Autonomic Dysfunction in Women

with Mitral Valve Prolapse SyndromeF. ANDREW GAFFNEY, M.D., ERLING S. KARLSSON, M.D., WILLIAM CAMPBELL, M.D.,

JAMES E. SCHUTTE, PH.D., J. V. NIXON, M.D., JAMES T. WILLERSON, M.D.,AND C. GUNNAR BLOMQVIST, M.D.

with the technical assistance of Willie E. Moore, Jr.

SUMMARY Autonomic cardiovascular regulation was evaluated in 35 women, 19 with mitral valve prolapseand 16 healthy controls. Heart rate responses to the diving reflex and to phenylephrine infusion werediminished in patients. Noninvasive measures of cardiac output, heart rate, blood pressure, forearm flow andleg volume during lower body negative pressure (LBNP) showed that patients had less lower extremity poolingof blood and had lower forearm conductance. Blood pressures during LBNP rose or remained unchangeddespite decreases in cardiac output of 20-25%. These data indicate that mitral valve prolapse patients have anincreased venous and arterial vasoconstrictor activity. Cardiac output at rest and echocardiographic indices ofcontractility were normal. Patients with a history of significant ventricular arrhythmias had higher heart ratesand lower forward stroke volumes than the other patients or controls.The combined data demonstrate autonomic dysfunction in women with the mitral valve prolapse syndrome

and suggest decreased parasympathetic, increased a- and normal j-adrenergic tone and responsiveness.

PATIENTS WITH MITRAL VALVE PROLAPSEsyndrome (MVPS) may have a variety of cardiac andnoncardiac abnormalities in addition to the charac-teristic valvular lesion with its mid-systolic click andlate systolic murmur. Included are skeletal abnormali-ties such as pectus excavatum and scoliosis, anasthenic build, symptoms such as atypical chest pain,easy fatiguability, abnormal cardiovascular and elec-trocardiographic responses to exercise, ST- and T-wave changes on resting ECGs, and a variety of atrialand ventricular arrhythmias.1 2 Attempts have beenmade to relate these findings with a single unifyingconcept, but no satisfactory explanation has beenpresented. The presence of chest pain, ST-T-waveabnormalities, and arrhythmias in the absence ofhemodynamically significant valvular, myocardial orcoronary artery disease suggests the possibility of afunctional disorder involving the autonomic nervoussystem. Many of the clinical features of the MVPS arefound in other conditions which have been attributedto some type of autonomic dysfunction, e.g., neuro-circulatory or vasoregulatory asthenia.3 To test thehypothesis that autonomic dysfunction is present inpatients with MVPS, we used a comprehensive set ofnoninvasive procedures to assess regulation of cardio-vascular function.

From the Pauline and Adolph Weinberger Laboratory for Car-diopulmonary Research and the Harry S. Moss Heart Center,Southwestern Medical School, University of Texas Health ScienceCenter at Dallas, Dallas, Texas.

Supported by NASA grant NSG9026, The American HeartAssociation, Inc., Texas Affiliate, and the Ischemic SCOR grant5-P50-HL1 17669.Address for reprints: F. Andrew Gaffney, M.D., University of

Texas Health Science Center at Dallas, Division of Cardiology,5323 Harry Hines Blvd., Dallas, Texas 75235.

Received August 22, 1978; revision accepted November 17, 1978.Circulation 59, No. 5, 1979.

Materials and MethodsPatients

The protocol was approved by the institutionalHuman Research Review Committee. The records ofthe Cardiographics Laboratory of the University ofTexas Southwestern Medical School, Dallas, Texaswere reviewed to obtain the names of patients with thediagnosis of mitral valve prolapse. Private car-diologists in the community were also contacted fornames of patients who might enter the study. Westudied only women because most of our MVPSpatients are women and because we wished to avoidproblems of data interpretation due to inherent car-diovascular differences between normal men andwomen. One patient was excluded because she re-quired propranolol for control of arrhythmias. Fouradditional patients were taking propranolol at thetime they agreed to participate, but all medicationswere discontinued for at least 72 hours before thisstudy. Nineteen female patients (mean age 30 ± 1.8years) were included.The diagnosis of MVPS was confirmed by echo-

cardiography using the criteria of DeMaria et al.4 orby phonocardiography demonstrating a nonejectionsystolic click-murmur complex which changed in thetypical fashion in response to amyl nitrite inhalationor the Valsalva maneuver.2 The laboratory andclinical findings of the patients are summarized intable 1.None of the patients had clinical findings suggestive

of hemodynamically significant mitral regurgitationor echo- or electrocardiographic evidence of left atrialor ventricular enlargement.We graded the severity and frequency of each

patient's symptoms and ventricular and supraventric-ular arrhythmias on a scale of 0-4+ (as defined intable 1) on entry to the study, and used the sum of

894

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

AUTONOMIC DYSFUNCTION IN MITRAL PROLAPSE/Gaffney et al.

TABLE 1. Clinical Information

Pt Skeletal Chest Arrhythmiasno Age/race abnormalities pain SOB Palpitations Ventricular Supraventricular Rest ECG

Group 1

1 34/W Pectus 4+ 2+ 3+ Recurrent V Tach 0 Minor IVCD2 31/W Pectus, scoliosis 3+ 3+ 4+ Frequent bigeminy Frequent S Tach Inferior Ts flat

V tach3 30/W Pectus, scoliosis 2+ 0 4+ Continual bigeminy 0 Diffuse NSSTTW

Rare V Tach changes4 36/W 0 2+ 2+ 4+ Frequent PVCs 0 Early repolarization

with couplets5 26/W Pectus 2+ 3+ 4+ Frequent PVCs Frequent S Tach Inferior ST changes

Rare V Tach6 28/W Pectus 0 2+ 3+ Frequent bigeminy Frequent S Tach Inferior ST changes7 23/W Pectus, scoliosis 4+ 3+ 3+ Occasional PVCs Frequent S Tach Inferior Ts flat8 36/W Pectus 3+ 0 3+ Frequent PVCs 0 Normal9 43/B 0 3+ 1+ - Occasional PVCs 0 Early repolarization

Group 2

10 37/W Pectus, scoliosis 2+ 2+ 3+ 0 Frequent S Tach Normal11 19/W Pectus, scoliosis 3+ 2+ 2+ 0 Frequent S Tach Inferior Qs12 37/W Scoliosis 2+ 1+ 2+ Rare PVCs 0 Inferior ST-T changes13 20/B Pectus, scoliosis 2+ 2+ 2+ 0 Frequent S Tach Early repolarization

Inferior Ts flat14 41/W 0 2+ 1+ 2+ Rare PVCs 0 Incomplete RBBB15 24/B 0 2+ 3+ 2+ 0 Rare S Tach16 31/W 0 2+ 2+ 2+ 0 Frequent S Tach Normal17 46/B Pectus, scoliosis 3+ 2+ 2+ 0 Occasional S Diffuse T flattening

Tach18 19/B Scoliosis 0 0 0 0 0 Inferior ST-T changes19 27/W Pectus, scoliosis 1+ 0 0 0 0 Inferior Ts flat

Patients are ranked in order of clinical severity. Group 1 patients had significant ventricular arrhythmias and generally moresevere symptoms. Group 2 patients were characterized by an absence of ventricular arrhythmias and had less severe symptoms.Symptoms were scored: 0 = never, 1+ rarely and 2+ occasionally mild symptoms, 3+ frequent or moderately severe symptoms,and 4+ at times disabling symptoms.

Abbreviations: Pectus = pectus excavatum; SOB = shortness of breath; V Tach = ventricular tachyeardia; IVCD intra-ventricular Conduction delay; S Tach = sinus tachyeardia; PVCs = premature ventricular contractions; NSSTTWC = non-specific ST- and T-wave changes; RBBB = right bundle branch block; + = present; 0 = absent; - = information not avail-able.

these scores to rank patients according to clinicalseverity. The patients are listed in table 1 according tothis ranking. A combination of the severity scores andexperimental findings separated the patients into twodistinct subgroups as described in the Results section.

Controls

Sixteen healthy female volunteers (medical schoolpersonnel, students and physicians' wives, mean age28 ± 1.3 years) served as controls. Neither patientsnor controls had hypertension, congestive heart failureor diabetes mellitus. No one was taking neuroactivemedication and all were asked to avoid products con-taining caffeine for at least 12 hours before the study.Six of the patients and two of the controls smoked, butno smoking was allowed within 12 hours of the study.

Monitoring ECGs (Frank's orthogonal leads) wererecorded and displayed continuously. Beat-to-beatheart rate was displayed by a Quinton Cardiotach-

ometer (Model 611). Indirect arterial blood pressurewas determined automatically with a Narco Bio-Systems Electro-Sphygmomanometer (PE-300). Per-formance data derived in our laboratory have beenpublished elsewhere.5We measured forearm blood flow by occlusion

plethysmography using the techniques of Siggaard-Andersen 6 with a specially designed air-filled latexcuff. Correct application of this cuff in each case wasverified by the presence of the usual vasoreactiveresponses to deep respiration and the Valsalvamaneuver. Forearm conductance was calculated as

Forearm flow (ml/min/ 100 g tissue)Mean arterial pressure (mm Hg) - 4

Leg volume changes were determined by a mercury-in-silastic Whitney strain gauge. The silastic tubingwas doubled with 1-cm spacers separating the two sec-

895

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 59. No 5, MAY 1979

tions of the strain gauge. This assembly was thenplaced around the calf at its maximum girth andchanges in calf circumference were converted to theequivalent volume change in the 1-cm-thick section ofcalf under the strain gauge. Changes are reported inpercent change from resting volume.We obtained cardiac outputs with an acetylene

rebreathing technique based on mass spectrometermeasurements, as described previously by thislaboratory.7 Comparison of the results to simulta-neously obtained dye dilution measurements over therange of 5-19 1/min demonstrated no systematicdifference and a linear correlation coefficient of 0.94.Lower body negative pressure (LBNP) was

produced with an airtight box sealed immediatelybelow the level of the iliac crests. The methods andcardiovascular effects of LBNP have been reviewedelsewhere.',

Protocol

Informed written consent was obtained from eachsubject. Care was taken to familiarize each personwith the laboratory personnel, equipment andprocedures to avoid the confounding effects of ex-cessive psychological stress.9 Each person was askedto eat a light meal at least 2 hours before beingstudied.No attempt was made to randomize the order of

testing. However, patients and subjects were testedconcurrently and the order of tests was varied at timesto accommodate requirements for optimal schedulingof patients and controls, computer time, laboratoryspace, etc. Serial resting measurements of heart rate,cardiac output, blood pressure and oxygen uptakewere performed before each intervention to establish asteady or basal resting state. A basal state was felt toexist when successive determinations of theseparameters varied by less than 10% and (for tests inthe supine position) were within 10% of the patient'sdata from previous tests performed as part of thisstudy.

LBNP

Subjects were placed in the LBNP device and rest-ing measurements of heart rate, blood pressure,forearm flow, calf volume, oxygen uptake and cardiacoutput were obtained as described above. LBNP wasthen begun according to the standard NASAprotocol" at levels of -8, -16, -32, and -40 mmHg, with repeat measures of heart rate, blood pressureand calf volume at each level and determination ofcardiac output and forearm flow at levels of -16 and-40 mm Hg. In no case did syncope occur and therewere no complications.

Diving ReflexResting heart rate was recorded in the sitting posi-

tion after a recovery period with repeat heart rate,blood pressure and cardiac output determinations toverify return to a steady resting state. The subject was

then asked to dunk her face in a basin of 00 C water.Each subject was told to take a deep breath before im-mersion but to avoid performing a Valsalvamaneuver. The heart rate before immersion and max-imal and minimal heart rates during immersion werenoted. Response to the dive is biphasic: A tachycardiacaused by anticipation of the test is followed by abradycardia caused mainly by increased vagal tonebut also by withdrawal of sympathetic tone."

Phenylephrine Infusion

After another recovery period and the return ofheart rate, blood pressure and cardiac output to steadystate resting levels, an infusion of phenylephrine HCl(40 A,g/ml) was begun intravenously. Blood pressureand heart rate were measured at 30-second intervalsand the infusion rate was adjusted to produce a rise insystolic blood pressure of 20-40 mm Hg. Heart rate(beats/min) and RR intervals (msec) were eachplotted against systolic, diastolic, mean and pulsepressures. The plots of RR intervals vs mean arterialpressures [ 1/3 (systolic - diastolic) + diastolic]yielded the highest correlation coefficient (mean 0.76for patients and 0.79 for controls). For this reason, weused those regression lines instead of the RR intervalsvs systolic pressure plots used by other authors."2 Anexample from one of our patients is shown in figure 1.The slope of each regression equation was consideredto be an index of baroreflex sensitivity and vagalresponsiveness.", 13

All electrocardiographic, forearm flow, blood pres-sure and leg volume data were recorded on an Elema-Schonander Mingograf eight-channel recorder. Car-diac outputs were computed with an on-line computer

cn

CY)

E

w

z

Cr-

1000

900

800

700

600

0

0 60 70 80 90 100MEAN ARTERIAL BP

(mmHg)FIGURE 1. An example of heart rate response to phenyl-ephrine infusion in patient 5. Linear regression. BP = bloodpressure.

_~~s*y** 0

Oy=1x - 106r=.88

i ILI I I L79 A--

896 CIRCULATION

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

AUTONOMIC DYSFUNCTION IN MITRAL PROLAPSE/Gaffney et al.

(PDP-12) or stored on magnetic tape for subsequentprocessing.7We obtained echocardiographic estimates of mean

left ventricular velocity of circumferential fibershortening (Vcf) at supine rest using proceduresdescribed by Cooper et al.14

All data were entered on punch cards and groupdifferences were analyzed with unpaired t-test or thenonparametric Mann-Whitney U test. Both tests areincluded in the Statistical Package for the SocialSciences program (SPSS 7.0).15 The slopes of regres-sion lines obtained during the phenylephrine infusionwere also calculated with the appropriate SPSSprogram.

ResultsThree of the patients had significant ventricular

arrhythmias before and during the tests, two requir-ing temporary treatment with antiarrhythmics afterthe formal test procedure. One of the control subjectsdeveloped ventricular bigeminy during the phenyl-ephrine infusion and after the diving reflex butreturned to normal sinus rhythm spontaneously. Thisoccurred at heart rates of 38-48 and probablyrepresented an escape mechanism. No other com-plications were encountered, and all patients and sub-jects completed the study protocol.To determine whether patients with MVPS

represented a relatively homogenous group or whethersubgroups were present, various statOiical analyseswere performed using the hemodynamic results andthe clinical data obtained before admission to the ex-perimental part of the study.

We found correlations between overall clinicalseverity and resting heart rate and the degree of vaso-constrictor responsiveness, suggesting that subgroupsdid exist. The first nine patients (group 1) are thosewith a greater frequency of both ventricular andsupraventricular arrhythmias as well as more severeclinical symptoms. Patients in group 2 had no signifi-cant ventricular arrhythmias and presented most oftenwith complaints of chest pain only. Because of bothhemodynamic and clinical homogeneity, data fromthe two groups are presented separately. The hemo-dynamic data from the LBNP studies are presented infigure 2. There were no significant differences betweenmean values of cardiac index, stroke index or bloodpressure at rest. Group 1 patients had higher restingheart rates and group 2 patients had lower forearmconductances. At -16 mm Hg, LBNP patients in bothgroups had significantly less venous pooling than con-trols (p < 0.002) as measured by changes in calfvolume, but group 2 patients nevertheless had lowerforearm conductance (p < 0.05). At -40 mm Hg,LBNP, mean blood pressure of the patients rose(group 2) or remained stable (group 1) while that ofthe controls fell. Calf volume changes remained lessmarked in patients than in controls (p < 0.01).Measurements of forearm conductance at this levelshowed a marked vasoconstrictor effect in all groups,but the controls still showed greater conductance thanthe patients (p < 0.05). Thus, the patients in bothgroups had significantly less venous pooling of bloodin the legs during LBNP at any given pressure level,but greater arterial vasoconstriction as measured byforearm conductance and blood pressures. The low

x

Cz

W

0Ir-Cn

x

0z

C-)4

cr

0 -16 -40mm Hg

0 -16 -40

mm Hg

E0

.0

E

E

-J

W

z4t

z0

IAt

50

0 -16 -40mm Hg

4.0

&53.02.52.00

0 -16 -40

mm Hg

00x

ObzEE

00

Nl

E

3 -F>i

0 -16 -40mm Hg

II--4

UA. ,o

0 -16 -40

mm Hg

Mitral Valve Prolopse, Group I 0.*Mitral Valve Prolopse, Group I *----aNormal Controls o0 o

* Significant difference- Group I vs Group+ Significant difference- Patient Group vs Controls

FIGURE 2. Hemodynamic datafrom lower body negative pressure studies. Values are mean ± SEM.

897

908580757065

908580750

W

< .ccr E

o

in

4 4

z

W

C,U)W

0-

W E

z4Lh

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 59, No 5, MAY 1979

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

MITRALVALVE

PROLAPSE

NORMAL

FIGURE 3. Distribution of individual slopes of regressionlines obtained during phenylephrine infusion. Mean values+ sEMare also indicated.

degree of venous pooling suggests diminished venouscompliance. A comparison based on absolute ratherthan relative leg volume changes would have producedeven greater differences between patients and controls.Absolute leg dimensions were significantly smaller inthe patients. All groups showed the expected progres-sive decrease in stroke volume and cardiac output, butthe changes were less marked in group 2.

Heart rate responses to phenylephrine infusion weremarkedly attenuated in the MVPS patients (fig. 3).Both patients and normal subjects had similar restingmean arterial blood pressure (82 ± 2.4, 76 ± 2.0 mmHg, p = NS). The patient RR/BP regression lineslopes were significantly different from the controls(1I ± 1.3, 19 ± 1.5; RR msec/BP mm Hg ± SEM).There was minimal overlap between the patients andnormal subjects with only three patients' values ap-

TABLE 2. Heart Rate Response to the Diving Maneuver(Beats/min, mean - SEM).

Pre-Dive Maximum Minimum Range*

Group 1 93=6 102 6 63=8 39 - 8

Group 2 80 4 92= 3t 57=4 35d4tControls 91=6 108l 5 51= 3 57 i 5

*Difference between maximal and minimal heart rate.tP <0.01 for the difference group 2 vs controls.

proaching the mean - 2 SEM of the control patients'values. No statistical relationship between slope andage or resting blood pressure was found among con-trols or patients, indicating that these factors did notinfluence the results. Groups 1 and 2 had virtuallyidentical mean slopes.The patients' response to facial immersion (diving

reflex) was also different from that of the normals.Both patients and normals were able to maintain im-mersion for an average of 21 seconds and no one hadan immersion time of less than 14 seconds. The resultsare given in table 2. The difference between maximaland minimal heart rates during this procedure wassmaller in the patients (37 ± 4 beats/min for thegroups 1 and 2 combined, compared with 57 ± 5 inthe control group, p = 0.005). Baseline and peak rateswere similar in the control group and in group 1. Bothmaximal and minimal heart rates were lower in group2 but the difference between the rates was of the samemagnitude in both patient groups whether measuredas beats/min (table 2) or as percent of pre-dive heartrate. The difference corresponded to 41.9% in group 1and 43.8% in group 2 compared with 62.6% in the con-trol group (p < 0.02 and p < 0.01). Minimal heartrates tended to be lower in the control group, but thedifference between patients and controls only ap-proached significance (p < 0.07) when rates were ex-pressed as percent of pre-dive heart rates.

Echocardiography demonstrated that the patients'Vcfs and shortening fractions were within normallimits for this laboratory and did not differsignificantly from those of the controls. Patients'mean Vcf was 0.90 ± 0.08 diameters/sec and meanshortening fraction was 32 ± 2%.

DiscussionThe present study examined vasoregulatory func-

tion in patients with MVPS. Because vasoregulatoryresponses often involve activation and/or inhibition ofboth major components of the autonomic nervoussystem, it is difficult in the intact human subject to at-tribute a response exclusively to a single component.However, the various test procedures will becategorized, whenever possible, by their majorautonomic effect.The patients' response to phenylephrine infusion is

consistent with diminished vagal responsiveness. Thepatients showed significantly less heart rate decreasefor a given rise in blood pressure. Sympatheticwithdrawal may contribute to the response, but

tLU0L0-J

zcn

z0C,)Uf)ZOcrnDlJ

0

- 0

* ** 18.6±146

* 0

_ T0

7111.2±1.27

00

0

0a

898 CIRCULATION

I

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

AUTONOMIC DYSFUNCTION IN MITRAL PROLAPSE/Gaffney et al.

human data supporting a major parasympatheticcomponent have been presented by Pickering et al.16This procedure has not been previously studied inpatients with MVPS but has been used as a test ofvagal function in a variety of other cardiac diseasestates. Bristow et al.12 demonstrated diminished baro-reflex sensitivity in patients with hypertension.Eckberg et al.'3 examined patients with congestiveheart failure of diverse etiologies and similarly showedmarkedly diminished baroreflex sensitivity which theyinterpreted as evidence of abnormal parasympatheticfunction. The younger normal controls in those studieshad a heart rate response to phenylephrine similar tothat of our controls, but their older controls (con-siderably older than our subjects) had diminishedresponsiveness. Age, hypertension and congestiveheart failure can be excluded as causes of the findingsin our patients. Patients and controls had nearly iden-tical mean ages. There were no signs of congestiveheart failure, and the blood pressures of both thepatients and controls were in the low-normal range.The patients' response to facial immersion supports

the hypothesis of altered vagal function. The divingreflex evokes an extremely potent vagal response andcan be used clinically to treat supraventricular tachy-arrhythmias.'7 The typical bradycardic response canbe elicited even during the strong ,B-adrenergic drivepresent during near-maximal exercise.'8 MVPSpatients were less responsive to the diving maneuver,with substantially less decrease in heart rate suggest-ing diminished vagal responsiveness. However, theresults are not as clear-cut as those from thephenylephrine infusion because group 2 patients alsohad an attenuated initial tachycardic response to thedive. Diminished vagal response to the diving reflexhas also been reported in patients several weeks aftermyocardial infarction.'9While the responses to phenylephrine infusion and

facial immersion are likely to be primarily vagallymediated, the response to LBNP is considerably morecomplex. The primary hemodynamic stimulus duringLBNP is peripheral venous pooling with an acutedecrease in circulating blood volume. Previousstudies5 have shown that increasing levels of LBNPcause a progressive decrease in left ventricular end-diastolic volume and stroke volume. Arterial bloodpressure, left ventricular contractile state and heartrate are normally unchanged during moderate levels(-10 to -30 mm Hg). The fall in cardiac output is ini-tially compensated for by peripheral vasoconstric-tion, manifest as decreasing forearm conductance andlater by splanchnic vasoconstriction as well.Presumably, the initial vasoconstrictor responses areprimarily mediated by activation of low pressure baro-receptors5' 20 and occur before any measurablechanges in arterial pressure are induced. Pressuresbeyond -40 mm Hg are likely to cause a fall inarterial pressure, activate the carotid sinus mechanismand produce a compensatory tachycardia.

Baseline hemodynamic values for the two groupswere similar except for a relative tachycardia in group1 patients. Significant differences developed during

LBNP. The patients had less lower extremity venouspooling and lower forearm conductance values andhad a paradoxical rise in arterial blood pressure.Samueloff et al. have shown that there is very littleacute change in venous compliance in response toLBNP stress,2' so the diminished pooling in thepatient group probably resulted from lower basalvenous compliance caused by either structuralvascular differences or increased a-adrenergic tone inMVPS patients.

Patients with ventricular arrhythmias (group 1) hadchanges in stroke index and cardiac index similar tothose in controls, but had significantly higher heartrates despite enhanced peripheral vasoconstrictorresponses. Group 2 patients had a smaller strokevolume reduction and less tachycardia. Themechanisms are not clear but the differences may bethe result of a complex interaction between changes inthe left ventricular contractile state and in the degreeof prolapse and mitral regurgitation related to thedecrease in preload and left ventricular end-diastolicvolume during LBNP.5 Actual chamber size was notmeasured during this study.

Although the findings of an enhanced arterial vaso-constrictor response and decreased venous compliancesuggest abnormal a-adrenergic activity, there isevidence suggesting normal f-adrenergic activity.There were no significant differences in resting bloodpressure, cardiac output or stroke volume betweenpatients and controls and the patients' restingechocardiographic indices of contractility were nor-mal. It is unlikely that excessive psychological stressamong the patients caused the abnormal response toLBNP, since most of the patients had been in thelaboratory several times before the study and werefamiliar with the laboratory equipment and personnel.Many of the patients and controls dozed frequentlyduring the testing and appeared relaxed most of thetime (the diving reflex was a notable exception).Our findings indicating abnormal autonomic func-

tion are supported in part by a preliminary report ofCoghlan et al.22 They examined MVPS patients' heartrate response to the Valsalva maneuver and to 90°head-up tilting. The patients had less tachycardia thancontrol subjects early after return to the supine posi-tion during the tilt test and a prolonged and morepronounced phase of bradycardia after the Valsalvamaneuver. These results may at first seem to con-tradict ours. Differences between the studies inmethods and patient selection may account for thedifferent results. However, the combined data mayalso be interpreted as suggesting that some patientsmay be vagally hyporesponsive to low or moderatelevels of cardiovascular stimulation, but hyper-reactive to higher levels of stimulation, at least interms of the duration of the response and perhaps alsothe magnitude. This type of abnormality would tendto create an inefficient homeostatic control systemcharacterized by a high threshold and underdamping,or a vagal stimulus-response curve displaced to theright with an abnormally steep slope. The presence ofboth sinus and ectopic tachyarrhythmias and severe

899

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 59, No 5, MAY 1979

bradyarrhythmias23 in MVPS patients appears to beconsistent wih this explanation. None of the patientsin this study suffered from bradyarrhythmias andtherefore would not be expected to show the excessivevagotonia which may have been present in some ofCoghlan's patients.The common occurence of ST-T abnormalities at

rest and during stress in MVPS patients is consistentwith the concept of autonomic dysfunction. It is wellknown that both central and peripheral autonomicalterations can produce these electrocardiographicchanges. The autonomic origin of ST- and T-wavechanges seen in cerebrovascular accidents has beenwell documented.24 Orthostatic ST-T abnormalitiesoften occur in patients with other functional cardio-vascular abnormalities25 and can be produced by smallintravenous doses of epinephrine.26 ST-T abnor-malities may reflect inhomogeneities with respect toelectrical recovery and state of excitation, i.e., con-ditions promoting the development of ventriculararrhythmias.27 Several investigators have attempted tolink the presence of ST-T abnormalities with a highincidence of arrhythmias in patients with MVPS.28-3'Indirect support for a relationship between autonomicdysfunction and arrhythmias in MVPS patients maybe derived from the data of Combs et al., who studiedthe electrocardiographic responses to a standardizedpsychological stress situation.32 Several patientsdeveloped significant arrhythmias.Most authors have attributed the arrhythmias and

chest pain of MVPS to localized papillary muscleischemia created by excessive tension of the mitralchordae tendineae33 or to a diffuse cardiomyopathy.34Our thallium-201 perfusion studies35 and those ofMassie et al.36 probably rule out significant ischemiacaused by a fixed perfusion deficit, but these studies donot exclude intermittent spasm. The presence of ex-cessive peripheral vasoconstrictor activity in MVPSpatients raises the possibility of coronary arteryspasm, but the exact relationship has not been deter-mined. LeWinter et al. produced the typical chest painin MVPS patients, but infused phenylephrine in dosessubstantially larger than used in this study.37 They in-terpreted this as being caused by increases in myocar-dial tension causing ischemic pain. A direct a-adrenergic coronary vascular response is an alternateexplanation. The patients in our study with the mostsevere chest pain (group 1) did not seem to have highervasoconstrictor activity than the other patients (group2) in response to acute volume depletion duringLBNP, but other stimuli may produce more powerfula-adrenergic responses.The clinical symptoms of MVPS resemble those of

the "vasoregulatory asthenia" patients of Holmgrenet al.3 25 Their patients were hyperkinetic and hadlarge cardiac outputs and narrow arteriovenous 02differences at rest and during exercise. While MVPSpatients clearly do not have these hyperkineticchanges, both groups commonly complain of atypicalchest pain, shortness of breath, easy fatiguability andpalpitations. Holmgrens' patients showed substantialimprovement after treatment designed to increase

vagal and decrease sympathetic tone, i.e., physicaltraining.38 Whether MVPS patients would respondfavorably to physical training remains to be studied.Further tests, including both vagal and a- and f-adrenergic blockade, are needed before the clinicalsymptomatology and electrocardiographic abnormali-ties of MVPS can be attributed to malfunction ofspecific components of the autonomic nervous system.

In conclusion, the results provide strong evidencefor autonomic dysfunction in patients with MVPS,probably involving both the sympathetic and para-sympathetic systems. The exact nature of theautonomic defects remains to be established. Whetherthese defects occur simply as another aspect of thesyndrome, i.e., as an expression of a generalizedgrowth disorder, also manifest as skeletal abnor-malities and an anthropometrically distinct MVPShabitus (Schutte JE, Gaffney FA, Blomqvist CG: un-published data), or as an abnormal response to thevalvular regurgitation itself is not known. Theseautonomic abnormalities may be a result of defectivesensing, inadequate central processing and output oraltered end-organ responsiveness, with or without un-derlying structural abnormalities of the nervoussystem.

References1. Devereux RB, Perloff JK, Reichek N, Josephson ME: Mitral

valve prolapse. Circulation 54: 3, 19762. Jeresaty RM: Mitral valve prolapse-click syndrome: etiology,

clinical findings, and therapy. Cardiovasc Med 3: 597, 19783. Holmgren A, Jonsson B, Levander M, Linderholm H,

Sjostrand T, Strom G: Low physical working capacity insuspected heart cases due to inadequate adjustment ofperipheral blood flow (vasoregulatory asthenia). Acta MedScand 158: 413, 1957

4. DeMaria AN, Neumann A, Lee A, Mason DT: Echocardio-graphic identication of the mitral valve prolapse syndrome. AmJ Med 62: 819, 1977

5. Ahmad M, Blomqvist CG, Mullins CB, Willerson JT: Left ven-tricular function during lower body negative pressure. AviatSpace Environ Med 48: 512, 1977

6. Siggaard-Anderson J: Venous occlusion plethysmography onthe calf. Danish Med Bull 17 (suppl 1): 1, 1970

7. Triebwasser JH, Johnson RL, Burpo RC, Campbell JC,Reardon WC, Blomqvist CG: Non-invasive determination ofcardiac output by a modified acetylene rebreathing procedureutilizing mass spectrometer measurements. Aviat Space En-viron Med 48: 203, 1977

8. Wolthuis RA, Bergman SA, Nicogossian AE: Physiologicaleffects of locally applied reduced pressure in man. Physiol Rev54: 566, 1974

9. Epstein SE, Beiser GD, Stampfer M, Braunwald E: Role ofvenous system in baroreceptor-mediated reflexes in man. J ClinInvest 47: 139, 1968

10. Johnson RL, Hoffler GW, Nicogossian AE, Bergman SA,Jackson MM: Lower body negative pressure: third mannedSkylab mission. In BioMedical Results from Skylab, edited byJohnston RS, Dietlein LF. Washington, DC, NASA Sci Pub,1977, pp 284-313

11. Strauss MB: Physiological aspects of mammalian breath-holddiving: a review. Aerospace Med 41: 1362, 1970

12. Bristow JD, Honour AJ, Pickering TG, Sleight P, Smyth HS:Diminished baroreflex sensitivity in high blood pressure. Cir-culation 39: 48, 1969

13. Eckberg DL, Drabinsky M, Braunwald E: Defective cardiacparasympathetic control in patients with heart disease. N EnglJ Med 285: 877, 1971

14. Cooper R, O'Rourke JS, Karliner RA: Comparison of ultra-

900 CIRCULATION

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

AUTONOMIC DYSFUNCTION IN MITRAL PROLAPSE/Gaffney et al.

sound and cineangiographic measuerements of Vcf in man. Cir-culation 46: 914, 1972

15. Nie NH: SPSS: Statistical Package for the Social Sciences, 2nded. New York, McGraw-Hill, 1975

16. Pickering TG, Gribbin B, Petersen ES, Cunningham DJC,Sleight P: Effects of autonomic blockade on the baroreflex inman at rest and during exercise. Circ Res 30: 177, 1972

17. Wildenthal K, Leshin SJ, Atkins JM, Skelton CL: The divingreflex used to treat paroxysmal atrial tachycardia. Lancet 1: 12,1975

18. Bergman SA, Campbell JK, Wildenthal K: "Diving reflex" inman: its relation to isometric and dynamic exercise. J ApplPhysiol 33: 27, 1972

19. Ryan C, Hollenberg M, Harvey DB, Gwynn R: Impaired para-sympathetic responses to patients after myocardial infarction.Am J Cardiol 37: 1013, 1976

20. Zoller RP, Mark AL, Abboud FM, Schmidt A, Heistad DD:The role of low pressure baroreceptors in reflex vasoconstrictorresponses in man. J Clin Invest 51: 2967, 1972

21. Samueloff SL, Browse NL, Shepherd JT: Response of capacityvessels in human limbs to head-up tilt and suction on lowerbody. J Appl Physiol 21: 47, 1966

22. Coghlan HC, Irwin PK, Cowley MJ, Copley DP, Gillis DR,James TN: Abnormal heart rate response in mitral valveprolapse syndrome: results with valsalva and tilt testing. (abstr)Clin Res 25: 45A, 1976

23. DeMaria AN: The syndrome of mitral valve prolapse:problems and perspectives. (Editorial) Ann Intern Med 85: 525,1976

24. Surawicz B: The pathogenesis and clinical significance ofprimary T-wave abnormalities. In Advances in Electrocardiog-raphy, edited by Schlant RC, Hurst JW. New York, Grune andStratton, 1972, pp 377-421

25. Holmgren A, Jonsson B, Levander M, Linderholm H,Sj6strand T, Str6m G: ECG changes in vasoregulatory astheniaand the effect of physical training. Acta Med Scand 165: 259,1959

26. Mitchell JH, Shapiro AP: The relationship of adrenaline and T-wave changes in the anxiety state. Am Heart J 48: 323, 1954

27. Lathers CM, Kelliher GJ, Roberts J, Beasley AB: Non-uniformcardiac sympathetic nerve discharge. Circulation 57: 1058, 1978

28. Swartz MH, Teicholz LE, Donoso E: Mitral valve prolapse: areview of associated arrhythmias. Am J Med 62: 377, 1977

29. Jeresaty RM: Sudden death in the mitral valve prolapse-clicksyndrome. Am J Cardiol 37: 317, 1976

30. Campbell RWF, Goodman MG, Fiddler GI, Marquis RM,Julian DA: Ventricular arrhythmias in syndrome of balloondeformity of mitral valve: definition of possible high risk group.Br Heart J 38: 1053, 1976

31. Winkle RA, Lopes MG, Popp RL, Hancock EW: Life-threatening arrhythmias in the mitral valve prolapse syndrome.Am J Med 60: 961, 1976

32. Combs R, Shah PM, Shulman R, Klorman R, Sylvester L:Effects of psychological stress on click and rhythm in mitralvalve prolapse. (abstr) Circulation 56: (suppl III): 111-1 11, 1977

33. Aranda JM, Befeler B, El-Sherif N, Castellanos A, Lazzara R:Mitral valve prolapse: recent concepts and observations. Am JMed 60: 997, 1976

34. Gulotta SJ, Gulco L, Padmanabhan V, Miller S: The syndromeof systolic click, murmur, and mitral valve prolapse- a cardio-myopathy? Circulation 49: 717, 1974

35. Gaffney FA, Wohl AJ, Blomqvist CG, Parkey RW, WillersonJT: Thallium-201 myocardial perfusion: studies in patients withthe mitral valve prolapse syndrome. Am J Med 64: 21, 1978

36. Massie B, Botvinick EH, Shames D, Taradash M, Werner J,Schiller N: Myocardial perfusion scintigraphy in patients withmitral valve prolapse: its advantage over stress electrocardiog-raphy in diagnosing associated coronary artery disease and itsimplications for the etiology of chest pain. Circulation 57: 19,1978

37. LeWinter MM, Hoffman JR, Shell WE, Karliner JS,O'Rourke RA: Phenylephrine-induced atypical chest pain inpatients with prolapsing mitral valve leaflet. Am J Cardiol 34:12, 1974

38. Holmgren A, Jonsson B, Levander M, Linderholm H,Mossfeldt F, Sj6strand T, Str6m G: Physical training ofpatients with vasoregulatory asthenia. Acta Med Scand 158:437, 1957

901

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from

BlomqvistF A Gaffney, E S Karlsson, W Campbell, J E Schutte, J V Nixon, J T Willerson and C G

Autonomic dysfunction in women with mitral valve prolapse syndrome.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1979 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.59.5.894

1979;59:894-901Circulation. 

http://circ.ahajournals.org/content/59/5/894the World Wide Web at:

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

  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 information about this process is available in the

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

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

by guest on June 16, 2018http://circ.ahajournals.org/

Dow

nloaded from