Variability in Blood Pressure: Just a blip?

What is your approach to the patient with occasional high, or variable, blood pressure readings, or measurements that bounce around above and below the target range?
In such situations, the American Heart Association guidelines on measurement of blood pressure propose that:

“it is generally agreed that conventional clinic readings, when made correctly, are a surrogate marker for a patient’s true blood pressure, which is conceived as the average over long periods of time, and which is thought to be the most important component of blood pressure in determining its adverse effects.”


Similarly, the Joint European guidelines state that, as long as blood pressure is not dangerously elevated, repeat measurements:

“should be obtained over several months to define the patient’s usual blood pressure as accurately as possible”

Implicit in these statements is the largely untested concept that there exists a “true” blood pressure that causes vascular disease, and that visit-to-visit variability in readings represents random noise that gets in the way of knowing a patients “true” blood pressure (i.e. mean BP over time). While there is no doubt that the mean BP over time is an important predictor of vascular disease, it is also possible that other aspects of BP are important, such as variability or surges/spikes in BP. In reality, the mechanisms by which raised blood pressure causes vascular events are poorly understood.

This question was recently addressed by a clutch of papers by Rothwell et. al in the Lancet. The main paper used data from 2000 patients in the UK-TIA study, with replication of those findings in 4 similar TIA study cohorts totalling over 7500 people. Finally, the generalizability of their findings were tested in almost 18,500 patients without cerebrovascular disease from the ASCOT-BPLA study, which included data from 24-hour BP monitoring. The study aims were to establish the prognostic significance of:

1. Visit-to visit BP variability (defined as SD of within-individual BP measures)
2. Maximum BP
3. Episodic HTN
4. Residual BP variability in those already on BP meds

In the main UK-TIA cohort, 1324 (66%) survived to follow-up and there were 270 vascular events. Here are 9 powerful findings I learned from this excellent study:

1. Visit-to-visit variability in SBP was a much stronger predictor of stroke than mean BP (HR 12·08; 95% CI 7·40–19·72; p<0·0001)
2. Visit-to-visit variability in SBP is a powerful predictor of stroke and coronary events independent of mean SBP
3. Maximum SBP is more predictive of stroke than mean SBP (on clinic readings or on ABPM)
4. People whose SBP remains variable on treatment have a poor prognosis compared to those who don’t.
5. Stable hypertension has a better prognosis than does episodic hypertension (i.e. "normal" BP with peaks/surges into hypertensive range)
6. Variability in SBP was a stronger predictor of stroke in patients whose mean SBP was less than the cohort median (143 mmHg) i.e. those patients sitting in the borderline range.
7. Amlodipine-based treatment was associated with less BP variability than atenolol-based treatment
8. Visit-to-visit variability is not the same as “white-coat” hypertension. “White-coat” HTN –defined as the variability between first reading vs. the mean of the second two readings at a single visit- did not predict vascular events in this study
9. You can’t diagnose “variability” on 2 readings! You need to take several over time, and precision improves up to the tenth recording.

These results fundamentally challenge the prevailing wisdom regarding blood-pressure and how we diagnose, treat, and monitor patients with hypertension, and I can see them having far-reaching implications in the future.