Beta blockers for hypertension: End of an era?

October 19, 2005

Oct 19, 2005

Umea, Sweden - Beta blockers are not as effective as other antihypertensive drugs in reducing stroke, a new meta-analysis has found [ 1 ]. The Swedish authors conclude that beta blockers should not remain first-choice agents in the treatment of primary hypertension or be used as reference drugs in future randomized controlled trials of hypertension.

The meta-analysis, published online in the Lancet on October 18, 2005, showed a 16% increased risk in stroke with beta blockers compared with other antihypertensive drugs and a 3% increase in all-cause mortality.

The authors, led by Dr Lars Lindholm (Umea University Hospital, Sweden), conclude: "Switching hypertension treatment from beta blockers to other low-cost antihypertensive drugs in patients without heart disease should have a major health effect without increasing cost. Such a change, however, should be carried out slowly and under a doctor's supervision."

The results of this meta-analysis build on those of the LIFE and ASCOT trials, which showed better results with other classes of antihypertensives than with beta blockers. An accompanying editorial by ASCOT investigator Dr Gareth Beevers (City Hospital, Birmingham, UK) says that this meta-analysis heralds the end of the era of beta blockers for hypertension [ 2 ]. "Surely, the era of beta blockers for hypertension is over," he writes

In an interview with renal wire , Lindholm addressed the question of why it has taken so long to establish that beta blockers are not as good as other antihypertensives. "I have been around for a long time, and I may be partly to blame, as I have done many studies, including STOP 1 and STOP 2, which helped to establish beta blockers as effective medications for hypertension. But most trials included beta blockers as conventional therapy, often in combination with diuretics, and as diuretics work very well, the not-so-impressive effects of beta blockers have been missed," he explained. "We are now saying it is still good to start with a diuretic if you want, but don't add a beta blocker as the next drug. Instead, add an ACE inhibitor, angiotensin receptor blocker, or calcium [channel] blocker," he added.

Lindholm also pointed out that the larger studies that showed that beta blockers were not as good as other antihypertensives (LIFE, ASCOT) have been published only in the past few years, and so the data haven't been available until relatively recently. But he stressed that beta blockers are still good drugs for other indications, such as secondary prevention post MI, heart failure, and atrial fibrillation.

In the Lancet paper, the researchers point out that beta blockers have been widely used in the treatment of hypertension and are recommended as first-line drugs in hypertension guidelines, but a preliminary analysis has previously shown that atenolol is not very effective in hypertension [ 3 ]. To gain more data on the whole class of beta blockers, they conducted a meta-analysis of all major randomized trials found in the literature comparing beta blockers with other antihypertensive drugs. They identified 13 such trials, involving a total of 105,951 patients. This included the recent ASCOT study. The authors also included seven studies comparing beta blockers with placebo.

Data were analyzed for all beta blockers and for three subgroups?atenolol, nonatenolol beta blockers, and the combination of beta blockers and diuretics, when more than 50% of patients were started on a beta blocker. The ALLHAT trial was not included because beta blockers were not an initial treatment in this study.

The main results showed a 16% increase in the risk of stroke for beta blockers compared with other antihypertensive drugs. Beta blockers were not associated with an increase in MI but were associated with a small increase in mortality.

Meta-analysis of trials comparing beta blockers with other antihypertensives

Relative risk with beta blockers
95% CI
All-cause mortality

The results for atenolol were much more convincing than those for the other beta blockers, but Lindholm noted that there were not enough data on the other beta blockers to make any definitive statement?documentation was poor, and there were surprisingly few studies and clinical events, and it was therefore prudent to be cautious. "Atenolol does look worse than other beta blockers, but we do not have enough information on the other beta blockers to be conclusive, the confidence limits are enormous. We should therefore be cautious and assume they all behave similarly. . . . I would say that all beta blockers are suboptimal for the treatment of primary hypertension," he commented to renal wire .

Atenolol vs other antihypertensives

Relative risk with atenolol
95% CI
All-cause mortality

Nonatenolol beta blockers vs other antihypertensives

Relative risk with beta blockers
95% CI
All-cause mortality

The authors also point out that there was also "a strong tendency" toward an increased risk of stroke with the combination of beta blockers plus diuretics compared with other antihypertensives.

Combination of beta blockers plus diuretics vs other antihypertensives

Relative risk with beta blockers
95% CI
All-cause mortality

When the effect of beta blockers was compared with that of placebo or no treatment, the relative risk of stroke was reduced by 19% for all beta blockers. The authors note that this is only about half that seen in previous trials of other antihypertensive agents (the meta-analysis by Collins et al [ 4 ] frequently referred to in hypertension guidelines showed a 38% reduction in stroke). In addition, there was no reduction in MI or total mortality in the current meta-analysis. Lindholm et al comment: "To say that beta blockers do not have an effect on patients with primary hypertension would be incorrect, but clearly their effect is suboptimum."

Meta-analysis of trials comparing beta blockers with placebo or no treatment

Relative risk with beta blockers
95% CI
All-cause mortality

To download tables as slides, click on slide logo below

The authors comment: "Altogether, one must conclude that beta blockers in primary hypertension are not as effective as other antihypertensive medication, and we see no reason to limit this conclusion to atenolol."


On the question of mechanism, they point out that beta blockers have negative effects on both glucose and lipid metabolism, but that these are no more pronounced than those seen with thiazide diuretics. But they also note that although beta blockers reduce brachial blood pressure effectively, they do not lower central systolic blood pressure as much as ACE inhibitors, diuretics, or calcium antagonists, and regression of left ventricular hypertrophy is more closely correlated with central blood pressure than with brachial blood pressure.

Guidelines to be changed?

In his editorial, Beevers notes that many guidelines committees are going to have to rethink their endorsement of beta blockers as reasonable first-line drugs for the treatment of hypertension and that the US National Heart, Lung, and Blood Institute will have to rethink its proposals to include beta blockers in a long-term outcome study of the treatment of systolic hypertension.

But he warns that beta blockers should not be discontinued in all patients, and if they are discontinued, this should be done slowly. "Many patients with hypertension have overt coronary heart disease, and some may have coronary disease that is not clinically evident because they are taking a beta blocker. Sudden discontinuation, particularly from high doses, might lead to rebound angina and precipitate a myocardial infarction. Rebound hypertension is another concern," he points out. He adds that some patients also derive other benefits from beta blockers, such as those who are hyperadrenergic or hyperanxious, and therefore there will remain some hypertensive patients who require beta blockers as first-line therapy, although they should be a minority.

Addressing the issue of whether all beta blockers are affected, Beevers notes that some newer drugs in this class have other possible benefits (such as vasodilation with carvedilol and nebivolol), but they have not been subjected to long-term outcome trials in the treatment of uncomplicated hypertension, and it is unlikely that they ever will be.

Bakris: Some concerns

Commenting on the papers for renal wire , Dr George Bakris (Rush University Medical Center, Chicago, IL), hypertension editor for thekidney.org, said he agreed that beta blockers should not be used as first-line therapy in uncomplicated hypertension, but he questioned whether most of the studies in the meta-analysis were actually conducted in uncomplicated-hypertension patients. "Uncomplicated hypertension means no other risk factors. And patients with uncomplicated hypertension are actually quite a rare breed. Most of the patients in the studies included in this meta-analysis had other risk factors and were therefore not uncomplicated," he said.

He also says that using atenolol at a once-daily dose (which is what most of the studies did) does not give optimal blood-pressure reduction. He noted that this was fair enough, because that was the recommended dosage schedule, but the drug actually performed much better with a twice-daily dosage. "It is not the fault of the investigators, because the company that marketed atenolol introduced it with a phenomenal marketing ploy based on its once-daily dosage, but blood-pressure control is much better when it is given twice daily," Bakris commented.

But despite these concerns, Bakris agrees that beta blockers probably aren't the best first-choice agents, even in hypertension complicated by other risk factors. He adds, however, that they do have a role in younger hypertensive patients with high sympathetic drive (high resting pulse) and no other major issues.

Dr Franz Messerli (St Luke's-Roosevelt Hospital, New York, NY) pointed out that he had conducted a study in 1998 that questioned the effectiveness of beta blockers for hypertension, despite lowering blood pressure, they did not reduce MI or stroke [ 5 ]. "It is just amazing to me that after we clearly documented the lack of efficacy of beta blockers, it took another seven years to clearly reemphasize the same findings," he commented to renal wire . "During this time, millions of patients were exposed to the cost, inconvenience, and side effects of beta blockers without having any benefit. Clearly, as we've said before, this demands a change in the guidelines," he said. But he added that some of the newer beta blockers such as carvedilol and nebivolol, may conceivably have a more hypertension-friendly hemodynamic profile than the traditional agents such as metoprolol and atenolol



  1. Lindholm LH, Carlberg B, and Samuelsson O. Should  blockers remain first choice in the treatment of primary hypertension? A meta-analysis. Lancet 2005; 366(9496):1545-1553.

  2. Beevers DG. The end of  blockers for uncomplicated hypertension? Lancet 2005; 366(9496):1510-1512.

  3. Carlberg B, Samuelsson O, Lindholm LH. Atenolol in hypertension: Is it a wise choice? Lancet 2004; 364:1684-1689.

  4. Collins R, Peto R, MacMahon S, et al. Blood pressure, stroke, and coronary heart disease. Part 2, Short-term reductions in blood pressure: overview of randomized drug trials in their epidemiological context. Lancet 1990; 335:827-838.

  5. Messerli FH, Grossman E, Goldbourt U. Are -blockers efficacious as first-line therapy for hypertension in the elderly? JAMA 1998; 279:1903-1907.