A large analysis of antihypertensive trials suggests some blood pressure drugs are easier to live with than others – and combination therapy need not mean more side effects.
Angiotensin receptor blockers have emerged as the best-tolerated blood pressure drugs in a large sweeping analysis of more than 700 randomised trials, challenging the assumption that adding antihypertensives inevitably adds side effects.
The network meta-analysis, published in JAMA, included 716 double-blind randomised trials involving 159,362 participants and compared adverse effects and treatment discontinuation across the five major classes of blood pressure-lowering drugs and their combinations.
“Blood pressure (BP) control rates among individuals with hypertension remain poor globally,” the researchers wrote.
“Concerns about the tolerability of BP-lowering drugs are a major contributor to undertreatment and treatment inertia. These concerns often result in reluctance to start therapy, especially combination therapy, and to intensify therapy.
“Additionally, these concerns are frequently cited as a reason to stop treatment.”
The standout performer was an angiotensin II receptor blocker (ARB) combined with a calcium channel blocker (CCB), which ranked as the best-tolerated regimen. Four of the five highest-ranked regimens contained an ARB.
Some combination treatments were better tolerated than monotherapy and some regimens were associated with fewer withdrawals for adverse events than placebo.
Side effects are an important contributor to antihypertensive non-adherence, which remains a major barrier to effective blood pressure control. Hypertension is a leading cardiovascular risk factor and lowering blood pressure substantially reduces cardiovascular events, yet treatment rates remain poor.
An accompanying JAMA editorial noted that between 30% and 80% of people newly prescribed an antihypertensive stop the medication during the first year.
“Among all patients treated with antihypertensive medications, adherence rates are 75% to 90%, based on pharmacy refill data. Because hypertension is an asymptomatic condition and blood pressure medications do not confer immediate benefits, there are fewer incentives for patients experiencing adverse effects to adhere to antihypertensive medications,” wrote Dr Mary McDermott and Dr Stephen Persell of the Northwestern University Feinberg School of Medicine.
“Nonadherence to antihypertensive medications is more common among people who report adverse effects and is associated with poorer blood pressure control.”
The researchers, led by Dr Nelson Wang of NSW-based The George Institute for Global Health, examined trials of ACE inhibitors, ARBs, beta blockers, CCBs, thiazide and thiazide-like diuretics and combinations of these drugs.
Trials lasted between four and 26 weeks, with an average follow-up of 8.6 weeks. Participants had a mean age of 54.6 years, 44% were women and their mean baseline blood pressure was 158/100mmHg.
The primary outcome was treatment discontinuation because of an adverse event, rather than simply the occurrence of a symptom. Secondary outcomes included headache, dizziness, oedema, and cough.
Compared with placebo, ARB monotherapy was associated with significantly fewer treatment discontinuations due to adverse events, with an odds ratio of 0.73 and an absolute risk difference of -0.8%.
The combination of an ARB and CCB did even better, with an odds ratio of 0.61 and an absolute risk difference of -1.2%.
In contrast, CCB monotherapy was associated with significantly more treatment discontinuation than placebo, with an odds ratio of 1.43 and an absolute increase of 1.2%.
ACE inhibitor plus CCB treatment was also associated with more discontinuations (OR 1.46), as was beta blocker plus thiazide therapy (OR 1.58).
When the researchers ranked treatments according to tolerability, ARB plus CCB came first, followed by ARB plus beta blocker, ARB monotherapy, CCB plus thiazide, and ARB plus thiazide.
The findings also complicated the conventional picture of antihypertensive “side effects”, the researchers noted.
All regimens significantly increased dizziness compared with placebo, with the risk generally greater with combination treatment than monotherapy. But almost all antihypertensive regimens also reduced headaches.
Headache occurred in about 10% of placebo-treated participants, and ARB plus CCB therapy was associated with an absolute 3.9 percentage-point reduction compared with placebo.
The researchers said that tolerability therefore reflected the net balance between symptoms caused and symptoms relieved by treatment.
“Virtually all” regimens increased dizziness and hypotension-related symptoms, while individual classes carried characteristic adverse effects such as cough with ACE inhibitors and oedema with CCBs. At the same time, all regimens apart from CCB monotherapy reduced headache.
Oedema was particularly associated with CCB-containing regimens, while treatments containing a diuretic generally ranked lowest for oedema. CCB tolerability also appeared dose-dependent, with discontinuation due to adverse events increasing at higher doses.
Related
The researchers said the results suggested clinicians should not necessarily assume that combination treatment would be harder for patients to tolerate.
Indeed, ARB plus CCB therapy was associated with substantially fewer adverse-event discontinuations than CCB monotherapy, with an odds ratio of 0.43 and an absolute risk difference of -2.4%.
Analyses using data from US FDA drug approval submissions also supported lower discontinuation rates with both ARB monotherapy and ARB plus CCB compared with placebo.
But the accompanying editorial urged clinicians not to turn the tolerability rankings into a new prescribing hierarchy.
Dr McDermott and Dr Persell said differences in overall tolerability should not override indications for particular drugs in patients with conditions where specific classes had established benefits. These included ACE inhibitors or ARBs in patients with diabetes and albuminuria or heart failure with reduced ejection fraction.
They also stressed that the meta-analysis did not assess cardiovascular outcomes and included trials lasting only four to 26 weeks, meaning longer-term adverse-effect profiles could differ.
Nor should doctors switch patients who are doing well on existing treatment simply because another regimen ranked more highly.
“Results from the study by Wang et al. can help inform clinicians’ selection of antihypertensive therapies for patients initiating medications for hypertension, particularly when comorbidities, such as the presence of diabetes with microvascular disease, do not warrant a specific therapy,” Dr McDermott and Dr Persell wrote.
“The study by Wang et al. should not prompt changes in therapies for patients taking stable antihypertensive therapies without adverse effects and for whom blood pressure is controlled.
“Similarly, the study by Wang et al. should not alter antihypertensive therapy selection for patients with comorbid diseases known to benefit from specific classes of antihypertensive therapy, such as those with atrial fibrillation and rapid ventricular response.”
Current guidelines recommend thiazide diuretics, ACE inhibitors, ARBs, or CCBs as initial treatment options for hypertension.
But for a patient starting therapy without a comorbidity dictating drug choice, the editorialists said the new evidence suggested an ARB may reduce the chance that treatment will be stopped.
For patients particularly prone to headache or lower-extremity oedema, they suggested thiazide diuretics, ACE inhibitors, or ARBs may be preferable to CCBs.
The bigger message, they said, was less about crowning a winning drug than paying attention to what patients experienced after the prescription was written.
“The work by Wang et al. serves as a reminder to ask patients about adverse effects that may be associated with antihypertensive therapy, particularly because adverse effects can impair adherence, which in turn may reduce the ability of the antihypertensive medications to prevent cardiovascular events,” they concluded.
JAMA, May 2026 (research)
JAMA, June 2026 (editorial)
