Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Effects of blood-pressure-lowering treatment in hypertension

9. Discontinuations for adverse events attributed to different classes of antihypertensive drugs

meta-analyses of randomized trials

Thomopoulos, Costas; Parati, Gianfranco; Zanchetti, Alberto

doi: 10.1097/HJH.0000000000001052

Background: Choice of antihypertensive drugs is also based on the expected burden of adverse events associated with each class of agents, and we have recently identified treatment discontinuation for adverse events as a measure of treatment tolerability frequently reported in randomized controlled trials (RCTs).

Objectives: To investigate whether all classes of blood pressure (BP) lowering drugs increase discontinuations for adverse events when compared with placebo and whether risk of discontinuation is similar for all classes when compared in head-to-head RCTs.

Methods: RCTs of BP-lowering treatment were subdivided in groups according to class of drug compared with placebo or with other classes. Risk ratios and 95% confidence intervals of major cardiovascular events and of treatment discontinuations for adverse events were calculated (random-effects model).

Results: Thirty-eight placebo-controlled RCTs (147 788 patients) and 37 head-to-head RCTs (242 481 patients) provided comparative information on discontinuations for adverse events. All classes of drugs significantly increased discontinuations for adverse events over those occurring on placebo: risk ratio diuretics 2.23 (1.32–3.76), beta-blockers 2.88 (1.58–5.28), calcium antagonists 2.03 (1.17–3.56), angiotensin-converting enzyme inhibitors 2.78 (1.37–5.47), central agents 1.74 (1.24–2.45), with the single exception of angiotensin receptor blockers, which did not significantly increase adverse events over placebo [risk ratio 1.13 (0.78–1.62)]. Similarly, in head-to-head comparison RCTs with other classes, angiotensin receptor blockers were the only class associated with a significantly lower risk of adverse events [risk ratio 0.71 (0.58–0.87)] when head-to-head compared with other classes. Regression analysis also shows that incidence of discontinuations for adverse events is proportional to the number of antihypertensive and other cardiovascular drugs, which accounts for the high incidence of this outcome often found in groups randomized to placebo.

Conclusion: Reduction of cardiovascular events by all classes of BP-lowering drugs is accompanied by increased treatment discontinuations for adverse events, except when angiotensin receptor blockers are used. Treatment discontinuations are also related to treatment often accompanying antihypertensive agents.

aDepartment of Cardiology, Helena Venizelou Hospital, Athens, Greece

bDepartment of Cardiovascular, Neural and Metabolic Sciences, San Luca Hospital, Istituto Auxologico Italiano IRCCS

cDepartment of Medicine and Surgery, University of Milan Bicocca

dScientific Direction, Istituto Auxologico Italiano IRCCS

eCentro Interuniversitario di Fisiologia Clinica e Ipertensione, Università degli Studi di Milano, Milan, Italy

Correspondence to Prof Alberto Zanchetti, Direzione Scientifica, Istituto Auxologico Italiano, Via L. Ariosto, 13, I-20145 Milano, Italy. Tel: +39 02 619112237; fax: +39 02 619112901; e-mail:

Abbreviations: ACE, angiotensin-converting enzyme; AE, adverse events; BP, blood pressure; CHD, coronary heart disease; CI, confidence interval; NNT, number needed to treat; RCT, randomized controlled trial; RR, risk ratio

Received 13 May, 2016

Revised 16 June, 2016

Accepted 28 June, 2016

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.