Lifestyle modifications include sodium restriction to less than 1.5–2 g/day. The major source of sodium consumed in the United States comes from processed food including bread and rolls, pasta, cured meat, and snacks. Table salt contributes to only 20% of total sodium consumed. Avoiding table salt alone will not be an effective strategy in limiting sodium intake, and educating patients to read nutritional label is a key to success. A diet rich in fruits and vegetables also reduces BP independent of sodium content.
Antihypertensive treatment should be initiated in patients with stage 1 uncomplicated hypertension who fail a trial of lifestyle modification alone. Pharmacologic intervention should be offered promptly without any delay for patients with stage 2 hypertension or those with stage 1 hypertension in the presence of target organ complication, diabetes mellitus, renal failure, or high cardiovascular risks in conjunction with lifestyle modification. The list of antihypertensive drugs available in the United States is shown in Table 2–1.
Table 2–1. Antihypertensive Drugs Available in the United States
Thiazide diuretics are considered to be the first-line drug therapy for hypertension according to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines due to their efficacy in reducing BP and low cost. Thiazide diuretics promote natriuresis by inhibiting sodium chloride cotransport at the distal convoluted tubule. Among thiazide diuretics, chlorthalidone has the longest half-life of between 40 and 60 hours compared to 8–15 hours for hydrochlorothiazide. Head-to-head comparison studies showed that chlorthalidone is at least two times more potent than hydrochlorothiazide in reducing BP. Thiazide diuretics are known to cause a number of metabolic side effects such as hypokalemia, elevated uric acid, increased fasting triglyceride, and, most importantly, increased risk of diabetes mellitus. Thiazide-induced dysglycemia or insulin resistance is mediated both by hypokalemia and potassium-independent mechanisms, which can be minimized by concomitant administration of the mineralocorticoid receptor antagonist spironolactone and, to a lesser extent, by ACEIs or ARBs.
Loop diuretics promote diuresis by inhibiting Na+/K+/Cl– cotransport at the thick ascending limb of Henle's loop. Although loop diuretics are more potent as diuretic agents than thiazide diuretics, the short half-life limits their efficacy in reducing BP. Furosemide has a half-life of only 1.5–2.5 hours and should be given every 6 hours to have sustained effects on BP; hence the name “Lasix.” Thus, it is not recommended for treatment of hypertensive patients with normal renal function. However, it is useful for patients with fluid retention from congestive heart failure, renal failure, or nephrotic syndrome. Among loop diuretics, furosemide has the least predictable absorption, whereas torsemide has the longest half-life.
Mineralocorticoid receptor (MR) antagonists, such as spironolactone and eplerenone, are known to reduce BP by inhibiting MR-dependent activation of the epithelial sodium channel in the cortical collecting duct. Furthermore, inhibition of the sympathetic nervous system and improvement in vascular endothelial function are thought to contribute to the BP-lowering effects of this class of drugs. MR antagonists are effective in lowering BP when either used alone or as add-on therapy for resistant hypertension despite use of three or more drugs. On a milligram-to-milligram basis, eplerenone is less potent than spironolactone in reducing BP in both patients with primary hypertension and PA, and thus, at least a twofold higher dose is often required to achieve BP control. However, spironolactone carries antiandrogenic side effects, including gynecomastia and testicular atrophy, and long-term use at the dose of more than 25 mg/day should be avoided in men. In these patients, the use of eplerenone, which has more than 100-fold lower affinity to androgen receptors than MRs, is preferable. Hyperkalemia is the main side effect of both agents, and serum potassium should be closely monitored in patients with CKD, diabetic patients with hyporeninemic hypoaldosteronism, or those on concomitant treatment with ACEIs, ARBs, or nonsteroidal anti-inflammatory agents. MR antagonists have been shown to reduce mortality in patients with systolic heart failure even with mild symptoms, and they should be considered in treatment of hypertension in patients with impaired ventricular function.
In addition to MR antagonists, direct inhibitors of the epithelial sodium channels (ENaCs), such as amiloride and triamterene, are also effective in lowering BP in patients with low renin essential hypertension. They are particularly useful in patients with gain-in-function mutation in the ENaC, causing excessive renal sodium absorption and renal K wasting, known as Liddle syndrome. Amiloride is also useful in patients with primary aldosteronism who cannot tolerate MR antagonists. Both amiloride and triamterene are often used in a combination pill formulation with hydrochlorothiazide (see Table 2–1).
ACEIs reduce BP by inhibiting conversion of angiotensin (Ang) I to potent vasoconstrictor Ang II. ACEIs also prevent degradation of the vasodilator bradykinin, which further contributes to BP reduction. ACEIs are particularly effective in reducing BP in younger patients and Caucasians and tend to be less effective in the setting of the low renin form of hypertension frequently observed in the elderly and black populations. However, the combination of ACEIs with diuretics eliminates racial and age differences in BP response to ACEIs. In addition to BP reduction, ACEIs reduce mortality in patients with congestive heart failure and improve cardiovascular outcomes in high-risk hypertensive patients without heart failure. ACEIs also slow the decline in renal function in patients with diabetic and nondiabetic kidney disease. Thus, they should be used in hypertensive patients with these comorbid conditions. The main side effects of this class of drugs are cough and angioedema from presumably excessive bradykinin accumulation. Hyperkalemia is another side effect, which may limit its use in some patients with severe CKD or bilateral renal artery stenosis.
ARBs reduce BP by blocking the vasoconstrictor effects of Ang II on angiotensin receptor subtype 1 (AT1R) in the vascular smooth muscle. This class of drugs has remarkably minimal side effects and is generally well tolerated. It avoids side effects of cough and angioedema associated with ACEIs. In placebo-controlled clinical trials, ARBs reduced mortality in patients with systolic heart failure who were intolerant to ACEIs and offered renal protection in patients with diabetic nephropathy. Thus, they should be considered as part of regimen in hypertensive patients with these clinical features. The combination of ARBs with maximal doses of ACEIs usually yields minimal additional BP reduction and is generally not helpful in improving cardiovascular mortality. Although combination therapy offers synergistic or additive effect in reducing proteinuria, dual renin-angiotensin system blockade is associated with more renal dysfunction than either class of drugs alone. Therefore, it is not recommended as a strategy to control BP in hypertensive patients and should be reserved only for patients with proteinuria from primary renal disease such as immunoglobulin A nephropathy.
Calcium Channel Blockers (CCBs)
CCBs are divided into two major classes: the dihydropyridine (DHP) and nondihydropyridine (non-DHP) CCBs. DHP CCBs include amlodipine, felodipine, nisoldipine, and nicardipine and cause peripheral vasodilation with minimal effects on the heart rate or myocardial contractility. In contrast, non-DHP CCBs (diltiazem and verapamil) exert negative inotropic and chronotropic effects. Non-DHP CCBs are contraindicated in patients with systolic heart failure because of increased cardiovascular mortality and should be avoided in patients with symptomatic bradyarrhythmias or advanced heart block. DHP CCBs have neutral effect on survival in congestive heart failure and may be considered if BP remains elevated despite ACEIs or ARBs in systolic heart failure. All CCBs increase myocardial blood flow and are antianginal. Meta-analysis of clinical trials showed that CCBs are more effective than ACEIs in preventing stroke beyond BP reduction, but less effective than ACEIs and ARBs in preventing coronary heart disease or heart failure. DHP CCBs preferentially dilate afferent arterioles, causing increase in intraglomerular hypertension and are not as effective as ACEIs or ARBs in preventing decline in renal function in diabetic or nondiabetic kidney diseases. Therefore, they should not be used without RAAS blockade in hypertensive patients with CKD.
β-Adrenergic Receptor Blockers (BBs)
BBs reduce BP by decreasing heart rate and myocardial contractility, leading to decreased cardiac output at least initially via inhibition of β1-adrenergic receptor (AR). Inhibition of renin release from the juxtaglomerular cells via β1-AR and reduction of norepinephrine release from sympathetic nerve terminals via inhibition of presynaptic β2-AR also contribute to BP reduction. BBs are divided into two major classes: selective (β1-specific) and nonselective (β1- and β2-AR blockers), as shown in Table 2–2. Activation of β2-AR leads to peripheral vasodilation and bronchodilation. Thus, nonselective BBs should not be used in patients with asthma or in the setting of sympathetic overactivity, such as clonidine withdrawal or cocaine intoxication because of unopposed α-AR–mediated vasoconstriction. However, selectivity to β1 versus β2 is dose-dependent and not an all-or-none phenomenon. At higher doses, even selective BBs such as metoprolol, bisoprolol, atenolol, or esmolol may cause bronchospasm and should be used with caution. Certain BBs exert direct vasodilating effects, which enhance antihypertensive efficacy independent of β-AR blockade. For example, labetalol and carvedilol exert α-AR–blocking properties, whereas nebivolol promotes nitric oxide release and possesses antioxidant properties. Although BBs reduce mortality in patients with systolic heart failure and myocardial infarction, older generation BBs, particularly atenolol, are not as effective as other classes of drugs in preventing stroke or all-cause mortality in hypertensive patients with cardiovascular diseases. Thus, atenolol should not be used as initial therapy in patients with uncomplicated essential hypertension, particularly in patients with low renin hypertension, elderly patients, or African Americans.
Table 2–2. Selective versus Nonselective β-Adrenergic Receptor Blockers
α-AR blockers exert antihypertensive effect by inhibiting postsynaptic α1-ARs in the vascular smooth muscle. α-AR blockers that selectively inhibit α1-AR antagonists include prazosin, terazosin, and doxazosin, whereas the nonselective blockers, which inhibit both α1- and α2-ARs, include phentolamine and phenoxybenzamine. Side effects of α-blockers are orthostatic hypotension, nasal congestion, and volume expansion. Nonselective α-blockers tend to cause more tachycardia because of inhibition of presynaptic α2-ARs, which normally exert inhibitory influence on norepinephrine release from sympathetic nerve terminals. α-Blockers should not be used as first-line therapy for hypertension because of increased risk of congestive heart failure. However, they are useful as an adjunctive treatment as the fourth- or fifth-line agent when BP fails to reach target goal. They should be considered as the first line of treatment in the setting of pheochromocytoma and are also useful in patients with hypertension related to cocaine intoxication or conditions associated with sympathetic overactivity such as drug or alcohol withdrawal.
Central Sympatholytic Drugs
Central sympatholytic drugs reduce BP mainly by stimulating presynaptic α2-ARs in the brainstem centers and sympathetic nerve terminals, thereby inhibiting sympathetic nerve discharge and neuronal release of norepinephrine to the heart and peripheral circulation. Drugs in this class include methyldopa, clonidine, guanfacine, and guanabenz. Moxonidine and rilmenidine are also centrally acting drugs used in England and other European countries but are not available in the United States. α-Methyldopa is converted to α-methylnorepinephrine to activate central α2-ARs in the brainstem, whereas clonidine, guanfacine, and guanabenz are direct α2-AR agonists. In contrast, moxonidine and rilmenidine predominantly reduce sympathetic nerve activity and BP by stimulating the imidazoline-1 (I1) receptor, rather than α2-AR. Because of several major side effects, including fatigue, sedation, and dry mouth, as well as lack of clear-cut cardiovascular benefit, these drugs should be used as fourth-line drug therapy for hypertension. Central sympatholytic drug use should be avoided in patients who are nonadherent to treatment because of precipitation of withdrawal symptoms upon abrupt drug discontinuation. Their use on an as-needed basis for treatment of episodes of asymptomatic BP surge in patients who are not on regular dosing of central α2-AR agonists should also be discouraged due to rebound hypertension.
Direct Renin Inhibitors (DRIs)
DRIs block catalytic action of circulating renin, thereby preventing formation of Ang I from angiotensinogen. DRIs also inhibit activity of the proenzyme prorenin, which can also cleave angiotensinogen into Ang I upon binding to the prorenin receptors. Inhibitory action of DRIs on renin and prorenin activity results in decreased production of Ang II and BP. Aliskiren is the only drug approved for clinical use in this class so far. Addition of aliskiren to ARBs reduces proteinuria in hypertensive patients with type 2 diabetes mellitus when compared to placebo. However, addition of aliskiren to ARBs or ACEIs is not more effective in reducing left ventricular (LV) mass in hypertensive patients or improving LV remodeling in patients after myocardial infarction than monotherapy with ARBs or ACEIs alone. Furthermore, a recent clinical trial showed that addition of aliskiren to ACEIs or ARBs in patients with diabetes mellitus and established cardiovascular disease failed to improve cardiovascular outcomes and even increased the risk of nonfatal stroke. Thus, the future role of DRIs in the treatment of hypertension is likely to be limited.
Hydralazine and minoxidil are the two main drugs in this class, which reduce BP by dilating resistant arterioles with minimal or no effect on the venous circulation. Hydralazine has short plasma half-life of 90 minutes, but its antihypertensive action lasts much longer than its plasma level. Thus, for the ease of administration, hydralazine should be given twice daily in hypertensive patients rather than three times a day, which is the schedule typically used in patients with heart failure. Side effects of hydralazine include lupus-like syndrome, particularly at the higher doses, reflex tachycardia, nausea, vomiting, and diarrhea. Minoxidil promotes vasodilation by activating adenosine triphosphate–sensitive potassium channels in the vascular smooth muscle. Similar to hydralazine, minoxidil-induced BP reduction is usually accompanied by reflex tachycardia and palpitation, and the addition of BBs is often needed to mitigate these side effects. Peripheral edema and volume expansion frequently occur, requiring concomitant diuretic treatment. Less common side effects include hirsutism and pericardial effusion.
Oral nitrates are nitric oxide donors, which produce venodilating effects at low doses and arterial vasodilating effects at higher doses. Although the benefits of nitrates as antianginal drugs are well validated, very few studies have addressed their efficacy as antihypertensive agents. In one small study, isosorbide mononitrate reduced systolic BP in elderly patients with resistant systolic hypertension without affecting diastolic BP. It is unknown whether these data can be extrapolated to other age groups with combined systolic-diastolic hypertension. Headache is the main side effect, and nitrate tolerance is the main limiting factor for long-term use.
Management of Uncomplicated Hypertension
In most patients, the goal BP should be < 140/90 mm Hg. Tight control of systolic BP below 120 mm Hg does not reduce cardiovascular death, but reduces the risk of stroke by 40% in patients with diabetes mellitus. Higher systolic BPs (< 150 mm Hg) may be more appropriate for patients > 80 years of age and patients with class 4 or greater CKD. In patients with mild uncomplicated hypertension, thiazide-type diuretics should be considered in patients who have a tendency to be salt sensitive, such as older patients and black patients. For younger patients, ACEIs, ARBs, and CCBs may be more suitable. The use of BBs, particularly older generation BBs such as atenolol, should be avoided in young patients with active lifestyle due to fatiguing side effects and modest efficacy in improving cardiovascular outcomes. BBs are more suitable for patients with compelling indication such as coronary artery disease or congestive heart failure. Treatment of hypertension is also beneficial in elderly patients even > 80 years old in terms of reducing stroke and cardiovascular mortality. In these patients, chlorthalidone or DHP CCB such as amlodipine should be part of the regimen given the proven benefit in reducing stroke and adverse cardiovascular events, particularly in patients with isolated systolic hypertension.
For patients with BP at least 20/10 mm Hg above the goal, the use of combination therapy is preferred over monotherapy to avoid side effects related to the use of a single agent at the high dose that is often needed to control BP. In patients with high cardiovascular risks, the combination of an ACEI and DHP CCB is superior for reducing adverse cardiovascular outcomes than the combination of an ACEI with thiazide-type diuretics. Thus, the addition of DHP CCBs should be considered in such patients whose BP reduction is not adequate with ACEIs alone.
Management of Hypertensive Crisis
Hypertensive crisis is a common condition encountered in the emergency department that can be classified as hypertensive emergency or hypertensive urgency. Hypertensive emergency is defined severe elevation in BP (often > 180/120 mm Hg) in the presence of acute target organ damage involving heart, brain, kidneys, or vascular system. Hypertensive urgency is defined as severe hypertension associated with symptoms, such as headache or chest pain, in the absence of any objective evidence of acute target organ damage. Thus, distinction between hypertensive emergency versus urgency is dependent of clinical presentation rather than severity of hypertension alone.
Once patients are identified to have hypertensive crisis, choice of therapy and rapidity of BP reduction are dependent on clinical presentation. In normotensive subjects, cerebral blood flow is maintained at a constant level despite variation in mean arterial BP between 60 and 120 mm Hg by a process known as cerebral autoregulation. At the lower end of this range, cerebral vessels dilate to maintain normal perfusion. Only when the mean BP is lower than 50–60 mm Hg, cerebral perfusion becomes pressure-dependent. Similarly, when BP rises to a higher level, myogenic vasoconstriction of the resistant arterioles prevents excessive increase in blood flow and cerebral edema. Patients with well-controlled hypertension have autoregulatory adjustment at the normal BP range similar to normotensives. However, in patients with severe hypertension, this BP range is shifted to the right between 110 and 180 mm Hg (Figure 2–1). Thus, BP lowering in patients with hypertensive encephalopathy should done cautiously by aiming for reduction by 20–25% in the first few hours but not below 140/90 mm Hg (mean BP around 107 mm Hg) in the first 24 hours. Patients with hypertensive urgency could be treated with oral antihypertensive medications in the emergency department and discharged home with close follow-up for titration of BP medications within a few days. Rapid normalization of BP with oral agents in the doctor's office or emergency department in patients with asymptomatic severe hypertension without any target organ damage may provoke hypoperfusion of the vital organs, such as brain and kidneys, and should be avoided.
Relationship between cerebral blood flow (CBF) and mean arterial pressure (MAP) in normotensive subjects (solid line). Normally, CBF is maintained constant at MAP between 60 and 120 mm Hg. However, in patients with uncontrolled hypertension, this relationship is shifted to the right. Therefore, precipitous reduction in BP in these individuals to normal range (MAP < 11 mm Hg in the first 24 hours) may lead to cerebral hypoperfusion.
In contrast to hypertensive encephalopathy, BP should be reduced to < 120/80 mm Hg as soon as possible in patients with hypertensive emergency in the setting of aortic dissection because delay in BP reduction may cause the dissection plane and false lumen to expand, which may jeopardize blood flow to the vital organs. Patients with acute pulmonary edema and acute coronary syndrome should also have more rapid BP reduction to less than 140/90 mm Hg as soon as possible to decrease LV wall stress, a major determinant of myocardial oxygen demand. Thrombolytic therapy should not be given in patients with ST elevation myocardial infarction until BP is reduced below 180/110 mm Hg to avoid the risk of intracranial hemorrhage. Patients who present with primary intracranial hemorrhage pose a challenge in management because cerebral perfusion pressure is determined both by systemic BP and intracranial pressure. Similarly, for patients with ischemic stroke, rigorous reduction in BP could further compromise the flow to the infarct area and, more importantly, the border zone. More conservative reduction in BP with the goal for BP reduction of approximately 15% within the first hour is generally recommended in these two neurologic emergencies. Treatment of hypertensive emergency in specific clinical settings is summarized in Table 2–3.
Table 2–3. Treatment of Hypertensive Emergency in Special Populations
Selection of antihypertensive agents in hypertensive emergency also depends on the clinical presentation (Table 2–4). Patients with hypertensive encephalopathy or ischemic stroke should be treated with labetalol, nitroprusside, or nicardipine. Nitroprusside and nitroglycerin should be avoided in patients with intracranial hemorrhage because they may increase intracranial pressure. Phentolamine should be considered in patients with pheochromocytoma or adrenergic crisis such as cocaine intoxication. However, this drug may not be available in many countries, including the United States, on a regular basis. Nicardipine is an alternative therapy for pheochromocytoma. Labetalol, although effective in reversing cocaine-induced increase in BP, should be used with caution in pheochromocytoma because it is still a predominant BB (β-AR to α-AR blockade ratio of 4:1) and may precipitate an unopposed α-AR–mediated vasoconstriction. Nitroprusside is a potent vasodilator but should be used with caution in patients with renal failure given its propensity to cause cyanide toxicity. Nitroglycerin is suitable for use in patients with acute pulmonary edema and acute coronary syndrome because of its antianginal property and ability to relieve pulmonary congestion. Nitrate tolerance may develop quickly with continuous infusion and limit its antihypertensive efficacy. Fenoldopam is a dopamine D1-like receptor agonist, which reduces BP by promoting peripheral vasodilation. Clevidipine is a newer CCB available in intravenous form with shorter half-life (1 minute) than nicardipine (30–40 minutes). The relatively high cost of these newer agents limits routine use in clinical practice.
Table 2–4. Intravenous Antihypertensive Drugs for Hypertensive Emergencies ||Download (.pdf)
Table 2–4. Intravenous Antihypertensive Drugs for Hypertensive Emergencies
Contraindications and Side Effects
Start at 0.25–1.0 mcg/kg/min, increase by 0.5 mcg/kg/min every 5 minutes until goal BP or maximal dose of 10 mcg/kg/min
Renal failure, cyanide toxicity, reflex tachycardia, methemoglobin
Start at 10–20 mcg/min, increase by 5 mcg/min every 5 minutes until goal BP or maximal dose of 200 mcg/min
Start at 5 mg/h, increase by 2.5 mg/h every 15–30 minutes until goal BP or maximal dose of 15 mg/h; thereafter decrease to 3 mg/h maintenance
Start at 1–2 mg/h, titrate every 90 seconds to 10-minute intervals up to 32 mg/h
Defective lipid metabolism such as pathologic hyperlipemia or acute pancreatitis
Start at 0.25–0.5 mg/kg at 2–4 mg/min until goal BP, then 5–20 mg/h maintenance
Bradycardia, second- or third-degree atrioventricular (AV) block, bronchospasm, systolic heart failure
0.5–1 mg/kg as bolus; 50–300 mcg/kg/min as continuous infusion plus sodium nitroprusside intravenous gtt; start at 0.25–1.0 mcg/kg/min, increase by 0.5 mcg/kg/min every 5 minutes until goal BP or maximal dose of 10 mcg/kg/min
Bradycardia, second- or third-degree AV block, bronchospasm, systolic heart failure
1–5 mg, repeat after 5–15 minutes until goal BP is reached; 0.5–1.0 mg/h as continuous infusion
Start at 0.03–0.1 mcg/kg/min, titrated at the increment of 0.05 to 0.1 mcg/kg/min up to 1.6 mcg/kg/min
Increased intraocular pressure, reflex tachycardia
Management of Resistant Hypertension
Resistant hypertension is defined as failure to achieve BP target goal despite three or more drugs, one of which should be a diuretic. The first simple step in managing resistant hypertension, after excluding WCE, nonadherence to medications, and secondary hypertension, is to determine whether patients are on an appropriate class of diuretics based on renal function. Patients with estimated glomerular filtration rate (eGFR) > 50 mL/min/1.73 m2 should be treated with thiazide diuretics, particularly chlorthalidone, rather than loop diuretics because of longer half-life and proven efficacy in lowering BP. Patients with an eGFR of 30–40 mL/min/1.73 m2 or less should be on loop diuretics because the ability of thiazide diuretics to promote diuresis diminishes with impaired renal function. The use of an appropriate drug combination that provides synergistic effect on BP could minimize the number of medications needed to control hypertension. Assessment of hemodynamic variables is also helpful in deciding appropriate drug combination. For example, the use of BBs and a central sympatholytic drug generally yields minimal incremental benefit and is prohibited in patients with bradycardia or heart block. These patients should be treated with vasodilators such as DHP CCBs, ACEIs, ARBs, or hydralazine (Figure 2–2). Patients with elevated resting heart rate are more likely to derive large BP reduction with BBs, diltiazem, or verapamil because elevated heart rate is usually a good indicator for hyperkinetic circulation in hypertensive patients.
Algorithm in management of resistant hypertension. BP, blood pressure; CCB, calcium channel blocker; eGFR, estimated glomerular filtration rate; HR, heart rate.
Addition of spironolactone should also be considered in patients with resistant hypertension despite adjustment of medications, as mentioned earlier. An increasing body of evidence suggests that low-dose spironolactone between 12.5 and 25 mg/day, which is not likely to produce a major diuretic effect, causes a dramatic fall in BP on average of 25/12 mm Hg, when used as add-on therapy in patients with uncontrolled hypertension. Antihypertensive effect of spironolactone is observed even in patients with essential hypertension without an elevated aldosterone-to-renin ratio. Combination of DHP and non-DHP CCBs appears to have additive effects on peripheral vasodilation and BP, possibly due to binding to different sites of the receptors, and should also be considered in these patients. In contrast, addition of an ARB to ACEI has modest effects on BP, on average of only 5/3 mm Hg. The addition of long-acting nitrates may be considered in patients with isolated systolic hypertension who are refractory to treatment because it has been shown to be beneficial in one small study.
ACCORD Study Group, et al. Effects of intensive blood-pressure control in type 2 diabetes mellitus. N Engl J Med.
Egan BM, et al. Uncontrolled and apparent treatment resistant hypertension in the United States, 1988 to 2008. Circulation.
HYVET Study Group. Treatment of hypertension in patients 80 years of age or older. N Engl J Med.
Krause T, et al. Management of hypertension: summary of NICE guidance. BMJ.
Mann JF, et al. Renal outcomes with telmisartan
, or both, in people at high vascular risk (the ONTARGET study): a multicentre, randomised, double-blind, controlled trial. Lancet.