Extracorporeal Ultrafiltration inHeart Failure and Cardio-Renal Syndromes

Maria Rosa Costanzo, MD,* Mario Cozzolino, MD,† Nadia Aspromonte, MD,‡ Flavio Mistrorigo, MD,§

Roberto Valle, MD,� and Claudio Ronco, MD¶,#

Summary: Acute decompensated heart failure and fluid overload are the most common causes of hospitalizationin heart failure patients and they often contribute to disease progression. Initial treatment encompasses intrave-nous diuretics, although there might be a percentage of patients refractory to this pharmacologic approach. Newtechnologies have been developed to perform extracorporeal ultrafiltration in fluid-overloaded patients. Newersimplified devices permit ultrafiltration to be performed with peripheral venous access and low blood flows, makingultrafiltration feasible at most hospitals and acute care settings. Extracorporeal ultrafiltration then is prescribed andconducted by specialized teams and fluid removal is planned to restore a status of hydration close to normal.Recent clinical trials, and European and North American practice guidelines, suggest that ultrafiltration is reason-able for patients with refractory congestion not responding to medical therapy and assigned to this recommen-dation a class IIa, level of evidence B. It seems that a close collaboration between nephrologists and cardiologistsmay be the key for a collaborative therapeutic effort in heart failure patients. Further studies suggest that wearabletechnologies might become available soon to treat patients in ambulatory and de-hospitalized settings. These newtechnologies may help to cope with the increasing demand for care of chronic heart failure patients.Semin Nephrol 32:100-111 © 2012 Elsevier Inc. All rights reserved.Keywords: Heart failure, cardio-renal syndrome, ultrafiltration, fluid overload, refractory edema, acute kidneyinjury, oliguria, hemodialysis, hemofiltration

vwfl

Utcc

Cardiorenal syndromes (CRS) represent an impor-tant chapter in the expenditure of health careplans worldwide. In the United States more than

1 million patients are hospitalized for heart failure (HF)each year.1 In most cases, dyspnea as a result of fluidoverload dominates the clinical picture.2 The early treat-ment and resolution of congestion and fluid-excess symp-toms is associated with improved outcomes.3-7 Acutedecompensated HF (ADHF) generally is treated withintravenous (IV) diuretics, which have limited efficacyespecially in patients with underlying chronic kidney

*Midwest Heart Foundation, Lombard, IL.†Renal Division, S. Paolo Hospital, University of Milan, Milan, Italy.‡Department of Cardiology, St Filippo Neri Hospital, Rome, Italy.§Department of Cardiology, St. Bortolo Hospital, Vicenza, Italy.�Heart Failure Unit, Ospedale Civile, Chioggia, Venezia, Italy.¶Department of Nephrology Dialysis and Transplantation, St. Bortolo

Hospital, Vicenza, Italy.#International Renal Research Institute, Vicenza, Italy.Financial support for this work: none.Financial disclosure and conflict of interest statement: Dr. Claudio

Ronco receives speaking honoraria from Alere, Abbot, Gambro andPfizer; Dr. Mario Cozzolino received in the past honoraria fromAbbott, Amgen, Shire, Sanofi, and Roche; Dr. Maria Rosa Costanzoreceives consulting and speaking honoraria from Gambro.

Address reprint requests to Maria Rosa Costanzo, MD, FACC, FAHA,Medical Director, Midwest Heart Specialists Heart Failure andPulmonary Arterial Hypertension Programs, Medical Director, Ed-ward Hospital Center for Advanced Heart Failure, Edward HeartHospital, 4th Floor, 801 South Washington St, PO Box 3226, Na-perville, IL 60566. E-mail: mcostanzo@midwestheart.com

0270-9295/ - see front matter© 2012 Elsevier Inc. All rights reserved.

doi:10.1016/j.semnephrol.2011.11.013

100

disease (CKD).2,8 In these circumstances, extracorporealtechniques of fluid removal may become an importantrescue therapy. The history of extracorporeal ultrafiltra-tion began in the mid-1970s8-10 but significant progresshas occurred in recent years.11,12 Newer simplified de-ices today permit performance of ultrafiltration (UF)ith low extracorporeal priming volumes and low bloodows, making it feasible at most hospitals.13

EXTRACORPOREALULTRAFILTRATION AND RELATED TECHNIQUES

Extracorporeal UF uses a semipermeable membrane toachieve plasma water separation from whole blood.14

Water is transported in response to a transmembranepressure gradient and the process is defined as UF.15-17 If

F occurs during dialytic techniques, additional soluteransport occurs by two main processes: diffusion andonvection.14-17 UF contains crystalloids but not cells orolloids, being iso-osmotic to plasma water.17-19 If a

dialysate solution is present on the other side of themembrane, convection and diffusion continuously inter-act in a dynamic equilibrium,20,21 resulting in possibleosmotic differences in the final composition of the ultra-filtrate. Different membranes are mounted on specificdevices designed to optimize either diffusion (hemodial-ysis predominantly operates with a countercurrent flowof blood and dialysate) or convection (hemofiltrationoperates predominantly by convection).22 New tech-niques such as hemodiafiltration can be performed, com-bining the advantages of diffusion and convection tooptimize clearances for a wide spectrum of solutes. This

is made possible by new partially hydrophilic synthetic

Seminars in Nephrology, Vol 32, No 1, January 2012, pp 100-111

gtadmafiptamb

idmssms

ffia

Extracorporeal ultrafiltration in heart failure 101

membranes with reduced wall thickness.23,24 Differenttechniques may be indicated for different patients andsituations according to fluid and solute removal needs.Blood flow is an important determinant of membraneperformance.25 High blood flows reduce blood viscosityand the chance of clotting, and removes proteins andstagnation layers from the inner side of the membrane. Inclinical practice, high wall shear rates (wall shearstress � velocity gradient at the membrane interface) areachieved with high blood flows and adequate devicedesign, resulting in a higher UF rate (Qf) and soluteclearances.25-28 For slow continuous ultrafiltration(SCUF), very low blood flows can be prescribed andalthough double-lumen catheters are the rule, peripheralveins can be used occasionally as a vascular access.29,30

The extracorporeal typically has pressure sensors andalarms to prevent clotting and circuit explosion as well ashemolysis or air embolism.11,12 Extracorporeal tech-niques can be performed intermittently (�3 times/wk),daily, or continuously.31,32 UF can be performed either incombination with blood-cleansing therapies, sequentially(first UF, then dialysis), or alone when used only toremove fluid.33 SCUF uses low blood flows (Qb � 50-100 mL/min) and low UF rates (Qf � 100-300 mL/h),requiring only small surface area filters and less intensesystemic anticoagulation to maintain circuit patency.33 Ifblood cleansing is needed then continuous veno-venoushemofiltration (CVVH) is used.33 With CVVH, UF ex-ceeds the requested fluid removal in the patient and theinfusion of replacement solutions is needed to maintainthe desired fluid balance. With CVVH the sodium bal-ance can be altered by adjusting the sodium concentra-tion in the replacement solution.34,35 Thus, the dissocia-tion of sodium and water removal by CVVH caninfluence total body sodium (extracellular and intracellu-lar) in a manner not achievable with any other pharma-cologic or extracorporeal therapies. SCUF or CVVHtypically last several days, whereas single sessions of UFor hemofiltration can be conducted intermittently forshorter time periods (hours).33 In critically ill patients orpatients with ADHF, these therapies generally are per-formed continuously over a prolonged period of time andare referred to as continuous renal replacement therapies(CRRT).36 With such therapies, fluid removal rates arelow enough to prevent hemodynamic instability com-monly observed during more aggressive therapies.

UF HARDWARE AND TREATMENTMONITORING

Most dialysis and continuous renal replacement therapymachines can perform isolated UF. More recently, dedi-cated equipment (Aquadex System 100; CHF Solutions,Brooklyn Park, MN and Dedyca; Bellco, Mirandola,Italy) has been designed specifically for SCUF, espe-cially in patients with HF and fluid overload. A machinefor UF in neonates recently was developed in Vicenza,Italy, at the International Renal Research Institute called

CARPEDIEM (Cardio Renal Pediatric Dialysis Emer- t

ency Machine (Fig. 1). Such machines provide data onhe pressures generated in the arterial line, the filter itself,nd the return line, thus allowing early detection of filterysfunction and access-related issues.11,37-39 Further-ore, the UF rate is controlled volumetrically, allowingprecise regulation of filtration fraction and net fluid loss

rom the patient. To prevent line thrombosis, heparin isnfused in the arterial line to achieve a local activatedartial thomboplastin time (aPTT) of 72 to 105. Directhrombin inhibitors can be used in patients with heparinllergies. Contemporary UF circuits also include a he-atocrit sensor that provides an estimate of changes in

lood volume and adequacy of intravascular refilling.40,41

An accurate monitoring of the patient during treatments advised. The hydration status of the patient should beetermined carefully and while fluid is removed, stableaintenance of blood pressure and tissue perfusion

hould be ensured. Although various methods presentignificant limitations, all strategies should be imple-ented to achieve an adequate estimation of hydration

tatus.42-44 Natriuretic peptides (NPs) have become im-portant tools in the diagnosis, treatment, and prognosticassessment of patients with HF.45-48 HF patients mayhave a dry and a wet NP levels, both of which vary withinthe same individual over time and clinical evolution.49-54

B-type natriuretic peptide (BNP) or NT-proBNP levels inADHF patients on admission are excellent diagnostictools and their decrease during the hospital stay correlatewith good outcomes.55 Furthermore, patients with a dis-charge BNP level less than 250 pg/mL generally have thehighest event-free survival.56-64 In HF patients, acutekidney injury (AKI) may occur owing to excessive or toofast UF (CRS type 1).7 This has spurred interest in newAKI biomarkers such as neutrophil gelatinase-associatedlipocalin (NGAL) and kidney injury molecule-1. Becausethese markers become abnormal earlier than serum cre-atinine, AKI can be detected earlier and further renaldamage caused by excessive UF can be prevented. Thesebiomarkers also may have an important role in guidingextracorporeal fluid removal therapies and enhancingtheir safety.65,66 Several online hematocrit sensors (Crit-line; Hemametrics, Salt Lake City, UT; Hemoscan; Gam-bro, Lund, Sweden; and Dedyca, Bellco) permit contin-uous estimation of blood volume and relative bloodvolume changes during UF. The sensor can show whencritical thresholds are reached (5%-7%), allowing mod-ulation of the UF rate according to the speed of intravas-cular refilling.67 Measurement of bioelectrical impedancebefore UF can help to determine the fluid status of thepatient. Subsequent sequential measurements during UFcan contribute further to guide therapy, allowing estab-lishment of a target for extracorporeal fluid removal(Cardio-EFG; EFG, Dublin, Ireland) (Fig. 2).68 Resultsrom recent studies evaluated the sensitivity and speci-city of bioimpedance vector analysis (BIVA) in thessessment of fluid volume status in HF patients and used

he results in combination with BNP to establish criteria

NetTroe

ergeo).

102 M.R. Costanzo et al

for hospital discharge.69-71 Studies in overhydrated pa-tients have confirmed the reliability of this parameter totarget a patient’s dry weight.72 Extracorporeal UF shouldin fact remove fluid to improve symptoms of congestionwhile maintaining renal and tissue blood flow. Toachieve this goal the amount and rate of fluid removalmust be clearly established. If UF rates are too high,hemodynamic instability occurs because the refilling ofthe intravascular space from the interstitium cannot keeppace with the reduction in intravascular volume. LowerUF rates, such as those that can be achieved with SCUFor CVVH, allow gradual intravascular refilling from theinterstitial space without inducing intravascular volumedepletion (Fig. 3).73-75 Clinical experience suggests thatextracorporeal fluid removal is better tolerated when con-ducted with low UF rates over a prolonged period oftime.76-78 In conclusion, the absolute amount to be re-moved and the rate of UF necessary to achieve dryweight with minimal or absent hemodynamic instabilitymust be prescribed. The combination of body weight andbalance, blood pressure, biomarkers, blood volume, and

Figure 1. Recent equipment specifically designed for SCUF (Aqumachine is called CARPEDIEM (Cardio-Renal Pediatric Dialysis EmResearch Institute of Vicenza for neonates and small children (Bellc

BIVA may permit optimization of the fluid status of the

HF patient, suggesting the overall amount of fluid to beremoved and the speed of fluid removal.

In Figures 4 and 5 we show the example of the use ofGAL as an early marker for detection of AKI and the

valuation of the NGAL curve to guide UF in conjunc-ion with BNP, blood volume, blood pressure, and BIVA.he implementation of this biomarker curve in clinical

outines might be the key factor in preventing the devel-pment of iatrogenic CRS type 1 induced by too fast orxcessive UF.

UF IN HF: CLINICAL RESULTS

Studies of Hemofiltration

Extracorporeal fluid removal has been used to treat con-gestion in HF patients for almost 4 decades.79 The inter-pretation of the results of studies conducted in the 1970sand 1980s is hindered by the small patient numbers, theretrospective or observational nature, and lack of a con-trol group.79-85 Background therapies did not include theneurohormonal blockers that are the mainstay of contem-

System 100, CHF Solutions, Inc, and Dedyca, Bellco). The thirdncy Machine) and it has been designed by the International Renal

adex

porary optimal medical therapy for HF. Diuretic types,

cuid

epeattegie

Extracorporeal ultrafiltration in heart failure 103

route of administration, and continuous versus intermit-tent extracorporeal therapies were variable.79-85 The def-inition of diuretic resistance was inconsistent. The vari-ables used to assess response to therapy differed betweenstudies. Often renal function was not assessed. Follow-upevaluation generally was short, thus preventing evalua-tion of the effects of fluid removal on long-term out-comes.79-85 Duration of extracorporeal fluid removalranged from a single, few-hour treatment to daily therapyrepeated over prolonged periods of time. Differentpumps, membranes, and types of therapy were used.Different blood flows (200-400 mL/min) and fluid re-moval rates (200-1,200 mL/h) were used for UF, accom-plished with either hemofiltration or continuous arterio-

Figure 2. Description of the simple and easy-to-use technology foris small and light and it can be carried and moved easily. The resultsallowing the construction of a vector perfectly describing the stainterpretation a hydration scale is reported by the printed module. Rto assess effects of therapy or even to guide fluid management stra

venous hemofiltration (CAVH).79-85 However, these early

studies yielded some consistent and valuable findings.The subjects generally had advanced HF of differentetiologies and severe renal impairment with refractoryfluid overload. However, extracorporeal fluid removalconsistently was associated with substantial weight loss(3.6-11.5 kg), improvement of congestion, and increaseddiuresis at lower diuretics doses.79-85

However, overly aggressive UF in ADHF patients canonvert nonoliguric renal dysfunction into oliguric fail-re by increasing neurohormonal activation and decreas-ng renal perfusion pressure (RPP), which may lead toialysis dependence.86-88 HF patients with mild renal

impairment may benefit from isolated UF whereas thosewith moderate to severe renal dysfunction also require

pedance vector analysis in patients with heart failure. The machined by the measurements are reactance, resistance, and phase angle,f hydration and the nutritional condition of the patient. For easyed measurement can offer a trend owing to vector migration usables.

bioimofferetus o

blood cleansing. In one study 36 New York Heart Asso-

ar1g(au4o

n bey

104 M.R. Costanzo et al

ciation classes II and III patients were assigned randomlyto either receive hemofiltration or to a control group: onlyhemofiltration-treated patients had a significant reductionin interstitial lung water associated with rapid improve-ment of dyspnea, cardiac function, pulmonary gas ex-change, radiologic signs of congestion, peripheral edema,ascites, and pleural and pericardial effusions.89 Hemofil-tration simultaneously reduced cardiac filling pressuresand optimized circulating blood volume without affect-ing heart rate, blood pressure, cardiac output, and sys-temic vascular resistance. During hemofiltration the cir-culating volume—the true cardiac preload—is preservedby fluid refilling it from the interstitium. The paralleldecrease in right and left ventricular filling pressuresreflects the reduction of intrathoracic pressure and theincrease in lung compliance owing to the reabsorption ofthe extravascular fluid. Similar effects occurred in 24stable HF patients with radiologic pulmonary conges-tion.90 Compared with 12 controls, only the 12 patients

Rela

�ve

Chan

ges

Blood VolumeBlood Volume

Figure 3. Typical hemodynamic behavior of a patient undergoing eowing to two main causes: (1) the occurrence of an � angle in the bloa sudden decrease in circulating blood volume (this is managed within the trajectory depending on excessive UF and patient dehydratiomeasurements (lower panels). ECFV, extracellular fluid volume.

who had 2 L of fluid removed by hemofiltration had t

improvement in left ventricular diastolic function and asustained recovery of diuretic responsiveness, one of theeffects responsible for the long-term clinical improve-ment observed after hemofiltration. The mechanisms ofdiuresis restoration were explored further in 32 HF pa-tients (New York Heart Association classes II-IV) withdifferent degrees of fluid overload in whom neurohor-mones and RPP were measured while fluid was removedby hemofiltration (500 mL/h) to achieve a 50% reductionin right atrial pressure.87 Response to fluid removal sep-rated the patients into three groups: in 10 patients withefractory congestion and a daily urine output of less than,000 mL/h, hemofiltration was associated with a 40% orreater reduction in norepinephrine, plasma renin activityPRA), and aldosterone levels, a 16% increase in RPP,nd a 493% increase in diuresis; in 9 patients with a dailyrine output of greater than 1,000 mL PRA increased by0% and RPP declined by 12%; in 13 patients withoutbvious congestion and a daily urine output of greater

1

orporeal ultrafiltration is reported. Blood pressure instability may beolume trajectory caused by rapid UF exceeding refilling capacity anduced speed of fluid removal), and (2) the occurrence of an �1 angleond adequate levels. This can be documented by sequential BIVA

xtracod va red

han 1,000 mL, neurohormone levels increased by 50%

uifi

6

hp(t

saeapasmnt

Extracorporeal ultrafiltration in heart failure 105

or more, and RPP and urine output declined, respectively,by 7% and 45%. There was a significant inverse corre-lation (P � .0001) between neurohormone level anddiuresis.87 Thus, depending on volume status, the re-sponse to hemofiltration may vary from neurohormonalactivation and oliguria to neurohormonal down-regula-tion and potentiation of sodium and water excretion. Thefavorable hemodynamic, neurohormonal, and ventilatoryresponses produced by hemofiltration in 8 clinically sta-ble HF patients were not observed in 8 matched subjectstreated with IV furosemide to achieve equivalent fluidremoval.87 In the latter group, norepinephrine, PRA, andaldosterone concentrations remained increased greaterthan pretreatment values for several days. In contrast, inhemofiltration-treated patients a sustained decrease inneurohormonal levels began 48 hours after therapy andstill was detectable at 90 days.87 Pulmonary congestionand functional measures were improved significantly at48 hours only in the hemofiltration. Because hemofiltra-tion and isolated UF remove fluid isotonic with plasma,approximately 150 mmoL of sodium are removed witheach liter of ultrafiltrate. In contrast, the sodium presentin the urine of furosemide-treated HF patients rarelyexceeds 100 mmoL. The different amount of sodiumremoved in a similar fluid volume may account for thediverging neurohumoral responses elicited by the hemo-filtration compared with IV furosemide.91 In patients withbaseline VO2 below 18.5 mL/min/kg, improvement infunctional capacity after hemofiltration persisted at 90days.91 These benefits can be ascribed to a reduction inlung water, which reduces lung stiffness, as indicated bythe concomitant improvement in spirometry values at restand lung mechanics during exercise. The latter also wasassociated with a reduction in heart size and in restingand exercise Doppler indices of hemodynamic restric-tion.91

These results were obtained with hemofiltration in a

Figure 4. Didactic representation of a time course of BNP, UFvolume, and creatinine over several days after hospitalization forADHF and the beginning of extracorporeal UF.

small number of subjects housed in the intensive care

nit. The use of UF for ADHF has become the focus ofncreased interest owing to the introduction of simpli-ed portable devices.86-91

Studies of Simplified Isolated Ultrafiltration

In 21 fluid-overloaded HF patients, removal of an aver-age of 2,611 � 1,002 mL (range, 325-3,725 mL) over.43 � 1.47 hours reduced weight from 91.9 � 17.5 kg

to 89.3 � 17.3 kg (P � .0001), and improved congestionwithout changes in heart rate, blood pressure, electrolytelevel, or hematocrit.11 UF was initiated within 4.7 � 3.5ours of hospitalization and before IV diuretics in 20 HFatients with volume overload and diuretic resistanceage, 74.5 � 8.2 y; 75% ischemic disease; ejection frac-ion, 3% � 15%), and continued until euvolemia.92 An

average of 8,654 � 4,205 mL were removed with 2.6 �1.2 eight-hour UF courses. Twelve patients (60%) weredischarged in 3 days or fewer. One patient was re-admitted in 30 days and two patients were re-admitted in90 days. Weight (P � .006), Minnesota Living withHeart Failure scores (P � .003), and Global Assessment(P � .00003) were improved after UF at 30 and 90 days.Levels of BNP were decreased after UF (from 1,236 �747 pg/mL to 988 � 847 pg/mL) and at 30 days (816 �494 pg/mL) (P � .03). Blood pressure, renal function,and medications were unchanged.92 The results of thistudy suggest that in HF patients with volume overloadnd diuretic resistance, early UF before IV diureticsffectively and safely decreases length of stay and re-dmissions and is associated with clinical benefit stillresent at 90 days. Compared with 20 subjects randomlyssigned to IV diuretics, 20 patients randomized to aingle, 8-hour UF session had greater median fluid re-oval (2,838 versus 4,650 mL; P � .001).93 There were

o adverse hemodynamic or renal effects. The results ofhis study show that an early UF strategy in ADHF

Figure 5. Didactic representation of a time course of BNP, UFvolume, and creatinine over several days after hospitalization for

ADHF and the beginning of extracorporeal UF. The management ismodified based on the NGAL curve.

emfl

dw

fh

ittc

t

106 M.R. Costanzo et al

patients results in greater fluid removal and improvementof congestion than attained with traditional diuretic ther-apies.

The Ultrafiltration vs. Intravenous Diuretics for Pa-tients Hospitalized for Acute Decompensated Heart Fail-ure (UNLOAD) trial was aimed at comparing the safetyand efficacy of veno-venous UF and standard IV diuretictherapy for ADHF patients with 2 or more signs ofhypervolemia.94 Two hundred patients (63 � 15 y; 69%men; 71% ejection fraction, �40%) were randomized toUF or IV diuretics. At 48 hours, weight (5.0 � 3.1 versus3.1 � 3.5 kg; P � .001) and net fluid loss (4.6 versus 3.3L; P � .001) were greater in the UF group.94 Dyspneascores were similarly improved in the two groups (Fig.2). At 90 days, the UF group had fewer patients rehos-pitalized for HF (16 of 89 [18%] versus 28 of 87 [32%];P � .037), HF rehospitalizations (0.22 � 0.54 versus0.46 � 0.76; P � .022), rehospitalization days per patient(1.4 � 4.2 versus 3.8 � 8.5; P � .022), and unscheduledvisits (14 of 65 [21%] versus 29 of 66 [44%]; P � .009).Changes in serum creatinine were similar in the 2 groupsthroughout the study. The percentage of patients withincreases in serum creatinine levels greater than 0.3mg/dL was similar in the UF and standard care group at24 hours (13 of 90 [14.4%] versus 7 of 91 [7.7%]; P �.528), at 48 hours (18 of 68 [26.5%] versus 15 of 74[20.3%]; P � .430), and at discharge (19 of 84 [22.6%]versus 17 of 86 [19.8%]; P � .709).94 Serum potassiumlevel less than 3.5 mEq/L occurred in 1 of 77 (1%)patients in the UF and in 9 of 75 (12%) patients in thediuretic group (P � .018). Episodes of hypotension dur-ing 48 hours after randomization were similar (4 of 100[4%] versus 3 of 100 [3%]).94,95 Thus, the UNLOAD trialshowed that in ADHF, UF safely produces greater weightand fluid loss than IV diuretics, reduces 90-day HFrehospitalizations, and is an effective alternative therapy.Despite the lack of statistical difference in weight andfluid loss by UF and continuous infusion IV diuretics,patients treated with UF had fewer rehospitalizations andunscheduled HF office or emergency department visits.These findings support the hypothesis that removal ofisotonic fluid by UF, rather than hypotonic urine by IVdiuretics, may contribute to the prolongation of the clin-ical benefits of UF. Among 15 ADHF patients treatedfirst with a furosemide IV bolus and then with UF be-cause of persistent congestion, sodium concentration inthe ultrafiltrate sampled after 8 hours of therapy wassignificantly higher than that in the urine after the IVfurosemide bolus (134 � 8.0 versus 60 � 47 mmol/L;P � .000025).96 Although the sodium concentration ofthe ultrafiltrate was similar in all 15 patients, urinarysodium concentration after IV furosemide was highlyvariable, ranging from less than 20 mmol/L to 100mmol/L in 13 (87%) of the subjects.96

Volume overload in HF patients inevitably is relatedto an increase and abnormal distribution of total body

sodium. A treatment strategy that is simultaneously more

ffective in reducing total body sodium and excess fluiday improve outcomes more than removal of hypotonicuid by diuretics or free water by vasopressin V2-recep-

tor blockers.95,97 It also is possible that prehospitalizationiuretic use itself may reduce the natriuresis achievableith IV loop diuretics.98 Notably, in the study by Ali et

al,96 urine potassium concentration in response to IVurosemide was higher than that in the ultrafiltrate after 8ours of therapy (41 � 23 versus 3.7 � 0.6 mmol/L; P �

.000017). The avoidance of electrolyte abnormalities alsomay account for the improved outcomes associated withUF. In the UF, IV bolus, and continuous infusion diureticgroups there was no correlation between net fluid lossduring hospitalization and the number of times patientswere rehospitalized for HF. Thus, the composition of thefluid removed may have a greater effect than its quantityin improving outcomes of congested HF patients.99

Edema in fluid-overloaded HF patients is isotonic, andtherefore eunatremic patients with edema have apprecia-ble total body sodium excess.100 Thus, loop diuretic–nduced diuresis of hypotonic fluid will reduce excessotal body water while only partially reducing excessotal body sodium. This may explain the recurrence ofongestion in patients treated with IV loop diuretics.101

A small substudy of 19 UNLOAD subjects showedhat diuretics and UF had similar renal outcomes.102

Quantitative measures of renal blood flow, glomerularfiltration rate, and filtration fraction did not differ be-tween treatments, suggesting neither renal benefit norharm of UF compared with conventional diuretics.

The economic impact of UF as an initial strategy fordecompensated HF was not addressed in the UNLOADtrial. Although diuretic therapy is less expensive thanextracorporeal fluid removal, it may not be the most costeffective if it is associated with unacceptably high rehos-pitalization rates. On the other hand, although the cost ofa filter is approximately $900, a single filter can be usedfor prolonged periods of time (�72 h). Furthermore, asUF therapy gains greater acceptance, the price of thefilter may be expected to decline. In addition, simplifiedUF can be performed in a cardiac telemetry unit. Thus,although the costs associated with UF during the indexhospitalization may exceed those of IV diuretics, totalcost over time may be lower because of decreased re-source use for HF.

CLINICAL USE OF UF IN HF: CURRENT STATUS,LIMITATIONS, AND REMAINING QUESTIONS

European and North American practice guidelines statethat UF is reasonable for patients with refractory conges-tion not responding to medical therapy and assign to thisrecommendation a class IIa, level of evidence: B.103-105 Inpatients with severe renal dysfunction or edema refrac-tory to standard care, UF or HFL may be needed toachieve adequate control of fluid retention.103-105 Theguidelines further acknowledge that in these patients UF

can produce clinical benefits, may restore responsiveness

cbsU

Extracorporeal ultrafiltration in heart failure 107

to conventional doses of loop diuretics, and correct hy-ponatremia.103-105 Neither practice guidelines nor clinicaltrial data provide guidance as to which clinical variablesshould trigger initiation of extracorporeal therapies. TheUNLOAD trial tested a strategy of early UF and showedthat this approach is associated with greater fluid loss andreduced 90-day hospitalizations compared with standardcare with IV diuretics. A frequently asked but still unan-swered question is which degree of congestion should betreated with UF rather than IV diuretics. A recent con-sensus statement proposed that congestion be gradedaccording to a combination of clinical and laboratoryparameters.106 The expert consensus suggests that a con-gestion grade higher than 12 together with low urineoutput (�1,000 mL/24 h) should trigger the use of ex-tracorporeal fluid removal. Another important question isthe degree of renal impairment that requires bloodcleansing in addition to UF. In the absence of data fromcontrolled clinical trials, clinicians often have used UFalone in patients with serum creatinine levels of 3 mg/dLor less and CVVH in those with serum creatinine levelsgreater than 3 mg/dL.94 Remaining questions also includethe amount and rate of fluid removal, and determinationof when euvolemia has been achieved. Unfortunately,none of the currently available methods of volume statusassessment, including biochemical markers, echocardio-graphic measurement of vena cave diameter, bioimped-ance, and blood volume monitoring, can guide mechan-ical fluid removal accurately.43 A frequently usedpractical approach is to estimate fluid excess by compar-ing the patient’s current weight with that measured in theabsence of signs and symptoms of congestion, and re-move at least 50% to 60% of this excess fluid without

Figure 6. A wearable/portable UF system for the treatment of ovecongestive heart failure. The system consists of a vest in which a cwaste bags are integrated. This prototype has been created in the In

Rand (Mirandola Modena) and Dainese (Vicenza, Italy). The system is intuse 24 h/d.

ausing worsening renal function or hemodynamic insta-ility. Although the amount of urine produced in re-ponse to IV diuretics is not predictable, fluid removal byF is completely controllable and adjustable.107 As de-

scribed earlier, online hematocrit sensors available incontemporary UF systems permit continuous estimationof blood volume. These devices can be programmed sothat fluid removal is stopped if the increase in hematocritexceeds the threshold set by the treating physician (3%-7%) and resumed when the hematocrit value decreases toless than the prespecified limit, which indicates that ad-equate refilling of the intravascular volume from theinterstitial space has occurred.

Patients should not be considered for UF if the fol-lowing conditions exist: venous access cannot be ob-tained; there is a hypercoagulable state; systolic bloodpressure is less than 85 mm Hg or there are signs orsymptoms of cardiogenic shock; patients require intrave-nous pressors to maintain an adequate blood pressure; orthere is end-stage renal disease, as documented by arequirement for dialysis approaches. UF can be per-formed in patients with hematocrit levels greater than40% only if it can be proven that hypovolemia is absent.

Of the UF approaches described in this review, themost practical are veno-venous UF techniques in whichisotonic plasma is propelled through the filter by anextracorporeal pump. These approaches avoid an arterialpuncture, remove a predictable amount of fluid, and arenot associated with significant hemodynamic instability.

To date, 60% of approximately 33,000 UF treatmentsperformed with the Aquadex System 100 (CHF Solu-tions, Inc) have been conducted with peripheral venousaccess. The conversion from a peripheral to a central

ation in patients with oliguria refractory to diuretics, AKI, CKD, andete extracorporeal circuit, blood and UF pumps, batteries, and fluidtional Renal Research Institute of Vicenza and in the laboratories of

rhydromplterna

ended to be used for ambulatory UF from a few hours to continuous

108 M.R. Costanzo et al

venous access is uncommon, with rates lower than 5%(data on file, CHF Solutions, Inc). Traditionally, a muchhigher percentage of patients were treated with centralvenous catheters and in some cases they still are.108

Nevertheless, simplified isolated veno-venous UF ap-pears to be safe. According to the Manufacturer and UserFacility Device Experience Food and Drug Administra-tion website, 8 adverse events, all unrelated to the device,have been reported to date in approximately 33,000 treat-ments conducted with the Aquadex System 100 (CHFSolutions, Inc). The reason for the paucity of adverseevents is the presence of safety features in contemporaryUF devices. Air emboli are prevented by the air detectorsensor and alarm. Line disconnection or pump or tubingmalfunctions are detected by the pressure sensors, whichstop the therapy until the mechanical problem has beenaddressed and corrected. Hemolysis and hemorrhage areless than 1%, as found in the UNLOAD study.94

EXPECTATIONS FOR THE FUTURE

The number of patients with advanced HF will continueto increase because of the aging of the population, theimproved outcomes after acute cardiac diseases, and theincreased survival of patients with chronic HF from ad-vances in medical and device therapy. Therefore, a grow-ing number of patients with HF will develop hypervol-emia refractory to conventional diuretic therapy. Clinicaltrials have shown that UF is a viable option to treathypervolemia in ADHF patients. In the future it is pos-sible that these extracorporeal therapies may be movedfrom the hospital to the home setting using transportableor wearable technologies.109,110 Patients could have ahome-based system for BIVA and biomarkers measure-ments. Patients may undergo placement of an in-dwellingdual-lumen catheter that easily can be connected with theUF device. Therapy then could be performed one or moretimes weekly to maintain euvolemia and prevent rehos-pitalizations resulting from recurrent ADHF. Preliminaryclinical studies have shown the feasibility of using wear-able, battery-powered, miniaturized UF systems.109,110

The first human feasibility study with the prototype of awearable UF device has yielded encouraging results.110

Six volume-overloaded patients were treated for 6 hourswith a wearable UF system. Blood flow was approxi-mately 116 mL/min and UF rates ranged between 120and 288 mL/h. With these settings an average of 151mmol of sodium was removed with each treatment. Nopatient had symptoms and signs of hemodynamic insta-bility because the device removes fluid continuously atrates slower than those of intermittent therapies. A newvest for ambulatory UF was generated in the laboratoriesof the International Renal Research Institute of Vicenza(Fig. 6). Moving from the experimental phase to a routineapplication of this technology would transform UF in HFpatients from an emergency-based rescue therapy to ascheduled elective therapy to prevent frequent rehospi-

talizations.

CONCLUSIONS

Despite the nearly universal use of loop diuretics, deathand rehospitalization rates for ADHF patients remainunacceptably high. The addition of inotropes, vasodila-tors, NPs, adenosine antagonists, arginine vasopressinantagonists, and invasive hemodynamic monitoring donot appear to improve outcomes in this patient popula-tion. For more than a century sodium has been recog-nized as the major determinant of extracellular fluidvolume in HF. Therefore, excess total body sodiumshould be the principal target for the therapy of patientshospitalized for worsening symptoms and signs of con-gestion. Loop diuretics, although often effective in re-moving water, are inherently incapable of consistentlyreducing total body sodium. In contrast, because of itsmechanism, UF predictably reduces total body sodium.UF and diuretic holiday may restore diuresis and natri-uresis. Because patients with ADHF are not a uniformpatient population, their therapy should be individual-ized. A trial of diuretic therapy is warranted if patientsare diuretic naive or have minimal diuretic resistance, asevidenced by low doses of outpatient diuretics, and anormal or minimally reduced serum creatinine clearance,particularly if the patient is only mildly or moderatelyvolume overloaded. In diuretic-resistant, severely fluid-overloaded HF patients, UF may have greater efficacyand cost effectiveness because it extracts the greatestamount of sodium per milliliter of fluid removed and it isassociated with fewer rehospitalizations than usual di-uretic therapy at 90 days.

Many questions regarding the use of UF in HF patientsremain unanswered and must be addressed in futurestudies. These include optimal fluid removal rates inindividual patients, effects of UF on cardiac remodeling,influence of low oncotic pressure in cachectic patients onplasma refill rates, and the economic impact of UF todetermine whether the expense of disposable filters isoffset by the cost savings resulting from reduced rehos-pitalization rates. The ongoing National Heart, Lung, andBlood Institute–sponsored Cardio-Renal Rescue Study inAcute Decompensated Heart Failure trial (NCT00608491) will shed further light on the effects of UF inADHF patients with refractory congestion who developworsening renal function with conventional diuretic ther-apies. Future studies should evaluate the safety and effi-cacy of wearable devices that permit extracorporeal fluidremoval in the ambulatory setting.

REFERENCES1. Lloyd-Jones D, Adams RJ, Brown TM, et al. Heart disease and

stroke statistics—2010 update. A report from the AmericanHeart Association statistics. Circulation. 2010;121:e46-215.

2. Adams KF, Fonarow GC, Emerman CL, et al. Characteristicsand outcomes of patients hospitalized for heart failure in theUnited States: rationale, design, and preliminary observationsfrom the first 100,000 cases in the Acute Decompensated HeartFailure National Registry (ADHERE). Am Heart J. 2005;149:

209-16.

Extracorporeal ultrafiltration in heart failure 109

3. Jain P, Massie BM, Gattis WA, et al. Current medical treatmentfor exacerbation of chronic heart failure resulting in hospitaliza-tion. Am Heart J. 2003;145:S3-17.

4. Drazner MH, Rame JE, Stevenson LW, Dries DL. Prognosticimportance of elevated jugular venous pressure and a third heartsound in patients with heart failure. N Engl J Med. 2001;345:574-81.

5. Androne AS, Katz SD, Lund L, et al. Hemodilution is commonin patients with advanced heart failure. Circulation. 2003;107:226-9.

6. Heywood JT, Fonarow GC, Costanzo MR, Mathur VS,Wigneswaran JR, Wynne J. High prevalence of renal dysfunc-tion and its impact on outcome in 118,465 patients hospitalizedwith acute decompensated heart failure: a report from the AD-HERE database. J Card Fail. 2007;13:422-30.

7. Ronco C, Haapio M, House A, Anavekar N, Bellomo R. Cardio-Renal syndrome. J Am Coll Cardiol. 2008;52:1527-39.

8. Felker GM, Lee KL, Bull DA, et al. Diuretic strategies evalua-tion in patients with acute decompensated heart failure. N EnglJ Med. 2011;364:797-805.

9. Silverstein ME, Ford CA, Lysaght MJ, Henderson LW. Treat-ment of severe fluid overload by ultrafiltration. N Engl J Med.1974;291:747-51.

10. Henderson LW, Lilley JJ, Ford CA, Stone RA. Hemodiafiltra-tion. J Dial. 1977;1:211-38.

11. Nalesso F, Garzotto F, Ronco C. Technical aspects of extracor-poreal ultrafiltration: mechanisms, monitoring and dedicatedtechnology. Contrib Nephrol. 2010;164:199-208.

12. Ricci Z, Bonello M, Salvatori G, Ratanarat R, Brendolan A, DanM, et al. Continuous renal replacement technology: from adap-tive technology and early dedicated machines towards flexiblemultipurpose machine platforms. Blood Purif. 2004;22:269-76.

13. Jaski BE, Ha J, Denys BG, Lamba S, Trupp RJ, Abraham WT.Peripherally inserted veno-venous ultrafiltration for rapid treat-ment of volume overloaded patients. J Card Fail. 2003;9:227-31.

14. Ronco C, Levin NW. Mechanisms of solute transport in extra-corporeal therapies. Contrib Nephrol. 2005;149:10-7.

15. Ronco C, Ricci Z, Brendolan A, Bellomo R, Bedogni F. Ultra-filtration in patients with hypervolemia and congestive heartfailure. Blood Purif. 2004;22:150-63.

16. Ronco C, Clark WR. Factors affecting hemodialysis and perito-neal dialysis efficiency. Semin Dialysis. 2001;14:257-62.

17. Ronco C, Ballestri M, Cappelli G. Dialysis membranes in con-vective treatments. Nephrol Dial Transplant. 2000;15 Suppl2:31-6.

18. Ronco C, Breuer B, Bowry SK. Hemodialysis membranes forhigh-volume hemodialytic therapies: the application of nano-technology. Hemodial Int. 2006;10:S48-50.

19. Ronco C, Crepaldi C, Brendolan A, Bragantini L, D’Intini V,Inguaggiato P, et al. Evolution of synthetic membranes for bloodpurification: the case of the Polyflux family. Nephrol Dial Trans-plant. 2003;18:10-20.

20. Clark WR, Gao D, Ronco C. Membranes for dialysis. Compo-sition, structure and function. Contrib Nephrol. 2002;137:70-7.

21. Ronco C. Hemodiafiltration: evolution of a technique towardsbetter dialysis care. Contrib Nephrol. 2011;168:19-27.

22. Ronco C, Brendolan A, Crepaldi C, Rodighiero M, Scabardi M.Blood and dialysate flow distributions in hollow-fiber hemodia-lyzers analyzed by computerized helical scanning technique.J Am Soc Nephrol. 2002;13:S53-61.

23. Ronco C, Bowry SK, Brendolan A, Crepaldi C, Soffiati G,Fortunato A, et al. Hemodialyzer: from macro-design to mem-brane nanostructure; the case of the FX-class of hemodialyzers.Kidney Int. 2002;61:S126-42.

24. Clark WR, Hamburger RJ, Lysaght MJ. Effect of membranecomposition and structure on solute removal and biocompatibil-ity in hemodialysis. Kidney Int. 1999;56:2005-15.

25. Ronco C, Ghezzi PM, Metry G, Spittle M, Brendolan A,

Rodighiero MP, et al. Effects of hematocrit and blood flow

distribution on solute clearance in hollow-fiber hemodialyzers.Nephron. 2001;89:243-50.

26. Kim JC, Kim JH, Sung J, Kim HC, Kang E, Lee SH, et al.Effects of arterial port design on blood flow distribution inhemodialyzers. Blood Purif. 2009;28:260-7.

27. Ronco C, Brendolan A, Lupi A, Metry G. Effects of a reducedinner diameter of hollow fibers in hemodialyzers. Kidney Int.2000;58:809-17.

28. Fiore G, Guadagni G, Lupi A, Ricci Z, Ronco C. A newsemiempirical mathematical model for prediction of internalfiltration in hollow fiber hemodialyzers. Blood Purif. 2006;24:555-68.

29. Ethier J, Mendelssohn DC, Elder SJ, Hasegawa T, Akizawa T,Akiba T, et al. Vascular access use and outcomes: an interna-tional perspective from the Dialysis Outcomes and PracticePatterns Study. Nephrol Dial Transplant. 2008;23:3219-26.

30. Bonello M, Levin NW, Ronco C. History and evolution of thevascular access for hemodialysis. Contrib Nephrol. 2004;142:1-13.

31. De Maria E, Pignatti F, Patrizi G, Benenati PM, Ricci S, Cap-pelli S. Ultrafiltration for the treatment of diuretic-resistant,recurrent, acute decompensated heart failure: experience in asingle center. J Cardiovasc Med (Hagerstown). 2010;11:599-604.

32. Kroop IG. America’s initial application of the Kolff artificialkidney. Hemodial Int. 2010;14:346-7.

33. Bellomo R, Ronco C, Metha R. Nomenclature for continuousrenal replacement therapies. Am J Kidney Dis. 1996;28:S2-7.

34. Mehrotra R, Bellomo R, Ronco C. Ultrafiltration in congestiveheart failure. Cardiology. 2001;96:121-6.

35. Clark WR, Paganini E, Weinstein D, Bartlett R, Sheinfeld G,Ronco C. Extracorporeal ultrafiltration for acute exacerbation ofchronic heart failure: report from the acute dialysis qualityinitiative. Int J Artif Organs. 2005;28:466-76.

36. Ronco C, Bellomo R. Continuous renal replacement therapies:the need for a standard nomenclature. Contrib Nephrol. 1995;116:28-33.

37. Ronco C, Bellomo R. Principles of solute clearance duringcontinuous renal replacement therapy. In: Ronco C, Bellomo R,editors. Critical care nephrology. Kluwer Academic Publishers;1998. p. 1213-22.

38. Clark WR, Macias WL, Molitoris BA, et al. Plasma proteinadsorption to highly permeable hemodialysis membranes. Kid-ney Int. 1995;48:481-8.

39. Ward RA, Ronco C. Improvements in technology: a path to saferand more effective hemodialysis. Blood Purif. 2009;27:6-10.

40. Boyle A Ultrafiltration for acute decompensated heart failure.Expert Rev Med Devices. 2005;2:689-97.

41. Dahle TG, Sobotka PA, Boyle AJ. A practical guide to ultrafil-tration in acute decompensated heart failure. Congest Heart Fail.2008;14:83-8.

42. Cerda J, Sheinfeld G, Ronco C. Fluid overload in critically illpatients with acute kidney injury. Blood Purif. 2010;29:11-8.

43. Jaeger JQ, Mehta RL. Assessment of dry weight in hemodialy-sis: an overview. J Am Soc Nephrol. 1999;10:392-403.

44. Mehta RL. Fluid management in CRRT. Contrib Nephrol. 2001;132:335-48.

45. Maisel AS, Krishnaswamy P, Nowak RM, et al. Rapid measure-ment of B-type natriuretic peptide in the emergency diagnosis ofheart failure. N Engl J Med. 2002;347:161-7.

46. Maisel A, Hollander JE, Guss D, et al. Primary results of therapid emergency department heart failure outpatient trial (RED-HOT): a multicenter study of B-type natriuretic peptide levels,emergency department decision making, and outcomes in pa-tients presenting with shortness of breath. J Am Coll Cardiol.2004;44:1328-33.

47. Mueller C, Schoeler A, Laule-Kilian K, et al. Use of B-typenatriuretic peptide in the evaluation and management of acute

dyspnea. N Engl J Med. 2004;350:647-54.

110 M.R. Costanzo et al

48. Januzzi JL Jr, Camargo CA, Anwaruddin S, et al. The N-termi-nal pro-BNP investigation of dyspnea in the emergency depart-ment (PRIDE) study. J Am Coll Cardiol. 2005;95:948-54.

49. Maisel AS. Use of BNP levels in monitoring hospitalizedheart failure patients with heart failure. Heart Fail Rev. 2003;8:339-44.

50. O’Hanlon R, O’Shea P, Ledwidge M, O’Loughlin C, Lange S,Conlon C, et al. The biologic variability of B-type natriureticpeptide and N-terminal pro-B-type natriuretic peptide in stableheart failure patients. J Card Fail. 2007;13:50-5.

51. Rehman SU, Januzzi JL Jr. Natriuretic peptide guided heartfailure management. Compr Ther. 2008;34:32-40.

52. Iwanaga Y, Nishi I, Furuichi S, Noguchi T, Sase K, Kihara Y, etal. B-type natriuretic peptide strongly reflects diastolic wallstress in patients with chronic heart failure: comparison betweensystolic and diastolic heart failure. J Am Coll Cardiol. 2006;47:742-8.

53. Troughton RW, Prior DL, Pereira JJ, et al. Plasma B-typenatriuretic peptide levels in systolic heart failure: importance ofleft ventricular diastolic function and right ventricular systolicfunction. J Am Coll Cardiol. 2004;43:416-22.

54. Kazanegra R, Cheng V, Garcia A, Krishnaswamy P, Gardetto N,Clopton P, et al. A rapid test for B-type natriuretic peptidecorrelates with falling wedge pressures in patients treated fordecompensated heart failure: a pilot study. J Card Fail. 2001;7:21-9. 5.

55. Bettencourt P, Ferreira S, Azevedo A, et al. Preliminary data onthe potential usefulness of B-type natriuretic peptide levels inpredicting outcome after hospital discharge in patients with heartfailure. Am J Med. 2002;113:215-9.

56. Cioffi G, Tarantini L, Stefenelli C, et al. Changes in plasmaN-terminal proBNP levels and ventricular filling pressures dur-ing intensive unloading therapy in elderly with decompensatedcongestive heart failure and preserved left ventricular systolicfunction. J Card Fail. 2006;12:608-15.

57. Di Somma S, Mgrini L, Mazzone M, et al. Decrease in NT-proBNP plasma levels indicates clinical improvement of acutedecompensated heart failure. Am J Emerg Med. 2007;25:335-9.

58. Yasumura Y, Miyatake K, Okamoto H, et al. Rationale for theuse of combination angiotensin-converting enzyme inhibitor andangiotensin II receptor blocker therapy in heart failure. Circ J.2004;68:361-6.

59. Anand IS, Fisher LD, Chiang YT, et al. Changes in brainnatriuretic peptide and norepinephrine over time and mortalityand morbidity in the Valsartan Heart Failure Trial (Val-HeFT).Circulation. 2003;107:1278-83.

60. Davis ME, Richards M, Nicholls MG, et al. Introduction ofmetoprolol increases plasma B-type cardiac natriuretic peptidesin mild, stable heart failure. Circulation. 2006;113:977-85.

61. Richards AM, Doughty R, Nicholls MG, et al. Plasma N-termi-nal pro-brain natriuretic peptide and adrenomedullin: prognosticutility and prediction of benefit from carvedilol in chronic isch-emic left ventricular dysfunction. Australia-New Zealand HeartFailure Group. J Am Coll Cardiol. 2001;37:1781-7.

62. Anwaruddin A, Lloyd-Jones DM, Baggish A, et al. Renal func-tion, congestive heart failure, and amino-terminal pro-brain na-triuretic peptide. J Am Coll Cardiol. 2006;48:1060-1.

63. McCullough PA, Duc P, Omland T, et al. B-type natriureticpeptide and renal function in the diagnosis of heart failure: ananalysis from the Breathing Not Properly Multinational Study.Am J Kidney Dis. 2003;41:571-9.

64. Tapolyai M, Uysal A, Maeweathers G, et al. B-type natriureticpeptide–directed ultrafiltration improves care in acutely hospi-talized dialysis patients. Congest Heart Fail. 2000;15:131-5.

65. Ronco C. Cardio-renal syndromes: from foggy bottoms to sunnyhills. Heart Fail Rev. [cited 2011 Jan 23]. PubMed PMID:21259069.

66. Haapio M, House AA, de Cal M, Cruz DN, Lentini P, Giavarina

D, et al. Heart-kidney biomarkers in patients undergoing cardiac

stress testing. Int J Nephrol. [cited 2010 Oct 20]. PubMed PMID:21151536; PubMed Central PMCID: PMC2989651.

67. Ronco C, Brendolan A, Bellomo R. On-line monitoring in con-tinuous renal replacement therapies. Kidney Int. 1999;56 Suppl72:S8-S14.

68. Piccoli A. Whole body—single frequency bioimpedance. Con-trib Nephrol. 2005;149:150-61.

69. Aspromonte N, Cruz DN, Valle R, Ronco C. Management andmonitoring of haemodynamic complications in acute heart fail-ure. Heart Fail Rev. [cited 2011 Feb 8]. PubMed PMID:21301959.

70. Valle R, Aspromonte N, Ronco C. Optimizing Fluid Manage-ment in patients with Acute Decompensated Heart Failure(ADHF): the emerging role of combined measurement of bodyhydration status and brain natriuretic peptide (BNP) levels. HeartFail. 2011. In press.

71. House AA, Haapio M, Lentini P, Bobek I, de Cal M, Cruz DN,et al. Volume assessment in mechanically ventilated critical carepatients using bioimpedance vectorial analysis, brain natriureticpeptide, and central venous pressure. Int J Nephrol. [cited 2010Dec 2]. PubMed PMID: 21151535; PubMed Central PMCID:PMC2997502.

72. Bonello M, House AA, Cruz D, Asuman Y, Andrikos E, PetrasD, et al. Integration of blood volume, blood pressure, heart rareand bioimpedance monitoring for the achievement of optimaldry body weight during chronic hemodialysis. Int J Artif Organs.2007;30:1098-108.

73. Ronco C, Bellomo R. Continuous versus intermittent renal re-placement therapy in the treatment of acute renal failure. Neph-rol Dial Transplant. 1998;13:79-85.

74. Bellomo R, Ronco C. Acute renal failure in the ICU: adequacyof dialysis and the case for continuous therapies. Nephrol DialTransplant. 1996;11:424-8.

75. Salvatori G, Ronco F, Bonello M, Bottero M. Management offluid overload in congestive heart failure: learning from a casereport. Int J Artif Organs. 2006;29:187-96.

76. Bellomo R, Palevsky PM, Bagshaw SM, Gibney N, McAlisterFA, Honore PM, et al. Recent trials in critical care nephrology.Contrib Nephrol. 2010;165:299-309.

77. RENAL Replacement Therapy Study Investigators, Bellomo R,Cass A, Cole L, Finfer S, Gallagher M, Lo S, et al. Intensity ofcontinuous renal-replacement therapy in critically ill patients.N Engl J Med. 2009;361:1627-38.

78. VA/NIH Acute Renal Failure Trial Network, Palevsky PM,Zhang JH, O’Connor TZ, Chertow GM, Crowley ST, Choud-hury D, et al. Intensity of renal support in critically ill patientswith acute kidney injury [Erratum in N Engl J Med. 2009;361:2391.]. N Engl J Med. 2008;359:7-20.

79. Asaba H, Bergstrom J, Furst P. et al. Treatment of diuretic-resistant fluid retention with ultrafiltration. Acta Med Scand.1978;204:145-9.

80. Fauchald P, Forfang K, Amlie J. An evaluation of ultrafiltrationas treatment of therapy-resistant cardiac edema. Acta MedScand. 1986;219:47-52.

81. Donato L, Biagini A, Contini C, et al. Treatment of end-stagecongestive heart failure by extracorporeal ultrafiltration. Am JCardiol. 1987;59:379-80.

82. Rimondini A, Cipolla CM, Della Bella P, et al. Hemofiltration asshort-term treatment for refractory congestive heart failure. Am JMed. 1987;83:43-8.

83. Di Leo M, Pacitti A, Bergerone S, et al. Ultrafiltration in thetreatment of refractory congestive heart failure. Clin Cardiol.1988;11:449-52.

84. Simpson IA, Simpson K, Rae AP, et al. Ultrafiltration in diureticresistant heart failure. Renal Fail. 1987;10:115-9.

85. Akiba T, Taniguchi K, Marumo F, Matsuda O, et al. Clinicalsignificance of renal hemodynamics in severe congestive heartfailure: responsiveness to ultrafiltration therapies. Jap Circ J.

1989;53:191-6.

Extracorporeal ultrafiltration in heart failure 111

86. Cipolla CM, Grazi S, Rimondini A, et al. Changes in circulatingnorepinephrine with hemofiltration in advanced heart failure.Am J Cardiol. 1990;66:987-94.

87. Marenzi G, Grazi S, Giraldi F, et al. Interrelation of humoralfactors, hemodynamics, and fluid and salt metabolism in con-gestive heart failure: effects of extracorporeal ultrafiltration.Am J Med. 1993;94:49-56.

88. Pepi M, Marenzi G, Agostoni PG, et al. Sustained cardiacdiastolic changes elicited by ultrafiltration in patients with mod-erate congestive heart failure: pathophysiological correlates. BrHeart J. 1993;70:135-40.

89. Agostoni PG, Marenzi GC, Pepi M, et al. Isolated ultrafiltrationin moderate congestive heart failure. J Am Coll Cardiol. 1993;21:424-31.

90. Marenzi CG, Lauri G, Assanelli E, et al. Circulatory response tofluid overload removal by extracorporeal ultrafiltration in refrac-tory congestive heart failure. J Am Coll Cardiol. 2001;38:963-8.

91. Agostoni PG, Marenzi GC, Lauri GF, et al. Sustained improve-ment in functional capacity after removal of body fluid withisolated ultrafiltration in chronic cardiac insufficiency: failure offurosemide to provide the same result. Am J Med. 1994;96:191-9.

92. Costanzo MR, Saltzberg M, O’Sullivan J, Sobotka P. Earlyultrafiltration in patients with decompensated heart failure anddiuretic resistance. J Am Coll Cardiol. 2005;4611:2047-51.

93. Bart BA, Boyle A, Bank AJ, et al. Randomized controlled trialof ultrafiltration versus usual care for hospitalized patients withheart failure: relief for acutely fluid overloaded patients withdecompensated congestive heart failure. J Am Coll Cardiol.2005;46:2043-6.

94. Costanzo MR, Guglin ME, Saltzberg MT, et al. Ultrafiltrationversus intravenous diuretics for patients hospitalized for acutedecompensated heart failure. J Am Coll Cardiol. 2007;49:675-83.

95. Costanzo MR, Saltzberg MT, Jessup M, et al. Ultrafiltration isassociated with fewer re-hospitalizations than continuous di-uretic infusion in patients with decompensated heart failure:results from UNLOAD. J Cardiac Fail. 2010;16:277-84.

96. Ali SS, Olinger CC, Sobotka PA, et al. Loop diuretics can causeclinical natriuretic failure: a prescription for volume expansion.Congest Heart Fail. 2009;15:1-4.

97. Konstam MA, Gheorghiade M, Burnett JC, et al for the Efficacyof Vasopressin Antagonism in Heart Failure Outcome Studywith Tolvaptan (EVEREST) Investigators. Effects of oraltolvaptan in patients hospitalized for worsening heart failure.

The EVEREST outcome trial. JAMA. 2007;297:1319-31.

98. Abdallah JG, Schreier RW, Edelstein C, et al. Loop diuretic infu-sion increases thiazide-sensitive Na�/Cl�-cotransporter abundance:role of aldosterone. J Am Soc Nephrol. 2001;12:1335-41.

99. He XR, Greenberg S, Briggs J, Schnermann J. Effects of furo-semide and verapamil on the NaCl dependency of macula densa-mediated secretion. Hypertension. 1995;26:137-42.

100. Nielsen S, Knepper M, Kwon T, et al. Regulation of waterbalance. In: Schrier RW, editor. Diseases of the kidney andurinary tract. 7th ed. New York: Lippincott Williams andWilkins; 2001. p. 109-34.

101. Fang J, Mensah GA, Croft JB, Keenan NL. Heart failure-relatedhospitalization in the U.S., 1979 to 2004. J Am Coll Cardiol.2008;52:428-34.

102. Rogers HL, Marshall J, Bock J, Dowling TC, Feller E, RobinsonS, et al. A randomized, controlled trial of the renal effects ofultrafiltration as compared to furosemide in patients with acutedecompensated heart failure. J Card Fail. 2008;14:1-5.

103. Jessup M, Abraham WT, Casey DE, et al. 2009 focused update:ACCF/AHA guidelines for the diagnosis and management ofheart failure in adults: a report of the American College ofCardiology Foundation/American Heart Association Task Forceon Practice Guidelines. Circulation. 2009;119:1977-2016.

104. Howlett JG, McKelvie RS, Costigan J, et al. The 2010 CanadianCardiovascular Society guidelines for the diagnosis and man-agement of heart failure update: heart failure in ethnic minoritypopulations, heart failure and pregnancy, disease management,and quality improvement/assurance programs. Can J Cardiol.2010;26:185-202.

105. Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC guidelinesfor the diagnosis and treatment of acute and chronic heart failure2008: the Task Force for the Diagnosis and Treatment of Acuteand Chronic Heart Failure 2008 of the European Society ofCardiology. Eur Heart J. 2008;29:2388-442.

106. Bart BA. Ultrafiltration is the only rational initial treatment ofvolume overload in decompensated heart failure. Circ Heart Fail.2009;2:499-504.

107. Bart BA. Ultrafiltration is the only rational initial treatment ofvolume overload in decompensated heart failure. Circ Heart Fail.2009;2:499-504.

108. Ricci Z, Ronco C, D’amico G, et al. Practice patterns in themanagement of acute renal failure in the critically ill patient: aninternational survey. Nephrol Dial Transplant. 2006;21:690-6.

109. Ronco C, Davenport A, Gura V. The future of the artificialkidney: moving towards wearable and miniaturized devices.Nefrologia. 2011;31:9-16.

110. Ronco C, Davenport A, Gura V. A wearable artificial kidney:

dream or reality? Nat Clin Pract Nephrol. 2008;4:604-5.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具