynard S, Epstein FH, Karumanchi SA. Preeclampsia andangiogenic imbalance. Annu Rev Med. 2008;59:61–78.

89. Easterling TR, Benedetti TJ, Schmucker BC, Millard SP.Maternal hemodynamics in normal and preeclamptic preg-nancies: A longitudinal study. Obstet Gynecol. 1990;76:1061–1069.

90. Bosio PM, McKenna PJ, Conroy R, O’Herlihy C. Maternalcentral hemodynamics in hypertensive disorders of pregnan-cy. Obstet Gynecol. 1999;94:978–984.

91. Messerli FH, De Carvalho JGR, Christie B, Frohlich ED.Systemic and regional hemodynamics in low, normal and highcardiac output borderline hypertension. Circulation. 1978;58:441–448.

92. Easterling TR, Brateng D, Schmucker B, Brown Z, MillardSP. Prevention of preeclampsia: A randomized trial of ateno-lol in hyperdynamic patients before onset of hypertension.Obstet Gynecol. 1999;93:725–733.

93. Hall JE. Renal and cardiovascular mechanisms of hyperten-sion in obesity. Hypertension. 1994;23:381–394.

94. Hibbard JU, Korcarz CE, Nendaz-Giardet G, LindheimerMD, Lang RM, Shroff SG. Arterial system in preeclampsiaand chronic hypertension with superimposed preeclampsia.Br J Obstet Gynecol. 2005;112:897–903.


95. Hibbard JU, Shroff SG, Lang RM. Cardiovascular changes inpreeclampsia. Semin Nephrol. 2004;24:580–587.

96. Elvan-Taspinar A, Franx A, Bots ML, Bruinse HW, KoomansHA. Central hemodynamics of hypertensive disorders in preg-nancy. Am J Hypertens. 2004;17:941–946.

97. Tihtonen KMH, Koobi T, Uotila JT. Arterial stiffness in pre-eclamptic and chronic hypertensive pregnancies. Eur J ObstetGynecol Reprod Biol. 2006;128:180–186.

98. Thadhani R, Ecker JL, Kettyle E, Sandler L, Frigoletto FD.Pulse pressure and risk of preeclampsia: a prospective study.Obstet Gynecol. 2001;97:515–520.

99. Oyama-Kato M, Ohmichi M, Takahashi K, Suzuki S, HenmiN, Yokoyama Y, Kurachi H. Change in pulse wave velocitythroughout normal pregnancy and its value in predicting preg-nancy-induced hypertension: A longitudinal study. Am JObstet Gynecol 2006:195;464–9.

100. Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA,McLaughlin MK. Preeclampsia: an endothelial cell disorder.Am J Obstet Gynecol. 1989;161:1200–1204.

101. Agatisa PK, Ness RB, Roberts JM, Costantino JP, Kuller LH,McLaughlin MK. Impairment of endothelial function inwomen with a history of preeclampsia: an indicator of car-diovascular risk. Am J Physiol Heart Circ Physiol. 2004;286:H1389–H1393.

102. Lampinen KH, Ronnback M, Kaaja RJ, Groop PH. Impairedvascular dilation in women with a history of pre-eclampsia.J Hypertens. 2006;24:751–756.

103. Spaanderman MEA, Ekhart THA, van Eyck J, Cheriex EC,deLeeuw PW, Peeters LLH. Latent hemodynamic abnormal-ities in symptom-free women with a history of preeclampsia.Am J Obstet Gynecol. 2000;182:101–107.

104. Spaanderman ME, Schippers M, vander Graaf F, Thijssen HJ,Liem IH, Peeters LL. Subclinical signs of vascular damagerelate to enhanced platelet responsiveness among nonpreg-nant formerly preeclamptics women. Am J Obstet Gynecol.2006;194:855–860.

105. Haukkamaa L, Salminen M, Laivuori H, Leinonen H, Hii-lesmaa V, Kaaja R. Risk for subsequent coronary arterydisease after preeclampsia. Am J Cardiol. 2004;93:805–808.

106. Ronnback M, Lampinen K, Groop PH, Kaaja R. Pulse wavereflection in currently and previously preeclamptic women.Hypertens Pregnancy. 2005;24:171–180.

107. Assali NS, Holm LM, Parker HR. Systemic and regionalhemodynamic alterations in toxemia. Circulation. 1956;29,30(Suppl II):53–59.

108. Wallenburg HCS. Hemodynamics in hypertensive pregnancy.In: Rubin PC, ed. Handbook of Hypertension. Volume 10:Hypertension in Pregnancy. Amsterdam: Elsevier; 1988:66–101.

109. Heesch CM, Rogers RC. Effects of pregnancy and progester-one metabolites on regulation of sympathetic outflow. ClinExp Pharmacol Physiol. 1995;22:136–142.

110. Pan ZR, Lindheimer MD, Bailin J, Barron WM. Regula-tion of blood pressure in pregnancy: Pressor system block-ade and stimulation. Am J Physiol. 1990;258:H1559–H1572.

111. O’HaganKP, Casey SM.Arterial baroreflex during pregnancyand renal sympathetic nerve activity during parturition inrabbits. Am J Physiol. 1998;274:H1635–H1642.

112. Schobel HP, Fischer T, Heuszer K, Geiger H, Schmieder RE.Preeclampsia – a state of sympathetic overactivity. N Eng JMed. 1996;335:1480–1485.

113. Ekholm EM, Piha SJ, Antila KJ, Erkkola RU. Cardiovascularautonomic reflexes in mid-pregnancy. Br J Obstet Gynaecol.1993;100:177–182.

114. Leduc L, Wasserstrum N, Spillman T, Cotton DB. Baroreflexfunction in normal pregnancy. Am J Obstet Gynecol. 1991;165:886–890.

115. Masilamani S, Heesch CM. Effects of pregnancy and proges-terone metabolites on arterial baroreflex in conscious rats. AmJ Physiol. 1997;272:R924–R934.

116. Brooks VL, Keil LC. Changes in the baroreflex during preg-nancy in conscious dogs: Heart rate and hormonal responses.Endocrinology. 1994;135:1894–1901.

117. Ekholm EMK, Erkkola RU. Autonomic cardiovascular con-trol in pregnancy. Eur J Obstet Gynecol Reprod Biol. 1996;64:29–36.

118. Conrad KP, Russ RD. Augmentation of baroreflex-mediatedbradycardia in conscious pregnant rats. Am J Physiol. 1992;262:R472–R477.

119. Gilson GJ, Mosher MD, Conrad KP. Systemic hemody-namics and oxygen transport during pregnancy in chroni-cally instrumented, conscious rats. Am J Physiol. 1992;H1911–H1918.

120. Stock MK, Metcalfe J. Maternal physiology during gestation.In: The Physiology of Reproduction, 2nd ed. NewYork: RavenPress; 1994:947–973.

121. Poston L, McCarthy AL, Ritter JM. Control of vascular resis-tance in the maternal and feto-placental arterial beds. Phar-macol Ther. 1995;65:215–239.

122. Nisell H, Hjemdahl P, Linde B. Cardiovascular responses tocirculating catecholamines in normal pregnancy and in preg-nancy induced hypertension. Clin Physiol. 1985;5:479–493.

123. Debrah DO, Novak JE, Matthews JE, Ramirez RJ, Shroff SG,Conrad KP. Relaxin is essential for systemic vasodilation andincreased global compliance during early pregnancy in con-scious rats. Endocrinology. 2006;147:5126–5131.

124. Conrad KP, Debrah DO, Novak J, Danielson LA, Shroff SG.Relaxin modifies systemic arterial resistance and compliancein conscious, nonpregnant rats. Endocrinology. 2004;145:3289–3296.

125. Osol G, Cipolla M. Pregnancy-induced changes in the three-dimensional mechanical properties of pressurized rat utero-placental (radial) arteries. Am J Obstet Gynecol. 1993;168(Pt 1):268–274.

126. Cipolla M, Osol G. Hypertrophic and hyperplastic effects ofpregnancy on the rat uterine arterial wall. Am J Obstet Gyne-col. 1994;171:805–811.

127. Hammer ES, Cipolla MJ. Arterial wall hyperplasia is in-creased in placental compared with myoendometrial radialuterine arteries from late-pregnant rats. Am J Obstet Gynecol.2005;192:302–308.

128. Griendling KK, Fuller EO, Cox RH. Pregnancy-inducedchanges in sheep uterine and carotid arteries. Am J Physiol.1985;248:H658–H665.

129. Mackey K, Meyer MC, Stirewalt WS, Starcher BC,McLaughlin MK. Composition and mechanics of mesentericresistance arteries from pregnant rats. Am J Physiol. 1992;263:R2–R8.


130. Williams JK, Adams MR, Herrington DM, Clarkson TB.Short-term administration of estrogen and vascular responsesof atherosclerotic coronary arteries. J Am Coll Cardiol.1992;20:452–457.

131. Gilligan DM, Badar DM, Panza JA, Quyyumi AA, CannonRO. Effects of estrogen replacement therapy on peripheralvasomotor function in postmenopausal women. Am J Cardiol.1995;75:264–268.

132. Farhat MY, Lavigne MC, Ramwell PW. The vascular protec-tive effects of estrogen. FASEB J. 1996;10:615–624.

133. Gilligan DM, Bader DM, Panza JA, Quyyumi AA, CannonRO 3rd. Acute vascular effects of estrogen in postmenopausalwomen. Circulation. 1994;90:786–791.

134. Kublickiene K, Svedas E, Landgren BM, CrisbyM, Nahar N,Nisell H, Poston L. Small artery endothelial dysfunction inpostmenopausal women: in vitro function, morphology, andmodification by estrogen and selective estrogen receptormodulators. J Clin Endocrinol Metab. 2005;90:6113–6122.

135. Masi CM, Hawkley LC, Berry JD, Cacioppo JT. Estrogenmetabolites and systolic blood pressure in a population-basedsample of postmenopausal women. J Clin Endocrinol Metab.2006;91:1015–1020.

136. da Costa LS, de Oliveira MA, Rubim VS, et al. Effects ofhormone replacement therapy or raloxifene on ambulatoryblood pressure and arterial stiffness in treated hypertensivepostmenopausal women. Am J Cardiol. 2004;94:1453–1456.

137. Kalantaridou SN, Naka KK, Papanikolaou E, et al. Impairedendothelial function in young women with premature ovarianfailure: normalization with hormone therapy. J Clin Endocri-nol Metab. 2004;89:3907–3913.

138. Corbacho AM, Eiserich JP, Zuniga LA, Valacchi G, Villa-blanca AC. Compromised aortic vasoreactivity in maleestrogen receptor-alpha-deficient mice during acute lipopoly-saccharide-induced inflammation. Endocrinology. 2007;148:1403–1411.

139. Novak J, Danielson LA, Kerchner LJ, SherwoodOD, RamirezRJ, Moalli PA, Conrad KP. Relaxin is essential for renal

vasodilation during pregnancy in conscious rats. J Clin Invest.2001;107:1469–1475.

140. Danielson LA, Sherwood OD, Conrad KP. Relaxin is a potentrenal vasodilator in conscious rats. J Clin Invest. 1999;103:525–533.

141. Debrah DO, Conrad KP, Danielson LA, Shroff SG. Effects ofrelaxin on systemic arterial hemodynamics and mechanicalproperties in conscious rats: sex dependency and dose re-sponse. J Appl Physiol. 2004;98:1013–1020.

142. Conrad KP, Novak J. Emerging role of relaxin in renal andcardiovascular function. Am J Physiol Regul Integr CompPhysiol. 2004;287:R250–R261.

143. Unemori EN, EriksonME, Rocco SE, Sutherland KM, ParsellDA, Mak J, Grove BH. Relaxin stimulates expression of vas-cular endothelial growth factor in normal human endometrialcells in vitro and is associated with menometrorrhagia inwomen. Hum Reprod. 1999;14:800–806.

144. Carbillon L, Uzan M, Uzan S. Pregnancy, vascular tone, andmaternal hemodynamics: A crucial adaptation. Obstet Gyne-col Surv. 2000;55:574–581.

145. Williams JG, Rincon-Skinner T, Sun D, Wang Z, Zhang S,Zhang X, Hintze TH. Role of nitric oxide in the coupling ofmyocardial oxygen consumption and coronary vascular dy-namics during pregnancy in the dog. Am J Physiol Heart CircPhysiol. 2007;293:H2479–H2486.

146. Brust AA, Assali NS, Ferris EB. Evaluation of neurogenic andhumoral factors in blood pressure maintenance in normal andtoxemic pregnancy using tetraethylammonium chloride.J Lab Clin Med. 1948;27:717–726.

147. Assali NS, Prystowsky H. Studies on autonomic blockade: I.Comparison between the effects of tetraethylammonium chlo-ride (TEAC) and high selective spinal anesthesia on bloodpressure of normal and toxemic pregnancy. J Clin Invest.1950;29:1354–1366.

148. Suzuki Y, Yamamoto T, Mabuchi Y, et al. Ultrastructuralchanges in omental resistance artery in women with pre-eclampsia. Am J Obstet Gynecol. 2003;189:216–221.

Documents

Chesley's Hypertensive Disorders in Pregnancy || Cardiovascular Alterations in Normal and Preeclamptic Pregnancy