DRUG THERAPY: SPECIAL CONSIDERATIONS IN DIALYSIS PATIENTS

Antihypertensive Medications in End-Stage Renal Disease Matthew G. Denker and Debbie L. Cohen Renal, Electrolyte and Hypertension Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

ABSTRACT Hypertension is almost universal in end-stage renal disease (ESRD) and contributes to the substantial cardiovascular (CV) morbidity and mortality observed in these patients. The management of blood pressure (BP) in ESRD is complicated by a number of factors, including missed dialysis treatments, intradialytic changes in BP, medication removal with dialysis, and poor correlation of BPs obtained in the dialysis unit with those at home and with CV outcomes. Control of extracellular volume with ultrafiltration and dietary sodium restriction represents the principal strategy to manage hypertension in ESRD, and antihypertensive medications are subsequently added if this strategy is inadequate. While reduction in BP with medication improves CV outcomes, few head-to-head clinical trials have been performed to firmly establish the superiority of one antihypertensive medication class over another. Therefore, individualization of therapy is neces-

sary, and patient comorbidities must be considered. Angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and beta-blockers are reasonable first-line agents for most patients. ACE inhibitors and ARBs exert cardioprotective effects that are independent of BP reduction. Medications that are removed with dialysis may be preferred in patients who are prone to develop intradialytic hypotension. Intradialytic hypertension can be managed with challenging the patient’s dry weight and using nondialyzable medications. Within a class of antihypertensive medications, there may be large variability in drug removal with dialysis, which must be considered upon medication selection. Studies demonstrate that even thrice-weekly dosing of medication after dialysis has robust BP-lowering effects, which may be a useful regimen in nonadherent patients.

As kidney disease advances to end-stage renal disease (ESRD), hypertension becomes more problematic, affecting almost 90% of patients requiring chronic hemodialysis (HD) (1). Hypertension in ESRD involves diverse pathways, including perturbations in sodium and water homeostasis, sympathetic nervous system (SNS) activity, and the renin–angiotensin–aldosterone system (RAAS). As in the general population, hypertension contributes to cardiovascular (CV) morbidity and mortality—the most common identifiable cause of death among patients with ESRD (2). Increased blood pressure (BP) is associated with the subsequent development of cardiac structural abnormalities and congestive heart failure (CHF) in dialysis patients (3), which at least partially contribute to an increased risk of death in these patients (4). These risks underscore the importance of hypertension management with volume control

through ultrafiltration (UF) and the use of antihypertensive medications. Treatment of hypertension leads to regression of left ventricular hypertrophy (LVH) (5) and reduction in CV morbidity and mortality (6). However, practitioners are presented with a number of challenges in treating this particular patient population including missed dialysis treatments, excessive interdialytic weight gain (IDWG), medication nonadherence, medication clearance with dialysis, intradialytic BP changes, and which BP (i.e., predialysis, postdialysis, home) to use as the appropriate target of therapy. Herein, we address some these challenges and provide recommendations for approaching the treatment of hypertension in patients with ESRD. BP Target Based on expert opinion and weak evidence, the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF KDOQI) guidelines suggest that pre-HD and post-HD BP should be less than 140/90 and 130/80 mmHg, respectively (7). Patients requiring in-center HD have their BP measured before, numerous times during, and after each treatment. However, multiple studies demon-

Address correspondence to: Matthew Denker, MD, Renal, Electrolyte and Hypertension Division, Hospital of the University of Pennsylvania, 1 Founders Building, 3400 Spruce Street, Philadelphia, PA 19104, Tel.: +215-662-7934, Fax: 215-6150349, or e-mail: [email protected]. Conflicts of interest: None. Seminars in Dialysis—2015 DOI: 10.1111/sdi.12369 © 2015 Wiley Periodicals, Inc. 1

2

Denker and Cohen

strate that pre- and post-HD BP measurements are poorly correlated with BPs obtained in the interdialytic period, thus complicating hypertension management (8,9). In addition, home BPs correlate better with CV outcomes and thus have more robust prognostic value (10). Inter-individual variability in the amount of UF performed, rate of vascular refilling, and cardiac function further obscures the interpretation of BP readings in the dialysis center. Therefore, when considering antihypertensive therapy in these patients, we recommend also obtaining home BPs (i.e., ambulatory BP monitoring or patient self-measurement of BP) as they represent a more accurate depiction of the patient’s overall BP burden. Though not widely available and impractical to perform in the majority of patients with ESRD, ambulatory BP monitoring has the added benefit of revealing the patient’s dipping pattern during sleep. Role of Extracellular Volume Control in the Management of Hypertension As kidney function declines, patients become more salt sensitive, such that a given increase in dietary sodium intake results in a greater rise in BP (11). With ESRD comes the loss of excretory capacity of the kidneys, and UF represents the mainstay of BP management. The overwhelming majority of patients requiring HD in the United States undergo treatments thrice weekly for 3–4 hours per treatment, but accomplishing extracellular volume (ECV) control with these time constraints is challenging. The dialysis experience in Tassin, France (12) and the evolving trends of dialysis treatment time in the United States since the 1960s perhaps best demonstrate this phenomenon. Short dialysis treatments lead to higher UF rates, possibly exceeding vascular refilling rate (i.e., the rate at which fluid is mobilized from the interstitium). Potential consequences include intradialytic hypotension, cramping, and increased postdialysis fatigue and recovery time. High UF rates are also thought to be a risk factor for myocardial stunning and may contribute to the considerable CV risk observed in these patients (13). Longer treatment time allows for lower UF rates, leading to fewer hypotensive events; also, treatment time is inversely correlated with predialysis BP (14). In Tassin, 8hour thrice-weekly nocturnal HD treatments are considered standard therapy. These treatments, in combination with dietary sodium restriction, result in BP control in almost all patients without the need for antihypertensive medication (15). Further, the effectiveness of reducing a patient’s dry weight in lowering BP often exceeds the BP reduction seen with antihypertensive medication (16). Numerous studies in both HD and peritoneal dialysis (PD) patients reinforce the beneficial impact of sodium restriction on BP and IDWG (17–19).

We advocate utilizing longer dialysis duration (i.e., nocturnal HD or at least 4 hours of thriceweekly HD) and avoiding sodium modeling as it increases patient thirst and exacerbates IDWG. During the initial probing for a patient’s dry weight, antihypertensive medications may need to be withdrawn to avoid hypotension, and lack of edema should not preclude further challenging of a hypertensive patient’s dry weight. Regardless of the volume of fluid removed during a standard dialysis treatment, the ECV will increase during the interdialytic period and lead to a rise in BP, which usually necessitates the use of antihypertensive medications. Despite these measures, hypertension control, based on the aforementioned BP targets, remains poor in ESRD (1). Reduction in BP with Antihypertensive Medications and Associations with CV Outcomes In a 2009 meta-analysis of eight randomized controlled trials (RCTs) performed in dialysis patients, Heerspink et al. reported that active treatment to lower BP was associated with statistically significant reductions in CV events, CV mortality, and allcause mortality (6). Three of these trials assessed the effects of an angiotensin receptor blocker (ARB), two angiotensin-converting enzyme (ACE) inhibitors, two beta-blockers, and one calcium channel blocker (CCB). There was great heterogeneity in inclusion criteria among the studies. Some trials included patients with cardiomyopathy or LVH, while others did not. One trial was performed in PD patients. Only two study protocols targeted a specific BP goal. We are unable to distinguish between the beneficial effects attributed to BP reduction and those derived from a particular medication class. As in the general hypertensive population, compelling indications (e.g., cardiomyopathy, arrhythmia) should be considered when selecting an antihypertensive agent. Aside from these indications, the Kidney Disease: Improving Global Outcomes 5D report states that “there is no compelling evidence to recommend one class of antihypertensive agents over another” (20). However, it is important to consider medication cost, adherence issues, drug removal with HD (Table 1) (removal with PD is usually not a concern), and simplification of medication regimens as these patients are already overburdened by medication loads. The following sections are devoted to a discussion of individual drug classes in ESRD. ACE Inhibitors and ARBs Renin–angiotensin–aldosterone system activation is common in ESRD, and angiotensin II and aldosterone negatively impact endothelial function and cardiac fibrosis (23,24). Hence, blockade of

3

ANTIHYPERTENSIVE MEDICATIONS IN ESRD

TABLE 1. Antihypertensive medications and extent of drug removal with hemodialysis (7,20–22) Medication a-blockers Doxazosin Prazosin Terazosin ACE-Is Benazepril Enalapril Fosinopril Lisinopril Perindopril Quinapril Ramipril Trandolapril ARBs Candesartan Irbesartan Losartan Olmesartan Telmisartan Valsartan b-blockers Atenolol Carvedilol Labetalol Metoprolol Nadolol Propranolol CCBs Amlodipine Diltiazem CD Felodipine Nifedipine XL Verapamil Diuretics Bumetanide Ethacrynic acid Furosemide Metolazone MRAs Eplerenone Spironolactone Others Clonidine (oral) Hydralazine Minoxidil

Usual dose range

Removal with hemodialysis (%)

1–16 mg daily 1–10 mg twice daily 1–20 mg daily

None None None

5–40 mg daily 2.5–10 mg q12 hours 10–80 mg daily 2.5–40 mg q24–48 hours 2–8 mg daily 10–20 mg daily 2.5–10 mg daily 0.5–4 mg daily

20–50 35–50 None 50 50 25 20–30 30

4–32 mg daily 75–300 mg daily 50–100 mg daily 10–40 mg daily 40–80 mg daily 80–320 mg daily

None None None None None None

25–100 mg q24–48 hours 6.25–25 mg twice daily 100–1200 mg twice daily 50–200 mg twice daily 40–240 mg daily 40–160 mg twice daily

50–75 None None High 50 None

2.5–10 mg daily 180–360 mg daily 2.5–10 mg daily 30–120 mg daily 180–360 mg daily

None None None None None

0.5–2 mg twice daily 50–200 mg twice daily 40–160 mg twice daily 5–10 mg daily

None None None None

50–100 mg daily 25–100 mg daily

None None

0.1–0.4 mg thrice daily 10–100 mg thrice daily 2.5–30 mg daily

5 None Yes

ACE-Is, angiotensin-converting enzyme inhibitors; ARBs, angiotensin receptor blockers; CCBs, calcium channel blockers; MRAs, mineralocorticoid receptor antagonists.

this pathway represents a logical therapeutic target. In the aforementioned meta-analysis by Heerspink et al., there was no statistically significant risk reduction observed in the active treatment arm in both studies that used an ACE inhibitor. In the larger study, the Fosinopril in Dialysis, FOSIDIAL, study enrolled 397 patients with LVH but was underpowered due to an event rate that was lower than anticipated (25); and 40% of patients were normotensive at baseline. Fosinopril is unique among ACE inhibitors in that the medication is not dialyzable. However, in a small RCT, candesartan significantly reduced the risk of CV events despite similar follow-up BP between the trial arms (26). As B-type natriuretic peptide levels rose in the control arm but remained unchanged in the candesartan arm, the reduction in CV risk with the use of an

ARB may be attributed to effects on cardiac remodeling. Additional studies have shown benefits of ACE inhibitors or ARBs on cardiac remodeling and vascular stiffness. London et al. reported that patients randomized to perindopril demonstrated greater reductions in LV mass vs. those randomized to the CCB nitrendipine, despite similar BP reductions (27). In a study by Shibasaki et al., 6 months of losartan resulted in greater LV mass reduction than that seen with enalapril or amlodipine (28); and Ichihara et al. showed that use of an ACE inhibitor or ARB improved pulse wave velocity (29). Further, the reported benefits of ACE inhibitors and ARBs on cardiac structure and vascular stiffness occurred in the absence of BP differences between groups, thus suggesting drug- or class-specific effects of these agents rather than the effects of BP-lowering.

4

Denker and Cohen

Interestingly, in a double-blind, placebo-controlled crossover trial, the ACE inhibitor enalapril reduced thirst and IDWG compared to placebo (30). The mechanism for this effect may relate to a reduction in angiotensin II, a hormone that stimulates thirst. ACE inhibitors and ARBs are generally welltolerated, without a significant rise in serum potassium. Moreover, they confer cardioprotective benefits independent of BP-lowering and thus, are reasonable first-line antihypertensive agents. With the exception of fosinopril, ACE inhibitors are removed with HD, which may be useful in patients with intradialytic hypotension; but ARBs are preferred in those with intradialytic hypertension. Combination therapy with ACE inhibitor plus ARB was compared to each of these classes as monotherapy in a small RCT that enrolled 33 HD patients with diabetes and LVH (31). Other drug classes were added as needed to achieve goal BP