Current issues of ACP Journal Club are published in Annals of Internal Medicine


Moderate exercise failed to reduce blood pressure

ACP J Club. 1992 Jan-Feb;116:15. doi:10.7326/ACPJC-1992-116-1-015

Source Citation

Blumenthal JA, Siegel WC, Appelbaum M. Failure of exercise to reduce blood pressure in patients with mild hypertension. Results of a randomized controlled trial. JAMA. 1991 Oct 16;266:2098-104.



To determine whether aerobic or strength and flexibility training is effective in reducing blood pressure (BP) in patients with mild hypertension.


Randomized, controlled trial of 16 weeks duration.


Volunteers were recruited from outpatient clinics and through media advertisements.


Men and women 29 to 59 years of age, weighing < 120% of ideal body weight, were included if they had a history of essential hypertension, were not doing regular aerobic exercise, and agreed to stop taking antihypertensive drugs for 4 weeks before the study. Exclusion criteria were systolic BP < 140 or > 180 mm Hg, diastolic BP < 90 or > 105 mm Hg, cardiovascular or respiratory disease, secondary hypertension, and contraindications to exercise. 92 of 99 randomized participants completed the study.


41 participants were randomly assigned to aerobic training (45 min, 3 times per week); 35 were assigned to a strength and flexibility training group (50 min, 2 to 3 times per week); and 23 were assigned to a "waiting list" control group. All participants were asked not to change their eating or outside-of-study exercise habits.

Main outcome measures

Changes in BP, measured as the mean of 4 sphygmomanometric measurements made in the clinic on 3 separate days; as the mean of at least 25 ambulatory recordings made over a normal day at work and at home; and as measurements made during treadmill exercise testing and laboratory mental stress testing. The study power to detect differences of 5 mm Hg between groups was 71% for systolic BP and 98% for diastolic BP.

Main results

Mean BP, measured in the clinic, decreased in all 3 groups (P < 0.001). The reduction for the aerobic group was from 141/95 mm Hg to 133/89 mm Hg; strength and flexibility group, 143/95 mm Hg to 136/89 mm Hg; waiting-list group, 142/95 mm Hg to 133/90 mm Hg. Groups did not differ for these measurements, for ambulatory BP, or for response during treadmill exercise. There were some small differences in BP response to mental stress.


A reduction in blood pressure in patients with mild hypertension and normal body weight was not related to participation in a 16-week program of moderate aerobic exercise.

Source of funding: National Institutes of Health.

Address for article reprint: Dr. J.A. Blumenthal, Department of Psychiatry, Duke University Medical Center, Center for Living, Box 3119, Durham, NC 27710, USA.


Several studies have concluded that exercise reduces blood pressure in patients with mild hypertension (1-4). How can one reconcile the negative findings of Blumenthal and colleagues with previous conclusions?

227 young to middle-aged persons with mild hypertension participated in the 5 relevant trials. Most were mildly to moderately obese healthy men (80%). Participants in the current trial possessed no obvious demographic or clinical differences from other studies, other than the inclusion of more women (38%).

Study exercise protocols have been comparable. They have consisted of 3 or 4 half-hour to hour sessions per week for 10 to 16 weeks, and have included walking, jogging, cycling, bicycle ergometry, and cross-country skiing. Control groups have either not exercised or have performed strength and flexibility calisthenics. A stratified analysis in this study found that individuals achieving the greatest improvements in aerobic capacity, regardless of their group assignment, experienced significant blood pressure reductions. This relationship could not be evaluated by the other trials because of small sample sizes and relative homogeneity in exercise intensity.

Blumenthal and colleagues' trial was the largest, had a drop-out rate of less than 10%, found no change in exercise capacity or eating habits in the control group, and controlled for potential confounders such as weight change. Although this trial is the best designed, no obvious differences in population or interventions can explain the varied findings.

As we await further large studies, clinicians treating mildly hypertensive persons might temper their enthusiasm concerning aerobic exercise, basing their recommendations more on other potential benefits and risks of exercise.

Cynthia D. Mulrow, MD
Audie L. Murphy Memorial Veterans Administration HospitalSan Antonio, Texas, USA

Cynthia D. Mulrow, MD
Audie L. Murphy Memorial Veterans Administration Hospital
San Antonio, Texas, USA


1. Duncan JJ, Farr JE, Upton SJ, et al. JAMA. 1985;254:2609-13.

2. Kukkonen K, Rauramaa R, Voutilainen E, Lansimies E. Ann Clin Res. 1982; 14(Suppl 34):139-45.

3. Martin JE, Dubbert PM, Cushman WC. Circulation. 1990;81:1560-7.

4. Urata H, Tanabe Y, Kiyonaga A, et al. Hypertension. 1987;9:245-52.