Atenolol in hypertension: is it a wise choice?

Bo Carlberg, Ola Samuelsson, Lars Hjalmar Lindholm


Source - Lancet 2004; 364: 1684-89


Department of Public Health and Clinical Medicine, Umeå University Hospital, SE 901 85 Umeå, Sweden (B Carlberg MD, Prof L H Lindholm MD); and Department of Nephrology, Sahlgrenska University Hospital, Göteborg, Sweden (O Samuelsson MD)


Correspondence to: Prof Lars H Lindholm LarsHLindholm@fammed. umu.se

Summary

Background Atenolol is one of the most widely used ßblockers clinically, and has often been used as a reference drug in randomised controlled trials of hypertension. However, questions have been raised about atenolol as the best reference drug for comparisons with other antihypertensives. Thus, our aim was to systematically review the effect of atenolol on cardiovascular morbidity and mortality in hypertensive patients.

Methods Reports were identified through searches of The Cochrane Library, MEDLINE, relevant textbooks, and by personal communication with established researchers in hypertension. Randomised controlled trials that assessed the effect of atenolol on cardiovascular morbidity or mortality in patients with primary hypertension were included.

Findings We identified four studies that compared atenolol with placebo or no treatment, and five that compared atenolol with other antihypertensive drugs. Despite major differences in blood pressure lowering, there were no outcome differences between atenolol and placebo in the four studies, comprising 6825 patients, who were followed up for a mean of 4·6 years on all-cause mortality (relative risk 1·01 [95% CI 0·89-1·15]), cardiovascular mortality (0·99 [0·83-1·18]), or myocardial infarction (0·99 [0·83-1·19]). The risk of stroke, however, tended to be lower in the atenolol than in the placebo group (0·85 [0·72-1·01]). When atenolol was compared with other antihypertensives, there were no major differences in blood pressure lowering between the treatment arms. Our meta-analysis showed a significantly higher mortality (1·13 [1·02-1·25]) with atenolol treatment than with other active treatment, in the five studies comprising 17671 patients who were followed up for a mean of 4·6 years. Moreover, cardiovascular mortality also tended to be higher with atenolol treatment than with other antihypertensive treatment. Stroke was also more frequent with atenolol treatment.

Interpretation Our results cast doubts on atenolol as a suitable drug for hypertensive patients. Moreover, they challenge the use of atenolol as a reference drug in outcome trials in hypertension.