79 Nitroprusside in Preeclampsia Circulatory Distress and Paradoxical Bradycardia Nathan Wasserstrum In severe preeclampsia, short-term peripartum management of hypertension with hydralazine is complicated by relatively prolonged hypotensive episodes, resulting in fetal distress. We hypothesized that nitroprusside's rapid onset and brief antihypertensive action would permit more controlled blood pressure reduction. Nitroprusside was infused into 10 invasively monitored subjects until mean arterial pressure either 1) was gradually reduced 10-20% or 2) fell abruptly. Subjects fell into two groups, defined by whether the hypotensive effect of nitroprusside was accompanied by a fall in heart rate (group A, n=$) or a rise (group B, n=2). Group B showed the expected sinoaortic baroreceptor reflex elevations in heart rate (+17 ±6 beats/min) in response to moderate falls in mean arterial pressure (—32±9 mm Hg) elicited by moderate doses (1.03±0.23 /tg/kg/min). However in group A, steep reductions in mean arterial pressure (—75±22 mm Hg, p<0.0001), significantly greater than in group B (p<0.05), occurred at much lower doses (0.35±0.23 /ig/kg/min;/?<0.05) and were accompanied by falls in heart rate ( - 2 1 ± 7 beats/min). The apparently paradoxical falls in heart rate and extreme hypotensive responses in group A indicate severe circulatory compromise, corresponding to the cardiac and vasomotor depression that characterizes severe hemorrhage and other forms of acute/severe hypovolemic hypotension. This hemodynamic pattern represents a cardiopul- monary baroreceptor reflex presumably related to the Bezold-Jarisch reflex. The appearance of this pattern in the present study probably reflects the imposition of nitroprusside's prominent venous dilator action on the relatively reduced blood volume that generally characterizes severe preeclampsia. (Hypertension 1991;18:79-84) P reeclampsia, a hypertensive disorder unique to pregnancy, occurs in about 7% of pregnan- cies that continue beyond the first trimester. 1 Severe preeclampsia is distinguished from the mild form of the disease by the magnitude of hypertension (severe, systolic pressure greater than 160 mm Hg or diastolic pressure greater than 110 mm Hg) and the existence or severity of proteinuria, visual symptoms, pulmonary edema, epigastric pain, or other accom- panying disturbances. 1 - 2 Severe preeclampsia is a major cause of maternal and fetal morbidity and mortality. Its etiology has not been established. The reduced sinoaortic baroreceptor reflex sensitiv- ity that characterizes severe preeclampsia results in hemodynamic instability. 3 This instability, manifested as markedly enhanced responses to antihypertensive drugs, is of concern with hydralazine because of the From the Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine and the Department of Medicine, Baylor College of Medicine, Houston, Tex. Address for reprints: Nathan Wasserstrum, PhD, MD, Depart- ment of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Baylor College of Medicine, One Baylor Plaza, Hous- ton, TX 77030. Received October 2, 1990; accepted in revised form March 21, 1991. relatively long and variable intervals to its maximum effect and its relatively long durations of action. When cumulative doses of the drug exert their full effects without baroreceptor reflex buffering, severe hypoten- sion and fetal distress frequently results. 4 - 5 Intravenous nitroprusside has become the drug of choice for parenteral treatment of most hypertensive crises in nonpregnant adults and has largely sup- planted diazoxide and hydralazine in that setting. 6 We hypothesized that nitroprusside, because of its rapid onset and short duration of action, would lead to more controlled short-term reduction in blood pressure in severe preeclampsia than is possible with hydralazine. Our results, including identification of a character- istic pattern of severe hypotension and apparently paradoxical fall in heart rate, pertain not only to the drug's potential clinical use but also yield insights into the pathophysiology of circulatory control in severe preeclampsia. Methods Women suffering from severe preeclampsia in the third trimester of pregnancy with a mean arterial pressure (MAP) greater than 130 mm Hg and with greater than 1+ proteinuria were studied. Patients by guest on May 24, 2018 http://hyper.ahajournals.org/ Downloaded from
Transcript
79
Nitroprusside in PreeclampsiaCirculatory Distress and Paradoxical Bradycardia
Nathan Wasserstrum
In severe preeclampsia, short-term peripartum management of hypertension with hydralazineis complicated by relatively prolonged hypotensive episodes, resulting in fetal distress. Wehypothesized that nitroprusside's rapid onset and brief antihypertensive action would permitmore controlled blood pressure reduction. Nitroprusside was infused into 10 invasivelymonitored subjects until mean arterial pressure either 1) was gradually reduced 10-20% or 2)fell abruptly. Subjects fell into two groups, defined by whether the hypotensive effect ofnitroprusside was accompanied by a fall in heart rate (group A, n=$) or a rise (group B, n=2).Group B showed the expected sinoaortic baroreceptor reflex elevations in heart rate (+17 ±6beats/min) in response to moderate falls in mean arterial pressure (—32±9 mm Hg) elicited bymoderate doses (1.03±0.23 /tg/kg/min). However in group A, steep reductions in mean arterialpressure (—75±22 mm Hg, p<0.0001), significantly greater than in group B (p<0.05),occurred at much lower doses (0.35±0.23 /ig/kg/min;/?<0.05) and were accompanied by fallsin heart rate ( -21±7 beats/min). The apparently paradoxical falls in heart rate and extremehypotensive responses in group A indicate severe circulatory compromise, corresponding to thecardiac and vasomotor depression that characterizes severe hemorrhage and other forms ofacute/severe hypovolemic hypotension. This hemodynamic pattern represents a cardiopul-monary baroreceptor reflex presumably related to the Bezold-Jarisch reflex. The appearance ofthis pattern in the present study probably reflects the imposition of nitroprusside's prominentvenous dilator action on the relatively reduced blood volume that generally characterizes severepreeclampsia. (Hypertension 1991;18:79-84)
Preeclampsia, a hypertensive disorder unique topregnancy, occurs in about 7% of pregnan-cies that continue beyond the first trimester.1
Severe preeclampsia is distinguished from the mildform of the disease by the magnitude of hypertension(severe, systolic pressure greater than 160 mm Hg ordiastolic pressure greater than 110 mm Hg) and theexistence or severity of proteinuria, visual symptoms,pulmonary edema, epigastric pain, or other accom-panying disturbances.1-2 Severe preeclampsia is amajor cause of maternal and fetal morbidity andmortality. Its etiology has not been established.
The reduced sinoaortic baroreceptor reflex sensitiv-ity that characterizes severe preeclampsia results inhemodynamic instability.3 This instability, manifestedas markedly enhanced responses to antihypertensivedrugs, is of concern with hydralazine because of the
From the Department of Obstetrics and Gynecology, Division ofMaternal-Fetal Medicine and the Department of Medicine, BaylorCollege of Medicine, Houston, Tex.
Address for reprints: Nathan Wasserstrum, PhD, MD, Depart-ment of Obstetrics and Gynecology, Division of Maternal-FetalMedicine, Baylor College of Medicine, One Baylor Plaza, Hous-ton, TX 77030.
Received October 2, 1990; accepted in revised form March 21,1991.
relatively long and variable intervals to its maximumeffect and its relatively long durations of action. Whencumulative doses of the drug exert their full effectswithout baroreceptor reflex buffering, severe hypoten-sion and fetal distress frequently results.4-5
Intravenous nitroprusside has become the drug ofchoice for parenteral treatment of most hypertensivecrises in nonpregnant adults and has largely sup-planted diazoxide and hydralazine in that setting.6
We hypothesized that nitroprusside, because of itsrapid onset and short duration of action, would leadto more controlled short-term reduction in bloodpressure in severe preeclampsia than is possible withhydralazine.
Our results, including identification of a character-istic pattern of severe hypotension and apparentlyparadoxical fall in heart rate, pertain not only to thedrug's potential clinical use but also yield insightsinto the pathophysiology of circulatory control insevere preeclampsia.
MethodsWomen suffering from severe preeclampsia in the
third trimester of pregnancy with a mean arterialpressure (MAP) greater than 130 mm Hg and withgreater than 1+ proteinuria were studied. Patients
*p<0.0001, tp<0.001, paired t test, comparing base with nadir.
were excluded if 1) they had been undergoing chronicantihypertensive therapy or had taken any antihyper-tensive or potentially vasoactive agent in the previous24 hours; 2) they had preexisting cardiac disease; 3)there was evidence at entry of fetal distress orimminent delivery. Consecutive patients who wereadmitted to the Obstetric Intensive Care Unit andsatisfied entry criteria were studied.
The mean age of the 10 women enrolled was22.7±4.4 years; the gestational age of the pregnan-cies was 32.9±3.9 weeks. Seven of the 10 patientswere nulliparous. The protocol was approved by theBaylor Institutional Review Board for Human Re-search and was used only after informed consent ofthe individual subject was obtained. Each subjectreceived an initial loading dose followed by continu-ous intravenous infusion of magnesium sulfate at 2-3g/hr. At least 2 hours passed between the loadingdose and collection of data. Five percent dextrose inlactated Ringer's solution was administered at a rateof 75 ml/hr.
The radial artery was cannulated. Percutaneouscatheterization of the pulmonary artery with a flow-directed catheter (Swan-Ganz, Edwards Laborato-ries, Anasco, Puerto Rico) was performed. Catheterposition during insertion was determined by pressurewaveform analysis and subsequently confirmed by asingle anterior-posterior chest x-ray. As customary, awedge was kept under the patient's right hip through-out the study to avoid vena caval compression. Themid axillary line served as the zero reference level.
Systolic and diastolic arterial pressure, MAP, cen-tral venous pressure (CVP), pulmonary capillarywedge pressure (PCWP), and heart rate (HR) wererecorded on a polygraph (Hewlett-Packard, PaloAlto, Calif.). Pulse pressure was calculated as systolicminus diastolic pressure.
An intravenous infusion of nitroprusside was be-gun at 0.02 /xg/kg/min, with incremental increases indosage until a 10-20% reduction in MAP wasreached. Infusion was immediately stopped at theonset of any abrupt reduction in pressure. Data atthe baseline and blood pressure nadir were analyzed.
At baseline, steady-state measurements of systemicarterial blood pressure and HR were averaged over aninterval of greater than 1 minute. At the blood pressurenadir, these measurements were averaged over a vari-able interval determined by the duration of the nadir(Figure 2). For PCWP measurements, the catheterballoon was inflated, and measurements were recorded
onto the polygraph. For these and CVP measurements,respiratory fluctuations were identified on the poly-graph record and measurements were taken at endexpiration. Unless otherwise specified, results are re-ported as mean±SD.
Statistical analysis of paired data was by paired ttest. For part of the analysis, patients were dividedinto two subgroups (group A, n=8; group B, n=2),defined by a fall (group A) or rise (group B) in HRduring nitroprusside infusion. Statistical comparisonsbetween the two groups were made via the two-sample Wilcoxon rank-sum test.7 A value of p< 0.05was taken as significant for both tests.
ResultsNitroprusside lowered blood pressure in all 10 sub-
jects (Table 1). Regression analysis of the data for thefull set of subjects indicated that changes in MAP andHR were directly (not inversely) related (r=0.7652;p<0.05, t test). Similarly, the lower an individual pa-tient's nadir MAP, the lower was her concomitant HR(Figure 1) (r=0.8365;/j<0.005, t test).
Among the 10 subjects, a subset of eight (group A)experienced frank reductions in HR (-21 ±7 beats/min,p<0.0001, t test) in association with the falls inblood pressure. This subset, defined by reductions inHR, will be referred to as group A. The other twosubjects, in whom heart rate rose during the hypoten-sive episode (+17 ±6 beats/min), comprised group B.
Subjects in group A (defined by reductions in HR),experienced very steep falls in blood pressure
40 68 SB
Nac* MAP
FIGURE 1. Scatterplot shows direct correlation of nadirmean arterial pressure (MAP) with concomitant heart rate(HR).
Wasserstrum Paradoxical Bradycardia in Preeclampsia 81
200-1
« £ 150H
ix
50-
250-,
Control15 min
of Infusion
NitroprussideStopped
15 sec.
FIGURE 2. Representative re-
cordings illustrate time course offalls in systemic arterial bloodpressure and heart rate in a sub-ject from group A receiving nitro-prusside at 0.50 ng/kglmin. Asnoted, drug infusion was discon-tinued after onset of a markedfall in blood pressure. In intervalpreceding abrupt depressor re-sponse, heart rate was higherthan its control, preinfusionvalue.
(-75±22 mmHg, /?<0.0001, / test) (Figure 2). Incontrast, in group B the falls in MAP were moremoderate (-32±9 mmHg) and significantly smallerthan those in group A (p<0.05, Wilcoxon).
Group B demonstrated the smoothly controlledreductions in blood pressure sought in the study. Incontrast, hypotensive episodes in group A beganabruptly. In some subjects in group A, heart rateduring the interval preceding the abrupt depressorresponse was higher than its control, preinfusionvalue (Figure 2).
The marked hypotensive responses in group A oc-curred at very low doses of nitroprusside (0.35±0.23/ug/kg/min). In contrast, the milder hypotensive re-sponse in group B occurred at doses (1.03±0.23 fig/kg/min) that were significantly higher than those adminis-tered in group A (p<0.05, Wilcoxon, Table 2).
For the full set of 10 patients, baseline CVPaveraged 6.1 ±3.8 and PCWP 11.5±4.7 (Table 1).Nadir values of MAP and the concomitant HR andthe magnitudes of change in MAP and HR were notrelated to baseline CVP or PCWP. Groups A and Bwere not distinguishable on the basis of their respec-tive baseline hemodynamic measurements. When theinfusion was stopped, MAP and HR returned rapidlytoward control values in both groups.
Four patients in group A experienced severe nau-sea near the MAP nadir. No patient in either grouprequired cesarean section for fetal distress related toa hypotensive episode.
DiscussionIn nonpregnant subjects, the fall in MAP produced
by vasodilator administration elicits a compensatory
TABLE 2. Statistical Comparison of Group A With Group B
BaseStats MAPGroup A (n=8)
Mean± 152SD 12
Group B (n=2)Mean± 143SD 8
NadirMAP
78*
18
112
17
BasePuPr
84
11
74
7
PuPr atMAP nadir
42
14
51
21
Basesyst
20819
19313
Nadirsyst
101*
26
14530
Basedias
124
10
119
6
Nadirdias
59*
21
95
10
BaseHR
10015
818
HRatMAP nadir
79
12
98
48
BaseCVP
6.7
3.7
2.0
BasePCWP
11.75.0
10.0
Dose(jtg/kg/min)
0.35*0.23
1.030.22
Inclusion in group A is defined by a fall in heart rate. Inclusion in group B is defined by a rise in heart rate. Standard deviations (SD)for CVP and PCWP are absent for group B because baseline central pressure measurements could only be obtained in one of the twopatients in this group. MAP, mean arterial pressure (mm Hg); PuPr, pulse pressure (mm Hg); syst, systolic arterial pressure (mm Hg); dias,diastolic arterial pressure (mm Hg); HR, heart rate (beats/min); CVP, central venous pressure (mm Hg); PCWP, pulmonary capillary wedgepressure (mm Hg).
*p<0.05, two sample Wilcoxon rank sum test comparing group A with group B. For MAP, PuPr, syst, dias, and HR, the change (i.e., thedifference between base and nadir) produced by infusion is compared for the two groups. For CVP, PCWP, and Dose, absolute value iscompared for the two groups.
baroreceptor reflex rise in HR. Thus, HR variesinversely with MAP. Similarly, in severe preeclamp-sia, the hypotensive responses to vasodilators such ashydralazine and diazoxide are accompanied by ele-vations in HR.3-8 Because sinoaortic baroreceptorreflex sensitivity910 is reduced in severe preeclamp-sia,3 the reflex elevations in HR in response to acutevasodilator-induced hypotension may be very small.However, frank reductions in HR in response toarterial vasodilators have not been reported in severepreeclampsia.
In contrast, in the present study, eight subjects(group A) of 10 exhibited frank reductions in HRduring the hypotensive response to nitroprusside.Such "paradoxical" reductions in HR are not expli-cable in terms of the sinoaortic baroreceptor reflex.
Despite the abrupt and steep hypotensive responseproduced by nitroprusside, MAP and HR returnedrapidly toward control levels when the infusion wasstopped. This reflects the drug's short half-life. Thus,unlike the experience reported with hydralazine,34
no patient in the present study required cesareansection for fetal distress related to a hypotensiveepisode.
In nonpregnant patients in hypertensive crisis,titration of nitroprusside dosage generally permitswell-controlled reductions in blood pressure. A finaldose of 1.0-3.0 /ig/kg/min is usually required.6 Incontrast, in group A the mean dose at which abrupthypotensive episodes occurred was only 0.35 tiglkglmin.
Group B showed results more consistent with theliterature. Moderate, well-controlled reductions inMAP occurred at significantly higher doses than ingroup A and were accompanied by the expectedbaroreceptor reflex increases in HR.
MechanismThe apparently anomalous results in group A help
elucidate the circulatory pathophysiology of severepreeclampsia. Their interpretation is facilitated byconsideration of two points.
1) There are some well-defined circumstances un-der which acute hypotension is accompanied by a fallin HR. These include the supine hypotension syn-drome of pregnancy,11 the Bezold-Jarisch reflex,12
and reflexive cardiac and vasomotor depression elic-ited by severe hemorrhage and other forms of acute/severe hypovolemic hypotension.13"18
2) Although hydralazine and diazoxide relax onlyarterial resistance vessels, nitroprusside also relaxesvenous capacitance vessels.19 Thus, although all threedrugs produce hypotensive responses, only nitroprus-side can, under appropriate conditions, also producefunctional hypovolemia.
Supine HypotensionThe supine hypotensive syndrome is defined by
marked reductions in MAP and HR11 that occur in asmall percentage of late gestation gravida when theyadopt the supine position. The position produces
caval compression and a reduction in venous return,thereby leading to a fall in MAP. Shortly after MAPstarts to fall, reflex reduction in HR begins.11
Depending on baseline hemodynamic status, nitro-prusside may increase or decrease cardiac output.20'21
In the context of the abnormally reduced bloodvolume characteristic of untreated severe preeclamp-sia,22 nitroprusside-induced venodilation decreasesvenous return and cardiac output. This potentiatesthe fall in MAP produced by the drug's arterialvasodilator action. Thus, the fall in HR we observedin group A may reflect a mechanism similar to thatresponsible for the supine hypotensive syndrome.
Of interest, in the supine hypotensive syndrome,HR frequently rises before beginning its abrupt fall.11
This is analogous to the responses of some subjectstreated with nitroprusside (Figure 2). In both in-stances, the initial rise in HR probably reflects acti-vation of the sinoaortic baroreceptor reflex beforethe latter is overpowered by an increasingly severehypovolemic and hypotensive stimulus.
Cardiac and Vasomotor Depression Due toVentricular Reflexes
Strictly defined, Bezold-Jarisch effects are reflexresponses elicited from inferior ventricular receptorsby ischemia or by various compounds, includingprostaglandins, that are locally secreted in responseto ischemia.12 The reflex results in hypotension (va-somotor depression) and a fall in HR (cardiac de-pression). The Bezold-Jarisch reflex, even defined sostrictly, cannot be excluded as the mechanism under-lying the fall in HR and severe hypotension demon-strated by group A.
Indeed, the bradycardia associated with nitroprus-side may be at least in part due to stimulation ofprostaglandin biosynthesis in the heart. In pentobar-bital-anesthetized rats, retrograde injection of nitro-prusside in the thoracic aorta produced hypotensionand bradycardia. Indomethacin abolished the brady-cardia without reducing nitroprusside's hypotensiveeffect.23 Thus, nitroprusside appeared to stimulatemyocardial prostaglandin biosynthesis and therebyinitiate a reflex bradycardia. Severe preeclampsiamay be accompanied by a disturbance in cardiacprostaglandin synthesis that contributes to thegreater sensitivity to nitroprusside noted in the pres-ent study. More generally, the importance of phar-macological effects on cardiac inhibitory reflexes hasrecently been emphasized.24-26
Depressor reflexes also originate at other ventric-ular sites and in response to stimuli not necessarilyrelated to ischemia.27 Data from a variety of modelsindicate that severe hemorrhage and other forms ofsevere hypovolemia and hypotension stimulate ven-tricular receptors (including mechanoreceptors) thatelicit centrally mediated cardiac and vasomotor de-pressor reflexes.13-15-27-28
When rats were bled to reach a predeterminedMAP within 2 minutes, HR rose as long as MAPremained above 100 mm Hg. However, when MAP
Wasserstrum Paradoxical Bradycardia in Preeclampsia 83
fell below 90 mm Hg, HR fell in proportion to thedepth of hypotension.14 In humans, Barriot andRiou16 found that paradoxical bradycardia developedin 20 of 273 patients with acute hemorrhagic shock.Compared with the other patients, these 20 had morerapid and severe hemorrhage. Sander-Jensen et al29
reported that in patients monitored during resuscita-tion from hemorrhagic shock, mean HR was lowerduring the shock phase than during the steady statereached after volume repletion. This group of inves-tigators also demonstrated that in healthy men, cen-tral hypovolemia induced via exposure to lower bodynegative pressure initially led to mild hypotensionand relative tachycardia. After further exposure, amuch greater fall in blood pressure was accompaniedby a slowing of HR back to control levels.30 Thesedepressor reflexes are sometimes categorized undermore broadly defined Bezold-Jarisch effects. Re-cently, their role in cardiovascular regulation invarious physiological and pathological states has be-come more widely appreciated.25
Nitroprusside's venodilator action combines withthe abnormally reduced baseline blood volume ofsevere preeclampsia to produce severe functionalhypovolemia and hypotension. This hemodynamicpattern is analogous to, and elicits the depressorreflex characteristic of, severe hemorrhage. This de-pressor pattern is reflected in the paradoxical fall inHR and abrupt, severe fall in MAP encountered inthe present study.
Venous ToneThe maintenance of relatively normal central pres-
sures and cardiac function in the face of a variablyreduced blood volume22-31 is consistent with theconcept of increased venous tone and central redis-tribution of blood volume in preeclampsia.32 In thepresence of minimal baseline venomotor tone, avenodilator has minimal hemodynamic effect. Con-versely, the marked response to nitroprusside-in-duced venodilation in the present study stronglysupports the concept of high baseline venous tone insevere preeclampsia.
According to this view, patients in group A had alower blood volume and greater venous tone thanthose in group B. This view is not contradicted by theabsence of differences between groups A and B incentral pressure or other baseline hemodynamics, orby the absence of any correlation between centralpressures and the hemodynamic response to nitro-prusside. Rather, it must be emphasized that thebaseline hemodynamic indexes usually measured,including central pressures, simply do not yield anaccurate assessment of venous tone.32
Previous Reports of Venodilators in Preeclampsia
In contrast to our findings, previous investigatorshave reported control and stabilization of MAP withnitroprusside in severe preeclampsia that was appar-ently uncomplicated by abrupt, steep hypotensiveepisodes.33-34 Unfortunately, HR measurements for
comparison with our results were not reported. Theapparent conflict with our results is resolved whenone notes that all five patients in the aforementionedstudies were clinically in acute pulmonary edemabefore nitroprusside was administered. PCWP mea-surements reported on four of the five patientsranged from 20 to 33 mm Hg, consistent with in-creased rather than decreased baseline intravascularvolume. Thus, these patients would not be expectedto show the steep hypotension and paradoxical fall inHR that accompanies severe hypovolemic hypoten-sion. In contrast, none of our patients showed evi-dence of elevated cardiac filling pressures or pulmo-nary edema.
Although not quite as powerful a vasodilator asnitroprusside, nitroglycerin acts more specifically onthe venous capacitance beds. Thus, to the extent thatthe venodilator action of nitroprusside is responsiblefor the steep hypotension and paradoxical fall in HRwe found in severe preeclampsia, one would expectto find analogous results with nitroglycerin. Carefulreading of the published results on the hemodynamiceffects of nitroglycerin in severe preeclampsia revealsa subset of patients with a pattern of hypotension andparadoxical fall in heart rate analogous to those weencountered with nitroprusside.35
In conclusion, clinically the current results empha-size how poorly patients with preeclampsia compen-sate for hypovolemic stimuli such as venodilators orperioperative hemorrhage. Secher and Bie36 warnedthat in hemorrhage, "the concept of tachycardia,being the only deviation of heart rate . . . may be fatalto the patient if as a consequence a decrease in heartrate is interpreted as an improvement of the patient'sgeneral condition." Our results indicate that thiswarning applies as well to the patient with severepreeclampsia exposed to nitroprusside, hemorrhage,or other hypovolemic/hypotensive stimuli. Clearly,the use of nitroprusside in severe preeclampsia mustbe approached with caution and only after consider-ation of the patient's blood volume.
The hemodynamic response to nitroprusside in thepresent study emphasizes that severe preeclampsia ischaracterized by decreased blood volume and in-creased venous tone. Thus, concomitant venous andarterial dilatation with nitroprusside frequently re-sult in hypovolemic hypotension and severe circula-tory compromise. The hypovolemic hypotension inturn elicits a pattern of reflex cardiac and vasomotordepression analogous to that seen in severe hemor-rhage. The result is a fall in HR, and an abrupt,further fall in MAP. This reflects much more severecirculatory compromise than that caused by the de-creased baroreceptor reflex sensitivity that character-izes severe preeclampsia.3
References1. Roberts JM: Pregnancy related hypertension, in Creasy RK,
Resnik R (eds): Maternal Fetal Medicine: Principles and Prac-tice,^!. Philadelphia, WB Saunders Co, 1989
2. Cunningham FG, MacDonald PC, Gant NF: Williams Obstet-rics, ed 18. Norwalk, Conn, Appleton & Lange, 1989
3. Wasserstrum N, Kirshon B, Rossavik IK, Willis RS, Moise KJ,Cotton DB: Implications of sino-aortic baroreceptor reflexdysfunction in severe preeclampsia. Obstet Gynecol 1989;74:34-39
4. Spinnato JA, Sibai BM, Anderson GD: Fetal distress afterhydralazine therapy for severe pregnancy-induced hyperten-sion. South Med J 1986;79:559-562
5. Kirshon B, Wasserstrum N, Cotton DB: Should continuoushydralazine infusions be utilized in severe pregnancy inducedhypertension. Am J Painatol (in press)
6. Dunagan WC, Ridner ML (eds): Manual of Medical Therapeu-tics of the Department of Medicine, Washington UniversitySchool of Medicine, ed 26. Boston, Little, Brown & Co, 1989,p87
7. Rosner B: Fundamentals of Biostatistics, ed 3. Boston, PWS-Kent Publishing, 1990
8. Morris JA, Aice JJ, Hamilton CJ, Davidson EC, Maidman JE,Clark JH, Bloom RS: The management of severe preeclampsiaand eclampsia with intravenous diazoxide. Obstet Gynecol1977;49:675-680
9. Wasserstrum N, Herd JA: Barorefiexive depression of oxygenconsumption in the squirrel monkey at 10°C. Am J Physiol1977;232:H451-H458
10. Wasserstrum N, Herd JA: Heart rate and oxygen uptakeresponse to angiotensin in the squirrel monkey at 10°C. Am JPhysiol 1977;233:H10-H14
11. Kerr MG: The mechanical effects of the gravid uterus in latepregnancy. Obstet Gynecol Br Commonwealth 1965;72:513-529
12. Zucker IH: Left ventricular receptors: Physiological control-lers or pathological curiosities? Basic Res Cardiol 1986;81:539-557
13. Feuerstein G, Zerbe RL, Meyer DK, Kopin IJ: Alteration ofcardiovascular, neurogenic, and humoral responses to acutehypovolemic hypotension by administered prostacyclin. J Car-diovasc Pharmacol 1982;4:246-253
14. McCafferty JP, Roesler JM, Pendleton RG: Studies on themechanisms involved in the cardiovascular response of the ratto hemorrhage. Arch Int Pharmacodyn Ther 1981;254:58-69
15. Kostreva DR, Castaner A, Pedersen DH, Kampine JP: Non-vagalty mediated bradycardia during cardiac tamponade orsevere hemorrhage. Cardiology 1981;68:65-79
16. Barriot P, Riou B: Hemorrhagic shock with paradoxical brady-cardia. Intensive Care Med 1987;13:203-207
17. Sanders JS, Ferguson DW: Profound sympathoinhibition com-plicating hypovolemia in humans. Ann Intern Med 1989;111:439-441
18. Murray RH, Thompson LJ, Bowers JA, Albright CD: Hemo-dynamic effects of graded hypovolemia and vasodepressorsyncope induced by lower body negative pressure. Am Heart J1968;76:799-811
19. Blaschke TF, Melmon KL: Antihypertensive agents and thedrug therapy of hypertension, in Gilman AG, Goodman LS,Gilman A (eds): Goodman and Gilman's The PharmacologicalBasis of Therapeutics, ed 6. New York, Macmillan, 1980, pp793-818
20. Pagani M, Vatner SF, Braunwald E: Hemodynamic effects ofintravenous sodium nitroprusside in the conscious dog. Circu-lation 1978^7:144-151
21. Shepherd AM, Irvine NA: Differential hemodynamic andsympathoadrenal effects of sodium nitroprusside and hydral-azine in hypertensive subjects. J Cardiovasc Pharmacol 1986;8:527-533
22. Hayes PM, Cruikshank DP, Dunn LJ: Plasma volume deter-mination in normal and preeclamptic pregnancies. Am J ObstetGynecol 1985;151:958-966
23. Lukacsko P: Paradoxical bradycardia following the adminis-tration of sodium nitroprusside in the rat is abolished byindomethacin. Cardiovasc Res 1982;16:95-98
24. Semple PF, Thoren P, Lever AF: Vasovagal reactions tocardiovascular drugs: The first dose effect. / Hypertens 1988;6:601-606
25. Mark AL: The Bezold-Jarisch reflex revisited: Clinical impli-cations of inhibitory reflexes originating in the heart. JAm CollCardiol 1983;l:9O-102
26. Mark AL: Sensitization of cardiac vagal afferent reflexes at thesensory receptor level: An overview. Fed Pmc 1987;46:36-40
27. Thoren P: Depressor reflexes from the heart during severehemorrhage, in Hainsworth R, McWilliam PN, Mary DASG(eds): Cardiogenic Reflexes. New York, Oxford UniversityPress, Inc, 1987, pp 389-401
28. Secher NH, Sander-Jensen K, Werner C, Warburg J, Bie P:Bradycardia during severe but reversible hypovolemic shock inman. Ore Shock 1984;14:267-274
29. Sander-Jensen K, Secher NH, Bie P, Warberg J, SchwartzTW: Vagal slowing of the heart during hemorrhage: Observa-tions from 20 consecutive hypotensive patients. Br Med J1986;292:364-366
31. Chesley LC: Hypertensive Disorders in Pregnancy. New York,Appleton-Century-Crofts, 1978, p 203
32. Mabie WC, Ratts TE, Sibai BM: The central hemodynamics ofsevere preeclampsia. Am J Obstet Gynecol 1989;161:1443-1448
33. Shoemaker CT, Meyers M: Sodium nitroprusside for controlof severe hypertensive disease of pregnancy: A case report anddiscussion of potential toxicity. Am J Obstet Gynecol 1984;149:171-173
34. Stempel JE, O'Grady JP, Morton MJ, Johnson KA: Use ofsodium nitroprusside in complications of gestational hyperten-sion. Obstet Gynecol 1982;60:533-538
35. Cotton DB, Longmire S, Jones MM, Dorman KF, Tessem J,Joyce TH III: Cardiovascular alterations in severe pregnancy-induced hypertension: Effects of intravenous nitroglycerincoupled with blood volume expansion. Am J Obstet Gynecol1986;154:1053-1059
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/01.HYP.18.1.79
1991;18:79-84Hypertension.
http://hyper.ahajournals.org/content/18/1/79the World Wide Web at:
The online version of this article, along with updated information and services, is located on
http://hyper.ahajournals.org//subscriptions/
is online at: Hypertension Information about subscribing to Subscriptions:
http://www.lww.com/reprints Information about reprints can be found online at: Reprints:
document. Permissions and Rights Question and Answer information about this process is available in the
is located, click Request Permissions in the middle column of the Web page under Services. Further requestedthe Editorial Office. Once the online version of the published article for which permission is being
can be obtained via RightsLink, a service of the Copyright Clearance Center, notHypertensionpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:
