●

mtdtpaaps(pso©

Ie

Cprsmtstrrseatc

W

M

o

Mi

9

Effect of Spironolactone on Blood Pressure and theRenin-Angiotensin-Aldosterone System in Oligo-Anuric

Hemodialysis Patients

Evan Gross, MD, Marcos Rothstein, MD, Susan Dombek, RN, and Henrikas Irmantas Juknis, MD

Background: Through its actions on nonepithelial tissues, including brain, blood vessels, and heart, aldosteroneay mediate hypertension, cardiac hypertrophy, and fibrosis. Whether aldosterone has a direct pathogenic role in

he development of cardiovascular complications in patients with end-stage renal disease is unknown. Oligo-anuricialysis patients provide a clinical setting to study the effects of the mineralocorticoid receptor blocker spironolac-one that are independent of the diuretic properties of the drug. We performed a randomized, double-blinded,lacebo-controlled, crossover study to assess the effect of spironolactone on blood pressure and the renin-ngiotensin-aldosterone system in oligo-anuric hemodialysis patients. Methods: Eight hemodialysis patients weredministered either spironolactone, 50 mg, or placebo orally twice daily for 2 weeks, followed by a 3-week washouteriod, after which patients crossed over in their treatment arms for 2 more weeks. Results: Administration ofpironolactone for 2 weeks decreased predialysis systolic blood pressure from 142.0 � 19.6 to 131.4 � 18.2 mm HgP < 0.05). Compared with placebo, a 2-week course of spironolactone had no effect on predialysis and postdialysislasma potassium or aldosterone concentrations or renin activity. Conclusion: When administered for 2 weeks,pironolactone, 50 mg twice daily, reduced predialysis systolic blood pressure, but did not produce hyperkalemia inligo-anuric hemodialysis patients. Am J Kidney Dis 46:94-101.2005 by the National Kidney Foundation, Inc.

NDEX WORDS: Spironolactone; hemodialysis (HD); aldosterone; hypertension; potassium; end-stage renal dis-

ase (ESRD).

at

afpvndhatrpFutssosistdbs

ARDIOVASCULAR complications arecommon in dialysis patients and com-

rise up to 50% of deaths in the end-stageenal disease (ESRD) population.1 Hyperten-ion and left ventricular hypertrophy occur inore than 70% of patients undergoing long-

erm hemodialysis therapy, and both contributeignificantly to mortality and morbidity.2-4 Inhe presence of normal renal function, theenin-angiotensin-aldosterone system (RAAS)egulates blood pressure and sodium and potas-ium balance. Through mechanisms outside itsffects on epithelial cell transport of sodiumnd potassium, aldosterone may mediate hyper-ension and cardiac fibrosis.5-7 Mineralocorti-oid receptors in the brain,8-11 vasculature,12,13

From the Department of Internal Medicine, Renal Division,ashington University School of Medicine, St Louis, MO.Received January 20, 2005; accepted in revised formarch 15, 2005.Originally published online as doi:10.1053/j.ajkd.2005.03.005

n April 29, 2005.Address reprint requests to Henrikas Irmantas Juknis,D, 4205 Forest Park Ave, St Louis, MO 63108. E-mail:

juknevi@im.wustl.edu© 2005 by the National Kidney Foundation, Inc.0272-6386/05/4601-0012$30.00/0

sdoi:10.1053/j.ajkd.2005.03.005

American Journa4

nd heart14 provide possible venues for nonepi-helial actions of aldosterone.

Hyperaldosteronism independent of reninctivity accompanies various stages of renalailure.15,16 Whether aldosterone has a directathogenic role in the development of cardio-ascular complications in the ESRD setting isot known, but it may directly contribute to theevelopment of hypertension, left ventricularypertrophy, and cardiac fibrosis. Oliguric di-lysis patients provide an ideal clinical settingo study the effects of the mineralocorticoideceptor blocker spironolactone that are inde-endent of the diuretic properties of the drug.urthermore, if safe, such drugs may proveseful in treating the cardiovascular complica-ions of patients with ESRD. Although sometudies addressed the issue of the safety ofpironolactone in patients with ESRD,17-19 tour knowledge, none examined the relation-hip between spironolactone and hypertensionn this population. Schohn et al20 showed thathort-term cardiovascular effects of spironolac-one in hemodialysis patients are dose depen-ent. We performed a randomized, double-linded, placebo-controlled, crossover pilottudy to assess the effect of a high dose of

pironolactone (50 mg twice a day) on blood

l of Kidney Diseases, Vol 46, No 1 (July), 2005: pp 94-101

pd

P

pCsmceehppttstasnpp

S

ds5fpmfia

wWmOopdcpppiuhcmweatto

io

S

db3taabpr2ntossttD

S

asaslaSc

cpAts0k(�ici(hs

ts

SPIRONOLACTONE IN HEMODIALYSIS 95

ressure and the RAAS in oligo-anuric hemo-ialysis patients.

METHODS

articipantsWe selected study participants from adult hemodialysis

atients treated thrice weekly at Barnes-Jewish Dialysisenter, St Louis, MO. Men and women qualified for the

tudy if they were on hemodialysis therapy for more than 3onths, had an average predialysis plasma potassium con-

entration less than 6 mEq/L (mmol/L) at the time ofnrollment, were not administered angiotensin-convertingnzyme inhibitors or angiotensin II receptor blockers, andad nil or minimal urine output (�500 mL/24 h). Allarticipating women of childbearing age had a negativeregnancy test result before entering into the study. Addi-ional exclusion criteria were a known allergy to spironolac-one; any acute illness; hypotension, defined as a predialysisystolic blood pressure less than 100 mm Hg; severe hyper-ension (predialysis systolic blood pressure �180 mm Hgnd/or diastolic blood pressure �100 mm Hg); decompen-ated heart failure; inability to give informed consent; andoncompliance. The Institutional Review Board of the De-artment of Medicine approved the study protocol. Allarticipants gave informed consent.

tudy DesignWe enrolled 8 patients in a prospective, randomized,

ouble-blinded, placebo-controlled, crossover study. At studytart, participants were administered either spironolactone,0 mg, or a placebo tablet orally twice daily for 2 weeks,ollowed by a 3-week washout period. After the washouteriod, patients crossed over in their treatment arms for 2ore weeks. Patients administered spironolactone for therst 2 weeks were administered placebo for the last 2 weeks,nd vice versa. All patients served as their own controls.

Participants received outpatient hemodialysis 3 times aeek using Fresenius 2008H machines (Fresenius USA, Inc,alnut Creek, CA). All patients used Fresenius polysulfoneembranes: 7 participants used F7NR and 1 participant usedptiflux 200 (high-flux) dialyzers. To control for the effectsf volume on blood pressure and the RAAS, we kept targetostdialysis weight (dry weight) for each patient constanturing the study. Dialysate sodium concentrations were heldonstant at 140 mEq/L (mmol/L) for all patients. Dialysateotassium concentrations were 2 mEq/L (mmol/L) for 5atients and 3 mEq/L (mmol/L) for 3 patients. No partici-ant was administered potassium-binding resins. We did notmpose additional dietary potassium restrictions beyond thesual recommendations for patients with ESRD treated withemodialysis. We made no changes to dialysate potassiumoncentrations or antihypertensive regimens. Because of aandate not to alter the antihypertensive regimen, targeteight, and dialysis potassium bath during the study, we

xcluded patients with hypotension, severe hypertension,nd hyperkalemia. We excluded patients administered angio-ensin-converting enzyme inhibitors or angiotensin II recep-or blockers to avoid compounding the effects of these drugs

n the RAAS. To evaluate the effect of spironolactone t

ndependent of its diuretic action, we included only oliguricr anuric patients (urine output �500 mL/24 h).

tudy VariablesWe determined predialysis and postdialysis weights, inter-

ialytic weight gains (IDWGs), and systolic and diastoliclood pressures for each patient by calculating an average ofmeasurements obtained during 1 week before the adminis-

ration of any study medication (baseline), during the firstnd second weeks of placebo and spironolactone treatment,nd during the last week of the washout period. At theeginning and end of each treatment period, we measuredredialysis and postdialysis plasma potassium and aldoste-one levels, renin activity (PRA), and, on a nondialysis day,4-hour urine excretion of sodium, potassium, and creati-ine. We monitored predialysis plasma potassium concentra-ions weekly in all subjects. Predialysis blood samples werebtained after cannulation of the vascular access before thetart of the dialysis session. We obtained postdialysis bloodamples from a predialyzer blood sample port at the end ofhe hemodialysis session after slowing the blood pump to 50o 100 mL/min. The Barnes-Jewish Hospital Laboratoryepartment performed all laboratory tests.

tatistical MethodsResults are presented as mean � SD. To compare clinical

nd laboratory parameters, Student t-test was used for pairedamples. Linear regression and correlation were used tossess the relationship between 2 variables. Stepwise regres-ion was used to identify predictors of a single variable. Pess than 0.05 defines statistical significance. We performedll statistical analyses using MINITAB 14.1 (Minitab, Inc,tate College, PA) and Microsoft Excel for Windows (Mi-rosoft Corp, Redmond, WA) software packages.

RESULTS

Table 1 lists baseline patient demographic andlinical characteristics, and Table 2 lists antihy-ertensive regimens for each individual patient.fter dialysis, as expected, we observed a statis-

ically significant decline in systolic blood pres-ure (143 � 20 versus 130 � 22 mm Hg; P �.05), weight (89.6 � 26.1 versus 86.6 � 25.4g; P � 0.05), and potassium concentration4.5 � 0.4 versus 3.6 � 0.5 mEq/L [mmol/L]; P

0.05; Table 1). There was no significant changen diastolic blood pressure or plasma aldosteroneoncentration (P � 0.05) after dialysis. PRAncreased from 1.8 � 1.7 to 6.1 � 6.4 ng/mL/h0.5 � 0.5 to 1.7 � 1.8 ng/L/s) after dialysis;owever, this increase did not reach statisticalignificance (P � 0.05; Table 1).

Table 3 lists weights and blood pressures ofhe placebo and spironolactone groups at thetart and 1 and 2 weeks of therapy. Administra-

ion of spironolactone for 2 weeks decreased

p2Bmdclns

2ocnef(bd

svspapa0P

isgtr

AMS

R

C

A�CVN

P

P

P

P

PPIPPPPPPUUUU

oeden(i0mbu

*P � 0.05 compared with predialysis value.

GROSS ET AL96

redialysis systolic blood pressure from 142 �0 to 131 � 18 mm Hg (P � 0.05; Table 3).ecause predialysis and postdialysis weights re-ained the same throughout the entire study, the

ifference in predialysis systolic blood pressureannot be attributed to a difference in extracellu-ar fluid volume (Table 3). Spironolactone didot affect predialysis diastolic or postdialysisystolic or diastolic blood pressures (Table 3).

As listed in Table 4, compared with placebo, a-week course of spironolactone had no effectsn predialysis and postdialysis plasma potassiumoncentrations, aldosterone levels, PRA, or uri-ary excretion of sodium and potassium. Asxpected, spironolactone had no diuretic effect inunctionally anephric hemodialysis patientsTable 4). Therefore, spironolactone decreasedlood pressure by a mechanism that is indepen-ent of the drug’s diuretic properties.Figure 1 shows mean predialysis plasma potas-

ium concentrations with 95% confidence inter-als in the spironolactone and placebo groups attudy start (spironolactone, 4.6 � 0.7 versuslacebo, 4.7 � 0.4 mEq/L [mmol/L]; P � 0.05)nd after 1 (spironolactone, 4.5 � 0.7 versuslacebo, 4.8 � 0.4 mEq/L [mmol/L]; P � 0.05)nd 2 weeks of treatment (spironolactone, 5.0 �.8 versus placebo, 4.7 � 0.5 mEq/L [mmol/L];� 0.05).At baseline, multivariate stepwise regression

dentified IDWG as a single predictor of predialy-is plasma aldosterone concentration. Linear re-ression between predialysis aldosterone concen-ration and IDWG yielded the following

Table 2. Antihypertensive Medications in IndividualPatients

Patient No. Antihypertensive Medications

1 No medication2 �-Blocker (metoprolol), nondihydropyridine

calcium channel blocker3 No medication4 No medication5 �-Blocker (metoprolol)6 �-Blocker (metoprolol), nondihydropyridine

calcium channel blocker7 �-Blocker (metoprolol), nondihydropyridine

calcium channel blocker, vasodilator(minoxidil)

8 No medication

Table 1. Baseline Patient Characteristics

Patient Characteristics N � 8

ge (y) 53 � 10onths on hemodialysis 20 � 14exMen 3 (37.5)Women 5 (62.5)aceAfrican American 7 (87.5)White 1 (12.5)ause of ESRDHypertension 3 (37.5)Diabetes mellitus 3 (37.5)Bone marrow transplant 1 (12.5)Obstructive uropathy 1 (12.5)

ntihypertensive medications/patient 1.0 � 1.2-Blockers 4 (50.0)alcium channel blockers 3 (37.5)asodilators 1 (12.5)o. of antihypertensives0 4 (50.0)1 1 (12.5)2 2 (25.0)3 1 (12.5)

redialysis systolic blood pressure(mm Hg)

143.3 � 20.1

redialysis diastolic blood pressure(mm Hg)

80.7 � 14.1

ostdialysis systolic blood pressure(mm Hg)

130.4 � 22.5*

ostdialysis diastolic blood pressure(mm Hg)

77.2 � 12.1

redialysis weight (kg) 89.6 � 26.1ostdialysis weight (kg) 86.6 � 25.4*

DWG (kg) 3.0 � 1.3redialysis plasma potassium (mEq/L) 4.5 � 0.4ostdialysis plasma potassium (mEq/L) 3.6 � 0.5*redialysis plasma aldosterone (ng/dL) 29.9 � 50.6ostdialysis plasma aldosterone (ng/dL) 24.1 � 32.3redialysis PRA (ng/mL/h) 1.8 � 1.7ostdialysis PRA (ng/mL/h) 6.1 � 6.4rine volume (mL/24 h) 187.5 � 215.1rine creatinine (mg/24 h) 278.9 � 357.8rine sodium (mEq/24 h) 16.4 � 18.9rine potassium (mEq/24 h) 5.9 � 7.4

NOTE. Values expressed as number of patients (percent)r mean � SD. In patients with normal renal function, refer-nce ranges for plasma aldosterone concentration and PRAepend on sodium intake and position. Aldosterone refer-nce ranges: normal-sodium diet, supine, 1 to 16 ng/dL;ormal-sodium diet, upright, 4 to 31 ng/dL; low-sodium diet�2 g/d), 2- to 5-fold increase greater than normal-sodiumntake values; for PRA: supine, 0.3 to 3.0 ng/mL/h; upright,.4 to 8.8 ng/mL/h. To convert potassium in mEq/L to mmol/L,ultiply by 1; plasma aldosterone in ng/dL to nmol/L, multiplyy 0.02774; PRA in ng/mL/h to ng/L/s, multiply by 0.2778;rine creatinine in mg/24 h to �mol/24 h, multiply by 8.84.

egression equation: predialysis aldosterone �

�artsasccaPd

pl0anrc0ttt4

PPP

P

P

P

P

2

at

SPIRONOLACTONE IN HEMODIALYSIS 97

78.8 � 36.5 IDWG in kg (P � 0.001). Predi-lysis aldosterone concentration closely and di-ectly correlated with IDWG (Pearson correla-ion coefficient, 0.927; P � 0.001). Figure 2hows this relationship. Traditional regulators ofldosterone, including predialysis plasma potas-ium concentration and PRA, did not predict ororrelate with predialysis plasma aldosterone con-entration (Pearson correlation coefficients, 0.492nd �0.343; P � 0.216 and 0.406, respectively).redialysis systolic and diastolic blood pressuresid not predict or correlate with predialysis

Table 3. Predialysis and Postd

Start

Placebo Spironolactone

redialysis weight (kg) 89.3 � 25.6 90.1 � 25.7ostdialysis weight (kg) 86.4 � 25.0 87.0 � 25.1redialysis blood

pressure (mm Hg)Systolic 146.0 � 20.8 142.4 � 30.1Diastolic 79.3 � 15.4 77.5 � 14.8

ostdialysis bloodpressure (mm Hg)

Systolic 126.3 � 27.4 131.3 � 27.4Diastolic 75.8 � 12.3 72.8 � 12.7

NOTE. Values expressed as mean � SD.*P � 0.05.

Table 4. Effect of Spironolactone on Plasma PotassiSodium a

Sta

Placebo

lasma potassium (mEq/L)Predialysis 4.7 � 0.4Postdialysis 3.7 � 0.6

lasma aldosterone (ng/dL)Predialysis 47.1 � 100.2Postdialysis 38.3 � 67.8

RA (ng/mL/h)Predialysis 2.2 � 1.9Postdialysis 5.3 � 6.4

4-Hour urineVolume (mL/24 h) 175.6 � 199.6Urine sodium (mmol/24 h) 15.6 � 18.1Urine potassium (mmol/24 h) 5.1 � 6.0Urine creatinine (mg/24 h) 259.5 � 331.5

NOTE. Values expressed as mean � SD. All P � 0.05.ldosterone in ng/dL to nmol/L, multiply by 0.02774; PRA in

o �mol/24 h, multiply by 8.84.

lasma aldosterone concentration (Pearson corre-ation coefficients, �0.550 and �0.505; P �.158 and 0.202, respectively). Similarly, postdi-lysis systolic and diastolic blood pressures didot predict or correlate with predialysis aldoste-one concentration (Pearson correlation coeffi-ients, �0.454 and �0.570; P � 0.258 and.140, respectively). The same relationship be-ween IDWG and predialysis aldosterone concen-ration persisted after 2 weeks of spironolactoneherapy (predialysis aldosterone � �82.4 �5.3 IDWG; P � 0.014). The regression coeffi-

Weights and Blood Pressures

1 Week 2 Weeks

cebo Spironolactone Placebo Spironolactone

� 25.8 89.6 � 26.3 89.7 � 25.9 89.1 � 25.8� 25.1 86.5 � 25.2 86.4 � 24.8 86.4 � 25.0

� 16.1 134.6 � 23.8 142.0 � 19.6 131.4 � 18.2*� 10.8 75.0 � 15.7 78.7 � 14.2 72.3 � 11.7

� 25.0 133.0 � 20.5 130.6 � 22.9 130.8 � 29.1� 14.1 75.8 � 12.9 75.7 � 12.0 74.3 � 14.6

Aldosterone Levels, PRA, and Urinary Excretion oftassium

2 Weeks

ironolactone Placebo Spironolactone

.6 � 0.7 4.7 � 0.5 5.0 � 0.8

.5 � 0.7 3.5 � 0.6 3.7 � 0.5

.6 � 51.1 47.6 � 98.6 41.5 � 73.9

.6 � 34.9 72.1 � 162.6 57.9 � 128.8

.5 � 1.7 1.6 � 1.4 1.9 � 1.6

.6 � 5.5 7.2 � 9.2 5.9 � 9.5

.3 � 170.0 202.5 � 240.6 161.3 � 231.9

.5 � 13.4 17.7 � 22.6 14.8 � 22.2

.8 � 8.2 5.8 � 7.8 4.4 � 6.8

.8 � 399.1 321.6 � 545.7 250.7 � 411.1

vert potassium in mEq/L to mmol/L, multiply by 1; plasma/h to ng/L/s, multiply by 0.2778; urine creatinine in mg/24 h

ialysis

Pla

89.186.3

143.878.0

129.675.7

um andnd Po

rt

Sp

43

2824

14

156125

262

To conng/mL

cd

ttpsFtalst

sashsnttpsthrsl

dcblfaTn0

GROSS ET AL98

ient after 2 weeks of spironolactone did notiffer from baseline (P � 0.05).

DISCUSSION

Results of our study indicate that administra-ion of spironolactone decreases predialysis sys-olic blood pressure in oligo-anuric hemodialysisatients by a nondiuretic mechanism. There areeveral plausible explanations for this finding.irst, aldosterone mediates hypertension cen-

rally.8-11 In a rat model, intracerebral infusion ofldosterone at doses that do not increase serumevels to greater than normal produces hyperten-ion.8,21 Blockade of mineralocorticoid recep-ors in the brain by means of intracerebral infu-

Fig 2. Scatter plot of pre-ialysis plasma aldosteroneoncentration and IDWG ataseline. X-axis, IDWG in ki-

ograms; Y-axis, log-trans-ormed scale of plasmaldosterone concentration.o convert aldosterone in

g/dL to nmol/L, multiply by.02774.

ion of a selective mineralocorticoid antagonistt doses that are ineffective when administeredubcutaneously inhibits the development of theypertension produced by the subcutaneous infu-ion of aldosterone or deoxycorticosterone inormotensive rats.22 These animal data supporthe notion that aldosterone causes hypertensionhrough its central effects by mechanisms inde-endent of sodium retention. Neuroendocrinetudies of humans showed that brain mineralocor-icoid receptors participate in regulation of theypothalamic-pituitary-adrenal axis and that spi-onolactone can modulate that activity.23,24 Thesetudies provided indirect evidence that spirono-actone can penetrate the blood-brain barrier and

Fig 1. Effects of spirono-lactone on predialysis plasmapotassium concentrations.Bars represent mean predialy-sis plasma potassium con-centrations (mmol/L [equiva-lent to mEq/L]), and rangesrepresent 95% confidence in-tervals. Spiro, spironolactone-treated group. P > 0.05 for allcomparisons between groups.

aocsmcsc

hwcttvcihcasppc

orqsamattck

rc(dsctlpccficl

icTpodadsran

psr(otisbpwrtsidltarpstdcpwo

tnapIpis

SPIRONOLACTONE IN HEMODIALYSIS 99

ct on central mineralocorticoid receptors. Sec-nd, vasculature contains mineralocorticoid re-eptors and a protective enzyme, 11�-hydroxy-teroid dehydrogenase (HSD2), necessary forineralocorticoid responsiveness.12,13 It is not

lear whether aldosterone can mediate hyperten-ion through its action on vascular mineralocorti-oid receptors.

Activation of the sympathetic nervous systemas a distinct role in hypertension of patientsith chronic renal failure,25 and angiotensin-

onverting enzyme inhibition decreases sympa-hetic activity.26 Huang and Leenen27 reportedhat intracerebral infusion of spironolactone pre-ented sympathetic hyperactivity and improvedardiac function in a rat heart failure/myocardialnfarction model. When administered to oliguricemodialysis patients, spironolactone inhibitedardiovascular reactivity to norepinephrine andngiotensin II.20 In our study, it is possible thatpironolactone mediated the decline in bloodressure through its effects on sympathetic out-ut or by modulating vascular reactivity to vaso-onstrictors.

The heart is another main nonepithelial targetf aldosterone. It possesses mineralocorticoideceptors and a protective HSD2 enzyme re-uired for direct aldosterone action.14 In large-cale studies of heart failure, aldosterone block-de had salutary effects on morbidity andortality.28,29 In hemodialysis patients, Sato et

l7 showed a direct correlation between left ven-ricular mass and plasma aldosterone concentra-ion. Whether spironolactone improves cardiovas-ular outcomes in the ESRD population is notnown.In addition to binding to mineralocorticoid

eceptors, spironolactone may interfere with glu-ocorticoid receptor occupancy by its ligandsglucocorticoids or aldosterone when plasma al-osterone levels are high). It is not known whetherpironolactone interferes with glucocorticoid re-eptor occupancy in vivo. Compared with renalubular cells, nonepithelial target tissues haveower levels of the protective HSD2 enzyme,otentially exposing those mineralocorticoid re-eptors to excessive stimulation by glucocorti-oids.30 In addition, enzyme HSD2 may be modi-ed in hemodialysis patients. Decreased renalonversion of cortisol into the inactive metabo-

ite cortisone may enhance the glucocorticoid s

mpact on mineralocorticoid receptors and glu-ocorticoid receptors in nonepithelial tissues.herefore, the decline in blood pressure in theresence of spironolactone may be related notnly to antagonism of central, vascular, or car-iac mineralocorticoid receptors, but also to alter-tions in plasma glucocorticoid levels or therug’s effect on glucocorticoid receptors. Thatpironolactone decreased blood pressure in oligu-ic dialysis patients reinforces the concept thatldosterone mediates hypertension by acting onontraditional target tissues.Results of the study indicate that hemodialysis

atients do not have suppressed predialysis aldo-terone levels. Mean predialysis plasma aldoste-one concentration was 29.9 � 50.6 ng/dL0.83 � 1.4 nmol/L). Additionally, we did notbserve a change in plasma aldosterone concen-ration before and after dialysis. Although PRAncreased after dialysis, it did not reach statisticalignificance, perhaps secondary to the low num-er of patients in our study. Five of 8 studyarticipants were women, and 7 of 8 patientsere African American. It is possible that innate

acial or sex differences could have influencedhe observed effect of dialysis on PRA and aldo-terone concentrations. Other studies showed thatn patients without diabetes, PRA increases withialysis, but aldosterone level does not.31 Theack of change in plasma aldosterone concentra-ions during dialysis in our study suggests anltered relationship between volume and aldoste-one levels in patients with ESRD. Patients withreserved renal function develop increased aldo-terone levels after administration of spironolac-one, perhaps because of the attendant volumeepletion.32,33 Spironolactone produced nohange in aldosterone levels in oliguric dialysisatients, although one should interpret the resultsith caution given the low number of patients inur pilot study.PRA and potassium concentration, the tradi-

ional regulators of plasma aldosterone level, didot predict or correlate with predialysis plasmaldosterone concentration. The only predictor ofredialysis plasma aldosterone concentration wasDWG: The higher the IDWG, the higher theredialysis aldosterone concentration. This find-ng reverses the normal physiological relation-hip between volume and aldosterone level. Our

tudy suggests that volume gain, perhaps in ex-

carmewcs

tddtrlapestHasrcsitsrirE

atporrsh

RK

A1

en

Is

auH

DtdoM

rst

ap1

mt

ShJ

Hah

1c

r6

NMd

aI

rh

Maa

la

M

GROSS ET AL100

ess of normal, directly correlates with plasmaldosterone concentration. Treatment with spi-onolactone did not alter this relationship. Theechanism of this association is not clear. How-

ver, it may tie high interdialytic volume gainith the development of cardiovascular compli-

ations in hemodialysis patients through thetimulation of aldosterone.

Last, when administered for 2 weeks, a rela-ively high dose of spironolactone (50 mg twiceaily) did not produce hyperkalemia. The shorturation of treatment and small number of pa-ients in our pilot study limit the applicability ofesults with regard to long-term risk for hyperka-emia. Michea et al17 showed that spironolactonet a dose of 50 mg 3 times weekly did notroduce hyperkalemia, but blocked aldosteroneffects in nonepithelial tissues. Saudan et al18

howed that low-dose spironolactone (25 mghrice weekly) was safe in hemodialysis patients.uman studies of anatomically anephric hemodi-

lysis patients showed no significant change intool potassium excretion during high-dose spi-onolactone therapy (100 mg orally 3 times/d)ompared with the control group.34 In the sametudy, spironolactone did not affect the degree ofncrease in plasma potassium levels in responseo chronic potassium loading.34 The issue ofpironolactone safety in hemodialysis patientsemains important because future long-term stud-es using aldosterone blockade in an attempt toeduce cardiovascular mortality in patients withSRD are likely to be undertaken.In summary, our pilot study shows that when

dministered for 2 weeks, spironolactone, 50 mgwice daily, reduced predialysis systolic bloodressure, but did not produce hyperkalemia inligo-anuric hemodialysis patients. IDWG di-ectly correlated with predialysis plasma aldoste-one concentration, suggesting an altered relation-hip between aldosterone level and volume inemodialysis patients.

REFERENCES

1. US Renal Data System: USRDS 2001 Annual Dataeport: The incidence and prevalence of ESRD. Am Jidney Dis 38:S17-S36, 2001 (suppl 3)2. Salem MM: Hypertension in hemodialysis population:survey of 649 patients. Am J Kidney Dis 26:461-468,

995 N

3. Foley RN, Parfrey PS, Harnett JD, et al: Clinical andchocardiographic disease in patients starting end-stage re-al disease therapy. Kidney Int 47:186-192, 19954. Silberberg JS, Barre PE, Prichard SS, Sniderman AD:

mpact of left ventricular hypertrophy on survival in end-tage renal disease. Kidney Int 36:286-290, 1989

5. Sato A, Suzuki Y, Saruta T: Effects of spironolactonend angiotensin-converting enzyme inhibitor on left ventric-lar hypertrophy in patients with essential hypertension.ypertens Res 22:17-22, 19996. Frierdich G, Schuh J, Brown M, Nochoviak D, Blehm

, Delyani JA: Effect of selective mineralocorticoid recep-or antagonist, eplerenone, in a model of aldosterone in-uced hypertension and cardiac fibrosis. American Societyf Hypertension 13th Scientific Meeting, New York, NY,ay 13-16, 1998 (abstr)7. Sato A, Funder JW, Saruta T: Involvement of aldoste-

one in left ventricular hypertrophy of patients with end-tage renal failure treated with hemodialysis. Am J Hyper-ens 12:863-873, 1999

8. Peysner K, Henry CA, Malvin RL: Central infusion ofldosterone increases blood pressure by mechanisms inde-endent of Na retention. Clin Exp Hypertens A 12:399-414,9909. De Kloet ER, Van Acker SA, Sibug RM, et al: Brainineralocorticoid receptors and centrally regulated func-

ions. Kidney Int 57:1329-1336, 200010. Young EA, Lopez JF, Murphy-Weinberg V, Watson

J, Akil H: The role of mineralocorticoid receptors inypothalamic-pituitary-adrenal axis regulation in humans.Clin Endocrinol Metab 83:3339-3345, 199811. Deuschle M, Weber B, Colla M, Muller M, Kniest A,

euser I: Mineralocorticoid receptor also modulates basalctivity of hypothalamus-pituitary-adrenocortical system inumans. Neuroendocrinology 68:355-360, 199812. Hatakeyama H, Inaba S, Takeda R, Miyamori I:

1Beta-hydroxysteroid dehydrogenase in human vascularells. Kidney Int 57:1352-1357, 2000

13. Takeda R, Hatakeyama H, Takeda Y, et al: Aldoste-one biosynthesis and action in vascular cells. Steroids0:120-124, 199514. Lombes M, Alfaidy N, Eugene E, Lessana A, Farman

, Bonvalet JP. Prerequisite for cardiac aldosterone action.ineralocorticoid receptor and 11 beta-hydroxysteroid dehy-

rogenase in the human heart. Circulation 92:175-182, 199515. Hene RJ, Boer P, Koomans HA, Mees EJ: Plasma

ldosterone concentrations in chronic renal disease. Kidneynt 21:98-101, 1982

16. Ratge D, Augustin R, Wisser H: Catecholamines,enin, aldosterone and arterial pressure in patients on chronicemodialysis treatment. Int J Artif Organs 6:255-260, 198317. Michea L, Vukusich A, Gonzalez M, Zehnder C,arusic ET: Effect of spironolactone on K(�) homeostasis

nd ENaC expression in lymphocytes from chronic hemodi-lysis patients. Kidney Int 66:1647-1653, 2004

18. Saudan P, Mach F, Perneger T, et al: Safety ofow-dose spironolactone administration in chronic hemodi-lysis patients. Nephrol Dial Transplant 18:2359-2363, 2003

19. Hussain S, Dreyfus DE, Marcus RJ, Biederman RWW,cGill RL: Is spironolactone safe for dialysis patients?

ephrol Dial Transplant 18:2364-2368, 2003

c4

r8

Ir

ShJ

Hah

hS

tw1

ch

AD

ssI

ASra2

M2

prw

lt

Jf

p

SPIRONOLACTONE IN HEMODIALYSIS 101

20. Schohn DC, Jahn HA, Pelletier BC: Dose-relatedardiovascular effects of spironolactone. Am J Cardiol 71:0A-45A, 199321. Gomez-Sanchez EP: Intracerebral infusion of aldoste-

one induces hypertension in rats. Endocrinology 118:819-23, 198622. Gomez-Sanchez EP, Fort CM, Gomez-Sanchez CE:

ntracerebroventricular infusion of RU28318 blocks aldoste-one-salt hypertension. Am J Physiol 258:E482-E484, 1990

23. Young EA, Lopez JF, Murphy-Weinberg V, WatsonJ, Akil H: The role of mineralocorticoid receptors inypothalamic-pituitary-adrenal axis regulation in humans.Clin Endocrinol Metab 83:3339-3345, 199824. Deuschle M, Weber B, Colla M, Muller M, Kniest A,

euser I: Mineralocorticoid receptor also modulates basalctivity of hypothalamus-pituitary-adrenocortical system inumans. Neuroendocrinology 68:355-360, 199825. Koomans HA, Blankestijn PJ, Joles JA: Sympathetic

yperactivity in chronic renal failure: A wake-up call. J Amoc Nephrol 15:524-537, 200426. Ligtenberg G, Blankestijn PJ, Oey PL, et al: Reduc-

ion of sympathetic hyperactivity by enalapril in patientsith chronic renal failure. N Engl J Med 340:1321-1328,99927. Huang BS, Leenen FH: Blockade of brain mineralo-

orticoid receptors or Na� channels prevents sympathetic

yperactivity and improves cardiac function in rats post-MI. 3

m J Physiol Heart Circ Physiol [Epub ahead of print],ecember 22, 200428. Pitt B, Zannad F, Remme WJ, et al: The effect of

pironolactone on morbidity and mortality in patients withevere heart failure. Randomized Aldactone Evaluation Studynvestigators. N Engl J Med 341:709-717, 1999

29. Pitt B, Remme W, Zannad F, et al: Eplerenone Post-cute Myocardial Infarction Heart Failure Efficacy andurvival Study Investigators. Eplerenone, a selective aldoste-one blocker, in patients with left ventricular dysfunctionfter myocardial infarction. N Engl J Med 348:1309-1321,00330. Farman N, Bocchi B: Mineralocorticoid selectivity:olecular and cellular aspects. Kidney Int 57:1364-1369,

00031. Sasamura H, Suzuki H, Takita T, et al: Response of

lasma immunoreactive active renin, inactive renin, plasmaenin activity, and aldosterone to hemodialysis in patientsith diabetic nephropathy. Clin Nephrol 33:288-292, 199032. DeCarvalho JG, Emery AC Jr, Frohlich ED: Spirono-

actone and triamterene in volume-dependent essential hyper-ension. Clin Pharmacol Ther 27:53-56, 1980

33. Ylostalo P, Heikkinen J, Kauppila A, Pakarinen A,arvinen PA: Low-dose spironolactone in the treatment ofemale hirsutism. Int J Fertil 32:41-45, 1987

34. Sugarman A, Brown RS: The role of aldosterone inotassium tolerance: Studies in anephric humans. Kidney Int

4:397-403, 1998