International Journal of Cardiology 167 (2013) 34–40

Contents lists available at SciVerse ScienceDirect

International Journal of Cardiology

j ourna l homepage: www.e lsev ie r .com/ locate / i j ca rd

Hypertonic saline solution for prevention of renal dysfunction in patients withdecompensated heart failure☆,☆☆

Victor S. Issa a,⁎, Lucia Andrade b, Silvia M. Ayub-Ferreira a, Fernando Bacal a, Ana C. de Bragança b,Guilherme V. Guimarães a, Fabiana G. Marcondes-Braga a, Fátima D. Cruz a, Paulo R. Chizzola a,Germano E. Conceição-Souza a, Irineu T. Velasco c, Edimar A. Bocchi a

a Heart Institute (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, Brazilb Department of Nephrology, Laboratory of Basic Research, Faculdade de Medicina da Universidade de São Paulo, Brazilc Emergency Department, Faculdade de Medicina da Universidade de São Paulo, Brazil

☆ The authors declare no conflicts of interest to dis☆☆ Trial registration — NCT00555685 at www.clinica⁎ Corresponding author at: Dr. Eneas de Carvalho A

Paulo (SP), Brazil. Tel./fax: +55 11 3069 5419.E-mail address: victor.issa@incor.usp.br (V.S. Issa).

0167-5273/$ – see front matter © 2011 Elsevier Irelanddoi:10.1016/j.ijcard.2011.11.087

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 16 August 2011Received in revised form 30 September 2011Accepted 27 November 2011Available online 12 January 2012

Keywords:Heart failureRenal failureRenal dysfunctionHypertonic saline solutionAquaporin-2

Background: Renal dysfunction is associated with increased mortality in patients with decompensated heartfailure. However, interventions targeted to prevention in this setting have been disappointing. We investigatedthe effects of hypertonic saline solution (HSS) for prevention of renal dysfunction in decompensated heartfailure.Methods: In a double-blind randomized trial, patients with decompensated heart failure were assigned toreceive three-day course of 100 mL HSS (NaCl 7.5%) twice daily or placebo. Primary end point was an increasein serum creatinine of 0.3 mg/dL ormore. Main secondary end point was change in biomarkers of renal function,including serum levels of creatinine, cystatin C, neutrophil gelatinase-associated lipocalin—NGAL and the urinaryexcretion of aquaporin 2 (AQP2), urea transporter (UT-A1), and sodium/hydrogen exchanger 3 (NHE3).Results: Twenty-two patients were assigned to HSS and 12 to placebo. Primary end point occurred in two (10%)patients in HSS group and six (50%) in placebo group (relative risk 0.3; 95% CI 0.09–0.98; P=0.01). Relative to

baseline, serum creatinine and cystatin C levels were lower in HSS as compared to placebo (P=0.004 and 0.03,respectively). NGAL level was not statistically different between groups, however the urinary expression ofAQP2, UT-A1 and NHE3 was significantly higher in HSS than in placebo.Conclusions: HSS administration attenuated heart failure-induced kidney dysfunction as indicated by improve-ment in both glomerular and tubular defects, a finding with important clinical implications. HSS modulated theexpression of tubular proteins involved in regulation of water and electrolyte homeostasis.

© 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Heart failure is a condition associated with poor prognosis [1], andepisodes of decompensation requiring hospital care are frequent [2].In these circumstances, the occurrence of renal dysfunction carries amajor prognostic burden [3] and even small changes in serumcreatinine are associated with higher mortality rate [4]. Mechanismsinvolved in the simultaneous occurrence of cardiac and renaldysfunctions are scarcely understood, and therapies targeted toprevention of renal dysfunction in patients with decompensatedheart failure have shown disappointing results [5,6].

In patients with heart failure, serum creatinine and estimations ofglomerular filtration rate (GFR) are usually used to evaluate renal

close.ltrials.org.guiar, 44, CEP 05403-000, São

Ltd. All rights reserved.

function. However, neither creatinine nor GFR completely representsthe function of the kidney, which comprises glomerular function,tubular function, along with other specific metabolic and hormonalfunctions [7]. In this sense, markers other than creatinine are beingused, such as cystatin C, which has been shown to be superior toserum creatinine in different patient populations [8], and neutrophilgelatinase-associated lipocalin (NGAL), an earlier marker of acuterenal injury [9,10]. In addition, tubular cell transporters, such asaquaporin-2, can be detected in urine, and have been described asmarkers of tubular function in the setting of acute renal injury[11–13].

Hypertonic saline solutions have been studied in different forms ofcardiovascular collapse since 1917 [14], and data from experimentalshock models demonstrate that the infusion of 7.5% NaCl producesvasodilatation and increased regional blood flow to coronary [15],renal [16], intestinal and skeletal muscle [17] circulation. Additionallysaline hypertonic improves renal function and myocardial contractility,a finding that is attributed to a direct cardiac inotropic effect induced byhypertonicity [18,19]. In patients, the infusion of 7.5% NaCl has been

35V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

successfully used in cardiogenic shock due to right ventricularinfarction [20].

Small volumes of saline solutions have also been tested in patientswith heart failure, [21,22] and most studies focused on safety andeffectiveness aspects. A randomized trial reported as a secondaryfinding in a selective population of patients highly resistant todiuretics, that the infusion of saline solution with different tonicitieswas associated with lower creatinine levels [23]. However no previousstudy specifically examined the effects of saline solution over renalfunction in patientswith decompensated heart failure, andmechanismsrelated to improved renal function in this setting remain unexplored.

Thus, we hypothesized that the infusion of hypertonic salinesolution to patients with decompensated heart failure could preventthe occurrence of renal dysfunction. The aim of this study was todetermine the effects of hypertonic saline solution on renal functionin this setting, and study possible mechanisms. The primary outcomewas increase in serum creatinine, and secondary outcomes includednewer markers of both glomerular and tubular function, namelycystatin C, NGAL and tubular cell transporters.

2. Methods

2.1. Study design

The present study is a single-center, randomized, double-blind, placebo-controlledtrial performed in a tertiary hospital dedicated to cardiology, and designed to evaluatethe effects of the administration of hypertonic saline solution (NaCl 7.5%) to patientswith decompensated heart failure for primary and secondary prevention for renaldysfunction. The study protocol was approved by the institutional Ethics Committee,and all patients gave written informed consent before enrollment. The authors of thismanuscript have certified that they comply with the Principles of Ethical Publishingin the International Journal of Cardiology [43].

Patients assigned to intervention group (HSS group) received a three-day course of100 mL of hypertonic saline solution (NaCl 7.5%) infused during 1 h, twice daily.Patients assigned to placebo group received a three-day course of 100 mL of NaCl0.9% infused during 1 h, twice daily. In both groups, the saline infusion was followedby an intravenous bolus dose of furosemide; the initial dose of furosemide wasestimated considering the dose previously administered to patient, renal functionand body weight; the initial dose could be modified along the course of the protocol,according to the initial patient response with the pre-established objective to achievea weight loss of 500–1000 g per day.

2.2. Patients

Patients 18 years of age or older, admitted with the diagnosis of decompensatedheart failure in the presence of congestive phenomena, with an ejection fraction ofno more than 40% as measured by transthoracic echocardiography were consideredeligible (Fig. 1). Patients were required to be receiving standard therapy for heart

Fig. 1. Flow of participan

failure, as determined to be appropriate by their physicians, and according to currentguidelines [24,25]. Criteria for the exclusion of patients were: patient refusal, signs ofhypoperfusion, alcohol abuse, primary valvular disease, myocardial infarction or unstableangina within 6 months before randomization, cardiac surgery or angioplasty within6 months before randomization, restrictive cardiomyopathy, chronic obstructivepulmonary disease, immunosuppressive therapy, malignant tumors, acute pulmonaryembolism, surgical interventions or infections in the last 30 days, serum creatinine over3.0 mg/dL, serum potassium over 5.5 mg/dL, any severe systemic disease expected toimpair survival, pregnancy or childbearing potential. Intravenous inotropic therapy wasnot considered an exclusion criterion, but the dose of the inotrope was required to beunchanged for 48 h before randomization, in the absence of clinical parameters ofhypoperfusion at the time of enrollment. In two patients, exclusion criteria occurredafter randomization: one patient developed cardiogenic shock and another patientdeveloped creatinine elevation up to 4.2 mg/dL after randomization but beforeintervention. The decision to exclude these patients was based on the fact thatunder these circumstances, the intervention was not safe to patients, and theoccurrence of the primary outcome was compromised.

All eligible patients were recruited from June 2008 and August 2010 and werefollowed until December 2010.

2.3. Randomization and masking

For allocation of the participants, a computer-generated list of random numbers inblocks of three was used, in a proportion of two interventions to one placebo. Patients,investigators and care providers were blind to allocation; 7.5% NaCl and 0.9 NaClsolutions were identical in appearance, and prepared in bags consecutively identifiedby concealed codes, according to the randomization schedule. Randomization listwas known exclusively by a pharmacist and a nurse not directly involved in the careof patients; they stored the randomization list, assigned patients to each groupaccording to randomization schedule, and delivered the study medication. Investigators,patients and care providers were kept blind after assignment.

2.4. Outcome measures

The pre-specified primary outcome was an increase in serum creatinine of0.3 mg/dL [4,6] or more during the study period, since this subtle creatinine elevationhas been associated with higher mortality in the setting of acute decompensated heartfailure.

Additionally, serum creatinine, urine output, body weight, serum sodium, urea,cystatin C, and neutrophil gelatinase-associated lipocalin (NGAL) were measured theday before the protocol was initiated, every day during the 3-day protocol, and 24 hafter the end of the protocol. Markers of neurohumoral and inflammatory activationwere measured before and after intervention. In order to further explore renal tubularfunction we measured the expression of renal tubular membrane proteins in urine(namely aquaporin 2 [AQP2] sodium-hydrogen exchanger 3 [NHE3] and ureatransporter A1 [UTA1]) before and after intervention. These transporters were selectedbecause they are involved in mechanisms of urine concentrations, and also becausethey are present in different locations along the nephron.

Samples containing sera from patients were centrifuged and stored under −80 °Cuntil analysis was performed with commercially available kits. Cystatin C and NGALwere determined in the serum of patients using enzyme linked immunosorbentassay (ELISA) kits (BioVendor LLC, North Carolina, USA); the limit of detection was

ts through the trial.

Table 1Baseline characteristics of patients.

HSSN (%)/mean±CI

PlaceboN (%)/mean±CI

P value

Number of patients 20 12Age (years) 53.3±13 41.5±13.1 0.02Gender

Male 19 (95%) 7 (58.3%) 0.01Female 1 (5%) 5 (41.7%)

Body mass index 28.4±4.8 27±7.4 0.513Etiology

Ischemic 6 (30%) 1 (8.3%) 0.417Idiopathic 3 (15%) 3 (25%)Hypertension 1 (8.3%)Chagas' disease 6 (30%) 5 (41.7%)Others 5 (25%) 2 (16.7%)

Diabetes mellitus 5 (20%) 1 (8.3%) 0.242Left ventricle ejection fraction 20.8±5.1 26.9±7.4 0.73Left ventricle diastolic diameter 74.5±12.7 68.6±6.6 0.131Systolic blood pressure 103.2±14.5 100±16.5 0.570Diastolic blood pressure 66.1±8.2 64.2±10 0.565Ascites 9 (45%) 7 (58.3%) 0.465Atrial fibrillation 9 (45%) 5 (41.7%) 0.854CRT 2 (10%) –

Cardiac defibrillator – 1 (8.3%) –

Sodium (mEq/L) 137.6±3.5 135.7±3.4 0.068Potassium (mEq/L) 4.3±0.7 4.1±0.5 0.264Creatinine 1.72±0.47 1.58±0.48 0.43Urea 80.8±35.6 83.7±34.6 0.655Serum osmolarity 293.8±11.2 286.9±12.1 0.114Type-B natriuretic peptide 2205.4±1326.1 1603.6±1260.7 0.233Hemoglobin (g/dL) 11.9±1.7 11.4±1.6 0.284Medication

Beta-blocker 15 (75%) 10 (83.3%) 0.581ACEI/AT2 inhibitor 11 (55%) 7 (66.7%) 0.515Hydralazine 14 (70%) 4 (33.3%) 0.043Spironolactone 9 (45%) 5 (41.7%) 0.854Digoxin 7 (35%) 2 (16.7%) 0.264Dobutamine 6 (30%) 5 (41.7%) 0.501

Furosemide (mg) — median(quartile range)

80 (60–100) 80 (40–102) 0.814

Fig. 2. Serum level of creatinine in hypertonic saline solution (HSS) group and inplacebo group.

36 V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

0.2 ng/mL for cystatin C, and 0.1 ng/mL for NGAL. Radioimmunoassay tests wereperformed for detection for aldosterone, renin, angiotensin II (Simens Health Care,Pennsylvania, USA) as well as vasopressin (DiaSource, Nivelles, Belgium); TNF alpha,interleukin 6 and BNP were measured with chemiluminescent kits (Immulite, SimensHealth Care, Pennsylvania, USA).

To analyze the urinary protein excretion of AQP2, UT-A1 and NHE3, creatinineequivalents of urine samples were analyzed by Western blot. We obtained thepeptide-derived polyclonal antibodies specific to AQP2 and NHE3 from Santa CruzBiotechnology (Santa Cruz, CA). The peptide-derived polyclonal antibody specific toUT-A1 was kindly supplied by Dr. Jeff Sands (Emory University, Atlanta, GA).

An independent data and safety monitoring board periodically reviewed theefficacy and safety data. In-hospital mortality and need of renal replacement therapywere explored.

2.5. Statistical analysis

Continuous variables are expressed as mean±standard deviation or median(interquartile range) as appropriate, and categorical variables are expressed aspercentages. All variable were tested for Gaussian distribution with D'Agostino–Pearsontest, and a P value of less than 0.10 was considered as evidence of Gaussian distribution.Variables without Gaussian distribution were submitted to logarithmic transformation.Baseline characteristics were compared by independent-samples Student T-test or Fisherexact test. For the analysis of repetitive measures we performed double factor ANOVA. Allanalyses and graphs were performed with SPSS statistical software version 11.5 andGraphPad Prism software version 4.02. A sample size of 50 patients was estimated inorder to detect a 25% variation in the value of serum creatinine with a two-sided 5%significance level and a power of 80%.

3. Results

Between June 2008 and December 2010, 34 patients wererandomized and 22 patients were assigned to HSS group and 12 toplacebo group (Fig. 1). Two patients from HSS were excluded afterrandomization but before intervention. The trial was stopped on therecommendation of the independent data and safety monitoringboard in a programmed interim analysis; the decision to stop thetrial was based on the fact that the primary end-point had beenreached. The median follow-up time was 69.5 (32.2–164.7) days,and no patient was lost to follow-up. Ten (50%) patients in HSSgroup died during hospital admission, as well as 4 (30%) patients inplacebo group. Baseline characteristics of patients are depicted inTable 1.

3.1. Primary outcome

The primary end point occurred in 8 (25%) patients — two (10%)patients in HSS group and six (50%) in placebo group; there was asignificant 70% reduction in the occurrence of creatinine elevation(relative risk 0.3; 95% CI 0.09–0.98; P=0.01) during the interventionperiod.

3.2. Secondary outcomes

3.2.1. Biomarkers of renal functionRelative to baseline, serum levels of creatinine and cystatin C

levels were lower in HSS as compared to placebo (Fig. 2 — Table 2);the peak creatinine occurred 72 h after intervention, whereas thepeak cystatin C occurred 48 h after intervention. Serum level of bothbiomarkers of glomerular function increased in HSS group 24 h afterintervention was stopped. Serum NGAL level decreased in HSSgroup during the protocol, whereas serum NGAL increased in placebogroup, but differences were not statistically significantly differentbetween groups (Fig. 2). After intervention, the urinary proteinexpression of AQP2 was significantly higher in the HSS group thanin the placebo group (3.5±0.33 vs. 1.6±0.38 arbitrary units, P=0.004) as well as the expression of UT-A1 (1.9±0.12 vs. 1.1±0.12arbitrary units, P=0.001) and the expression of the NHE3 wassignificantly higher in the HSS group than in the placebo group(1.37±0.19 vs. 0.46±0.04 arbitrary units, P=0.008) (Fig. 3). Therewas a correlation between serum creatinine and cystatin C (correlation

coefficient 0.5, CI95% 0.17–0.73, P=0.004), as well as between serumcreatinine and NGAL (correlation coefficient 0.45, CI95% 0.11–0.70,P=0.01). No statistically significant correlation was found betweenurinary proteins and cystatin C or NGAL (Figs. 4 and 5).

In the follow-up, at 2 weeks after intervention was stopped serumcreatinine was 1.8±0.7 mg/dL in HSS group and 1.9±0.7 mg/dL in

Table 2Values of clinical and laboratory variables during the protocol.

Variable (mean±sd) Baseline Day 1 Day 2 Day 3 24 h post intervention P factor P time P interaction

Weight (kg) 0.07 b0.001 0.62HSS 83.8±18.4 82.8±18.0 82.0±17.8 81.3±17.6 80.9±18.3Placebo 79.0±28.0 78.2±27.7 78.6±27.6 77.9±27.8 76.4±27.6

Urine output (mL/kg/h) 0.07 b0.001 0.11HSS 0.93±0.55 1.15±0.44 1.29±0.46 1.12±0.41 0.83±0.32Placebo 0.73±0.33 1.09±0.82 0.91±0.43 1.25±0.74 1.28±0.80

Sodium (mEq/L) 0.024 0.03 0.92HSS 137.6±3.5 138.7±3.7 139.1±3.9 139.5±4.6 137.8±4.3Placebo 134.4±5.6 135.0±5.5 134.9±6.1 136.0±5.0 134.9±5.2

Osmolarity (mOsm/kgH2O) 0.13 0.005 0.93HSS 293.8±11.3 294.0±10.9 297.8±12.5 296.1±12.4 294.8±13.0Placebo 286.9±12.1 288.5±10.9 292.5±13.5 289.8±9.60 289.4±11.7

Creatinine (mg/dL) 0.58 0.06 0.004HSS 1.72±0.47 1.71±0.44 1.65±0.44 1.66±0.55 1.88±0.68Placebo 1.58±0.48 1.81±1.01 1.96±1.06 1.89±0.77 1.90±0.76

Cystatin C (ng/mL) 0.30 0.23 0.03HSS 1531.8±400.5 1498.8±390.8 1435.1±318.5 1476.7±301.0 1638.6±469.3Placebo 1452.5±507.8 1731.1±639.5 1654.0±539.8 1706.5±464.9 1707.2±557.9

NGAL (ng/mL) 0.05 0.27 0.68HSS 72.4±36.4 65.0±33.6 62.7±27.4 62.0±28.6 67.9±29.6Placebo 81.0±27.2 77.1±32.3 87.1±51.1 88.0±48.9 93.0±58.8

Urea (mg/dL) 0.66 0.94 0.66HSS 80.8±35.6 78.6±33.3 77.6±32.8 77.0±38.6 83.2±45.2Placebo 83.7±.43.6 82.7±43.6 85.9±46.5 85.7±42.0 86.5±41.3

Na excretion fraction 0.71 b0.001 0.48HSS 1.67±2.04 3.19±2.34 3.19±2.31 2.86±1.80 2.21±2.05Placebo 1.42±2.04 2.22±1.93 2.32±1.20 3.27±2.08 3.06±2.66

Urinary osmolarity 0.005 b0.001 0.50HSS 432.4±98.5 361.8±46.0 369.8±44.1 385.6±66.8 395.3±57.7Placebo 372.8±56.6 336.9±28.9 337.0±31.6 332.5±39.3 331.6±34.8

Free water clearance 0.08 0.46 0.15HSS −0.42±0.17 −0.21±0.65 −0.37±0.37 −0.40±0.45 −0.36±0.30Placebo −0.23±0.14 −0.21±0.14 −0.13±0.39 −0.02±0.63 −0.18±0.39

NGAL: neutrophil gelatinase-associated lipocalin; HSS: hypertonic saline solution group; sd: standard deviation.

37V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

placebo group; at 30 days after intervention was stopped, creatininelevelswere respectively 1.6±0.7 mg/dL and 1.9±1.4 mg/dL; differenceswere not statistically significant.

3.2.2. Clinical variablesBoth groups increased urine output and reached the pre-

established weight loss of 500–1000 g per day (Table 2); the dose offurosemide in HSS group was smaller than in placebo group (HSSdose 120 mg [80–120 mg]; placebo dose 160 mg [120–180 mg] —

Fig. 3. Serum level of cystatin C relative to baseline in hypertonic saline solution (HSS)group and in placebo group.

Pb0.001). Serum sodium and serum osmolarity increased in bothgroups during intervention, and the changes did not differ betweengroups. Changes in urinary osmolarity and sodium excretion fractionwere similar between groups; there was a tendency towards morepronounced free water clearance of in HSS group. No serious eventspossibly related to the intervention were reported; nine (45%)patients in HSS group and three (25%) patients in control groupreported pain during infusion of saline solutions. Phlebitis occurredin three (15%) patients in HSS group and none in control group.

Fig. 4. Serum level of neutrophil gelatinase-associated lipocalin (NGAL) in hypertonicsaline solution (HSS) group and in placebo group.

Fig. 5. Urinary expression of renal tubular membrane proteins in hypertonic saline solution (HSS) group and in placebo group. AQP2 – aquaporin 2; UTA1 – urea transporter1; NHE3 - sodium-hydrogen exchanger 3; HSS: hypertonic saline solution.

38 V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

3.2.3. Inflammatory and humoral activationInterleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-α)

values did not significantly change during the protocol, in neithergroup. Similarly, the levels of BNP along with other markers ofhumoral activation remained stable during the protocol (Table 3).

4. Discussion

We demonstrate for the first time in a specifically designed trialthat the administration of HSS to patients with decompensatedheart failure can prevent the occurrence of renal dysfunction.Importantly, HSS administration attenuated heart failure-inducedkidney dysfunction as indicated by improvement in both renalglomerular and tubular defects, a finding that has potential clinicalimplications. HSS also modulated the expression of renal tubularproteins involved in regulation of water and electrolyte homeostasis.

Table 3Neurohumoral and inflammatory variables before and after the intervention.

HSSmedian(interquartile range)

P value Placebomedian(interquartile range)

P value

Renin 0.44 0.47Baselinea 20.6 (5.4–35.4) 32.9 (17.9–42.6)Postintervention

13.7 (5.6–35.1) 23.8 (17.8–45.1)

Aldosterone 0.27 0.86Baselinea 10 (3.7–17.2) 9.7 (6.2–41.6)Postintervention

12 (4.6–19.7) 7.6 (3.1–79.9)

Vasopressin 0.15 0.51Baselinea 2.4 (1.9–3.1) 2.7 (2–3.4)Postintervention

2.6 (2–4.8) 2.4 (1.9–3.3)

Angiotensin II 0.38 0.77Baselinea 25.6 (13.4–44.9) 39.4 (12.6–71.1)Postintervention

28.2 (15–45.9) 40.2 (12.5–70.1)

BNP 0.11 0.36Baselinea 2077 (1046–3353) 1219 (574–2336)Postintervention

1728 (1067–3574) 1384 (682–1816)

Interleukin 6 0.35 0.24Baseline† 17.1 (8.9–24.5) 26.4 (12.9–57.7)Postintervention

23.6 (10.4–30.3) 22.9 (12.3–32.6)

Alpha TNF 0.63 0.66Baselinea 16.5 (13.6–27.7) 23.2 (14.6–38.5)Postintervention

16.7 (12.5–31.1) 22.7 (15.8–43)

HSS: hypertonic saline solution group; TNF: tumor necrosis factor; BNP: B-typenatriuretic peptide.

a No difference in baseline values between HSS and placebo.† P=0.03.

Importantly, patients in HSS group improved the effectiveness ofdiuretic therapy, as they received lower doses of diuretics, butpresented similar amount of urine output and weight loss.

We defined the occurrence of renal dysfunction, as an increase inserum creatinine in 0.3 mg/dL, since this is the smallest creatinineelevation associated with worse prognosis, and has been used inother clinical trials. However, it should be noticed that it has beensuggested that this phenomenon may be related to transient modifi-cations in renal perfusion, and may not represent established kidneyinjury [26].

In accordance with our results, a single previous study suggested,as a secondary outcome, that HSS was associated with reduction inthe creatinine level in patients with decompensated heart failure;[23] in this double-blind, randomized study, only patients withdiuretic resistance were included, and an unusual high dose offurosemide was given in association with different HSS tonicities,ranging from 1.4 to 4.6%. Our results expand the use of hypertonicsolutions to a broader scenario of heart failure patients, closer tocurrent clinical practice, and indicate that adjustments in solutiontonicity based on serum sodium may be unnecessary.

HSS can improve glomerular function, as indicated by lower cystatinC levels, a marker that can more accurately and rapidly detect changesin GFR when compared to creatinine [27]. This hypothesis is supportedby the reported effects of HSS over the cardiovascular and renalsystems. HSS can increase in up to 30% [28] the plasma volume, andthus offers protection against episodes of hypovolemia, that have beenreported to occur transiently or persistently in patients with heartfailure, [29,30] especially in those under diuretic therapy. HSS can alsoincrease renal perfusion, inducing vasodilatation in renal [31] andmesenteric [32] territories. Additionally, a direct inotropic effect hasbeen described [33], and possible mechanisms include reduction ofmyocardial edema, increased myocardial uptake of calcium, andrestoration of transmembrane potential [34,32]. Taken together, thesehemodynamic and cardiac effects of HSS offer a rationale for thefindings reported in our study.

HSS can also improved tubular function, as indicated by augmentedurinary expression of AQP2, UT-A1 and NHE3, tubular cell transportersthat are located in different portions of the nephron. AQP2 is a waterchannel of collecting ducts regulated by vasopressin release; inresponse to vasopressin, AQP2 is translocated from cytoplasmic vesiclesto apical plasma membranes, and increases water permeability. It hasbeen shown that APQ2 plays an important role in water homeostasisin heart failure [35,13], and its expression is reduced in models ofacute renal injury. To the best of our knowledge, there are no previousreports regarding the expression of UT-A1 and NHE3 in patients withheart failure.

In contrast, NGAL is a very early marker of tubular damage thatincreases in patients with acute and chronic renal injury [36,37], andin patients with heart failure [7] and its levels did not significantly

39V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

differ between groups, afinding possibly related to the limited number ofpatients studied; otherwise, it should be noticed that NGAL has beenmore closely associated with pathophysiological conditions of tissuedamage and inflammation; [38,39] however, the effects of HSS overrenal function are probably related tomodulation of renal hemodynamicsas well as mechanisms of water and electrolyte homeostasis.

We found no effect of HSS over humoral and inflammatorymarkers, an aspect that has been scarcely studied in patients withheart failure [40]. Immune-modulator effects of HSS have beendescribed in other scenarios [41,42] and include modulation ofneutrophils, macrophages, T-cells' activation, reduction of TNF-alphaproduction, increase in anti-inflammatory cytokines, modulation ofapoptosis, and influence on the transcriptional, functional and proteinexpression of genes related to matrix metalloproteinases. Ourfindings further point to a predominantly hemodynamic and vasculareffect of HSS. BNP levels remained unchanged; it is possible thatdifferent variables influenced the BNP levels, such as the deteriorationof renal function that occurred in placebo group, and a possibletransient increase in plasma volume in patients that received HSS. Incontrast with our findings, reduction of BNP value with HSS has beendescribed [23,43]. Finally, it should be noticed that all neurohormonaland inflammatory measurements were made 24 h before interventionand 24 h after intervention it was terminated. Interestingly, most ofthe effects observedwith the infusion of hypertonic salinewere presentduring the 3 day infusion, butwere lost after infusionwas terminated. Itis possible that if these measurements were performed earlier, moresubtle differences could have been found. Our study included a limitednumber of patients, what may have prevented from detecting moresubtle differences in some of the secondary end-points, such as BNP,neurohumoral and inflammatory markers. Additionally, there weredifferences between groups at baseline — patients in HSS wereolder and had a higher proportion of male individuals; even thoughdifferences were not statistically significant, it should be acknowledgedthat a higher proportion of patients in placebo group were using beta-blockers, ACEI/AT2 inhibitors as well as dobutamine. Like most clinicalstudies in renal function, we did not perform a direct measure of theglomerular filtration rate and the limited duration of the intervention(3 days) may have been insufficient to detect longer term effects ofthe intervention. Two patients were not included in the analysis eventhough they were randomized into the trial, what may have influencedour results.

Taken together, our findings indicate that HSS is an effectivemeasure for prevention of renal dysfunction in patients with decom-pensated heart failure, a condition associated with a poor prognosis.These data point to a promising low-cost therapeutical strategy inpatients with decompensated heart failure, and warrant furtherinvestigation in multicenter trials.

Funding

This workwas supported by a research grant from the Fundação deAmparo à Pesquisa do Estado de São Paulo— FAPESP (2007/04048-7).

Acknowledgements

The authors of this manuscript have certified that they complywith the Principles of Ethical Publishing in the International Journalof Cardiology.

References

[1] Bocchi EA, Guimarães G, Tarasoutshi F, Spina G, Mangini S, Bacal F. Cardiomyopa-thy, adult valve disease and heart failure in South America. Heart 2009;95:181–9.

[2] Bocchi EA, Cruz F, Guimarães G, et al. Long-term prospective, randomized, con-trolled study using repetitive education at six-month intervals and monitoringfor adherence in heart failure outpatients: the REMADHE trial. Circ Heart Fail2008;1:115–24.

[3] Adams Jr KF, Fonarow GC, Emerman CL, et al. Characteristics and outcomes of pa-tients hospitalized for heart failure in the United States: rationale, design, andpreliminary observations from the first 100,000 cases in the Acute Decompen-sated Heart Failure National Registry (ADHERE). Am Heart J 2005;149:209–16.

[4] Gottlieb SS, Abraham WT, Butler J, et al. The prognostic importance of differentdefinitions of worsening renal function in congestive heart failure. J Card Fail2002;8:136–41.

[5] Massie BM, O'Connor CM, Metra M, et al. PROTECT Investigators and Committees.Rolofylline, an adenosine A1-receptor antagonist, in acute heart failure. N Engl JMed 2010;363:1419–28.

[6] Felker GM, Lee KL, Bull DA, et al. NHLBI Heart Failure Clinical Research Network.Diuretic strategies in patients with acute decompensated heart failure. N Engl JMed 2011;364:797–805.

[7] Damman K, Voors AA, Navis G, Veldhuisen DJ, Hillege HL. Current and novel renalbiomarkers in heart failure. Heart Fail Rev 2012 Mar;17(2):241–50.

[8] Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creat-inine as a marker of kidney function: a meta-analysis. Am J Kidney Dis 2002;40:221–6.

[9] Mishra J, Dent C, Tarabishi R, et al. Neutrophil gelatinase-associated lipocalin(NGAL) as a biomarker for acute renal injury after cardiac surgery. Lancet2005;365:1231–8.

[10] Mishra J, Ma Q, Prada A, et al. Identification of neutrophil gelatinase-associatedlipocalin as a novel early urinary biomarker for ischemic renal injury. J Am SocNephrol 2003;14:2534–43.

[11] Höcherl K, Schmidt C, Kurt B, Bucher M. Inhibition of NF-kappaB amelioratessepsis-induced downregulation of aquaporin-2/V2 receptor expression andacute renal failure in vivo. Am J Physiol Renal Physiol 2010;298:F196–204.

[12] Grinevich V, Knepper MA, Verbalis J, Reyes I, Aguilera G. Acute endotoxemia inrats induces down-regulation of V2 vasopressin receptors and aquaporin-2 con-tent in the kidney medulla. Kidney Int 2004;65:54–62.

[13] Funayama H, Nakamura T, Saito T, et al. Urinary excretion of aquaporin-2 waterchannel exaggerated dependent upon vasopressin in congestive heart failure.Kidney Int 2004;66:1387–92.

[14] Penfield WG. The treatment of severe and progressive hemorrhage by intrave-nous injections. Am J Physiol 1919;48:121–8.

[15] Crystal GJ, Gurevicius J, Kim SJ, Eckel PK, Ismail EF, SalemMR. Effects of hypertonicsaline solutions in the coronary circulation. Circ Shock 1994;42:27–38.

[16] Maningas P. Resuscitation with 7.5% NaCl in 6% dextran-70 during hemorrhagicshock in swine: effects on organ blood flow. Crit Care Med 1987;15:1121–6.

[17] Rocha-e-Silva M, Negraes GA, Soares AM, Pontieri V, Loppnow L. Hypertonic re-suscitation from severe hemorrhagic shock: patterns of regional circulation. CircShock 1986;19:165–75.

[18] Ing RD, Nazeeri MN, Zeldes S, Dulchavsky SA, Diebel LN. Hypertonic saline/dextranimproves septic myocardial performance. Am Surg 1994;60:507–8.

[19] Kien ND, Kramer GC. Cardiac performance following hypertonic saline. Braz J MedBiol Res 1989;22:2245–8.

[20] Ramires JA, Serrano Júnior CV, César LA, Velasco IT, Rocha e Silva Júnior M, PileggiF. Acute hemodynamic effects of hypertonic (7.5%) saline infusion in patientswith cardiogenic shock due to right ventricular infarction. Circ Shock 1992;37:220–5.

[21] Issa VS, Bacal F, Mangini S, et al. Hypertonic saline solution for renal failure preven-tion in patients with decompensated heart failure. Arq Bras Cardiol 2007;89:251–5.

[22] Paterna S, Pasquale P, Parrinello G, et al. Effects of high-dose furosemide andsmall-volume hypertonic saline solution infusion in comparison with a highdose of furosemide as a bolus, in refractory congestive heart failure. Eur J HeartFail 2000;2:305–13.

[23] Paterna S, Di Pasquale P, Parrinello G, et al. Changes in brain natriuretic peptidelevels and bioelectrical impedance measurements after treatment withhigh-dose furosemide and hypertonic saline solution versus high-dose furose-mide alone in refractory congestive heart failure: a double-blind study. J AmColl Cardiol 2005;45:1997–2003.

[24] Bocchi EA, Braga FG, Ferreira SM, et al. Sociedade Brasileira de Cardiologia. III Dir-etriz Brasileira de Insuficiência Cardíaca Crônica. Arq Bras Cardiol 2009;93:1–71.

[25] Jessup M, Abraham WT, Casey DE, et al. 2009 focused update: ACCF/AHA guide-lines for the diagnosis and management of heart failure in adults. J Am Coll Car-diol 2009;53:e1–90.

[26] Blair JE, Pang PS, Schrier RW, et al. Changes in renal function during hospitalizationand soon after discharge in patients admitted for worsening heart failure in theplacebo group of the EVEREST trial. Eur Heart J 2011;32(20):2563–72.

[27] Lassus JPE, Nieminen MS, Peuhkurinen K, et al. Markers of renal function andacute kidney injury in acute heart failure: definitions and impact on outcomesof the cardiorenal syndrome. Eur Heart J 2010;31:2791–8.

[28] Velasco IT, Pontieri V, Rocha e Silva M, Lopes OU. Hyperosmotic NaCl and severehemorrhagic shock. Am J Physiol 1980;239:H664–73.

[29] Feigenbaum MS, Welsch MA, Mitchell M, Vincent K, Braith RW, Pepine CJ. Con-tracted plasma and blood volume in chronic heart failure. J Am Coll Cardiol2000;35:51–5.

[30] Kalra PR, Anagnostopoulos C, Bolger AP, Coats AJS, Anker SD. The regulation andmeasurement of plasma volume in heart failure. J Am Coll Cardiol 2002;39:1901–8.

[31] Gazitua S, Scott JB, Chou CC, Haddy FJ. Effect of osmolarity on canine renal vascu-lar resistance. Am J Physiol 1969;217:1216-.

[32] Thompson R, Greaves I. Hypertonic saline-hydroxyethyl starch in trauma resusci-tation. J R Army Med Corps 2006;152:6–12.

[33] Rocha e Silva M, Velasco IT, Nogueira da Silva RI, Oliveira MA, Negraes GA, OliveiraMA. Hyperosmotic sodium salts reverse severe hemorrhagic shock: other solutesdo not. Am J Physiol 1987;253:H751–62.

40 V.S. Issa et al. / International Journal of Cardiology 167 (2013) 34–40

[34] Oliveira RP, Velasco I, Soriano F, Friedman G. Clinical review: hypertonic saline re-suscitation in sepsis. Crit Care 2002;6:418–23.

[35] Lutken SC, Kim SW, Jonassen T, et al. Changes of renal AQP2, ENaC, and NHE3 inexperimentally induced heart failure: response to angiotensin II AT1 receptorblockade. Am J Physiol Renal Physiol 2009;297:F1678–88.

[36] Ding H, He Y, Li K, et al. Urinary neutrophil gelatinase-associated lipocalin (NGAL)is an early biomarker for renal tubulointerstitial injury in IgA nephropathy. ClinImmunol 2007;123:227–34.

[37] Mori K, Lee HT, Rapoport D, et al. Endocytic delivery of lipocalin–siderophore–iron complex rescues the kidney from ischemia-reperfusion injury. J Clin Invest2005;115:610–21.

[38] Bolignano D, Donato V, Coppolino G, et al. Neutrophil gelatinase-associatedlipocalin (NGAL) as a marker of kidney damage. Am J Kidney Dis 2008;52:595–605.

[39] Haase M, Bellomo R, Devarajan P, Schlattmann P, Haase-Fielitz A; NGALMeta-analysis Investigator Group. Accuracy of neutrophil gelatinase-associated

lipocalin (NGAL) in diagnosis and prognosis in acute kidney injury: a systematicreview and meta-analysis. Am J Kidney Dis 2009;54:1012–24.

[40] Tuttolomondo A, Pinto A, Di Raimondo D, et al. Changes in natriuretic peptide andcytokine plasma levels in patients with heart failure, after treatment with highdose of furosemide plus hypertonic saline solution (HSS) and after a saline load-ing. Nutr Metab Cardiovasc Dis 2011;21:372–9.

[41] Rizoli SB, Rhind SG, Shek PN, et al. The immunomodulatory effects of hypertonicsaline resuscitation in patients sustaining traumatic hemorrhagic shock: a ran-domized, controlled, double-blinded trial. Ann Surg 2006;243:47–57.

[42] Junger WG, Coimbra R, Liu FC, et al. Hypertonic saline resuscitation. A tool tomodulate immune function in trauma patients. Shock 1997;8:235–41.

[43] Parrinello G, Paterna S, Di Pasquale P, et al. Changes in estimating echocardiogra-phy pulmonary capillary wedge pressure after hypersaline plus furosemide ver-sus furosemide alone in decompensated heart failure. J Card Fail 2011;17:331–9.