Abstract

Resistant hypertension is defined as blood pressure above the patient's goal despite the use of 3 or more antihypertensive agents from different classes at optimal doses, one of which should ideally be a diuretic. Evaluation of patients with resistive hypertension should first confirm that they have true resistant hypertension by ruling out or correcting factors associated with pseudoresistance such as white coat hypertension, suboptimal blood pressure measurement technique, poor adherence to prescribed medication, suboptimal dosing of antihypertensive agents or inappropriate combinations, the white coat effect, and clinical inertia. Management includes lifestyle and dietary modification, elimination of medications contributing to resistance, and evaluation of potential secondary causes of hypertension. Pharmacological treatment should be tailored to the patient's profile and focus on the causative pathway of resistance. Patients with uncontrolled hypertension despite receiving an optimal therapy are candidates for newer interventional therapies such as carotid baroreceptor stimulation and renal denervation.

1. Introduction

Hypertension is the most common chronic disease in the developed world affecting up to 25% of the adult population [1]. It remains the most important modifiable risk factor for coronary heart disease, stroke, congestive heart failure, renal disease, and peripheral vascular disease. Suboptimal blood pressure control is responsible for 62% of cerebrovascular disease, 49% of ischemic heart disease, and an estimated 7.1 million deaths a year [2]. Because of the associated morbidity, mortality and economic cost to society early diagnosis and treatment within the established guidelines is imperative. A sizeable percentage of the hypertensive population does not manage to achieve adequate control in spite of receiving 3 or more antihypertensive medications. These are the patients with resistant hypertension.

Resistant hypertension is defined by the Joint National Committee 7 as blood pressure that is above the patient’s goal despite the use of 3 or more antihypertensive agents from different classes at optimal doses, one of which should ideally be a diuretic [3]. Patients whose blood pressure is controlled but require 4 or more medications to do so should also be considered resistant to treatment. However, the definition does not include newly diagnosed hypertensives. Resistant hypertension is not synonymous with uncontrolled hypertension. The latter includes both patients with inadequately treated blood pressure due to poor adherence or inadequate treatment, as well as those with true resistant hypertension [3, 4]. The importance of resistant hypertension lies in the identification of patients who are at high risk of suffering complications from reversible causes of hypertension and patients who may benefit from a particular diagnostic or therapeutic approach [3].

The exact prevalence of resistant hypertension is unknown, in part because of its arbitrary definition. However, small studies estimate prevalence from 5% in general medical practice up to 50% in nephrology clinics [5]. In a prospective analysis of Framingham study data, a higher baseline systolic blood pressure along with older age, the presence of LVH and obesity (BMI > 30 kg/m2) were the strongest predictors of lack of blood pressure control [6, 7]. Results were similar in ALLHAT where the older, obese patients with higher baseline systolic blood pressure and LVH required 2 or more antihypertensive agents [8]. The strongest predictor however was serum creatinine over 1.5 mg/dL. Other patient characteristics associated with resistant hypertension include excessive salt ingestion, diabetes, black race, and female gender. Both studies showed greater difficulty in controlling systolic blood pressure compared to diastolic. Up to 92% of patients achieved target diastolic blood pressure while only 60%–67% achieved systolic blood pressure goals [6, 8]. It is likely that this condition will become increasingly common because of the aging population and a progressive increase in obesity and comorbidities such as diabetes.

There are also a few studies implicating gene mutations. A Finnish study found that certain variants of the β and γ subunits of the epithelial sodium channel gene ENaC were significantly more prevalent in patients with resistant hypertension [9]. Other studies associate the allele of the CYP3A5*1 enzyme with both higher blood pressure levels in normotensive people of black race, as well as with hypertension resistant to treatment [10, 11]. This particular enzyme is involved in the metabolism of cortisol and corticosterone. These and other genes that may be identified in the future hold the potential for the development of novel therapeutic targets.

2. Pseudoresistance Evaluation

The workup of patients with suspected resistant hypertension is summed up in Table 3.

The first step in evaluating a patient with uncontrolled blood pressure is to establish whether it is a case of true resistant hypertension or just pseudoresistance. The latter refers to a lack of blood pressure control despite receiving treatment without true resistance. This can be caused by easily reversible causes such as suboptimal blood pressure measurement technique, poor adherence to prescribed medication, suboptimal dosing of antihypertensive agents or inappropriate combinations, the white coat effect, and clinical inertia. It is important to exclude these causes before labeling a patient as having resistant hypertension.

Poor blood pressure measurement technique is quite common, usually the result of not letting the patient rest before measurement and using a small cuff [12]. Patients should always rest in a chair with their back supported for a minimum 5 minutes prior to measurement and the cuff’s air bladder must encircle at least 80% of the arm circumference. The average of two readings taken a minute apart represents the patient’s blood pressure.

Approximately 40% of newly diagnosed patients will discontinue their antihypertensive medication the first year of treatment [13, 14]. Eventually, less than 40% will continue taking their medication after 5 to 10 years [13, 15]. The most common causes are poor patient-physician communication concerning blood pressure goals and the importance of achieving them, potential side effects, high cost of treatment, and complex regimens [16, 17]. Adherence can be improved by choosing affordable agents with minimal side effects that are given once daily alone or in fixed dose combinations. Older patients with memory deficits or psychiatric illness can benefit from using pill boxes.

Clinical inertia can be described as a physician’s ignorance of treatment guidelines or reluctance to adhere to them due to lack of training or inexperience in antihypertensive medication, underestimation of cardiovascular risk, and overestimation of the treatment provided [18, 19]. This results in suboptimal dosing or inappropriate combinations of agents. A large part of this problem could be resolved if physicians familiarize themselves with one or two drugs in each class of antihypertensives. Proper training is imperative so that physicians realize the importance of treating to reach a goal blood pressure level of less than 140/90 mmHg and knowing when to refer patients to a hypertension specialist.

A white coat effect should be suspected in patients whose clinical blood pressure measurements are consistently and significantly higher than reliable out of office measurements. Other signs include repetitive symptoms of overtreatment such as orthostatic hypotension and persistent fatigue as well as absence of target organ damage including left ventricular hypertrophy, retinopathy, and chronic kidney disease [20, 21]. These cases must be confirmed with 24-hour ambulatory blood pressure monitoring. One study found that 20 to 30% of a patient population believed to have resistant hypertension was actually well controlled when measured by 24-hour ambulatory blood pressure monitoring [22]. Accurate home blood pressure values are the best guide for therapy. In elderly patients, especially diabetics, arterial stiffness may cause pseudoresistance because less compressible arteries cause falsely elevated blood pressure [23].

3. Concomitant Conditions

As was previously mentioned, obesity is associated with resistant hypertension. Obese patients have increased sympathetic activity, higher cardiac output, and a rise in peripheral vascular resistance due to reduced endothelium-dependent vasodilation. Plasma aldosterone and endothelin are also increased, while excessive surrounding adipose tissue results in increased intrarenal pressures and changes in renal architecture [24]. As the body mass index increases, progressively higher doses of antihypertensive drugs are required to control blood pressure [25]. Weight loss has been found to reduce both systolic and diastolic blood pressure [3, 26].

Another common concomitant condition in hypertensive patients is diabetes. Insulin resistance increases sympathetic nervous activity, vascular smooth muscle cell proliferation, and sodium retention leading to elevated blood pressure resistant to treatment [3]. The common comorbidities of obesity, hypertension, and diabetes induce renal dysfunction, further hindering blood pressure treatment.

Dietary factors include increased salt and alcohol consumption. Although small amounts of alcohol (2 drinks/day) have vasodilating effects and may lower blood pressure, consumption of more than 30 mL daily raises blood pressure and may increase cardiovascular risk. Older patients, patients of African origin, and patients with chronic kidney disease are particularly susceptible to salt intake [3]. Current guidelines recommend that dietary sodium for a hypertensive person should be under 100 mmol/day (2.4 g sodium or 6 g sodium chloride) and even lower in salt sensitive patients [27]. Excessive salt intake can be assessed by measuring sodium excretion in a 24-hour urine collection.

Several common medications can cause elevated blood pressure and hinder treatment. Perhaps the most common are nonsteroidal anti-inflammatories including COX-2 inhibitors and aspirin, decongestants (phenylephrine and pseudoephedrine), stimulant agents used for weight loss, narcolepsy or attention deficit disorder, contraceptives, cyclosporine, and erythropoietin [3]. Corticosteroids increase blood pressure through fluid retention, particularly but not limited to those with increased mineralocorticoid activity. Licorice and herbal medication that contains stimulants such as ephedra can also cause hypertension.

4. Assessment of Secondary Causes of Hypertension

Secondary causes of hypertension are common in patients with resistant hypertension, particularly in the elderly. These include obstructive sleep apnea, renal parenchymal disease, renal artery stenosis, and primary aldosteronism [28, 29] (Table 1).

Obstructive sleep apnea is particularly frequent in patients with resistant hypertension. A small study of 41 patients with resistant hypertension discovered that 83% suffered from sleep apnea [30]. The severity of sleep apnea is positively associated with the likelihood of resistant hypertension [31, 32]. Several mechanisms are believed to contribute to this effect. Intermittent hypoxemia and increased upper airway resistance induce a sustained increase in sympathetic nervous system activity [33, 34]. Also, there seems to be a significantly higher prevalence of primary aldosteronism in patients with sleep apnea [35, 36]. Alternatively, obesity may be the common factor that increases risk for both obstructive sleep apnea and excess aldosterone production [35]. A sleep study is indicated in patients with resistant hypertension and other signs and symptoms of sleep apnea including obesity, large neck size, excessive loud snoring, interrupted sleep, daytime somnolence, polycythemia, and carbon dioxide retention [37, 38]. Treatment with a continuous positive airway pressure device (CPAP) has been shown to reduce blood pressure and thus is beneficial in resistant hypertension patients with obstructive sleep apnea [3].

The recent studies suggest that primary aldosteronism is a much more common cause of hypertension than originally believed. Particularly in patients with resistant hypertension, the prevalence of primary aldosteronism has been found between 10% and 20% [3, 39, 40]. Aldosterone exerts a number of effects leading to increased systemic vascular resistance, such as endothelial dysfunction, vascular remodeling through collagen deposition, vascular damage, impairment of the baroreflex leading to loss of compensation for elevated blood pressure, and hypovolemia [4143]. It has been suggested that obesity is involved, causing a generalized activation of the renin-angiotensin-aldosterone system, perhaps through excretion of cytokines from adipocytes [44, 45].

Primary aldosteronism may be suggested by hypokalemia; however this is often a late manifestation preceded by the development of hypertension [4648]. Screening should be done with plasma renin and serum aldosterone ratio measurement (which has a high sensitivity but low specificity) and confirmed with sodium loading or fludrocortisone suppression testing [39, 49].

Renal parenchymal disease is both a cause and a complication of poorly controlled hypertension [50, 51]. As was previously mentioned, in ALLHAT serum creatinine above 1.5 mg/dL was the strongest predictor of failure to achieve goal blood pressure [8]. Resistant hypertension in chronic kidney disease is mainly due to activation of the renin-angiotensin system, sodium retention, and the resulting intravascular volume expansion [52]. Other factors include activation of the sympathetic nervous system due to decreased blood flow to the kidney, alterations in vasoconstrictor and vasodilator excretion from the endothelium, and increased arterial stiffness [52].

Renal artery stenosis is a relatively common finding in hypertensive patients undergoing cardiac catheterization with approximately 20% of patients having unilateral or bilateral stenosis above 70% [53]. However the causative role of these stenosis in hypertension remains unknown since only a few patients actually benefit from surgical or endovascular revascularization [54, 55]. The majority of cases (90%) are due to atherosclerotic lesions and are seen in older patients, smokers, patients with known atherosclerotic disease and unexplained renal insufficiency [56]. The other 10% are fibromuscular lesions, commonly in women under 50 years of age, and these are the patients that will usually improve blood pressure control after revascularization [3]. Bilateral renal artery stenosis should be suspected in patients with a history of “flash” pulmonary edema with preserved systolic heart function. Screening can be done using magnetic resonance angiography (MRA), computer tomographic angiography (CTA), Doppler ultrasonography, or angiotensin converting enzyme (ACE) inhibitor renography [49, 57].

Pheochromocytoma is a rare cause of resistant hypertension with a prevalence of 0.1%–0.6% among hypertensives [58, 59]. The average time between initial symptoms and diagnosis is 3 years, and many cases are missed altogether according to autopsy studies [60]. Clinical signs include episodic headaches, palpitations, and sweating. The best screening test for pheochromocytoma is 24-hour urinary metanephrines or plasma free metanephrines (normetanephrine and metanephrine) which carries a 99% sensitivity and an 89% specificity [61].

Hypertension is a common manifestation of Cushing’s syndrome. Up to 90% of patients are hypertensive and 17% have resistant hypertension [62, 63]. The primary mechanism is increased mineralocorticoid activity that leads to increased intravascular volume [64]. However, other factors such as obstructive sleep apnea and insulin resistance also contribute substantially [65, 66]. Target organ damage in Cushing’s syndrome is more severe than in primary hypertension because it is associated with many other cardiovascular risk factors such as diabetes, obstructive sleep apnea, obesity, and dyslipidemia [67, 68].

Thyroid and parathyroid dysfunction are common reversible causes of secondary hypertension. Patients with hyperthyroidism usually present with systolic hypertension and those with hypothyroidism have diastolic hypertension. Most patients with primary hyperparathyroidism are diagnosed because of routine findings of hypercalcemia [69, Table  1-2].

5. Pharmacological Treatment

Pharmacological treatment should be based on the most common causes of resistant hypertension and focused on blocking all the physiological pathways to blood pressure elevation [18, 70]. Antihypertensive agent doses should be titrated upward until blood pressure is controlled or the maximum recommended dosage is reached, unless the patient experiences dose related adverse effects. It is then appropriate to add a drug from another class that has additive or synergistic effects with the first drug. In general, a typical regimen should include a diuretic, an ACE inhibitor or angiotensin receptor blocker (ARB), a calcium channel blocker (CCB), and a β-blocker.

The timing of medication administration can also affect blood pressure control. Switching one of 3 or more medications from morning to bedtime administration can result normalization of blood pressure in 21.7%–37% of patients [71, 72]. This is particularly important in nondippers. Since volume overload is the most frequent underlying pathophysiology and suboptimal dosing the most frequent cause of resistant hypertension, adding, increasing or changing diuretic therapy is the key to successful treatment and will help over 60% of patients achieve target blood pressure [18, 7377].

Patients with normal kidney function should receive 12.5–25 mg/day of hydrochlorothiazide although some will benefit from doses up to 50 mg/day [73]. Chlorthalidone at 25 mg/day is an alternative that offers greater 24-hour blood pressure reduction than 50 mg/day hydrochlorothiazide with the greatest difference occurring overnight and may be preferred in certain patients with resistant hypertension [78]. When given at the same dose as hydrochlorothiazide, chlorthalidone will reduce blood pressure an additional 8 mmHg according to a small study [79]. Patients should be monitored for hyponatremia and hypokalemia, and caution is warranted in patients with a history of prediabetes and gout. A common pitfall is not realizing that the patient’s kidney function is deteriorating and not switching diuretic class when it does. Thiazide diuretics are not effective in chronic kidney disease. Therefore, they must be replaced with loop diuretics when the estimated glomerular filtration rate (eGFR) falls below 40 mL/min/1.73 m2 [27, 73, 80]. In a study of 12 elderly patients with hypertension whose blood pressure was uncontrolled on multidrug regimens the use of furosemide significantly improved blood pressure control [81]. Furosemide and bumetanide have a relatively short half-life and should be dosed twice daily in order to avoid reactive sodium retention due to intermittent natriuresis and consequent activation of the renin-angiotensin system [73, 80, 82].

When optimal diuresis fails, other medications should be considered. Since subclinical aldosteronism is a common occurrence in resistant hypertension, low doses of spironolactone (25–50 mg/day) or eplerenone can be particularly helpful. Patients most expected to benefit from mineralocorticoid blockade include those with primary hyperaldosteronism, the obese, and those suffering from obstructive sleep apnea. According to one study, the addition of spironolactone 12.5–25 mg/day to 76 patients with uncontrolled hypertension taking an average of 4 antihypertensive agents resulted in an average 25/12 mmHg reduction after 6 months [83]. The blood pressure lowering arm of the ASCOT study had similar results for patients who were unselected for aldosterone/plasma renin activity. When spironolactone was added as a fourth line agent, blood pressure dropped by 21.9/9.5 mmHg. At this dosage spironolactone is safe, well tolerated and provides significant additive blood pressure reduction [84, 85]. Eplerenone may be more suitable for patients requiring spironolactone doses above 25 mg/day because breast tenderness is a common adverse effect at higher doses [86]. Hyperkalemia is another risk that must be monitored.

Amiloride is an alternative indirect aldosterone antagonist that is better tolerated than spironolactone. One small study found that the addition of 2.5 mg/day amiloride decreased blood pressure by 31/15 mmHg [40]. However, it has been shown that 10 mg of amiloride has half the blood pressure reduction capability of 25 mg spironolactone and therefore should be considered only when spironolactone is not tolerated [87].

Blockade of the renin angiotensin system with ACE inhibitors or ARBs in patients that are intolerant of ACE inhibitors is particularly recommended in patients with diabetes mellitus, heart failure, postmyocardial infarction, chronic kidney disease, high coronary disease risk, and recurrent stroke prevention [27, 88]. Dosage should be increased to the maximum recommended dosage as long serum creatinine does not increase more than 35% above baseline and hyperkalemia does not develop [111]. Dual ACE and ARB therapy is no longer recommended in most cases because of the possibility of adverse renal outcomes [89, 90].

Aliskiren, the only available direct renin inhibitor, is at least as effective as ARBs in reducing end target organ damage but has not been directly tested in resistant hypertension. The ALLAY trial showed that aliskiren monotherapy was as effective as losartan in reducing LVMI, although the combination of both did not achieve a statistically significant further LVMI regression [91]. The addition of aliskiren to losartan did however seem to have additional renoprotective effects in another study, reducing the mean urinary albumin creatinine ratio by 20% in patients with diabetic nephropathy [92]. The additional blood pressure reduction was marginal and therefore the role of direct rennin inhibitors in resistant hypertension remains undetermined.

Polypharmacy is difficult to avoid because blood pressure can be controlled by using one drug in only about 50% of patients. Fixed dose combinations offer the convenience of taking fewer pills, combining antihypertensive agents with additive or synergistic effect and reducing dose-dependant adverse effects of individual components. The latter is evident in ACE inhibitor or ARB combinations with CCBs, as the ACE inhibitors/ARBs reduce the peripheral edema that frequently develops with dihydropyridine calcium channel blocker therapy. Another popular fixed dose combination with synergistic effects is ACE inhibitors/ARBs with diuretics, since the latter enhance the antihypertensive efficacy of all the other classes. The recent introduction of triple agent combinations containing dihydropyridine CCBs as well is expected to lessen the burden of polypharmacy and further improve adherence [86].

CCBs are particularly indicated in black and elderly patients. The ACCOMPLISH study suggested that combining ACE inhibitors with CCBs was more effective at preventing major cardiovascular and renal events than ACE inhibitors with diuretics despite achieving similar blood pressure control rates [93]. Both regimens are available as fixed dose combinations and are useful options in different circumstances. In cases of true resistant hypertension, there are also data to support adding a complementary non dihydropyridine CCB to a regimen including a RAS blocker, diuretic, and dihydropyridine CCB. Such a combination of complementary CCBs results in additive BP reduction with a low-side effect profile and makes pharmacological sense [94, 95].

Beta-blockers are indicated in the setting of coronary artery disease, congestive heart failure, and postmyocardial infarction. When adding on combinations already including a diuretic, an ACE inhibitor or ARB and a calcium channel blocker, vasodilating β-blockers should be preferred [73, 96]. Beta-blockers as well as loop diuretics are also usually necessary when administering direct vasodilators to overcome the reflex tachycardia and fluid retention, respectively [97]. If BP control is still not achieved with full doses of a 4-drug combination, use of other agents such as centrally acting alpha-agonists (methyldopa and clonidine) or vasodilators (hydralazine or minoxidil) is needed. These agents are very effective for lowering BP but have poor tolerability, require frequent dosing, and lack positive outcome data [27]. At this stage, the intervention of a hypertension specialist is warranted. Besides these established treatments, new antihypertensive agents are being developed.

Endothelin receptor antagonists are a new family of antihypertensive medications that are currently being evaluated. Darusentan is a selective antagonist for type A endothelin receptors, activation of which causes vasoconstriction and proliferation of vascular smooth muscle [98]. It has demonstrated significant dose-dependant reductions in both systolic and diastolic blood pressures and has been positively evaluated in resistant hypertension [98, 99]. Unfortunately, another unpublished phase 3 clinical trial failed to meet its coprimary end point and the drug’s future remains uncertain [100]. Atrasentan is another highly selective endothelin receptor antagonist that has shown positive results in blood pressure reduction for 72 patients [101]. Interestingly, it also had a positive influence on the patients metabolic profile. Another promising category under development is medication that combines inhibitors of vasoconstrictive mediators with drugs that potentiate vasodilating mediators by inhibiting their breakdown by neutral endopeptidases (NEPs). Omapatrilat is such an agent that has been evaluated favorably in the OCTAVE trial [102]. Vaccines targeting angiotensin I and II are also being developed and tested [103, 104].

In any case, treatment should be tailored to the patient’s profile, lifestyle, and comorbidities. Constructing a regimen that is acceptable to the patient, well tolerated and will maintain long-term compliance is important. Yet in some patients, optimal blood pressure control will not be achieved even with the most carefully designed regimen. In these cases, new device-based approaches for blood pressure control are being evaluated (Tables 1 and 2).

One of these devices is the Rheos device (CVRx, Maple Grove, Minn) which stimulates the carotid baroreceptors for better blood pressure control by taking advantage of chronic electrical activation of the afferent limb of the carotid baroreflex. The device consists of a pulse generator and bilateral perivascular carotid sinus leads that are implanted under narcotic anesthesia. According to the findings from the Device-Based Therapy of Hypertension (DEBuT-HT) study that were recently presented, after four years of treatment, Rheos reduced systolic blood pressure by an average of 53 mmHg (193 mmHg versus 140 mmHg). Blood pressure was reduced significantly each year, with the largest decrease occurring in year four. Many of these patients were able to reach their blood pressure goal and reduce the number of medications that patients were taking to treat their hypertension from an average of 5 at baseline to 3.4 medications at 4 years. Baroreflex activation therapy also improved functional capacity and reduced left ventricular mass without any evidence of carotid injury or stenosis [105].

Another target for the interventional treatment of resistant hypertension is catheter-based renal nerve ablation. Renal sensory afferent nerve activity directly influences sympathetic outflow to the kidneys and other highly innervated organs involved in cardiovascular control, such as the heart and peripheral blood vessels, by modulating posterior hypothalamic activity [106, 107]. All these components are stimulated in hypertension and contribute to blood pressure elevation.

Renal sympathetic nerve ablation is achieved percutaneously via the lumen of the renal artery, using a catheter connected to a radiofrequency generator. Treatment has been administered to 45 patients with resistant hypertension taking a median of 4.7 antihypertensive agents. Followup at 1 and 2 years has shown a sustained blood pressure reduction of 27/11 mmHg [108, 109]. The multicentre Simplicity HTN-1 study that was recently presented at the European Society of Hypertension’s 20th meeting included 108 patients with persistently elevated blood pressure despite treatment with an average of five medications. Catheter-based renal denervation produced a mean blood pressure reduction of 33/15 mmHg at 24 months without evidence of vascular or renal abnormalities. Results from the Simplicity HTN-2 study comparing renal denervation treatment to rigorous medical therapy are expected by the end of 2010.

6. Conclusions

Resistant hypertension remains a challenging clinical problem that will increasingly become more common. Causes of resistance should be considered when blood pressure does not respond satisfactorily to a rational triple antihypertensive regimen that includes a diuretic. The workup of patients with resistive hypertension should be a two-step approach (Table 3): first, confirmation that it is indeed true resistant hypertension by ruling out or correcting factors associated with pseudoresistance, and second, identification of the true factors involved in treatment resistance. The cornerstone of therapy remains a rigorous evaluation followed by correction of contributing causes and appropriate pharmacological treatment. Newer interventional therapies may become a viable option in the future for those patients with uncontrolled hypertension despite receiving an optimal multiple medication antihypertensive regimen and those who cannot tolerate the medication.