Managing Calcium Channel Blocker‐Related Peripheral Edema

A 69‐year‐old woman was referred to the hypertension clinic with refractory stage 2 hypertension. She had been taking a diuretic for 20 years. Mild pedal edema occurred a year previously but had gotten worse following prescription of nifedipine 2 months before the initial hypertension clinic visit. The patient denied exertional dyspnea or paroxysmal nocturnal dyspnea. She complained of dry mouth and fatigue. Past medical history included transient ischemic attacks with a normal brain computed tomography scan and echocardiogram, hemiparkinsonism with a right‐sided tremor, and hypothyroidism. Both of the patient's parents, her two brothers, and her two sisters had hypertension. Medications at the time of her first hypertension clinic visit were levothyroxine 0.075 mg q.d., pramipexole dihydrochloride 1.5 mg t.i.d., clonidine 0.3 mg t.i.d., clopidogrel 75 mg q.d., losartan 50 mg b.i.d., amantadine 100 mg b.i.d., nifedipine extended release 30 mg q.d., and furosemide 40 mg q.d. Blood pressure (BP) was 184/92 mm Hg in both arms supine and 168/94 mm Hg in the left arm standing. Creatinine was 1.2 mg/dL (normal range 0.7–1.3 mg/dL), potassium 4.0 mEq/L (normal range 3.5–5.0 mEq/L), thyroid‐stimulating hormone 3.3 IU/mL (normal range 0.4–4.0 IU/mL), and urinalysis was negative for protein. Twenty‐four‐hour urine collections for vanilmandelic acid, catecholamine, and metanephrine were normal.

Initial hypertension clinic plans were to switch the furosemide to chlorthalidone 25 mg q.d. and to wean her from clonidine. Fatigue and dry mouth resolved, but pedal edema continued to be bothersome; a follow‐up BP was 144/94 mm Hg. She refused to try a β blocker due to fear of alopecia. Nifedipine was changed to diltiazem extended release 120 mg q.d., and when seen in the hypertension clinic about 6 weeks later, her pedal edema had resolved and BP was 132/82 mm Hg.

CASE TWO

A 57‐year‐old man was referred to the hypertension clinic with refractory stage 2 hypertension. Four years previously, when hypertension was diagnosed, atenolol 50 mg q.d. had been initiated followed by the addition of hydrochlorothiazide 25 mg q.d. and lisinopril 40 mg q.d. Systolic BP had been running in the 150–160+ mm Hg range and in the prior few months had gone up to the 170+ mm Hg range. He had a childhood right‐sided head injury that he blamed for chronic right‐sided headaches responsive to ibuprofen 200 mg q.d., and he complained of new posterior cervical headaches that he attributed to elevated BP. Creatinine was 0.9 mg/dL, and an echocardiogram revealed mild left ventricular hypertrophy.

Hypertension clinic BP was 176/84 mm Hg supine and 174/88 mm Hg standing; heart rate was 64 bpm. His posterior cervical headaches were attributed to cervical osteoarthritis. Nifedipine extended release 30 mg q.d. was added as a fourth antihypertensive drug, and 6 weeks later his BP had improved to 148/84 mm Hg. A walking program was advised, but another 6 weeks later his BP was 144/74 mm Hg and 1+ pedal edema was noted on exam. With the patient expressing belief that his edema predated the nifedipine, the dose was increased to 60 mg q.d. Edema worsened and the patient stopped the nifedipine on his own followed by nearly complete resolution of edema. Felodipine 5 mg q.d. was initiated in addition to the hydrochlorothiazide, lisinopril, and atenolol. Eight weeks later the patient's BP had improved to 134/88 mm Hg, and he maintained a heart rate of 68 bpm without edema.

DISCUSSION

Calcium channel blockers (CCBs) are potent agents and may be useful additions to the regimen of patients with resistant hypertension, as they were in the two cases presented here. The most common class‐specific side effect responsible for intolerance and patient non‐adherence to therapy is peripheral edema. Peripheral edema is more likely to occur with dihydropyridine CCBs. One study noted 272/2000 (13.6%) patients reporting edema with CCB monotherapy, slightly but not significantly more common in women than men (15.6% vs. 11.8%), ¹ with the varying prevalence according to the specific agent (Table). Two major hypertension trials involving CCBs have reported edema rates: the second Swedish Trial in Old Patients with Hypertension (STOP‐2) noted that 25.5% of patients treated with either felodipine or isradipine had edema; ⁵ and Intervention as a Goal in Hypertension Treatment (INSIGHT) study patients reported a 28% incidence of edema while taking nifedipine long‐acting GastroIntestinal Transport System (GITS), 8% of whom discontinued that drug due to this side effect. ⁶ Additionally, CCB edema rates are both dose and duration related. In the Leonetti et al. study, ³ increasing edema incidence was noted past 3 months of same‐dose administration of a CCB.

CCB edema is unlike fluid retention, which occurs with the nonspecific vasodilators minoxidil and hydralazine and is not diuretic responsive. CCBs are natriuretic, a property that has been attributed to a direct tubular effect. ⁷ A water displacement study demonstrated simultaneously increased foot volume and sodium excretion with nifedipine. ⁸ The CCB edema mechanism is attributed to direct precapillary arteriolar dilatation leading to increased hydrostatic pressure and fluid shift into the interstitial compartment of gravitationally dependent sites. ⁹ This mechanism is described as creating a hypertensive state at the arteriocapillary connection in the absence of compensatory postcapillary venous dilation resulting in a leak syndrome. ¹⁰ This pathophysiology accounts for the observation of dependent edema, which is often asymmetrical due to variable local arteriolar and venous anatomic features in older hypertensive populations. CCB effects on lymphatic return or arteriolar vascular permeability may play lesser roles. ⁹ , ¹⁰

Both patients described preexisting dependent edema, which worsened with nifedipine, and in case 2 was not initially thought to be a drug effect. Whereas the spectrum of non‐drug‐related dependent edema in older persons due to venous insufficiency is most often mild, CCB‐related edema may be more severe and painful. CCB‐associated peripheral edema may also be warm, erythematous, and petechial as a result of red blood cell transudation. ¹¹ The severity may be more pronounced in patients with preexisting venous insufficiency, but there has not as yet been a prospective study looking at venous insufficiency as a risk factor for CCB‐related edema. Edema on CCB agents also merits consideration of alternative etiologies such as heart, renal, or hepatic failure, sleep apnea, ⁹ or concurrent use of nonsteroidal antiinflammatory agents. Most often a clinical rather than a test‐driven approach suffices to obviate these conditions.

Treatment options for CCB‐related peripheral edema include reducing the dose; switching preferably to a nondihydropyridine CCB, or to another dihydropyridine CCB; or providing a venodilator drug to reduce the venous hypertension that characterizes the phenomenon. ⁹ The incidence of CCB edema is dose related and a dose decrease may lead to patient tolerability. As indicated in the Table, variable edema rates for individual CCBs merit an attempt with an alternative agent in a patient with refractory hypertension. Third‐generation lipophilic CCBs (lercanidipine, lacidipine, manidipine) are marked by their cell membrane domain with sustained site of action access and long receptor half‐life. ¹² An associated dilatory effect upon efferent arterioles appears to explain the reduced incidence of edema compared with first‐ and second‐generation CCBs. The Leonetti et al. study ³ specifically examined comparative CCB tolerability and looked at examiner‐verified edema in addition to less reproducible patient‐reported complaints (Figure 1).

Figure 1.

Incidence of peripheral edema after 1, 2, 3, and 6 months of treatment and at study end (N=828). Adapted from Am J Hypertens. 2002;15:932–940. ⁴

Venodilator drugs include angiotensin‐converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and nitrates. Reports indicate that concomitant ACE inhibitor use can ameliorate CCB‐related edema. ⁹ Most of the time the edema is reduced rather than resolved. Occasionally, an ACE inhibitor‐associated BP decrease may permit dose reduction of the CCB and therefore improve edema in this fashion. One report stated that the incidence of CCB edema was “improved in 85% of patients experiencing the problem.” ¹³ The Safety of Lotrel vs. Amlodipine in a Comparative Efficacy (SOLACE) trial showed that patients randomized to the combination tablet Lotrel (Novartis Pharmaceuticals Corp., East Hanover, NJ) containing both the ACE inhibitor benazepril and amlodipine experienced peripheral edema significantly less frequently than those randomized to amlodipine alone ( 12.6% vs. 23.0%, p=0.01). ¹⁴ ARBs and nitrates have not been as well studied in this area.

Both patients presented here had resistant hypertension despite taking four antihypertensive drugs and were referred to a hypertension clinic. Nifedipine‐related dependent edema occurred despite an ACE inhibitor or an ARB in addition to a diuretic leading to patient intolerance and discontinuation. The second patient stopped nifedipine on his own without notifying a physician. Both patients also observed a gradual worsening of prenifedipine venous insufficiency edema becoming more prominent over several months at the same nifedipine dosage (case 1) or following a dose increase (case 2). Exchanging the nifedipine for extended‐release diltiazem (case 1) or felodipine (case 2) led to patient tolerability and BP control. Learning to be a CCB jockey may occasionally enhance the management of difficult hypertension.