INTRODUCTION — There is general agreement as to which antihypertensive drugs are appropriate for initial therapy in most patients with hypertension. The major options are:

●Thiazide-type diuretics

●Angiotensin-converting enzyme (ACE) inhibitors/angiotensin II receptor blockers (ARBs)

●Calcium channel blockers

Recommendations concerning the use of particular agents for the initial treatment of hypertension will be presented here. The diagnosis and evaluation of hypertension, as well as a discussion of goal blood pressure in the treatment of hypertension, are presented separately:

●(See "Overview of hypertension in adults".)

●(See "Initial evaluation of the hypertensive adult".)

●(See "Goal blood pressure in adults with hypertension".)

●(Related Pathway(s): Hypertension: Initial management.)

RELATIVE EFFICACY OF ANTIHYPERTENSIVE DRUGS — Each of the antihypertensive therapy classes is roughly equally effective in lowering the blood pressure, producing a good antihypertensive response in 30 to 50 percent of patients (figure 1A-B) [1-5]. There is, however, wide interpatient variability as many patients will respond well to one drug but not to another. In addition, there are few clinical parameters that reliably predict individual responses to one drug or another. (See 'Initial monotherapy' below.)

Importance of attained blood pressure — Meta-analyses, the 2017 American College of Cardiology/American Heart Association statement on the treatment of blood pressure [6], the 2020 International Society of Hypertension (ISH) Global Hypertension Practice Guidelines [7], and the 2018 European Society of Hypertension/European Society of Cardiology guidelines on the management of hypertension [8] all concluded that the amount of blood pressure reduction is the major determinant of reduction in cardiovascular risk in both younger and older patients with hypertension, not the choice of antihypertensive drug (assuming that the patient does not have an indication for a particular drug, such as diltiazem, verapamil, or a beta blocker for rate control in atrial fibrillation) [1,2,9-13].

This conclusion was based upon the finding in a number of large randomized trials that, at the same level of blood pressure control, most antihypertensive drugs provide the same degree of cardiovascular protection. As an example, the CAPPP, STOP-Hypertension-2, NORDIL, UKPDS, and INSIGHT trials found little overall difference in outcomes between older (such as diuretics and beta blockers) and newer antihypertensive drugs (such as angiotensin-converting enzyme [ACE] inhibitors and calcium channel blockers) [3,14], and the CAMELOT trial found no significant difference in outcomes with amlodipine and enalapril, two newer antihypertensive drug classes at that time [15]. Similar conclusions have been reached in patients at increased cardiovascular risk.

When differences in outcomes have been noted in trials comparing different antihypertensive drugs, the treatment strategy that produced better outcomes had also produced better blood pressure control. As examples:

●The ASCOT trial found a lower rate of cardiovascular disease and death with a calcium channel blocker (amlodipine) compared with a beta blocker (atenolol). However, patients in the amlodipine arm had a significantly lower mean blood pressure at the end of the study (3/2 mmHg) [16].

●Ramipril and perindopril produced better outcomes than placebo in the HOPE and EUROPA trials of patients at increased cardiovascular risk, but the blood pressure was significantly lower in the treated patients: 3.3/1.4 mmHg (with a greater difference overnight) in HOPE and 5/2 mmHg in EUROPA [17,18].

●In the VALUE trial of over 15,000 patients who had either prior atherosclerotic cardiovascular disease or at least one cardiovascular risk factor, amlodipine produced better outcomes than valsartan but also greater blood pressure reduction [19]. When 5000 pairs were matched exactly for systolic blood pressure and other risk factors, the two groups had nearly identical rates of cardiovascular events [20].

Possible exceptions to these general findings were thought to come from the ALLHAT and ACCOMPLISH trials.

ALLHAT trial — The ALLHAT trial randomly assigned over 41,000 hypertensive patients (mean blood pressure 146/84 mmHg) with at least one other coronary risk factor to one of four initial regimens: chlorthalidone (12.5 to 25 mg/day), amlodipine, lisinopril, or doxazosin; the doxazosin arm was prematurely terminated due to an increased risk of heart failure [21]. At a mean follow-up of 4.9 years, the primary outcome (fatal coronary heart disease or nonfatal myocardial infarction) was the same in the three arms (figure 2) [21]. However, the chlorthalidone arm had a significantly lower rate of heart failure than amlodipine and lisinopril (figure 3) and a significantly lower rate of combined cardiovascular disease outcomes than lisinopril (figure 4).

It seems likely that the benefits seen with chlorthalidone were due at least in part to an earlier and greater degree of blood pressure reduction, similar to the findings with amlodipine in the VALUE trial described in the preceding section [20]. Chlorthalidone was associated with a small but significantly lower systolic pressure over the course of the study than amlodipine or lisinopril (133.9 versus 134.7 and 135.9 mmHg, respectively) and a higher proportion of patients who attained the blood pressure goal of less than 140/90 mmHg (68.2 versus 66.3 and 61.2 percent, respectively). The difference in mean systolic blood pressure was most pronounced in the first two years (136.4 versus 137.8 and 139.2 mmHg, respectively).

Twenty-four-hour blood pressure monitoring was not obtained in ALLHAT, which may have been important since chlorthalidone is long acting, while the effect of lisinopril may diminish toward the end of the day, especially at doses of 10 mg/day. One cannot exclude the possibility that the observed benefits with chlorthalidone in ALLHAT were due at least in part to the lower attained blood pressure, which is in keeping with the findings in other trials cited in the preceding section.

Looking at the data in another way, there was no difference in the primary end point among the three treated groups (figure 2), despite the lower attained blood pressure with chlorthalidone. This raises the possibility that thiazide diuretics may actually be associated with worse outcomes at the same attained blood pressure.

Thiazide-like versus thiazide-type diuretics — Chlorthalidone and indapamide, commonly used thiazide-like diuretics, are significantly more potent antihypertensive agents than hydrochlorothiazide, a commonly used thiazide-type diuretic, at similar dose levels [22-27]. In a meta-analysis of 14 trials that compared the blood pressure reduction with one of three dose levels of hydrochlorothiazide (low, intermediate, high) to a similar dose level of one of the thiazide-like diuretics, systolic pressure reduction was greater with chlorthalidone and indapamide (by 3.6 and 5.1 mmHg, respectively) [22].

This may not be so important for efficacy, since the dose-response curve for thiazide diuretics in the treatment of primary hypertension (formerly called "essential" hypertension) is relatively flat (figure 5) [2,28,29]. However, metabolic complications, such as hypokalemia, glucose intolerance, and hyperuricemia increase with dose (figure 6) [2,28,29]. In two major trials of low-dose chlorthalidone (12.5 to 25 mg/day), treatment for hypokalemia was required in 7 to 8 percent of patients [21,30]. (See "Use of thiazide diuretics in patients with primary (essential) hypertension" and "Causes of hypokalemia in adults", section on 'Diuretics'.)

A possibly more important difference than potency is the longer duration of action of chlorthalidone and indapamide (24 or more hours versus 6 to 12 hours with hydrochlorothiazide) (table 1) [23-25]. This may not affect office blood pressure if the medication is taken in the morning but may result in a greater fall in nighttime blood pressure with chlorthalidone compared to baseline (eg, -13.5 mmHg with 12.5 mg/day [force titrated to 25 mg/day] versus -6.4 mmHg with 25 mg/day [force titrated to 50 mg/day] of hydrochlorothiazide in a small randomized, crossover trial) [24].

The lesser efficacy of hydrochlorothiazide throughout the day was also demonstrated in a study in which both 24-hour ambulatory monitoring and office blood pressure were performed in 228 hypertensive patients at baseline and after four weeks of therapy with hydrochlorothiazide (25 mg/day) [31]. The mean four-week decrease in blood pressure was significantly greater when measured with office readings than with 24-hour ambulatory monitoring (14/8 versus 10/6 mmHg).

Efficacy in preventing cardiovascular events — Nearly all cardiovascular outcome trials in hypertension that examined diuretics compared chlorthalidone or indapamide with placebo or active comparators. However, there are no randomized trials that directly compare cardiovascular outcomes in hypertensive patients treated with hydrochlorothiazide (or other thiazide-type diuretics such as chlorothiazide) versus chlorthalidone (or other thiazide-like diuretics such as indapamide). In the absence of such head-to-head trials, a multiple-treatment (network) meta-analysis of 21 trials including approximately 120,000 patients was conducted in which thiazide-type diuretics (such as hydrochlorothiazide) and thiazide-like diuretics (such as chlorthalidone) were indirectly compared by evaluating their efficacy against placebo or common comparator drugs (eg, ACE inhibitors were compared with hydrochlorothiazide in ANBP2 and with chlorthalidone in ALLHAT) [32]. After controlling for achieved blood pressure, thiazide-like but not thiazide-type diuretics reduced the risk of cardiovascular events compared with placebo (relative risk 0.88, 95% CI 0.79 to 0.98) and heart failure (relative risk 0.71, 95% CI 0.57 to 0.89). Compared with thiazide-type diuretics, thiazide-like diuretics significantly lowered the relative risk of cardiovascular events by 12 percent and heart failure by 21 percent.

The results of this meta-analysis are consistent with an earlier network meta-analysis [33] and with observational data from the Multiple Risk Factor Intervention Trial (MRFIT) [34]. Among hypertensive men in MRFIT, 2392 were treated with chlorthalidone and 4049 were treated with hydrochlorothiazide. During six years of follow-up, cardiovascular events (defined as myocardial infarction, stroke, coronary artery bypass surgery, heart failure, left ventricular hypertrophy, peripheral artery disease, or angina) were significantly less common with chlorthalidone compared with hydrochlorothiazide (hazard ratio 0.79, 95% CI 0.68 to 0.92). Through the course of the study, systolic blood pressure and LDL cholesterol levels were also lower with chlorthalidone compared with hydrochlorothiazide.

By contrast, other observational studies suggest that chlorthalidone and hydrochlorothiazide lead to similar rates of cardiovascular events but that chlorthalidone increases the risk of adverse metabolic effects [35,36]. As an example, in a retrospective study of more than 700,000 hypertensive patients prescribed one of these two drugs, chlorthalidone was not associated with a reduced risk of myocardial infarction, stroke, or heart failure (hazard ratio 1.0, 95% CI 0.85-1.17) compared with hydrochlorothiazide, but was associated with an increased risk of hypokalemia (hazard ratio 2.72, 95% CI 2.38-3.12), hyponatremia (hazard ratio 1.31, 95% CI 1.16-1.47), and acute kidney injury (hazard ratio 1.37, 95% CI 1.15-1.63) [35]. Nevertheless, the trial evidence supporting the efficacy of low-dose thiazide diuretics in the management of hypertension comes primarily from those using chlorthalidone, such as ALLHAT [21]. There is little if any trial evidence that hydrochlorothiazide alone in a dose of 12.5 to 25 mg/day reduces cardiovascular events [37-39], and the blood pressure may not be as well controlled overnight [24,40].

Choice between thiazide-like and thiazide-type diuretics — Based upon the above observations, we and other experts suggest that thiazide-like diuretics (such as chlorthalidone, 12.5 to 25 mg/day) are preferred to thiazide-type diuretics [25,37-39,41,42]. However, the choice may vary with the clinical setting:

●In most patients not previously treated with a thiazide diuretic, we suggest 12.5 to 25 mg/day of chlorthalidone (or 1.25 to 5 mg/day of indapamide), rather than hydrochlorothiazide. However, among older patients who are less than 10 mmHg above goal blood pressure, low-dose hydrochlorothiazide is a reasonable alternative.

●Among patients already treated with low-dose hydrochlorothiazide, the optimal approach has not been defined. Some experts would switch all patients to chlorthalidone or indapamide at their next visit, with the possible exception of those who monitor their blood pressure at home and have values below goal at the end of the drug's dosing period.

Monitoring for hypokalemia — Chlorthalidone produced hypokalemia requiring therapy in 7 to 8 percent of patients in large clinical trials including ALLHAT and SHEP [21,30]. It is possible that hypokalemia is more common with chlorthalidone than hydrochlorothiazide [26], given its longer duration of action. Concurrent use of a low-salt diet will both contribute to blood pressure lowering and reduce the risk of hypokalemia [43,44]. (See "Causes of hypokalemia in adults", section on 'Diuretics'.)

Monitoring for the development of hypokalemia is warranted with all thiazide diuretics. In stable patients on a fixed dose of either chlorthalidone or hydrochlorothiazide, potassium loss, like other diuretic-induced fluid and electrolyte complications, occurs only during the first two weeks of therapy before a new steady state is established (figure 7). Thus, a stable patient with a normal serum potassium concentration at three weeks is not at risk of late hypokalemia unless the diuretic dose is increased, extrarenal potassium losses increase, or dietary potassium intake is reduced. (See "General principles of disorders of water balance (hyponatremia and hypernatremia) and sodium balance (hypovolemia and edema)", section on 'The steady state' and "Time course of loop and thiazide diuretic-induced electrolyte complications".)

Issues with thiazide-like diuretics — The basic principles of monitoring for hypokalemia with chlorthalidone and indapamide are identical to those with hydrochlorothiazide, as described in the preceding section.

There are two other potential limitations associated with chlorthalidone:

●There is no 12.5 mg chlorthalidone tablet. Thus, 25 mg tablets of generic chlorthalidone need to be cut in half.

●In patients who require combination therapy, fixed dose combination pills of chlorthalidone with ACE inhibitors and long-acting calcium channel blockers are not available (in contrast to hydrochlorothiazide). However, chlorthalidone (at a dose of either 12.5 or 25 mg) is available in combination with azilsartan medoxomil, an ARB. (See 'Combination therapy' below.)

ACCOMPLISH trial — Only one major trial, ACCOMPLISH, directly compared different combination regimens in hypertensive patients who require two drugs [45].

The ACCOMPLISH trial included 11,506 patients with hypertension who were at high risk for a cardiovascular event and, despite prior antihypertensive therapy in 97 percent (most requiring two or more drugs), had a mean baseline blood pressure of 145/80 mmHg [46]. The patients were randomly assigned to initial combination therapy with benazepril (20 mg/day) plus either amlodipine (5 mg/day) or hydrochlorothiazide (12.5 mg/day). Benazepril was increased to 40 mg/day in both groups at one month. If goal blood pressure was not attained, the amlodipine dose was increased to 10 mg/day and the hydrochlorothiazide dose to 25 mg/day.

The primary end point was measured as the time to the first event, which was a composite of death from cardiovascular causes, nonfatal myocardial infarction, nonfatal stroke, hospitalization for angina, resuscitation after sudden cardiac death or coronary revascularization. The trial was terminated early upon recommendation of the Data Safety Monitoring Board at a mean follow-up of 36 months when a substantial disadvantage associated with the hydrochlorothiazide arm was noted and the prespecified stopping rule was exceeded.

The primary end point was achieved significantly less often in the benazepril-amlodipine group (9.6 versus 11.8 percent, hazard ratio 0.80, 95% CI 0.72-0.90). However, in a prespecified, post-hoc analysis, the superiority of amlodipine-based therapy was most pronounced in nonobese individuals; both combinations produced similar outcomes among obese patients [47].

Benazepril-amlodipine therapy led to a similar reduction in the secondary end point of cardiovascular death or nonfatal myocardial infarction or stroke (5 versus 6.3 percent, hazard ratio 0.79). These benefits increased progressively over the duration of the trial. The development of chronic kidney disease (mostly defined as doubling of the serum creatinine) was also less common with benazepril-amlodipine (2 versus 3.7 percent) [48].

The mean office blood pressure was slightly (approximately 1 mmHg) but significantly lower in the benazepril-amlodipine group (131.6/73.3 compared to 132.5/74.4 mmHg). However, in contrast to all other major randomized trials that compared the outcomes with different antihypertensive drugs, ACCOMPLISH included 24-hour blood pressure monitoring in a subset of 573 patients [49]. The 24-hour average blood pressure was higher (1.6/0.3 mmHg) in the benazepril-amlodipine group, although this was not statistically significant. Similar trends were noted with daytime and nighttime average blood pressures.

Thus, the clinical benefits observed with the benazepril-amlodipine combination cannot be explained by better blood pressure control. The difference in outcome can be explained in one or both of two ways: there is a beneficial effect of benazepril-amlodipine or an adverse effect of benazepril-hydrochlorothiazide. There is no way to distinguish between these possibilities. In addition, it is not known if amlodipine plus benazepril is associated with better outcomes than other combinations.

Implications for practice — The ACCOMPLISH trial was large, well designed, and without apparent flaws. In addition, it compared combination regimens using the three classes of antihypertensive drugs that are preferred for initial monotherapy in the absence of a specific indication for a particular drug class. (See 'Initial monotherapy' below and 'Clinical reasons for specific drugs' below.)

Some have suggested that confirmatory trials are required before practice recommendations are changed. However, such information is unlikely to be available for many years.

We and our peer reviewers feel that the results of ACCOMPLISH should not be ignored and that they raise the following questions about the use of a long-acting ACE inhibitor/ARB and a long-acting dihydropyridine calcium channel blocker:

●Is the long-acting dihydropyridine calcium channel blocker combination the preferred regimen in previously untreated patients who require two drugs because they are more than 20/10 mmHg above goal?

●Should patients being treated with the combination of an ACE inhibitor/ARB and a thiazide diuretic who are at goal and doing well be switched to long-acting dihydropyridine calcium channel blocker? Approximately 75 percent of patients in ACCOMPLISH had previously been treated with two or more antihypertensive drugs [46].

●In patients being treated with and having responded to a thiazide diuretic who require a second antihypertensive drug, should the thiazide be discontinued and the patient started on long-acting dihydropyridine calcium channel blocker?

●Does ACCOMPLISH affect the choice of monotherapy, with a long-acting ACE inhibitor/ARB or a long-acting dihydropyridine calcium channel blocker being preferred so that the second class can be added if the patient responds but does not reach goal with the initial drug?

These issues will be discussed below. (See 'Combination therapy' below and 'Summary and recommendations' below.)

INITIAL MONOTHERAPY — Initial monotherapy is successful in many patients with mild primary hypertension. However, single-drug therapy is unlikely to attain goal blood pressure in patients whose blood pressures are more than 20/10 mmHg above goal. In such patients, initial combination therapy using two drugs is recommended [50]. (See 'First-line combination therapy' below.) (Related Pathway(s): Hypertension: Initial management.)

Goal blood pressure is discussed in detail elsewhere. (See "Goal blood pressure in adults with hypertension".)

General principles — Each of the antihypertensive agents is roughly equally effective in lowering the blood pressure, producing a good antihypertensive response in 30 to 50 percent of patients (figure 1A-B) [1-5]. There is, however, wide interpatient variability as many patients will respond well to one drug but not to another. There are some predictable differences, such as Black patients and older patients generally responding better to monotherapy with a thiazide diuretic or calcium channel blocker and relatively poorly to an angiotensin-converting enzyme (ACE) inhibitor or beta blocker (figure 1B). (See 'Monotherapy based upon age and race' below.) (Related Pathway(s): Hypertension: Initial management.)

In addition to these general observations, the following findings were noted in a 2009 meta-analysis of randomized trials [2]:

●Defining the standard dose of a class of drugs as the usual maintenance dose in reference pharmacopoeias, the largest reduction in blood pressure was seen at a half-standard dose with only modestly greater reductions in systolic and diastolic blood pressures at standard or twice-standard doses. As examples, the average fall in systolic blood pressure over 24 hours with half-standard, standard, and twice-standard doses was 7.1, 9.1, and 10.9 mmHg with data from all classes combined and 7.4, 8.8, and 10.3 mmHg with a thiazide diuretic. (See 'Drug dosing and drug frequency' below.)

●With thiazide diuretics, calcium channel blockers, and beta blockers, the rate of symptomatic and metabolic adverse effects increased significantly with standard or twice-standard doses compared to half-standard doses. Similar findings have been noted in other studies (figure 6). By contrast, there was a very low rate of side effects with ACE inhibitors and angiotensin II receptor blockers (ARBs) with no dose dependence.

Thus, after the initial dose, going to higher doses produced on average relatively small further reductions in blood pressure at the price of an increasing rate of adverse effects. As a result, we generally limit dose titration to one step with a given antihypertensive drug (eg, 12.5 to 25 mg of chlorthalidone and 5 to 10 mg of amlodipine).

These observations suggest that two or even three drugs at half-standard doses might have greater antihypertensive efficacy and less toxicity than one drug at standard or twice-standard doses and might produce better patient outcomes [2]. Randomized trials to validate this hypothesis have not been performed.

Choice of drug — Some patients have clinical reason to take a specific drug or drugs that is unrelated to primary hypertension (eg, a nondihydropyridine calcium channel blocker or beta blocker for rate control in patients with atrial fibrillation). (See 'Clinical reasons for specific drugs' below.) (Related Pathway(s): Hypertension: Initial management.)

In the absence of a specific clinical reason, there are three main classes of drugs that have been used for initial monotherapy: thiazide diuretics, long-acting calcium channel blockers (most often a dihydropyridine), and ACE inhibitors or ARBs (algorithm 1). Each of these classes of drugs has been equally effective in monotherapy trials if the attained blood pressure is similar. Beta blockers are not commonly used for initial monotherapy in the absence of a specific indication since they may have an adverse effect on some outcomes compared with other agents, particularly in older patients. (See 'Clinical reasons for specific drugs' below and 'Importance of attained blood pressure' above and 'Beta blockers' below.)

Among patients in whom there is a reasonable likelihood of requiring a second drug (eg, more than 10/5 mmHg above goal), some clinicians prefer initial therapy with a long-acting ACE inhibitor/ARB or a long-acting dihydropyridine calcium channel blocker since the second class can be added if additional therapy is required to achieve the desired combination regimen (algorithm 1). As described in the next section, the choice between these drug classes may be influenced by age and race. (See 'ACCOMPLISH trial' above.)

This is different from the current practice of many clinicians. Low-dose hydrochlorothiazide (12.5 to a maximum of 25 mg/day) is widely used and, after publication of the ALLHAT trial, was recommended as initial monotherapy in most patients with mild primary hypertension [51,52]. However, hydrochlorothiazide was not examined in ALLHAT and appears to be less effective and has a shorter duration of action than chlorthalidone, and there is little, if any, evidence that low-dose hydrochlorothiazide alone reduces cardiovascular events as opposed to the evidence with chlorthalidone. (See 'Thiazide-like versus thiazide-type diuretics' above.)

Thus, when a diuretic is used either as initial therapy or in combination with other antihypertensive therapies, we and others suggest indapamide (1.25 to 2.5 mg/day) or chlorthalidone (12.5 to a maximum of 25 mg/day) [37,38], which produced the best outcomes in ALLHAT, rather than hydrochlorothiazide at the same doses (algorithm 1). Clinicians should be aware that chlorthalidone is associated with somewhat greater risks of hypokalemia, glucose intolerance, and new onset diabetes mellitus than hydrochlorothiazide [53].

Monotherapy based upon age and race — The likelihood of a good clinical blood pressure response is increased when two simple clinical characteristics, age and race, are utilized to determine drug treatment. The following patients respond best to different types of antihypertensive agents used as monotherapy [54,55]:

●Younger patients (eg, <50 years) respond best to ACE inhibitors or ARBs and beta blockers. However, beta blockers are not commonly used for initial monotherapy in the absence of a specific indication because they appear to provide inferior protection against stroke risk. (See 'Beta blockers' below.)

Support for this differential antihypertensive response in younger patients is supported by a study of 56 young (22 to 51 years) hypertensive White patients who were treated in a crossover rotation with the four main classes of antihypertensive drugs: ACE inhibitor, thiazide diuretic, long-acting dihydropyridine CCB, and beta blocker [54]. Significantly greater responses in both systolic and diastolic blood pressure levels were noted with the ACE inhibitor and beta blocker than with the CCB or diuretic.

●Black patients (figure 1B) and older adults (ie, age ≥60 years) often respond best to a thiazide diuretic or long-acting calcium channel blocker [5,11,56,57]. However, many older hypertensive patients have a specific indication for an ACE inhibitor or ARB, including heart failure, prior myocardial infarction, and proteinuric chronic kidney disease. In the absence of such indications, thiazide diuretics or long-acting calcium channel blockers are reasonable initial drugs. (See "Treatment of hypertension in Black individuals" and "Treatment of hypertension in older adults, particularly isolated systolic hypertension" and 'Clinical reasons for specific drugs' below.) (Related Pathway(s): Hypertension: Initial management.)

Sequential monotherapy — In patients who can be initially treated with monotherapy (ie, who are less than 20/10 mmHg above goal blood pressure) and who do not respond well to a moderate dose of antihypertensive therapy, we recommend a strategy of sequential monotherapy. This recommendation differs from the approach outlined by the eighth Joint National Committee (JNC-8); the JNC-8 panel suggested pushing the initial drug in such patients to the maximum recommended dose or adding a second drug, rather than determining the potential efficacy of sequential monotherapy [57].

Each of the recommended first-line agents will normalize the blood pressure in 30 to 50 percent of patients with mild hypertension (figure 1A-B) [1-5]. A patient who is relatively unresponsive to one drug has an almost 50 percent likelihood of becoming normotensive on a second drug [58]. Thus, in a patient who has little or no fall in blood pressure after an adequate dose of drug 1, switching to (rather than adding) drug 2 and, if this is ineffective, switching to drug 3 may allow as many as 60 to 80 percent of patients with stage 1 hypertension to initially be controlled with a single agent [54,58].

There are no strict guidelines as to how to perform sequential monotherapy, although the following approach is reasonable:

●The blood pressure response to initial monotherapy should be assessed in four to six weeks.

●If there is an inadequate response, the dose can be increased and the blood pressure response assessed after another four-to-six-week period. As the dose is increased with most antihypertensive drugs, the antihypertensive response attenuates and side effects become more prominent with the relative exception of ACE inhibitors and ARBs in patients with normal kidney function (figure 8) [59]. As a result, we generally limit dose titration to one step with a given drug (eg, 12.5 to 25 mg of chlorthalidone and 5 to 10 mg of amlodipine). Using higher doses may produce a lesser blood pressure response and more toxicity than switching to an initial dose of a second drug. (See 'General principles' above.)

●Thus, if after a single dose increase fails to produce an adequate blood pressure response, we switch to a different agent.

This process of trying to find the one drug to which the patient is most responsive may minimize side effects, maximize patient compliance, and is as effective as some forms of combination therapy. However, over time, more than one drug will be needed in many patients who are initially controlled. In ALLHAT, for example, the proportion of patients treated with more than one drug increased from 26 to 33 percent at one year to 40 to 43 percent at five years [21]. (See 'Addition of a second drug' below.)

Drug dosing and drug frequency — Although stepped care therapy has emphasized pushing initial therapy, as necessary, to the maximum recommended dose, the steepest part of the dose-response curve is typically seen at lower doses: good responders generally respond to low doses with few side effects, while higher doses produce more side effects often with little further reduction in blood pressure (figure 8) [2,60]. (See 'General principles' above.)

As examples, patients who are treatment "responders" may experience a similar blood pressure response with 12.5 or 25 mg/day of hydrochlorothiazide (or its equivalent) as they do with 50 mg of the drug (figure 5); similar results might be seen with 50 mg of atenolol compared with 100 mg, 10 compared with 20 mg of enalapril, and 50 compared with 100 mg of losartan [28,61-65]. In addition to its efficacy, low-dose diuretics are less likely to produce the metabolic abnormalities that are often seen at higher doses (figure 6) [28]. (See "Use of thiazide diuretics in patients with primary (essential) hypertension".)

The issue of dose frequency relates to the possible absence of 24-hour efficacy with shorter-acting drugs [66-68]. Once daily dosing with such drugs gives a greater peak response, but the blood pressure tends to return toward baseline at night or during the early morning hours, well before the next dose. This is a potential concern since a greater daily blood pressure load, nocturnal hypertension, and early morning elevations in blood pressure can increase cardiovascular risk. Thus, drugs that are longer acting are preferred [69]. Giving one-half the dose twice a day is an alternative that produces a lesser peak effect but a more sustained response; however, patient compliance may be reduced [68,70]. (See "Out-of-office blood pressure measurement: Ambulatory and self-measured blood pressure monitoring".)

Clinical reasons for specific drugs — The general recommendations for initial therapy should be amended in patients with specific underlying conditions in whom specific agents might offer particular benefit independent of blood pressure control (table 2) [8,10,51,71]. These indications include the demonstration that ACE inhibitors improve outcomes in a number of high-risk settings and that beta blockers improve survival in patients with heart failure with reduced ejection fraction (HFrEF) or a prior myocardial infarction [10]. Further information on specific disorders is available in separate topic reviews.

ACE inhibitors — Angiotensin-converting enzyme (ACE) inhibitors are first-line therapy in all patients who have HF or asymptomatic LV systolic dysfunction, in all patients who have had an ST elevation MI, in patients with a non-ST elevation MI who have had an anterior infarction, diabetes, or systolic dysfunction, and in patients with proteinuric chronic kidney disease. (See "Angiotensin converting enzyme inhibitors and receptor blockers in acute myocardial infarction: Recommendations for use" and "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults" and "Initial pharmacologic therapy of heart failure with reduced ejection fraction in adults", section on 'ACE inhibitor'.)

It has been suggested that ACE inhibitors and ARBs have a cardioprotective effect independent of blood pressure lowering in patients at high risk for a cardiovascular event. However, as mentioned above and described in detail elsewhere, the available evidence suggests that the attained blood pressure, not the drug used, is of primary importance in such patients. (See 'Importance of attained blood pressure' above.)

Angiotensin II receptor blockers — The specific indications for and efficacy of angiotensin II receptor blockers (ARBs) are similar to those with ACE inhibitors. (See "Angiotensin converting enzyme inhibitors and receptor blockers in acute myocardial infarction: Recommendations for use" and "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults".)

There is at least one setting in which ARBs have specific benefit and which a similar trial has not been performed with ACE inhibitors: severe hypertension with ECG evidence of left ventricular hypertrophy in the LIFE trial [61]. An ARB can be used instead of an ACE inhibitor in such patients, although it is highly likely that an ACE inhibitor is equally effective. We would not switch such a patient who is already receiving and tolerating an ACE inhibitor to an ARB.

An ARB is particularly indicated in patients who do not tolerate ACE inhibitors (primarily because of cough). (See "Differences between angiotensin-converting enzyme inhibitors and receptor blockers".)

Thiazide diuretics — The preferred thiazide diuretic in patients with primary hypertension is chlorthalidone since major trials such as ALLHAT have shown benefit with this regimen. Indapamide, another thiazide-like diuretic, may be used in place of chlorthalidone. There is little, if any, evidence that hydrochlorothiazide improves cardiovascular outcomes. Hydrochlorothiazide is both less potent and shorter acting than chlorthalidone and indapamide. (See 'Thiazide-like versus thiazide-type diuretics' above and 'Initial monotherapy' above.)

One problem with low-dose chlorthalidone is that there is no 12.5 mg tablet. Thus, 25 mg tablets of generic chlorthalidone need to be cut in half; however, these tablets are not scored, and attempts to halve them may result in uneven dosing. In addition, in patients who require combination therapy, fixed-dose combination pills of chlorthalidone with ACE inhibitors and long-acting calcium channel blockers are not available (in contrast to hydrochlorothiazide). However, chlorthalidone (at a dose of either 12.5 or 25 mg) is available in combination with azilsartan medoxomil, an ARB. Indapamide, an alternative to chlorthalidone, has both a low-dose option available (1.25 mg) and a fixed-dose combination with an ACE inhibitor (perindopril).

If monotherapy is appropriate in a patient with hypertension and osteoporosis, thiazide-like diuretics may have advantages over ACE inhibitors, ARBs, and calcium channel blockers. These drugs stimulate distal tubular reabsorption of calcium, leading to a decrease in urinary calcium excretion. As a result, thiazide diuretics may have a beneficial effect on bone mineral density. (See "Drugs that affect bone metabolism", section on 'Thiazide diuretics'.)

The rates of hip or pelvic fractures among patients treated with thiazide-like diuretics, ACE inhibitors, and calcium channel blockers were compared in a post-hoc analysis of the ALLHAT trial (discussed above) [72]. At approximately five years, those randomly assigned chlorthalidone had significantly fewer hip or pelvic fractures as compared with those assigned either lisinopril or amlodipine (1.3 versus 1.7 percent). (See 'ALLHAT trial' above.)

Diuretics should also be given for volume control in patients with heart failure or chronic kidney disease, with or without nephrotic syndrome; these settings usually require loop diuretics. In addition, a mineralocorticoid receptor antagonist (spironolactone or eplerenone) is indicated in patients with HF who have relatively preserved kidney function and for the prevention or treatment of hypokalemia. (See "Secondary pharmacologic therapy in heart failure with reduced ejection fraction (HFrEF) in adults", section on 'Mineralocorticoid receptor antagonist'.)

Calcium channel blockers — There are no absolute indications for calcium channel blockers in patients with hypertension. Long-acting dihydropyridines are most commonly used. Like beta blockers, the nondihydropyridine calcium channel blockers (verapamil, diltiazem) can be given for rate control in patients with atrial fibrillation or for control of angina in patients with coronary disease and normal left ventricular systolic function [73]. Calcium channel blockers also may be preferred in patients with obstructive airways disease. (See "Treatment of hypertension in asthma and COPD".)

Beta blockers — A beta blocker without intrinsic sympathomimetic activity should be given after an acute myocardial infarction and to stable patients with heart failure or asymptomatic left ventricular dysfunction (beginning with very low doses to minimize the risk and degree of initial worsening of myocardial function). The use of beta blockers in these settings is in addition to the recommendations for ACE inhibitors in these disorders. (See "Acute myocardial infarction: Role of beta blocker therapy" and "Initial pharmacologic therapy of heart failure with reduced ejection fraction in adults", section on 'Beta blocker'.)

Beta blockers are also given for rate control in patients with atrial fibrillation, for control of angina, and for symptom control in a number of other disorders (table 2).

In the absence of such indications, we and others recommend that beta blockers not be used as first-line therapy, particularly in patients over age 60 years [6,50,53,74-76]. Compared with other antihypertensive drugs in the primary treatment of hypertension, beta blockers (not all trials used atenolol) may be associated with inferior protection against stroke risk and all-cause mortality [76-80]. These effects are primarily seen in patients over age 60 years [79,81-83]. Beta blockers are also associated with impaired glucose tolerance and an increased risk of new onset diabetes [53], with the exception of vasodilating beta blockers such as carvedilol and nebivolol [84,85]. (See "Treatment of hypertension in patients with diabetes mellitus".)

Alpha blockers — The ALLHAT trial cited above included a doxazosin arm that was terminated prematurely because of a significantly increased risk of heart failure compared to chlorthalidone (relative risk 2 after adjusting for a 3 mmHg higher in-trial systolic pressure with doxazosin) noted during an interim analysis [86] and a higher rate of cardiovascular events [87]. Thus, an alpha blocker is not recommended for initial monotherapy, with the possible exception of older men with symptoms of prostatism and if they are not at high cardiovascular risk. (See "Medical treatment of benign prostatic hyperplasia".)

COMBINATION THERAPY — Two major issues related to combination therapy include the use of combination therapy as first-line therapy and addition of a second drug when the goal blood pressure is not achieved with monotherapy. The following discussion assumes that the patient does not have a clinical reason to use a particular drug. (See 'Clinical reasons for specific drugs' above.) (Related Pathway(s): Hypertension: Initial management.)

Recommendations for combination therapy were made by the 2018 European Society of Hypertension/European Society of Cardiology (ESH/ESC) and by the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Guideline for High Blood Pressure in Adults; each clearly supports initial combination therapy for those 20/10 mmHg above the goal [6,8]. By contrast, the 2020 International Society of Hypertension (ISH) guidelines suggest combination therapy for most patients with blood pressure >140/90 mmHg [88].

First-line combination therapy — Administering two drugs as initial therapy should be considered when the blood pressure is more than 20/10 mmHg above goal, as recommended by the ESH/ESC and ACC/AHA panels (algorithm 1) [6,8]. This strategy may increase the likelihood that target blood pressures are achieved in a reasonable time period [89]. Fixed-dose combination preparations are available that may improve patient compliance, blood pressure control, and, if both drugs are given at lower doses, reduce side effects [2,59,90-95].

Supine and standing pressures should be measured prior to the initiation of combination therapy in patients at increased risk for orthostatic (postural) hypotension, such as older adult patients and those with diabetes. Orthostatic hypotension is diagnosed when, within two minutes of quiet standing, one or more of the following is present (see "Mechanisms, causes, and evaluation of orthostatic hypotension"):

●At least a 20 mmHg fall in systolic pressure

●At least a 10 mmHg fall in diastolic pressure

●Symptoms of cerebral hypoperfusion, such as dizziness

Based upon the results of the ACCOMPLISH trial [45], we recommend the use of a long-acting dihydropyridine calcium channel blocker plus a long-acting angiotensin-converting enzyme (ACE) inhibitor/ARB (such as amlodipine plus benazepril as used in ACCOMPLISH) (algorithm 1). In addition, in nonobese patients already being treated with and doing well on the combination of a thiazide diuretic and a long-acting angiotensin inhibitor, we suggest replacing the thiazide diuretic with a long-acting dihydropyridine calcium channel blocker. In obese patients, the combination of a thiazide diuretic and a long-acting angiotensin inhibitor can be continued [47]. (See 'ACCOMPLISH trial' above.)

However, a different combination of antihypertensive drugs may be preferred (ie, other than the combination of a long-acting dihydropyridine calcium channel blocker plus a long-acting ACE inhibitor/ARB) if the patient has a specific clinical indication. As examples, appropriate initial combination therapy may include a beta blocker plus a long-acting ACE inhibitor/ARB in patients who have had a recent myocardial infarction or a loop diuretic plus an ACE inhibitor/ARB or mineralocorticoid receptor antagonist in edematous patients with heart failure. (See 'Beta blockers' above and "Acute myocardial infarction: Role of beta blocker therapy" and "Angiotensin converting enzyme inhibitors and receptor blockers in acute myocardial infarction: Recommendations for use" and "Treatment and prognosis of heart failure with preserved ejection fraction" and "Overview of the management of heart failure with reduced ejection fraction in adults".)

Combination therapy with more than two agents — Some studies have compared triple combination therapy with dual combination therapy or usual care in hypertensive adults [96-99]. Most have compared standard doses of an ACE inhibitor (or ARB), amlodipine, plus a thiazide diuretic with dual combination therapy selected from these three options. In general, patients treated with triple combination therapy attained lower blood pressures and were more likely to achieve their blood pressure goal at 12 to 24 weeks; by contrast, those treated with triple therapy were more likely to have treatment-related adverse events. One small trial examined the effects of single-pill quadruple (rather than triple) therapy, with each agent given at one quarter the normal starting dose in an attempt to limit side effects (ie, irbesartan at 37.5 mg, amlodipine at 1.25 mg, hydrochlorothiazide at 6.25 mg, and atenolol at 12.5 mg) [100]. Compared with placebo, quadruple therapy reduced 24-hour systolic pressure by 19 mmHg and increased the proportion attaining goal blood pressure (100 versus 33 percent). The small sample size and lack of an active comparator limits the relevance of this study [101].

Addition of a second drug — As noted above, each of the recommended first-line agents will normalize the blood pressure in up to 30 to 50 percent of patients with mild hypertension [5]. In the patient who is relatively unresponsive to one drug, sequentially trying different agents may allow 60 to 80 percent of patients with mild hypertension to be initially controlled with a single agent [54,58]. These issues are discussed in detail above. (See 'Initial monotherapy' above.)

We generally limit dose titration to one step with a given drug (eg, 12.5 to 25 mg of chlorthalidone or 5 to 10 mg of amlodipine). Using higher doses generally produces a lesser blood pressure response and more toxicity than switching to an initial dose of a second drug (figure 8) [5,59,102].

Over time, more than one drug will be needed in many patients who are initially controlled. In ALLHAT, for example, the proportion of patients treated with more than one drug increased from 26 to 33 percent at one year to 40 to 43 percent at five years [21].

As noted above, we suggest that combination therapy consist of a long-acting dihydropyridine calcium channel blocker plus a long-acting ACE inhibitor/ARB (such as amlodipine plus benazepril). Thus, if the patient is being treated with one of the drugs, add the other. In patients being treated with a thiazide diuretic, we suggest discontinuing the thiazide and starting combination therapy. Approximately 75 percent of patients in ACCOMPLISH had previously been treated with two or more antihypertensive drugs. (See 'ACCOMPLISH trial' above.)

Beta blockers are now used rarely as initial therapy except for patients with another indication for their use. The preferred second drugs in patients who are treated with a beta blocker are a thiazide diuretic or a dihydropyridine calcium channel blocker [71]. An alpha blocker would be chosen only if there is another reason for its use, such as symptomatic benign prostatic hyperplasia.

An ACE inhibitor or ARB is likely to be less effective in patients treated with a beta blocker since beta blockers reduce renin secretion and therefore angiotensin II formation [103], and a beta blocker should be used with caution in combination with verapamil and to a lesser degree diltiazem. These drugs can potentiate the cardiac depressant effect of the beta blocker, possibly leading to or exacerbating bradycardia or heart block, particularly if one of the drugs is given intravenously.

BEDTIME VERSUS MORNING DOSING — The average nocturnal blood pressure is approximately 15 percent lower than daytime values. Failure of the blood pressure to fall by at least 10 percent during sleep is called "nondipping," and is a stronger predictor of adverse cardiovascular outcomes than daytime blood pressure. (See "Out-of-office blood pressure measurement: Ambulatory and self-measured blood pressure monitoring".)

In some studies, shifting at least one antihypertensive medication from the morning to the evening both restored the normal nocturnal blood pressure dip [104-108] and reduced 24-hour mean blood pressure [109,110]. However, several blinded and unblinded trials found no difference in 24-hour mean or nocturnal blood pressure with evening versus morning dosing of antihypertensive therapy and no effect on the proportion of dipping [107,111-114].

A research group from Spain has published several trials that reported very large benefits from shifting one or more antihypertensive drugs from the morning to bedtime (eg, 50 percent or greater relative reductions in stroke, myocardial infarction, and cardiovascular death) [108,115]. Effects of this magnitude are rarely if ever observed in rigorous cardiovascular trials; in addition, the biologic rationale (a modest reduction in nighttime blood pressure without a major difference in 24-hour blood pressure) does not support such large effects, and the findings have not been independently confirmed [116].

Other trials also employed nocturnal therapy, although they were not specifically designed to compare bedtime versus morning dosing of antihypertensive therapy:

●In the HOPE and EUROPA trials, angiotensin-converting enzyme (ACE) inhibitor therapy given at bedtime reduced the incidence of cardiovascular events as compared with placebo.

●In the CONVINCE trial, sustained release verapamil (given at bedtime) did not reduce cardiovascular events compared with either hydrochlorothiazide or atenolol (given in the morning) [117].

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".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Medicines for high blood pressure (The Basics)")

●Beyond the Basics topics (see "Patient education: High blood pressure in adults (Beyond the Basics)" and "Patient education: High blood pressure treatment in adults (Beyond the Basics)" and "Patient education: High blood pressure, diet, and weight (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS — The American Heart Association and the European Society of Hypertension/European Society of Cardiology (ESH/ESC), as well as various meta-analyses, all concluded that the amount of blood pressure reduction is the major determinant of reduction in cardiovascular risk in both younger and older patients with hypertension, not the choice of antihypertensive drug. This conclusion also applies to patients at increased cardiovascular risk as shown in the ALLHAT, VALUE, and CAMELOT trials. (See 'Importance of attained blood pressure' above and 'ALLHAT trial' above.)

However, there may be exceptions when using combination therapy. In the ACCOMPLISH trial, amlodipine plus benazepril was associated with a 20 percent lower rate of cardiovascular events compared to hydrochlorothiazide plus benazepril, despite slightly higher 24-hour blood pressures in the amlodipine arm. (See 'ACCOMPLISH trial' above.)

Some hypertensive patients have underlying conditions for which specific antihypertensive drugs might offer particular benefit independent of blood pressure control, such as diltiazem, verapamil, or a beta blocker for rate control in atrial fibrillation. The following recommendations do not apply to such patients. (See 'Clinical reasons for specific drugs' above.)

Monotherapy — Patients with hypertension who are less than 20/10 mmHg above goal can initially be treated with monotherapy. Among such patients who do not have an indication for a specific drug, the major classes of drugs that have been used for monotherapy are a low-dose thiazide diuretic, long-acting angiotensin-converting enzyme (ACE) inhibitor/angiotensin II receptor blocker (ARB), or a long-acting dihydropyridine calcium channel blocker (algorithm 1).

●Given the preference for an ACE inhibitor/ARB plus a dihydropyridine calcium channel blocker in patients requiring combination therapy, we suggest use of one of these drug classes as initial therapy so that the other can be added, if necessary (Grade 2C). If this approach is chosen, an ACE inhibitor/ARB may be more effective in younger patients, and a dihydropyridine calcium channel blocker may be more effective in older adult and Black patients. (See 'Initial monotherapy' above and 'Monotherapy based upon age and race' above.)

●If a diuretic is chosen, we suggest chlorthalidone or indapamide rather than hydrochlorothiazide (Grade 2B). The basic principles of monitoring for hypokalemia with these drugs are identical to those with hydrochlorothiazide. (See 'Choice between thiazide-like and thiazide-type diuretics' above and 'Issues with thiazide-like diuretics' above and 'Monitoring for hypokalemia' above.)

●We recommend that patients who have a minimal or no response to the initial antihypertensive drug be treated with sequential monotherapy (Grade 1B). (See 'Sequential monotherapy' above.)

Combination therapy

●Among patients who have an untreated office (or clinic) blood pressure more than 20/10 mmHg above goal, we recommend therapy with the combination of a long-acting ACE inhibitor/ARB plus a long-acting dihydropyridine calcium channel blocker (algorithm 1) (Grade 1B). (See 'ACCOMPLISH trial' above and 'First-line combination therapy' above.)

●Among nonobese patients who are already being treated with an ACE inhibitor/ARB plus a thiazide diuretic and have attained goal blood pressure, we suggest stopping the thiazide and switching to a long-acting dihydropyridine calcium channel blocker (Grade 2B). Among obese patients, the combination of an ACE inhibitor/ARB plus a thiazide diuretic can be continued. We suggest continuing therapy in patients who are well controlled on combinations other than an ACE inhibitor/ARB plus a thiazide (Grade 2C).

●Among patients being treated with a thiazide diuretic as monotherapy who have responded but have not attained goal blood pressure, we suggest stopping the thiazide and switching to a long-acting ACE inhibitor/ARB plus a long-acting dihydropyridine calcium channel blocker (Grade 2B).

ACKNOWLEDGMENT — The editorial staff at UpToDate would like to acknowledge Norman M Kaplan, MD, who contributed to an earlier version of this topic review.