Journal of Human Hypertension (2009) 23, 267–273 & 2009 Macmillan Publishers Limited All rights reserved 0950-9240/09 $32.00 www.nature.com/jhh ORIGINAL ARTICLE Exercise training favourably affects autonomic and pressure responses during mental and physical stressors in African-American men

V Bond1, MN Bartels2, RP Sloan3, RM Millis4, AS Zion5, N Andrews6 and RE De Meersman2,5 1Department of HHPLS, Howard University, Washington, DC, USA; 2Department of Rehabilitation Medicine, Columbia University, New York, NY, USA; 3Department of Psychiatry, Columbia University, New York, NY, USA; 4Department of Physiology and Biophysics, Howard University, Washington, DC, USA; 5Department of Biobehavioral Sciences, Teachers College, Columbia University, New York, NY, USA and 6Department of Cardiology, Howard University, Washington, DC, USA

Aerobic exercise is a powerful mechanism by which compared with a matched sedentary control group, cardiovascular and autonomic parameters may be im- aerobic capacity of the trained group significantly proved. We sought to quantify the extent of benefit that increased by 16%. Autonomic modulations, arterial com- could be achieved by a short-term monitored exercise pliance and BP responses significantly improved during regimen on several autonomic parameters during recog- some of the stressors, whereas no such improvements nized mental and physical stressors in young normoten- were seen in the control group. Attenuated responses, sive African-American men matched for a family history of mediated through a favourable shift in sympathovagal hypertension, a group at high risk for the development of balance and enhanced arterial compliance, provide hypertension. Autonomic modulations were derived mechanistic evidence of how certain variables may be using spectral decomposition of the electrocardiogram improved due to aerobic conditioning in a population at and beat-to-beat blood pressures (BPs). Arterial compli- high risk for the development of hypertension. ance was obtained using contour analysis of the radial Journal of Human Hypertension (2009) 23, 267–273; artery pulse wave. The analysis of variance revealed that doi:10.1038/jhh.2008.125; published online 9 October 2008

Keywords: aerobic training; stress reactivity; autonomic responses; blood pressure; African-American men

Introduction and BP have been documented following endurance training in animals and humans.8–12 Our prior study It has been hypothesized that repeated mental and/ has demonstrated an augmentation in parasympa- or physical excitation of the sympathetic nervous thetic or vagal modulation and enhanced baroreflex system, incongruent with metabolic needs, plays a sensitivity (BRS) and arterial compliance (AC) dominant role in the establishment and mainte- 13–21 1–4 following aerobic training. Not surprisingly, nance of high blood pressure (BP). To date, few of these studies assessed these effects in stress–response studies have demonstrated that African-American men, not in populations other African-American (AA) men and individuals with than whites. a positive family history of hypertension have a Accordingly, our aim was to determine the extent greater propensity towards augmented BP reactivity 4,5 of benefit that could be achieved in young normo- compared with non-AA men. These augmented tensive AA men, matched for a family history of BP responses appear to be mediated predominantly 5–7 hypertension. Specifically, our hypotheses were that through changes in autonomic modulation. following aerobic training, the trained group would Favourable modifications in autonomic modulation have an improved autonomic modulation profile, an enhanced arteriolar compliance and attenuated BP responses when compared with the non-trained Correspondence: Dr RE De Meersman, Rehabilitation Medicine, control group. These responses were compared Columbia University, 630 West 168th Street, Box 38, NYC, NY, during physical and mental stressors before and 10032, USA. E-mail: [email protected] following an aerobic training programme in Received 18 May 2008; revised 4 September 2008; accepted an experimental (exercise-trained) and control 6 September 2008; published online 9 October 2008 (sedentary) group. In addition to both groups being Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 268 matched on body habitus, we also controlled for an training programme or no family history of hypertension, a largely ignored exercise programme (sedentary lifestyle). confounder, yet potent mediator, of exaggerated stress responses.22 Study design Participants were scheduled to the laboratory four Methods times, at which time exercise and autonomic tests Participants were obtained. Visit 1: maximal-performance ex- Thirty-two healthy, normotensive AA men between ercise test using a cycling protocol was employed. 18 and 26 years of age met the inclusion criteria for Visit 2: autonomic tests: cold pressor, colour conflict participation and were enrolled in the study, which and deep breathing. Visits 3 and 4 were identical was approved by the University’s Human Partici- repeats of visits 1 and 2 following an 8-week aerobic pants Institutional Review Board of The Howard training or sedentary interval. Protocols for visits 1 University Institutional Review Board. The nature, and 2 as well as 3 and 4 were carried out within 1 purpose and risks of the study were explained to week, but at least 48 h apart. All post-testing was each participant before written informed consent carried out between 48 and 96 h following the last was obtained. An initial screening interview indi- training session. Descriptive data on participants are cated that all participants were non-smokers, ab- presented in Table 1. stained from use of alcohol (less than two standard alcohol drinks a day), took no prescription medica- tions, did not participate in regular exercise during Haemodynamic measures the previous year and were normotensive. Normo- Systolic BP and DBP were measured in the right arm tension was based upon two to three seated BP by a Colin STBP-680 automated BP monitor (Colin measurements during the initial screening inter- Medical Instruments, San Antonio, TX, USA), view. Participants were recruited through advertise- which meets established criteria for SBP and DBP ments placed throughout the university buildings. measures during rest and dynamic exercise.24 Mean All had resting systolic (SBP) and diastolic BP (DBP) arterial BP (MABP) was calculated using the con- within the recognized normotensive-to-pre-hyper- ventional formula: DBP þ 1/3 pulse pressure. tensive BP range of o140/90 mm Hg. As familial rate was determined from a CM5 electrocardio- influence of hypertension is a strong risk modulator graphic tracing by the R-R interval measurement of the autonomic nervous system,22 equal numbers using the Colin STBP-680 monitor. In addition to of both positive and negative family history of these two monitors, participants were instrumented hypertension were included in the study. Partici- with an applanation tonometer (Colin NIBP ton- pants were assigned to either an experimental or a ometer; Colin Instruments) attached to the partici- control group with half of each group having a pant’s left radial artery and supported by an positive and/or negative family history of hyperten- adjustable table to the level of the atria. Following sion. Family history of hypertension was obtained the equilibration period, data were collected and all from the initial interview. When one family member signals, including electrocardiogram and beat-to- was taking antihypertensive medication, this re- beat radial BP waves, were collected at 500 Hz and sulted in the classification of positive family history, channelled into a computer (Dell computers, Aus- whereas no use resulted in a negative classification. tin, TX, USA) through an analog-to-digital conver- The Godin leisure time exercise questionnaire was sion board (ATMIO-16; National Instruments, used to characterize participants’ lifestyle/physical Austin, TX, USA). Following all data-acquisition activity level.23 Data obtained from the activity protocols, autonomic analyses were carried out questionnaire were converted to metabolic equiva- off-line in accordance with the recommendations 25 lents (METs) (1 MET ¼ 3.5 ml O2 per kg per min) and of the European and North American Task Force a weekly score of 20 was defined as ‘sedentary’. using LabView programs (LabView 7.0; National Each group was further randomly assigned to either Instruments).

Table 1 Participant characteristics

2 . Group Age (years) Height (cm) Weight (kg) BMI (kg/m ) Peak VO2 (ml O2 per kg per min)

Pre-treatment Post-treatment

Controls (n ¼ 16) 22.4 (2.2) 179.2 (6.5) 80.5 (12) 25.0 32 (7.1) 33.6 (5.3) Trained (n ¼ 16) 20.4 (1.9) 179.8 (4.7) 78.4 (11) 24.2 31.8 (6) 40 (3.8)*

Abbreviation: BMI, body mass index. Data are means and standard deviations (s.d.). *Significantly different from pre-training values at Po0.05.

Journal of Human Hypertension Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 269 Stress protocols participants performed a stroop CCT of 5 min in Rest: Following equilibration and attachment of duration. The CCT is designed to elicit responses at instruments to the subject, participants continued the cardiac autonomic level. to rest 15 min. Heart rates and beat-by-beat arteriolar Deep breathing: Participants were instructed to BPs were then extracted from the middle 5 min of breathe at six breaths per minute (0.1 Hz) for 5 min resting data (o5% fluctuations in heart rates and while following a visual cue (Labtimer). Throughout BPs). These data were considered resting data and these 5 min, participants’ heart rates and beat-by- entered as such in the analyses. Sufficient time was beat BPs were collected. The deep breathing causes allowed for vital signs to return to resting values large fluctuations in SBP and as such stimulates between stressors. No recording of the recovery baroreflexes. times was made; however, resting values were Arterial compliance measurement: The method recorded and when reached the next test was employed utilized the area under the DBP waveform implemented. of the radial artery, rather than on the waveform Oxygen uptake measurement: During the first itself. We used the well-defined incisura and the laboratory session, peak aerobic fitness of each end of diastole to perform the integration of this participant was assessed by a graded exercise test area,26,27 a valid and reliable estimate of AC.28 on an Ergoline 800S electronic-braked cycle erg- Arterial compliance data used in the analyses were ometer (SensorMedics Corp., Anaheim, CA, USA). normalized for height and . Pre- and post- Each participant began exercising at 30 watts (W) for tests were carried out by the same investigator who 3 min, and the study rate was continually increased was an AA individual. by 30 W every 3 min until volitional exhaustion was reached. Minute ventilation was measured by a pneumotachometer connected to a FLO-1 volume Autonomic analyses transducer module (PHYSIO-DYNE Instrument Parasympathetic or vagal modulation measurement: Corp., Quogue,. NY, USA). Percentage. of expired The digitized electrocardiogram signal was in- oxygen (VO2) and carbon dioxide (VCO2) was spected for artefact correction and rejection follow- measured using paramagnetic O2 and infrared CO2 ing established procedures, and an R-R interval analysers connected to a computerized system series was extracted.29,30 Spectral decomposition of (MAX-1; PHYSIO-DYNE Instrument Corp.). These the R-R series was performed: power spectra within analysers were calibrated. against known medical the 0.04–0.15 Hz range were defined as low-fre- grade gases. The highest VO2 value attained. during quency components, whereas those at the frequency the graded exercise test was considered VO2 peak. of 0.15–0.4 Hz were defined as high-frequency Similar test procedures were carried out at the end components and representative of parasympathetic of. the training programme to determine changes in modulation.30 VO2 peak. Sympathetic modulation measurement: BP peaks Aerobic exercise training protocol: The aerobic were detected through an established peak-detec- exercise training programme consisted of 8 weeks of tion algorithm; the radial pressure waveforms were bicycling three. times per week, 30 min per session, visually inspected for missed systolic peaks. The at 70% of VO2 peak. Warm-up and cool-down power of the systolic peaks was calculated by periods of 3–5 min were allowed before and sub- measuring the area (integration) under the peak of sequent to each training session, respectively. All 24 the spectra. Power spectra within the 0.04–0.15 Hz training sessions were attended by one of the range were defined as low-frequency components investigators, during which they monitored and and representative of sympathetic vasomotor mod- recorded the training progression of all participants. ulation.30 Exercise. intensity (heart rates obtained at 70% of Baroreflex sensitivity assessment: The sponta- VO2 peak) for all training sessions was verified neous and dynamic beat-by-beat interactions of through a heart rate monitor (radial pulse). As SBP and R-R interval reflect true baroreflex events training progresses, maintenance of this heart rate rather than random interactions. These baroreceptor intensity requires increased work. assessments were performed on the identical heart Cold pressor test (foot) (CPFt): Participants per- rate and BP epochs that were spectrally analysed. formed a CPFt after rest. The right foot was placed in Our method is based on the calculation of the a container of ice water that was kept at a constant modulus of the transfer function between SBP and temperature of 41C for 4 min. During these 4 min, pulse interval powers (a-index) in the low-frequency participants were instructed to avoid gasping and to (LF, near 0.1 Hz) bandwidth.31 maintain a normal breathing pattern. This 4-min epoch was extracted and used for analyses. The CPFt is designed to elicit exaggerated sympathetic re- Statistical analyses sponses primarily affecting vascular responses and to obtain sufficient low-frequency modulation data. Statistical analyses were conducted using SPSS Stroop colour conflict test (CCT): Once heart version 11.0. Results are presented as means and rates and BPs returned to their pre-test values, s.d. for each task. A two (group: control versus

Journal of Human Hypertension Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 270 experimental) by two (time: pre- and post-treatment) Results by four (tasks: baseline, cold pressor, colour conflict and deep breathing) repeated measures analysis of By design, there were no differences in body variance was performed for each physical variable habitus, autonomic and cardiovascular variables with the four conditions (baseline, cold pressor, between the control and trained groups at baseline. colour conflict and deep breathing) as the repeated The trained group significantly improved by 16% measures factor. Newman–Keuls post hoc compar- compared with their pre-training values ± isons were used to identify significant differences (x¯ ¼ 33.6 5.3 ml O2 per kg per min) and. post- among means. All post hoc tests were performed training (x¯ ¼ 40±3.8 ml O2 per kg per min) VO2 peak using an a-level of 0.05. All analyses were carried out (Po0.01).. No changes were seen between the pre- by an independent technician who was blinded to all and post-VO2 peak values of the control group. The testing conditions. In addition, a second technician results between the pre- and post-values following 8 analysed a random third of all files and comparisons weeks of exercise for AC, parasympathetic modula- between analyses were carried out. The coefficient of tion (RRHF), sympathetic modulation (SBPLF) and variation for all dependent variables between the two a-index for low-frequency modulation are presented analyses was o3%. Throughout all protocols, breath- in Table 2. The cardiovascular variables, heart rate ing rates were derived and compared using an and SBP, are presented in Table 3. As expected, established and validated method,32 and no signifi- cardiovascular and autonomic parameters improved cant differences were found in breathing rates in the group who trained. No significant differences between the pre- and post-tests for any of the were seen in any of the dependent variables protocols. All post hoc tests (Newman–Keuls) were for the pre-testing conditions between the two performed using an a-level of 0.05. groups.

Table 2 Mean and s.d. in normalized units (nu) for autonomic variables between pre- and post-treatment for controls and trained groups for rest, CPFt, CCT and BRTH

RRHF (nu) SBPLF (nu) a-index (LF)

Pre-treatment Post-treatment Pre-treatment Post-treatment Pre-treatment Post-treatment

Trained Rest 29 (14) 31 (13) 70 (7) 67 (12)* 8 (3) 14 (3)* CPFt 28 (16) 27 (17) 57 (13) 53 (17)* 9 (7) 16 (4)* CCT 22 (9) 25 (8)* 54 (12) 54 (12) 8 (4) 13 (7)* BRTH 11 (9) 14 (10)* 88 (6) 84 (16)* 13 (7) 18 (9)*

Controls Rest 31 (11) 29 (10) 69 (16) 70 (12) 9 (7) 10 (14) CPFt 28 (16) 29 (17) 54 (19) 53 (18) 7 (5) 8 (9) CCT 24 (7) 23 (8) 50 (14) 52 (13) 9 (5) 7 (11) BRTH 10 (6) 11 (13) 86 (23) 88 (20) 14 (6) 12 (5)

Abbreviations: BRTH, breathing protocol; CCT, colour conflict test; CPFt, cold pressure test. *Significantly different from pre-training values at Po0.05.

Table 3 Mean and s.d. for arterial compliance, SBP and HR between pre- and post-treatment for controls and trained groups for rest, CPFt, CCT and BRTH

Arterial compliance (mm Hg s) Heart rate (b.p.m.) Blood pressure (mm Hg)

Pre-treatment Post-treatment Pre-treatment Post-treatment Pre-treatment Post-treatment

Trained Rest 6 (4.4) 6 (5.8) 68 (6.2) 67 (9.8) 118 (17) 118 (16) CPFt 4.6 (3) 6 (3)* 86 (14) 82 (16) 143 (21) 134 (24)* CCT 4 (1.6) 5 (2)* 80 (12) 72 (11)* 139 (20) 133 (25)* BRTH 5 (2.5) 4.7 (3.1) 71 (6.6) 71 (9.1) 114 (14) 113 (17)

Controls Rest 5 (3.4) 5.7 (4.7) 67 (9.5) 68 (8.3) 122 (23) 123 (15) CPFt 4.3 (4) 3.8 (2.5) 82 (14) 84 (16) 150 (21) 153 (15) CCT 3.8 (1.4) 3.6 (1.9) 80 (9.2) 81 (8.7) 143 (17) 145 (19) BRTH 5 (2.2) 4.1 (1.7) 70 (10) 71 (8) 115 (13) 114 (8)

Abbreviations: BRTH, breathing protocol; CCT, colour conflict test; CPFt, cold pressure test; HR, heart rate; SBP, systolic blood pressure. *Significantly different from pre-values at Po0.05 level.

Journal of Human Hypertension Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 271 Discussion cold pressor and the CCT when compared with the pre-training testing protocol. No attenuation in The major finding in this investigation demonstrates sympathetic modulation was seen between pre- that significant improvements in aerobic capacity testing and post-testing for any of the stressors in improve autonomic and non-autonomic BP control the control group. In addition, SBP rises during the systems. Specifically, autonomic modulation, baror- cold pressure test were significantly less for the eflex sensitivity and AC were favourably altered post-testing condition in the trained group com- in young AA men during some of the physical pared with the pre-testing condition. These observa- and mental stressors. As respiration has been tions support the notion that improvements in shown to modify the amplitude of heart rate aerobic capacity following endurance training sig- and BP variability, and is frequently ignored, we nificantly buffer sudden elevations in BP that occur verified that no differences were present in respira- during mental and physical stressors. Furthermore, tory rates between and within groups during pre- 33,34 our results indicate that this attenuated reactivity is and post-tests for any of the stressors. Prior in part due to a decrease in sympathetic modulation. investigations on endurance training have been 35 Taken together, our findings demonstrate a signifi- shown to improve autonomic modulations and cant and favourable shift (lowering) in the auto- BPs in human and animal populations after a nomic balance following aerobic training. myocardial infarction,36 with borderline hyperten- sion,37 and in patients with chronic heart fail- ure.16,38,39 Therefore, it appears that these Baroreflex sensitivity favourable autonomic changes occur in both healthy The baroreflex sensitivity in the trained group was and diseased populations. significantly improved following the training pro- gramme for all four conditions. These findings of significant improvements in BRS for all conditions Parasympathetic modulation emphasize the importance of baroreflex modulation It has been proposed that the increased prevalence following aerobic training, as the BRS plays a key and severity of hypertension in AAs is due in part to role as a mediator of autonomic responses and an exaggerated reactivity leading to autonomic overall cardiovascular morbidity and mortality.14–18 dysfunction.40,41 Our findings demonstrate improve- Moreover, prior research has shown improved BRS ments ranging from 13% during the CCT to 27% after exercise training in hypertensive partici- during the deep-breathing test in parasympathetic pants.47–49 The improvement in BRS in this inves- modulation following a 16% improvement in aero- tigation could be in part due to a lesser sympathetic bic capacity in the trained group compared with the modulation and a well-known mediator of vascular matched control group, in which no such improve- stiffness.20,50,51 ments were seen. These improvements demonstrate less of a loss of cardiac parasympathetic activity Arterial compliance during mental and physical stressors. These im- Loss of AC during sympathetic stimuli leads to provements were seen as well in heart rates that exaggerated increases in an a2-adrenergic-mediated were significantly lower following the training vasoconstriction and/or decreases in b2-adrenergic- period during the CCT test in the trained. Together, mediated vasodilation, both of which contribute to these findings strongly suggest the presence of an increases in arterial stiffness. Augmented and/or attenuated heart rate-reactivity response following exaggerated sympathetic activity has been shown to aerobic training, which may be explained by a lesser further reduce distensibility of small, medium and withdrawal of parasympathetic modulation during large arteries,49–53 and is a common denominator for mental (CCT) and physical (CPFt and breathing risk factors related to hypertension.54–56 Our prior protocol) stressors. study, and that of others, has. shown that a strong relationship exists between VO2 peak and AC, even among older individuals who take part in endurance Sympathetic modulation 19–21 Plasma norepinephrine values and microneuro- training. These findings are evident in the graphic techniques have provided convincing evi- present investigation in which AC, in a highly dence for exaggerated sympathetic modulation in sympathetically innervated artery, was significantly hypertension.42 Moreover, this exaggerated response greater following aerobic training during the CPFt also appears to be present in pre-hypertensive and and CCT tests. hypertensive patients with different degrees of BP elevation and of different ages.43,44 Chronic aerobic Integration training has been shown to lower plasma catecho- BP is the product of cardiac output (Q)and lamines and lower BPs in humans and animals.44–46 peripheral vascular resistance. Our results demon- Conceivably, these observations are corroborated in strate that aerobic exercise training confers a protec- the current investigation as SBPLF, a non-invasive tive effect on both autonomic and non-autonomic marker of sympathetic vasomotor modulation, regulatory components during physical and mental was lower in our trained group for the resting, stressors. Specifically, enhanced parasympathetic

Journal of Human Hypertension Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 272 modulation during the CCT has a slowing effect on References heart rates. Moreover, an attenuation of sympathetic modulation during the CPFt may help explain the 1 Brod J, Fend V, Hejl Z, Jorka G. Circulatory changes augmented AC and, as such, a lessening in periph- underlying blood pressure elevation during acute emotional stress (mental arithmetic) in normotensive eral vascular resistance. All of these parameters and hypertensive participants. Clin Sci 1959; 23: clearly suggest that aerobic training confers cardio- 339–349. protection through autonomic and non-autonomic 2 Hollenberg NK, Williams GH, Adams DF. Essential mechanisms. hypertension: abnormal renal vascular and endocrine In summary, 8 weeks of aerobic training at a responses to a mild psychological stimulus. Hyperten- sufficient intensity such that oxygen uptake im- sion 1981; 3: 11–17. proved by 16% resulted in enhanced autonomic 3 Anderson NB, Lane JD, Marunaka M, Willams RB, modulation, baroreflex sensitivity and AC changes Houseworth SJ. Racial differences in blood pressure during several physical and mental stressors. These and forearm vascular responses to the cold face improvements were seen in a group of young stimulus. Psychosom Med 1988; 50: 57–63. 4 Dimsdale JE, Graham RM, Ziegler MG, Zusman RM, normotensive AA men. Finally, these findings Berry CC. Age, race diagnosis, sodium effects on the emphasize the prominent role of regular aerobic pressor response to infused norepinephrine. Hyperten- exercise as a prominent mediator in the regulation of sion 1987; 10: 564–569. autonomic and non-autonomic BP control mechan- 5 Murphy JK, Alpert BS, Moses DM, Somes GW. Race isms in young AA men, a population at augmented and cardiovascular reactivity: a neglected relationship. risk for the development of hypertension. Hypertension 1986; 8: 1075–1083. 6 Sherwood A, Hinderliter AL. Responsiveness to alpha-and beta-adrenergic receptor agonists: effects Limitations of race in borderline hypertensive compared to One of the limitations associated with the study is normotensive men. Am J Hypertens 1993; 6: 630–635. 7 Dimsdale JE, Ziegler M, Mills P, Delehanty SG, Berry C. the small number of subjects completing the study; a Effects of salt, race and hypertension on reactivity to more large-scale intervention would most likely stressors. Hypertension 1990; 16: 573–580. yield more significant results. A closer monitoring 8 Fagard RH. Prescription and results of physical of the physical activity of the sedentary control activity. J Cardiol Pharmacol 1995; 25(suppl 1): S20–S27. group could have provided further verification that 9 World Hypertension League. Physical exercise in the the effects were due solely to the intervention. management of hypertension. Bull World Health Finally, in spite of the a priori control for a family 1991; 69: 149–153. history of hypertension and considering the multi- 10 Veras-Silva AS, Mattos KC, Gava NS, Brum PC, Negrao factorial nature of hypertension, there is no guaran- CE, Krieger EM. Low-intensity exercise training de- tee that there may not have been a history of creases cardiac output and hypertension in sponta- neously hypertensive rats. Am J Physiol 1997; 273: hypertension. H2627–H2631. 11 Duncan JJ, Far JE, Upton SJ, Hagen RD, Oglesby ME, What is known about this topic Blair SN. The effect of aerobic exercise on plasma K Chronic aerobic training has a beneficial effect on blood catecholamine and blood pressure in patients pressure, autonomic modulation and arterial compliance. with mild essential hypertension. JAMA 1985; 254: K Whether this type of training has a beneficial effect in young 2609–2613. African-American men, a population at a greater risk for the 12 Pagani M, Somers V, Furlan R, Dell’Orto S, Conway J, development of hypertension, is uncertain. Baselli G et al. Changes in autonomic regulation What this study adds induced by physical training in mild hypertension. K To our knowledge, this is the first longitudinal study in Hypertension 1988; 12: 600–610. African-American men, in which family history of 13 De Meersman R. Heart rate variability and aerobic hypertension, a potent mediator of stress reactivity, is fitness. Am Heart J 1993; 125: 726–731. controlled for. 14 La Rovere MT, Mortara A, Sandrone G, Lombardi F. K Aerobic exercise training confers cardioprotective effects in Autonomic nervous system adaptations to short-term African-American men during psychological and physiological exercise training. Chest 1992; 101: 299S–303S. stressors. 15 Somers VK, Conway J, Johnston J, Sleight P. Effects of endurance training on baroreflex sensitivity and blood pressure in borderline hypertension. Lancet 1991; 337: 1363–1368. Acknowledgements 16 Coats AJ, Adamapoulos S, Radaelli A, McCance A, Supported by National Institutes of Health, Grant Meyer TE, Bernardi L et al. Controlled trial of physical no. S08016-30 (VB) and VIDDA Foundation (RED) training in chronic heart failure. Circulation 1993; 85: 2119–2131. and RO1 HL61287 (RPS). 17 Pagani M, Somers V, Furlan R, Dell’Orto S, Conway J, Baselli G et al. Changes in autonomic regulation induced by physical training in mild hypertension. Disclosure Hypertension 1988; 12: 600–610. 18 Silva GIJ, Brum PC, Negrao CE, Krieger EM. Acute None. and chronic effects of exercise on baroreflexes in

Journal of Human Hypertension Exercise training and autonomic and blood pressure responses in African-American men V Bond et al 273 spontaneously hypertensive rats. Hypertension 1997; 36 Shindo M, Tanaka H, Ohara S, Tokuyama I. Training

30: 714–719. 55% O2max on healthy middle-aged men by bicycle 19 Vaitkevicius PV, Fleg JL, Engel JH, O’Connor FC, ergometer. Rep Res Cent Phys Educ 1974; 2: 139–152. Wright JG, Lakatta LE et al. Effects of age and aerobic 37 Tipton CM, Matthes RD, Marcus KD, Rowlett KA, capacity on arterial stiffness in healthy adults: inter- Leininger RJ. Influences of exercise intensity, age, and relationships between hypertension, left ventricular medication on resting systolic blood pressure of SHR hypertrophy, and coronary heart disease. Circulation populations. 1983; 55: 1305–1310. 1993; 88: 1456–1462. 38 Winder WW, Beattie MA, Holman RT. Endurance 20 De Meersman R. Aging: a modulator of respiratory training attenuates stress hormone responses to exercise sinus arrhythmia. JGerontolBiolSc1993; 48(2): B74–B78. in fasted rats. Am J Physiol 1982; 243: R179–R184. 21 Gallagher D, Terenzi T, De Meersman R. Autonomic 39 Somers VK, Conway J, Johnston J, Sleight P. effects of outflow in smokers, sedentary and aerobically trained endurance training on baroreflex sensitivity and blood individuals. Clin Auton Res 1992; 2(6): 383–387. pressure in borderline hypertension. Lancet 1991; 337: 22 Grim CE, Luft FC, Weinberger MH. Genetic, familial 1363–1368. and racial influences on BP control systems in man. 40 Anderson N, Lane J, Monou H. Racial differences in Aust NZ J Med 1984; 14: 453–457. cardiovascular reactivity to mental arithmetic. Int J 23 Godin G, Shephard RJ. A simple method to assess Psychophysiol 1988; 6: 161–164. exercise behavior in the community. Can J Appl Sport 41 Clark R. Perceptions of interethnic group racism Sci 1985; 10: 141–146. predict increased vascular reactivity to a laboratory 24 Bond V, Basset Jr DR, Howley ET, Lewis J, Walker AJ, challenge in college women. Ann Behav Med 2000; Swan PD et al. Evaluation of the Colin STBP-680 at rest 22(3): 214–222. and during exercise: an automated blood pressure 42 Grassi G. role of the sympathetic nervous system monitor using R-wave gating. Br J Sports Med 1993; 27: in human hypertension. JHypertens1998; 16: 1979–1987. 107–109. 43 Esler MD, Lambert G, Jennings G. Regional norepi- 25 Task Force of the European Society of Cardiology and nephrine turnover in human hypertension. Clin Exp North American Society of Pacing and Electrophysio- Hypertens 1989; 1(Suppl): 75–89. logy. Heart rate variability, standards of measurement, 44 Valbo AB, Hagbarth KE. Activity from skin mechan- physiological interpretation, and clinical use. Circula- oreceptors recorded percutaneously in awake human tion 1996; 93(5): 1043–1065. participants. Exp Neurol 1968; 21: 270–289. 26 Kannel WB, Wolf PA, McGee DL, Dawber TR, McNa- 45 Ferrier C, Esler MD, Eisenhofer G. Increased norepi- mara P, Castelli PW. Systolic blood pressure, arterial nephrine spillover into cerebrovascular circulation in rigidity, and risk of stroke. The Framingham study. essential hypertension: evidence of high central JAMA 1981; 245: 1225–1229. nervous system norepinephrine turnover. 27 De Meersman RE. New noninvasive computerized Hypertension 1992; 19: 62–69. method for the area measurement of the dicrotic notch. 46 Licht KC, Obrist PA, Sherwood A, James S, Strogatz D. Comp Biol Med 1989; 19(3): 189–195. Effects of race and marginally elevated blood pressure 28 Liu Z, Brinn KP, Yin CPF. Estimation of total on cardiovascular responses to stress in young men. arterial compliance: an improved method and evalua- Hypertension 1987; 10: 555–569. tion of current methods. Am J Physiol 1986; 251: 47 La Rovere MT, Mortara A, Sandrone G, Lombardi F. H588–H600. Autonomic nervous system adaptations to short-term 29 De Meersman RE, Reisman S, Daum M, Zorowitz R, exercise training. Chest 1992; 101: 299S–303S. Leifer M, Findley T. Influence of respiration on 48 Somers VK, Conway J, Johnston J, Sleight P. Effects of metabolic, hemodynamic, psychometric and R-R inter- endurance training on baroreflex sensitivity and blood val power spectral parameters. Am J Physiol 1995; 269: pressure in borderline hypertension. Lancet 1991; 337: H1437–H1440. 1363–1368. 30 Pagani M, Lombardi F, Guzetti S, Rimoldi O, Furlan R, 49 Bengel DH. The dynamic elastic properties of the Pizzinelli P et al. Power spectral analysis of heart rate arterial wall. J Physiol 1961; 156: 458–469. and arterial pressure variabilities and a marker of 50 Cox RH. Effects of norepinephrine on mechanics of sympatho-vagal interactions in man and conscious arteries in vivo. Am J Physiol 1976; 231: 420–425. dog. Circ Res 1986; 59: 178–193. 51 Gerova M, Gero J. Range of sympathetic control of the 31 Bertinieri G, Dirienzo M, Cavallazzi A. Evaluation of dog femoral artery. Circ Res 1969; 24: 349–359. baroreceptor reflex by blood pressure monitoring in 52 Cox RH, Bagshaw RJ. Effects of pulsations on carotid unanesthetized cats. 1981; 254: H377–H383. sinus control of regional . Am J Physiol 32 De Meersman R, Zion AS, Teitelbaum S, Weir JP, 1980; 238: H182–H190. Lieberman JS, Downey JA. Deriving respiration from 53 Boutouyrie P, Lacolley P, Girard X, Beck L, Safar M, pulse wave: a new signal-processing technique. Am J Laurent S. Sympathetic activation decreases medium Physiol 1996; 270: H1671–H1675. sized arterial compliance in humans. Am J Physiol 33 Eckberg DL. Human sinus arrhythmia as an index 1994; 267: H1368–H1377. of vagal cardiac outflow. J Appl Physiol 1983; 54: 54 Masuo K, Mikami H, Ogihara T, Tuck ML. Sympathetic 961–966. nerve hyperactivity precedes hyperinsulinemia 34 Brown TE, Beightol LA, koh J, Eckberg DL. Important and blood pressure elevation in a young, nonobese influence of respiration on human RR-interval power Japanese population. Am J Hypertens 1997; 10: 77–83. spectra is largely ignored. J Appl Physiol 1993; 75(5): 55 Julius S, Esler MD, Randall OR. Role of autonomic 2310–2317. nervous system in mild hypertension. Am Heart 35 De Meersman R. Respiratory sinus arrhythmia altera- J 1975; 48: 243s–252s. tion following training in endurance athletes. Eur 56 Julius S. Sympathetic hyperactivity and coronary risk J Appl Physiol 1992; 64: 434–436. in hypertension. Hypertension 1993; 21: 886–893.

Journal of Human Hypertension