Cardiovascular regulation in hypertension Research Paper
Clinical Autonomic Research 2, 243-255 (1992).
PHYSIOLOGICALconsequences of altered peak left ventricular diastolic filling rate in hypertension have not yet been fully assessed. The hypothesis that altered left ventricular diastolic filling rate interferes with inhibitory cardiopulmonary reflexes was tested. Normalized peak left ventricular diastolic filling rate was calculated from radionuclide ventriculography. Haemodynamic changes during lower body negative pressure (--5 to --40mmHg) in nine hypertensive patients with slow normalized left ventricular filling rate (Group A) were compared with 16 hypertensive patients with normal normalized peak left ventricular diastolic filling rate and ten normal volunteers of the same age group. Baseline total peripheral resistance was higher in essential hypertension compared to normals but did not differ significantly between the two hypertensive groups. For data analysis, the levels of lower body negative pressure were grouped as low levels of - 5 to -10, and --15 to --20 mmHg, an intermediate level of --25 mmHg, and high levels of --30 to --40 mmFIg; the change in total peripheral resistance (from baseline) was less prominent in Group A compared to Group B and to normals (--1.4 + 1.7 [SE], --0.06 + 1.4, 1.1 + 1.2 and 4.5 + 2 u'M 2 in Group A at the four consecutive levels of lower body negative pressure vs. 0.9 + 0.7, 3.8 + 0.9, 7.2 + 1.6, and 8.2 + 1.4 in Group B, and 2.0 + 0.7, 3.3 + 0.8, 4.9 _+ 0.8, and 5.6_ 1.0 in normals). The reductions in central venous pressure and in pulmonary wedge pressure were not significantly different among the three groups at the different levels of lower body negative pressure, but the reduction in cardiac output was smaller in patients with reduced dv/dt ratio than in the other two groups. The responses to the cold pressor test were similar in all subjects. We conclude that patients with essential hypertension and diastolic dysfunction have impaired total peripheral resistance responses to lower body negative pressure. This abnormality may reflect an alteration in cardiac baroreflexes secondary to left ventricular diastolic dysfunction, an influence of baseline sympathetic activity on the observed vascular responsiveness to lower body negative pressure, or primary differences among groups in the changes in cardiac output induced by similar levels of lower body negative pressure.
Reduced vascular excitatory responses to cardiopulmonary unloading in hypertensive patients with left ventricular diastolic dysfunction M. Aziz M a d k o u r , MD 1, Luis Bedoya, MD 2 and Fetnat M. Fouad-Tarazi, MD 1"CA 1Department of Heart and Hypertension Research, and 2Department of Hypertension/Nephrology, The Cleveland Clinic Foundation, FF1-02b, 9500 Euclid Avenue, Cleveland, OH 44195-5069, USA. CaCorresponding A u t h o r
Key words: Hypertension, Diastole, Haemodynamics, Lower body negative pressure, Systemic resistance
Introduction Left ventricular filling rate was shown to be slow in hypertensive patients compared to age matched normals at a stage when resting systolic function was still normal; >4 however, the significance o f this left ventricular functional alteration has not yet been fully assessed. O n the other hand, cardiopulmonary reflexes have been reported to be impaired in established hypertension. This impairment of cardiopulmonary reflexes was demonstrated by direct recording of afferent neural activity in spontaneously hyptertensive rats with left ventricular hypertrophy, s In man, the peripheral
© Rapid Communications of Oxford Ltd.
arterial vasoconstrictive response to head-up tilt was previously found to be blunted in patients with severe established hypertension. 6 The possible pathophysiologic link between blunting o f cardiopulmonary reflexes and impairment o f left ventricular diastolic function in hypertension has not yet been reported. We hypothesized that the altered diastolic filling rate o f the left ventricle in hypertension interferes with the sensitivity o f the inhibitory cardiopulmonary reflexes. This study was designed to assess the haemodynamic changes in response to unloading the cardiopulmonary receptors by application o f lower body negative pressure (LBNP) in hypertensive
Clinical Autonomic Research-vol 2" 1992
243
M. A . Madkour, L. Bedoya and F. M. Fouad-Tarazi
patients with and without abnormalities in left ventricular diastolic filling characteristics. Material
and Methods
Patient population: The study included 25 patients with documented hypertension and ten normal volunteers. The hypertensive group included twelve women and 13 men with an age range of 20 to 61 years (mean 43.7-t-2.3 [SE]); body surface area averaged 1.97-t-0.05 in hypertensives and 1.95 -t- 0.05 in normals; those who were previously treated had their medications discontinued prior to the study. This drug-free time interval depended on the biological half-life of the individual medications; a minimum of 10 drug-free days was allowed in this study. These medications included calcium entry blockers in three patients, converting enzyme inhibitors in two, beta blockers in four, alpha methyldopa in one, and diuretics in five. At the time of the study, blood pressure was in the hypertensive range except in four patients where supine resting study blood pressure ranged from 123/88 to 136/89. They were considered to be labile hypertensives. Two fell in the group of hypertensives with abnormal left ventricular diastolic filling and two in the group of hypertensives with normal left ventricular diastolic filling. The normal volunteers included three women and seven men with an age range of 28 to 55 years (mean 39.5 _ 2.5). None had clinical or electrocardiographic evidence of coronary artery disease. The protocol of the study was approved by the Institutional Review Board of The Cleveland Clinic Foundation. All patients gave their written consent to participate in the study.
Protocol of the study: The protocol of the haemodynamic study is illustrated in Fig. 1. All studies were performed in the morning after an overnight fast. A No. 7.5 F triple lumen thermodilution Swan-Ganz catheter was introduced percutaneously (Seldinger technique under local anaesthesia with 1% lidocaine) via an antecubital vein. The tip of the Swan-Ganz catheter was positioned under fluoroscopy in the right or left pulmonary arterial branch and wedged whenever needed. A No. 5F polyethylene arterial catheter was introduced percutaneously (Seldinger technique under local anaesthesia) via the brachial artery; the tip of the arterial line was securely placed in the ascending aorta under fluoroscopy. Four patients refused the arterial catheterization; blood pressure was obtained in these patients by sphygmomanometer cuff method every minute during procedures and for a 10-min control period before every procedure. Cuff blood pressure in these patients was obtained by an experienced nurse well trained in precise haemodynamic monitoring. After placement of catheters, the subject's body below the iliac crest was introduced in the LBNP chamber, then the subject was positioned under a gamma camera (large field of view parallel hole collimator); the camera head was tilted across the chest to a left anterior oblique position of 45 ° with a 5 ° caudad angle so as to visualize the heart and large vessels clearly. After 30 min supine rest, blood volume (12sIRISA = radio-iodinated human serum albumin) was determined, then control supine haemodynamic measurements were recorded including: systemic arterial blood pressure, heart rate, pulmonary
PROTOCOL 4
HR; Right atrial, Pulmonary wedge & arterial pressures
BASm.E'+4 I CATHETERSI ~, and REST •
I •
I •
~
/-
L BN P 1
P
REST
+
~
= L__.__
ff 1 | ! !
IV PYRO PO4
BV CO EF (TD) (RN) P.Cats
CO (TD)
CO (TD)
CO (TD)
CO (TO)
! I I !
CO EF COLD PRESSOR (TD) VALSALVA P.Cats PE & AN TEST
FIG. 1. Protocol of the study showing time sequence of haemodynamic tests. AN = amylnitrite test; BV= blood volume; CO =cardiac output, EF=ejection fraction; HR = heart rate; IV= intravenous cannula; LBNP = lower body negative pressure; P. Cats = blood samples for plasma catecholamines (norepinephrine and epinephrine); PE = phenylalanine test; PyroPO4 = pyrophosphate; RN = injection of radionuclide (99mTechnetium); TD = thermodilution method. 244
Clinical Autonomic Research. vol 2.1992
Cardiovascular regulation in hypertension arterial pressure, pulmonary wedge pressure, right atrial pressure, and cardiac output (Edwards 9520A thermodilution computer). Systemic intra-arterial pressure and right heart pressures were continuously measured with micron pressure transducers (Micron Instruments, MP-15D) and recorded simultaneously with heart rate (ECG) on a direct writing Gould physiologic recorder (Gould Brush 200). Calibration of the instrument for pressure measurement was done with a mercury manometer before starting each study, and the stability of the pressure zero levels was checked frequently during the procedures. At baseline, cardiac output was determined in triplicate to quintuplet with the average value used for calculation. Also, radionuclide multigated ejection fraction was obtained. Following these measurements, eight levels of graded lower body negative suction were applied, from --5 to --40 mmHg to simulate orthostatic stress. Each level was maintained for 2 min. Blood pressure and heart rate, as well as pulsatile pulmonary artery pressure and right atrial pressure were monitored continuously. Measurements were made after 30 s of stabilization at each level. The remaining time allowed for only one set of measurements of mean right atrial pressure, thermodilution cardiac output, and mean pulmonary wedge pressure via the Swan-Ganz catheter. Blood samples for plasma catecholamine determinations were collected from a venous angiocatheter pre-positioned in a forearm vein at baseline and at the maximal level of LBNP. The LBNP was stopped before the predicted maximum level only if the subject became symptomatic (such as with dizziness, lightheadedness, or nausea) or if signs of impending vasovagal reaction or symptomatic orthostatic hypotension appeared (such as slowing of heart rate and/or decrease of both systolic and diastolic blood pressures). At the end of the study, the cold pressor test (for vascular reactivity), and the baroreceptor sensitivity test (phenylephrine and/or amylnitrite test) were performed. All subjects but one had an echocardiogram done prior to the haemodynamic evaluation to assess left ventricular dimensions and to rule out valvular or pericardial abnormalities.
Methods Lower body negative pressure: The LBNP chamber used in our laboratory was built in the Bioengineering Department of the University of Iowa; periodic testing has demonstrated its accuracy and reproducibility. It is safe and allows the study of response to graded venous pooling induced by suction. In order to create LBNP during the study, the
patient's body below the iliac crest was placed in the cylindrical chamber which was sealed tightly with plastic sheets (aluminized Mylar reinforced aerospace material) at its upper open end; the feet of the subject rested on a padded board in the lower end of the cylinder; this board provided support during the suction exposures.
Cardiac output determination: Measurements of cardiac output were obtained by thermodilution method. 7's Cardiac output was determined from injections into the right atrium of 10 ml of 5% glucose solution at 0°C. Manual injection over less than 3 s was employed. Incomplete or delayed injections were discarded. Edwards Laboratory Cardiac Output Computer (Model 9520A) provided digital display of cardiac output values. Correlation of this method with simultaneous measurements by .indocyanine green dye dilution method was previously reported from our laboratory to be 0.87 and 0.92 during two separate series (p < 0.001 for both). The coefficient of variation of the repeated cardiac output measurement at baseline in our study (pre-LBNP) was 4.6 ___3 (SD)%. Blood volume determination." Plasma volume was measured using 10/~cu of 125I-RISA injected i.v., and 10-min equilibration period. 9 Total blood volume was calculated from plasma volume and simultaneously determined venous haematocrit; values were calculated as per cent of normal for corresponding sex to allow averaging of data obtained from men and women. Normal values for our laboratory averaged 29.4 + 0.8 (SD) ml/cm height for men and 23.7 + 0.5 ml/cm height for women. Determination of cardiopulmonary blood volume: Cardiopulmonary blood volume was calculated as previously described in detail 1° from radionuclide 99mTechnetium-red blood cell derived cardiac output and pulmonary mean transit time. For this purpose, the first pass of the radioisotope dilution curve was analysed in duplicate. The ratio of cardiopulmonary volume to total blood volume (RISA, calculated immediately before the injection of 99mTechnetium bolus) was used as an indirect index of venous tone and preload volume. 1° Radionuclide left ventricular volume curve and derived left ventricular function indices: Left ventricular volume curves were obtained from gated blood pool studies 2'11'12 at equilibrium after i.v. injection of 99mTechnetium (12 mci), and in vivo red blood cell tagging. Analysis of left ventricular volume curves, after background subtraction, allowed determination of left ventricular ejection fraction and the peak rates of left ventricular ejection and filling. 2J1J2 Ejection Clinical Autonomic Research. vol 2-1992
245
M. A. Madkour, L. Bedoya and F. M. Fouad-Tarazi fraction was calculated as: (left ventricular end-diastolic c o u n t s - left ventricular end-systolic counts)/left ventricular enddiastolic counts The left ventricular peak rates of ejection and filling were obtained from a derivative curve constructed from the time-activity curve using standard Analytical Development Associates Corporation software adapted from Medical Data Systems A 2 software. 11 Since this curve represents the radioactivity counts as a mathematical function of time, the derivatives were expressed in counts per second. The negative derivatives corresponded to the decrease in counts or ejection phase (--dv/dt) and the positive derivatives corresponded to the increase in counts or filling phase ( + dv/dt). In this analysis, we specified the largest negative and maximum positive derivatives. These derivatives were scaled by the maximum absolute counts equivalent to the end diastolic c o u n t s 2'12-14" t o compensate for variation in the injected dose and efficiency of the acquisition system. In this way, it was possible to compare data from different patients. The units of these new variables were therefore expressed in units of fractional change per second or Hertz and were proportional to the change in volume per unit time (dr~dr). Moreover, the close and complex interaction of ventricular relaxation and early filling with cardiac inotropy, heart rate and end-diastolic diameter have been taken into account by calculating the ratio of the left ventricular filling rate (+dv/dt) to left ventricular ejection rate ( - d v / d t ) . 13'14 In this study, values for this ratio + d v / d t / - d v / d t have ranged from 0.86 to 1.49 in the normal volunteers. A methodological aspect in our study should be discussed at this point regarding the importance of framing rate during acquisition of left ventricular gated blood pool scanning. The left ventricular function curves obtained at 32 frames per R - R interval were noisy in many patients reflecting the smaller number of counts per frame; therefore, we acquired data at a rate of 16 frames per R - R interval similar to other previously published studies from our laboratory. 2'12 Left ventricular function indices derived from this acquisition module are in agreement with findings from other laboratories.15-17
Echocardiographic technique." The technique for echocardiographic recording and measurement of relevant indices was done according to the recommendation of the American Society of Echocardiography. 18 Excellent quality echocardiographic tracings were an essential prerequisite for inclusion in the study. The method has been applied both to patients and normal volunteers; only one 246
Clinical Autonomic Research.vol 2.1992
hypertensive patient did not have an echocardiogram done. In all cases, M-mode echocardiography was performed while monitoring the echo-beam direction by two-dimensional view on a TV screen using a Hewlett Packard phased array ultrasonic sector scanner (Model HP 7702 OA). A 3.5 MHz transducer was used. Strip chart recorder was run at a speed of 50 mm/s during routine tracings. The subject was usually examined in a tilted position (left lateral decubitus) in order to obtain both the septum and the posterior wall in the same view. Special attention was taken to avoid misleading angulation of the left ventricle against the echo beam. Measurements of left ventricular dimensions were made in diastole at the onset of the QRS complex and in systole as determined by the shortest left ventricular diameter. Measurements were obtained from an average of ten consecutive cardiac cycles using a programmable calculator (Numonics Clinical Graphics Analyzer, model 1239). Left ventricular mass was calculated using the method of Devereux et al. 19 Reported upper limits of normal for Devereux correction 19 are 110 g/m 2 for women and 134 g/m 2 for men. Left ventricular fractional shortening was calculated as the difference between left ventricular end-diastolic diameter and end-systolic diameter divided by left ventricular end-diastolic diameter.
Coldpressor test: The patient was asked to immerse the hand (contralateral to catheter site), up to the wrist, in ice-cold water for 1 min. The magnitude of the heart rate and blood pressure responses to somatic pain were examined. These responses do not depend on afferent pathways involved in LBNP; this test is used, therefore, to determine whether abnormal responses to LBNP were specific or were a manifestation of a generalized alteration in responsiveness to reflex stimuli. 2°'21 Arterial baroreceptor sensitivity test: According to the method of Smyth et al., 22 a bolus of phenylephrine (25, 50, or 75#g) was injected intravenously starting with the lower dose followed by the next higher to increase systolic blood pressure by at least 2 0 m m H g . In our study, the test could not be performed in five hypertensive patients because of high levels of baseline blood pressure; however, all of these five patients had amylnitrite inhalation test for baroreceptor sensitivity as described below. The sensitivity of the baroreceptors was quantitated by correlating the individual intra-arterial systolic blood pressure levels with the subsequent R - R interval obtained from a simultaneous electrocardiographic tracing recorded at a paper speed of 100 mm/s. Amylnitrite inhalation test was performed following the phenylephrine test whenever the two
Cardiovascular regulation in hypertension Table 1. Baseline haemodynamics in hypertensive and normal subjects All hypertensives n Heart rate (bpm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Cardiac index (I/min/m 2) Total peripheral resistance (u'm 2) Right atrial pressure (mmHg) Pulmonary wedge pressure (mmHg) Cardiopulmonary volume/total blood volume (%) Left ventricular ejection fraction (%) +dv/dt (Hz) -dv/dt (Hz) +dv/dt/-dv/dt (Hz)
67 161 96 3.1 39 9 14 11 59 2.4 2.5 0.96
25 -+ 2 ± 5 _+ 2 _+ 0.1 _+ 1.9 _+ 0.8 _+ 0.8 _+ 0.5 -+ 1.9 _+ 0.1 i 0.1 -t- 0.03
Normals
66 127 80 3.6 27 8 12 13 60 2.6 2.4 1.08
10 -+ 2 -+ 2 -t- 0.6 __+0.2 -+ 1.4 _+ 0.9 _+ 1.1 _+0.9 -+ 1.4 _+ 0.1 -I- 0.1 _+ 0.05
p -NS 0.0001 0.0001 0.012 0.0001 NS NS NS NS NS NS NS
Mean -t- SE.
dv/dt ratio = left ventricular filling characteristics expressed as the ratio of left ventricular peak filling rate (positive dv/dt) to peak ejection rate (negative dv/dt).
tests were done in sequence. While intra-arterial blood pressure was recorded continuously at a speed of 100 ram/s, the subject was asked to inhale deeply twice as the ampoule of amylnitrite was crushed close to the nostrils. The change in systolic blood pressure was examined immediately; if the drop in systolic blood pressure was less than 20 mmHg, the test was repeated after 5 min of rest. During both phenylephrine test and amylnitrite test, the subjects were instructed to avoid straining. Plasma catecholamines: Plasma norepinephrine and epinephrine were determined by radioenzymatic assay. 23 Normal supine values in our laboratory average 260-t-120 (SD) pg/ml for total plasma catecholamines, 218 -t:_92 for plasma noradrenaline and 42 _ 18 for plasma epinephrine. Statistical analysis: Standard techniques 24 were applied using the Prophet computer system. Analysis of variance was used for comparison. Statistical significance was defined by a p value of
Clinical Autonomic Research 2, 243-255 (1992).
PHYSIOLOGICALconsequences of altered peak left ventricular diastolic filling rate in hypertension have not yet been fully assessed. The hypothesis that altered left ventricular diastolic filling rate interferes with inhibitory cardiopulmonary reflexes was tested. Normalized peak left ventricular diastolic filling rate was calculated from radionuclide ventriculography. Haemodynamic changes during lower body negative pressure (--5 to --40mmHg) in nine hypertensive patients with slow normalized left ventricular filling rate (Group A) were compared with 16 hypertensive patients with normal normalized peak left ventricular diastolic filling rate and ten normal volunteers of the same age group. Baseline total peripheral resistance was higher in essential hypertension compared to normals but did not differ significantly between the two hypertensive groups. For data analysis, the levels of lower body negative pressure were grouped as low levels of - 5 to -10, and --15 to --20 mmHg, an intermediate level of --25 mmHg, and high levels of --30 to --40 mmFIg; the change in total peripheral resistance (from baseline) was less prominent in Group A compared to Group B and to normals (--1.4 + 1.7 [SE], --0.06 + 1.4, 1.1 + 1.2 and 4.5 + 2 u'M 2 in Group A at the four consecutive levels of lower body negative pressure vs. 0.9 + 0.7, 3.8 + 0.9, 7.2 + 1.6, and 8.2 + 1.4 in Group B, and 2.0 + 0.7, 3.3 + 0.8, 4.9 _+ 0.8, and 5.6_ 1.0 in normals). The reductions in central venous pressure and in pulmonary wedge pressure were not significantly different among the three groups at the different levels of lower body negative pressure, but the reduction in cardiac output was smaller in patients with reduced dv/dt ratio than in the other two groups. The responses to the cold pressor test were similar in all subjects. We conclude that patients with essential hypertension and diastolic dysfunction have impaired total peripheral resistance responses to lower body negative pressure. This abnormality may reflect an alteration in cardiac baroreflexes secondary to left ventricular diastolic dysfunction, an influence of baseline sympathetic activity on the observed vascular responsiveness to lower body negative pressure, or primary differences among groups in the changes in cardiac output induced by similar levels of lower body negative pressure.
Reduced vascular excitatory responses to cardiopulmonary unloading in hypertensive patients with left ventricular diastolic dysfunction M. Aziz M a d k o u r , MD 1, Luis Bedoya, MD 2 and Fetnat M. Fouad-Tarazi, MD 1"CA 1Department of Heart and Hypertension Research, and 2Department of Hypertension/Nephrology, The Cleveland Clinic Foundation, FF1-02b, 9500 Euclid Avenue, Cleveland, OH 44195-5069, USA. CaCorresponding A u t h o r
Key words: Hypertension, Diastole, Haemodynamics, Lower body negative pressure, Systemic resistance
Introduction Left ventricular filling rate was shown to be slow in hypertensive patients compared to age matched normals at a stage when resting systolic function was still normal; >4 however, the significance o f this left ventricular functional alteration has not yet been fully assessed. O n the other hand, cardiopulmonary reflexes have been reported to be impaired in established hypertension. This impairment of cardiopulmonary reflexes was demonstrated by direct recording of afferent neural activity in spontaneously hyptertensive rats with left ventricular hypertrophy, s In man, the peripheral
© Rapid Communications of Oxford Ltd.
arterial vasoconstrictive response to head-up tilt was previously found to be blunted in patients with severe established hypertension. 6 The possible pathophysiologic link between blunting o f cardiopulmonary reflexes and impairment o f left ventricular diastolic function in hypertension has not yet been reported. We hypothesized that the altered diastolic filling rate o f the left ventricle in hypertension interferes with the sensitivity o f the inhibitory cardiopulmonary reflexes. This study was designed to assess the haemodynamic changes in response to unloading the cardiopulmonary receptors by application o f lower body negative pressure (LBNP) in hypertensive
Clinical Autonomic Research-vol 2" 1992
243
M. A . Madkour, L. Bedoya and F. M. Fouad-Tarazi
patients with and without abnormalities in left ventricular diastolic filling characteristics. Material
and Methods
Patient population: The study included 25 patients with documented hypertension and ten normal volunteers. The hypertensive group included twelve women and 13 men with an age range of 20 to 61 years (mean 43.7-t-2.3 [SE]); body surface area averaged 1.97-t-0.05 in hypertensives and 1.95 -t- 0.05 in normals; those who were previously treated had their medications discontinued prior to the study. This drug-free time interval depended on the biological half-life of the individual medications; a minimum of 10 drug-free days was allowed in this study. These medications included calcium entry blockers in three patients, converting enzyme inhibitors in two, beta blockers in four, alpha methyldopa in one, and diuretics in five. At the time of the study, blood pressure was in the hypertensive range except in four patients where supine resting study blood pressure ranged from 123/88 to 136/89. They were considered to be labile hypertensives. Two fell in the group of hypertensives with abnormal left ventricular diastolic filling and two in the group of hypertensives with normal left ventricular diastolic filling. The normal volunteers included three women and seven men with an age range of 28 to 55 years (mean 39.5 _ 2.5). None had clinical or electrocardiographic evidence of coronary artery disease. The protocol of the study was approved by the Institutional Review Board of The Cleveland Clinic Foundation. All patients gave their written consent to participate in the study.
Protocol of the study: The protocol of the haemodynamic study is illustrated in Fig. 1. All studies were performed in the morning after an overnight fast. A No. 7.5 F triple lumen thermodilution Swan-Ganz catheter was introduced percutaneously (Seldinger technique under local anaesthesia with 1% lidocaine) via an antecubital vein. The tip of the Swan-Ganz catheter was positioned under fluoroscopy in the right or left pulmonary arterial branch and wedged whenever needed. A No. 5F polyethylene arterial catheter was introduced percutaneously (Seldinger technique under local anaesthesia) via the brachial artery; the tip of the arterial line was securely placed in the ascending aorta under fluoroscopy. Four patients refused the arterial catheterization; blood pressure was obtained in these patients by sphygmomanometer cuff method every minute during procedures and for a 10-min control period before every procedure. Cuff blood pressure in these patients was obtained by an experienced nurse well trained in precise haemodynamic monitoring. After placement of catheters, the subject's body below the iliac crest was introduced in the LBNP chamber, then the subject was positioned under a gamma camera (large field of view parallel hole collimator); the camera head was tilted across the chest to a left anterior oblique position of 45 ° with a 5 ° caudad angle so as to visualize the heart and large vessels clearly. After 30 min supine rest, blood volume (12sIRISA = radio-iodinated human serum albumin) was determined, then control supine haemodynamic measurements were recorded including: systemic arterial blood pressure, heart rate, pulmonary
PROTOCOL 4
HR; Right atrial, Pulmonary wedge & arterial pressures
BASm.E'+4 I CATHETERSI ~, and REST •
I •
I •
~
/-
L BN P 1
P
REST
+
~
= L__.__
ff 1 | ! !
IV PYRO PO4
BV CO EF (TD) (RN) P.Cats
CO (TD)
CO (TD)
CO (TD)
CO (TO)
! I I !
CO EF COLD PRESSOR (TD) VALSALVA P.Cats PE & AN TEST
FIG. 1. Protocol of the study showing time sequence of haemodynamic tests. AN = amylnitrite test; BV= blood volume; CO =cardiac output, EF=ejection fraction; HR = heart rate; IV= intravenous cannula; LBNP = lower body negative pressure; P. Cats = blood samples for plasma catecholamines (norepinephrine and epinephrine); PE = phenylalanine test; PyroPO4 = pyrophosphate; RN = injection of radionuclide (99mTechnetium); TD = thermodilution method. 244
Clinical Autonomic Research. vol 2.1992
Cardiovascular regulation in hypertension arterial pressure, pulmonary wedge pressure, right atrial pressure, and cardiac output (Edwards 9520A thermodilution computer). Systemic intra-arterial pressure and right heart pressures were continuously measured with micron pressure transducers (Micron Instruments, MP-15D) and recorded simultaneously with heart rate (ECG) on a direct writing Gould physiologic recorder (Gould Brush 200). Calibration of the instrument for pressure measurement was done with a mercury manometer before starting each study, and the stability of the pressure zero levels was checked frequently during the procedures. At baseline, cardiac output was determined in triplicate to quintuplet with the average value used for calculation. Also, radionuclide multigated ejection fraction was obtained. Following these measurements, eight levels of graded lower body negative suction were applied, from --5 to --40 mmHg to simulate orthostatic stress. Each level was maintained for 2 min. Blood pressure and heart rate, as well as pulsatile pulmonary artery pressure and right atrial pressure were monitored continuously. Measurements were made after 30 s of stabilization at each level. The remaining time allowed for only one set of measurements of mean right atrial pressure, thermodilution cardiac output, and mean pulmonary wedge pressure via the Swan-Ganz catheter. Blood samples for plasma catecholamine determinations were collected from a venous angiocatheter pre-positioned in a forearm vein at baseline and at the maximal level of LBNP. The LBNP was stopped before the predicted maximum level only if the subject became symptomatic (such as with dizziness, lightheadedness, or nausea) or if signs of impending vasovagal reaction or symptomatic orthostatic hypotension appeared (such as slowing of heart rate and/or decrease of both systolic and diastolic blood pressures). At the end of the study, the cold pressor test (for vascular reactivity), and the baroreceptor sensitivity test (phenylephrine and/or amylnitrite test) were performed. All subjects but one had an echocardiogram done prior to the haemodynamic evaluation to assess left ventricular dimensions and to rule out valvular or pericardial abnormalities.
Methods Lower body negative pressure: The LBNP chamber used in our laboratory was built in the Bioengineering Department of the University of Iowa; periodic testing has demonstrated its accuracy and reproducibility. It is safe and allows the study of response to graded venous pooling induced by suction. In order to create LBNP during the study, the
patient's body below the iliac crest was placed in the cylindrical chamber which was sealed tightly with plastic sheets (aluminized Mylar reinforced aerospace material) at its upper open end; the feet of the subject rested on a padded board in the lower end of the cylinder; this board provided support during the suction exposures.
Cardiac output determination: Measurements of cardiac output were obtained by thermodilution method. 7's Cardiac output was determined from injections into the right atrium of 10 ml of 5% glucose solution at 0°C. Manual injection over less than 3 s was employed. Incomplete or delayed injections were discarded. Edwards Laboratory Cardiac Output Computer (Model 9520A) provided digital display of cardiac output values. Correlation of this method with simultaneous measurements by .indocyanine green dye dilution method was previously reported from our laboratory to be 0.87 and 0.92 during two separate series (p < 0.001 for both). The coefficient of variation of the repeated cardiac output measurement at baseline in our study (pre-LBNP) was 4.6 ___3 (SD)%. Blood volume determination." Plasma volume was measured using 10/~cu of 125I-RISA injected i.v., and 10-min equilibration period. 9 Total blood volume was calculated from plasma volume and simultaneously determined venous haematocrit; values were calculated as per cent of normal for corresponding sex to allow averaging of data obtained from men and women. Normal values for our laboratory averaged 29.4 + 0.8 (SD) ml/cm height for men and 23.7 + 0.5 ml/cm height for women. Determination of cardiopulmonary blood volume: Cardiopulmonary blood volume was calculated as previously described in detail 1° from radionuclide 99mTechnetium-red blood cell derived cardiac output and pulmonary mean transit time. For this purpose, the first pass of the radioisotope dilution curve was analysed in duplicate. The ratio of cardiopulmonary volume to total blood volume (RISA, calculated immediately before the injection of 99mTechnetium bolus) was used as an indirect index of venous tone and preload volume. 1° Radionuclide left ventricular volume curve and derived left ventricular function indices: Left ventricular volume curves were obtained from gated blood pool studies 2'11'12 at equilibrium after i.v. injection of 99mTechnetium (12 mci), and in vivo red blood cell tagging. Analysis of left ventricular volume curves, after background subtraction, allowed determination of left ventricular ejection fraction and the peak rates of left ventricular ejection and filling. 2J1J2 Ejection Clinical Autonomic Research. vol 2-1992
245
M. A. Madkour, L. Bedoya and F. M. Fouad-Tarazi fraction was calculated as: (left ventricular end-diastolic c o u n t s - left ventricular end-systolic counts)/left ventricular enddiastolic counts The left ventricular peak rates of ejection and filling were obtained from a derivative curve constructed from the time-activity curve using standard Analytical Development Associates Corporation software adapted from Medical Data Systems A 2 software. 11 Since this curve represents the radioactivity counts as a mathematical function of time, the derivatives were expressed in counts per second. The negative derivatives corresponded to the decrease in counts or ejection phase (--dv/dt) and the positive derivatives corresponded to the increase in counts or filling phase ( + dv/dt). In this analysis, we specified the largest negative and maximum positive derivatives. These derivatives were scaled by the maximum absolute counts equivalent to the end diastolic c o u n t s 2'12-14" t o compensate for variation in the injected dose and efficiency of the acquisition system. In this way, it was possible to compare data from different patients. The units of these new variables were therefore expressed in units of fractional change per second or Hertz and were proportional to the change in volume per unit time (dr~dr). Moreover, the close and complex interaction of ventricular relaxation and early filling with cardiac inotropy, heart rate and end-diastolic diameter have been taken into account by calculating the ratio of the left ventricular filling rate (+dv/dt) to left ventricular ejection rate ( - d v / d t ) . 13'14 In this study, values for this ratio + d v / d t / - d v / d t have ranged from 0.86 to 1.49 in the normal volunteers. A methodological aspect in our study should be discussed at this point regarding the importance of framing rate during acquisition of left ventricular gated blood pool scanning. The left ventricular function curves obtained at 32 frames per R - R interval were noisy in many patients reflecting the smaller number of counts per frame; therefore, we acquired data at a rate of 16 frames per R - R interval similar to other previously published studies from our laboratory. 2'12 Left ventricular function indices derived from this acquisition module are in agreement with findings from other laboratories.15-17
Echocardiographic technique." The technique for echocardiographic recording and measurement of relevant indices was done according to the recommendation of the American Society of Echocardiography. 18 Excellent quality echocardiographic tracings were an essential prerequisite for inclusion in the study. The method has been applied both to patients and normal volunteers; only one 246
Clinical Autonomic Research.vol 2.1992
hypertensive patient did not have an echocardiogram done. In all cases, M-mode echocardiography was performed while monitoring the echo-beam direction by two-dimensional view on a TV screen using a Hewlett Packard phased array ultrasonic sector scanner (Model HP 7702 OA). A 3.5 MHz transducer was used. Strip chart recorder was run at a speed of 50 mm/s during routine tracings. The subject was usually examined in a tilted position (left lateral decubitus) in order to obtain both the septum and the posterior wall in the same view. Special attention was taken to avoid misleading angulation of the left ventricle against the echo beam. Measurements of left ventricular dimensions were made in diastole at the onset of the QRS complex and in systole as determined by the shortest left ventricular diameter. Measurements were obtained from an average of ten consecutive cardiac cycles using a programmable calculator (Numonics Clinical Graphics Analyzer, model 1239). Left ventricular mass was calculated using the method of Devereux et al. 19 Reported upper limits of normal for Devereux correction 19 are 110 g/m 2 for women and 134 g/m 2 for men. Left ventricular fractional shortening was calculated as the difference between left ventricular end-diastolic diameter and end-systolic diameter divided by left ventricular end-diastolic diameter.
Coldpressor test: The patient was asked to immerse the hand (contralateral to catheter site), up to the wrist, in ice-cold water for 1 min. The magnitude of the heart rate and blood pressure responses to somatic pain were examined. These responses do not depend on afferent pathways involved in LBNP; this test is used, therefore, to determine whether abnormal responses to LBNP were specific or were a manifestation of a generalized alteration in responsiveness to reflex stimuli. 2°'21 Arterial baroreceptor sensitivity test: According to the method of Smyth et al., 22 a bolus of phenylephrine (25, 50, or 75#g) was injected intravenously starting with the lower dose followed by the next higher to increase systolic blood pressure by at least 2 0 m m H g . In our study, the test could not be performed in five hypertensive patients because of high levels of baseline blood pressure; however, all of these five patients had amylnitrite inhalation test for baroreceptor sensitivity as described below. The sensitivity of the baroreceptors was quantitated by correlating the individual intra-arterial systolic blood pressure levels with the subsequent R - R interval obtained from a simultaneous electrocardiographic tracing recorded at a paper speed of 100 mm/s. Amylnitrite inhalation test was performed following the phenylephrine test whenever the two
Cardiovascular regulation in hypertension Table 1. Baseline haemodynamics in hypertensive and normal subjects All hypertensives n Heart rate (bpm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Cardiac index (I/min/m 2) Total peripheral resistance (u'm 2) Right atrial pressure (mmHg) Pulmonary wedge pressure (mmHg) Cardiopulmonary volume/total blood volume (%) Left ventricular ejection fraction (%) +dv/dt (Hz) -dv/dt (Hz) +dv/dt/-dv/dt (Hz)
67 161 96 3.1 39 9 14 11 59 2.4 2.5 0.96
25 -+ 2 ± 5 _+ 2 _+ 0.1 _+ 1.9 _+ 0.8 _+ 0.8 _+ 0.5 -+ 1.9 _+ 0.1 i 0.1 -t- 0.03
Normals
66 127 80 3.6 27 8 12 13 60 2.6 2.4 1.08
10 -+ 2 -+ 2 -t- 0.6 __+0.2 -+ 1.4 _+ 0.9 _+ 1.1 _+0.9 -+ 1.4 _+ 0.1 -I- 0.1 _+ 0.05
p -NS 0.0001 0.0001 0.012 0.0001 NS NS NS NS NS NS NS
Mean -t- SE.
dv/dt ratio = left ventricular filling characteristics expressed as the ratio of left ventricular peak filling rate (positive dv/dt) to peak ejection rate (negative dv/dt).
tests were done in sequence. While intra-arterial blood pressure was recorded continuously at a speed of 100 ram/s, the subject was asked to inhale deeply twice as the ampoule of amylnitrite was crushed close to the nostrils. The change in systolic blood pressure was examined immediately; if the drop in systolic blood pressure was less than 20 mmHg, the test was repeated after 5 min of rest. During both phenylephrine test and amylnitrite test, the subjects were instructed to avoid straining. Plasma catecholamines: Plasma norepinephrine and epinephrine were determined by radioenzymatic assay. 23 Normal supine values in our laboratory average 260-t-120 (SD) pg/ml for total plasma catecholamines, 218 -t:_92 for plasma noradrenaline and 42 _ 18 for plasma epinephrine. Statistical analysis: Standard techniques 24 were applied using the Prophet computer system. Analysis of variance was used for comparison. Statistical significance was defined by a p value of
