Management of blood pressure in primary careBMJ 2009; 338 doi: https://doi.org/10.1136/bmj.b940 (Published 19 May 2009) Cite this as: BMJ 2009;338:b940
- 1Primary Care Clinical Sciences, University of Birmingham, Birmingham B15 2TT
- 2General Practice and Primary Care Research Unit, University of Cambridge, Cambridge CB2 0SR
Two linked studies provide new insights into the monitoring (doi:10.1136/bmj.b1492) and control (doi:10.1136/bmj.b1665) of blood pressure.1 2 In recent years, emphasis has shifted away from treating hypertension as a separate entity towards treating it in the context of the risk of cardiovascular disease.3 This shift has informed randomised controlled trials, such as the PROGRESS trial, which have looked for (and found) benefits of treating people at high risk (in this case, those with a previous stroke or transient ischaemic attack) with blood pressure lowering agents regardless of their blood pressure.4 It has also influenced guidelines that have recommended lower blood pressure treatment targets for people at higher risk of cardiovascular events.5 6 Within this new paradigm, we continue to titrate the use of antihypertensive drugs against blood pressure measurements. The two linked studies both provide data to challenge this orthodoxy.
Hypertension is perhaps the most common form of chronic disease managed in primary care and routine review is costly. Follow-up is recommended every six to 12 months for people with controlled blood pressure, and in the United Kingdom pay for performance remuneration requires that blood pressure has been measured within the past nine months.5 6 7 Actual attendance seems to be even more frequent, however—a recent study found that patients receiving usual care for hypertension saw their general practitioner for such care more than four times in a year.8
Keenan and colleagues’ findings suggests that much of this activity is misguided. They used data from patients in the active, two drug, treatment arm of the PROGRESS study to evaluate variability in blood pressure from three months after the patients started treatment.1 Over the next 33 months they estimated long term and short term variability with the aim of differentiating true changes in blood pressure over time from random variation and measurement “noise.” They found that, if someone had a blood pressure of 130 mm Hg at baseline, there was only about a 1% chance that a subsequent blood pressure reading three months later of 140 mm Hg or above was a “true” increase. This figure rose with time, but even after 21 months there was only a 50% chance that such an increase was real.
These data suggest that we need to rethink how we monitor the control of blood pressure. We could use techniques that reduce the effect of the inherent variability of blood pressure and increase the chance that an observed increase in blood pressure is a true one. For example, self monitoring allows multiple measurements to be taken over several days.9 Recent European guidelines have described self monitoring schedules, but the evidence base to support these schedules or self monitored treatment targets is not strong.10 Intermittent use of self monitoring would require careful education of patients and doctors, but modern automated sphygmomanometers are simple enough to require minimal instruction. Ambulatory monitoring could be an alternative for those unable to self monitor, but it is expensive and not widely available in primary care. Alternatively, less frequent monitoring may be more appropriate and cost effective. Adverse effects of treatment rather than blood pressure control would then be the main driver of the frequency of monitoring.
Law and colleagues conducted a wide ranging meta-analysis of trials of blood pressure lowering, and their findings will contribute to debate on the management of hypertension in several areas.2 For example, in contrast to other reviews, the authors conclude that β blockers are as effective as other blood pressure lowering agents.5 11 Their finding that antihypertensive drugs have no major effects on cardiovascular risk independent of their effect on blood pressure concurs with previous analyses.12 They also found that the effect of lowering blood pressure on risk of cardiovascular disease was independent of the pretreatment blood pressure; this reinforces the view that treatment to lower blood pressure should be offered on the basis of risk, regardless of blood pressure.3
Perhaps the most controversial aspect of their analysis is their comparison of combination blood pressure therapy at half standard doses with combination therapy at standard dose. For example, they conclude that patients with a systolic blood pressure of 150 mm Hg would achieve a 20 mm Hg drop in pressure if they took three drugs at half standard dose, or a 16 mm Hg drop if they took two drugs at standard dose. No trials have tested combinations of three drugs at half standard dose, and their analysis assumes that the effects are additive. Taken at face value, these findings provide tacit support for the use of a “polypill” to lower the risk of cardiovascular disease in people likely to be at high risk (such as all people over the age of 55) without first checking their blood pressure.13
If antihypertensives differ little in their efficacy, then acceptability becomes a key driver for treatment choice, along with the presence of comorbidities, such as diabetes or benign prostatic hypertrophy. Law and colleagues argue that reduced dose combination therapy is likely to reduce dose related side effects, and that patient specific factors such as comorbidity are generally of minor importance in determining antihypertensive therapy—a view that runs counter to current guidelines.5 6 7
Taken together, these two studies may herald a new era of blood pressure management. The place of the sphygmomanometer in the doctor’s office for monitoring blood pressure lowering treatment no longer seems secure—it may perhaps be replaced by intermittent but infrequent blood pressure measurement at home supplemented by periodic enquiry about side effects as treatment choices become determined by tolerability rather than algorithm.
Cite this as: BMJ 2009;338:b940
Competing interests: RJM has received funding for research or conference expenses from Sanofi Aventis, Pfizer, A Menarini Pharma, and MSD. JM has undertaken consultancy for Pharmaswiss. JM and RJM are investigators on an NIHR funded stroke prevention programme, which includes exploration of the potential role of a polypill for cardiovascular risk reduction in primary care.
Provenance and peer review: Commissioned; not externally peer reviewed.