IIKJ

RcpuepbWddt(atus

nflammation and Nutrition in Renalnsufficiencyamyar Kalantar-Zadeh, Peter Stenvinkel, Luana Pillon, and

oel D. KoppleProtein-energy malnutrition (PEM) and inflammation are common in patients with chronic kidney

disease (CKD) and worsen as the CKD progresses toward the end-stage renal disease (ESRD). These

conditions are major predictors of poor clinical outcome in kidney failure, as reflected by a strong

association between hypoalbuminemia and cardiovascular disease (CVD). It has been suggested that

inflammation is the cause of both PEM and CVD and, hence, the main link among these conditions,

but these hypotheses are not well established. Increased release or activation of inflammatory

cytokines, such as interleukin-6 or tumor necrosis factor alpha, may suppress appetite, cause muscle

proteolysis and hypoalbuminemia, and may be involved in atherogenesis. Increasing serum levels of

proinflammatory cytokines caused by reduced renal function, volume overload, oxidative or carbonyl

stress, decreased levels of antioxidants, increased susceptibility to infection in uremia, and the

presence of comorbid conditions may lead to inflammation in CKD patients. In hemodialysis patients,

the exposure to dialysis tubing and dialysis membranes, poor quality of dialysis water, back-filtration

or back-diffusion of contaminants, and foreign bodies in dialysis access maybe additional causes of

inflammation. Similarly, episodes of overt or latent peritonitis, peritoneal dialysis (PD) catheter and its

related infections, and constant exposure to PD solution may contribute to inflammation in these

patients. The degree to which PEM in dialysis patients is caused by inflammation is not clear. Because

both PEM and inflammation are strongly associated with each other and can change many nutritional

measures and outcome concurrently in the same direction, the terms malnutrition-inflammation

complex syndrome (MICS) and/or malnutrition-inflammation-atherosclerosis (MIA) have been sug-

gested to denote the important contribution of both of these conditions to poor clinical outcome.

Maintenance dialysis patients who are underweight or who have low serum levels of cholesterol,

creatinine, or homocysteine may be suffering from the MICS/MIA and its subsequent poor outcome.

Consequently, obesity and hypercholesterolemia may appear protective, which is known as reverse

epidemiology. Although MICS/MIA may have a significant contribution in reversing the traditional

CVD risk factors in dialysis patients, it is not clear whether PEM or inflammation and their compli-

cations can be effectively managed in CKD and ESRD or whether their management improves clinical

outcome.

© 2003 by the National Kidney Foundation, Inc.

Index Words: Chronic kidney disease; End-stage-renal disease; Inflammation; Protein-energy malnu-

trition; Malnutrition-inflammation complex syndrome; Reverse epidemiology

mC

bcDKh

tr

MNH9

enal failure has been established as anindependent risk factor for cardiovas-

ular disease (CVD).1-4 Although theathophysiologic link between renal fail-re and CVD is not well understood, thelements of chronic inflammation and/orrotein-energy malnutrition (PEM) maye major contributors to this association.ith the progression of chronic kidney

isease (CKD), PEM and inflammationevelop and worsen as the CKD advances

oward the end-stage renal diseaseESRD).5-8 There appears to be a strongssociation between PEM and inflamma-ion in dialysis patients. In these individ-als both PEM and inflammation are as-

ociated with increased morbidity and

Advances in Renal Replacement Therapy, V

ortality, including enhanced risk ofVD and cardiovascular death.6,9-12 The

From the Division of Nephrology and Hypertension, Har-or-UCLA Medical Center, and David Geffen School of Medi-ine, University of California Los Angeles, Torrance, CA; andepartment of Clinical Science, Division of Renal Medicine,arolinska Institute, Huddinge University Hospital, Stock-olm, Sweden.

Supported by grant no. DK61162 from the National Insti-ute of Digestive, Diabetes and Kidney Diseases (KKZ), and aesearch grant from Amgen, Inc (KKZ, JDK).

Address correspondence to Kamyar Kalantar-Zadeh, MD,PH, David Geffen School of Medicine at UCLA, Division ofephrology and Hypertension, Harbor-UCLA Medical Center,arbor Mailbox 406, 1000 West Carson Street, Torrance, CA

0509-2910. E-mail: kamkal@ucla.edu© 2003 by the National Kidney Foundation, Inc.1073-4449/03/1003-0002$30.00/0

doi:10.1053/j.arrt.2003.08.008

ol 10, No 3 (July), 2003: pp 155-169 155

nabcTtaShmtfttafaktoCmmatmrpoooitsPcab

P

Tmcmdmc

rtstawasttacdcratrdptst

tkrtfpbhpe(amyptmtlmata

156 Kalantar-Zadeh et al

ature of the relationships between PEMnd inflammation and the relative contri-utions of these 2 entities to clinical out-ome have not been clearly elucidated.6,13

hese considerations are of major impor-ance because the annual mortality ratemong dialysis patients in the Unitedtates continues to remain unacceptablyigh (ie, approximately 20%), despiteany recent improvements in dialysis

echnology and treatment.14 The con-ounding effects of PEM and inflamma-ion on outcome appear to be so stark thathey may even reverse the conventionalssociations between such traditional riskactors as obesity or hypercholesterolemiand clinical outcome in individuals withidney failure.15-17 Hence, in contrast tohe general population in which elementsf overnutrition are strong predictors ofVD and death, decreased nutritionaleasures such as a low weight or bodyass or a reduced serum cholesterol, cre-

tinine, or possibly homocysteine concen-rations are correlated with increased

orbidity and mortality including higherisk of CVD in dialysis patients.15-18 Thisaradoxical phenomenon has also beenbserved with regard to the associationf low blood pressure, which is moreften observed in inflamed or malnour-

shed dialysis patients, and increased mor-ality.19-21 These inverse associations aretrongly suggestive of the central role ofEM and/or inflammation in poor out-ome of patients with chronic renal failurend known as reverse epidemiology (Ta-le 1).

EM

here are significant associations betweeneasures of PEM and such clinical out-

omes as increased rate of hospitalization,ortality, and worsened quality of life in

ialysis patients.11,12 Indicators of PEM inaintenance dialysis patients include de-

reased dietary protein and energy intake; 0

educed serum albumin, prealbumin,ransferrin, cholesterol, creatinine, and in-ulin-like growth factor-1 (IGF-1) concen-rations; decreased total body nitrogennd total body potassium; decreasedeight-for-height and body mass index;

nd reduced midarm muscle mass andkinfold thicknesses.6,11,12,16 Many otherools for assessment of the nutritional sta-us of the dialysis population are avail-ble. However, these methods are not ex-lusive detectors of PEM and may alsoetect conditions that are associated withhronic inflammation. For example, se-um albumin, transferrin, and prealbuminre negative acute phase reactants,13 andhe Subjective Global Assessment and itsefinements may also be a marker of theegree of sickness in maintenance dialysisatients.23,24 During acute catabolic states,

he urea nitrogen appearance may tran-iently increase independently of food in-ake.25

A low normalized protein equivalent ofotal nitrogen appearance (nPNA), alsonown as normalized protein catabolicate (nPCR), reflects the daily protein in-ake.26 The nPNA, which may be con-ounded because to its mathematical cou-ling with Kt/V, has been shown in someut not all studies to be a predictor ofospitalization and mortality in dialysisatients.6,22,27 A recent study showed thatven among maintenance hemodialysisMHD) patients who have a reportedlydequate Kt/V (�1.20), the nPNA stillaintains a strong association with dial-

sis outcome.28 In this study, 122 MHDatients with a delivered Kt/V �1.20 at

he start of the study were followed for 12onths. Serum albumin and nPNA were

he only variables with significant corre-ations with both mortality and three

easures of hospitalization (total days ofnd total number of hospitalizations andime to first hospital admission). The rel-tive risk of first hospitalization for each

.3 g/kg/d (standard deviation) decrease

ivri

qdla

nrhdmawa

T

NgAhp

157Nutrition/Inflammation in Renal Failure

n nPNA was 1.43 (95% confidence inter-al: 1.06-1.93, P � .02), and the relativeisk of death was 3.29 (95% confidencenterval: 1.57-6.91, P � .002).28

In another study using food frequencyuestionnaires, it was shown that hemo-ialysis patients consumed significantly

ower amounts of potassium, vitamin C,

able 1. Reverse Epidemiology of Cardiovascu

Risk Factors of CVD

Direction of the

General Po

Weight and body mass High BMI and oassociated wiCVD and mo

Serum cholesterol HyperchlesterolLDL, and lowdeleterious

BP Hypertension leincreased CVDmortality

Serum creatinine A mild to modein serum creaindependent rCVD

Plasma homocysteine A high total plahomocysteinerisk factor ofCVD and mo

Serum iron A high serum irrisk factor of(hemochromapoor outcome

Energy and/or proteinintake

A high energy aintake may bewith risk of oincreased mor

OTE. Effect and direction of traditional CVD riskeneral population.bbreviations: BMI, body mass index; CVD, cardigh density lipoprotein; BP, blood pressure; nPNrotein catabolic rate.

nd dietary fibers as well as some carote- a

oids.29 Thus, it is possible that prescribedestrictions in potassium in maintenanceemodialysis (MHD) patients lead to re-uced fruit and vegetable intake. Thisodification of dietary habits leaves meat

nd fats as the main source of calories,hich may contribute to atherosclerosis

nd increased cardiovascular morbidity

isk Factors in Maintenance Dialysis Patients

ciations Between Risk Factors and Outcomes

ion Maintenance Dialysis Patients

y arereased

High BMI and weight-for-height appear to beprotective; low BMI isassociated with increasedmortality

highare

Moderatehypercholesterolemia isprotective;hypocholesterolemiaincreases mortality

o A low (and not a high)predialysis BP correlateswith increased mortality

ncreaseis anctor of

An increased creatinine levelis associated with bettersurvival

is ased

Some recent studies havefound that a lowhomocysteine level isassociated with increasedrisk of CVD and mortalityin dialysis patients

vel is averloadand

A recent study has foundthat a low serum iron andtransferrin saturation isassociated with increasedrisk of hospitalization andmortality in hemodialysispatients

odciatedand

Increased protein intake andhigher nPNA (nPCR)improves survival

rs are compared between dialysis patients and the

ular disease; LDL, low density lipoprotein; HDL,malized protein catabolic rate; nPCR, normalized

lar R

Asso

pulat

besitth incrtality

emia,HDL

ads tand

rate itinineisk fa

smalevel

increartality

on leiron otosis)

nd foasso

besitytality

s facto

iovascA, nor

nd mortality in these patients. Further-

magfAsotmtoshtqc

I

Ialstbscctsalspmrc

iaecbatatm

plsddstmthetIfaramie

taabEoarmeesdMamds

ccaiata

158 Kalantar-Zadeh et al

ore, increased blood levels of thetheroprotective isoflavones such asenistein and daidzein may be beneficialor dialysis patients.30 Nonobese Asian-merican dialysis patients are found in

ome but not all studies to have a betterutcome compared with their obese coun-erparts (ie, reversal of the reverse epide-

iology, see later), which may be relatedo the fact that Asian Americans moreften ingest isoflavone-rich foods such asoy products.17 The effect of health-en-ancing diets on clinical outcome in main-

enance dialysis patients is an importantuestion that will require well designedlinical trials.

nflammation in Renal Failure

nflammation is a physiologic responsend contributes to defense mechanisms iniving organisms. An inflammatory re-ponse to an acute event, such as infec-ion, trauma, or toxic injury, helps theody to defend against pathological in-ults.31,32 However, if inflammation be-omes prolonged, it may lead to adverseonsequences such as a decline in appe-ite, increased rate of protein depletion inkeletal muscle and other tissues, musclend fat wasting, hypercatabolism, vascu-ar endothelial damage, and atherosclero-is.32 Thus, the acute-phase inflammatoryrocess that is a normal host defenseechanism may play a detrimental role in

enal patients and contribute to the in-reased risk for cardiovascular events.

Inflammatory processes are common inndividuals with both CKD and ESRDnd are closely related to PEM and accel-rated atherosclerosis.7,9,10,13,33 ESRD-asso-iated inflammation has been observed inoth North American,7,34,35 European,10,36

nd Asian37-39 dialysis patients, and 30%o 60% of ESRD patients in Europeannd North American countries are foundo have increased serum levels of inflam-

atory markers. Interestingly, dialysis c

atients in Asian countries may have aower prevalence of inflammation,38,39

uggesting that genetic factors and/orifferences in cultural practices (such asiet) may influence the inflammatory re-ponse. Moreover, recent data suggesthat female patients with chronic inflam-

atory response have a better outcomehan their male counterparts; female sexormones may have important protectiveffects that limit the impact of inflamma-ion on vascular injury in ESRD patients.40

n recent years, more attention has beenocused on the inflammatory processes aspossible cause of accelerated atheroscle-

osis as well as a wasting syndrome7,9,33

nd refractory anemia.41,42 Thus, inflam-ation may lead to a poor outcome in

ndividuals with underlying kidney dis-ase and renal insufficiency.

There is no uniform approach to assesshe severity of inflammation in individu-ls with kidney disease. Several serumnd blood measures of inflammation haveeen commonly used in both CKD andSRD patients as well as in other groupsf patients.33,43 Serum levels of “positivecute-phase reactants,” such as serum C-eactive protein (CRP) or ferritin, arearkers that are elevated during an acute

pisode of inflammation. The serum lev-ls of “negative acute-phase reactants,”uch as albumin or transferrin, decreaseuring an inflammatory process.31,32

any negative acute-phase reactants arelso traditionally known as nutritionalarkers because their serum levels are

ecreased with a decline in nutritionaltatus.

Occult or overt inflammation may behronic or recurrent; hence, they have oc-asionally been referred to as “chronic”cute-phase responses.23,44 Such chronicnflammatory states are associated withn elevation of serum acute phase pro-eins, including CRP, serum amyloid A,nd some proinflammatory cytokines in-

luding a variety of interleukins. Among

p(tienayial

C

Diamcibo

Tmratvnnmwriastrtssp

T

Aa

159Nutrition/Inflammation in Renal Failure

roinflammatory cytokines, interleukinIL)-6 is reported to play a central role inhe pathophysiology of adverse effects ofnflammation in patients with renal dis-ase.9,45-48 Increased serum levels of tumorecrosis factor � (TNF-�) and IL-6 may bessociated with reduced appetite in dial-sis patients.49 Renal failure may lead to

ncreased inflammatory responses throughnumber of mechanisms as presented

ater (Table 2).

KD and Inflammation

eteriorating renal function may promotenflammatory responses by several mech-nisms that are related to a reduced glo-erular filtration rate. Decreasing renal

learances of factors that are directly orndirectly involved in inflammation maye a main factor. The serum half-lives

able 2. Causes of Inflammation in Renal Fa

Causes of inflammation in CKD or bDecreased clearance of proinflammVolume overloadOxidative stressCarbonyl stressIncreased level of endotoxinsDecreased levels of antioxidantsDeteriorating protein-energy nutriIncreased susceptibility to infectionGenetic factors such as low produ

Coexistence of Comorbid Conditions1. Inflammatory diseases with kidn2. Increased prevalence of other co

Additional Inflammatory Factors RelHemodialysis:

1. Exposure to dialysis tubing2. Dialysis membranes with decr3. Impurities in dialysis water an4. Back-filtration or back-diffusio5. Foreign bodies (such as PTFE6. Intravenous catheter

Peritoneal Dialysis:1. Episodes of overt or latent pe2. PD catheter as a foreign body3. Constant exposure to PD solu

bbreviations: CKD, chronic kidney disease; GFR,tosus; HIV, human immuno-deficiency virus; PTF

f such proinflammatory cytokines as i

NF-� or IL-1 are greater in nephrecto-ized as compared to sham-operated

ats.50,51 Moreover, Descamps-Latscha etl52 have shown strong negative correla-ions between serum concentrations ofarious inflammatory markers, such aseopterin, TNF-� TNF-sR55, and creati-ine clearance. Declining renal functionay affect other inflammatory markers asell because serum CRP, IL-6 and hyalu-

onan levels have been found to have annverse correlation with creatinine clear-nce.53,54 Even among ESRD patients withome degree of residual renal function,hose with a lower glomerular filtrationate have higher serum CRP concentra-ions.55 In addition, accumulation of sub-tances that are products of carbonyltress such as advanced glycosylated endroducts in CKD patients may provoke

56-58

e of decreased GFRcytokines

state and food intakeremiaof anti-inflammatory cytokinesKD Patientsvolvement (SLE, HIV, and so on)id conditionsto Dialysis Treatment

biocompatiblility (eg, cuprophane)dialysate

contaminantsialysis access grafts

tisits related infections

rular filtration rate; SLE, systemic lupus erythem-y-tetra-fluoro-ethylene; PD, peritoneal dialysis.

ilure

ecausatory

tionalin u

ctionin Cey inmorbated

easedd/orn of

) in d

ritoniand

tion

glomeE, pol

nflammation. Fluid overload and

ctsgttioitclm

ttedflpmtsoccAspnia

prooubmtlmfl(dia

uocam

E

AegdoCtdgsmanraaMsMaTtiypftmbbtctfipacbb

160 Kalantar-Zadeh et al

ongestive heart failure may be other con-ributors to increased inflammatory re-ponses in CKD patients. It has been sug-ested that vascular congestion may alterhe permeability of the gastrointestinalract, leading to accumulation of endotox-ns such as lipopolysaccharides.33 More-ver, there is a greater susceptibility to

nfection in uremic patients as comparedo healthy controls. The foregoing pro-esses may in turn stimulate monocyteseading to increased release of proinflam-

atory cytokines.59,60

Oxidative stress, which occurs whenhere is an excessive free-radical produc-ion or low antioxidant levels, hasmerged as an important promoter of en-othelial dysfunction and possibly in-ammation and atherogenesis.61,62 Lowerlasmalogen levels have been found inalnourished and inflamed CKD pa-

ients, indicating increased oxidativetress.63 Hence, it may be speculated thatxidative stress could result in an in-reased production of proinflammatoryytokines and an acute-phase response.n acute-phase response also has been

hown to be associated with decreasedlasma levels of several antioxidants inonrenal patients.64 A lower intake of var-

ous antioxidants in ESRD patients mightlso contribute to oxidative stress.29

The common occurrence of superim-osed illnesses frequently encountered inenal patients plays a role in the etiologyf the hypercatabolic state and the devel-pment of inflammation in these individ-als.65 Hence, comorbid conditions maye the major contributors to the develop-ent of chronic inflammation in CKD pa-

ients. Underlying kidney disease (such asupus erythematosus or some of the glo-

erulonephritides associated with in-ammatory conditions), genetic factors

for example in diabetic patients), cardiacisease per se, unrecognized persistence

nfections, and atherosclerosis per se may

lso engender inflammation in renal fail- v

re. However, even in the absence ofvert clinical illness, these patients mayontinue to have inflammatory processes,lthough not as severely as observed inultimorbid patients.

SRD and Inflammation

n increased level of inflammatory mark-rs has been reported in patients under-oing both MHD and chronic peritonealialysis (CPD). In addition to the causesf inflammation that are described inKD patients (see earlier), proinflamma-

ory cytokines are overtly increased inialysis patients when compared to theeneral population. Kimmel et al43 mea-ured serum levels of several pro-inflam-atory cytokines in 230 MHD patients

nd compared the data with those in 28ondialytic controls and found that se-um levels of interleukins 1, 2, 4, 5, 6, 12,nd 13 as well as tumor necrosis factorlpha were significantly increased inHD patients.43 Kaysen et al44 have

hown that CRP levels varies over time inHD patients and that these fluctuations

re unrelated to the dialysis procedure.44

his may suggest that factors unrelated tohe dialysis procedure may be the mostmportant cause for inflammation in dial-sis patients. However, it should beointed out that various dialysis-related

actors may also contribute to inflamma-ion (Table 3). First, exposure to dialysis

embranes, especially those that are lessiocompatible (eg, cuprophane mem-ranes) as well as exposure to dialysisubing during dialysis treatment may be aause of inflammation.48,66 Second, alsohe quality of dialysis water67-69 and back-ltration or back-diffusion because of ex-osure to endotoxins that are possiblyvailable in the dialysate fluid may be aause of inflammation.70 Finally, it shoulde emphasized that implanted foreignodies in those who have an arterio-

enous graft, which may harbor chronic

onptfrmptsi

I

MtctlomMupChweEElaIpmtrI

ipvssaivD

mticesegCtmscpoteladtgdttdlt

asttfihtaep

arotic

161Nutrition/Inflammation in Renal Failure

r recurrent latent infection or intrave-ous catheter in those who do not have aeripheral dialysis access may contribute

o inflammation.71,72 In CPD patients, theollowing additional factors may play aole in the development of chronic inflam-ation: episodes of overt or subclinical

eritonitis, CPD catheter-related infec-ions,39,73 and constant exposure to PDolution, which may include bioincompat-ble substances or endotoxins (Table 3).73

nflammation and Outcome

arkers of inflammation and malnutri-ion predict poor outcome including in-reased mortality in CKD and ESRD pa-ients. Ironically, infection, which usuallyeads to an overt inflammatory response,r malnutrition per se are not the com-on causes of death in these patients.ost individuals with chronic renal fail-

re die of CVD.14 Inflammation predis-oses dialysis patients to atheroscleroticVD as well as to a net catabolic state andypoalbuminemia.9,10,13,74 ESRD patientsith coronary heart disease often have

levated levels of acute-phase reactants.10

pidemiological studies indicate that inSRD patients increased serum CRP is at

east as a strong predictor of morbiditynd mortality as is serum albumin.34,35

ncreased serum levels of IL-6 also is re-orted to be associated with increasedortality in both MHD74 and CPD pa-

ients.47 Progression of carotid atheroscle-osis during dialysis may be related toL-6 levels.45

It should be noted that the cascade ofnflammatory factors leading to an acute-hase reaction is counter-regulated byarious anti-inflammatory cytokines,uch as IL-10. Recently Girndt et al75

howed that the �1082A allele, which isssociated with a low production of IL-10,s related to an increased risk of cardio-ascular events in 300 MHD patients.

ata indicate that inflammatory processes c

ay promote proliferation and infiltra-ion of inflammatory cells into the tunicantima of small arteries, including theoronary arteries. This may result in ath-rosclerosis and stenosis of the blood ves-els and subsequent occlusive vascularvents.76,77 Epidemiological evidence sug-ests that inflammation may be linked toVD via some specific low-grade infec-

ions, such as is caused by Chlamydia pneu-oniae.76,77 C pneumoniae infection is

hown to predict adverse vascular out-ome in CPD patients,78 and elevated Cneumoniae IgA titres predict progressionf carotid atherosclerosis in ESRD pa-ients.45 Myeloperoxidase, an abundantnzyme secreted by neutrophils, may alsoink inflammation to oxidative stress andtherosclerosis in ESRD patients.79 In-eed, recent data have shown that a func-

ional variant of the myeloperoxidaseene is associated with cardiovascularisease in end-stage renal disease pa-

ients.80 It should be noted that inflamma-ion might also cause direct endothelialysfunction via a stimulation of intercel-

ular adhesion molecules in CKD pa-ients.62

The association between inflammationnd increased rate of CVD is not re-tricted to renal disease. Women with sys-emic lupus erythematosus are severalimes more likely to have myocardial in-arction than age-matched women,81 thencidence of myocardial infarction is 50%igher in individuals with rheumatoid ar-

hritis than in controls matched for agend sex,82 and accelerated coronary ath-rosclerosis has been observed in youngatients with HIV disease.83

Inflammation might also cause otherdverse outcomes in renal disease such asefractory anemia, laboratory signs of ironverload, or poor quality of life.11 Main-enance dialysis patients with evidence ofnflammation frequently display in-reased serum ferritin, which, under these

ircumstances, is not a reliable indicator

oirhcMotcbo

M

Tpamtaerttaiaia

Tdepphnfpisyibbisit

ttIPmTm

162 Kalantar-Zadeh et al

f increased iron burden but more a pos-tive acute-phase reactant.41,84 Serum fer-itin has been shown to correlate withospitalization rates, and a recent in-rease in serum ferritin concentration in

HD patients may carry a greater riskf death.84 Hence, the association be-ween a high serum ferritin and poor out-ome in dialysis patients may be entirelyecause of inflammation and not ironverload.

alnutrition-Inflammation Complex

he term malnutrition-inflammation com-lex syndrome (MICS) indicates the closessociation between PEM and inflam-ation in dialysis patients.6,17,85,86 Al-

ernatively, malnutrition-inflammation-therosclerosis (MIA) has been used tomphasize the importance of atheroscle-otic diseases as a major consequence ofhis disorder.87 Renal failure and uremicoxins, as well as comorbid conditionsnd dialysis related factors, may lead tonflammation and PEM, which per sere major determinants of the low qual-ty of life and increased hospitalization

nd mortality in dialysis patients (Fig 1). P

he theory of reverse epidemiology inialysis patients is compatible with thexistence of MICS or MIA and its inter-lay with the traditional risk factors;atients who are underweight or whoave a low serum cholesterol, creati-ine, or homocysteine may be sufferingrom the MICS/MIA and its subsequentoor outcome17 (Table 1). Several scor-

ng systems have been proposed to as-ess the degree of MICS or MIA in dial-sis patients, such as malnutrition-

nflammation score (MIS),24 which haseen found to correlate strongly withoth measures of nutritional status and

nflammation and anemia. The MIS isignificantly associated with hospital-zation rates and mortality in HD pa-ients.24

The association of PEM with inflamma-ion in ESRD patients may be an explana-ion for PEM associated mortality.6,7,9,10

ndeed, several investigators suggest thatEM is a consequence of chronic inflam-atory processes in ESRD patients.7,87,88

hus, chronic inflammation may be theissing link or factor that causally ties

Figure 1. A hypotheti-cal model of the com-plex interrelationshipsamong the predictors(inflammation and PEM)and outcomes (quality oflife, morbidity, and mor-tality) in individuals un-dergoing maintenancedialysis.

EM to morbidity and mortality. The fol-

liipkngsiflkpshedtbstpdwspIcawtitplmmean

mtisaaio

echiiaeiiorpsgrpactttpcqimdoniTshptamhtFscke

ct

163Nutrition/Inflammation in Renal Failure

owing arguments have been advancedn support of this thesis: inflammations associated with a rise in plasma androbably tissue levels of catabolic cyto-ines. TNF-�, also known as cachexin,ot only promotes catabolic processes, en-endering both protein degradation anduppression of protein synthesis, but alsonduces anorexia.49,89,90 Elevation of in-ammatory proteins and catabolic cyto-ines are observed in both nondialyzedatients with advanced chronic renal in-ufficiency and in dialysis patients whoave signs of PEM. Dialysis patients withvidence of inflammation are reported toevelop weight loss and a negative pro-

ein balance even with an intact appetiteecause there may be a shift in proteinynthesis from muscle to acute-phase pro-eins as renal function declines.8 Inflamedatients appear to lose more body weighturing dialysis when they are comparedith noninflamed patients.91 Evidence

uggests that albumin synthesis is sup-ressed when serum CRP is elevated.44,92

n CKD patients, serum albumin de-reases as renal function deteriorates,5nd this phenomenon occur concurrentlyith the accumulation of proinflamma-

ory cytokines.53-55 Serum albumin may benfluenced by factors other than malnutri-ion and high concentrations of acute-hase proteins, such as CRP, are corre-

ated with low serum albumin inalnourished dialysis patients.93 Inflam-ation may also lead to hypocholesterol-

mia, a strong mortality risk factor in di-lysis patients and a marker of poorutritional status.74

Although proinflammatory cytokinesay be one common link between malnu-

rition, inflammation, and atherosclerosis,t should be emphasized that other factorsuch as oxidative stress, carbonyl stress,nd uremic toxins may play a direct roles well.61,62 The extent to which PEM andnflammation may be a cause of each

ther and the degree to which they can m

ach independently induce adverse out-omes in patients with kidney diseasesas not been clearly defined. The follow-

ng arguments have questioned the role ofnflammation as a cause of PEM: serumlbumin and other indicators of protein-nergy nutritional status correlate withndicators of protein intake irrespective ofnflammatory status.87,93 During studiesf PEM induced in normal individuals byeducing their nutrient intake or in HDatients fed with low protein diets, theerum albumin decreases modestly, sug-esting clearly that serum albumin is aeflection of protein intake.94 In dialysisatients, the association of serum albuminnd CRP is not precise, and the reportedorrelation coefficients are usually lesshan 0.50.13,93 Serum albumin concentra-ions usually do not fluctuate on a month-o-month basis, whereas serum CRP ap-ears to do so.44 At least in some acute orhronic illnesses, the provision of ade-uate nutrition without treatment of the

nflammation improves hypoalbumine-ia and clinical outcome.95 Malnourished

ialysis patients may be deficient of anti-xidants such as vitamin C or carote-oids,29 which may lead to increased ox-

dative stress leading to inflammation.6here is evidence that certain nutrients,uch as arginine and glutamine, may en-ance the immune response.96 Moreover,reliminary data suggest that levocarni-

ine may protect against endotoxins andlso suppress elaboration of TNF-� fromonocytes.97 Thus, PEM may decrease

ost resistance and predispose to infec-ion, which is an inflammatory disorder.inally, the positive association of mea-ures of PEM with inflammation may beaused, in part, by the generation of cyto-ines in the setting of low protein andnergy intake.6

These considerations, although notonclusive, indicate that factors otherhan the catabolic consequences of inflam-

ation may also affect serum albumin

anfpptrno

MI

TtmiphsekdtmtmmrviosstpjaM

slctattp

oItscapegstcjtfpltbem

sccaifltittbflotp

1

2

164 Kalantar-Zadeh et al

nd other nutritional measures and thatutrient intake is almost certainly such aactor. In summary, although in dialysisatients PEM may be associated withoor outcome because of the primary con-

ribution of inflammation, existing dataemain consistent with the possibility thatutrient intake may independently affectutcome.

anagement of Malnutrition andnflammation in Kidney Failure

he evidence as to whether nutritionalreatment may improve morbidity and

ortality in dialysis patients is quite lim-ted. There are no large-scale randomizedrospective interventional studies thatave examined these questions. Amongtudies based on food intake, Kuhlmannt al98 reported that prescription of 45cal/kg/d and 1.5 g protein/kg/d in-uced weight gain and improved nutri-

ional status including serum albumin inalnourished hemodialysis patients. In

his study, weight change correlated withean dietary energy intake but not withean dietary protein intake. Leon et al95

eported that tailored nutritional inter-ention improved serum albumin levels

n 52 MHD patients, and this effect wasbserved even among patients with higherum CRP levels. Caglar et al99 recentlyhowed that oral nutritional supplemen-ation given during hemodialysis im-roves nutritional markers such as Sub-

ective Global Assessment and serumlbumin concentration in malnourishedHD patients.A limited number of retrospective

tudies evaluated the effect of intradia-ytic parenteral nutrition (IDPN) on clini-al outcome.100,101 Even though the correc-ion of hypoalbuminemia by IDPN wasssociated with improved outcome inhese studies, a number of other interven-ional studies fail to show evident im-

rovement of nutritional status or clinical

utcome in dialysis patients receivingDPN or other nutritional interven-ions.102 However, recently Pupim et al103

howed that IDPN promoted a large in-rease in whole-body protein synthesisnd a significant decrease in whole-bodyroteolysis in 7 MHD patients withoutvidence of inflammation. Many investi-ators of such studies used small sampleizes, failed to restrict study subjects tohose with PEM, did not control for con-urrent food intake, did not define or ad-ust appropriately for comorbid condi-ions, performed nutritional interventionsor short periods of time, and followed upatients for only short intervals. Until

arge-scale, prospective randomized in-erventional studies are conducted, it wille difficult to ascertain the potential ben-fits of increasing nutritional intake inalnourished dialysis patients.Although epidemiological evidence

trongly links inflammation to poor out-ome in individuals with renal insuffi-iency, it must be recognized that therere not yet randomized clinical trials tondicate improvement of outcomes by in-ammation-reducing measures. Hence,

here is currently no consensus concern-ng the management strategies for thereatment of chronic inflammation in pa-ients with renal failure. However, possi-le treatment modalities may target in-ammation directly, or they may focus onxidative stress or endothelial dysfunc-ion. The following therapeutic ap-roaches may be considered:

. Statins (HMG-CoA reductase inhibi-tors) have been suggested for chroni-cally inflamed patients. Statins areshown to decrease CRP levels indepen-dently of their effects on lipids andmay be associated with reduced mor-tality in ESRD patients.37,104

. Angiotensin-converting enzyme inhib-itors may have anti-inflammatory

properties in both the general popula-

3

4

cCnecfibibmp

C

Ittimsshtvpottmne

mqtccdaHsydifltoctpMb(ciocaftatr

R

165Nutrition/Inflammation in Renal Failure

tion and in CKD and ESRD patients105

and are associated with delayed pro-gression of chronic renal failure andimproved outcome in these individu-als.1

. Vitamin E may have anti-inflammatoryeffects, and vitamin E administration isassociated with a decreased risk forcardiovascular mortality in dialysis pa-tients.106

. Optimization of dialysis treatmentmay improve inflammatory status indialysis patients, and the type of dial-ysis membrane may have a bearing onmortality.107

The use of ultra-pure dialysate and bio-ompatible membranes may decreaseRP, even though such interventions mayot completely normalize the serum lev-ls of CRP.48,66 Of note, a major clini-al trial (the HEMO Study)108 recentlyailed to show an improvement in clin-cal outcomes by use of high-flux mem-ranes. However, this finding does not

pso facto refute the role of dialysis mem-rane biocompatibility in the develop-ent of chronic inflammation in ESRD

atients.

linical Implications and Future Steps

ndividuals with renal failure, especiallyhose undergoing maintenance dialysisreatment, are a group of highly selectedndividuals whose high rate of CVD and

ortality does not appear to be caused byuch traditional risk factors as hyperten-ion, obesity, hypercholesterolemia, oryperhomocysteinemia. Indeed these fac-

ors are associated with improved survi-al.17 Such paradoxical associations ap-ear to have major clinical implications inur patients.17 To that end, the extrapola-ion of data on traditional risk factors inhe general population to prognostic and

anagement considerations in mainte-ance dialysis patients without adequate

vidence may be flawed. Intense manage-

ent of traditional risk factors, which isuite important, does not appear to leado major improvements of CVD or poorlinical outcome in dialysis patients.14 In-reasing dialysis dose or use of high-fluxialysis membranes also failed to showny improvement in survival according toEMO study.108 Hence, it may be time to

hift the focus of clinical studies and dial-sis patient management to such nontra-itional factors as nutritional status and

nflammation. Although the PEM and in-ammation may confound the effect of

raditional risk factors on clinical outcomef maintenance dialysis patients, theseonditions in themselves may be amonghe main causes of death in this group ofatients. Hence, they deserve attention.aintenance dialysis patients appear to

e a unique population with their ownnonconventional) risk factors as is thease with the phenomenon of reverse ep-demiology.17 To that end, more epidemi-logical studies and well-designed clini-al trials are needed to verify the existencend precise nature of nontraditional riskactors in chronic renal insufficiency ando examine the role of nutritional supportnd anti-inflammatory interventions as areatment for high CVD and mortalityate in these individuals.

eferences

1. Mann JF, Gerstein HC, Pogue J, et al: Renalinsufficiency as a predictor of cardiovascularoutcomes and the impact of ramipril: TheHOPE randomized trial. Ann Intern Med 134:629-636, 2001

2. Ruilope LM, Salvetti A, Jamerson K, et al:Renal function and intensive lowering ofblood pressure in hypertensive participants ofthe hypertension optimal treatment (HOT)study. J Am Soc Nephrol 12:218-225, 2001

3. Shlipak MG, Fried LF, Crump C, et al: Cardio-vascular disease risk status in elderly personswith renal insufficiency. Kidney Int 62:997-1004, 2002

4. Shlipak MG, Simon JA, Grady D, et al: Renal

insufficiency and cardiovascular events in

166 Kalantar-Zadeh et al

postmenopausal women with coronary heartdisease. J Am Coll Cardiol 38:705-711, 2001

5. Kopple JD, Greene T, Chumlea WC, et al:Relationship between nutritional status andthe glomerular filtration rate: Results from theMDRD study. Kidney Int 57:1688-1703, 2000

6. Kalantar-Zadeh K, Kopple JD: Relative contri-butions of nutrition and inflammation to clin-ical outcome in dialysis patients. Am J KidneyDis 38:1343-1350, 2001

7. Stenvinkel P: Inflammation in end-stage renalfailure: Could it be treated? Nephrol DialTransplant 17:33-38; discussion 40, 2002(suppl)

8. Stenvinkel P, Alvestrand A: Inflammation inend-stage renal disease: Sources, conse-quences, and therapy. Semin Dial 15:329-337,2002

9. Stenvinkel P, Barany P, Heimburger O, et al:Mortality, malnutrition, and atherosclerosis inESRD: What is the role of interleukin-6? Kid-ney Int Suppl 80:103-108, 2002

10. Zimmermann J, Herrlinger S, Pruy A, et al:Inflammation enhances cardiovascular riskand mortality in hemodialysis patients. Kid-ney Int 55:648-658, 1999

11. Kalantar-Zadeh K, Kopple JD, Block G, et al:Association among SF36 quality of life mea-sures and nutrition, hospitalization, and mor-tality in hemodialysis. J Am Soc Nephrol 12:2797-2806, 2001

12. Kopple JD: Nutritional status as a predictor ofmorbidity and mortality in maintenance dial-ysis patients. ASAIO J 43:246-250, 1997

13. Kaysen GA: Malnutrition and the acute-phasereaction in dialysis patients-how to measureand how to distinguish. Nephrol Dial Trans-plant 15:1521-1524, 2000

14. United States Renal Data System: Bethesda,MD, US Department of Public Health andHuman Services, Public Health Service, Na-tional Institutes of Health, 2002

15. Iseki K, Yamazato M, Tozawa M, et al: Hypo-cholesterolemia is a significant predictor ofdeath in a cohort of chronic hemodialysis pa-tients. Kidney Int 61:1887-1893, 2002

16. Lowrie EG, Lew NL: Death risk in hemodial-ysis patients: The predictive value of com-monly measured variables and an evaluationof death rate differences between facilities.Am J Kidney Dis 15:458-482, 1990

17. Kalantar-Zadeh K, Block G, Humphreys MH,et al: Reverse epidemiology of cardiovascularrisk factors in maintenance dialysis patients.Kidney Int 63:793-808, 2003

18. Suliman ME, Qureshi AR, Barany P, et al:

Hyperhomocysteinemia, nutritional status,and cardiovascular disease in hemodialysispatients. Kidney Int 57:1727-1735, 2000

19. Zager PG, Nikolic J, Brown RH, et al: “U”curve association of blood pressure and mor-tality in hemodialysis patients. Medical Direc-tors of Dialysis Clinic, Inc. Kidney Int 54:561-569, 1998

20. Port FK, Hulbert-Shearon TE, Wolfe RA, et al:Predialysis blood pressure and mortality risk ina national sample of maintenance hemodialysispatients. Am J Kidney Dis 33:507-517, 1999

21. Klassen PS, Lowrie EG, Reddan DN, et al:Association between pulse pressure and mor-tality in patients undergoing maintenance he-modialysis. JAMA 287:1548-1555, 2002

22. Kalantar-Zadeh K, Kopple J: Malnutrition as acause of morbidity and mortality in dialysispatients, in Kopple J, Massry S, eds: Nutri-tional Management of Renal Disease (2nd ed).Philadelphia, PA, Lippincott, Williams &Wilkins, 2003

23. Kalantar-Zadeh K, Kleiner M, Dunne E, et al:total iron-binding capacity-estimated trans-ferrin correlates with the nutritional subjectiveglobal assessment in hemodialysis patients.Am J Kidney Dis 31:263-272, 1998

24. Kalantar-Zadeh K, Kopple JD, Block G, et al: Amalnutrition-inflammation score is correlatedwith morbidity and mortality in maintenancehemodialysis patients. Am J Kidney Dis 38:1251-1263, 2001

25. Grodstein GP, Blumenkrantz MJ, Kopple JD:Nutritional and metabolic response to cata-bolic stress in uremia. Am J Clin Nutr 33:1411-1416, 1980

26. Borah MF, Schoenfeld PY, Gotch FA, et al:Nitrogen balance during intermittent dialysistherapy of uremia. Kidney Int 14:491-500, 1978

27. Kalantar-Zadeh K, Kopple J: Nutritional man-agement of hemodialysis patients, in Kopple J,Massry S, eds: Nutritional Management ofRenal Disease (2nd ed). Philadelphia, PA, Lip-pincott, Williams & Wilkins, 2003

28. Kalantar-Zadeh K, Supasyndh O, Lehn RS, etal: Normalized protein nitrogen appearance iscorrelated with hospitalization and mortalityin hemodialysis patients with Kt/V greaterthan 1.20. J Renal Nutr 13:15-25, 2003

29. Kalantar-Zadeh K, Kopple JD, Deepak S, et al:Food intake characteristics of hemodialysispatients as obtained by food frequency ques-tionnaire. J Renal Nutr 12:17-31, 2002

30. Fanti P, Sawaya BP, Custer LJ, et al: Serum

levels and metabolic clearance of the isofla-

167Nutrition/Inflammation in Renal Failure

vones genistein and daidzein in hemodialysispatients. J Am Soc Nephrol 10:864-871, 1999

31. Streetz KL, Wustefeld T, Klein C, et al: Medi-ators of inflammation and acute phase re-sponse in the liver. Cell Mol Biol 47:661-673,2001

32. Suffredini AF, Fantuzzi G, Badolato R, et al:New insights into the biology of the acutephase response. J Clin Immunol 19:203-214,1999

33. Kalantar-Zadeh K, Kopple J: Inflammationin renal failure, in Rose B (ed): UpToDate.Wellesley, MA, UpToDate, Inc, 2003

34. Yeun JY, Levine RA, Mantadilok V, et al:C-Reactive protein predicts all-cause and car-diovascular mortality in hemodialysis pa-tients. Am J Kidney Dis 35:469-476, 2000

35. Owen WF, Lowrie EG: C-reactive protein asan outcome predictor for maintenance hemo-dialysis patients. Kidney Int 54:627-636, 1998

36. Zoccali C, Benedetto FA, Mallamaci F, et al:Inflammation is associated with carotid ath-erosclerosis in dialysis patients: Creed Inves-tigators. Cardiovascular Risk Extended Eval-uation in Dialysis Patients. J Hypertens 18:1207-1213, 2000

37. Chang JW, Yang WS, Min WK, et al: Effects ofsimvastatin on high-sensitivity C-reactive pro-tein and serum albumin in hemodialysis pa-tients. Am J Kidney Dis 39:1213-1217, 2002

38. Iseki K, Tozawa M, Yoshi S, et al: SerumC-reactive protein (CRP) and risk of death inchronic dialysis patients. Nephrol Dial Trans-plant 14:1956-1960, 1999

39. Noh H, Lee SW, Kang SW, et al: Serum C-re-active protein: A predictor of mortality in con-tinuous ambulatory peritoneal dialysis pa-tients. Perit Dial Int 18:387-394, 1998

40. Stenvinkel P, Wanner C, Metzger T, et al:Inflammation and outcome in end-stage renalfailure: Does female gender constitute a sur-vival advantage? Kidney Int 62:1791-1798,2002

41. Kalantar-Zadeh K, McAllister C, Lehn R, et al:Effect of malnutrition-inflammation complexsyndrome on erythropoietin hyporesponsive-ness in maintenance hemodialysis patients.Am J Kidney Dis (in press)

42. Stenvinkel P, Barany P: Anaemia, rHuEPOresistance, and cardiovascular disease in end-stage renal failure: Links to inflammation andoxidative stress. Nephrol Dial Transplant17(suppl):32-37, 2002

43. Kimmel PL, Phillips TM, Simmens SJ, et al:Immunologic function and survival in hemo-

dialysis patients. Kidney Int 54:236-244, 1998

44. Kaysen GA, Dubin JA, Muler HG, et al: Theacute-phase response varies with time andpredicts serum albumin levels in hemodialy-sis patients. The HEMO Study Group. KidneyInt 58:346-352, 2000

45. Stenvinkel P, Heimburger O, Jogestrand T:Elevated interleukin-6 predicts progressivecarotid artery atherosclerosis in dialysis pa-tients: Association with Chlamydia pneu-moniae seropositivity. Am J Kidney Dis 39:274-282, 2002

46. Pecoits-Filho R, Araujo MR, Lindholm B, et al:Plasma and dialysate IL-6 and VEGF concen-trations are associated with high peritonealsolute transport rate. Nephrol Dial Transplant17:1480-1486, 2002

47. Pecoits-Filho R, Barany P, Lindholm B, et al:Interleukin-6 is an independent predictor ofmortality in patients starting dialysis treat-ment. Nephrol Dial Transplant 17:1684-1688,2002

48. Memoli B, Minutolo R, Bisesti V, et al:Changes of serum albumin and C-reactiveprotein are related to changes of interleukin-6release by peripheral blood mononuclear cellsin hemodialysis patients treated with differentmembranes. Am J Kidney Dis 39:266-273, 2002

49. Kalantar-Zadeh K, Block G, McAllister C, etal. Appetite and inflammation, nutrition, ane-mia and clinical outcome in hemodialysis pa-tients. Am J Clin Nutr (in press)

50. Bemelmans MH, Gouma DJ, Buurman WA:Influence of nephrectomy on tumor necrosisfactor clearance in a murine model. J Immunol150:2007-2017, 1993

51. Porrini M, Simonetti P, Ciappellano S, et al:Thiamin, riboflavin and pyridoxine status inchronic renal insufficiency. Int J Vitam NutrRes 59:304-308, 1989

52. Descamps-Latscha B, Herbelin A, Nguyen AT,et al: Balance between IL-1 beta, TNF alpha,and their specific inhibitors in chronic renalfailure and maintenance dialysis: Relation-ships with activation markers of T cells, B cellsand monocytes. J Immunol 154:882-889, 1995

53. Panichi V, Migliori M, De Pietro S, et al: Creactive protein in patients with chronic renaldiseases. Renal Fail 23:551-562, 2001

54. Stenvinkel P, Heimburger O, Wang T, et al:High serum hyaluronan indicates poor sur-vival in renal replacement therapy. Am J Kid-ney Dis 34:1083-1088, 1999

55. Chung SH, Heimburger O, Stenvinkel P, et al:Association between inflammation andchanges in residual renal function and perito-

neal transport rate during the first year of

168 Kalantar-Zadeh et al

dialysis. Nephrol Dial Transplant 16:2240-2245, 2001

56. Sakata N, Imanaga Y, Meng J, et al: Increasedadvanced glycation end products in athero-sclerotic lesions of patients with end-stage re-nal disease. Atherosclerosis 142:67-77, 1999

57. Miyata T, Ueda Y, Horie K, et al: Renal catabo-lism of advanced glycation end products: Thefate of pentosidine. Kidney Int 53:416-422, 1998

58. Schwedler S, Schinzel R, Vaith P, et al: Inflam-mation and advanced glycation end productsin uremia: Simple coexistence, potentiation orcausal relationship? Kidney Int Suppl 78:S32-36, 2001

59. Hasper D, Hummel M, Kleber FX, et al: Sys-temic inflammation in patients with heart fail-ure. Eur Heart J 19:761-765, 1998

60. Sato Y, Takatsu Y, Kataoka K, et al: Serialcirculating concentrations of C-reactive pro-tein, interleukin (IL)-4, and IL-6 in patientswith acute left heart decompensation. ClinCardiol 22:811-813, 1999

61. Spittle MA, Hoenich NA, Handelman GJ, et al:Oxidative stress and inflammation in hemodi-alysis patients. Am J Kidney Dis 38:1408-1413,2001

62. Mezzano D, Pais EO, Aranda E, et al: Inflam-mation, not hyperhomocysteinemia, is relatedto oxidative stress and hemostatic and endo-thelial dysfunction in uremia. Kidney Int 60:1844-1850, 2001

63. Stenvinkel P, Holmberg I, Heimburger O, etal: A study of plasmalogen as an index ofoxidative stress in patients with chronic renalfailure: Evidence of increased oxidative stressin malnourished patients. Nephrol Dial Trans-plant 13:2594-2600, 1998

64. Langlois M, Duprez D, Delanghe J, et al: Se-rum vitamin C concentration is low in periph-eral arterial disease and is associated withinflammation and severity of atherosclerosis.Circulation 103:1863-1868, 2001

65. Keane WF, Collins AJ: Influence of co-morbid-ity on mortality and morbidity in patientstreated with hemodialysis. Am J Kidney Dis24:1010-1018, 1994

66. Schindler R, Boenisch O, Fischer C, et al: Effectof the hemodialysis membrane on the inflam-matory reaction in vivo. Clin Nephrol 53:452-459, 2000

67. Tielemans C, Husson C, Schurmans T, et al:Effects of ultrapure and non-sterile dialysateon the inflammatory response during in vitrohemodialysis. Kidney Int 49:236-243, 1996

68. Bambauer R, Walther J, Meyer S, et al: Bacte-

ria- and endotoxin-free dialysis fluid for use in

chronic hemodialysis. Artif Organs 18:188-192, 1994

69. Bambauer R, Schmidt R, Falkenhagen D, et al:Sterile and endotoxin free dialysis fluid forhemodialysis. Biomater Artif Cells Immobili-zation Biotechnol 19:71-83, 1991

70. Hosoya N, Sakai K: Backdiffusion rather thanbackfiltration enhances endotoxin transportthrough highly permeable dialysis mem-branes. ASAIO Trans 36:M311-313, 1990

71. Nassar GM, Ayus JC: Infectious complicationsof the hemodialysis access. Kidney Int 60:1-13,2001

72. Ayus JC, Sheikh-Hamad D: Silent infection inclotted hemodialysis access grafts. J Am SocNephrol 9:1314-1317, 1998

73. Yeun JY, Kaysen GA: Acute phase proteinsand peritoneal dialysate albumin loss are themain determinants of serum albumin in peri-toneal dialysis patients. Am J Kidney Dis 30:923-927, 1997

74. Bologa RM, Levine DM, Parker TS, et al: In-terleukin-6 predicts hypoalbuminemia, hypo-cholesterolemia, and mortality in hemodialy-sis patients. Am J Kidney Dis 32:107-114, 1998

75. Girndt M, Kaul H, Sester U, et al: Anti-inflam-matory interleukin-10 genotype protects dial-ysis patients from cardiovascular events. Kid-ney Int 62:949-955, 2002

76. Becker AE, de Boer OJ, van Der Wal AC: Therole of inflammation and infection in coronaryartery disease. Annu Rev Med 52:289-297,2001

77. Kaplan N: Risk factor for atherosclerotic dis-ease. In Braunwald EZD, Libby P, eds: Braun-wald: Heart Disease: A Textbook of Cardio-vascular Medicine. Philadelphia, PA, Saunders,2001, pp 1010-1039

78. Haubitz M, Brunkhorst R: C-reactive proteinand chronic Chlamydia pneumoniae infection:Long-term predictors for cardiovascular diseaseand survival in patients on peritoneal dialysis.Nephrol Dial Transplant 16:809-815, 2001

79. Daugherty A, Dunn JL, Rateri DL, et al: My-eloperoxidase, a catalyst for lipoprotein oxida-tion, is expressed in human atheroscleroticlesions. J Clin Invest 94:437-444, 1994

80. Pecoits-Filho R, Stenvinkel P, Marchlewska A,et al: A functional variant of the myeloperox-idase gene is associated with cardiovasculardisease in end-stage renal disease patients.Kidney Int Suppl 84:172-176, 2003

81. Manzi S, Meilahn EN, Rairie JE, et al: Age-specific incidence rates of myocardial infarc-tion and angina in women with systemic lu-

pus erythematosus: Comparison with the

1

1

1

1

1

1

1

1

1

169Nutrition/Inflammation in Renal Failure

Framingham Study. Am J Epidemiol 145:408-415, 1997

82. Gabriel SE, Crowson CS, O’Fallon WM: Mor-tality in rheumatoid arthritis: Have we madean impact in 4 decades? J Rheumatol 26:2529-2533, 1999

83. Tabib A, Leroux C, Mornex JF, et al: Acceler-ated coronary atherosclerosis and arterioscle-rosis in young human-immunodeficiency-vi-rus-positive patients. Coron Artery Dis 11:41-46, 2000

84. Kalantar-Zadeh K, Don BR, Rodriguez RA, etal: Serum ferritin is a marker of morbidity andmortality in hemodialysis patients. Am J Kid-ney Dis 37:564-572, 2001

85. Kalantar-Zadeh K, Ikizler A, Block G, et al.Malnutrition-inflammation complex syn-drome in dialysis patients: Causes and conse-quences. Am J Kidney Dis (in press)

86. Ifudu O, Uribarri J, Rajwani I, et al: Lowhematocrit may connote a malnutrition-in-flammation syndrome in hemodialysis pa-tients. Dial Transplantation 31:845-878, 2002

87. Stenvinkel P, Heimburger O, Lindholm B, etal: Are there two types of malnutrition inchronic renal failure? Evidence for relation-ships between malnutrition, inflammationand atherosclerosis (MIA syndrome). NephrolDial Transplant 15:953-960, 2000

88. Kaysen GA: Inflammation nutritional stateand outcome in end stage renal disease. MinerElectrolyte Metab 25:242-250, 1999

89. Flores EA, Bistrian BR, Pomposelli JJ, et al:Infusion of tumor necrosis factor/cachectinpromotes muscle catabolism in the rat: A syn-ergistic effect with interleukin 1. J Clin Invest83:1614-1622, 1989

90. McCarthy DO: Tumor necrosis factor alphaand interleukin-6 have differential effects onfood intake and gastric emptying in fastedrats. Res Nurs Health 23:222-228, 2000

91. Kaizu Y, Kimura M, Yoneyama T, et al: Inter-leukin-6 may mediate malnutrition in chronichemodialysis patients. Am J Kidney Dis 31:93-100, 1998

92. Kaysen GA, Dubin JA, Muller HG, et al: Re-lationships among inflammation nutritionand physiologic mechanisms establishing al-bumin levels in hemodialysis patients. KidneyInt 61:2240-2249, 2002

93. Kaysen GA, Chertow GM, Adhikarla R, et al:Inflammation and dietary protein intake exertcompeting effects on serum albumin and cre-atinine in hemodialysis patients. Kidney Int60:333-340, 2001

94. Ginn HE, Frost A, Lacy WW: Nitrogen bal-ance in hemodialysis patients. Am J Clin Nutr

21:385-393, 1968

95. Leon JB, Majerle AD, Soinski JA, et al: Can anutrition intervention improve albumin levelsamong hemodialysis patients? A pilot study.J Renal Nutr 11:9-15, 2001

96. Hulsewe KW, van Acker BA, von MeyenfeldtMF, et al: Nutritional depletion and dietarymanipulation: Effects on the immune re-sponse. World J Surg 23:536-544, 1999

97. De Simone C, Famularo G, Tzantzoglou S, etal: Carnitine depletion in peripheral bloodmononuclear cells from patients with AIDS:Effect of oral L-carnitine. AIDS 8:655-660, 1994

98. Kuhlmann MK, Schmidt F, Kohler H: Highprotein/energy vs: standard protein/energynutritional regimen in the treatment of mal-nourished hemodialysis patients. Miner Elec-trolyte Metab 25:306-310, 1999

99. Caglar K, Fedje L, Dimmitt R, et al: Therapeuticeffects of oral nutritional supplementation dur-ing hemodialysis. Kidney Int 62:1054-1059, 2002

00. Chertow GM, Ling J, Lew NL, et al: The asso-ciation of intradialytic parenteral nutrition ad-ministration with survival in hemodialysis pa-tients. Am J Kidney Dis 24:912-920, 1994

01. Capelli JP, Kushner H, Camiscioli TC, et al:Effect of intradialytic parenteral nutrition onmortality rates in end- stage renal diseasecare. Am J Kidney Dis 23:808-816, 1994

02. Wolfson M, Foulks CJ: Intradialytic parenteralnutrition: A useful therapy? Nutr Clin Pract11:5-11, 1996

03. Pupim LB, Flakoll PJ, Brouillette JR, et al: Intra-dialytic parenteral nutrition improves proteinand energy homeostasis in chronic hemodialy-sis patients. J Clin Invest 110:483-492, 2002

04. Seliger SL, Weiss NS, Gillen DL, et al: HMG-CoA reductase inhibitors are associated withreduced mortality in ESRD patients. KidneyInt 61:297-304, 2002

05. Stenvinkel P, Andersson P, Wang T, et al: DoACE-inhibitors suppress tumour necrosis fac-tor-alpha production in advanced chronic re-nal failure? J Intern Med 246:503-507, 1999

06. Boaz M, Smetana S, Weinstein T, et al: Second-ary prevention with antioxidants of cardiovas-cular disease in endstage renal disease(SPACE): Randomised placebo-controlledtrial. Lancet 356:1213-1218, 2000

07. Bloembergen WE, Hakim RM, Stannard DC,et al: Relationship of dialysis membrane andcause-specific mortality. Am J Kidney Dis 33:1-10, 1999

08. Eknoyan G, Beck GJ, Cheung AK, et al: Effectof dialysis dose and membrane flux in main-tenance hemodialysis. N Engl J Med 347:2010-

2019, 2002