Postexercise hypotension differs between white and black women

Postexercise hypotension differs between white and black women Linda S. Pescatello, PhD,a,b Lynn Bairos,a Jaci L. VanHeest, PhD,a C. M. Maresh, PhD,a N. R. Rodriguez, PhD, RD,a Niall M. Moyna, PhD,c Christopher DiPasquale, MS, PT,a Valerie Collins,a Cherie L. Meckes, MS,d Lori Krueger, MS,d and Paul D. Thompson, MD, FACCd Storrs, New Britain, and Hartford, Conn, and Dublin, Ireland

Background Because data are lacking, we examined the acute effect of exercise on ambulatory blood pressure (BP) in premenopausal white women (n ⫽ 18) and black women (n ⫽ 15) with normal (n ⫽ 21) and high (n ⫽ 12) BP. Methods Women performed 40 minutes of control and moderate-intensity exercise. BP and hormones were measured before, during, and after the control and exercise periods. By means of RMANCOVA (repeated measures analysis of covarience), we tested whether BP and hormones differed with time and between ethnic, BP, and experimental groups. Multiple regression analysis was used to examine hormonal mediators of the postexercise BP response.

Results Among white women with hypertension, average daytime systolic (S) and diastolic (D) BP decreased 11.0 ⫾ 3.3 mm Hg (⫺2.9, ⫺19.1; P ⫽ .017) and 8.2 ⫾ 2.8 mm Hg (⫺1.2, ⫺13.9; P ⫽ .000), from 142.6 ⫾ 5.8 mm Hg and 96.1 ⫾ 2.8 mm Hg, respectively, after exercise. Among black women with high BP, mean daytime SBP rose 12.5 ⫾ 5.2 mm Hg (⫺2.0, 27.1; P ⫽ .000) after exercise, from 121.8 ⫾ 6.1 mm Hg, whereas DBP was similar before and after exercise (81.4 ⫾ 4.3 mm Hg and 82.8 ⫾ 4.7 mm Hg, respectively). In white women without hypertension, daytime SBP and DBP were similar before and after exercise. In black women without hypertension, mean daytime SBP increased 6.3 ⫾ 2.6 mm Hg (0.4, 12.1; P ⫽ .000) after exercise from 103.6 ⫾ 1.4 mm Hg, and DBP did not change. In black women, hypertension (P ⫽ 0.000) and exercise-mediated insulin decreases (P ⫽ .005) explained 85.6% of the postexercise SBP response (P ⫽ .000). In white women, hypertension (P ⫽ .003) and baseline plasma renin (P ⫽ .049) accounted for 53.3% of the postexercise SBP response (P ⫽ .001). Exercise acutely reduced daytime BP in white women, but not in black women with high BP.

Conclusion Endurance exercise may adversely affect the BP of black women. (Am Heart J 2003;145:364-70.)

See related Editorial on page 206.

Hypertension has an earlier age of onset, is more prevalent, and is associated with greater comorbidities and mortality in black Americans than in white Americans. Of all ethnic groups in the United States, hypertension is most common in black women, and these women experience the highest death rates attributable to hypertension.1 Probable reasons for these ethnic disparities include augmented hemodynamic and neurohormone responses to stress,2,3 altered salt sensitiviFrom the aUniversity of Connecticut, Storrs, Conn, bNew Britain General Hospital, New Britain, Conn, cDublin City University, Dublin, Ireland, and dHartford Hospital, Hartford, Conn. Supported in part by the University of Connecticut Office for Sponsored Programs and the New Britain General Hospital Arthur E. Thornton Cardiopulmonary Fund. Submitted January 25, 2002; accepted April 4, 2002. Reprint requests: Linda S. Pescatello, PhD, Assistant Professor and Director, Center for Health Promotion, University of Connecticut, School of Allied Health, 358 Mansfield Rd, U-2101, Storrs, CT 06269-2101. E-mail: [email protected] Copyright 2003, Mosby, Inc. All rights reserved. 0002-8703/2003/$30.00 ⫹ 0 doi:10.1067/mhj.2003.107

ty,4 abnormal vascular reactivity,5-8 and a greater prevalence of being overweight and of physical inactivity9,10 because of interactions between genetic and environmental factors11,12 that remain to be identified. Participation in an endurance exercise training program of low to moderate intensity (40%-70% of maximal oxygen consumption [VO2 max]) decreases blood pressure (BP), with the greatest reductions seen in patients with hypertension.9 Few exercise studies have involved black women. Of those that have, Oluseye13 reported BP reductions in Nigerian women aged 20 to 50 years after a 12-week moderate- to vigorous-intensity endurance training program. In contrast, Headley et al14 found that moderate-intensity walking did not acutely lower the BP of middle-aged African American women who were not taking medication for their hypertension. The immediate and sustained BP-lowering effect of a single session of dynamic exercise, termed “post-exercise hypotension” (PEH), holds promise in the prevention and treatment of hypertension.15-18 The neurohormonal and vascular responses to endurance exercise that produce an immediate and prolonged BP-lowering

American Heart Journal Volume 145, Number 2

effect are not clear.18,19 Sympathetic withdrawal, baroreceptor resetting, and reduced vascular responsiveness to adrenergic receptor activation are postulated to be involved in PEH.18-23 We found that an acute bout of moderate-intensity exercise lowered ambulatory BP for most of the waking hours in unmedicated, middle-aged non-Hispanic white men16 and women17 with high-normal BP to stage I hypertension. The preliminary findings of Headley et al14 suggest that PEH may not occur in African American women as it does in white women with hypertension. The purpose of this study was to investigate whether ethnicity alters the duration and magnitude of PEH among premenopausal women. Because young, healthy black people demonstrate an increased peripheral vascular resistance response to pharmacologic and environmental stimuli,5-8 we hypothesized that the magnitude and duration of PEH would be less in black women than we previously observed in non-Hispanic white women of comparable age. We also sought to identify hormonal mediators of the postexercise BP response.

Methods Subjects Thirty-three premenopausal women between 19 and 45 years of age volunteered to participate. Of these women, 18 were non-Hispanic white and 15 were black. Twelve of the subjects had high-normal BP to stage I hypertension (systolic blood pressure [SBP] ⱖ130-ⱕ159 mm Hg and/or diastolic blood pressure [DBP] ⱖ85-ⱕ99 mm Hg), and 21 subjects had normal BP (SBP ⬍130 mm Hg and DBP ⬍80 mm Hg). Data from the white women have been presented previously17 and are included in this report as a means of comparing the postexercise BP response between ethnic groups. The same study protocols were followed for both groups of women, with the exception of different hormonal assessments and measurement time frames in identifying possible PEH mediators. Self-reported BP status was confirmed by means of multiple BP measurements made during familiarization visits to our laboratory and by means of ambulatory BP monitoring. No woman was taking medication to treat her hypertension or oral contraceptives. All subjects were sedentary, in good health, and had no physical limitations that precluded exercise participation. Their menstrual cycles were regular, as documented by means of menstrual diaries maintained for the duration of study. Subjects signed an informed consent form that was approved by the University of Connecticut and Hartford Hospital or New Britain General Hospital Human Studies Committees.

Experiment design Subjects were familiarized with all study procedures before participation. After resting for 5 minutes, SBP and DBP were measured in each arm, with the patient in the seated position, a minimum of 3 times at least 5 minutes apart on 3 sep-

Pescatello et al 365

arate occasions. Readings were averaged and used to classify the subjects as normotensive or hypertensive (Table I). Weight (kg) and height (cm) were measured, and body mass index (kg/m2) was calculated. Waist circumference was taken at the narrowest portion of the torso between the sternum and iliac crest, and maximal hip (gluteal) circumference was measured across the widest portion of the buttocks. Circumference measures were used as a means of determining the waist-to-hip circumference ratio. After the subjects fasted overnight, we assessed blood lipid-lipoprotein and glucose levels by use of a fingerstick with the Cholestech LDX system enzymatic technology (Cholestech Corporation, Hayward, Calif) to profile the study volunteers metabolically (Table I). Low-density lipoprotein level was estimated by use of the Friedewald equation.24 VO2 max was determined by means of a breath-by-breath analysis of expired gases during testing with an open circuit respiratory apparatus (New Britain General Hospital and the University of Connecticut: Med Graphics Cardiopulmonary Exercise System, St Paul, Minn; Hartford Hospital: Sensormedics Vmax 29 Metabolic Chart, SensorMedics, Yorba Linda, Calif) with an incremental resistance exercise test on a cycle ergometer (Monark Ergomedic 818E, Stockholm, Sweden). The exercise test consisted of continuous cycling at a constant cadence (60 rev/min), with the resistance increased by 0.5 kilopond every 2 minutes until volitional exhaustion after a 2-minute warm up of unloaded cycling. Exercise test termination criteria were: an overall rating of perceived exertion ⱖ18 on a scale ranging from 6 to 2025; a plateau in oxygen uptake; a respiratory exchange ratio ⬎1.1; achievement of age-predicted maximal heart rate; and/or an inability to maintain pedaling frequency. We used the results of the maximal exercise test to calculate the intensity of the experimental exercise sessions from a linear regression plot of heart and work rate as a function of VO2.

Experimental sessions The experiments occurred at the same time of day for each subject, between the hours of 8 and 11 AM, separated by a minimum of 2 days. All subjects were tested during the early follicular phase (days 2-8) of their menstrual cycle, as verified by means of diary recording. Volunteers were asked to refrain from formal exercise for 24 hours before and after the experimental sessions. Subjects consumed a light breakfast without caffeine that totaled approximately 360 Kcal on the morning of the experimental sessions. Otherwise, they were asked not to alter their normal diet. On arrival at the laboratory, subjects were weighed, and an indwelling catheter was inserted into an antecubital vein and kept patent with a solution of heparin and saline. Subjects sat quietly for a 15-minute baseline period. Heart rate was recorded every 4 minutes during the baseline period. SBP and DBP were measured with the subject in the seated position every other minute by means of auscultation in the arm opposite catheter insertion. The same investigator made all BP determinations for a subject. Baseline blood samples were obtained for aldosterone, cortisol, insulin, glucose, lactate, hematocrit, and hemoglobin levels in the black women, and catecholamines, plasma renin, and ␤-endorphin levels in the white women.

American Heart Journal February 2003

366 Pescatello et al

Table I. Cardiovascular risk factors (Mean ⫾ SEM) for the total sample (n ⫽ 33) and by blood pressure status and racial group Blood pressure status

Cardiovascular risk factors Age (y) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Total blood cholesterol (mg/dL) High-density lipoprotein (mg/dL) Total cholesterol/high-density lipoprotein ratio (U) Low-density lipoprotein (mg/dL) Triglycerides (mg/dL) Fasting blood glucose (mg/dL) Waist circumference (cm) Body mass index (kg/m2) Waist-to-hip ratio (U) VO2 (mL/min) VO2 (mL/kg⫺1 䡠 min⫺1)

Normotensive

Hypertensive

Total sample (n ⴝ 33)

Normotensive (n ⴝ 21)

Hypertensive (n ⴝ 12)

White (n ⴝ 11)

Black (n ⴝ 10)

White (n ⴝ 7)

Black (n ⴝ 5)

30.5 ⫾ 1.6 122.5 ⫾ 2.5 76.2 ⫾ 1.7 177.7 ⫾ 5.7 60.3 ⫾ 2.3 3.06 ⫾ 0.13

28.1 ⫾ 2.0 111.6 ⫾ 2.4 70.6 ⫾ 1.3 167.2 ⫾ 6.5 60.1 ⫾ 3.1 2.88 ⫾ 0.15

34.7 ⫾ 2.4* 137.1 ⫾ 2.8‡ 85.8 ⫾ 2.4‡ 195.9 ⫾ 9.0† 60.6 ⫾ 3.7 3.36 ⫾ 0.24

34.1 ⫾ 2.4 111.1 ⫾ 3.4 71.6 ⫾ 2.7 176.6 ⫾ 6.3 64.4 ⫾ 3.6 2.81 ⫾ 0.17

21.5 ⫾ 1.2¶ 112.1 ⫾ 3.4 71.0 ⫾ 0.8 157.0 ⫾ 11.2 55.4 ⫾ 4.9 2.96 ⫾ 0.26

38.1 ⫾ 1.6 136.6 ⫾ 3.7 89.7 ⫾ 2.7 188.9 ⫾ 13.3 57.3 ⫾ 5.0 3.43 ⫾ 0.35

29.8 ⫾ 4.9 137.9 ⫾ 4.4 84.2 ⫾ 4.8 205.8 ⫾ 10.9 65.2 ⫾ 5.5 3.26 ⫾ 0.34

98.2 ⫾ 4.9 95.9 ⫾ 7.6 89.6 ⫾ 3.0 78.3 ⫾ 2.4 26.3 ⫾ 1.0 0.76 ⫾ 0.01 1.58 ⫾ 0.05 22.9 ⫾ 5.9

89.5 ⫾ 5.4 88.1 ⫾ 5.6 85.9 ⫾ 3.2 72.6 ⫾ 2.0 24.0 ⫾ 0.8 0.74 ⫾ 0.01 1.58 ⫾ .06 24.9 ⫾ 1.2

113.4 ⫾ 7.8† 109.6 ⫾ 18.5 96.2 ⫾ 5.8 87.3 ⫾ 4.4† 30.4 ⫾ 1.8† 0.78 ⫾ 0.02 1.58 ⫾ 0.10 19.5 ⫾ 1.4†

95.0 ⫾ 6.0 85.6 ⫾ 7.6 75.6 ⫾ 3.1 71.2 ⫾ 2.9 24.0 ⫾ 1.0 0.73 ⫾ 0.03 1.48 ⫾ 0.09 24.3 ⫾ 2.0

83.4 ⫾ 9.4 90.9 ⫾ 8.6 97.2 ⫾ 3.2¶ 73.8 ⫾ 2.8 24.0 ⫾ 1.2 0.74 ⫾ 0.02 1.68 ⫾ 0.06 25.5 ⫾ 1.5

109.0 ⫾ 10.5 112.7 ⫾ 31.6 84.7 ⫾ 4.2 79.5 ⫾ 4.0† 26.8 ⫾ 1.1 0.76 ⫾ 0.07 1.62 ⫾ 0.14 21.9 ⫾ 1.7

119.6 ⫾ 13.1 105.2 ⫾ 12.0 112.2 ⫾ 8.7㛳 98.3 ⫾ 6.7㛳 35.4 ⫾ 3.0㛳 0.82 ⫾ 0.03 1.53 ⫾ 0.16 16.1 ⫾ 1.3㛳

Data on the white women were published previously.17 VO2, Maximal oxygen consumption. *P ⬍ .05 normotensive versus hypertensive women. †P ⬍ .01 normotensive versus hypertensive women. ‡P ⬍ .001 normotensive versus hypertensive women. 㛳P ⬍ .05 normotensive white versus black women or hypertensive white versus black women. ¶P ⬍ .001 normotensive white versus black women or hypertensive white versus black women.

Each subject completed 2 randomly assigned experimental sessions. The experimental sessions consisted of a control session of 40 minutes of rest that involved quiet upright sitting or 40 minutes of upright cycle exercise. The exercise bout included a 5-minute warm up followed by 30 minutes of cycling at 60% VO2 max and a 5-minute cool-down. Heart rate, SBP, and DBP were measured every 5 minutes during the experimental sessions. Blood samples were obtained after 35 minutes in the control and exercise sessions and immediately after and 15 minutes after the experimental sessions in the black women. Hormones were measured immediately after and 15 minutes after the control and exercise sessions in the white women. SBP, DBP, and heart rate were recorded before blood sampling and at every 5 minutes during the recovery period. After the experimental sessions, subjects were attached to the Accutracker automatic noninvasive ambulatory BP monitor (Suntech Medical Instruments, Raleigh, NC) until the next morning. The same investigator adjusted the ambulatory BP monitor for a given subject, who was in the seated position, until 2 successive measurements agreed within 5 mm Hg of auscultation values taken manually. The monitor was programmed to randomly record SBP, DBP, and heart rate approximately every 20 minutes. All subjects left the laboratory with instructions to proceed with their usual activities, not to engage in formal exercise until the monitor was removed the next morning, and to return the monitor the next day.

Blood sampling and analyses The hematocrit level was determined in triplicate from whole blood by use of the microcapillary technique. The he-

moglobin level was measured in triplicate by use of the cyanmethemoglobin technique (Sigma Diagnostics, St Louis, Mo). Percent plasma volume changes from baseline were calculated from alterations in hematocrit and hemoglobin levels.26 Plasma concentrations of insulin were measured in duplicate by means of radioimmunoassay with a solid-phase, single antibody assay (Coat-a-Count Insulin, TKIN2, Diagnostic Products Corporation, Los Angeles, Calif). A multilevel control (CON6 Multivalent Control Module, Diagnostic Procedures Corporation) was analyzed during each insulin run as a means of monitoring the accuracy of the tests. This assay has a sensitivity of 1.3 ␮IU/mL and an interassay coefficient of 6.8%. Plasma glucose concentration before, during, and after the experimental sessions was determined by an automated glucose oxidase method (Yellow Springs Instruments, model 2003, Yellow Springs, Ohio). Lactate concentrations were measured by enzymatic reaction simultaneously. Cortisol and aldosterone were measured in duplicate by use of a coated tube radioimmunoassay procedure (DSL2100, Diagnostic System Laboratory, Webster, Tex). These procedures have a sensitivity of 0.3 ␮g/dL and interassay coefficient of 5.1% for cortisol and 10 pg/mL and 6.8% for aldosterone, respectively. We previously described the methodology used as a means of quantifying hormones in the white women.17

Statistical analysis Means plus or minus SEM were calculated on all study parameters and tested for statistical differences between BP status (normotensive and hypertensive) and ethnic group (black and non-Hispanic white) with the independent samples t

American Heart Journal Volume 145, Number 2

Pescatello et al 367

Table II. Mean daytime blood pressure (mean ⫾ SEM) (95% CI) before and after control and exercise by blood pressure status and racial group

Group Hypertensive white (n ⫽ 7) Hypertensive black (n ⫽ 5) Normotensive white (n ⫽ 11) Normotensive black (n ⫽ 10)

Blood pressure parameter

Precontrol (mm Hg)

Postcontrol (mm Hg)

Preexercise (mm Hg)

Postexercise (mm Hg)

Systolic㛳 Diastolic‡

146.7 ⫾ 5.0 (134.4-159.0) 140.4 ⫾ 4.4* (129.5-151.2) 142.6 ⫾ 5.8 (128.5-156.7) 131.6 ⫾ 3.8* (122.4-140.7) 99.0 ⫾ 3.0 (91.6-106.4) 93.3 ⫾ 2.7* (86.6-100.0) 96.1 ⫾ 2.8 (89.3-103.0) 88.0 ⫾ 2.7† (81.3-94.6)

Systolic Diastolic

120.0 ⫾ 3.6 (109.9-130.1) 135.8 ⫾ 4.1† (124.5-147.2) 121.8 ⫾ 6.1 (104.9-138.7) 134.4 ⫾ 4.0† (123.4-145.4) 81.2 ⫾ 3.1 (72.5-89.9) 80.6 ⫾ 2.8 (72.8-88.4) 81.4 ⫾ 4.3 (69.4-93.4) 82.8 ⫾ 4.7 (69.9-95.8)

Systolic Diastolic

108.6 ⫾ 4.0 (99.8-117.5) 73.5 ⫾ 2.5 (67.9-79.2)

Systolic Diastolic‡

106.6 ⫾ 2.2 (101.5-111.7) 114.7 ⫾ 1.3† (111.6-117.7) 103.6 ⫾ 1.4 (100.5-106.8) 109.9 ⫾ 3.1† (103.0-116.8) 69.9 ⫾ 3.0 (63.2-76.6) 70.4 ⫾ 1.2 (67.7-73.1) 68.7 ⫾ 1.8 (64.7-72.7) 67.4 ⫾ 1.5 (64.0-70.7)

112.6 ⫾ 2.8 (106.4-118.7) 71.9 ⫾ 2.5 (66.4-77.4)

110.5 ⫾ 2.9 (104.0-116.9) 112.8 ⫾ 3.3 (105.5-120.0) 73.0 ⫾ 2.4 (67.7-78.3) 73.7 ⫾ 2.8 (67.4-80.0)

Data on the white women were published previously.17 *P ⬍ .05 pre- versus postexperiment. †P ⬍ .001 pre- versus postexperiment. ‡P ⬍ .05 BP change control versus exercise. 㛳P ⬍ .001 BP change control versus exercise.

test. RMANCOVA (repeated measures analysis of covarience) was used to determine whether BP and humoral factors differed with time and between experimental conditions (control and exercise), BP status (hypertensive and normotensive), and ethnic group (African American and non-Hispanic white). Covariates included age, overall and central adiposity, and VO2 max. When significant main effects, their interactions, or both were found, we used RMANCOVA to determine whether BP and humoral factors differed more before than after the control and exercise sessions for an ethnic and BP group. When significant effects were noted, 2-tailed paired t tests were used to determine at what points BP and hormones were different before, compared with after, control and exercise sessions. Stepwise multiple regression analyses were performed to determine which humoral substances related to the BP changes that occurred after control and exercise sessions. We performed all statistical analyses with the Statistical Package for Social Sciences MS Windows release 10.0.7,26 and a P value ⬍.05 was established as the level of significance.

Results The subjects in the study sample were overweight and sedentary, with an average age of (mean ⫾ SEM) 30.5 ⫾ 1.6 years (Table I). The non-Hispanic white women were older (35.7 ⫾ 1.6 vs 24.3 ⫾ 2.0 years; P ⫽ .000) and had lower mean fasting blood glucose level (79.1 ⫾ 2.7 vs 102.2 ⫾ 3.9 mg/dL; P ⫽ .000) than the black women. Other parameters did not differ between the ethnic groups. The women with highnormal BP to stage I hypertension were older, heavier, and displayed less favorable lipid profiles than the women with normal BP. The black women with normal BP were younger and had elevated fasting blood

glucose levels compared with their white counterparts. The black women with hypertension had more overall and abdominal adiposity, were less fit, and had higher fasting blood glucose levels than the black women with elevated BP.

Blood pressure In the white women with high BP, the mean SBP and DBP decreased 11.0 ⫾ 3.3 mm Hg (⫺2.9, ⫺19.1; P ⫽ .017) and 8.2 ⫾ 2.8 mm Hg (⫺1.2, ⫺13.9; P ⫽ .000) from pre-exercise averages of 142.6 ⫾ 5.8 mm Hg and 96.1 ⫾ 2.8 mm Hg, respectively, during the day after exercise (Table II). Average daytime SBP and DBP were also lowered 6.3 ⫾ 3.1 mm Hg (⫺13.9, 1.2; P ⫽ .022) and 5.7 ⫾ 1.2 mm Hg (⫺8.7, ⫺2.6; P ⫽ .033) from 146.7 ⫾ 5.0 mm Hg and 99.0 ⫾ 3.0 mm Hg, respectively, after the control session. Daytime SBP (P ⫽ .003) and DBP (P ⫽ .014) were reduced to a greater extent in the exercise session than in the control session in the white women with hypertension (Table II). In contrast, SBP increased and DBP was similar for the remainder of the day after the exercise and control sessions in the black women with high-normal BP to stage I hypertension (Table II). Mean daytime SBP rose 12.5 ⫾ 5.2 mm Hg (⫺2.0, 27.1; P ⫽ .000) from a baseline of 121.8 ⫾ 6.1 mm Hg after exercise. After the control session, average daytime SBP was greater by 15.8 ⫾ 5.1 mm Hg (1.7, 29.9; P ⫽ .000) compared to a control value of 120.0 ⫾ 3.6 mm Hg in the black women with elevated BP. SBP and DBP were similar after the exercise and control sessions for the remainder of the day in the white women with normal BP (Table II). In contrast,

American Heart Journal February 2003

368 Pescatello et al

Table III. Mean humoral responses before, during and after exercise (mean ⫾ SEM) between normotensive (n ⫽ 8) and hypertensive (n ⫽ 5) black women Variable Aldosterone† (␮mol/L) Cortisol (␮mol/L) Insulin† (␮IU/mL) Glucose (mmol/L) Lactate† (mmol/L) Plasma volume (% change)

Blood pressure status Normotensive Hypertensive Normotensive Hypertensive Normotensive Hypertensive Normotensive Hypertensive Normotensive Hypertensive Normotensive Hypertensive

Baseline

Peak

Immediate post

15 minutes post

330.4 ⫾ 42.0 259.2 ⫾ 42.4 381.3 ⫾ 82.5 259.1 ⫾ 106.7 28.2 ⫾ 8.0 17.9 ⫾ 9.7 86.0 ⫾ 5.0 84.3 ⫾ 42 1.7 ⫾ 0.3 2.4 ⫾ 0.4

681.3 ⫾ 149.2* 493.0 ⫾ 77.9* 353.1 ⫾ 79.6 234.7 ⫾ 91.5 11.3 ⫾ 4.2* 7.5 ⫾ 2.4* 76.6 ⫾ 3.0 76.2 ⫾ 3.4 2.4 ⫾ 0.4* 3.1 ⫾ 0.5* –7.4 ⫾ 1.9 –6.6 ⫾ 0.5

645.3 ⫾ 153.5* 439.1 ⫾ 70.0* 356.0 ⫾ 80.4 266.2 ⫾ 110.7 10.1 ⫾ 2.3* 8.5 ⫾ 2.2* 78.8 ⫾ 3.2 75.0 ⫾ 3.6 2.1 ⫾ 0.4* 2.6 ⫾ 0.6* –7.9 ⫾ 1.3 –6.0 ⫾ 1.3

479.7 ⫾ 114.5 320.6 ⫾ 64.6 301.0 ⫾ 60.4 230.0 ⫾ 76.4 14.2 ⫾ 4.3 6.0 ⫾ 1.7 78.9 ⫾ 2.5 82.2 ⫾ 4.0 1.7 ⫾ 0.3 2.0 ⫾ 0.4 –3.0 ⫾ 1.3 –0.0 ⫾ 1.9

*P ⬍ .05 versus baseline. †P ⬍ .05 exercise versus control.

in the black women with normal BP, mean daytime SBP increased by 6.3 ⫾ 2.6 mm Hg (0.4, 12.1; P ⫽ .000) from 103.6 ⫾ 1.4 mm Hg after exercise. SBP was also greater by 8.1 ⫾ 2.7 mm Hg (2.0, 14.1; P ⫽ .000) compared to 106.6 ⫾ 2.2 mm Hg during the day after the control session. In the black women with normal BP, daytime DBP was similar after the exercise and control sessions.

Humoral factors Table III contains the humoral response before, during, and after exercise for the black women. In the black women, changes in aldosterone (P ⫽ .000), lactate (P ⫽ .003), and insulin (P ⫽ .023) levels differed between the exercise and control sessions, regardless of BP classification. Aldosterone and lactate levels increased during exercise compared with baseline, from 259.2 ⫾ 42.4 ␮mol/L to 493.0 ⫾ 77.9 ␮mol/L (P ⫽ .002) and 1.6 ⫾ 0.2 ␮mol/L to 2.5 ⫾ 0.4 mmol/L (P ⫽ .002), respectively, and returned to pre-exercise levels after exercise. In contrast, aldosterone and lactate levels did not change after the control session in the black women (P ⬎ .05). Insulin level was decreased to a greater extent during the exercise session than during the control session, from 24.9 ⫾ 6.7 ␮IU/mL to 12.6 ⫾ 4.6 ␮IU/mL versus 21.8 ⫾ 6.0 ␮IU/mL to 19.4 ⫾ 9.8 ␮IU/mL (P ⫽ 0.045), respectively. The alterations in cortisol, glucose, and percent plasma-volume change were similar before and after the exercise and control sessions (P ⬎ .05) in the black women. The hormonal data for the white women have been presented previously.17 In brief, catecholamines, plasma renin, and ␤-endorphin levels increased with exercise and returned to baseline levels after exercise in the white women, regardless of BP classification.

In the black women, the presence of high BP (␤ ⫽ 0.918, T ⫽ 8.291, P ⫽ .000) and decreases in insulin level during exercise (␤ ⫽ 0.398, T ⫽ 3.590, P ⫽ .005) explained 85.7% of the daytime SBP response after exercise (P ⫽ .000). Similarly, the presence of hypertension (␤ ⫽ 0.761, T ⫽ 3.843, P ⫽ .003) and reductions in insulin level during exercise (␤ ⫽ 0.361, T ⫽ 1.823, P ⫽ .098) explained 54.2% of the variance in the daytime DBP response after exercise (P ⫽ .008). In the white women, BP classification (␤ ⫽ 0.599, T ⫽ 3.528, P ⫽ .003) and baseline plasma renin levels (␤ ⫽ ⫺0.364, T ⫽ ⫺2.140, P ⫽ .049) accounted for 53.3% of the variation in SBP during the day after exercise (P ⫽ .001).

Discussion The major finding of this study was that PEH occurred in the premenopausal non-Hispanic white women with high-normal BP to stage I hypertension, but not their black counterparts. SBP and DBP were lower during the day after exercise in the white women with elevated BP, whereas average daytime SBP and DBP increased after exercise in the black women with hypertension. Exercise did not acutely lower the BP of the white women with normal BP. In contrast, SBP rose for the remainder of the day after exercise in the black women with normal BP, a response similar to that seen in the black women with hypertension. Statistical adjustment for age, fitness level, adiposity, and/or baseline BP did not explain the differences in the acute exercise BP response that we observed between the white women and black women. A 12-week endurance-exercise training program reduced BP in Nigerian women with normal to high-normal BP,13 but

American Heart Journal Volume 145, Number 2

few other studies have been designed to examine the effect of aerobic exercise on BP in black women.14 We did not investigate the influence of exercise training on BP. Our results and those of Headley et al14 suggest that the BP response to acute exercise differs between white women and black women. The immediate beneficial effects of endurance exercise on BP that we observed in white women were not apparent in the black women with high-normal BP to stage I hypertension. Instead, exercise seemed to adversely affect the BP of the black women with normal and high BP. Because environmental factors that influence BP vary considerably between the Nigerian subjects studied by Oluseye13 and African American subjects27,28 our findings raise the possibility that there may also be differences between white American women and black American women in the BP response to endurance training. Lang et al3 demonstrated that young African American black men with normal BP had a reduction of BP to a lesser extent than their white peers after central sympathoinhibition with clonidine, which suggests that nonadrenergic mechanisms contributed more to BP control in black subjects than white subjects. Perregaux et al6 examined endothelial function with ultrasound scanning in healthy black and white men and women between 20 and 47 years of age. They found that endothelium-dependent vasodilation was impaired in the black subjects compared with the white subjects, regardless of sex. Increased vascular adrenergic vasoconstriction and decreased vasodilation to various stress perturbations because of defects in endothelial function are likely explanations for the diminished BP depressor influence of acute dynamic exercise in the black women with and without hypertension compared with the white women in our study. We found that the presence of diagnosed hypertension and exercise-induced decreases in insulin accounted for a significant portion of the daytime SBP and DBP response in the black women. Our observation is consistent with the prolonged effect that acute exercise has on both BP15-18 and insulin.29,30 Black women are more susceptible to the influence of insulin on resting BP than members of other ethnic groups,31 partially because of defects in endothelial function.2,3,5-8,29 It remains possible that the insulin alterations during exercise were simply reflective of the BP classification of the black women. Nonetheless, our finding that insulin appeared to be a mediator of the post-exercise BP response is intriguing and merits further investigation. In the white women, BP status and baseline plasma renin levels accounted for a significant proportion of the daytime SBP response after exercise, with the women with the highest baseline plasma renin levels manifesting the greatest postexercise decreases in SBP.

Pescatello et al 369

Paulev et al32 suggested that endurance exercise acutely diminishes the vasoconstrictor influence of the renin-angiotensin system on BP for sustained periods of time after exercise. Their findings32 and ours are consistent with the proposition that PEH results from competing neurohormonal influences on vascular tone that remain to be more clearly delineated.18-20 This study was subject to several limitations. The ethnic groups were not matched on characteristics known to influence BP (ie, initial BP, VO2 max, and adiposity); however, we adjusted for these parameters in the statistical analyses. Although the sample size was small, the effect-size estimates were moderate to large (eta-squared 0.262-0.555), which indicates sufficient power (0.872-1.000) to detect differences between the black women and white women. We confirmed the self-reported BP status of our subjects with multiple causal and ambulatory BP determinations performed by the same investigator. During the experimental baseline period of quiet, seated rest, the BP of the black women with high-normal BP to stage I hypertension approached normal levels. Misclassification of the BP status of the black women with elevated BP remains a possibility, but is unlikely. The average daytime BP of the black women with high-normal BP to stage I hypertension after the experiments was consistent with their BP status designation. Subjects were given explicit instructions not to alter their typical diets or engage in any formal exercise for 24 hours before and after the experimental sessions. Despite our efforts to control for behaviors that could affect BP, the women may have engaged in activities after the experiments that differentially affected their BP more than an acute endurance exercise bout. In conclusion, exercise acutely reduced BP for a major portion of the day in premenopausal white women, but not in black women with high-normal BP to stage I hypertension. Indeed, daytime BP rose after acute exercise in the black women with high and normal BP. Exercise-mediated decreases in insulin appear to be partially responsible for the postexercise BP response in the black women. Endurance exercise appears to adversely affect the BP of black women. We thank Dr Jonna Kulikovich for her statistical expertise and consultation.

References 1. American Heart Association. 2001 heart and stroke statistical update. Dallas (Tex): American Heart Association; 2000. 2. Cardillo C, Kilcoyne CM, Cannon RO. Racial differences in nitric oxide-mediated vasodilator response to mental stress in the forearm circulation. Hypertension 1998;31:1235-9. 3. Lang CC, Stein CM, He HB. Blunted blood pressure response to central sympathoinihibition in normotensive blacks: increased im-

370 Pescatello et al

4.

5.

6. 7. 8.

9.

10. 11. 12. 13. 14.

15.

16. 17.

18.

portance of nonsympathetic factors in blood pressure maintenance in blacks. Hypertension 1997;30:157-62. Svetkey LP, Chen YT, McKeown SP. Preliminary evidence of linkage of salt sensitivity in black Americans at the ␤2-adrenergic receptor locus. Hypertension 1997;27:918-22. Stein CM, Lang CC, Singh I. Increased vascular adrenergic vasoconstriction and decreased vasodilation in blacks: additive mechanisms leading to enhanced vascular reactivity. Hypertension 2000; 36:945-51. Perregaux D, Chaudhuri A, Rao S. Brachial reactivity in blacks. Hypertension 2000;36:866-71. Joyner MJ, Dietz NM. Nitric oxide and vasodilation in human limbs. J Appl Physiol 1997;83:1785-96. Gokce N, Holbrook M, Duffy SJ. Effects of race and hypertension on flow-mediated and nitroglycerin-mediated dilation of the brachial artery. Hypertension 2001;38:1349-54. American College of Sports Medicine. Position stand: physical activity, physical fitness, and hypertension. Med Sci Sports Exerc 1993;25:I-X. Cooper R, Rotimi C. Hypertension in blacks. Am J Hypertens 1997;7:804-12. Pickering TG. Why is hypertension more common in African Americans? J Clin Hypertens 2001;3:50-2. Abate NI, Mansour YH, Tuncel M. Overweight and sympathetic overactivity in black Americans. Hypertension 2001;38:379-83. Oluseye KA. Cardiovascular responses to exercise in Nigerian women. J Hum Hypertens 1990;4:77-9. Headley SA, Keenan TG, Manos TM. Renin and hemodynamic responses to exercise in borderline hypertensives. Ethn Dis 1998; 8:312-8. Pescatello LS, Kulikovich J. The after effects of dynamic exercise on ambulatory blood pressure. Med Sci Sports Exerc 2001;11:185561. Pescatello LS, Fargo AE, Leach CN. Short-term effect of dynamic exercise on arterial blood pressure. Circulation 1991;83:1557-61. Pescatello LS, Miller BE, Danias PG. Dynamic exercise normalizes resting blood pressure in mildly hypertensive women. Am Heart J 1999;138:916-21. Kenney MJ, Seals DR. Postexercise hypotension: key features, mechanisms, and clinical significance. Hypertension 1993;22:65364.

American Heart Journal February 2003

19. Halliwill JR. Mechanisms and clinical implications of post-exercise hypotension in humans. Exerc Sports Sci Rev 2001;29:65-70. 20. Hara K, Floras JS. Influence of naloxone on muscle sympathetic nerve activity, systemic and calf haemodynamics and ambulatory blood pressure after exercise in mild essential hypertension. J Hypertens 1994;13:447-61. 21. Kulics, JM, Collins HL, DiCarol SE. Postexercise hypotension is mediated by reductions in sympathetic nerve activity. Am J Physiol 1999;45:H27-32. 22. Chandler MP, DiCarlo SE. Sinoaortic denervation prevents postexercise reductions in arterial pressure and cardiac sympathetic tonus. Am J Physiol 1997;42:H2734-45. 23. Patil RD, DiCarlo SE, Collins HL. Acute exercise enhances nitric oxide modulation of vascular response to phenylephrine. Am J Physiol 1993;34:H1184-8. 24. Friewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502. 25. Noble BJ, Borg GAV, Jacobs J. A category-ratio perceived exertion scale: relationship to blood and muscle lactate and heart rate. Med Sci Sports Exerc 1983;15:523-6. 26. Dill DB, Costill DL. Calculation of percent changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol 1974; 37:247-8. 27. Cooper RS, Rotimi CN, Ward R. The puzzle of hypertension in African-Americans. Sci Am 1999;280:56-63. 28. Luke A, Guo X, Adeyemo AA. Heritability of obesity-related traits among Nigerians, Jamaicans and US black people. Int J Obes 2001;25:1034-41. 29. Baron AD. Hemodynamic actions of insulin. Am J Physiol 1994; 267(Endocrinol Metab 30):E187-202. 30. Thompson PD, Crouse SF, Goodpaster B. The acute versus the chronic response to exercise. Med Sci Sports Exerc 2001;33: S438-45. 31. Reaven GM, Lithell H, Landsberg L. Hypertension and associated metabolic abnormalities-the role of insulin resistance and the sympathoadrenal system. N Eng J Med 1996;334:374-81. 32. Paulev PE, Jordal R, Kristensen O. Therapeutic effect of exercise on hypertension. Eur J Appl Physiol 1984;53:180-5.