Contributing factors in resistant hypertension

Truly refractory disease is rarely found in a properly conducted workup

Donald G. Vidt, MD

VOL 107 / NO 5 / MAY 1, 2000 / POSTGRADUATE MEDICINE

CME learning objectives

• To identify clinical factors that contribute to resistant hypertension
• To understand the relevance of secondary forms of hypertension and recognize clues to their diagnosis
• To learn hemodynamic and humoral abnormalities that may contribute to resistant hypertension and classes of drugs that may counter these effects

This page is best viewed with a browser that supports tables

Preview: Many patients are considered to have resistant hypertension when their blood pressure remains elevated despite use of multiple antihypertensive agents. However, the prevalence of the problem is difficult to determine because of differences in diagnostic criteria, physicians' practice patterns, and therapeutic approach. Numerous factors can contribute to treatment resistance, so every patient should undergo a thorough, orderly, and carefully directed evaluation. In this article, Dr Vidt describes such an evaluation.
Vidt DG. Contributing factors in resistant hypertension: truly refractory disease is rarely found in a properly conducted workup. Postgrad Med 107(5):57-70

An estimated 50 million adults in the United States have arterial hypertension. Although trials conducted over the past 30 years have shown antihypertensive therapy to be effective in reducing the risks of stroke, coronary artery disease, congestive heart failure, renal failure, and progression of hypertension, in most patients receiving such therapy, blood pressure remains above recommended treatment goals (1).

Hypertension is usually considered resistant if blood pressure cannot be reduced to less than 140/90 mm Hg using an adequate and appropriate triple-drug regimen, including an oral diuretic, with all three drugs near the maximum recommended dose. In older patients with isolated systolic hypertension, resistance is defined as failure to achieve a systolic blood pressure of less than 160 mm Hg with a similar regimen. (Some clinical trials suggest that a lower systolic blood pressure--probably 150 mm Hg--would be more appropriate (2,3).)

Available studies have suggested a prevalence of resistant hypertension ranging from as low as 3% in carefully structured programs to as high as 29% in tertiary-care hypertension clinics (4-6). In clinical practice, the prevalence of true drug resistance is extremely low. However, because hypertension remains uncontrolled in three fourths of all patients who have the problem, significant numbers of patients are labeled as resistant to treatment and are referred for further clinical evaluation.

Common causes of resistant hypertension may be revealed by careful history taking and clinical examination, as outlined in the algorithm in figure 1 (not shown). The following text describes how to obtain answers to the questions in the algorithm and how to proceed when additional laboratory and diagnostic studies are required (7).

Is the regimen adequate?

Physicians' tendency to undertreat medical conditions pharmacologically is well recognized, and an inadequate regimen can contribute significantly to resistant hypertension. For example, in one study of 436 patients treated at a hypertension clinic (6), 91 (21%) patients had resistant hypertension. A cause was identified in 83 (91%) of the 91 patients, the most common being suboptimal therapy (figure 2: not shown), which was found in 39 (43%) patients. Blood pressure was brought under control or improved (ie, >15% decrease in diastolic pressure) in 58 (68%) of the 91 patients with modification of or addition to the regimen.

The following are important steps in evaluating the appropriateness of a patient's antihypertensive regimen:

• Look for inappropriately low dosages of prescribed medications and for inappropriate combinations (eg, two agents in the same class, or multiple agents but no oral diuretic).
• Ensure that dose and frequency of administration are suitable for the particular drugs being used. Most antihypertensive agents currently being marketed are approved for once-daily administration, but they may not provide full 24-hour effects when lower doses are prescribed.
• Consider patient characteristics that may influence drug choice. For example, a short-acting loop diuretic may be inappropriate in a patient with normal renal function, just as a thiazide diuretic would be less effective in a hypertensive patient with renal insufficiency.

Are drug interactions likely?

Most hypertensive patients who are middle-aged or older have comorbid conditions that require regular use of specific medications that may interact with antihypertensive agents.

For example, nonsteroidal antiinflammatory drugs (NSAIDs) can cause sodium retention, enhance vasoconstrictor responses to pressor hormones, and antagonize the effects of all antihypertensive agents, with the possible exception of calcium channel blockers (8). In addition, they can lead to increases in mean arterial pressure (average, 4 to 5 mm Hg). Over-the-counter nasal sprays, oral decongestants, and appetite suppressants may contain such vasoactive compounds as phenylpropanolamine hydrochloride, ephedrine, and pseudoephedrine, which can induce hypertension or interfere with pharmacologic treatment of hypertension when ingested regularly (9).

Women who take oral contraceptives (particularly women who are obese and who smoke) are two to three times more likely to have hypertension than those who do not. Tricyclic antidepressants can antagonize the hypotensive effect of selected adrenergic blocking agents, such as clonidine hydrochloride (Catapres) and guanethidine monosulfate (Ismelin). Regular use of corticosteroids for any clinical condition may promote salt and water retention, leading to volume-dependent hypertension. Cyclosporine (Neoral, Sandimmune, SangCya), erythropoietin (Epogen, Procrit), and anabolic steroids can affect blood pressure in selected patients.

In view of the prevalence of recreational drug use, cocaine abuse should not be overlooked in evaluation of patients with resistant hypertension. Consideration of injecting drug use is mandatory in any patient presenting to an emergency department with hypertension-related problems.

Is the patient compliant with therapy?

Noncompliance with recommended therapy is a costly and pervasive problem that lowers overall hypertension control rates and is a major contributor to resistant hypertension. Obviously, if patients are unable or unwilling to adhere to recommended lifestyle modifications or drug regimens, long-term control will be impaired.

Compliance problems are seen in all ages and tend to progress over time, particularly with chronic conditions such as hypertension. Education level, socioeconomic status, and sex are not particularly predictive of adherence to recommendations, whereas smoking, excessive alcohol intake, use of recreational drugs, and other unhealthy behaviors are closely linked to noncompliance.

Careful history taking should address prior medications and response to therapy and proceed to a review of current medications, dose, frequency of administration, and response. Such a drug history should elucidate inadequate dosages or inappropriate combinations of drugs. Direct and nonthreatening inquiry regarding adherence to prescribed drug regimens usually brings a forthright response. Addressing the complexity of the regimen, inconvenience of dosing, and cost of medication often uncovers treatment problems. The following are steps that are useful in evaluating patient compliance:

• Look for symptoms or complaints that may point to an adverse reaction that is contributing to irregular dosing or intermittent interruption in use of the prescribed agents.
• Try to determine whether the dosing frequency of each agent is consistent with its optimal duration of action.
• Ask about use of over-the-counter drugs (eg, NSAIDs, oral decongestants) and recreational drugs and whether taking them interferes with taking prescribed medications.
• Check whether the interval between appointments is long or the medicine regimen is complicated, because over time both tend to lead to missed appointments, skipped prescription refills, and decreased compliance.
• Inquire about beliefs, perceptions, or attitudes regarding adverse effects of medications that may lead to limitation in dosing or frank noncompliance. Subjective complaints probably exceed objective adverse effects actually experienced with most of today's antihypertensive medications.

If careful history taking has not clarified potential adherence problems, clinical clues should be sought on physical examination. The absence of reflexive tachycardia in a patient receiving a direct vasodilator (eg, minoxidil [Loniten]) but no beta blocker or the absence of bradycardia in a patient in whom a moderate dose of a beta blocker has been prescribed should raise the question of poor treatment adherence. The absence of dry mouth or drowsiness in a patient taking clonidine or the absence of pedal edema in a female patient taking a near-maximal dose of a dihydropyridine calcium antagonist should also raise warning flags.

Are associated conditions likely?

Excessive alcohol intake is the most common cause of reversible hypertension in US society and should always be suspected in patients with resistant hypertension. Average ingestion exceeding 60 mL (1.8 fl oz) per day as ethanol can have an impact on blood pressure. Heavy cigarette smoking has been associated with a persistent rise in blood pressure and can blunt the efficacy of beta blockers, particularly in African Americans.

Obesity and hyperinsulinemia diminish the effectiveness of antihypertensive drug therapy and can contribute to resistant hypertension. Insulin-induced hypertrophy of the smooth muscle in resistance vessels is one proposed explanation for the finding of increased vascular resistance in patients with refractory hypertension (10).

Obstructive sleep apnea, suggested by such findings as obesity, excessive drowsiness, and observed apneic episodes during sleep, also may contribute to hypertension (11). Appropriate sleep studies can confirm the presence of the condition.

Is pseudohypertension or office hypertension likely?

Pseudohypertension is found occasionally in older patients with generalized atherosclerosis leading to a stiffened brachial artery (12). In this condition, pressures obtained using a sphygmomanometer cuff are inappropriately high compared with intra-arterial pressures. Pseudohypertension should be considered in patients who have markedly elevated pressures in the absence of apparent target organ damage, radiologic evidence of pipe-stem calcifications in the brachial artery, severe and isolated systolic hypertension, or brachial artery pressures higher than lower extremity pressures.

In addition, watch for symptoms consistent with hypotension (ie, dizziness, fatigue) that occur in the absence of excessively low cuff measurements. In such cases, use of an infrasonic or oscillometric device appears to achieve blood pressure measurements that more closely approximate intra-arterial blood pressure than does indirect auscultatory sphygmomanometry.

In 20% to 50% of patients referred for hypertensive evaluation, blood pressure is lower, if not normal, on readings taken outside the medical environment (thus the label "white coat hypertension") (13). When faced with a history of variable blood pressure or out-of-office measurements that are lower or normal compared with office measurements, consider confirming the findings with home blood pressure recordings. Inexpensive, calibrated, digital devices and 24-hour ambulatory monitors are available for patients to take blood pressure recordings at home. However, physicians must feel comfortable offering treatment recommendations on the basis of readings patients have obtained at home.

Is volume overload present?

Volume expansion is a classic feature of primary hyperaldosteronism and a common secondary cause of resistant hypertension (14); in fact, volume overload is the most common physiologic cause of resistant hypertension. It should be considered in patients being treated for hypertension with three or more drugs, none of which is a diuretic. In particular, use of direct vasodilators (eg, minoxidil, hydralazine hydrochloride [Apresoline]) or adrenergic blocking drugs (beta blockers, alpha blockers) can lead to volume expansion and acquired treatment resistance (pseudoresistance) if not accompanied by an appropriate diuretic. Volume expansion should also be suspected in a patient with normal renal function whose regimen includes a single daily dose of a loop diuretic and in patients with a history of excessive dietary sodium intake.

A simple 24-hour urine sodium measurement often uncovers excessive dietary ingestion of sodium, which is suggested by sodium excretion exceeding 120 mEq/24 hours, including in patients receiving diuretic therapy. A plasma volume measurement can accurately assess the degree of volume overload and can serve as a guide to diuretic therapy. If these studies are not readily available, a trial of increased diuretic therapy can be rewarding.

In a patient already taking a thiazide diuretic, enhanced sodium restriction and addition of a loop diuretic may prove helpful. In patients with renal insufficiency who are receiving a loop diuretic, the addition of a long-acting diuretic, such as metolazone (Mykrox, Zaroxolyn), may be indicated.

Are causes of secondary hypertension present?

To this point, evaluation has been exceptionally cost-effective, requiring little or no laboratory testing, and 80% to 85% of recognized causes of resistant hypertension have been assessed. In addition, thorough history taking and clinical examination have already addressed the next step in the algorithm--uncovering clues to the presence of common secondary and potentially reversible causes of hypertension. Confirmation of clinical clues may require additional screening studies and tests (table 1) for appropriate diagnosis or localization of lesions.

Table 1. Screening and diagnostic studies for secondary hypertension
Possible cause	Study
Coarctation of aorta	Chest film* Two-dimensional echocardiogram Aortogram Magnetic resonance imaging (MRI)
Cushing's syndrome	Dexamethasone suppression test* 24-hr urinary free cortisol* Computed tomography (CT)
Primary aldosteronism	Plasma aldosterone:renin ratio* Aldosterone excretion rate during salt loading Adrenal CT
Pheochromocytoma	Plasma catecholamines* Urine catecholamines* (metanephrine, vanillylmandelic acid) Clonidine suppression test CT, MRI Iodine 131 metaiodobenzylguanidine scan
Renovascular disease	Captopril renography Renal duplex sonography* MRI Angiography Renal vein renin ratio
Thyroid disease	Thyrotropin level* Serum thyroid hormone level* Serum calcitonin level
Hyperparathyroidism	Calcium and phosphorus levels* Serum parathyroid hormone level
Renal parenchymal disease	24-hr urine protein and creatinine levels* Renal ultrasound* Iothalamate glomerular filtration rate Renal biopsy
*Preferred and readily available for additional screening studies.

A secondary cause of hypertension is found in 10% to 11% of patients undergoing evaluation for resistant hypertension, most often one of the following.

Coarctation of the aorta
Because coarctation of the aorta is often asymptomatic in adults, the medical history may be of little help unless an accompanying congenital malformation (eg, a bicuspid aortic valve, patent ductus, ventricular septal defect) is present. Absent or reduced pulses in the legs together with lower blood pressure in the legs than in the arms are a valuable clue to diagnosis. Upper extremity systolic blood pressure is elevated disproportionately to diastolic blood pressure, resulting in a wide pulse pressure and bounding pulses. A thrill may be felt in the suprasternal notch together with palpable pulsations or bruits heard over the intercostal arteries.

Cushing's syndrome
A history of considerable weight gain and recent change in facial appearance together with characteristic physical features (eg, truncal obesity, moon face, plethora, purplish skin striae) is indicative of Cushing's syndrome (15).

Primary aldosteronism
Clinical findings other than resistant hypertension may be of little help in identifying this disorder. Hypokalemia, whether spontaneous or provoked, provides the best clue to the presence of primary aldosteronism (16). In the absence of hypokalemia, aldosteronism may masquerade as essential hypertension.

Pheochromocytoma
A history of episodes of headache, palpitations, pallor, profuse perspiration, and unusual lability of blood pressure characterizes pheochromocytoma. It may present as accelerated or malignant hypertension and may be associated with multiple endocrine neoplasia type II in conjunction with pancreatic islet cell tumor, neurofibromatosis, von Hippel-Lindau disease, and medullary carcinoma of the thyroid or parathyroid gland.

Renovascular disease
The most useful clues to the presence of renovascular disease are patient age of less than 30 or more than 50 years, systolic or diastolic epigastric bruit, and renal insufficiency that is unexplained or induced by angiotensin-converting enzyme (ACE) inhibitor therapy. Hypertension is often accelerated or malignant, of abrupt onset, accompanied by diffuse vascular disease, and refractory to triple-drug therapy (17).

Thyroid and parathyroid disease
Thyroid dysfunction together with renovascular hypertension is the most common form of reversible secondary hypertension in patients over age 60. Patients with thyrotoxicosis have hyperdynamic hypertension and high cardiac output, seen predominately as elevated systolic blood pressure, whereas hypothyroidism is associated with diastolic hypertension. This difference can be a valuable clue in the elderly, in whom primary diastolic hypertension is rare.

Most patients with primary hyperparathyroidism are asymptomatic. The clinical diagnosis is suggested by the finding of hypercalcemia, which causes such side effects as polyuria, polydipsia, renal calculi, and peptic ulcer disease.

Renal parenchymal disease
Patients with renal parenchymal disease usually present with renal insufficiency, proteinuria, or hematuria that is easily detected with urinalysis and renal function tests (18). This disease is a common cause of secondary hypertension and, unfortunately, is often not reversible.

Should the regimen be modified?

If systematic evaluation has not uncovered a likely cause for resistance of blood pressure to treatment, re-evaluation of the combination of agents and the dosages being used is appropriate. By definition, patients with resistant hypertension should already be following a regimen of no less than three antihypertensive agents from different classes, including an oral diuretic in near maximum doses. At this point, regimen modification may consist of adding an agent from yet another class or adjusting currently used drugs to above the usually recommended maximum dose. The risk in increasing doses is the potential for adverse drug effects.

Because of the frequency with which volume expansion contributes to resistance in these patients, it is also appropriate to reassess the possibility of hypervolemia. This is especially important if a plasma volume measurement or 24-hour urine sodium test was not previously performed in a patient who is presumably adhering strictly to a sodium-restricted diet.

In selected patients, the direct-acting vasodilator minoxidil can be a potent addition to therapy. However, fluid retention is common with use of this agent, so a loop diuretic and beta blocker should be added to the regimen to prevent reflex tachycardia.

The central agonist clonidine or guanfacine hydrochloride (Tenex) can be beneficial if the patient can tolerate the associated drowsiness and dry mouth. If small doses prove effective, use of the transdermal preparation of clonidine (Catapres-TTS) may lessen adverse effects.

Most currently available antihypertensive drugs offer additive effects when used together (figure 3: not shown) (19). There may be an increased risk of cardiac conduction abnormalities when beta blockers and nondihydropyridine calcium antagonists are used in the same regimen. An ACE inhibitor and a beta blocker is a less effective combination than an ACE inhibitor used with other agents.

Is hemodynamic and neurohumoral investigation needed?

After completion of the evaluation steps discussed, resistance to antihypertensive therapy should be found in no more than 3% to 5% of patients. In these cases, careful hemodynamic and neurohumoral evaluation may be prudent, because a possible mechanism of drug resistance may be identified and a more targeted modification of the treatment regimen may be possible.

Any hemodynamic or neurohumoral aberrations that are uncovered may be a primary cause of the resistant hypertension or may represent a compensatory reaction to other drugs in the regimen. Regardless, such observations may facilitate drug therapy through removal of potential protagonist agents or addition of agents capable of counteracting a given mechanism. Table 2 lists hemodynamic and humoral measurements obtainable by careful evaluation and specific classes of agents that may be useful in modifying these mechanisms.

Table 2. Hemodynamic and humoral changes in resistant hypertension and classes of agents recommended for treatment
Elevated finding	Recommended treatment*
Cardiac output	Beta blocker Nondihydropyridine calcium antagonist
Peripheral resistance	ACE inhibitor, angiotensin-receptor blocker Dihydropyridine calcium antagonist Minoxidil (Loniten) Hydralazine HCl (Apresoline)
Plasma volume	Loop diuretic Strict sodium restriction
Plasma catecholamines	Clonidine HCl (Catapres) Guanfacine HCl (Tenex) Alpha blocker
Plasma renin activity	ACE inhibitor, angiotensin-receptor blocker Beta blocker
Plasma/urinary aldosterone	Spironolactone (Aldactone) Amiloride HCl (Midamor)
ACE, angiotensin-converting enzyme. *If patient is already taking drugs in this column, consider dosage increase. Adapted, with permission, from Oparil S, Weber MA, eds. Hypertension: a companion to Brenner and Rector's the kidney. Philadelphia: Saunders, 2000:564-72.

Summary

Systematic evaluation of each patient with refractory hypertension is necessary to identify conditions or altered mechanisms of hypertension that are amenable to targeted therapy. Issues that should be carefully investigated are appropriateness of the regimen, possible drug interactions, patient compliance, associated conditions (alcohol or recreational drug use, hyperinsulinemia), pseudohypertension, office hypertension, and volume overload. When these problems are eliminated, causes of secondary hypertension should be sought, the most common being coarctation of the aorta, Cushing's syndrome, primary aldosteronism, pheochromocytoma, renovascular disease, thyroid and parathyroid disease, and renal parenchymal disease. In a few cases, a careful hemodynamic and neurohumoral assessment is needed to direct treatment. When the recommended stepwise workup is followed, hypertension that is truly refractory to treatment is a rare finding.

References

1. The Joint National Committee. The sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med 1997;157(21):2413-46
2. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 1991;265(24):3255-64
3. Staessen JA, Fagard R, Thijs L, et al. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet 1997;350(9080):757-64
4. Alderman MH, Budner N, Cohen H, et al. Prevalence of drug resistant hypertension. Hypertension 1988;11(3 Pt 2):71-5
5. Swales JD, Bing RF, Heagerty A, et al. Treatment of refractory hypertension. Lancet 1982;1(8277):894-6
6. Yakovlevitch M, Black HR. Resistant hypertension in a tertiary care clinic. Arch Intern Med 1991;151(9):1786-92
7. Gifford RW Jr. An algorithm for the management of resistant hypertension. Hypertension 1988;11(3 Pt 2):101-5
8. MacFarlane LL, Orak DJ, Simpson WM. NSAIDs, antihypertensive agents and loss of blood pressure control. Am Fam Physician 1995;51(4):849-56
9. Bravo EL. Phenylpropanolamine and other over-the-counter vasoactive compounds. Hypertension 1988;11(3 Pt 2):7-10
10. Isaksson H, Cederholm T, Jansson E, et al. Therapy-resistant hypertension associated with central obesity, insulin resistance, and large muscle fibre area. Blood Press 1993;2(1):46-52
11. Hoffstein V, Chan CK, Slutsky AS. Sleep apnea and systemic hypertension: a causal association review. Am J Med 1991;91(2):190-6
12. Zuschke CA, Pettyjohn FS. Pseudohypertension. South Med J 1995;88(12):1185-90
13. Pickering T, for an American Society of Hypertension ad hoc panel. Recommendations for the use of home (self) and ambulatory blood pressure monitoring. Am J Hypertens 1996;9(1):1-11
14. Tarazi RC, Dustan HP, Frohlich ED, et al. Plasma volume and chronic hypertension: relationship to arterial pressure levels in different hypertensive diseases. Arch Intern Med 1970;125(5):835-42
15. Kaye TB, Crapo L. The Cushing syndrome: an update on diagnostic tests. Ann Intern Med 1990;112(6):434-44
16. Bravo EL. Management of hypercortisolism and hyperaldosteronism. In: Izzo JL Jr, Black HR, eds. Hypertension primer: the essentials of high blood pressure. 2d ed. Baltimore: Williams & Wilkins, 1999:449-53
17. National High Blood Pressure Education Program Working Group. 1995 update of the working group reports on chronic renal failure and renovascular hypertension. Arch Intern Med 1996;156(17):1938-47
18. Moore MA, Porush JG. Hypertension and renal insufficiency: recognition and management. Am Fam Physician 1992;45(3):1248-56
19. Moser M, Prisant LM. Low-dose combination therapy in hypertension. (Editorial) Am Fam Physician 1997;56(5):1275-82

Presented at a meeting of the Interstate Postgraduate Medical Association, Las Vegas.

Dr Vidt is a consultant, Cleveland Clinic Foundation, and professor of medicine, Ohio State University College of Medicine and Public Health, Columbus. Correspondence: Donald G. Vidt, MD, Department of Nephrology and Hypertension, Desk A-101, Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195. E-mail: vidtd@ccf.org.

Please send technical questions related to the Web site to Ann Harste