Resistant Hypertension Responding to Change From Furosemide to Thiazide: Understanding Calcium Channel Blocker–Related Edema

A 73-year-old man was referred to a hypertension clinic for hypertension resistant to three drugs, including a diuretic. Two years previously, hydrochlorothiazide (HCTZ) 25 mg increased to 50 mg was initiated and then followed by amlodipine 5 mg, then 10 mg daily. Due to the development of dependent peripheral edema, HCTZ was changed to furosemide 20 mg twice a day, then 40 mg twice a day. According to the patient, there was no appreciable change in the edema. Metoprolol 50 mg twice a day and losartan 50 mg daily were added prior to referral. At the time of the hypertension clinic evaluation, the patient’s blood pressures (BPs) were 146/82 mm Hg and 144/84 mm Hg and physical examination was remarkable for 1+ bilateral ankle and bipedal edema. Serum creatinine was 1.2 mg/dL (normal 0.6–1.2 mg/dL) with an estimated glomerular filtration rate (GMR) of 56 cc/min (normal >90 cc/min). Furosemide was changed to chlorthalidone 25 mg daily with unchanged mild edema, and follow-up BPs were 134/80 mm Hg and 132/82 mm Hg. The patient was assured that his mild edema was not medically detrimental, and daytime compression stockings and leg elevation were advised as needed. The common experience of two university referral clinics for resistant hypertension has been that a chance to select a diuretic appropriate for renal function can occasionally lead to control.1,2 In 50% of patients who responded to a medication change, initiation of a diuretic, advancing the diuretic dose, or change to a diuretic appropriate for renal function was highly successful.1 The Yale University Hypertension Clinic used a GFR <30 cc/min as a cutoff for use of a loop diuretic. Two patients were changed from a loop diuretic to a thiazide for GFR >30 cc/min, resulting in BP control.3 At least 11 small prospective randomized trials have compared antihypertensive efficacy of furosemide and thiazide in patients with “normal” renal function.4–9 When renal function was recorded in these studies conducted in the 1970s and 1980s, maximal serum creatinine was 1.3 mg/dL. In all six studies where HCTZ 50 mg once or twice a day or chlorthalidone 25 mg was compared with furosemide 40 mg twice a day, there was significantly greater BP reduction with thiazide.4,9 In studies where the comparison dose of HCTZ was <50 mg, there was no significant difference in antihypertensive efficacy compared with twice-a-day furosemide.4,6–8 Therefore, a summary of the diuretic comparison studies supports the experience of the university hypertension clinics: thiazide is a more effective antihypertensive agent than twice-a-day furosemide when HCTZ is used at a dose of at least 50 mg daily or chlorthalidone 25 mg is prescribed for patients with normal or near-normal renal function. Commentary and analysis following the Avoiding Cardiovascular Events in Patients Living With Systolic Hypertension (ACCOMPLISH) and the Second Australian National Blood Pressure Study (ANBP2) trials also argue that HCTZ 50 mg or chlorthalidone 25 mg is necessary because these represent the doses used in the successful thiazide trials,10,11 in addition to BP control. Treatment trials using less than 50-mg HCTZ have not consistently demonstrated cardiovascular disease end point benefit vs comparative antihypertensive agents. Because occult volume expansion can be a cause of resistant hypertension even in patients with normal renal function, a higher dose of thiazide is often necessary to overcome resistance.12 In a Veteran’s Administration Hospital study, 45% of HCTZ responders were controlled with 12.5-mg HCTZ, 68% were controlled with 25 mg, and the rest responded to 50 mg daily.13 The type of thiazide may also be important. There is expert opinion supporting the view that chlorthalidone 25 mg may be more effective than HCTZ 50 mg in resistant cases.12 As noted, small prospective randomized trials support higher-dose thiazide with a serum creatinine up to 1.3 mg/dL, corresponding to a GFR greater than approximately 50 cc/min. Between GFRs of 30 cc/min and 50 cc/min, the renal function cutoff for use of thiazide vs loop diuretic in treating resistant hypertension is subject to expert opinion. Some experts recommend use of a loop diuretic over thiazide when the GFR is <30 to 40 cc/min,14 some suggest a threshold of 40 cc/min,15,16 and others suggest a threshold of 30 to 50 cc/min or a serum creatinine >1.5 mg/dL.17 The Yale University hypertension referral clinic used 30 cc/min as a cutoff,3 and that is also the cutoff recommended by the Kidney Disease Outcomes Quality Initiative (DOQI).18 While thiazide therapy is considered generally ineffective when the GFR is <20 to 30 cc/min,4,19 at least one randomized controlled study showed no difference in antihypertensive efficacy comparing HCTZ 25 mg with long-acting furosemide 60 mg in patients with stage 4 or 5 chronic kidney disease.16 Probably the dose of furosemide was too low in this trial of 7 patients with GFRs ranging from 12 cc/min to 41 cc/min. Therefore, there is fair evidence, based on the unanimous conclusions of six small prospective randomized trials, to recommend HCTZ 50 mg or chlorthalidone 25 mg rather than furosemide to treat resistant hypertension when the GFR is >50 cc/min due to improved efficacy (Table I). There is expert opinion and hypertension clinic experience supporting thiazide over furosemide in patients with GFR 30 cc/min to 50 cc/min. For patients with a GFR <30 cc/min, furosemide is recommended. When furosemide is prescribed, the dose frequency needs to be twice daily rather than once daily to overcome a compensatory antinatriuretic effect, which would otherwise occur following the six-hour duration of action.14 More gradual and sustained diuresis with a thiazide diuretic induces less of a compensatory counter-regulatory response or “breaking phenomenon” than furosemide.14,20 Counter-regulation occurs via renin-angiotensin-aldosterone system activation, increasing angiotensin II, as well as sympathetic nervous system activation, which promotes increased proximal renal tubular sodium absorption and reduced delivery of sodium to loop of Henle and distal convoluted tubule areas, representing the major loop diuretic and thiazide action sites, respectively. More abrupt and transient loop diuretic diuresis, lasting only six hours, is followed by an antinatriuresis-breaking interval where counter-regulation predominates. A more even and persistent thiazide diuresis also leads to a secondary vasodilator effect, which is lacking with furosemide.20 At a GFR <30 cc/min, increased plasma volume becomes a factor leading to resistant hypertension and furosemide is more effective than thiazide, as long as furosemide is prescribed twice daily. Calcium channel blocker–related edema is not due to excess intravascular volume and therefore is not responsive to diuretics.21 Although peripheral edema formation is a class effect, whether an individual patient will experience edema is due to a combination of drug and inherent patient factors. Drug factors include the strength of arteriolar vasodilatation, post–capillary venular dilatation as a result of other drug influence, duration of therapy, sex, age, and recumbency. Nondrug causes of peripheral edema, such as right heart failure due to obstructive sleep apnea, need to be considered. Natriuresis with calcium channel blockers has been demonstrated simultaneously with edema development, shown by volume displacement of edematous feet in tanks of water,22 and may cause nocturia and urinary incontinence in the elderly.21 Rather than plasma volume retention, the pathophysiology of calcium channel blocker–related edema has to do with selective precapillary arteriolar dilatation. That effect causes increased intracapillary pressure, inducing leakage of fluid into dependent soft tissue, which results in edema formation. Dihydropyridine calcium channel blockers such as amlodipine, nifedipine, and felodipine have greater arteriolar dilatation potency and are more likely to cause edema than nondihydropyridine calcium blockers such as diltiazem and verapamil.21 Third-generation calcium blockers not currently available in the United States, such as lercanidipine, lacidipine, and manidipine, may be less likely to cause edema because of more balanced pre– and post–capillary dilatation.23 Concomitant use of angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) have been reported to reduce calcium channel blocker edema due to post–capillary venular relaxation reducing intracapillary pressure and capillary leakage21 (Figure). Another explanation for reduced calcium blocker edema with ACE inhibitors and ARBs is that the calcium blocker dose can be reduced. Table II lists drugs that cause peripheral edema. Scheme of capillary circuit illustrating primary vasodilatory action of calcium blockers on precapillary arteriolar sites and primary vasodilatory action of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) on postcapillary venular sites. ACEI/ARB action may reduce intracapillary pressure and edema, causing fluid leakage by unopposed calcium blockers, in addition to permitting reduction in calcium blocker dose. Patient factors are important because individuals who have edema with one agent may tolerate an alternative calcium blocker, particularly with a switch from a dihydropyridine to a nondihydropyridine calcium blocker. Increasing age and female gender as well as heat and prolonged upright stature increase edema risk. Interestingly, preexisting edema status such as venous insufficiency has not been reported to increase the risk of calcium channel blocker edema. Contrary to the treatment of edema with minoxidil, hydralazine, and methyldopa, which cause volume retention and are diuretic responsive, how is calcium channel blocker edema managed? Use of ACE inhibitors and ARBs may or may not be beneficial, but calcium blocker dose reduction and switching to another calcium blocker can be helpful. For mild to moderate edema, daytime compression stockings and leg elevation can be helpful. Many times, reassurance that this edema is not medically threatening is all that is necessary for continuation of therapy. Table III describes management options for calcium blocker–related edema. Misunderstanding of the mechanism of calcium channel blocker–related edema led to use of less-effective diuretic therapy in this patient with mild renal impairment, which was then the cause of uncontrolled hypertension. Switching from twice-daily furosemide to chlorthalidone 25 mg and advising daytime compression stockings for mild calcium blocker edema led to control of BP and a good quality-of-life outcome.