A high blood pressure simulation
A simulation of strokes, heart attacks, and heart failure in men with high blood pressure, with and without treatment.
What happens to 500 untreated men with high blood pressure over 20 years?
- Major cardiovascular event
- 0
- Fatal cardiovascular event
- 0
A stroke, heart attack, or heart failure.
About 4 in 10 cardiovascular events are fatal.
Model follows 500 hypothetical 55-year-old men (non-smokers, without diabetes) with blood pressure around 155/95. For women of the same age and blood pressure, the risk is about a third lower.
How it works
The simulation follows 500 men with untreated high blood pressure for 20 years, showing strokes, heart attacks, and heart failure as they happen. Each man is 55, does not smoke, does not have diabetes, and has a blood pressure of 155/95. The simulation then re-rolls the dice as if each man’s systolic blood pressure (the top number) had been lowered by 18 points with treatment.
This is a Monte Carlo simulation: for each man, we roll dice weighted by likelihood.
Baseline risk. Without treatment, over 20 years, each of these men has a 33% chance of a major cardiovascular event: a stroke, a heart attack, or heart failure. This figure is derived from the Framingham Heart Study’s risk formula. In the simulation, 40% of these events are fatal, either right away or within the next few years.
Treatment. A 2016 analysis that pooled blood pressure trials finds that each 10-point drop in systolic blood pressure lowers the risk of a major cardiovascular event by 20%. An 18-point drop therefore brings the 33% risk down to 22%. Well-controlled treatment typically lowers systolic pressure that much, often with more than one medication. The share of events that are fatal stays the same, so fewer events means fewer deaths.
Caveats. The simulation models one kind of patient. Besides blood pressure, a real person’s risk depends on their age and sex, whether they smoke, and whether they have diabetes or other health conditions. Women of the same age and blood pressure have a lower risk than the men in the simulation. The simulation ends after 20 years, and nobody in it dies of anything else, though in real life some men like these would die of other causes first. It counts only deaths that follow a stroke, heart attack, or heart failure. Treatment also prevents some other deaths, such as sudden cardiac death and death from a ruptured aorta (the body’s main artery), so the simulation understates the benefit of treatment. The simulation does not show side effects of medication, or the effects of diet, exercise, and weight loss, which can lower blood pressure further.
Sources
- Ettehad et al., Lancet 2016: Blood pressure lowering for prevention of cardiovascular disease and death
Meta-analysis of 123 randomized trials (613,815 participants). Per 10 mmHg SBP reduction: stroke 27%, CHD 17%, heart failure 28%, major CVD event composite ~20%, all-cause mortality 13%.
- D'Agostino et al., Circulation 2008: General cardiovascular risk profile for use in primary care (Framingham Heart Study)
Framingham 10-year general CVD risk equations. The simulation starts from a ~18% 10-year risk for an untreated 55-year-old man (non-smoker, no diabetes, total cholesterol ~210, HDL ~45) at SBP ~155 and extends it to ~33% over 20 years; the 20-year figure is our own extrapolation, not a number the paper reports.
- Lewington et al., Lancet 2002: Age-specific relevance of usual blood pressure to vascular mortality
Prospective Studies Collaboration meta-analysis of 1 million adults in 61 observational studies. Continuous BP-risk relationship down to at least 115/75 mmHg with no threshold: at ages 40-69, each 20 mmHg higher usual SBP (or ~10 mmHg DBP) is associated with about double the death rates from stroke, ischemic heart disease, and other vascular causes; the proportional effect is about half as steep at ages 80-89.
- Law, Wald, Morris, BMJ 2003: Value of low dose combination treatment with blood pressure lowering drugs (analysis of 354 randomised trials)
A single standard-dose BP medication produces ~9 mmHg SBP reduction.
- Lonn et al., NEJM 2016: Blood-pressure lowering in intermediate-risk persons without cardiovascular disease (HOPE-3 BP arm)
After an active run-in, dizziness / low blood pressure / lightheadedness led to stopping candesartan + HCTZ in 3.4% vs 2.0% on placebo. Active run-in caveat: HOPE-3 randomized only adherers, so these rates understate true first-start nuisance effects.
- ALLHAT Officers and Coordinators, JAMA 2002: Major outcomes in high-risk hypertensive patients randomized to ACE inhibitor or calcium channel blocker vs diuretic
Class-specific safety findings: low potassium was more common with chlorthalidone; ACE-inhibitor angioedema was rare but serious.