Exercise is one of the best nonpharmacological strategies for reducing blood pressure. Two other large-scale meta-analyses have confirmed the blood-pressure-lowering effects of exercise.

A meta-analysis published in 2012^[1] included 93 randomized controlled trials that investigated the effects of endurance, resistance, combined, and isometric exercise on blood pressure. With the exception of combined training for DBP, all modes of exercise were effective for reducing SBP and DBP. However — and aligning with findings from the summarized study — isometric exercise produced the largest reductions in SBP (−10 mmHg) and DBP (−6 mmHg) compared to the other modes of exercise, for which reductions ranged from 1 to 4 mmHg.

Another network meta-analysis published in 2018^[2] included 391 randomized controlled trials, about half of which compared exercise interventions (i.e., endurance, dynamic resistance, combined, and isometric exercise) to a control intervention. The second half of the trials compared antihypertensive medications (i.e., angiotensin-converting enzyme inhibitors, angiotensin-2 receptor blockers, beta-blockers, calcium channel blockers, and diuretics) to control interventions for the management of blood pressure.

Antihypertensive medications outperformed exercise by approximately 4 mmHg, though both exercise and medications were effective for reducing blood pressure. However, among people with hypertension, there were no differences found between most medications and endurance or dynamic resistance exercise for blood pressure. So while more research needs to be done directly comparing exercise and medication, exercise generally has comparable blood-pressure-lowering effects to medication, which is why exercise is included in the guidelines of many major societies as a first-line strategy to prevent/treat hypertension.^[3]^[4]

This was the first network meta-analysis to compare the effects of different exercise modes and submodes on blood pressure, and it revealed the distinct effects of each type of exercise.

Network meta-analysis 101

When there are many studies comparing just a couple of interventions or a single intervention to a control, how does a researcher choose the best out of all available interventions? A network meta-analysis^[5] helps generate practical evidence-based answers to these questions.

Clinical trials typically consist of a direct comparison between two interventions or an intervention compared to a control. A traditional meta-analysis can synthesize these trials only if the interventions and controls are similar. A network meta-analysis extends this concept by integrating direct comparisons (ones in which clinical trials have been done) while also using statistical modeling to yield indirect comparisons (comparisons where no trial has actually been performed). Here’s an oversimplified example: if trials comparing A to B exist, and there are other trials comparing A to C, network meta-analysis can calculate how B compares to C even though there aren’t trials that directly compare B to C.

One valuable number that can be obtained from network meta-analyses is called the surface under the cumulative ranking curve (SUCRA^[6]). Each treatment in a network meta-analysis can have a SUCRA value associated with it, which indicates the relative ranking of the treatment — or in other words, how effective that treatment is compared to the others. SUCRA values range from 0% to 100% and can be roughly thought of as the probability that the treatment is among the best of the bunch. The higher an intervention’s SUCRA value, the more likely it is that it ranks near the top of all treatments considered in the network meta-analysis. If a treatment is definitely the best of those considered, it would have a SUCRA of 100%. The concepts of indirect comparisons and SUCRA are illustrated in Figure 1.

When exercise modes were ranked for their effectiveness in reducing SBP, isometric exercise was the most effective, followed by combined training, resistance training, aerobic exercise training, and HIIT. Subgroup rankings were as follows: wall squats, leg extension, handgrip exercise, cycling, running, combined training, sprint interval training, resistance training, aerobic interval training, and walking.

When exercise modes were ranked for their effectiveness in reducing DBP, isometric exercise was also the most effective, followed by resistance training, HIIT, combined training, and aerobic exercise training. Subgroup rankings were as follows: running, wall squats, handgrip exercise, leg extension, cycling, sprint interval training, resistance training, aerobic interval training, combined training, and walking.

One novelty of this meta-analysis compared to previous ones was the addition of HIIT as an exercise mode. Although it is often considered to be aerobic exercise, HIIT is shorter and more intense and may have effects that are unique from those of traditional endurance training. In this study, HIIT led to clinically relevant reductions in SBP and DBP, though it was least effective among the interventions. The benefits of HIIT were driven primarily by sprint interval training (“all out” efforts), while aerobic interval training interventions weren’t effective. Along with the superior benefits of running and cycling compared to walking, this finding seems to indicate that higher-intensity aerobic exercise may be necessary to elicit the greatest blood pressure reductions.

It’s somewhat surprising that isometric exercise ranked the best among all interventions because this type of exercise is often given less attention than aerobic and resistance training. However, there are mechanisms to support its benefits. During isometric exercise, prolonged muscle contraction temporarily cuts off blood flow through the vasculature. When the exercise is over, blood flow returns with a high magnitude in a process called reactive hyperemia, leading to a large release of nitric oxide and the relaxation of blood vessels. Over time, this process improves the health and function of blood vessels and reduces blood pressure.^[7]

Performing isometric exercise with a larger muscle — wall squats or leg extension, for example — would have an even greater effect by causing more postexercise reactive hyperemia. This is likely why wall squats were ranked as the most effective submode of exercise for reducing blood pressure.

Although it’s nice to know which exercise is “best” for improving blood pressure, this study is a welcome reminder that any and all types of exercise are beneficial. If you enjoy running, then run. If you enjoy lifting weights, then do that. Better yet, do a bit of both.

The best types of exercise for reducing blood pressure Original paper