To examine the effect of aerobic exercise on ambulatory systolic and diastolic blood-pressure.

The studies were identified by searching MEDLINE, by cross-referencing, through handsearches, and through communication with experts. Only studies published in English language journals between January 1985 and December 1994 were included.

Study designs of evaluations included in the review

Studies that included a control group or a control period. A change in systolic and diastolic ambulatory blood-pressure must have been reported.

Specific interventions included in the review

Aerobic exercise had to be the only mode of training. The interventions included stationary cycling, walking, jogging and dynamic calisthenics. One trial of weight training was excluded. The included studies reported the following characteristics of the training programmes: the mean duration of training was 14 weeks (range: 3 to 58); the mean frequency was 4 days per week (range: 3 to 7); the mean intensity (percentage maximum oxygen consumption) was 65 (range: 50 to 75); the mean duration of the training session was 39 minutes (range: 30 to 60); and the mean adherence was 92% (range: 84 to 98).

Participants included in the review

The participants were not to be taking part in a regular exercise programme before participation in a study. Over 80% of the participants in the trials were male. The baseline mean systolic blood-pressure was 132 mmHg (standard deviation, SD=11) in the exercise group and 133 mmHg (SD=11) in the control group. The baseline mean diastolic blood-pressure was 88 mmHg (SD=8) in the exercise group and 89 mmHg (SD=9) in the control group.

Outcomes assessed in the review

The ambulatory systolic and diastolic blood-pressure were measured in a 24-hour period, and during waking and sleeping hours.

How were decisions on the relevance of primary studies made?

The author does not state how the papers were selected for the review, or how many of the reviewers performed the selection.

The studies were coded on the basis of randomisation, study size and duration. The author does not state how the papers were assessed for validity, or how many of the reviewers performed the validity assessment.

The data were extracted onto a coding sheet for up to 223 pieces of information. The data were classified and coded for study characteristics, participant characteristics, training programme characteristics, and ambulatory blood-pressure results.

How were the studies combined?

The change in ambulatory blood-pressure was calculated by subtracting the post-training group mean blood-pressure from the baseline pre-training blood-pressure, and then subtracting the change in the control group from the change in the exercise group. The 95% confidence intervals (CIs) were also calculated. The blood-pressure results were weighted by sample size. Outlier groups were identified and deleted from the analysis. A bootstrap simulation (10,000 random repeat samples from the pool of data) was used to examine the change in ambulatory blood-pressure (see Other Publications Of Related Interest).

How were differences between studies investigated?

Independent t-tests were used to assess the post-exercise change in ambulatory blood-pressure based on the following: whether the trials were randomised or not; whether there was crossover or no crossover; and whether blood-pressure was measured during waking or sleeping hours.

Six trials, with a total of 167 participants representing 12 exercise and 9 control groups, were included. The author reported that 137 were 'exercisers' and 118 controls were control; some participants served as both exercisers and controls. Four of the seven studies were randomised and/or used a crossover design.

During a 24-hour period, the ambulatory systolic and diastolic blood-pressure decreased by 1.72 mmHg (95% CI: 0.34, 3.10) and 2.56 mmHg (95% CI: 1.06, 4.01), respectively, as a result of training.

The ambulatory systolic and diastolic blood-pressure decreased by 1.25 mmHg (95% CI: -2.30, 4.80) and 1.32 mmHg (95% CI: -0.90, 3.54), respectively, during waking hours. During sleeping hours, the ambulatory systolic and diastolic blood-pressure decreased by 1.76 mmHg (95% CI: -1.42, 4.94) and 3.04 mmHg (95% CI: 0.26, 5.82), respectively.

The bootstrap simulation, adjusting for bias, confirmed the reliability of the data in all outcomes except for the change in ambulatory diastolic blood-pressure during sleeping hours (95% CI: 1.29, 5.27 mmHg). The change in 24-hour ambulatory blood-pressure was significantly greater for non-randomised studies than for randomised trials: the decreases were 3.50 mmHg (SD=0.70) and 0.83 mmHg (SD=1.47), respectively (t=2.90, p=0.02).

The results suggested that aerobic exercise results in a statistically-significant reduction in ambulatory systolic and diastolic blood-pressure, both through 24 hours and throughout sleeping hours. However, additional well-designed studies are needed.

The author acknowledged that although some of the results were statistically significant, they may not be clinically significant. The review included a good discussion section outlining the limitations of both the included trials and the review itself. Limitations of the studies included the following: the absence of data on the length of time that the participants had been sedentary prior to the intervention; the lack of data on women or differing ethnic backgrounds; and the lack of information on adherence to therapy. Limitations of the review, as reported by the author, included the exclusion of outliers and the meta-analytic approach itself. Further limitations were the search strategy, which was insufficient, and the decision to include only English language trials. In addition, the inclusion criteria for the participants (i.e. age, and whether they were normotensive, or hypertensive or a mixture) were not stated. The methods of the meta-analysis were also unclear. There appeared to be significant heterogeneity with respect to the study design, and the duration, length and frequency of the intervention; this suggested that it may have been inappropriate to combine the results. It was also unclear as to how the author assigned groups as 'exercisers' or controls, as some participants were included in both the control and intervention groups.

The author states that there is a need for additional well-designed studies on the topic.

Elfron B. Bootstrap methods: another look at the jackknife. Ann Stat 1979;7:1-26.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.