Lowering blood pressure: what movement actually changes

The cuff beeps, you write down two numbers, and one of them sits higher than you wanted. Maybe 138 over 86. Maybe well past that. The first impulse is to buy something that will fix the figure by Friday. The second is to put the monitor in a drawer and call the morning a one-off.

Both skip the thing the guidelines have been saying for years, and the thing you can start this week: move, on a dose you will actually repeat. How far the number moves is measured. It is not a miracle, and it is not a reason to drop a tablet someone prescribed.

When my weight is part of why the reading is up, I replace two main meals a day in that stretch with ProShape All-in-1 Wild Berries. The authorised effect is weight loss on an energy-restricted diet. Not a drop in blood pressure, and not a reason to stop a medicine.

What the number on the monitor means

The 2024 European guidelines, in the German Cardiac Society's commentary, sort an office reading more tightly than many of us still remember. Non-elevated is under 120 over 70. Elevated is 120 to 139 systolic, or 70 to 89 diastolic. They call it hypertension from 140, or from 90. One high morning is not a diagnosis. For that, the cardiologists want a standardised home series: a validated upper-arm cuff, five minutes seated, two readings one to two minutes apart, morning and evening, on at least three days and preferably seven.

Why the fuss is worth it comes from an observational pooling, not from a training trial. The Prospective Studies Collaboration, Lewington and colleagues, pulled together data on about a million adults with no vascular disease recorded at the start. At ages 40 to 69, a difference of 20 mmHg in usual systolic pressure, or about 10 mmHg diastolic, went with more than a doubling of the stroke death rate, and with a doubling of the death rate from ischaemic heart disease. The link ran down to at least 115 over 75, with no comfortable floor where it stopped mattering.

That is an association across years. It is not a promise that 5 mmHg off a training plan will halve your risk. It is why cardiologists take small, lasting shifts seriously. And why a cuff you only use after a tense phone call will lie to you.

What the training trials actually measured

The densest comparison of training types is a 2023 paper in the British Journal of Sports Medicine. Edwards and colleagues pooled 270 randomised trials, 15,827 people, in pairwise analyses. Against control groups, resting blood pressure fell like this:

• aerobic training: 4.49 mmHg systolic and 2.53 diastolic
• dynamic resistance training: 4.55 and 3.04
• the two combined: 6.04 and 2.54
• high-intensity interval training: 4.08 and 2.50
• isometric training, holding work without the joint moving: 8.24 and 4.00

Inside the aerobic group, walking was the smaller shift, at minus 2.85 mmHg systolic (credible interval about minus 4.41 to minus 1.29). Running was minus 6.83, cycling minus 6.88. Inside the isometric group, the wall squat was minus 10.47 mmHg systolic (about minus 13.36 to minus 7.63), and the leg-extension version minus 7.96. People who started higher had a little more room to fall: about 0.10 mmHg more change for each 1 mmHg of higher baseline systolic pressure.

A second line of German guidance points the same way, with different figures. The national care guideline, as cited on leitlinien.de, puts brisk walking at minus 4.11 systolic and minus 1.79 diastolic across 73 randomised trials. The German Hypertension League's patient page talks about up to about 5 mmHg from aerobic or dynamic strength work, and up to about 8 mmHg from isometric work. Same neighbourhood as Edwards. It is the plain-language version, not the tighter trial arithmetic.

Two brakes belong in the same paragraph, or the wall squat starts to sound like a drug. First, the trials were uneven, and there were signs of publication bias for the aerobic effect and for the diastolic isometric effect. Second, a ranking of modes is not a head-to-head proof. The network analysis did not have direct comparative trials for that. People who turn up and finish the sessions are also not the same as someone who stops after a fortnight.

And the tablet. In 2018, in the same journal, Naci and colleagues set 391 randomised trials side by side: 197 on exercise (10,461 people) and 194 on blood-pressure drugs (29,281 people). Nobody tested them directly against each other. Across all groups, the drugs lowered systolic pressure more than exercise, by 3.96 mmHg on average. Only among people who already had high pressure was there no detectable difference between endurance or dynamic resistance work and several drug classes, including ACE inhibitors, angiotensin receptor blockers, beta-blockers and diuretics. The exercise trials were messier, mostly because you cannot blind a training plan. The fair reading: movement is its own lever. It is not evidence for taking the box out of the drawer.

When you should not just start

The German Hypertension League puts practical limits on its patient page, and they belong before the first wall sit, not after. Under 140 over 90 it sees no extra restriction. Between 140 and 159 over 90 to 99 it advises against maximal strength work and against breath-holding under strain. Moderate aerobic work is fine there. From 160 over 100, strength training waits for a doctor's clearance, and intervals are off the plan. From 180 over 110: no training until the reading is controlled. Chest pressure, dizziness, a severe headache or vision changes mean stop, and get medical advice. Not another round.

That is one society's patient guidance, not a licence I am handing you, and not a ban I get to impose from here. If you already take medicine, the dose stays a clinical question. A good walk does not change that.

The plan the guideline actually recommends

The 2024 ESC guideline, in the German Cardiac Society commentary and in the comparison table, gets specific. Aerobic work of at least 150 minutes a week at moderate effort, or 75 minutes harder. Isometric holds on two or three days: three sets of one to two minutes, for example a hand grip, a plank or a wall sit. Dynamic strength on two or three days: three sets of 10 to 15 repetitions at 40 to 60 percent of the weight you could lift once. Their examples are squats, press-ups and sit-ups.

My week looks like this. It is a translation of that guideline, not a protocol from a paper of my own.

Five brisk sessions on foot or on the bike, adding up to the 150 minutes. Brisk means you can talk. Singing would already be a push. If 30 minutes in one go is too much today, start shorter and add time once the walk feels easy. That is household sense, not a trial arm. What walking did in the large analysis is above: less than cycling or running, and not zero. How to fit the steps into a full day is in 10,000 steps a day. The easy base that makes the harder sessions worth doing is in endurance training.

On three days, the wall sit, the same hold the trials called a wall squat. Back to the wall, feet far enough forward that the thighs work and you can still breathe evenly. Three holds, a minute to start, rest between them, building toward two minutes if the first week was clean and nobody has restricted your training. In the Edwards trials the common pattern was four holds of two minutes, three times a week, with effort that ran from moderate on the first hold to hard but controlled on the last. The guideline is a little shorter: three sets of one to two minutes. I use the guideline as the week, and the trial version as a reminder that "leaning a bit" is not the dose. Breathe through the hold. Do not trap the air. Straining on a held breath is exactly what the League does not want once the resting reading is already up.

On two days, a short strength piece: squats or press-ups, three sets, 10 to 15 repetitions, heavy enough that the last two are work and the form still holds. That is the kitchen version of 40 to 60 percent. Not the load you fight for a photo.

I am keeping intervals small in this piece. They did lower pressure in the Edwards analysis, by about 4 mmHg systolic. The Hypertension League does not want them while the resting reading is 160 over 100 or higher. If you already train there and you have clearance, the plain version is in HIIT.

Weight: the second lever, with a narrow limit

The guideline puts a stable, healthy body weight next to the training. Two separate reviews have looked at how far pressure moves with the kilos.

In 2003, in Hypertension, Neter and colleagues pooled 25 randomised trials, 4,874 people. A net loss of 5.1 kg, from eating less, moving more, or both, went with a drop of 4.44 mmHg systolic and 3.57 diastolic. Per kilogram that was about 1.05 and 0.92 mmHg.

The 2016 Cochrane review by Semlitsch filtered harder: only people with hypertension, only weight-reducing diets against no diet, at least 24 weeks. The difference was minus 4.49 mmHg systolic (interval minus 7.20 to minus 1.78), minus 3.19 diastolic, and minus 3.98 kg. They graded the evidence low, because of bias risk and wide intervals. Direction and size look like Neter. That review is not proof that the diet prevents heart attacks. It measured pressure and weight.

In the evening, when the alternative is a late takeaway, I make the shake and count it as the meal, not as pudding: ProShape All-in-1 Wild Berries. The label says what it is allowed to say. Substituting two of the main daily meals of an energy-restricted diet with meal replacements contributes to weight loss. Other food stays part of the diet, and you need enough calorie-free drink, at least two litres a day. A shake beside a normal dinner is an extra meal. That does not lower anything.

How narrow a meal replacement is, outside this, is in meal-replacement shakes. Alcohol is its own lever on the same guideline table, and we wrote that separately: alcohol and muscle building. Less salt is on the guideline too. The German Cardiac Society commentary mentions a potassium recommendation only for a narrow group: hypertension, no moderate or advanced kidney disease, and a sodium intake above 5 g a day. That is a conversation at the surgery, not a powder off a shelf. Extra potassium can do harm if the kidneys are not clearing it.

How you tell whether it is working

Not by how you feel after one good walk. Pressure can sit lower for a while after a session and be back the next morning. The Edwards trials ran at least two weeks. The Hypertension League's patient page talks about three to six months before the home reading shifts in a way you can see. Measure the way the cardiologists describe it for diagnosis: upper arm, seated, two readings, morning and evening, across several days, and compare week with week. Not Tuesday with the Tuesday after an argument.

If the reading climbs instead of falling, or you go dizzy in the hold, that is not a cue to make the holds longer. That is a cue to call the surgery before you harden the plan.

The takeaway

A week that matches the guideline: 150 minutes of brisk walking or cycling, wall sits on three days with steady breathing, a moderate strength set on two days. The tablet stays until someone with your notes says otherwise. The shake is not a blood-pressure drug.

The walks and the wall sits stay the lever for me, even in a stretch where the weight needs to come down. ProShape All-in-1 Wild Berries only comes along when I am actually replacing two main meals. If the shake sits next to a full plate, it has missed the job.

Photo: Soroush Karimi, CC0.