Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/12241781

SeverelyImpairedBaroreflex-BufferinginPatientsWithMonogenicHypertensionandNeurovascularContact

ARTICLEinCIRCULATION·NOVEMBER2000

ImpactFactor:14.43·DOI:10.1161/01.CIR.102.21.2611·Source:PubMed

CITATIONS

48

READS

29

15AUTHORS,INCLUDING:

HakanToka

PartnersHealthCare

35PUBLICATIONS1,058CITATIONS

SEEPROFILE

OkanToka

Children´sHostpital,UniversityofErlange…

48PUBLICATIONS1,068CITATIONS

SEEPROFILE

JensTank

HannoverMedicalSchool

161PUBLICATIONS3,145CITATIONS

SEEPROFILE

RaminNaraghi

BundeswehrkrankenhausUlm

49PUBLICATIONS769CITATIONS

SEEPROFILE

Availablefrom:RaminNaraghi

Retrievedon:04February2016

Schuster, Hans P. Schobel, Hermann Haller and Friedrich C. LuftDiedrich, Christine Stabroth, Mandy Stoffels, Ramin Naraghi, Wolfgang Oelkers, Herbert

Jens Jordan, Hakan R. Toka, Karsten Heusser, Okan Toka, John R. Shannon, Jens Tank, AndreNeurovascular Contact

Severely Impaired Baroreflex-Buffering in Patients With Monogenic Hypertension and

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2000 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.102.21.2611

2000;102:2611-2618Circulation. 

http://circ.ahajournals.org/content/102/21/2611World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation 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 this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

Severely Impaired Baroreflex-Buffering in Patients WithMonogenic Hypertension and Neurovascular Contact

Jens Jordan, MD; Hakan R. Toka, MD; Karsten Heusser, MD; Okan Toka, MD;John R. Shannon, MD; Jens Tank, MD; Andre Diedrich, MD; Christine Stabroth, MD;

Mandy Stoffels, RN; Ramin Naraghi, MD; Wolfgang Oelkers, MD; Herbert Schuster, MD;Hans P. Schobel, MD; Hermann Haller, MD; Friedrich C. Luft, MD

Background—We identified a family with a monogenic syndrome of hypertension, brachydactyly, and neurovascularcontact of the brain stem. Neurovascular contact of the ventrolateral medulla may lead to arterial hypertension byinterfering with baroreflex function.

Methods and Results—In 5 patients with monogenic hypertension (18 to 34 years old), we conducted detailed autonomicfunction tests. Blood pressure during complete ganglionic blockade was 13464.9/8264.1 mm Hg and 9066/4962.4mm Hg in patients and in control subjects, respectively. During ganglionic blockade, plasma vasopressin concentrationincreased 24-fold in control subjects and,2-fold in patients. In patients, cold pressor testing, hand-grip testing, andupright posture all increased blood pressure excessively. In contrast, muscle sympathetic nerve activity was notincreased at rest or during cold pressor testing. The phenylephrine dose that increased systolic blood pressure 12.5mm Hg was 8.062.0mg in patients and 135635 mg in control subjects before ganglionic blockade and 5.460.4 mg inpatients and 1364.8mg in control subjects during ganglionic blockade.

Conclusions—In patients with monogenic hypertension and neurovascular contact, basal blood pressure was increasedeven during sympathetic and parasympathetic nerve traffic interruption. However, sympathetic stimuli caused anexcessive increase in blood pressure. This excessive response cannot be explained by increased sympathetic nerve trafficor increased vascular sensitivity. Instead, we suggest that baroreflex buffering and baroreflex-mediated vasopressinrelease are severely impaired.(Circulation. 2000;102:2611-2618.)

Key Words: baroreceptorsn geneticsn hypertensionn nervous system, autonomicn receptors

Bilginturan et al1 first described autosomal dominanthypertension and brachydactyly. The gene responsible

for the syndrome was mapped to chromosome 12p2. Affectedindividuals exhibit a profoundly accelerated increase in bloodpressure with age.3 We showed earlier that affected individ-uals are not salt sensitive and that plasma renin activity andaldosterone respond normally to volume expansion and de-pletion.3 We subsequently showed by MR tomography that all15 affected family members tested had a left-sided loop of theposterior inferior cerebellar artery (PICA) impinging on therostroventrolateral medulla, compared with none of 11 non-affected family members.4 Pulsatile compression of the ros-troventrolateral medulla may contribute to essential hyperten-sion.5,6 This hypothesis is supported by the observation thatneurosurgical decompression of the brain stem improvesblood pressure control.7 Animal studies suggest that pulsatilecompression of the rostroventrolateral medulla increasesblood pressure through sympathetic activation.8 Whether

similar autonomic abnormalities occur in humans is notknown. We tested the hypothesis that in patients with mono-genic hypertension, brachydactyly, and neurovascular con-tact, the hypertension is mediated through sympathetic acti-vation. Furthermore, we tested whether these changes werecaused by increased sympathetic nerve traffic, increasedvascular sensitivity, or impaired baroreflex buffering.

Methods

SubjectsFive younger family members with hypertension and brachy-dactyly traveled to Germany to participate in this study (3men, 2 women, 2663.4 years old, 5764.0 kg, 15262.4 cmtall). Their results were compared with those of healthynormotensive control subjects of similar age. Written in-formed consent was obtained. All studies were approved bythe institutional review board.

Received June 6, 2000; revision received July 5, 2000; accepted July 7, 2000.From the Clinical Research Center, Franz Volhard Clinic, Medical Faculty of the Charité, Humboldt-University, and the Department of Endocrinology,

Klinikum Benjamin Franklin, Free University, Berlin, and the Department of Internal Medicine and Nephrology and Department of Neurosurgery,Friedrich-Alexander University, Erlangen, Germany; and the Autonomic Dysfunction Service, Vanderbilt University, Nashville, Tenn.

Correspondence to Jens Jordan, MD, Clinical Research Center, Franz-Volhard-Clinic, Humboldt University, Wiltbergstraße 50, 13125 Berlin,Germany. E-mail jordan@fvk-berlin.de

© 2000 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org

2611

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

ProtocolThree days before the study, volunteers received a 150-mmolsodium diet free of substances that could interfere withcatecholamine measurements. All vasoactive medicationswere discontinued$5 half-lives before testing. The patientsunderwent a physical examination, MRI imaging of the brainstem,4 and a series of autonomic tests on separate days.

Posture StudyAfter the patients had remained overnight in the supineposition, venous samples for plasma renin activity andaldosterone concentrations were obtained from a heparin lockplaced $30 minutes before the first blood draw. Bloodsamples were again obtained after 30 minutes in the standingposition. Automated measurements of blood pressure andheart rate were made (Dinamap, Criticon). On a separate day,the study was repeated after an intravenous infusion of 1000mL normal saline in 30 minutes.

Autonomic Reflex TestingSinus arrhythmia was assessed during controlled breathing(5-second inhalation and 5-second exhalation for 90 seconds).The sinus arrhythmia ratio was calculated as the ratio of thelongest to the shortest RR interval during these 90 seconds.Patients performed a Valsalva maneuver (40 mm Hg pressurefor 15 seconds). Blood pressure and heart rate responses toisometric handgrip (30% maximum contraction for 1 minute)and cold pressor testing were also determined.9

Pharmacological testing was conducted with the subjectrecumbent$2.5 hours after the last meal. Heart rate wasdetermined with a continuous ECG, blood pressure by anindwelling catheter in the radial artery, and changes in strokevolume by impedance cardiography.10 Leg blood flow wasdetermined before and during complete ganglionic blockade byocclusion plethysmography (EC 5R, Hokanson). Responses toincremental intravenous bolus doses of nitroprusside and phen-ylephrine were evaluated before ganglionic blockade. Thereaf-ter, we infused the ganglionic blocker trimethaphan (CambridgePharmaceuticals) starting at 1 mg/min and increasing at6-minute intervals until the efferent arc of the baroreflex was

completely blocked.11 Bolus doses of phenylephrine were thenadministered just as before blockade.

The baroreflex slope was determined at the linear portionof the sigmoidal relation between phenylephrine-inducedblood pressure changes and changes in RR interval. Thespontaneous baroreflex slope was determined by the se-quence technique.12 The doses of each drug that would changeblood pressure by 12.5 mm Hg were determined by extrapo-lation from individual dose-response curves.

MicroneurographyDuring the microneurography study, heart rate was deter-mined by continuous ECG and beat-by-beat blood pressureby photoplethysmography (Finapres, Ohmeda). Microneu-rography recordings were obtained as described previously.13

Analytical MethodsPlasma and urinary catecholamines were determined by amodification of a high-performance liquid chromatographicmethod.14 Plasma vasopressin concentration was determinedby a radioimmunoassay.

StatisticsAll data are expressed as mean6SEM. Intraindividual andinterindividual differences were compared by paired andunpairedt tests, respectively. ANOVA testing for repeatedmeasures was used for multiple comparisons. If necessary,data were logarithmically transformed before analysis. Rela-tionships between parameters were assessed by linear regres-sion analysis. A value ofP,0.05 was considered significant.

ResultsClinical CharacteristicsAll 5 patients had severe arterial hypertension. None hadcardiovascular complications. All had a normal creatinineclearance, and none had microalbuminuria. Echocardiogra-phy showed normal systolic function. One patient had mildleft ventricular hypertrophy. However, none had signs ofimpaired left ventricular filling. MRI of the brain stemshowed a left-sided loop of the PICA in all affected subjects(Figure 1).4

Posture Studies and CatecholaminesFigure 2 illustrates supine and upright blood pressure andheart rate. Without volume loading, blood pressure was13566.7/8164.6 mm Hg and increased to 14863.2/9965.7mm Hg after 5 minutes in the standing position (P50.1 forsystolic blood pressure andP50.01 for diastolic bloodpressure). This increase in blood pressure was associated witha change in heart rate from 5963.2 bpm supine to 8765.5bpm upright. Before volume loading, supine blood pressurewas 14066.0/8466.0 mm Hg and supine heart rate was6364 bpm. Immediately after volume loading, supine bloodpressure was 14065.8/8464.0 mm Hg and supine heart ratewas 6665.0 bpm. With volume loading, the increase in bloodpressure with standing was attenuated. After 5 minutes ofstanding, blood pressure was 14163.9/9464.4 mm Hg andheart rate was 7164.2 bpm. Plasma norepinephrine andepinephrine concentrations were not excessively increased in

Plasma Catecholamines

Norepinephrine,nmol/L

Epinephrine,nmol/L

Mean SEM Mean SEM

Without volume loading

Supine 0.59 0.13 0.22 0.07

Upright 1.40 0.21 0.49 0.21

After volume loading

Supine 0.51 0.06 0.23 0.05

Upright 1.30 0.09 0.32 0.05

Ganglionic blockade 0.13 0.01 0.16 0.04

Plasma catecholamines were determined after patients had been in thesupine position overnight and again after 30 minutes standing. The test wasrepeated on a different day after volume loading with intravenous salinesolution. In addition, plasma catecholamines were determined after completeganglionic blockade was achieved.

2612 Circulation November 21, 2000

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

the supine or upright position (Table). Urinary excretion ofnorepinephrine was 112624 nmol/d (normal range 136 to621 nmol/d). Urinary epinephrine excretion was 3868.2nmol/d (normal range 22 to 109 nmol/d).

Autonomic Reflex TestingRespiratory sinus arrhythmia was 1.360.055. During Valsalvamaneuver phase II, blood pressure increased 12611/2669.8mm Hg. Blood pressure increased 45612/2466.3 mm Hg dur-ing phase IV. The Valsalva heart rate ratio was 1.460.19. Withhand-grip testing, blood pressure increased 3665.0/2462.7mm Hg. After 1 minute of cold pressor testing, blood pressureincreased 4668.8/2765.5 mm Hg. In 4 patients, blood pressureincreased profoundly with cold pressor testing (range 45 to66/22 to 41 mm Hg); 1 patient had a clearly abnormal response(9/7 mm Hg change) but had no pain during the test. The pressorresponses to hand-grip exercise and to cold were exaggeratedcompared with control subjects.15

Responses to Complete Ganglionic BlockadeBefore trimethaphan infusion, blood pressure was 15064.9/8661.8 mm Hg and heart rate was 6766.3 bpm. Figure 3illustrates changes in blood pressure, heart rate, and strokevolume with increasing trimethaphan infusion. When completeganglionic blockade was achieved, blood pressure and heart rate

were 13464.9/8264.1 mm Hg and 9062.6 bpm, respectively.In control subjects,15 blood pressure in the supine position was12964.0/6563.0 mm Hg before trimethaphan. With completeganglionic blockade, blood pressure decreased to 9066/4962.4mm Hg (P,0.01 compared with patients). Decreases in bothsystolic (P,0.05) and diastolic blood pressure (P,0.01) weresmaller in patients than in control subjects. In patients, thedecrease in blood pressure was associated with a 7.565.3%decrease in cardiac output. Stroke volume decreased 3165.0%.Leg blood flow increased from 5.960.6 mL·100 mL-1·s-1 atbaseline to 1162 mL·100 mL-1·s-1 during complete ganglionicblockade. Plasma norepinephrine concentration decreased pro-foundly during ganglionic blockade (Table). In patients, plasmavasopressin concentration was 0.4760.02 pg/mL at baseline and0.8460.13 pg/mL during complete ganglionic blockade (Figure4). In control subjects, in contrast, plasma vasopressin concen-tration increased profoundly during ganglionic blockade(1.660.17 pg/mL at baseline, 39613 pg/mL during ganglionicblockade).

Sensitivity to Phenylephrine and Nitroprusside andBaroreflex SlopeBefore ganglionic blockade, patients were extremely hyper-sensitive to the pressor effect of phenylephrine (Figure 5).The phenylephrine dose that increased systolic blood pressure

Figure 1. T2-weighted high-resolution MR tomography (MRT) of medulla oblongata in a nonaffected subject (A, B) and a patient withbrachydactyly and hypertension (C, D). A, In nonaffected subject, axial scan does not show signs of neurovascular contact (NVC).There is no vascular attachment to surface of ventrolateral medulla (VLM) and root entry zone of cerebral nerves 9 and 10. B, Similarly,there was no evidence of vascular compression of VLM in coronal plane. C, In a patient with brachydactyly and hypertension, axialT2-weighted MRT showed a loop of PICA on left side impinging on VLM and root entry zone of cerebral nerves 9 and 10. D, Coronalview shows course of loop in front of VLM causing NVC at left. VA indicates vertebral artery.

Jordan et al Baroreflex Buffering in Monogenic Hypertension 2613

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

12.5 mm Hg was 8.062.0mg in patients and 135635 mg incontrol subjects (Figure 6). During ganglionic blockade, thephenylephrine dose that increased systolic blood pressure12.5 mm Hg was 5.460.4mg in patients and 1364.8mg incontrol subjects (P,0.0001 between groups,P,0.001 be-tween interventions,P,0.01 for the interaction betweengroup and intervention). Ganglionic blockade potentiated thepressor effect by factors of 1.560.5 and 1564.3 in patientsand control subjects, respectively (P,0.01). Without gangli-onic blockade, patients were hypersensitive to the depressoreffect of nitroprusside (Figure 7). The nitroprusside dose thatdecreased systolic blood pressure 12.5 mm Hg without gan-glionic blockade was 0.2060.066mg/kg in patients and1.860.50 in control subjects (P,0.01) (Figure 8). Thebaroreflex slope determined by phenylephrine bolus applica-tion was 9.461.6 ms/mm Hg. Baroreflex slope determined bythe sequence technique was 1262.0 ms/mm Hg. In 3 patients,baroreflex slope was above the 90th percentile for age, and in2 patients, it was between the 90th and 95th percentiles.12

MicroneurographyRepresentative microneurography recordings are illustratedin Figure 9. Baseline sympathetic activity in the supine

position was 1564.6 bursts/min. In a previous study, restingmuscle sympathetic nerve activity was 1266 bursts/min inyoung normotensive control subjects and 1469 bursts/min inyoung borderline hypertensives.13 Microneurography inci-dence was 2266.5 bursts/100 heartbeats. With cold pressortesting, muscle sympathetic nerve activity increased 7.566.8bursts/min (11610 bursts/100 heartbeats). By comparison,cold pressor testing elicited an increase in muscle sympa-thetic nerve activity by 14610 bursts/min in normotensivecontrol subjects and by 1168.0 bursts/min in patients withborderline hypertension.13 With lower-body negative pres-sure, muscle sympathetic nerve activity increased from1965.8 bursts/min (3168.6 bursts/100 heartbeats) at 0mm Hg to 2466.5 bursts/min (3669.0 bursts/100 heartbeats)at 210 mm Hg suction (n54). This increase in sympatheticnerve traffic during lower-body negative pressure was asso-ciated with a decrease in forearm blood flow from 8.761.4mL/100 mL to 6.660.8 mL/100 mL. Blood pressure was13164.6/7462.6 mm Hg at 0 mm Hg and 13463.3/7662.0mm Hg at210 mm Hg lower-body negative pressure.

DiscussionOur main finding was that in patients with monogenichypertension, brachydactyly, and neurovascular contact,

Figure 2. Changes in systolic blood pressure (SBP), diastolic bloodpressure (DBP), and heart rate (HR) with standing. Patients stoodup at 0 minutes. Without volume administration (top), blood pres-sure and heart rate increased substantially with standing. Intrave-nous infusion of 1000 mL normal saline attenuated blood pressureand heart rate increase with standing (bottom).

Figure 3. Top, Changes in systolic blood pressure (SBP), dia-stolic blood pressure (DBP), and heart rate (HR) with incremen-tal infusion of trimethaphan. Blood pressure decreased substan-tially with trimethaphan. Bottom, Changes in stroke volume withincremental infusion of trimethaphan.

2614 Circulation November 21, 2000

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

blood pressure during complete ganglionic blockade was 44mm Hg greater than in control subjects. However, sympa-thetic stimuli caused an excessive increase in blood pressure.This increase was not due to increased sympathetic nervetraffic. The patients were extremely hypersensitive to phen-ylephrine and nitroprusside. With ganglionic blockade, thepressor effect of phenylephrine was only slightly augmented

in patients but increased dramatically in control subjects.Furthermore, with ganglionic blockade, plasma vasopressinconcentration increased profoundly in control subjects butchanged little in patients.

We used trimethaphan to determine the contribution ofautonomic nervous system activity to blood pressure con-trol.16 Ganglionic blockade at the level achieved in this studyabolishes sympathetic and parasympathetic modulation ofheart rate and blood pressure.15 Even with substantial changesin blood pressure, heart rate does not change and musclesympathetic nerve activity is abolished.11 The blockade isassociated with decreased plasma norepinephrine concentra-tions as low as seen in patients with severe pure autonomicfailure.15 Plasma catecholamines and muscle sympatheticnerve activity were low normal in our patients, suggestingthat the sympathetic contribution to supine blood pressurewas not increased. In this regard, our patients resemblepatients with secondary hypertension rather than patients withessential hypertension.17 Because sympathetic nerve traffic is

Figure 4. Plasma vasopressin concentra-tions at baseline and during ganglionicblockade with trimethaphan in patientsand in young control subjects. In controlsubjects, ganglionic blockade is associ-ated with a profound baroreflex-mediated increase in plasma vasopressinconcentration. A similar increase wasobserved in a previous study by Baylisand Robinson.32 In patients, plasmavasopressin concentration increasedslightly or not at all during ganglionicblockade.

Figure 5. Changes in systolic blood pressure with incrementalbolus doses of phenylephrine before (top) and during (bottom)ganglionic blockade in patients with monogenic brachydactylyand hypertension (HTN) and in young normotensive control sub-jects. Before ganglionic blockade, patients were extremelyhypersensitive to pressor effect of phenylephrine. During gangli-onic blockade, sensitivity to phenylephrine was not significantlydifferent between patients and control subjects.

Figure 6. Doses of phenylephrine that increased systolic bloodpressure 12.5 mm Hg before and during ganglionic blockadewith trimethaphan in patients with monogenic brachydactyly andhypertension (HTN) and in young normotensive control subjects.In control subjects, sensitivity to pressor effect of phenylephrineincreased 15-fold after interruption of baroreflex with tri-methaphan. Patients were extremely hypersensitive to pressoreffect of phenylephrine before ganglionic blockade; with gangli-onic blockade, sensitivity to phenylephrine increased only1.5-fold.

Jordan et al Baroreflex Buffering in Monogenic Hypertension 2615

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

not evenly distributed throughout the body,18 muscle sympa-thetic nerve activity may not necessarily reflect overallsympathetic tone. Yet, blood pressure decreased with com-plete ganglionic blockade to a lesser degree than in controlsubjects,15,19which further supports there being no increase inbasal sympathetic tone. Blood pressure during ganglionicblockade was still greater in patients than in control subjects.Thus, the increase in basal blood pressure is not entirelyexplained by sympathetic nervous system activation. Wecannot completely exclude chronic sympathetic stimulationand vascular remodeling. Stroke volume, on the other hand,decreased sharply in our patients, which may be due to adecrease in cardiac preload or cardiac contractility.20

Several lines of evidence suggest that the autonomic nervoussystem contributes to hypertension in our patients. Perhaps thestrongest evidence is the profound increase in blood pressurewith standing. The increase was attenuated with volume loading,

suggesting that the excessive increase in blood pressure was dueto overcompensation of a reduction in cardiac preload. Similarly,leg compression prevents the orthostatic increase in bloodpressure in patients with orthostatic hypertension.21 Moreover,stimuli such as cold pressor testing and hand-grip exercise led toa greater than normal pressor response. Although blood pressureresponses to these stimuli were excessive, the increase in musclesympathetic nerve activity was moderate. Thus, a smaller in-crease in sympathetic nerve traffic may elicit a greater pressorresponse.

If a smaller increase in sympathetic nerve traffic is associatedwith an excessive pressor response, the amount of norepineph-rine with a given stimulus must be increased or the same amountof norepinephrine elicits a greater response. Plasma and urinarynorepinephrine concentrations were not increased in our pa-tients, evidence against increased norepinephrine release. Lesslikely is the explanation that increased norepinephrine releasecould be masked by changes in norepinephrine clearance.22

Therefore, the same amount of norepinephrine may elicit agreater pressor response. Indeed, our patients were extremelyhypersensitive to the phenylephrine pressor response beforeganglionic blockade. Ganglionic blockade can be used to studyvascular responses in the absence of baroreflex-mediatedchanges in autonomic tone.11 The fact that the sensitivity tophenylephrine was similar in patients and in control subjectsduring ganglionic blockade challenges the concept that increasedvascular sensitivity underlies the hypersensitivity tophenylephrine.

The most likely explanation for phenylephrine hypersensi-tivity in our patients is impaired baroreflex buffering. Themain purpose of the baroreflexes is to buffer changes incardiac preload or vascular tone. In patients with damage tothe afferent arc (baroreflex failure), the sensitivity to phenyl-ephrine and nitroprusside is dramatically increased.23 Simi-larly, impaired function of the efferent arc, due to eitherneuronal degeneration (autonomic failure)24–27 or ganglionicblockade,11,28,29 causes a profound increase in sensitivity tovasodilators and vasoconstrictors. In our patients, phenyleph-rine sensitivity was similar to the sensitivities observed inpatients with baroreflex failure23 or autonomic failure24 or incontrol subjects during ganglionic blockade.11 Furthermore,the sensitivity to phenylephrine increased only slightly duringcomplete ganglionic blockade, suggesting that the restrainingeffect of the autonomic nervous system was profoundlyimpaired. The finding that plasma vasopressin concentrationdid not increase despite a marked depressor response isfurther evidence for impaired baroreflex function.30,31 We15

and others32 have shown that in healthy subjects, ganglionicblockade with trimethaphan leads to a profound baroreflex-mediated increase in vasopressin concentrations. Paradoxi-cally, the baroreflex control of heart rate was only mildlyimpaired, suggesting that the baroreflex dysfunction involvedprimarily control of vascular tone and of vasopressin release.The wide spontaneous fluctuations in blood pressure typicalfor patients with baroreflex failure23,33 were not observed inour patients. These findings suggest that noninvasive orinvasive determination of the baroreflex–heart rate slope34,35

does not give sufficient information on baroreflex control ofvascular tone and vasopressin release.36 Reliance on this

Figure 7. Changes in systolic blood pressure with incrementalbolus doses of nitroprusside before ganglionic blockade withtrimethaphan in patients with monogenic brachydactyly andhypertension (HTN) and in young normotensive control subjects.Patients were hypersensitive to depressor effect ofnitroprusside.

Figure 8. Doses of nitroprusside that decreased systolic bloodpressure 12.5 mm Hg before ganglionic blockade. Patients were9-fold more hypersensitive to depressor effect than controlsubjects.

2616 Circulation November 21, 2000

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

parameter alone in our subjects would have led to faultyconclusions regarding their baroreceptor reflex function.

The location of the lesion to the baroreflex in our patientsis difficult. Absence of orthostatic hypotension and the resultsof autonomic reflex testing suggest that the efferent limb ofthe baroreflex is at least in part intact. Impaired baroreflex-mediated vasopressin release suggests that the lesion isproximal to sympathetic efferent neurons in the rostroventro-lateral medulla. Because baroreflex control of heart rate wasonly slightly impaired, whereas control of vascular tone wasseverely affected, we suggest that the lesion must be locatedat a place where heart rate and blood pressure control are inpart separated.

All patients studied and other affected family members hada left-sided vascular PICA loop impinging on the rostroven-trolateral medulla, whereas nonaffected family members hadno such loops.4 Animal studies have shown that pulsatilecompression of the rostroventrolateral medulla exerted by aballoon increases blood pressure, which may be mediatedthrough alteration of afferent neurons traveling from barore-ceptors to the nucleus tractus solitarius or through directactivation of efferent sympathetic neurons.8,37 Our studydemonstrates that in humans, neurovascular contact of therostroventrolateral medulla can be associated with baroreflexdysfunction and changes in sympathetic nervous systemactivation. However, neurosurgical decompression of thebrain stem would be necessary to prove a causal relationshipbetween neurovascular contact and baroreflex dysfunction inthese patients.5,38,39

We conclude that in patients with monogenic hypertension,brachydactyly, and neurovascular contact of the rostroventro-lateral medulla, basal blood pressure was increased evenduring interruption of sympathetic and parasympathetic nerve

traffic. However, sympathetic stimuli such as standing andcold pressor testing caused an excessive increase in bloodpressure. This excessive response cannot be explained byincreased sympathetic nerve traffic or increased vascularsensitivity. Instead, we suggest that the ability of the barore-flex to buffer changes in vascular tone is severely impaired inthese subjects.

AcknowledgmentThis study was supported by a grant-in-aid from AstraZeneca,

Wedel, Germany.

References1. Bilginturan N, Zileli S, Karacadag S, et al. Hereditary brachydactyly

associated with hypertension.J Med Genet. 1973;10:253–259.2. Schuster H, Wienker TE, Bahring S, et al. Severe autosomal dominant

hypertension and brachydactyly in a unique Turkish kindred maps tohuman chromosome 12.Nat Genet. 1996;13:98–100.

3. Schuster H, Wienker TF, Toka HR, et al. Autosomal dominant hyper-tension and brachydactyly in a Turkish kindred resembles essential hyper-tension.Hypertension. 1996;28:1085–1092.

4. Naraghi R, Schuster H, Toka HR, et al. Neurovascular compression at theventrolateral medulla in autosomal dominant hypertension and brachy-dactyly.Stroke. 1997;28:1749–1754.

5. Jannetta PJ, Segal R, Wolfson SKJ. Neurogenic hypertension: etiologyand surgical treatment, I: observations in 53 patients.Ann Surg. 1985;201:391–398.

6. Morimoto S, Sasaki S, Miki S, et al. Neurovascular compression of therostral ventrolateral medulla related to essential hypertension.Hyper-tension. 1997;30:77–82.

7. Geiger H, Naraghi R, Schobel HP, et al. Decrease of blood pressure byventrolateral medullary decompression in essential hypertension.Lancet.1998;352:446–449.

8. Jannetta PJ, Segal R, Wolfson SKJ, et al. Neurogenic hypertension:etiology and surgical treatment, II: observations in an experimentalnonhuman primate model.Ann Surg. 1985;202:253–261.

9. Robertson D. Pharmacology: assessment of autonomic function. In:Baughman KL, Greene BM, eds.Clinical Diagnostic Manual for theHouse Officer. Baltimore, Md: Williams & Wilkins; 1981:86–101.

Figure 9. Spontaneous muscle sympathetic nerve activity in 2 patients with monogenic brachydactyly and hypertension (top) and in anormotensive and a hypertensive control subject of similar age (bottom). Muscle sympathetic nerve activity was not increased inpatients.

Jordan et al Baroreflex Buffering in Monogenic Hypertension 2617

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from

10. Hokanson E. An electrical calibrated plethysmograph for direct mea-surement of limb blood flow.Trans Biomed Eng. 1999;22:25–29.

11. Shannon JR, Jordan J, Black B, et al. Uncoupling of the baroreflex byNN-cholinergic blockade in dissecting the components of cardiovascularregulation.Hypertension. 1998;32:101–107.

12. Tank J, Baevski RM, Fender A, et al. Reference values of indices ofspontaneous baroreceptor reflex sensitivity.Am J Hypertens. 2000;13:268–275.

13. Schobel HP, Heusser K, Schmieder RE, et al. Evidence against elevatedsympathetic vasoconstrictor activity in borderline hypertension.J Am SocNephrol. 1998;9:1581–1587.

14. Goldstein DS, Eisenhofer G, Stull R, et al. Plasma dihydroxyphenylglycoland the intraneuronal disposition of norepinephrine in humans.J ClinInvest. 1988;81:213–220.

15. Jordan J, Shannon JR, Black BK, et al. NN-Nicotinic blockade as an acutehuman model of autonomic failure.Hypertension. 1998;31:1178–1184.

16. Tarazi RC, Dustan HP. Neurogenic participation in essential and reno-vascular hypertension assessed by acute ganglionic blockade: correlationwith haemodynamic indices and intravascular volume.Clin Sci. 1973;44:197–212.

17. Grassi G, Cattaneo BM, Seravalle G, et al. Baroreflex control of sympa-thetic nerve activity in essential and secondary hypertension.Hyper-tension. 1998;31:68–72.

18. Esler MD. Clinical application of noradrenaline spillover methodology:delineation of regional human sympathetic nervous responses.PharmacolToxicol. 1993;73:243–253.

19. White M, Leenen FH. Aging and cardiovascular responsiveness to beta-agonist in humans: role of changes in beta-receptor responses versusbaroreflex activity.Clin Pharmacol Ther. 1994;56:543–553.

20. Aviado DM. Hemodynamic effects of ganglion blocking drugs.Circ Res.1960;8:304–314.

21. Streeten DH, Auchincloss-JH J, Anderson-GH J, et al. Orthostatic hyper-tension: pathogenetic studies.Hypertension. 1985;7:196–203.

22. Meredith IT, Eisenhofer G, Lambert GW, et al. Plasma norepinephrineresponses to head-up tilt are misleading in autonomic failure.Hyper-tension. 1992;19:628–633.

23. Jordan J, Shannon JR, Black B, et al. Malignant vagotonia due toselective baroreflex failure.Hypertension. 1997;30:1072–1077.

24. Robertson D, Hollister AS, Carey EL, et al. Increased vascular beta2-adrenoreceptor responsiveness in autonomic dysfunction.J Am CollCardiol. 1984;3:850–856.

25. Jordan J, Shannon JR, Biaggioni I, et al. Contrasting actions of pressoragents in severe autonomic failure.Am J Med. 1998;105:116–124.

26. Mohring J, Glanzer K, Maciel JJ, et al. Greatly enhanced pressor responseto antidiuretic hormone in patients with impaired cardiovascular reflexesdue to idiopathic orthostatic hypotension.J Cardiovasc Pharmacol. 1980;2:367–376.

27. Jordan J, Shannon JR, Pohar B, et al. Contrasting effects of vasodilatorson blood pressure and sodium balance in the hypertension of autonomicfailure. J Am Soc Nephrol. 1999;10:35–42.

28. Page IH, McCubbin JW. Buffer reflexes, tolerance to ganglioplegics andtheir relationship to enhanced pressor responsiveness.Am J Physiol.1959;197:217–222.

29. Haas E, Goldblatt H. Effects of various ganglionic blocking agents onblood pressure and on activity of pressor agents.J Physiol. 1960;198:1023–1028.

30. Kaufmann H, Oribe E, Miller M, et al. Hypotension-induced vasopressinrelease distinguishes between pure autonomic failure and multiple systematrophy with autonomic failure.Neurology. 1992;42:590–593.

31. Zerbe RL, Henry DP, Robertson GL. Vasopressin response to orthostatichypotension: etiologic and clinical implications.Am J Med. 1983;74:265–271.

32. Baylis PH, Robertson GL. Osmotic and non-osmotic stimulation of vaso-pressin release.J Endocrinol. 1979;83:39P–40P.

33. Robertson D, Hollister AS, Biaggioni I, et al. The diagnosis and treatmentof baroreflex failure.N Engl J Med. 1993;329:1449–1455.

34. deBoer RW, Karemaker JM, Strackee J. Hemodynamic fluctuations andbaroreflex sensitivity in humans: a beat-to-beat model.Am J Physiol.1987;253:H680–H689.

35. Taylor AA. Autonomic control of cardiovascular function: clinical eval-uation in health and disease.J Clin Pharmacol. 1994;34:363–374.

36. Sato T, Kawada T, Inagaki M, et al. New analytic framework for under-standing sympathetic baroreflex control of arterial pressure.Am J Physiol.1999;276:H2251–H2261.

37. Morimoto S, Sasaki S, Miki S, et al. Pulsatile compression of the rostralventrolateral medulla in hypertension.Hypertension. 1997;29:514–518.

38. Jannetta PJ, Gendell HM. Clinical observations on etiology of essentialhypertension.Surg Forum. 1979;30:433–435.

39. Naraghi R, Gaab MR, Walter GF, et al. Arterial hypertension and neu-rovascular compression at the ventrolateral medulla: a comparativemicroanatomical and pathological study.J Neurosurg. 1992;77:103–112.

2618 Circulation November 21, 2000

by guest on June 5, 2013http://circ.ahajournals.org/Downloaded from