Neurohumoral and hemodynamic effects of lower body negative pressure in patients with congestive heart failure

Neurohumoral and hemodynamic effects of lower body negative pressure in patients with congestive heart failure Baroreflex modulation of forearm vascular resistance (FVR) has been reported to be abnormal in patients with congestive heart failure (CHF). However, the neurohumoral mechanisms for this impairment are not defined. We assessed the responses of arterial pressure, FVR, plasma norepinephrine, and plasma renin activity to lower body negative pressure in 29 patients with compensated CHF (New York Heart Association class Ill and IV) and in 11 normal age-matched control subjects. Baseline mean arterial pressure (83 + 2 vs 84 I 2 mm Hg) and mean arterial pressure during LBNP (-10, -20, and -40 mm Hg) were not significantly different in the two groups. Basal FVR (43.7 i 4 vs 27 + 2 units), plasma norepinephrine (805 i- 81 vs 155 ? 8 pg/ml), and plasma renin activity (8.3 + 1.7 vs 1.2 + 0.2 ng/ml/hr) were significantly (p < 0.01) higher in patients with CHF. The relative increases in FVR responses during LBNP of -10, -20, and -40 mm Hg (10 t 4% vs 70 F 12%, 17 F 6% VI 106 f 21%, and 24 t 9% vs 152 i 28%) were markedly attenuated in patients with CHF compared to control subjects. Plasma norepinephrine and plasma renin activity responses during LBNP were also attenuated in patients with heart failure. Our results suggest that baroreflex control of FVR and plasma norepinephrine and plasma renin activity is impaired in CHF because of the inability of the cardiopulmonary baroreceptors to alter sympathetic outflow. (AM HEART J 1989;118:78.)

Pramod K. Mohanty, MD, James A. Arrowood, MD, Kenneth Marc D. Thames, MD. Richmond, Vu.

Cardiovascular reflexes that are mediated by sensory receptors in the heart and blood vessels regulate the cardiovascular system by neural and humoral mechanisms.1-8 In normal states most of these sensory endings function as part of a classical negative feedback loop for control of arterial pressure, heart rate, peripheral vascular resistance, and the release of several humoral substances.11 4 In congestive heart failure (CHF) the receptor function and the efferent mechanisms may be altered resulting in abnormal reflex function. Cardiac dysfunction in experimental animals has been reported to cause impairment of both arterialgel and cardiopulmonary baroreceptor121 I3 control of circulation. These abnormalities may lead to impaired baroreflex-mediated inhibition of the vasoFrom the Hunter Holmes McGuire Veterans ter, and the Medical College of Virginiaflirginia sity.

Administration Medical CenCommonwealth Univer-

Supported by National Heart, Lung, and Blood and by funds from the Veterans Administration. Received

for publication

Oct. 27, 1988;

accepted

Reprint requests: Pramod K. Mohanty, MD, VA Medical Center, Richmond, VA 23249.

78

Institute

grant

March

1, 1989.

Cardiology

Division

HL-30506

(lllA),

A. Ellenbogen,

MD, and

motor center and consequent increases in neurohumoral drive. Increased vascular resistance and neurohumoral drive are prominent features in patients with advanced heart failure.2, 8 Although baroreflex modulation of forearm vascular resistance (FVR) also has been reported to be abnormal,14 abnormalities of neurohumoral mechanisms have not been studied extensively. Moreover, there have been no reports in which forearm vascular responses to lower body negative pressure (LBNP) in patients with heart failure have been correlated with simultaneously measured plasma norepinephrine and renin levels to further assess the status of efferent sympathetic mechanisms. We hypothesized that the presence of significant left ventricular dysfunction in humans results in attenuation of the tonic inhibitory influences of cardiopulmonary baroreflexes under basal conditions and that this appears as impairment of cardiopulmonary baroreflex control of sympathetic outflow as reflected in plasma norepinephrine and renin levels. To test this hypothesis we used the technique of LBNP, which unloads mainly cardiopulmonary baroreceptors.

Volume Number

118 1

METHODS Patient

Cardiopulmonary

The study population included 29 patients with end-stageheart diseasewith severeleft ventricular dysfunction and CHF (New York Heart Association functional classIII or IV). These patients all had left ventricular systolic dysfunction characterized by a left ventricular ejection fraction of lessthan 20%. Most (21 of 29) of the patients were referred to our medical center for cardiac transplantation. All patients were men, aged 19to 55 years (43 -+ 3 years [mean f SE]), who had clinical, hemodynamic,radionuclide, and/or echocardiographicevidence of impaired left ventricular function. The underlying causeof left ventricular dysfunction wasdilated cardiomyopathy in 12 (41%) and ischemicheart diseasein 17 (59%). None of the patients had diabetes mellitus, renal failure, or significant hepatic dysfunction. None of the patients with CHF had acute myocardial infarction during the 6 months immediately precedingthe reflex physiologic studies.All patients were hemodynamically stable, in sinus rhythm, and had no significant ventricular or atria1 tachyarrhythmias. Eleven normal healthy volunteers formed the control group. All were men between the agesof 29 and 57 years (39 + 4 years [mean * SEMI). They had no cardiovascular diseaseaccording to medical history and results of physical examination, ECG, and echocardiographicstudies and were receiving no medication. Written informed consent was obtained from all patients and control subjects. The researchprotocol wasapproved by the Committee on the Conduct of Human Research of Virginia Commonwealth University and the McGuire Veterans Administration Medical Center. Experimental procedures. Reflex physiologic studies were conducted with subjects in the postabsorptive state without premeditation. Caffeine and cigarettes were all prohibited during the 4 hours before the study. In patients with CHF digitalis and vasodilators were withheld for at least 48 hours and diuretics for at least 24 hours before the study. All studies were carried out in a temperaturecontrolled laboratory at 22” C. On the morning of testing the subjectswere brought to the human physiology laboratory. A peripheral venous cannula wasinserted into the left antecubital vein 30 minutes before blood sampling was begun. In 11 patients internal jugular venouscannulation wasaccomplished,and right atria1and pulmonary artery pressuresweremeasured continuously via a 5F Swan-Ganz catheter connected to a Statham P23ID pressuretransducer and recorded simultaneously with heart rate (ECG) and forearm blood flow on a direct-writing Gould physiologic recorder (2800s). Pulmonary artery diastolic pressurewas taken as an index of left ventricular filling pressure.In all subjectspulmonary artery diastolic pressurecorrelated well (within k 3 mm Hg) with pulmonary capillary wedgepressure(measuredat baseline),which is consideredto be an optimal reflection of left ventricular filling pressure.Pulmonary artery diastolic pressurewas thus chosenas an index of left ventricular filling pressurebecauseof the easeand consistencyin continuous recording of the pulmonary arterial pressurecurves selection.

barorefEexes in CHF

79

during LBNP. Systemic arterial pressurewasmeasuredby an automatic cuff method (Critikon model 1160,Tampa, Fla.). Forearm blood flow was measuredby venous occlusion plethysmography with a mercury-in-silicone rubber strain gaugeand plethysmograph (EC-4; D. E. Hokanson, Inc., Seattle, Wash.). The technique of venousocclusion plethysmography has been described previously in detail.15*l6 The strain gaugewasplaced approximately 5 cm below the antecubital creaseof the right arm. The arm was elevated and supported sothat the proximal part of the forearm was approximately 10 cm above the anterior chest wall. The pressureof the venous occlusion or congestingcuff of the arm was 40 to 50 mm Hg. Circulation to the hand wasarrested by inflating a cuff around the wrist to suprasystolic pressuresduring determinations of forearm blood flow. FVR wascalculated by dividing the meanarterial pressure (diastolic plus one third of pulse pressurein mm Hg) by forearm blood flow (expressedasml/min/lOO ml of forearm volume). Values for FVR are expressedas units. Echocardiography.Two-dimensionalechocardiography was performed with a commercially available mechanical sectorscanner (MK-300, Advanced Technology Laboratories, Inc., Bothell, Wash.) with a real-time, high-resolution 2.5 MHz transducer, a go-degreesector angle, displays at tissue depths of 16 and 21 cm, and a videotape recording speedof 30 frames/set. These studies were performed to determine whether the levels of suction used reduced the dimensionsof the cardiac chamber and thus the stimulus to cardiac mechanoreceptors.Echocardiography was performed with subjectsin the supineposition. Standard apical four-chamber and two-chamber views were used for calculation of left ventricular volumes. The end-diastolic frame was chosenat the peak of the R wave of the simultaneously recorded ECG. The frame in which the lowest left ventricular volume wasrecorded was identified asend systole. Volumes (normalized for body surface area) were calculatedby modified Simpson’srule methodI l8with the useof a microcomputer-basedimage analysissystem (Microsonic, Inc., Indianapolis, Ind.). M-mode left atria1 echocardiographictracings were derived from the parasternal short-axis view of the twodimensional echogramat the level of the aortic-tricuspid valve plane. Left atria1 end-systolic dimension(normalized for body surface area)wasmeasuredaccordingto the standard methodlg at baseline,during eachlevel of LBNP, and during the recovery phaseafter eachintervention. Baseline values were reestablishedduring each recovery phase. Interventions. LBNP was used to unload receptors in the cardiopulmonary region, which normally results in reflex sympathetic activation.20 This was accomplishedby positioning each subject in a chamber that encasedthe body below the iliac crest. The chamber was sealedand connectedto an adjustable vacuum source.Graded LBNP was applied at -10, -20, and -40 mm Hg. LBNP was maintained at each level for 5 minutes after which a lo-minute recovery period wasinterposed before application of the next level of LBNP. Measurementsof forearm blood flow were madeevery 15seconds.Values of the fore-

80 Table

Mohanty

et al.

I. Systemic

July 1989 Heart Journal

American

hemodynamic

responses

(mean rt SEM)

to LBNP LBNP

Baseline

Hemodynamics

HR (beats/min) CHF (n = 29) Control (n = 11) SAP (mm Hg) CHF (n = 28) Control in = 11) DAP (mm Hg) CHF (n = 29) Control (n = 11) MAP (mm Hg) CHF (n = 29) Control (fl = 11) FBF (ml/min/dl) CHF (n = 29) Control (n = 11) FVR (units) CHF (n = 29) Control (n = 11) HR = heart rate; SAP = systolicarterial vascular

resistance; CHF *p < 0.05 “S control. tp < 0.01 vs baseline.

= congestive

-10

-20

mm Hg

mm Hg

-10

mm Hg

79 + 3*

78 i 3

78 -t 3

81 * :I

62

60 Ik 2

63

71

of- 4

111 zk 3* 119 IL 3.5

f

2

+ 3f

109 i 3 117 + 3

108 I 3 116 z!z4

107 L 3 113 :I 3-I

70 zk 2.5

73 i 2 68 + 3

71 It 2 69 k 4

73 +_ 2 73 + 3

86 84

85 t 2 85 +- 3

84 84

84

f

87

-t 3

73 f

2.3 3.4

2

i- 2 f 2

2.2

i 0.2 2 + 0.2t

+ 0.2* I!z 0.3

43 +- 4* 27 + 2 pressure; DBP heart failure.

= diastolic

blood

47 45.6

i- 4.6 zt 4t

pressure;

MAP

arm blood flow weretaken asthe averageof flows (five flow curves) during the last 60 secondsof each control or intervention period. The order of application of the three levels of suction was randomized. Responsesto cold pressorstimulation were determined after immersion of the subject’sleft hand in ice water for 90 secondswhile all hemodynamic parameters including forearm blood flow were determined as described previously. The cold pressortest wasusedto assess responsivenessto another stimulus for reflex sympathetic activation and thus to determine the relative specificity of abnormal responsesto LBNP. Study protocol. The protocol wasbegun a minimum of 30 minutes after preparation of the patient. Baseline hemodynamic and blood sampling for hormonal measurements was then initiated. We studied the hemodynamic, plasmanorepinephrine, and renin (plasmarenin activity) responsesto graded levels of LBNP at -10, -20, and -40 mm Hg and the responsesto cold pressortesting. Blood samples(6 ml) were obtained before and during the last minute of each level of LBNP. Samplesfor determination of catecholamineand renin levels were collected in prechilled EDTA and heparinized tubes, respectively, and plasmawasstored at -75’ C until assay.Plasmanorepinephrine levels were measured by high-performance liquid chromatography with electrochemical detection.21 Detection in this system wascalorimetric and the sensitivity was <5 pg/ml. Plasma renin activity was measuredby radioimmunoassayof angiotensinI after titration to pH 7.4 with phosphatebuffer and incubation at 37” C in the presenceof angiotensinaseinhibitors.22Sensitivity of this assay is 0.2 ng/ml/hr and intraassaycoefficient of variation is 4 %. Statistical analysis. The results of the experiments are

+ 2 + 3

2.1 + 0.2 1.5 k 0.2t

2 i 0.2 1.2 -+ 0.2t

49.7 ?I 4.5 65 i 8t = mean

arterial

pressure;

FBF

2

= forearm

blood

52 i

5.5

91 i

11t

flow; FVR

= forearm

presentedas the mean + SEM. The data before and during LBNP were comparedby multivariate analysisof variancethat alsopermitted determination of whether the responsesweredifferent betweenthe two groups.Differences in basalvaluesbetweenthe groupswere assessed by paired t test with the level of significanceadjusted for the number of comparisonsaccording to Bonferroni. Probability levels lessthan 0.05 were consideredsignificant. RESULTS Characteristics

of patients. The presence of severe left ventricular dysfunction was reconfirmed by results of gated radionuclide left ventriculography, echocardiography, or both within the 2 weeks preceding the study. A resting ejection fraction of 20% or less (16 + 4% [mean f SEMI) was documented in all patients. Results of right-sided catheterization of the heart, performed as part of a routine pretransplant evaluation 2 to 15 days before the physiologic studies, showed a cardiac index of 2.3 + 0.2 L/min/ m2 and a pulmonary capillary wedge pressure of 21 +- 4.5 mm Hg. Hemodynamic responses to LBNP. Table I shows the baseline values and the responses of heart rate, blood pressure (systolic, diastolic, and mean), forearm blood flow, and FVR to graded LBNP. Baseline heart rate and FVR were higher (p < 0.05) and systolic blood pressure and forearm blood flow were lower (p < 0.05) in patients with heart failure compared to normal control subjects. In both groups there were no significant changes in heart rate and blood pressure

Volume

118

Number

1

Cardiopulmonary

Table II. Pulmonary artery diastolic responses (mean 2 SEM) to LNBP and patients with CHF

180

pressure (mm Hg) in normal subjects

r

Subjects CHF (n = 11) Normal (n = 4)

27.8 +- 2

-10

mm Hg

24.5 i

-20

mm Hg

-40

Normals(n

mm Hg

1.3*

23.2 i 1*

20 i

2*

11 I 1.5*

10.5 t 2*

7.3 * 2*

*p < 0.05 vs baseline. tp < 0.001 vs CHF.

at low levels of suction (-10 and -20 mm Hg). However, heart rate increased and systolic blood pressure decreased at high levels of suction (-40 mm Hg) only in normal subjects. Table I and Fig. 1 show the changes in FVR (absolute change) and the percentage of change in FVR (relative change), respectively, in both patients and control subjects. In normal subjects graded LBNP (-10, -20, and -40 mm Hg) produced significant increases in FVR (Table I) at all three levels of suction. In contrast the vasoconstrictor responses observed in patients with heart failure were nearly aboli.shed. Eight of 29 patients had paradoxic vasodilation with a decrease in FVR (-5 f 2, -6 + 3, and -8 * 3 units) during cardiopulmonary baroreceptor unloading with LBNP at -10, -20, and -40 mm Hg. Inasmuch as baseline FVR was significantly higher in patients with heart failure, we normalized the FVR data (as a percentage of change from baseline), which is shown in Fig. 1. The relative responses (10 & 4 % vs 70 * 12%,17 -t 6% vs106 f 21%,and24 & 9% vs 152 + 28%) respectively, at -10, -20, and -40 mm Hg of LBNP) also remained markedly attenuated in patients with CHF. The absence or marked attenuation of vasoconstrictor responses (defined as 10% or more reduction in FVR) in patients with CHF was observed despite a significant (p < 0.01) decrease in cardiac filling pressure (pulmonary artery diastolic pressure) from the baseline value of 28 ? 2 mm Hg to 24.5 f 1.3, 23 -+ 1, and 20 + 2.0 mm Hg during LBNP of -10, -20, and -40 mm Hg, respectively (Table II). These forearm vascular responses in patients with CHF were noted without significant change in mean arterial pressure or heart rate at any level of suction. Effects of LBNP on left atrial dimension and calculated left ventricular volume. In six normal subjects and six

patients with heart failure, we obtained two-dimensional echocardiographic estimates of left ventricular end-diastolic volumes and M-mode-derived left atria1 end-systolic size at baseline and during LBNP.

P

=r

s 2 g

80

% 2

40

a

20

0 0 0

81

=ll)

T /

T fiJ 100

14 k 2t

in CHF

Heart Failure Patients (n =29) *p
LBNP Baseline

e 4

barorefiexes

/

0 000 !

,/

60

F t

0 1 0

t

I

I

-10

-20

-40

LBNP (mmHg) Fig. 1. Relative (percentage change) FVR responses (mean k SEM) of normal subjects and patients with heart failure to LBNP at -10, -20, and -40 mm Hg.

These results are summarized in Fig. 2. Left atria1 dimension decreased significantly in both normal subjects and patients with heart failure, although atria1 size was significantly higher in patients with heart failure during both the control period and LBNP. However, in patients with heart failure left ventricular end-diastolic volumes did not change with LBNP (-20 and -40 mm Hg), whereas a significant decrease in volumes was observed in normal subjects. Hormonal responses to LBNP. Fig. 3 shows the plasma norepinephrine and renin (plasma renin activity) values at baseline and their responses to LBNP in 12 patients with CHF and 11 normal subjects. The baseline values for plasma renin activity (8.3 & 1.7 vs 1.2 + 0.2 ng/ml/hr) and plasma norepinephrine (605 + 81 vs 155 + 8 pg/ml) were significantly higher (p < 0.001) in patients with CHF. Plasma norepinephrine levels increased significantly in response to LBNP in normal subjects but not in patients with CHF. The norepinephrine levels in patients with CHF tended to decrease at LBNP of -40 mm Hg. Plasma renin activity increased during LBNP at -40 mm Hg in normal subjects but did not change in the patients with heart failure. Responses to cold pressor test. Fifteen of 29 patients with heart failure and 9 of 11 normal subjects underwent cold pressor stimulation to assess the vasoconstrictor responses to another stimulus. Fig. 4 summarizes the vasoconstrictor responses (absolute

82

Mohanty

et al.

American

C

-20

-40

C

LBNP

ii

-20

July 1989 Heart Journal

-40 LEINP

Fig. 2. Responseof left atria1 dimension and left ventricular end-diastolic volume in patients with CHF

and normal subjects. * = p < 0.05 vs control subjects;t = p < 0.01 vs patients with CHF. B j\,

*p
Narmals(n=lli

(HR="ryy""

Patlent

tp
Normal

VI Heart Falure

05, "I oarei,ne

A

B

LBNP (mmHg)

LBNP(mmHg)

Fig. 3. Plasmanorepinephrine and renin responses(mean F SEM) of’normal subjects heart failure to LBNP at -10, -20, and -40 mm Hg.

change and percentage change in FVR) to cold pres-

sor stimuli in both normal subjects and patients with CHF. In patients with heart failure the cold pressor stimuli caused significant increases in FVR and were no different from those observed in normal subjects, although the normalized changes tended to be smaller than in the group with heart failure. It should be noted that there were striking differences in the mechanisms for the increases in calculated vascular resistance in the two groups. Normal subjects had large decreases in forearm blood flow (-1.3 * 0.4 ml/min/lOO cc forearm volume) and modest increases in mean arterial pressure (10 -t 5 mm Hg). In contrast, patients with heart failure had smaller reductions in flow (-0.7 rt 0.1 ml/min/lOO cc forearm volume) and larger increases in mean arterial pressure (14 & 5 mm Hg). These differences in responses be-

and patients with

tween normal subjects and patients with CHF to the cold pressure test were statistically significant (p < 0.05). DISCUSSION

The major finding of our study is that the cardiopulmonary baroreflex control of FVR, plasma norepinephrine, and plasma renin activity is impaired in patients with severe, chronic, stable CHF. These abnormalities in reflex control may be attributable to abnormalities in the afferent limb, central nervous system, or neuroeffector mechanisms. In the paragraphs that follow we will discuss our data in relation to this concept. Ours is not the first study to show impaired vasoconstrictor responses to LBNP in patients with CHF. Ferguson et a1.i4 reported previously that patients

Volume

118

Number

1

Cardiopulmonary

NOWIld (n = 9)

Normal (n = 9)

baroreflexes

in CHF

83

Heart Failure (n =12)

Fig. 4. Absolute (AFVR) and relative (AFVR%) forearm vascular resistanceresponses(mean ? SEM) of

normal subjectsand patients with heart failure to cold pressortest. There were no significant differences in calculated FVR responses(both absolute and relative) between groups.

with CHF had impaired forearm vasoconstrictor responses and suggested that there was a selective impairment of cardiopulmonary baroreflexes in CHF. Levine et a1.3 and Kasis” demonstrated impaired vasoconstriction or frank vasodilatation of forearm vessels during head-up tilt in patients with CHF. This is in striking contrast to the vasoconstriction seen during tilt in normal subjects. Ferguson et a1.14 did not report changes in plasma norepinephrine levels, but Levine et al.3 observed impaired norepinephrine responses to head-up tilt in their subjects. Thus our data are in agreement with results of prior studies that indicated impairment of cardiopulmonary baroreflex control of FVR. Ours is the first study to demonstrate impaired responses of plasma norepinephrine and renin activity to LBNP in patients with CHF. Taken together these three indices (vascular resistance, plasma norepinephrine, and plasma renin) indicate impairment of the cardiopulmonary baroreflexes in CHF. As noted in the results, the changes in calculated FVR that were observed in response to cold pressor testing in the patients with heart failure were no different from those of normal subjects. The fact that we were able to observe reductions in forearm blood flow during cold pressor testing indicates that the forearm can respond to cold pressor sympathetic activation with reductions in flow and increases in vascular resistance. Despite the ability of the forearm circulation to respond to sympathetic activation, we detected very little if any response of the forearm to LBNP. The failure of the forearm to respond to LBNP along with the failure to increase plasma norepinephrine and plasma renin activity is

consistent with the view that there is an abnormality in the afferent limb of the central nervous system, which also contributes to impaired cardiopulmonary baroreflexes in CHF. A question could be raised as to the adequacy of the change in the stimulus to the cardiopulmonary baroreceptors that was induced by LBNP. For this reason we measured changes in filling pressure (pulmonary artery diastolic pressure) and cardiac dimensions (echocardiography). Our data show that there are reductions in cardiac filling pressure during LBNP that are associated with large decreases in left atrial dimension but not in calculated left ventricular volumes. Similar data have not been published previously. These echocardiographic and pressure measurement data may provide an important clue as to the mechanism for the impairment in cardiopulmonary baroreflexes in CHF. We reported recently that patients who undergo cardiac transplantation have impaired forearm vasoconstrictor and plasma norepinephrine responses to LBNP.15 The technique of cardiac transplantation that is used results mainly in ventricular rather than total cardiac deafferentation, since the regions of the atria that contain most of the atrial receptors (the vein-atria1 junctions) remain in situ and innervated. Thus the impaired responses observed in transplant patients are the result of ventricular deafferentation. Patients with CHF do not change their calculated ventricular volume during LBNP. Thus the degree of stretch of the ventricular endings was changed little if any during LBNP. If, as we have suggested previously, the ventricular receptors are the most important ones in mediating the responses to LBNP, then it is not surprising that

84

Mohanty et al.

patients with CHF would not vasoconstrict and raise their plasma norepinephrine levels, since the stimulus to the ventricular receptors appears to change little during LBNP in patients with CHF. We noted frank forearm vasodilation in eight patients with CHF during LBNP. All of these patients were in New York Heart Association functional class IV, and this subgroup of patients had the highest cardiac filling pressures (31 ? 3 mm Hg) in the entire study. Ferguson et a1.14 and Levine et a1.3 also observed vasodilation during LBNP and head-up tilt, respectively. The mechanism for this paradoxic response is not clear and cannot be determined from our data. All eight subjects who had vasodilation also had decreased plasma norepinephrine levels during LBNP, thus suggesting a reduction in sympathetic outflow to the heart and peripheral circulation. This could be the result of either withdrawal of an excitatory reflex or activation of an inhibitory reflex. Abboud et a1.23 have speculated that decreases in cardiac filling pressure and chamber dimension would result in a paradoxic activation of vagal afferents in patients with heart failure. On the other hand, stimulation of pulmonary c fibers could result in sympathoexcitation, and reduction in pulmonary congestion may lead to a withdrawal of this excitatory response.24 Further studies are needed to resolve this issue. Our observation of paradoxic vasodilation in sicker (New York Heart Association class IV) patients with much higher (31 + 3 mm Hg) pulmonary capillary wedge pressures are similar to those of Ferguson et a1.,14 who noted frank vasodilation with LBNP in 5 of 11 patients with heart failure in their study. Similar vasodilatory response@ and decreased plasma norepinephrine responses8 to orthostatic tilt have been noted in some patients with heart failure who had higher cardiac filling pressures. This, together with attenuated vasoconstrictor responses, suggests that there may be a spectrum of altered reflex response varying from blunted vasoconstriction to overt paradoxic vasodilation when patients with heart failure are subjected to LBNP or head-up tilt. It is likely that this spectrum of responses is dependent on the severity and/or degree of compensation of CHF. In conclusion, our data provide evidence that patients with CHF have impaired cardiopulmonary baroreflex control of FVR and plasma norepinephrine and renin activity. We have interpreted our data to indicate that there are abnormalities of both neuroeffector mechanisms and the afferent limb or in the central nervous component of the cardiopulmonary baroreflex in these patients.

American

July 1989 Heart Journal

We thank Carolyn McNamara. RN, and Ketan Kapadia for technical assistance. Debra Butler for typing this manuscript, and Deborah Ravin of the Veterans Administration Medical (‘enter Medical Media Services for the illustrations. REFERENCES

1. Ahhoud FM. Integration of reflex responses in the control of blood pressure and vascular resistance. Am ,J Cardiol 1979; 44:903-11. 9-. Cody RJ, Franklin KW, Kluger .J, Laragh JH. Mechanisms governing the postural responses and haroreceptor ahnormalities in chronic congestive heart failure: effect of acute and long term converting enzyme inhibition. Circulation 1982;66:13542. 3. Levine ET, Francis GS, Goldsmith SR, Cohn JR. The neurohumoral and hemodynamic responses to orthostatic tilt in patients with congestive heart failure. Circulation 1983;67: 1070-5. 4. Donald DE, Shepherd J. Cardiac receptors: normal and disturbed function. Am J Cardiol 1979;44:873-9. 5<. Kasis E, Amtorp 0. Cardiovascular and neurohumoral postural responses and baroreceptor abnormalities during a course of adjunctive vasodilator therapy with felodipine for congestive heart failure. Circulation 1987;75:1204-13. responses to orthostatic tilt in pa6. Kasis E. Cardiovascular tients with severe congestive heart failure. Cardiovasc Res 1987;21:362-8. 7. Cohn JN, Levine TB, Francis GS, Goldsmith SR. Neurohumoral control mechanisms in congestive heart failure. AM HEART J 1981;102:509-14. SR, Simon AB, Cohn JN. 8. Levine BT, Francis GS, Goldsmith Activity of sympathetic nervous system and renin-angiotensin system assessed by plasma hormone levels and their relation to hemodynamic abnormalities in congestive heart failure. Am J Cardiol 1982;49:1659-66. CB. Vatner SF, Eckherg DL, Braunwald E. Alter9. Higgins ations in the haroreceptor reflex in conscious dogs with heart failure. J Clin Invest 1972;51:715-24. LO. Vatner SF, Boettcher DH, Heindricks GR, McRitchie RJ. Reduced baroreflex sensitivity with volume loading in conscious dogs. Circ Res 1975;37:236-42. 11. White CW. Abnormalities in haroreflex control of heart rate in canine heart failure. Am J Physiol 1981;240:H793-9. 12. Riegger AJG, Leihau G. The renin-angiotensin-aldosterone system, anti-diuretic hormone and sympathetic nerve activity in an experimental model of congestive heart failure. Clin Sci 1982;62:645-9. 13. Zucker IH, Gorman AJ, Cornish KG, Lang M. Impaired atria1 receptors modulation of renal nerve activity in dogs with chronic volume overload. Cardiovasc Res 1985;19:411-8. 14. Ferguson DW, Ahboud FM, Mark AL. Selective impairment of baroreflex mediated vasoconstrictor responses in patients with ventricular dysfunction. Circulation 1984;69:451-60. 15. Mohanty PK, Thames MD, Arrowood JA, Sowers JR, McNamara C, Szentpetery S. Impairment of cardiopulmonary baroreflex after cardiac transplantation in humans. Circulation 1987;75:914-21. 16. Greenfield ADM, Whitney RT, Mowhray J. Methods of investigation of peripheral blood flow. Br Med Bull 1963: 19:101-9. 17. Rogers EW, Feigenbaum H, Weyman AE. Echocardiography for quantitation of cardiac chamber. In: Progress in cardiology. Yu PN, Goodwin JF, eds. Philadelphia: Lea & Febiger, Puhlishers, 1979:8:1-28. 18. Schiller NP, Acquatella H, Posts TA, Drew D, Goerve J, Ringert H, Silverman NH, Brundage B, Botvinick EH, Boswell R, Carlsson E, Parmley WW. Left ventricular volume from hiolane two dimensional echocardiogranhv. 1979; I _ - Circulation 66547-52. 19. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results

Volume Number

1 IS 1

of survey of echocardiographic measurements. Circulation 1978;58:1072-6. 20. Mohanty PK, Sowers JR, McNamara C, Thames MD. Reflex effects of prolonged cardiopulmonary baroreceptor unloading in humans. Am J Physiol 1988;254:R320-4. 21. Keller R, Ove A, Medford I, Adamas RN. Liquid chromatographic analysis of catecholamines. Life Sci 1976;19:995-1004. 22. Sowers JR, Gloub MS, Eggena PH, Catagnia RA. Influence of sodium hemostasis on dopamine modulation of aldosterone

Cardiopulmonary

baroreflexes in CHF

85

renin and prolactin in man. J Clin Endocrinol Metab 1982;54:121-6. 23. Abboud FM, Thames MD, Mark AL. Role of cardiac afferent nerves in regulation of circulation during coronary occlusion and heart failure. In: Abbound FM, Fozzard HA, Gilmore JP, Reis DJ, eds. Baltimore: The Williams & Wilkins Company, 1981:65-86. 24. Paintal AS. Vagal sensory receptors and their reflex effects. Physiol Rev 1973;53:159-227.