Intended for healthcare professionals


Cochrane review finds no proved benefit in drug treatment for patients with mild hypertension

BMJ 2012; 345 doi: (Published 14 August 2012) Cite this as: BMJ 2012;345:e5511
  1. Jeanne Lenzer
  1. 1New York

Treating patients with stage 1 (mild) hypertension has no benefit, a Cochrane review of studies conducted in the United Kingdom, Australia, and the United States has found.1

Data from four randomised controlled trials, involving 8912 patients with stage 1 hypertension (systolic blood pressure 140-159 mm Hg or diastolic 90-99 mm Hg, or both) and treated for four to five years, found that drug treatment did not reduce total mortality (risk ratio 0.85 (95% confidence interval 0.63 to 1.15)), coronary heart disease (1.12 (0.8 to 1.57), or stroke (0.51 (0.24 to 1.08)). Patients with pre-existing cardiovascular disease were excluded from the study.

David Cundiff, one of the reviewers, said that he believes that the analysis should lead to dramatic changes in the way doctors treat mild hypertension, allowing patients to throw away their blood pressure pills and focus instead on far more effective as well as evidence based approaches, such as exercising, smoking cessation, and eating a DASH (diet against systolic hypertension) or Mediterranean diet.

Cundiff told the BMJ that “in light of the negative results of the trials in the literature” further clinical trials of drug treatment for mild hypertension should be conducted only if patients first undergo lifestyle changes and those efforts fail.

James Wright, coordinating editor of the Cochrane Hypertension Group, told the BMJ that until now it has simply been assumed that treating mild hypertension, which is what most hypertensive patients have, is beneficial. He said that doctors and guideline writers have based their opinions on a combination of assumptions and data from clinical trials in which patients with mild hypertension were not analysed separately.

The reviewers were able to obtain patient level data for three of the four trials and conducted a pooled analysis of those patients with mild hypertension. They then combined those data with a fourth trial that enrolled participants who almost exclusively had mild hypertension.

The lack of benefit is perhaps even more significant, because the Cochrane review did not exclude patients with risk factors such as diabetes and probably included patients with target organ damage or a 10 year risk of cardiovascular disease of 20% or more, as these parameters were not recorded.

Guidelines on treating hypertension differ among countries, but practice may be the same.

A spokesperson for the US National Heart, Lung, and Blood Institute, which appoints experts to the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (known as the JNC), told the BMJ in a written statement that its previous guidelines (JNC7) recommend lifestyle modification and drug treatment to achieve a target blood pressure goal below 140/90 mm Hg for patients with stage 1 hypertension. The spokesperson said that the committee is “currently reviewing the evidence and deliberating on various topics, including blood pressure treatment thresholds and goals.”

The institute plans to pass the Cochrane analysis along to the committee for consideration.

Professional US and European guidelines are more aggressive in recommending drug treatment for uncomplicated, mild hypertension than UK guidance, which recommends treatment for stage 1 hypertension only if the patient has additional risk factors.2 3 4 However, according to the Cochrane reviewers and others, prescribing in the UK and Europe is likely to be similar to that in the US, as many doctors do not follow guidelines.

Des Spence, a general practitioner in Glasgow, told the BMJ, “Risk is abstract, and doctors and patients struggle with the concept and may not follow the guidelines. Doctors see a blood pressure of 156/98 and they prescribe. It’s easier to treat than not treat—you never get blamed for overtreating.”

A spokesperson for the UK National Institute for Health and Clinical Excellence told the BMJ that it was unable to comment on how many patients with stage 1 hypertension without the risk criteria specified by NICE are currently taking antihypertensive drugs.

Jerome Hoffman, emeritus professor of medicine at UCLA and an expert in critical appraisal of medical literature, said, “We’ve long known that almost all benefit from treating severe hypertension comes with lowering BP [blood pressure] just a little. On the other hand, efforts to lower BP to ‘normal,’ typically requiring multiple drugs, are not only usually unsuccessful but produce more harm than good, since adverse effects of intensive treatment outweigh the minimal marginal benefit of a little more BP ‘control.’ Drug treatment of mild hypertension, like intensive treatment of severe hypertension, may be of great value to drug makers, but it was almost predictable that it would provide little or no benefit for patients.”

However, not all doctors are convinced that these results should change practice. Mark Ebell, associate professor of epidemiology in the University of Georgia’s College of Public Health, told the BMJ that most patients came from a single study and that many of the patients in that study were taking only a β blocker. Recent systematic reviews have indicated that β blockers do not provide the same mortality benefit as other antihypertensives. Also, the total number of events in the studies was relatively small, suggesting that a large, well designed study comparing thiazides or angiotensin converting enzyme inhibitors (or both) with placebo is needed to definitively answer this question.

Wright said that he agrees that a well designed trial is needed. However, he added, “The fact that we don’t know hopefully should change practice.” He said that doctors should explain to patients the lack of evidence supporting drug treatment for mild hypertension and engage the patient in choosing whether to continue the pills or not.


Cite this as: BMJ 2012;345:e5511


View Abstract