Baroreflex regulation of forearm vascular resistance after exercise in hypertensive and normotensive humans JEAN CLEROUX, N’GUESSAN KOUAME, ANDRE NADEAU, DENIS COULOMBE, AND YVES LACOURCIERE Hypertension Research Unit and Department of Cardiology, Centre Hospitalier de l’Universit6 Lava1 Research Center, Lava1 University, Quebec Gl V 4G2, Canada Cl&roux, Jean, N’guessan Kouam6, Andr6 Nadeau, Denis Coulombe, and Yves Lacourcih-e. Baroreflex regulation of forearm vascular resistance after exercise in hypertensive and normotensive humans. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H1523-H1531, 1992.-The mechanisms underlying the antihypertensive period following a bout of exercise are not well understood. We examined the aftereffects of exercise on the linear relationship between forearm vascular resistance (FVR) and estimated central venous pressure (CVP) during leg raising and lower body negative pressure to determine whether an alteration of the baroreflex control of FVR was associated with the decreased blood pressure. Blood pressure, forearm blood flow (FBF), and estimated CVP were obtained in 13 hypertensive and 9 normotensive subjects evaluated in a randomized crossover fashion after 30 min of cycle ergometer exercise and after a nonexercise control period. In hypertensive subjects, the reduced blood pressure was accompanied by an increased baseline FBF after exercise. This resulted in a downward shift of the FVR-CVP relationship, while the slope was unchanged. In normotensive subjects, blood pressure and baroreflex control of FVR were unaffected by prior exercise. Four of the hypertensive subjects performed an additional study in which forearm skin was vasodilated with local heating to FVR levels similar to those observed after exercise. Results suggested that the aftereffects of exercise could not be attributed to changes in cutaneous blood flow. We speculate that modulation of the baroreflex control of FVR after exercise contributes to its antihypertensive effect. blood pressure; forearm blood flow; central venous pressure; ventricular internal diameter in diastole
left
BOUT of exercise, blood pressure is usually reduced in hypertensive patients. This period may extend over several hours when the patients remain seated or supine in the laboratory (2, 6, 13, 18, 34). The mechanisms involved in this antihypertensive period are not well understood. In a recent paper (6), we reported that a fall in total peripheral resistance accompanied by a significant vasodilatation in the forearm was associated with the reduced blood pressure after exercise. In the present paper, we examine the aftereffects of exercise on baroreflex control of forearm vascular resistance (FVR) in an attempt to gain more insight into the mechanism(s) underlying forearm vasodilatation. In addition to arterial baroreceptors (25), cardiopulmonary receptors with vagal afferents play an important role in the reflex regulation of circulation. Cardiopulmonary reflexes have been shown to be involved in the control of skeletal muscle vascular resistance by restraining sympathetic activity to systemic arterioles via tonic inhibition of central vasomotor centers (24, 27, 31). Inhibitory cardiopulmonary baroreflexes were shown to decrease systemic vascular resistance after exercise in dogs (10). Furthermore, Bennett et al. (2) reported that reflex vasoconstriction in the forearm in
AFTER A SINGLE
0363-6135/92
$2.00
Copyright
response to lower body negative pressure (LBNP) was enhanced after exercise in hypertensive patients. Taken together, these results suggest that altered baroreflexes could contribute to decreased blood pressure after exercise by decreasing sympathetic nervous outflow to skeletal muscle vascular beds. However, Bennett et al. (2) did not determine whether central blood volume, which is thought to be an important stimulus to cardiopulmonary receptors, was similar before and after exercise. This raises the possibility that the increased reflex vasoconstriction in response to baroreceptor unloading found by these authors may not have been related to an increased sensitivity of the baroreflex per se (defined as the slope of the relationship between FVR and central blood volume), but to other factors such as a change in venous return, in vascular reactivity, and/or in the distribution of blood flow to forearm tissues after exercise. The present study was therefore undertaken to examine the aftereffects of exercise on baroreflex control of FVR while taking these factors into account. METHODS Subjects Thirteen patients with uncomplicated, established mild to moderate hypertension and nine normotensive subjects gave informed consent to participate in this randomized crossover study, which was approved by our institutional ethics committee on human research. Secondary hypertension was ruled out by clinical and laboratory evaluation. Patients with labile hypertension and women of childbearing potential were excluded from the study. None of the subjects was receiving antihypertensive therapy or any other medication during this study. Antihypertensive therapy was gradually withdrawn over 2 wk in patients previously treated. They remained without treatment during an additional 6-wk period before entering the study. All subjects were seen weekly during 3 wk preceding entry into the study to measure office blood pressure. Office blood pressure was taken as the mean of three measurements not differing by >5 mmHg in the sitting position at 5-min intervals. Hypertensive patients qualified for the study if their diastolic blood pressure was 90-114 mmHg during this period. Normotensive subjects with diastolic pressure ~85 mmHg qualified for the study. During screening, blood pressure was measured in both arms to ensure that the levels did not differ by >5 mmHg. Maneuvers
Used to Study Baroreflex
Control
of FVR
Baroreflex control of FVR was studied during maneuvers that activated or deactivated mainly cardiopulmonary receptors. Activation of cardiopulmonary receptors was obtained by increasing venous return to the heart and intrathoracic blood volume via passive elevation of the legs of the supine subjects to 60” (5, 15, 16, 26, 29). Deactivation of cardiopulmonary receptors was obtained by reduction of venous return and intrathoracic blood volume via the technique of LBNP, whereby the legs and pelvis of the supine subjects are enclosed within a rigid Plexiglas box
0 1992 the American
Physiological
Society
H1523
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
H1524
EXERCISE
AND
REFLEX
CONTROL
up to the level of the anterosuperior iliac crests (5, 15, 16, 21, 26). The pressure within the box was monitored with a pressure transducer (P23 Id, Gould) and was reduced 15 mmHg below atmospheric pressure with a vacuum (LBNP15). Leg raising (LR) and LBNP up to LBNP15 have been shown to increase and decrease central blood volume with little or no change in blood pressure and heart rate, thereby activating and deactivating, respectively, cardiopulmonary receptors without involving arterial baroreceptors, and to induce reflex vasodilatation and vasoconstriction, respectively, in the forearm (5, 15, 16, 20, 26, 35).
Reflex vasoconstriction during LBNP at -40 mmHg (LBNP40) was also examined. During this level of LBNP, arterial baroreceptors are deactivated in addition to cardiopulmonary receptors (1, 5, 15, 16). The FVR response to this maneuver therefore represents integrated baroreflex control of FVR. The sensitivity of baroreflex control of FVR was taken as the slope of the relationship between FVR and central venous pressure (CVP) (23), as well as between FVR and left ventricular internal diameter in diastole (LVIDD). Slopes were calculated while taking into account the FVR response during LR and LBNP15 maneuvers only, and also while considering the results during LBNP40. Main
Study Protocol
In this study, a randomized crossover design was used to examine the reflex control of FVR after exercise and after a nonexercise control situation. This design was chosen to measure control values over the same time period to that required for the postexercise evaluation. This also allowed avoiding an unduly extended stay of the subjects in the laboratory. The two evaluations were performed 1 wk apart at the same time of day, i.e., 0830 h. Each evaluation required 3.5 h. The first hour was used for exercise or rest; the next 1.5 h for measuring heart rate, blood pressure, and forearm blood flow (FBF) at baseline and during the maneuvers; and the last hour for measuring heart rate, blood pressure, and CVP and LVIDD at baseline and during the same maneuvers. On the exercise day, the first hour comprised 30 min of upright exercise on a cycle ergometer at 50% of maximal oxygen uptake, followed by 15 min of recovery sitting in a chair, and 15 min for setting the FBF, heart rate, and blood pressure measurement equipment with the subjects lying on a bed. During exercise, oxygen uptake was monitored on-line and work rate was adjusted to elicit 50% of the maximal individual value determined 2 wk before the first evaluation. On the nonexercise control day, the subjects remained seated for 45 min before moving to the bed. Forearm vascular reflex responses were evaluated over the following 1.5 h, i.e., from 30 to 120 min after exercise or seated rest. Alternate periods of baseline and stimulus application, each for 15 min, were used. LR at an angle of 60”, LBNP15 and LBNP40 were applied in a “random” order (LR at random with LBNP, and LBNP15 at random with LBNP40). The same sequence was always used for a given subject. A 60-s cold pressor test, during which one hand was immersed in iced water, was performed after the maneuvers to examine the FVR response to a stimulus different from that originating from the cardiopulmonary region (33). During the last hour, i.e., between 120 and 180 min after exercise or seated rest, the sequence was repeated with alternating lo-min baseline and stimulus periods to allow estimation of changes in CVP via an antecubital vein catheter (14, 23), and measurement of LVIDD with echocardiography during the maneuvers (5, 15). All measurements were performed during the last 5 min of LR and LBNP. During the cold pressor test, measurements were performed before and during the last 30 s of
OF FOREARM
VASCULAR
RESISTANCE
immersion. On study days, all subjects were instructed to have a light breakfast and to abstain from caffeine and smoking. On a separate occasion, we obtained data in four hypertensive and four normotensive subjects showing that baseline estimated CVP and LVIDD and their changes in response to the maneuvers were stable over the 2.5-h measurement period during the control evaluation and after exercise. The data collected during LBNP15 are provided for illustrative purposes. Thus, during this maneuver, the mean decrease in CVP between 30 and 120 min was similar to the decrease between 120 and 180 min after seated rest in the control study (-2.64 t 0.51 and -2.70 t 0.64 mmHg, respectively) and after exercise (-2.57 & 0.72 and -2.48 & 0.74 mmHg, respectively) in hypertensive subjects. The respective CVP decreases were -2.75 & 0.43 and -2.75 t 0.72 mmHg in the control study and -3.13 & 0.55 and -3.00 & 0.85 mmHg after exercise in normotensive subjects. With regard to LVIDD measurements, mean decreases in response to LBNPl5 during the 30- to 120- and 120- to 180-min time frames were -2.44 & 0.7 and -2.81 * 0.8 mm, respectively, in the control study and -2.82 & 0.83 and -3.00 * 0.53 mm after exercise in hypertensive subjects, and -3.30 ? 0.95 and -3.00 t 0.91 mm in the control study and -3.10 & 0.63 and -3.25 t 0.75 mm after exercise in normotensive subjects. Additional Study Protocol To investigate whether the aftereffects of exercise on cardiopulmonary reflex control of FVR could be related to an increased skin blood flow in the forearm, an additional study was performed with four hypertensive subjects having participated in the main study. In this study, reflex vasomotor responses during LR and LBNP15 were examined in control conditions and during local heat-induced vasodilatation of forearm cutaneous vessels. We reasoned that if skin vasodilatation was involved in modifying the FVR response to the maneuvers after exercise, then selectively heating the forearm would result in a similar response. Heat was applied with two 250-W infrared light bulbs placed 60 cm above and below the forearm (500 W total), with the subject shielded from the light source with a curtain. At this distance, each light source covered a surface of 0.5 rn? The total infrared radiation was therefore 500 W/m”, or about one-half that of the sun (3). Alternate 15-min baseline and stimulus periods were used in the control situation and during heating of the forearm for a total study duration of 2 h. Skin surface temperature was measured at I-min intervals throughout the study with an infrared thermometer (model 310, Everest Interscience, Tustin, CA). Estimation of CVP and measurement of LVIDD were not performed in this study. Measurements Oxygen uptake was measured with Sensormedics’ Energy Expenditure Unit 2900 (Anaheim, CA), which analyzes expired gases drawn at 250 ml/min from a 7-liter mixing chamber with infrared absorption and zirconium cell analyzers for carbon dioxide and oxygen, respectively. The analyzers and mass flowmeter are interfaced with an IBM PS-2 computer to give 20-s averages of values sampled at 100 Hz. Maximal oxygen uptake was measured during a progressively increasing work load test on a cycle ergometer (Ergomedic 8293, Monark, Sweden) with increments of 50 W every 2 min. Verbal encouragement was given when the subjects demonstrated signs of fatigue. Maximal values were considered to be attained when the subjects were unable to maintain a given power output for 2 min, oxygen uptake increased less or no further during the last steps, heart rate reached maximal age-predicted value, and respiratory exchange ratio was X.10. Blood pressure was measured with a standard mercury sphygmomanometer, taking the first and the fifth Korotkoff sounds
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
EXERCISE
AND
REFLEX
CONTROL
as the systolic and diastolic values, respectively, with a cuff of appropriate size. Mean blood pressure was calculated as diastolic plus one-third pulse pressure. Heart rate was measured with a tachograph triggered by the R wave of the electrocardiogram recorded in lead III (7P4, Grass Instruments), and both traces were recorded on polygraph paper. CVP was estimated by measuring venous pressure from a small antecubital vein catheter (20 g, 32 mm, Cathlon IV, Critikon, Canada) connected via a saline-filled line to a pressure transducer (P23 Id, Gould). The transducer was positioned at midsternal level in a lateral decubitus position while letting the arm hang down on the side of the bed. The right arm (14, 23) was used in 8 subjects and the left arm in 14 subjects. No differences were observed between venous pressure tracings recorded from the right or left arm. Echocardiography (ATLMark III Ultrasonograph, Seattle, WA) was performed in a left lateral decubitus position simultaneously with CVP estimation in 14 subjects (8 hypertensive and 6 normotensive individuals). LVIDD was determined in M mode after locating the parasternal long axis in two-dimensional mode (5, 15) over three consecutive cardiac cycles when heart rate was stable. The intrasubject variability (SD) of LVIDD measurements 1 wk apart was &2.3%. Left ventricular mass was calculated from left ventricular septal wall thickness, internal diameter, and posterior wall thickness according to the Penn convention and the formula of Devereux and Reichek (1 l), and standardized per unit of body surface area. Blood samples were taken at midpoint during the control evaluation and after exercise for hematocrit and hemoglobin determinations, which were used to evaluate plasma volume changes (8). FBF was measured by venous occlusion plethysmography (Hokanson EC-4, Issaquah, WA) using a mercury-in-Silastic strain gauge (17) applied around the arm contralateral to that employed for blood pressure measurements. The strain gauge was placed 4-5 cm distal from the antecubital crease and the forearm elevated 10 cm above the anterior chest wall. Measurements were made at constant room temperature (23-24°C) while circulation to the hand was arrested by inflating a cuff around the wrist at suprasystolic pressure. Measurements were performed on tracings recorded during the last minute of each 2-min hand-occluded period to avoid the first-minute period after inflation of the arresting cuff at the wrist during which arterial inflow to the forearm is disturbed (17). Measurements were derived from the average of three consecutive values. FBF variability (SD) calculated on two sequential averages was +5% in agreement with previous reports (5, 16). FVR was calculated by dividing mean arterial pressure (mean of 2 measurements) by FBF. Statistical
Analysis
Results are expressed as means t SE. The aftereffects of exercise on the variables at baseline and during the various maneuvers were examined with a two-way (hypertensive and normotensive individuals) analysis of variance (ANOVA) for repeated measurements (control and postexercise evaluations). The slopes and intercepts of least-square regression equations calculated individually for each subject between FVR and estimated CVP, as well as between FVR and LVIDD, were also compared with two-way ANOVAs for repeated measurements. FVR intercepts were compared at estimated CVP of 0 mmHg and at LVIDD of 50 mm based on the observation that estimated CVP and LVIDD were linearly related and intercepted at LVIDD of 50 mm (see RESULTS). When a significant (P < 0.05) F ratio was observed, Duncan’s test was used to locate significant differences (30). Data from the study on the effects of heating were analyzed with Student’s paired t test.
OF FOREARM
VASCULAR
H1525
RESISTANCE
RESULTS
Both groups were characterized by similar age, weight, height, maximal oxygen uptake, LVIDD, and left ventricular mass index (Table 1). Office blood pressure (mean of 3 weekly visits) was higher in hypertensive (151 t 3/99 t 1 mmHg) than in normotensive subjects (118 t 2/79 t 4 mmHg). Similarly, during the control evaluation, supine systolic and diastolic blood pressure levels were higher in hypertensives (Table 1). With regard to regional hemodynamics, FBF was lower and FVR was higher in hypertensive than in normotensive subjects (Table 1). Cycle ergometer exercise during 30 min induced identical increases in heart rate and percent maximal oxygen uptake in both groups (Table 2). The power output required to induce similar relative exercise intensities was not significantly different in both groups. During exercise, systolic blood pressure increased to higher values in hypertensives. Diastolic pressure remained unchanged or decreased slightly from resting levels, but remained greater in hypertensives than in normotensives. Aftereffects of Exercise Baseline results. Figures 1 and 2 show baseline results obtained after rest and after exercise in both groups of subjects. In hypertensive subjects, systolic (-11 t 2 mmHg) and diastolic (-4 t 1 mmHg) blood pressures were significantly lower after exercise than during the control evaluation. This antihypertensive effect was sustained during the entire period of measurements, as shown by similar changes in blood pressure 30 min (-13 -+ 3/-4 -+ 2 mmHg) and 3 h after exercise (-9 t 2/-4 t 2 mmHg). In contrast, exercise produced no aftereffect on the blood pressure of normotensive subjects. A moderate but significant tachycardia was detected both in hypertensive (+5 t 1 beats/min) and normotensive subjects (+6 t 1 beats/min) after exercise compared with control. In hypertensive subjects, FBF increased by 34 t 11% after exercise, whereas it remained similar in normotensive subjects. FVR decreased by 25 t 6% after exercise compared with control in hypertensives, whereas it was unchanged in normotensives.
Table 1. Characteristics of hypertensive and normotensive subjects participating in the study Variable
n Age, yr Weight, kg Height, cm VO ml. kg-- ’ . min- l Cardiac mass, g/m” LVIDD, mm Systolic pressure, mmHg Diastolic pressure, mmHg Heart rate, beats/min FBF, ml 100 gg’ smin-’ FVR, units 2
IlilX9
Hypertensives
13 44&Z 75t3 175t2 35tl 87k6 5022 14e4* 95t2* 59tl 1.8&0.1* 64.6t2.7*
Normotensives
9 41k2 73+6 173t3 36+3 89t5 51+1 106+2 77t3 58t4 2.4k0.2 37.2t2.7
Values are means f. SE. VO, tn;,x, maximal oxygen uptake; LVIDD, left ventricular internal diameter in diastole; FBF and FVR, forearm blood flow and forearm vascular resistance. * P c 0.05 vs. normotensives.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
H1526
EXERCISE
AND
REFLEX
CONTROL
OF FOREARM
Table 2. Cardiovascular variables and work load during the 30-min period of exercise in hypertens ive and no rmoltensive subjects Variable
Hypertensives
Normotensives
Heart rate, beats/min 128k5 Systolic pressure, mmHg 170t5* Diastolic pressure, mmHg 90t2* Mean arterial pressure, mmHg 117t4” Work load, W 84k6 53&l %G tni\x Values are means f: SE. %i70L! tlIRX, percent of maximal take. * P < 0.05 vs. normotensives.
133xk4 138k4 68k4 91t4 91tll 52tl oxygen
up-
150 140
a CONTROL 69 AFTER EXERCISE
130 =120 0 M E 110 I
g100 90 80 70 i
-
80 70
E 3.0 -' ii0 LJ- g
[L s60 > 'E 50 k3 -40
2.5 2.0
9
1.5
30
*
1.0
20
-
--
I-II
I I
NT
I
Fig. 1. Blood pressure (BP), heart rate (HR), forearm blood flow (FBF), and forearm vascular resistance (FVR) during control evaluation and after exercise in hypertensive (HT) and normotensive (NT) subjects. In the BP panel, systolic and diastolic values correspond to top and bottom of each bar, respectively, and mean arterial pressure is indicated within each bar. * P < 0.05 vs. control.
Identical hemoglobin concentrations were found after exercise and during the control evaluation in either group (144 t 3 g/l), while no differences were found in hematocrit in hypertensive (42 t 1 and 41 t 1%) and normotensive subjects (43 t 1 and 42 t 1%). The calculated changes in plasma volume were not significant in hypertensives (-0.9 t 1.7%) and in normotensives (-0.4 t 1.0%) after exercise compared with control. Estimated CVP also did not vary significantly after exercise compared with control in hypertensives (-0.2 t 0.2 mmHg; NS), but decreased in normotensives (-0.6 t 0.2 mmHg; P < 0.05). LVIDD was unchanged in hypertensives (+0.9 t 0.8 mm; NS), but increased slightly in normotensives (+l.l t 0.4 mm; P c 0.05) after exercise compared with control. Response to maneuvers. Figure 3 shows that LR induced a significant increase in estimated CVP, as well as in LVIDD, while LBNP15 and LBNP40 induced graded falls in estimated CVP and in LVIDD. The changes in estimated CVP and in LVIDD were similar during the control evaluation and after exercise (see also Fig. 2). In addition, the relationship between CVP and LVIDD was similar in normotensive and hypertensive subjects. In the group of subjects in which both measurements were per-
VASCULAR
RESISTANCE
formed (n = 14, 8 hypertensive and 6 normotensive subjects), CVP was linearly related to LVIDD by the equation LVIDD (mm) = 1.33 [estimated CVP (mmHg)] + 50.0. LR and LBNPl5 did not induce any significant change in blood pressure or heart rate (data not shown). In contrast, LBNP40 induced a significant decrease in systolic blood pressure in hypertensive (-9 t 1 and -10 t 3 mmHg, during the control evaluation and after exercise, respectively) and in normotensive subjects (-6 t 1 and -7 t 1 mmHg). This was associated with reflex tachycardia both in hypertensive (+13 t 2 and +l3 t 2 beats/ min) and normotensive subjects (+l7 t 2 and +20 t 4 beats/min). Thus arterial baroreceptors were additionally deactivated during this maneuver. LR induced a decrease in FVR, while graded LBNP induced progressive increases in FVR in both groups of subjects. The FVR-estimated CVP and FVR-LVIDD relationships are illustrated in Fig. 4. Statistical comparison of control and postexercise regression lines of FVRestimated CVP and FVR-LVIDD relationships yielded identical conclusions whether the data from LBNP40 was included or not with the data from LBNP15 and LR. For clarity, only the relationships taking into account the data from LBNP40 will be presented. The FVR-estimated CVP relationship was shifted downward after exercise compared with control in hypertensive subjects (at estimated CVP = 0, FVR = 55.0 t 8.4 and 70.4 t 6.5 units, respectively; P < 0.05), whereas no significant change was noted in normotensive subjects (at estimated CVP = 0, FVR = 40.3 t 6.2 and 45.2 t 5.0 units; NS). The slopes were not significantly different after exercise compared with control in hypertensive subjects (-5.21 t 0.76 and -4.18 t 0.64 units/mmHg, respectively) or in normotensive subjects (-3.79 t 0.61 and -4.49 t 0.56 units/mmHg). Similar findings were made concerning the slopes of the FVR-LVIDD relationships after exercise compared with control in hypertensive (-4.78 t 0.87 and -3.80 t 0.93 units/mm; NS) and in normotensive subjects (-2.67 t 0.79 and -2.68 t 0.86 units/mm; NS). However, as for the FVR-estimated CVP relationship, the FVR-LVIDD relationship was shifted downward after exercise compared with control in hypertensive (at LVIDD = 50, FVR = 53.1 t 7.9 and 65.8 t 6.9 units; P < 0.05) but not in normotensive subjects (at LVIDD = 50, FVR = 40.8 t 4.2 and 41.0 t 5.3 units; NS). Cold Pressor Test As shown in Table 3, immersion of one hand in iced water induced a moderate tachycardia and a marked pressor response and increase in FVR that were unaffected by prior exercise in hypertensive as well as in normotensive subjects. Effects of Local Heat Application Heating the forearm induced a rapid rise in skin surface temperature that plateaued within the first 5 min. Mean skin temperature increased up to a stable level of 33.6 t 0.4OC during heating from 32.0 t 0.5”C in the control situation (P < 0.05). Heating was associated with significant (P < 0.05) increases in FBF (2.7 t 0.8 vs. 2.1 t 0.8
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
EXERCISE
AND
REFLEX
CONTROL
OF FOREARM
VASCULAR
RESISTANCE
H1527
A Af=rER
CONTROL
CVP (mm Hg) LVIDD
; -5 I
HR (b.min -1)
45.3
50.2
46.0
m/105
163/100
135195
136193
;; 4.
Hg)
(ml.lOOcc FBF-l.mln * -l) FVR
50.7
(mm)
BP (mm
EXERCISE
m
Ld.n 79.2
86.4
49.2
56.5
BASAL
lBNP15
BASAL
LBNP15
(Units)
B CONTROL
Af7ER
EXERCISE
Fig. 2. Original tracings from a hypertensive (A) and a normotensive (B) subject showing mean central venous pressure (CVP, measured from a peripheral vein), heart rate (HR), and forearm blood flow (FBF) recordings, along with left ventricular internal diameter in diastole (LVIDD), blood pressure (BP), and calculated forearm vascular resistance (FVR) at baseline and during lower body negative pressure at -15 mmHg (LBNP15) during control evaluation and af-
LLduuLld
LVIDD
50.0
(mm)
HR (b.min -1) BP (mm
ii 3
---C./---L
Hg)
(Units)
49.6
47.3
9417-I
98/69
95166
56.2
41.4
50.4
7I------94169
(ml.lOOcc FBF-l.mln - -l) lP--c-d FVR
46.9
100
LL-iYkY 42.9
BASAL
LBNP15
BASAL
ml.100 ml-‘.rnin-‘) and decreases in FVR (39.7 t 11.2 vs. 55.3 t 14.2 units) compared with the control situation, respectively. The changes in baseline levels were similar to, i.e., not significantly different from, those found in the same four subjects after exercise compared with the nonexercise control evaluation (FBF, 3.0 t 0.7 vs. 1.8 t 0.1 ml. 100 ml-’ emin-‘; FVR, 41.0 t 7.9 vs. 63.0 t 3.2 units; both P < 0.05; see also mean group values, Fig. 1). However, reflex FVR responses during LR and LBNP15 were different after exercise and during heating. Figure 5 illustrates that the reflex changes in FVR were unaffected after exercise but decreased significantly during heating. DISCUSSION
In agreement with previous studies in hypertensive subjects, the present results indicate that a statistically and clinically significant antihypertensive period was observed after mild physical exercise (2, 13, 18, 34). In our study, a control group of normotensive subjects was additionally evaluated and no significant change in blood pressure was found after exercise. Comparison of the effects of exercise on the baroreflex control of FVR in hypertensive and normotensive subjects therefore allows
LBNPlS
identification of alterations that are associated with the reduced blood pressure. The results of this study show that the baroreflex regulation of FVR is modified during the postexercise period in hypertensive subjects. During maneuvers that increased and decreased venous return, the relationships between FVR and estimated CVP, as well as between FVR and LVIDD, were shifted downward, i.e., to lower FVR values, after exercise in hypertensive subjects. This was not simply related to a lower mean arterial blood pressure. In fact, the increase in baseline FBF (~30%) was more pronounced than the fail in blood pressure (
left
BOUT of exercise, blood pressure is usually reduced in hypertensive patients. This period may extend over several hours when the patients remain seated or supine in the laboratory (2, 6, 13, 18, 34). The mechanisms involved in this antihypertensive period are not well understood. In a recent paper (6), we reported that a fall in total peripheral resistance accompanied by a significant vasodilatation in the forearm was associated with the reduced blood pressure after exercise. In the present paper, we examine the aftereffects of exercise on baroreflex control of forearm vascular resistance (FVR) in an attempt to gain more insight into the mechanism(s) underlying forearm vasodilatation. In addition to arterial baroreceptors (25), cardiopulmonary receptors with vagal afferents play an important role in the reflex regulation of circulation. Cardiopulmonary reflexes have been shown to be involved in the control of skeletal muscle vascular resistance by restraining sympathetic activity to systemic arterioles via tonic inhibition of central vasomotor centers (24, 27, 31). Inhibitory cardiopulmonary baroreflexes were shown to decrease systemic vascular resistance after exercise in dogs (10). Furthermore, Bennett et al. (2) reported that reflex vasoconstriction in the forearm in
AFTER A SINGLE
0363-6135/92
$2.00
Copyright
response to lower body negative pressure (LBNP) was enhanced after exercise in hypertensive patients. Taken together, these results suggest that altered baroreflexes could contribute to decreased blood pressure after exercise by decreasing sympathetic nervous outflow to skeletal muscle vascular beds. However, Bennett et al. (2) did not determine whether central blood volume, which is thought to be an important stimulus to cardiopulmonary receptors, was similar before and after exercise. This raises the possibility that the increased reflex vasoconstriction in response to baroreceptor unloading found by these authors may not have been related to an increased sensitivity of the baroreflex per se (defined as the slope of the relationship between FVR and central blood volume), but to other factors such as a change in venous return, in vascular reactivity, and/or in the distribution of blood flow to forearm tissues after exercise. The present study was therefore undertaken to examine the aftereffects of exercise on baroreflex control of FVR while taking these factors into account. METHODS Subjects Thirteen patients with uncomplicated, established mild to moderate hypertension and nine normotensive subjects gave informed consent to participate in this randomized crossover study, which was approved by our institutional ethics committee on human research. Secondary hypertension was ruled out by clinical and laboratory evaluation. Patients with labile hypertension and women of childbearing potential were excluded from the study. None of the subjects was receiving antihypertensive therapy or any other medication during this study. Antihypertensive therapy was gradually withdrawn over 2 wk in patients previously treated. They remained without treatment during an additional 6-wk period before entering the study. All subjects were seen weekly during 3 wk preceding entry into the study to measure office blood pressure. Office blood pressure was taken as the mean of three measurements not differing by >5 mmHg in the sitting position at 5-min intervals. Hypertensive patients qualified for the study if their diastolic blood pressure was 90-114 mmHg during this period. Normotensive subjects with diastolic pressure ~85 mmHg qualified for the study. During screening, blood pressure was measured in both arms to ensure that the levels did not differ by >5 mmHg. Maneuvers
Used to Study Baroreflex
Control
of FVR
Baroreflex control of FVR was studied during maneuvers that activated or deactivated mainly cardiopulmonary receptors. Activation of cardiopulmonary receptors was obtained by increasing venous return to the heart and intrathoracic blood volume via passive elevation of the legs of the supine subjects to 60” (5, 15, 16, 26, 29). Deactivation of cardiopulmonary receptors was obtained by reduction of venous return and intrathoracic blood volume via the technique of LBNP, whereby the legs and pelvis of the supine subjects are enclosed within a rigid Plexiglas box
0 1992 the American
Physiological
Society
H1523
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
H1524
EXERCISE
AND
REFLEX
CONTROL
up to the level of the anterosuperior iliac crests (5, 15, 16, 21, 26). The pressure within the box was monitored with a pressure transducer (P23 Id, Gould) and was reduced 15 mmHg below atmospheric pressure with a vacuum (LBNP15). Leg raising (LR) and LBNP up to LBNP15 have been shown to increase and decrease central blood volume with little or no change in blood pressure and heart rate, thereby activating and deactivating, respectively, cardiopulmonary receptors without involving arterial baroreceptors, and to induce reflex vasodilatation and vasoconstriction, respectively, in the forearm (5, 15, 16, 20, 26, 35).
Reflex vasoconstriction during LBNP at -40 mmHg (LBNP40) was also examined. During this level of LBNP, arterial baroreceptors are deactivated in addition to cardiopulmonary receptors (1, 5, 15, 16). The FVR response to this maneuver therefore represents integrated baroreflex control of FVR. The sensitivity of baroreflex control of FVR was taken as the slope of the relationship between FVR and central venous pressure (CVP) (23), as well as between FVR and left ventricular internal diameter in diastole (LVIDD). Slopes were calculated while taking into account the FVR response during LR and LBNP15 maneuvers only, and also while considering the results during LBNP40. Main
Study Protocol
In this study, a randomized crossover design was used to examine the reflex control of FVR after exercise and after a nonexercise control situation. This design was chosen to measure control values over the same time period to that required for the postexercise evaluation. This also allowed avoiding an unduly extended stay of the subjects in the laboratory. The two evaluations were performed 1 wk apart at the same time of day, i.e., 0830 h. Each evaluation required 3.5 h. The first hour was used for exercise or rest; the next 1.5 h for measuring heart rate, blood pressure, and forearm blood flow (FBF) at baseline and during the maneuvers; and the last hour for measuring heart rate, blood pressure, and CVP and LVIDD at baseline and during the same maneuvers. On the exercise day, the first hour comprised 30 min of upright exercise on a cycle ergometer at 50% of maximal oxygen uptake, followed by 15 min of recovery sitting in a chair, and 15 min for setting the FBF, heart rate, and blood pressure measurement equipment with the subjects lying on a bed. During exercise, oxygen uptake was monitored on-line and work rate was adjusted to elicit 50% of the maximal individual value determined 2 wk before the first evaluation. On the nonexercise control day, the subjects remained seated for 45 min before moving to the bed. Forearm vascular reflex responses were evaluated over the following 1.5 h, i.e., from 30 to 120 min after exercise or seated rest. Alternate periods of baseline and stimulus application, each for 15 min, were used. LR at an angle of 60”, LBNP15 and LBNP40 were applied in a “random” order (LR at random with LBNP, and LBNP15 at random with LBNP40). The same sequence was always used for a given subject. A 60-s cold pressor test, during which one hand was immersed in iced water, was performed after the maneuvers to examine the FVR response to a stimulus different from that originating from the cardiopulmonary region (33). During the last hour, i.e., between 120 and 180 min after exercise or seated rest, the sequence was repeated with alternating lo-min baseline and stimulus periods to allow estimation of changes in CVP via an antecubital vein catheter (14, 23), and measurement of LVIDD with echocardiography during the maneuvers (5, 15). All measurements were performed during the last 5 min of LR and LBNP. During the cold pressor test, measurements were performed before and during the last 30 s of
OF FOREARM
VASCULAR
RESISTANCE
immersion. On study days, all subjects were instructed to have a light breakfast and to abstain from caffeine and smoking. On a separate occasion, we obtained data in four hypertensive and four normotensive subjects showing that baseline estimated CVP and LVIDD and their changes in response to the maneuvers were stable over the 2.5-h measurement period during the control evaluation and after exercise. The data collected during LBNP15 are provided for illustrative purposes. Thus, during this maneuver, the mean decrease in CVP between 30 and 120 min was similar to the decrease between 120 and 180 min after seated rest in the control study (-2.64 t 0.51 and -2.70 t 0.64 mmHg, respectively) and after exercise (-2.57 & 0.72 and -2.48 & 0.74 mmHg, respectively) in hypertensive subjects. The respective CVP decreases were -2.75 & 0.43 and -2.75 t 0.72 mmHg in the control study and -3.13 & 0.55 and -3.00 & 0.85 mmHg after exercise in normotensive subjects. With regard to LVIDD measurements, mean decreases in response to LBNPl5 during the 30- to 120- and 120- to 180-min time frames were -2.44 & 0.7 and -2.81 * 0.8 mm, respectively, in the control study and -2.82 & 0.83 and -3.00 * 0.53 mm after exercise in hypertensive subjects, and -3.30 ? 0.95 and -3.00 t 0.91 mm in the control study and -3.10 & 0.63 and -3.25 t 0.75 mm after exercise in normotensive subjects. Additional Study Protocol To investigate whether the aftereffects of exercise on cardiopulmonary reflex control of FVR could be related to an increased skin blood flow in the forearm, an additional study was performed with four hypertensive subjects having participated in the main study. In this study, reflex vasomotor responses during LR and LBNP15 were examined in control conditions and during local heat-induced vasodilatation of forearm cutaneous vessels. We reasoned that if skin vasodilatation was involved in modifying the FVR response to the maneuvers after exercise, then selectively heating the forearm would result in a similar response. Heat was applied with two 250-W infrared light bulbs placed 60 cm above and below the forearm (500 W total), with the subject shielded from the light source with a curtain. At this distance, each light source covered a surface of 0.5 rn? The total infrared radiation was therefore 500 W/m”, or about one-half that of the sun (3). Alternate 15-min baseline and stimulus periods were used in the control situation and during heating of the forearm for a total study duration of 2 h. Skin surface temperature was measured at I-min intervals throughout the study with an infrared thermometer (model 310, Everest Interscience, Tustin, CA). Estimation of CVP and measurement of LVIDD were not performed in this study. Measurements Oxygen uptake was measured with Sensormedics’ Energy Expenditure Unit 2900 (Anaheim, CA), which analyzes expired gases drawn at 250 ml/min from a 7-liter mixing chamber with infrared absorption and zirconium cell analyzers for carbon dioxide and oxygen, respectively. The analyzers and mass flowmeter are interfaced with an IBM PS-2 computer to give 20-s averages of values sampled at 100 Hz. Maximal oxygen uptake was measured during a progressively increasing work load test on a cycle ergometer (Ergomedic 8293, Monark, Sweden) with increments of 50 W every 2 min. Verbal encouragement was given when the subjects demonstrated signs of fatigue. Maximal values were considered to be attained when the subjects were unable to maintain a given power output for 2 min, oxygen uptake increased less or no further during the last steps, heart rate reached maximal age-predicted value, and respiratory exchange ratio was X.10. Blood pressure was measured with a standard mercury sphygmomanometer, taking the first and the fifth Korotkoff sounds
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
EXERCISE
AND
REFLEX
CONTROL
as the systolic and diastolic values, respectively, with a cuff of appropriate size. Mean blood pressure was calculated as diastolic plus one-third pulse pressure. Heart rate was measured with a tachograph triggered by the R wave of the electrocardiogram recorded in lead III (7P4, Grass Instruments), and both traces were recorded on polygraph paper. CVP was estimated by measuring venous pressure from a small antecubital vein catheter (20 g, 32 mm, Cathlon IV, Critikon, Canada) connected via a saline-filled line to a pressure transducer (P23 Id, Gould). The transducer was positioned at midsternal level in a lateral decubitus position while letting the arm hang down on the side of the bed. The right arm (14, 23) was used in 8 subjects and the left arm in 14 subjects. No differences were observed between venous pressure tracings recorded from the right or left arm. Echocardiography (ATLMark III Ultrasonograph, Seattle, WA) was performed in a left lateral decubitus position simultaneously with CVP estimation in 14 subjects (8 hypertensive and 6 normotensive individuals). LVIDD was determined in M mode after locating the parasternal long axis in two-dimensional mode (5, 15) over three consecutive cardiac cycles when heart rate was stable. The intrasubject variability (SD) of LVIDD measurements 1 wk apart was &2.3%. Left ventricular mass was calculated from left ventricular septal wall thickness, internal diameter, and posterior wall thickness according to the Penn convention and the formula of Devereux and Reichek (1 l), and standardized per unit of body surface area. Blood samples were taken at midpoint during the control evaluation and after exercise for hematocrit and hemoglobin determinations, which were used to evaluate plasma volume changes (8). FBF was measured by venous occlusion plethysmography (Hokanson EC-4, Issaquah, WA) using a mercury-in-Silastic strain gauge (17) applied around the arm contralateral to that employed for blood pressure measurements. The strain gauge was placed 4-5 cm distal from the antecubital crease and the forearm elevated 10 cm above the anterior chest wall. Measurements were made at constant room temperature (23-24°C) while circulation to the hand was arrested by inflating a cuff around the wrist at suprasystolic pressure. Measurements were performed on tracings recorded during the last minute of each 2-min hand-occluded period to avoid the first-minute period after inflation of the arresting cuff at the wrist during which arterial inflow to the forearm is disturbed (17). Measurements were derived from the average of three consecutive values. FBF variability (SD) calculated on two sequential averages was +5% in agreement with previous reports (5, 16). FVR was calculated by dividing mean arterial pressure (mean of 2 measurements) by FBF. Statistical
Analysis
Results are expressed as means t SE. The aftereffects of exercise on the variables at baseline and during the various maneuvers were examined with a two-way (hypertensive and normotensive individuals) analysis of variance (ANOVA) for repeated measurements (control and postexercise evaluations). The slopes and intercepts of least-square regression equations calculated individually for each subject between FVR and estimated CVP, as well as between FVR and LVIDD, were also compared with two-way ANOVAs for repeated measurements. FVR intercepts were compared at estimated CVP of 0 mmHg and at LVIDD of 50 mm based on the observation that estimated CVP and LVIDD were linearly related and intercepted at LVIDD of 50 mm (see RESULTS). When a significant (P < 0.05) F ratio was observed, Duncan’s test was used to locate significant differences (30). Data from the study on the effects of heating were analyzed with Student’s paired t test.
OF FOREARM
VASCULAR
H1525
RESISTANCE
RESULTS
Both groups were characterized by similar age, weight, height, maximal oxygen uptake, LVIDD, and left ventricular mass index (Table 1). Office blood pressure (mean of 3 weekly visits) was higher in hypertensive (151 t 3/99 t 1 mmHg) than in normotensive subjects (118 t 2/79 t 4 mmHg). Similarly, during the control evaluation, supine systolic and diastolic blood pressure levels were higher in hypertensives (Table 1). With regard to regional hemodynamics, FBF was lower and FVR was higher in hypertensive than in normotensive subjects (Table 1). Cycle ergometer exercise during 30 min induced identical increases in heart rate and percent maximal oxygen uptake in both groups (Table 2). The power output required to induce similar relative exercise intensities was not significantly different in both groups. During exercise, systolic blood pressure increased to higher values in hypertensives. Diastolic pressure remained unchanged or decreased slightly from resting levels, but remained greater in hypertensives than in normotensives. Aftereffects of Exercise Baseline results. Figures 1 and 2 show baseline results obtained after rest and after exercise in both groups of subjects. In hypertensive subjects, systolic (-11 t 2 mmHg) and diastolic (-4 t 1 mmHg) blood pressures were significantly lower after exercise than during the control evaluation. This antihypertensive effect was sustained during the entire period of measurements, as shown by similar changes in blood pressure 30 min (-13 -+ 3/-4 -+ 2 mmHg) and 3 h after exercise (-9 t 2/-4 t 2 mmHg). In contrast, exercise produced no aftereffect on the blood pressure of normotensive subjects. A moderate but significant tachycardia was detected both in hypertensive (+5 t 1 beats/min) and normotensive subjects (+6 t 1 beats/min) after exercise compared with control. In hypertensive subjects, FBF increased by 34 t 11% after exercise, whereas it remained similar in normotensive subjects. FVR decreased by 25 t 6% after exercise compared with control in hypertensives, whereas it was unchanged in normotensives.
Table 1. Characteristics of hypertensive and normotensive subjects participating in the study Variable
n Age, yr Weight, kg Height, cm VO ml. kg-- ’ . min- l Cardiac mass, g/m” LVIDD, mm Systolic pressure, mmHg Diastolic pressure, mmHg Heart rate, beats/min FBF, ml 100 gg’ smin-’ FVR, units 2
IlilX9
Hypertensives
13 44&Z 75t3 175t2 35tl 87k6 5022 14e4* 95t2* 59tl 1.8&0.1* 64.6t2.7*
Normotensives
9 41k2 73+6 173t3 36+3 89t5 51+1 106+2 77t3 58t4 2.4k0.2 37.2t2.7
Values are means f. SE. VO, tn;,x, maximal oxygen uptake; LVIDD, left ventricular internal diameter in diastole; FBF and FVR, forearm blood flow and forearm vascular resistance. * P c 0.05 vs. normotensives.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
H1526
EXERCISE
AND
REFLEX
CONTROL
OF FOREARM
Table 2. Cardiovascular variables and work load during the 30-min period of exercise in hypertens ive and no rmoltensive subjects Variable
Hypertensives
Normotensives
Heart rate, beats/min 128k5 Systolic pressure, mmHg 170t5* Diastolic pressure, mmHg 90t2* Mean arterial pressure, mmHg 117t4” Work load, W 84k6 53&l %G tni\x Values are means f: SE. %i70L! tlIRX, percent of maximal take. * P < 0.05 vs. normotensives.
133xk4 138k4 68k4 91t4 91tll 52tl oxygen
up-
150 140
a CONTROL 69 AFTER EXERCISE
130 =120 0 M E 110 I
g100 90 80 70 i
-
80 70
E 3.0 -' ii0 LJ- g
[L s60 > 'E 50 k3 -40
2.5 2.0
9
1.5
30
*
1.0
20
-
--
I-II
I I
NT
I
Fig. 1. Blood pressure (BP), heart rate (HR), forearm blood flow (FBF), and forearm vascular resistance (FVR) during control evaluation and after exercise in hypertensive (HT) and normotensive (NT) subjects. In the BP panel, systolic and diastolic values correspond to top and bottom of each bar, respectively, and mean arterial pressure is indicated within each bar. * P < 0.05 vs. control.
Identical hemoglobin concentrations were found after exercise and during the control evaluation in either group (144 t 3 g/l), while no differences were found in hematocrit in hypertensive (42 t 1 and 41 t 1%) and normotensive subjects (43 t 1 and 42 t 1%). The calculated changes in plasma volume were not significant in hypertensives (-0.9 t 1.7%) and in normotensives (-0.4 t 1.0%) after exercise compared with control. Estimated CVP also did not vary significantly after exercise compared with control in hypertensives (-0.2 t 0.2 mmHg; NS), but decreased in normotensives (-0.6 t 0.2 mmHg; P < 0.05). LVIDD was unchanged in hypertensives (+0.9 t 0.8 mm; NS), but increased slightly in normotensives (+l.l t 0.4 mm; P c 0.05) after exercise compared with control. Response to maneuvers. Figure 3 shows that LR induced a significant increase in estimated CVP, as well as in LVIDD, while LBNP15 and LBNP40 induced graded falls in estimated CVP and in LVIDD. The changes in estimated CVP and in LVIDD were similar during the control evaluation and after exercise (see also Fig. 2). In addition, the relationship between CVP and LVIDD was similar in normotensive and hypertensive subjects. In the group of subjects in which both measurements were per-
VASCULAR
RESISTANCE
formed (n = 14, 8 hypertensive and 6 normotensive subjects), CVP was linearly related to LVIDD by the equation LVIDD (mm) = 1.33 [estimated CVP (mmHg)] + 50.0. LR and LBNPl5 did not induce any significant change in blood pressure or heart rate (data not shown). In contrast, LBNP40 induced a significant decrease in systolic blood pressure in hypertensive (-9 t 1 and -10 t 3 mmHg, during the control evaluation and after exercise, respectively) and in normotensive subjects (-6 t 1 and -7 t 1 mmHg). This was associated with reflex tachycardia both in hypertensive (+13 t 2 and +l3 t 2 beats/ min) and normotensive subjects (+l7 t 2 and +20 t 4 beats/min). Thus arterial baroreceptors were additionally deactivated during this maneuver. LR induced a decrease in FVR, while graded LBNP induced progressive increases in FVR in both groups of subjects. The FVR-estimated CVP and FVR-LVIDD relationships are illustrated in Fig. 4. Statistical comparison of control and postexercise regression lines of FVRestimated CVP and FVR-LVIDD relationships yielded identical conclusions whether the data from LBNP40 was included or not with the data from LBNP15 and LR. For clarity, only the relationships taking into account the data from LBNP40 will be presented. The FVR-estimated CVP relationship was shifted downward after exercise compared with control in hypertensive subjects (at estimated CVP = 0, FVR = 55.0 t 8.4 and 70.4 t 6.5 units, respectively; P < 0.05), whereas no significant change was noted in normotensive subjects (at estimated CVP = 0, FVR = 40.3 t 6.2 and 45.2 t 5.0 units; NS). The slopes were not significantly different after exercise compared with control in hypertensive subjects (-5.21 t 0.76 and -4.18 t 0.64 units/mmHg, respectively) or in normotensive subjects (-3.79 t 0.61 and -4.49 t 0.56 units/mmHg). Similar findings were made concerning the slopes of the FVR-LVIDD relationships after exercise compared with control in hypertensive (-4.78 t 0.87 and -3.80 t 0.93 units/mm; NS) and in normotensive subjects (-2.67 t 0.79 and -2.68 t 0.86 units/mm; NS). However, as for the FVR-estimated CVP relationship, the FVR-LVIDD relationship was shifted downward after exercise compared with control in hypertensive (at LVIDD = 50, FVR = 53.1 t 7.9 and 65.8 t 6.9 units; P < 0.05) but not in normotensive subjects (at LVIDD = 50, FVR = 40.8 t 4.2 and 41.0 t 5.3 units; NS). Cold Pressor Test As shown in Table 3, immersion of one hand in iced water induced a moderate tachycardia and a marked pressor response and increase in FVR that were unaffected by prior exercise in hypertensive as well as in normotensive subjects. Effects of Local Heat Application Heating the forearm induced a rapid rise in skin surface temperature that plateaued within the first 5 min. Mean skin temperature increased up to a stable level of 33.6 t 0.4OC during heating from 32.0 t 0.5”C in the control situation (P < 0.05). Heating was associated with significant (P < 0.05) increases in FBF (2.7 t 0.8 vs. 2.1 t 0.8
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on September 16, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
EXERCISE
AND
REFLEX
CONTROL
OF FOREARM
VASCULAR
RESISTANCE
H1527
A Af=rER
CONTROL
CVP (mm Hg) LVIDD
; -5 I
HR (b.min -1)
45.3
50.2
46.0
m/105
163/100
135195
136193
;; 4.
Hg)
(ml.lOOcc FBF-l.mln * -l) FVR
50.7
(mm)
BP (mm
EXERCISE
m
Ld.n 79.2
86.4
49.2
56.5
BASAL
lBNP15
BASAL
LBNP15
(Units)
B CONTROL
Af7ER
EXERCISE
Fig. 2. Original tracings from a hypertensive (A) and a normotensive (B) subject showing mean central venous pressure (CVP, measured from a peripheral vein), heart rate (HR), and forearm blood flow (FBF) recordings, along with left ventricular internal diameter in diastole (LVIDD), blood pressure (BP), and calculated forearm vascular resistance (FVR) at baseline and during lower body negative pressure at -15 mmHg (LBNP15) during control evaluation and af-
LLduuLld
LVIDD
50.0
(mm)
HR (b.min -1) BP (mm
ii 3
---C./---L
Hg)
(Units)
49.6
47.3
9417-I
98/69
95166
56.2
41.4
50.4
7I------94169
(ml.lOOcc FBF-l.mln - -l) lP--c-d FVR
46.9
100
LL-iYkY 42.9
BASAL
LBNP15
BASAL
ml.100 ml-‘.rnin-‘) and decreases in FVR (39.7 t 11.2 vs. 55.3 t 14.2 units) compared with the control situation, respectively. The changes in baseline levels were similar to, i.e., not significantly different from, those found in the same four subjects after exercise compared with the nonexercise control evaluation (FBF, 3.0 t 0.7 vs. 1.8 t 0.1 ml. 100 ml-’ emin-‘; FVR, 41.0 t 7.9 vs. 63.0 t 3.2 units; both P < 0.05; see also mean group values, Fig. 1). However, reflex FVR responses during LR and LBNP15 were different after exercise and during heating. Figure 5 illustrates that the reflex changes in FVR were unaffected after exercise but decreased significantly during heating. DISCUSSION
In agreement with previous studies in hypertensive subjects, the present results indicate that a statistically and clinically significant antihypertensive period was observed after mild physical exercise (2, 13, 18, 34). In our study, a control group of normotensive subjects was additionally evaluated and no significant change in blood pressure was found after exercise. Comparison of the effects of exercise on the baroreflex control of FVR in hypertensive and normotensive subjects therefore allows
LBNPlS
identification of alterations that are associated with the reduced blood pressure. The results of this study show that the baroreflex regulation of FVR is modified during the postexercise period in hypertensive subjects. During maneuvers that increased and decreased venous return, the relationships between FVR and estimated CVP, as well as between FVR and LVIDD, were shifted downward, i.e., to lower FVR values, after exercise in hypertensive subjects. This was not simply related to a lower mean arterial blood pressure. In fact, the increase in baseline FBF (~30%) was more pronounced than the fail in blood pressure (
