INTRODUCTION — The large majority of patients with hypertension will require lifelong antihypertensive drug therapy to control their blood pressure. However, some patients with hypertension have well-controlled blood pressures, occasionally needing just a single medication. After a year below the target blood pressure, many wonder whether antihypertensive drug therapy can be gradually diminished or even discontinued. The issue of discontinuation of drug therapy also arises in patients who develop symptoms related to low blood pressure.
The discontinuation of antihypertensive drug therapy in patients with controlled blood pressure is discussed in this topic. Withdrawal syndromes that may develop as a result of antihypertensive drug discontinuation and the general approach to drug deprescribing are presented elsewhere. (See "Withdrawal syndromes with antihypertensive drug therapy" and "Deprescribing".)
IDENTIFYING PATIENTS FOR ANTIHYPERTENSIVE DRUG WITHDRAWAL
Identifying candidates for drug withdrawal — Diminishing or discontinuing antihypertensive medication may be possible among patients with hypertension that is well controlled for at least one year [1-4].
Among such patients in one study, approximately 40 percent remained free from antihypertensive drugs at one year after discontinuation [1,4], and approximately 25 percent remained off therapy at two years [4]. A larger fraction can successfully decrease the number and/or dose of medications taken [1,3,5].
Higher success rates for cessation of antihypertensive therapy are seen in individuals with the following characteristics:
●Lower pretreatment blood pressure (which correlates with fewer and lower doses of antihypertensive medications). Successful discontinuation is most likely among those treated with only one medication.
●Initiation of and adherence to lifestyle modifications (eg, weight loss and sodium restriction). (See "Overview of hypertension in adults", section on 'Nonpharmacologic therapy'.)
●Younger age; although successful discontinuation is less likely in older compared with younger adults, some may, nonetheless, be able to stop drug therapy (18 percent in one study) [5].
Individuals without any of these characteristics have a significantly lower likelihood of successfully discontinuing and staying off of drug therapy [2].
The value of stopping drug therapy is uncertain; medication reduction is associated with a lower incidence and/or severity of drug-induced adverse effects. However, the cost of frequent follow-up (even using home blood pressure monitoring) that is needed after discontinuation may exceed the cost of continuing a well-tolerated generic medication. This is an opportunity for "shared decision making" as some patients are eager to reduce and/or discontinue medication and are willing to undertake home blood pressure monitoring and/or more frequent office visits [6].
Several observational studies have suggested that withdrawal of antihypertensive therapy in older adults with cognitive decline may improve cognitive function and prevent progression to dementia [7,8]. However, in a randomized trial of 385 patients 75 years or older with mild cognitive impairment, no significant cardiovascular disease, and a mean blood pressure of 148/81 mmHg, gradual discontinuation of antihypertensive medications led to an increase in blood pressure at 16 weeks (by 7/3 mmHg) but no improvement in cognition, functional status, or quality of life compared with not withdrawing medications [9].
In addition, withdrawing antihypertensive therapy in older adults may lead to negative cardiovascular outcomes. In a subgroup analysis of the Hypertension in the Very Elderly Trial (HYVET), withdrawing antihypertensive therapy in already treated subjects was associated with an increase in total death, cardiovascular death, and cardiovascular events [10,11].
How to diminish or discontinue antihypertensive drug therapy — The specific strategy for diminishing or discontinuing antihypertensive drugs often depends upon the specific medication:
●For agents with a long serum elimination half-life (eg, amlodipine, chlorthalidone), taking the usual dose every other day (such as on odd-numbered days rather than daily) can be effective if the patient can successfully adhere to this regimen without forgetting doses. If the patient has difficulty remembering to take the medication on alternate days, then half the previous (usual) dose can be taken on a daily basis.
●For other drugs with shorter half-lives (eg, lisinopril), halving the dose (either with a new prescription or by using a pill cutter) is easier to implement.
We closely monitor blood pressure (eg, every two to four weeks) after reducing or discontinuing antihypertensive medications. Ideally, patients self-monitor their blood pressure at home (see "Out-of-office blood pressure measurement: Ambulatory and self-measured blood pressure monitoring"); however, when home monitoring is not feasible, office visits with a medical assistant, nurse, or clinician may be required. Some health care centers have established electronic mechanisms of relaying home blood pressures to the clinician whereas others rely on telephone communication or office visits.
If blood pressure consistently exceeds the target, then we resume the previous dose of antihypertensive medication. If, after one to three months, blood pressure remains well controlled, then we fully discontinue the medication and closely monitor the blood pressure for another one to three months.
Nonsynchronous, uncontrolled observational studies suggest that clinicians' recommendations to initiate "step-down therapy" with close follow-up is associated with less cardiovascular risk than unsupervised stopping of antihypertensive agents, especially in people with diabetes [12,13].
MECHANISMS OF SUCCESSFUL ANTIHYPERTENSIVE DRUG WITHDRAWAL — The mechanism of persistent normotension with less intensive drug treatment is incompletely understood. Long-term blood pressure control may reverse hypertension-induced arteriolar hyperplasia, thereby reducing vascular resistance directly or by reducing the sensitivity to vasoconstrictors such as angiotensin II and norepinephrine [14].
An alternate explanation is based upon observations from the first Medical Research Trial after the study was concluded [2]. Following control of hypertension for five to six years during the trial, thiazide or beta blocker therapy was either continued or replaced by placebo (figure 1). Within 9 to 12 months, patients who switched to placebo experienced a rise in blood pressure to a level equivalent to the trial's original placebo group, who remained mostly untreated. However, approximately 45 percent of patients in both of these placebo-treated groups were normotensive.
These findings suggest that effective therapy may not alter the course of the disease but that many patients being treated for mild hypertension are, in fact, normotensive or become normotensive by adopting nonpharmacologic therapies.
Thus, several factors may underlie the successful discontinuation or diminution of antihypertensive drug therapy:
●Initial misdiagnosis (of normotensive individuals) as hypertensive (eg, white coat hypertension) and initiation of unnecessary drug treatment
●Institution and maintenance of lifestyle modifications, which can lower blood pressure
●Initial use of excessive drug dosing, which could explain successful lowering of antihypertensive doses
Initial hypertension misdiagnosis — Many patients (perhaps 20 to 25 percent) who are at first diagnosed as having stage 1 hypertension are actually normotensive outside the medical office and have no hypertensive end-organ damage.
In such patients, the elevation in blood pressure reflects an acute stress response induced by visiting the clinician. This phenomenon is called "white coat" hypertension and is diagnosed using out-of-office blood pressure measurements. (See "Out-of-office blood pressure measurement: Ambulatory and self-measured blood pressure monitoring" and "White coat and masked hypertension".)
Effective lifestyle modifications — Some patients who are initially hypertensive can, with lifestyle modifications, lower their blood pressure into the normal range. The triad of dietary sodium restriction, weight reduction in the obese, and avoidance of excess alcohol intake may be particularly important when step-down therapy is undertaken [15-18].
Persistent normotension following drug withdrawal may be most likely among nonobese patients who restrict sodium and among overweight patients who lose weight [15]. One trial, for example, evaluated the effect of these lifestyle changes following the discontinuation of antihypertensive medications after several years of excellent control. At four years, 39 percent of patients assigned lifestyle modifications remained normotensive, compared with less than 5 percent of those in the control group (figure 2) [16]. In another trial of antihypertensive drug discontinuation in older adults, the composite endpoint of cardiovascular event-free survival plus persistent normotension (off of medications) at two years was twice as likely among those assigned weight loss and sodium reduction compared with those assigned usual care (50 versus 24 percent) [19]. (See "Diet in the treatment and prevention of hypertension" and "Salt intake, salt restriction, and primary (essential) hypertension" and "Overweight, obesity, and weight reduction in hypertension".)
Excessive antihypertensive medication — Years ago, initial dosing recommendations for many antihypertensive drugs were too high, leading to an increased incidence of adverse effects while producing little, if any, further reduction in blood pressure.
As an example, hydrochlorothiazide was formerly given at doses 50 to 100 mg daily (four times higher than antihypertensive doses used presently) (figure 3) [3,20-22]. Although there is often a greater diuresis with higher doses of hydrochlorothiazide, the ensuing rise in renin and angiotensin II leads to an increase in vascular resistance that may counteract the fall in blood volume. Low-dose diuretic therapy also reduces the incidence and severity of hypokalemia, hyperuricemia, hypercholesterolemia, and hyperglycemia (figure 4) [21,22]. (See "Use of thiazide diuretics in patients with primary (essential) hypertension".)
WITHDRAWAL SYNDROMES — Abrupt cessation of therapy with a short-acting beta blocker (eg, propranolol) or an alpha-2 agonist (eg, clonidine, guanfacine, guanabenz) can lead to a potentially fatal withdrawal syndrome [23]. This syndrome is characterized by increased sympathetic activity (due to adrenergic receptor upregulation that occurs during treatment, when there is decreased sympathetic activity), rebound hypertension, and possible accelerated angina or myocardial infarction. Gradual tapering and eventual discontinuation of these agents over a period of weeks should prevent this problem. This issue is discussed elsewhere. (See "Withdrawal syndromes with antihypertensive drug therapy".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hypertension in adults".)
SUMMARY
●The large majority of patients with hypertension will require lifelong antihypertensive drug therapy to control their blood pressure. However, diminishing or discontinuing antihypertensive medication may be possible in some patients with hypertension that is well controlled for at least one year. Among such patients, cessation of antihypertensive therapy is more likely to be successful in younger individuals, those with lower pretreatment blood pressure (which correlates with fewer and lower doses of antihypertensive medications), and those who initiate and adhere to lifestyle modifications. Individuals without any of these characteristics have a significantly lower likelihood of successfully stopping drug therapy. (See 'Identifying candidates for drug withdrawal' above.)
●When attempting to reduce and then discontinue antihypertensive medication, we take the following approach (see 'How to diminish or discontinue antihypertensive drug therapy' above):
•For agents with a long serum elimination half-life (eg, amlodipine, chlorthalidone), taking the usual dose every other day (such as on odd-numbered days rather than daily) can be effective if the patient can adhere to this regimen without forgetting doses.
•For other drugs with shorter half-lives (eg, lisinopril), halving the dose (either with a new prescription or by using a pill cutter) is easier to implement.
•We closely monitor blood pressure (eg, every two to four weeks) after reducing or discontinuing antihypertensive medications. Ideally, patients self-monitor their blood pressure at home; however, when home monitoring is not feasible, office visits with a medical assistant, nurse, or clinician may be required. If blood pressure consistently exceeds the target, then we resume the previous dose of antihypertensive medication. If, after one to three months, blood pressure remains well controlled, then we fully discontinue the medication and closely monitor the blood pressure for another one to three months.
●Several factors may underlie the successful discontinuation or diminution of antihypertensive drug therapy (see 'Mechanisms of successful antihypertensive drug withdrawal' above):
•Initial mislabeling (of normotensive individuals) as hypertensive (eg, white coat hypertension) and initiation of unnecessary drug treatment (see 'Initial hypertension misdiagnosis' above)
•Institution and maintenance of lifestyle modifications, which can lower blood pressure (see 'Effective lifestyle modifications' above)
•Initial use of excessive drug dosing, which could explain successful lowering of antihypertensive doses (see 'Excessive antihypertensive medication' above)
●Abrupt cessation of therapy with a short-acting beta blocker (eg, propranolol) or an alpha-2 agonist (eg, clonidine, guanfacine, guanabenz) can lead to a potentially fatal withdrawal syndrome. This issue is discussed elsewhere. (See "Withdrawal syndromes with antihypertensive drug therapy".)
ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Norman M Kaplan, MD, who contributed to an earlier version of this topic review.
