Intradialytic Parenteral Nutrition Does Not Improve Survival in Malnourished Hemodialysis Patients: A 2-Year Multicenter, Prospective, Randomized Study
Noël J.M. Cano*,
Denis Fouque,
Hubert Roth,
Michel Aparicio,
Raymond Azar||,
Bernard Canaud¶,
Philippe Chauveau**,
Christian Combe,**,
Maurice Laville,
Xavier M. Leverve the French Study Group for Nutrition in Dialysis
* Service d’Hépatogastroenterologie et Nutrition, Clinique Résidence du Parc, Marseille, Service de Néphrologie, Hôpital Edouard Herriot, Lyon, INSERM-E0221 Bioénergétique Fondamentale et Appliquée, Grenoble, Université Bordeaux2 and CHU de Bordeaux and ** AURAD Aquitaine, Bordeaux, || Service de Néphrologie, Centre Hospitalier, Dunkerque, and ¶ Service de Néphrologie, Hôpital Lapeyronnie, Montpellier, France
Correspondence: Dr. Noël J.M. Cano, CRNH Auvergne, 58 Rue Montalembert, BP 321, 63009 Clermont-Ferrand cedex 1, France. Phone: +33-4-7360-8250; Fax: +33-4-7360-8255; E-mail: ncano{at}clermont.inra.fr
Received for publication February 11, 2007.
Accepted for publication May 3, 2007.
Although intradialytic parenteral nutrition (IDPN) is a methodused widely to combat protein-calorie malnutrition in hemodialysispatients, its effect on survival has not been thoroughly studied.We conducted a prospective, randomized trial in which 186 malnourishedhemodialysis patients received oral nutritional supplementswith or without 1 year of IDPN. IDPN did not improve 2-yearmortality (primary end point), hospitalization rate, Karnofskyscore, body mass index, or laboratory markers of nutritionalstatus. Instead, both groups demonstrated improvement in bodymass index and the nutritional parameters serum albumin andprealbumin (P < 0.05). Multivariate analysis showed thatan increase in prealbumin of >30 mg/L within 3 months, amarker of nutritional improvement, independently predicted a54% decrease in 2-year mortality, as well as reduced hospitalizationsand improved general well-being as measured by the Karnofskyscore. Therefore, although we found no definite advantage ofadding IDPN to oral nutritional supplementation, this is thefirst prospective study demonstrating that an improvement inprealbumin during nutritional therapy is associated with a decreasein morbidity and mortality in malnourished hemodialysis patients.
Despite the continuing progress in hemodialysis therapy, themortality rate of maintenance dialysis patients is unacceptablyhigh. In this population, protein-calorie malnutrition is independentlyassociated with an increase in morbidity and mortality.1,2 Severemalnutrition has been reported in 25% of maintenance hemodialysispatients2,3 and found to be associated with a yearly mortalityrate of approximately 30%.2,4,5 Intradialytic parenteral nutrition(IDPN) has been proposed to improve patient nutritional statusand outcome. The interest of IDPN has been assessed in termsof metabolic effect and nutritional benefit6: IDPN has beenshown to improve energy and protein balance, albumin synthesisrate, and, in randomized trials, nutritional parameters.7–11Retrospective studies have suggested that IDPN may improve survivalin hypoalbuminemic hemodialysis patients.4,5,12 However, todate, the effects of IDPN on patient morbidity and mortalityhave not been assessed in a prospective, randomized manner.More precise, the impact of albumin and prealbumin changes duringnutritional therapy on survival has not been addressed.13,14This investigator-initiated, prospective, randomized, controlled,French Intradialytic Nutrition Evaluation study (FineS) wasdesigned to evaluate in an intention-to-treat analysis the effectsof a 1-yr IDPN given in addition to oral supplements on the2-yr survival and morbidity. Moreover, this study aimed to definethe determinants of the outcome in malnourished maintenancedialysis patients who receive nutritional therapies.
Patients
Between January 2001 and December 2002, 186 patients were randomlyassigned to receive either IDPN plus oral supplements (IDPNgroup, n = 93) or oral supplements alone (control group, n =93) during 1 yr, then followed during a subsequent year (Figure 1).The two groups were similar with respect to baseline characteristics(Table 1). At months 3, 6, and 12, IDPN provided the equivalentof 6.6 ± 2.6, 6.4 ± 2.1, and 6.1 ± 2.2kcal and 0.26 ± 0.08, 0.25 ± 0.09, and 0.24 ±0.10 g protein/kg per d, respectively. Patients who actuallyreceived IDPN at months 3, 6, and 12 represented 87, 79, and67%, respectively, of the IDPN group. Causes for IDPN discontinuationwere nutritional status improvement after 86 to 275 d in sevenpatients, wish of patient in nine cases, adverse events in 11patients (nausea, 3; muscle pain, 2; hypertriglyceridemia, 1;arteriovenous fistula pain, 1; other adverse events, 4), andother reasons in five patients. At months 3, 6, and 12, oralsupplements provided 5.9 ± 2.6, 5.8 ± 2.5, and5.6 ± 2.7 kcal and 0.39 ± 0.18, 0.38 ±0.18, and 0.37 ± 0.18 g protein/kg per d, without between-groupdifference. In control and IDPN groups, mean compliance to oralsupplementation was respectively 72 and 69% after 3 mo, 68 and75% after 6 mo, and 70 and 61% after 12 mo (NS).
Figure 1. Number of patients who entered the study, were assigned to intradialytic parenteral nutrition (IDPN) or control group, completed the protocol, and were included in intention-to-treat analysis.
Table 1. Baseline data in control and IDPN groupsa
Primary Outcome
Thirty-six patients died during the 2-yr follow-up in the controlgroup and 40 in the IDPN group. Causes of death did not differbetween groups (Table 2). Mean cumulative survival was 0.77and 0.58 after 1 and 2 yr, respectively, similar in IDPN andcontrol groups (Figure 2). No difference appeared after adjustmentfor diabetes, serum C-reactive protein (CRP), and comorbidities.As compared with control subjects, patients with diabetes showeda similar survival at month 12 (2 = 0.38, P = 0.538), then tendedto have a decreased survival from months 12 to 24 (2 = 3.77,P = 0.052).
Figure 2. Kaplan-Meier survival analysis in control (black line) and IDPN (gray line) groups (NS). The number of patients on day 0 and months 3, 6, 12, 18, and 24 was, respectively, 93, 83, 70, 64, and 56 in the control group and 93, 80, 67, 55, and 46 in the IDPN group. One patient in the control group (day 530) and three patients in the IDPN group (days 208, 259, and 299) received a kidney transplant.
Secondary Outcomes
Karnofsky score was 66 ± 17 and 66 ± 14 in controland IDPN groups, respectively, on day 0 and remained unaffectedand similar in the two groups during the follow-up (data notshown). All patients but 27 were hospitalized at least onceduring the follow-up. The median hospitalization length was21 d. Hospitalization rate was 0.06 ± 0.10 and 0.06 ±0.15 in control and IDPN groups from day 0 to month 12 and 0.06± 0.11 and 0.08 ± 0.16 from month 12 to month24, respectively. Cardiovascular diseases represented 22.3%of hospitalization causes, infections 18.9%, digestive diseases13.4%, disability 7.9%, neurologic complications 7.1%, fallsand fractures 4.6%, cancer 4.6%, and other causes 21%, withoutbetween-group statistical difference.
Figure 3 shows nutritional data from day 0 to month 24. As comparedwith control subjects, IDPN patients exhibited lower spontaneousprotein intake at month 12 and higher total energy intake atmonths 3 and 6. No between-group difference was observed regardingother nutritional data. More than 50% of patients received nutritionalsupport after the 1-yr randomized treatment as per physician'sdecision. At months 18 and 24, oral supplements were given to44 and 55% of control subjects and 54 and 50% of IDPN patients,respectively; IDPN was administered in 6 and 9% of control patientsand in 13 and 17% of IDPN patients (NS). In both groups, spontaneousintake was stable during the follow-up. Paired tests showedthat in both groups, total energy, total protein intake, andnormalized protein nitrogen appearance (nPNA) increased fromday 0 to months 3, 6, and 12. Nutritional support was followedby an increase in body weight at months 3, 6, and 12 in IDPNpatients and at month 3 in control subjects. In addition, bothgroups showed an increase in serum albumin and prealbumin fromday 0 to month 3. Serum albumin remained elevated until month18 and serum prealbumin until the end of the follow-up (month24). In the two groups, serum CRP did not vary during the follow-up.
Figure 3. Changes in spontaneous (spont.) and total energy and protein intakes, body mass index (BMI), serum albumin, prealbumin, and normalized protein nitrogen appearance (nPNA) during the 2-yr follow-up in control (black line) and IDPN (gray line) groups (means ± SEM). Between-group differences: *P < 0.05; **P < 0.01. Nutritional therapies were followed by a significant increase in BMI at months 3, 6, and 12 in the IDPN group (P < 0.01) and at month 3 in the control group (P < 0.05). In both groups, nutritional support induced an increase in serum albumin at months 3, 6, 12, and 18 (P < 0.01) and in serum prealbumin at months 3 to 24 (P < 0.02).
Predictors of Primary Outcome
Because the two groups did not differ with respect to outcomes,the predictors of primary and secondary end points were studiedin all patients as a single group. Univariate analysis showedthat diabetes and the number of comorbidities increased themortality risk, whereas dialysis vintage and baseline valuesof Karnofsky score, serum albumin, and creatinine were negativelycorrelated with mortality. Mean cumulative survival was 0.74and 0.37 after 1 and 2 yr in patients with diabetes and 0.78and 0.64 in patients without diabetes (P < 0.01 at 2 yr).The increase in Karnofsky score and serum prealbumin from day0 to month 3 predicted an improved survival. Baseline serumCRP did not predict any of the outcomes. Multivariate analysisshowed four independent predictors of the 2-yr mortality (Figure 4):Number of comorbidities (odds ratio [OR] 1.53; 95% confidenceinterval [CI] 1.14 to 2.05 per comorbidity), baseline serumalbumin (OR 0.93; 95% CI 0.89 to 0.98 per g/L) and creatinine(OR 0.98; 95% CI 0.97 to 0.99 per 10 µmol/L), and serumprealbumin increase >30 mg/L from day 0 to month 3 (OR 0.46;95% CI 0.27 to 0.79).
Figure 4. Independent predictors of mortality: Multivariate Cox regression analysis.
Predictors of Secondary Outcomes
The activity score weakly correlated with serum prealbumin changesfrom day 0 to month 3 (r = 0.286, P < 0.01 for Karnofskyscore change from day 0 to month 18). Multivariate logisticregression showed that only a serum prealbumin increase independentlypredicted the risk to be hospitalized for >21 d (OR 0.24;95% CI 0.10 to 0.59 for a serum prealbumin increase >30 mg/L).Albumin and prealbumin changes from day 0 to month 3 were negativelycorrelated with CRP changes during the same period (r = –0.474,P < 0.001, and r = –0.461, P < 0.001, respectively).However, at uni- and multivariate logistic regression analyses,CRP changes did not predict the serum albumin increase overthe critical threshold of 35 g/L or the prealbumin increaseby >30 mg/L from day 0 to month 3. Only nPNA determined the30 mg/L increase of serum prealbumin from day 0 to month 3 (OR1.34; 95% CI 1.16 to 1.55 per 0.1 g/kg per d). The increasein serum albumin and prealbumin during the follow-up was independentfrom baseline serum CRP. At day 0, 50% of patients presentedwith serum CRP >10 mg/L. When all patients were separatedinto two groups according to their baseline serum CRP levels,it seemed that patients with inflammation, as defined by serumCRP >10 mg/L, were characterized by lower baseline serumalbumin and prealbumin concentrations. However, the increasein serum albumin and prealbumin during nutritional support wasobserved independent from baseline serum CRP (Figure 5). Atmonth 18, the increase in serum albumin was even greater inpatients with inflammation.
Figure 5. Serum albumin and prealbumin changes from day 0 to month 24 in patients with serum C-reactive protein (CRP) <10 mg/L (n = 88, blue line) or 10 mg/L (n = 86, red line). Baseline albumin and prealbumin were lower in patients with serum CRP 10 mg/L (P < 0.05). A significant increase in serum albumin and prealbumin was observed irrespective of baseline CRP. At month 18, the increase in serum albumin was greater in patients with the higher baseline CRP concentrations.
At baseline, patients with diabetes exhibited a higher bodymass index (BMI) than patients without diabetes (24.1 ±0.7 versus 22.3 ± 0.3; P < 0.05) and similar serumalbumin (31.1 ± 4.2 versus 31.7 ± 3.7 g/L) andprealbumin (235 ± 52 versus 241 ± 52 mg/L). Asshown in Figure 6, similar to patients without diabetes, patientswith diabetes exhibited a significant increase in serum albuminand prealbumin from day 0 to months 3 and 6. After month 6,serum albumin but not prealbumin decreased in patients withdiabetes as compared with patients without diabetes.
Figure 6. Serum albumin and prealbumin changes from day 0 to month 24 in patients without (n = 141, blue line) and with (n = 45, red line) diabetes. Between-group differences: *P < 0.05; **P < 0.01. Nutritional therapies were followed by a significant increase in serum albumin, from months 3 to 24 in patients without diabetes (P < 0.01) and at months 3 and 6 in patients with diabetes (P < 0.01). Serum prealbumin increased from months 3 to 24 in the two groups.
Adverse Events
The most frequent adverse effects occurred with a similar frequencyin the two groups: Digestive symptoms, hypotension, and musclecramps (Table 2). They were responsible for IDPN discontinuationin nine cases. No between-group difference was observed accordingto liver function tests, vascular access–related symptoms,and plasma triglycerides. Arteriovenous fistula pain and hypertriglyceridemiacaused IDPN discontinuation in one case each.
This study is the first prospective, randomized, controlledtrial to address in an intention-to-treat design the effectof IDPN on mortality and morbidity in malnourished hemodialysispatients. Ninety-three patients were randomly assigned to receiveIDPN at each hemodialysis session for 1 yr, and 93 were consideredas control subjects and did not receive IDPN. The randomizationprocedure resulted in comparable study groups, although an NStrend to more comorbidities was observed in the IDPN group.Both control and IDPN groups received oral supplements. Bothgroups exhibited a similar improvement in nutritional status.Mortality rate was not different between the two groups (42%over 2 yr). Similarly, hospitalization rate and changes in Karnofskyscore were not influenced by the addition of IDPN.
At first analysis, these negative results question the studypower. Given the number of inclusions achieved and the initialhypothesis, the study power (1 – risk) was calculatedto be 78%. Because no tendency to a lesser morbidity and mortalitywas noticed in the IDPN group, it seems unlikely that a highernumber of patients would have allowed us to show a beneficialeffect of IDPN. Indeed, a tentative estimation of sample sizeusing the 1-yr efficacy observed in this study showed that aminimum of 1364 patients in each group would have been necessaryto observe potentially a difference in the treatment effectwith a power of 80%. Such a large 2700-patient study is unlikelyto be planned in this area. The lack of effect of adding IDPNto oral supplements was consistent with the nutritional response:No nutritional benefit was noted at each time point from day0 to month 24. The tendency to a decrease in survival from months12 to 24 in patients with diabetes from IDPN group may havebeen due to a deleterious effect of IDPN-induced hyperglycemia,as reported in intensive care unit patients15 and during totalparenteral nutrition.16
In both groups, BMI, serum albumin and prealbumin increasedduring oral supplementation without additional effect of IDPN.For ethical reasons, no control group without nutritional supportwas studied. However, the beneficial effect of nutritional supplementationis supported by the analysis of body weight and serum albuminchanges during the 6 mo before inclusion and during nutritionaltherapies: Although the two parameters significantly worsenedbefore inclusion, they strikingly improved during supplementation(Figure 7). This dramatic spontaneous degradation of nutritionalstatus before intervention, which was similar in control andIDPN-randomized patients, argues in favor of the design of ourstudy, which included a minimal nutritional support in the controlgroup. Previous randomized studies showed that IDPN was ableto improve body weight, serum albumin, and prealbumin in malnourishedhemodialysis patients.9–11 In this study, protein andenergy requirements were obtained apart from the administrationof IDPN. Such an efficacy of oral supplements may explain thelack of benefit from IDPN addition.
Figure 7. Body weight (blue line) and serum albumin (red line, n = 121) changes before, during, and after nutritional therapies. Both body weight and serum albumin decreased from month –6 to day 0 then increased after nutritional support initiation (P < 0.05).
A limitation of this study is that urine collection was notperformed. However, because the number of patients with urineoutput >500 ml/d was similar in the two groups and representedonly four patients in each group (Table 1), it is unlikely thatthis limitation affected the between-group comparison of nPNAvalues. IDPN was discontinued in seven patients who were randomlyassigned to IDPN as a result of an improvement of nutritionalstatus. Such decision by physicians in charge of patients correspondsto a "real life" situation that should be considered in theintention-to-treat analysis. The follow-up of these seven patientsled to two deaths, one transplantation, and four living patientsat 2 yr. Because such a death rate was slightly lower than forall patients, it seems unlikely that IDPN discontinuation influencedthe comparisons between the two groups.
In all patients, mean changes in serum albumin and prealbuminfrom day 0 to month 3 were, respectively, 1.6 g/L and 30 mg/L.Among the 167 patients who presented with serum albumin <35g/L, one of the criteria for inclusion, 62 reached albumin levels35 g/L at month 3. Similarly, among the 159 patients who presentedwith serum prealbumin <300 mg/L, another criterion for inclusion,62 reached prealbumin levels 300 mg/L at month 3. It is interestingthat during the same period, 80 (43%) patients exhibited a serumprealbumin increase of >30 mg/L, associated with a two-foldimprovement of the 2-yr survival. This study is the first toaddress the predictors of mortality in hemodialysis patientswho receive nutritional therapy. Four independent parametersassociated with mortality were identified. Besides the well-documentedinfluence of comorbidities, baseline serum albumin, and creatinine,1,17it is worth emphasizing that the early increase in serum prealbuminduring nutritional support independently predicted survival.This finding demonstrates that nutritional therapy was associatedwith increased survival when nutritional status, as assessedby serum prealbumin, was improved. Moreover, these results allowphysicians to identify patients in whom an improvement of survivalcould be expected, depending on the early nutritional response.Previous reports showed that hospitalization risk is determinedby nPNA and serum albumin.18–20 In this study, serum prealbuminincrease during nutritional support also appeared as an independentpredictor of hospitalization. Serum prealbumin is now widelyaccepted as a sensitive and reliable marker of nutritional statusin maintenance hemodialysis patients.20–23 These dataexhibit the prognostic value of serum prealbumin changes duringnutritional support. They also show the major role of proteinintake because only nPNA determined serum prealbumin increase.
Besides comorbidities, causes of protein loss that may haveinfluenced the response to nutritional therapy include acidosis,dialysis procedure, inflammation, and diabetes.24,25 Our datadid not make it possible to evaluate the role of acidosis inthe response to nutritional support. Kt/V values (1.7 ±0.3 on day 0) did not vary during the follow-up, attesting todialysis adequacy. In these conditions, dialysis procedure,as assessed by Kt/V, the use of high-permeability membranes,hemodialysis, or hemodiafiltration, did not influence nutritionaland outcome parameter changes during the 2-yr follow-up. Inhemodialysis patients, inflammation was reported to decreaseappetite and to induce protein catabolism.26 In 79 well-dialyzedpatients without nutritional intervention, Kaysen et al.27 demonstratedthat inflammation and reduced albumin synthesis were the principalcause of decrease in serum albumin, whereas protein intake remainedstable. Conversely, protein energy supplementation by IDPN wasdemonstrated to enhance albumin synthesis8 and to increase serumalbumin as well as prealbumin.9,11 The effect of inflammationon the response to nutritional therapy is poorly documented.In a pilot study, Leon et al.28 reported that serum CRP didnot alter the response to nutritional intervention, as assessedby serum albumin. In this study, inflammation, as assessed byserum CRP on day 0, did not alter the nutritional response tonutritional support. Although albumin and prealbumin changeswere negatively correlated with CRP changes from day 0 to month3, logistic regression analyses failed to show a predictivevalue of CRP changes for the serum albumin increase over thecritical threshold of 35 g/L or the prealbumin increase by >30mg/L. Furthermore, the beneficial effect of the prealbumin increaseon mortality was independent from baseline serum CRP concentration.These data strongly argue for the provision of a nutritionalsupplementation in malnourished hemodialysis patients, irrespectiveof their inflammatory status.
Diabetes did not alter the early response to nutritional supportbut was associated with a less sustained increase in serum albumin.Patients with diabetes were characterized by an increased mortalityduring the second year of follow-up. It is noticeable that,opposite to the picture observed in patients without diabetes,the increase in serum prealbumin by >30 mg/L from day 0 tomonth 3 did not predict an improvement of survival in patientswith diabetes. These data are consistent with a previous reportshowing that, despite a higher prevalence of protein malnutrition,survival of patients with diabetes was independent from nutritionalstatus.29 Besides insulin resistance,30 increased inflammatoryprocess and oxidative stress have been advocated as possiblecauses of lower survival in hemodialysis patients with diabetes.In this study, serum CRP was not influenced by diabetes. Inpatients with diabetes, glycosylated hemoglobin was 6.98 ±1.16% on day 0 and not influenced by oral supplementation orIDPN.
This study showed three main findings: (1) In malnourished hemodialysispatients, the intention-to-treat analysis failed to show anyadvantage of adding IDPN to oral supplements; (2) in patientswithout diabetes, nutritional supplementation was associatedwith a dramatic and sustained improvement in nutritional status;and (3) the increase in serum prealbumin during nutritionaltherapy was an independent predictor of mortality and hospitalizationrisk during a 2-yr follow-up.
Study Design
Our first objective was to determine the effects of IDPN onsurvival, morbidity, and nutritional status of malnourishedhemodialysis patients receiving oral supplements. The additionalobjective was to identify the factors that determine primaryand secondary end points. The 3583 patients from the 38 hemodialysiscenters belonging to the French Study Group of Nutrition inDialysis were screened. Inclusion criteria were age between18 and 80 yr; hemodialysis vintage >6 mo; and two of thefollowing markers of malnutrition: BMI <20 kg/m2, body weightloss within 6 mo >10%, serum albumin <35 g/L, and serumprealbumin <300 mg/L. Exclusion criteria were weekly dialysistime <12 h; urea Kt/V <1.2; comorbidities compromisingthe 1-yr survival (evolutive cancer and AIDS); treatment byoral, enteral, or parenteral feeding during the past 3 mo; andhospitalization at time of randomization.
Two arms were considered: A treated group, receiving IDPN during1 yr, and a control group. For ethical reasons, given the pooroutcome associated with malnutrition in maintenance dialysis,both control and IDPN groups received oral supplements duringthe same period. Standard oral supplements were given on a basisof 500 kcal/d and 25 g/d protein, according to each physician'susual practice, and the compliance was assessed during dietaryinterviews at each time point (see End Points). Rules for IDPNdelivery were given to physicians who cared for patients: (1)The nonprotein energy and protein supply should fulfill thedifference between spontaneous intakes as estimated by dietaryinterview and recommended intakes (i.e. 30 to 35 kcal/d and1.2 g protein/kg per d31; (2) a standard lipid emulsion shouldrepresent 50% and glucose 50% of nonprotein energy supply; (3)nitrogen supply should be a standard amino acid solution; and(4) the rate of infusion should be constant and not exceed 125ml/h during the first week, then 250 ml/h during the dialysissession. The amount of fluid infused was fully compensated byultrafiltration. Four grams of sodium chloride was added perliter of IDPN solution to compensate Na losses as a result ofultrafiltration.4
End Points
The primary end point was all-cause mortality decrease in theIDPN group, and secondary end points were hospitalization rate,Karnofsky score,32 BMI, serum albumin, and prealbumin. Mortality,causes of death, causes and length of hospitalizations, Karnofskyscore, nutritional parameters, and adverse events were collectedat day 0 and after 3, 6, 12, 18, and 24 mo by two clinical datamonitors. The hospitalization rate was defined as the ratioof the number of days of hospitalization per day of follow-up.Spontaneous dietary intake was determined at each time pointby a 3-d food report including one dialysis day using the SU-VI-MAXfood picture book.33 Data were computed (Bilnut 4.0 SCDA Nutrisoft,Le Hallier, Cerelles, France) using the French Data Base CIQUAL(Centre Informatique sur la Qualité des Aliments, AgenceFrançaise de Sécurité Sanitaire des Aliments).Pre- and postdialysis body mass was recorded from a single midweekdialysis session. On the same day, predialysis hemoglobin, serumCRP, albumin, prealbumin, alanine amino transferase, glutamyltransferase, triglycerides, cholesterol, and pre- and postdialysisurea and creatinine concentrations were determined by the usuallaboratories of the different centers using conventional autoanalyzers.Immunonephelometry was used for serum CRP, albumin, and prealbuminmeasurements. This method was reported to exhibit the lowerintercenter variations and to be the most reproducible methodfor measuring these plasma proteins.34 Dialysis adequacy wasestimated by urea Kt/V.35,36 nPNA, a reflection of protein intakein stable conditions, was calculated from urea generation rateafter measurements of pre- and postdialysis plasma urea.37
Statistical Analyses
Sample sizes were determined according to the primary end point.Considering a spontaneous yearly mortality rate of 30% and and error types of 5 and 20%, respectively, the number of patientsrequired to show a 10% reduction of mortality rate (from 30down to 20%) was 102 in each group (NCSS-PASS software, Kaysville,UT). Randomization was stratified by center: Each center receivedone or more blocks of six sequentially numbered opaque sealedenvelopes.
Data are presented as means ± SD. OR are given with a95% CI. Statistical tests were realized with Stata8 (Stata Corp.,College Station, TX). The level of significance was set at 0.05.Between-group comparisons were performed in intention-to-treatanalysis using t test for continuous variables, 2 tests forcategorical variables, and generalized estimating equationsmethod for longitudinal data. Survival analysis was performedusing Kaplan-Meier graphs and log-rank tests.
Age; gender; presence of diabetes; serum CRP; number of comorbidities;baseline nutritional parameters; energy and protein suppliesthrough oral supplementation, IDPN, and whole energy and proteinintakes; number of IDPN sessions; and changes in BMI, serumalbumin, prealbumin, and nPNA during nutritional support weretested for their predictive value of survival using the univariateCox proportional hazard model. Variables exhibiting a significant(with P < 0.20) predictive value of survival in univariateanalysis were then tested in a multivariate Cox model integratingIDPN, diabetes, and age. The same parameters were tested fortheir predictive value of secondary outcomes using simple regressionor uni- and multivariate logistic regression studies.
Ethics
The protocol was approved by the Ethics Committee of Grenoble,France, and registered on clinicaltrials.gov (NCT00314834).Data were analyzed by H.R. and N.J.M.C.
Grants were obtained from INSERM, Société Francophonede Nutrition Entérale et Parentérale; SociétéFrancophone de dialyze; and the following companies: Amgen,Baxter, Braun, DHN-Celia, Fresenius Kabi, Fresenius MedicalCare, Gambro, Hospal, Meditor, Nestlé Clinical Nutrition,Novartis, Nutricia, Ortho-Biotech, Roche, Sigma-tau, and Sorin-Bellco.
We thank the clinical data monitors Leslie Berrerd and KatiaMaurizi and the following physicians from the French Study Groupfor Nutrition in Dialysis, in charge of patient recruitmentand care: R. Fraissinet and J. Legros, Centre Hospitalier Général,Aix-En Provence; M. Al Adib, J.M. Marc, Centre Hospitalier,Annonay; D. Erbilgin, Centre d'hémodialyse, Arles; F.Degroc, Clinique Delay, Bayonne; E. Mac Namara, Centre Hospitalier,Béthune; F. Maurice, Centre de Néphrologie duBitterois, Béziers; V. de Précigout, C. Lasseur,K. Moreau, Centre Hospitalo-Universitaire Pellegrin, Bordeaux;J.L. Boucher, P. Martin-Dupont, CTMR-St. Augustin, Bordeaux;P. Bataille, Centre Hospitalier Duchenne, Boulogne sur Mer;J.C. Glachant, Centre Hospitalier, Bourg en Bresse; A. Abokasem,J.F. Cabanne, Centre Hospitalier, Chalon sur Saone; J. Potier,Centre Hospitalier Pasteur, Cherbourg; P. Collin, Centre Hospitalier,Cholet; R. Azar, Centre Hospitalier Général, Dunkerque;M. Achache, A.G.D.U.C., La Tronche; P. Henri, Centre HospitalierE. Bisson, Lisieux; Combarnous, Hôpital E. Herriot, Lyon;P. Leitienne, Centre d’hémodialyse, Pinel, Lyon;A. Heyani, A. Nefti, Centre Hospitalier, Macon; M. Lankester,Centre d’hémodialyse de la Résidence duParc, Marseille, P. Brunet, Hôpital Ste-Marguerite, Marseille;T. Zerrouki, Centre Hospitalier, Meaux; H. Leray, HôpitalLapeyronie, Montpellier; F. Maurice, Centre d’Hémodialysedu Languedoc Méditerranéen, Montpellier; C. Delcroix,Centre Hospitalo-Universitaire, Nantes; A. Cremeau, Centre Hospitalier,Nevers; A. Kolko, Centre Médical E. Rist, Paris; C. Bony,A.U.R.A., Paris; N. Pertuiset Centre Hospitalier, Poissy; M.Maheut, Centre Hospitalo-Universitaire, Reims; A. Haddj-Elrabet,V. Joyeux, Centre Hospitalo-Universitaire Régional Pontchaillou,Rennes; J.P. Guy, Centre Hospitalier, Saint-Claude; L. Azzouz,C. Broyet, Centre Hospitalo-Universitaire, Saint-Etienne; C.Ngohou, Centre Hospitalier, Saint-Herblain; C. Chazot, F. Maazoun,Centre du Rein Artificiel, Tassin; B. Birmele, J. Pengloan,Centre Hospitalo-Universitaire, Tours; R. Binaut, V. Lemaitre,Centre Hospitalier, Valenciennes.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
N.J.M.C. and D.F. contributed equally to the study.
Owen WF, Lew NL, Liu Y, Lowrie EG, Lazarus JM: The urea reduction rate and serum albumin concentration as predictors of mortality in patients undergoing hemodialysis.
N Engl J Med 329
: 1001
–1006, 1993[Abstract/Free Full Text]
Combe C, Chauveau P, Laville M, Fouque D, Azar R, Cano N, Canaud B, Roth H, Leverve X, Aparicio M: Influence of nutritional factors and hemodialysis adequacy on the survival of 1,610 French patients.
Am J Kidney Dis 37[Suppl 2]
: S81
–S88, 2001
Aparicio M, Cano N, Chauveau P, Azar R, Canaud B, Laville M, Leverve XM: Nutritional status of hemodialysis patients: A French National Cooperative Study.
Nephrol Dial Transplant 14
: 1679
–1686, 1999[Abstract/Free Full Text]
Capelli JP, Kushner H, Camiscioli TC, Chen SM, Torres MA: Effect of intradialytic parenteral nutrition on mortality rates in end-stage renal disease care.
Am J Kidney Dis 23
: 808
–816, 1994
Chertow GM, Ling J, Lew NL, Lazarus JM, Lowrie EG: The association of intradialytic parenteral nutrition administration with survival in hemodialysis patients.
Am J Kidney Dis 24
: 912
–920, 1994
Cano N: Intradialytic parenteral nutrition: Where do we go from here?
J Ren Nutr 14
: 3
–5, 2004[CrossRef][Medline]
Pupim LB, Flakoll PJ, Brouillette JR, Levenhagen DK, Hakim RM, Ikizler TA: Intradialytic parenteral nutrition improves protein and energy homeostasis in chronic hemodialysis patients.
J Clin Invest 110
: 483
–492, 2002[CrossRef][Medline]
Pupim LB, Flakoll PJ, Ikizler TA: Nutritional supplementation acutely increases albumin fractional synthetic rate in chronic hemodialysis patients.
J Am Soc Nephrol 15
: 1920
–1926, 2004[Abstract/Free Full Text]
Cano N, Labastie-Coeyrehourcq J, Lacombe P, Stroumza P, di Costanzo-Dufetel J, Durbec JP, Coudray-Lucas C, Cynober L: Predialytic parenteral nutrition with lipids and amino-acids in malnourished hemodialysis patients.
Am J Clin Nutr 52
: 726
–730, 1990[Abstract/Free Full Text]
Navarro JF, Mora C, Leon C, Martin-Del Rio R, Macia ML, Gallego E, Chahin J, Mendez ML, Rivero A, Garcia J: Amino acid losses during hemodialysis with polyacrylonitrile membranes: Effect of intradialytic amino acid supplementation on plasma amino acid concentrations and nutritional variables in nondiabetic patients.
Am J Clin Nutr 71
: 765
–773, 2000[Abstract/Free Full Text]
Cano NJ, Saingra Y, Dupuy AM, Lorec-Penet AM, Portugal H, Lairon D, Cristol JP, Come A, Le Brun A, Atlan P, Leverve XM: Intradialytic parenteral nutrition: Comparison of olive oil versus soybean oil-based lipid emulsions.
Br J Nutr 95
: 152
–159, 2006[CrossRef][Medline]
Foulks CJ: The effect of intradialytic parenteral nutrition on hospitalization rate and mortality in malnourished hemodialysis patients.
J Ren Nutr 4
: 5
–10, 1994[Medline]
Chauveau P, Nguyen H, Combe C, Chene G, Azar R, Cano N, Canaud B, Fouque D, Laville M, Leverve X, Roth H, Aparicio M: Dialyzer membrane permeability and survival in hemodialysis patients.
Am J Kidney Dis 45
: 565
–571, 2005[CrossRef][Medline]
Kalantar-Zadeh K, Kilpatrick RD, Kuwae N, McAllister CJ, Alcorn H Jr, Kopple JD, Greenland S: Revisiting mortality predictability of serum albumin in the dialysis population: Time dependency, longitudinal changes and population-attributable fraction.
Nephrol Dial Transplant 20
: 1880
–1888, 2005[Abstract/Free Full Text]
Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P, Bouillon R: Intensive insulin therapy in critically ill patients
N Engl J Med 345
: 1359
–1367, 2001[Abstract/Free Full Text]
Cheung NW, Zaccaria C, Napier B, Fletcher JP: Hyperglycemia is associated with adverse outcomes in patients receiving total parenteral nutrition.
Diabetes Care 28
: 2367
–2371, 2005[Abstract/Free Full Text]
Goodkin DA, Bragg-Gresham JL, Koenig KG, Wolfe RA, Akiba T, Andreucci VE, Saito A, Rayner HC, Kurokawa K, Port FK, Held PJ, Young EW: Association of comorbid conditions and mortality in hemodialysis patients in Europe, Japan, and the United States: The Dialysis Outcomes and Practice Patterns Study (DOPPS).
J Am Soc Nephrol 14
: 3270
–3277, 2003[Abstract/Free Full Text]
Rocco MV, Soucie JM, Reboussin DM, McClellan WM: Risk factors for hospital utilization in chronic dialysis patients. Southeastern Kidney Council (Network 6).
J Am Soc Nephrol 7
: 889
–896, 1996[Abstract]
Kalantar-Zadeh K, Supasyndh O, Lehn RS, McAllister CJ, Kopple JD: Normalized protein nitrogen appearance is correlated with hospitalization and mortality in hemodialysis patients with Kt/V greater than 1.20.
J Ren Nutr 13
: 15
–25, 2003[CrossRef][Medline]
Chertow GM, Goldstein-Fuchs DJ, Lazarus JM, Kaysen GA: Prealbumin, mortality, and cause-specific hospitalization in hemodialysis patients.
Kidney Int 68
: 2794
–2800, 2005[CrossRef][Medline]
Avram MM, Sreedhara R, Fein P, Oo KK, Chattopadhyay J, Mittman N: Survival on hemodialysis and peritoneal dialysis over 12 years with emphasis on nutritional parameters.
Am J Kidney Dis 37[Suppl 2]
: S77
–S80, 2001
Cano NJ: Metabolism and clinical interest of serum transthyretin (prealbumin) in dialysis patients.
Clin Chem Lab Med 40
: 1313
–1319, 2002[CrossRef][Medline]
Kopple JD, Mehrotra R, Suppasyndh O, Kalantar-Zadeh K: Observations with regard to the National Kidney Foundation K/DOQI clinical practice guidelines concerning serum transthyretin in chronic renal failure.
Clin Chem Lab Med 40
: 1308
–1312, 2002[CrossRef][Medline]
Mitch WE: Robert H. Herman Memorial Award in Clinical Nutrition Lecture, 1997. Mechanisms causing loss of lean body mass in kidney disease.
Am J Clin Nutr 67
: 359
–366, 1998[Abstract]
Mitch WE: Malnutrition: A frequent misdiagnosis for hemodialysis patients.
J Clin Invest 110
: 437
–439, 2002[CrossRef][Medline]
Kalantar-Zadeh K, Block G, McAllister CJ, Humphreys MH, Kopple JD: Appetite and inflammation, nutrition, anemia, and clinical outcome in hemodialysis patients.
Am J Clin Nutr 80
: 299
–307, 2004[Abstract/Free Full Text]
Kaysen GA, Dubin JA, Muller HG, Rosales L, Levin NW, Mitch WE: Inflammation and reduced albumin synthesis associated with stable decline in serum albumin in hemodialysis patients.
Kidney Int 65
: 1408
–1415, 2004[CrossRef][Medline]
Leon JB, Majerle AD, Soinski JA, Kushner I, Ohri-Vachaspati P, Sehgal AR: Can a nutrition intervention improve albumin levels among hemodialysis patients? A pilot study.
J Ren Nutr 11
: 9
–15, 2001[Medline]
Cano NJ, Roth H, Aparicio M, Azar R, Canaud B, Chauveau P, Combe C, Fouque D, Laville M, Leverve XM: Malnutrition in hemodialysis diabetic patients: Evaluation and prognostic influence.
Kidney Int 62
: 593
–601, 2002[CrossRef][Medline]
Pyorala M, Miettinen H, Laakso M, Pyorala K: Plasma insulin and all-cause, cardiovascular, and noncardiovascular mortality: The 22-year follow-up results of the Helsinki Policemen Study.
Diabetes Care 23
: 1097
–1102, 2000[Abstract/Free Full Text]
National Kidney Foundation. Kidney Disease Outcomes Quality Initiative. Clinical practice guidelines for nutrition in chronic renal failure. I: Adult guidelines. A. Maintenance dialysis.
Am J Kidney Dis 35[Suppl 2]
: S17
–S55, 2000
Rettig RA, Sadler JH, Meyer KB, Wasson JH, Parkerson GR Jr, Kantz B, Hays RD, Patrick DL: Assessing health and quality of life outcomes in dialysis: A report on an Institute of Medicine workshop.
Am J Kidney Dis 30
: 140
–155, 1997
Hercberg S, Preziosi P, Briancon S, Galan P, Triol I, Malvy D, Roussel AM, Favier A: A primary prevention trial using nutritional doses of antioxidant vitamins and minerals in cardiovascular diseases and cancers in a general population: The SU.VI.MAX study—Design, methods, and participant characteristics. SUpplementation en VItamines et Mineraux AntioXydants.
Control Clin Trials 19
: 336
–351, 1998[CrossRef][Medline]
Cardenas D, Blonde-Cynober F, Ziegler F, Cano N, Cynober L: Should a single centre for the assay of biochemical markers of nutritional status be mandatory in multicentric trials?
Clin Nutr 20
: 553
–558, 2001[CrossRef][Medline]
Garred L, Barichello D, DiGiuseppe B, McCready W, Canaud B: Simple Kt/V formulas based on urea mass balance theory.
ASAIO J 40
: 997
–1004, 1994[CrossRef][Medline]
Daugirdas JT, Greene T: Dialysis dose as a determinant of adequacy.
Semin Nephrol 25
: 76
–80, 2005[CrossRef][Medline]
Garred LJ, Canaud BC, McCready WG: Optimal dialysis. The role of quantification.
Semin Dial 7
: 236
–245, 1994
This article has been cited by other articles:
D. Fouque, J. McKenzie, R. de Mutsert, R. Azar, D. Teta, M. Plauth, N. Cano, and the Renilon Multicentre Trial Study Group Use of a renal-specific oral supplement by haemodialysis patients with low protein intake does not increase the need for phosphate binders and may prevent a decline in nutritional status and quality of life
Nephrol. Dial. Transplant.,
September 1, 2008;
23(9):
2902 - 2910.
[Abstract][Full Text][PDF]
T. A. Ikizler, P. Stenvinkel, and B. Lindholm Resolved: Being Fat Is Good for Dialysis Patients: The Godzilla Effect: Pro
J. Am. Soc. Nephrol.,
June 1, 2008;
19(6):
1059 - 1064.
[Abstract][Full Text][PDF]
P. Stenvinkel, J. J. Carrero, J. Axelsson, B. Lindholm, O. Heimburger, and Z. Massy Emerging Biomarkers for Evaluating Cardiovascular Risk in the Chronic Kidney Disease Patient: How Do New Pieces Fit into the Uremic Puzzle?
Clin. J. Am. Soc. Nephrol.,
March 1, 2008;
3(2):
505 - 521.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
2 = 0.38, P = 0.538), then tended to have a decreased survival from months 12 to 24 (
risk) was calculated to be 78%. Because no tendency to a lesser morbidity and mortality was noticed in the IDPN group, it seems unlikely that a higher number of patients would have allowed us to show a beneficial effect of IDPN. Indeed, a tentative estimation of sample size using the 1-yr efficacy observed in this study showed that a minimum of 1364 patients in each group would have been necessary to observe potentially a difference in the treatment effect with a power of 80%. Such a large 2700-patient study is unlikely to be planned in this area. The lack of effect of adding IDPN to oral supplements was consistent with the nutritional response: No nutritional benefit was noted at each time point from day 0 to month 24. The tendency to a decrease in survival from months 12 to 24 in patients with diabetes from IDPN group may have been due to a deleterious effect of IDPN-induced hyperglycemia, as reported in intensive care unit patients15 and during total parenteral nutrition.16 
35 g/L at month 3. Similarly, among the 159 patients who presented with serum prealbumin <300 mg/L, another criterion for inclusion, 62 reached prealbumin levels 
glutamyl transferase, triglycerides, cholesterol, and pre- and postdialysis urea and creatinine concentrations were determined by the usual laboratories of the different centers using conventional autoanalyzers. Immunonephelometry was used for serum CRP, albumin, and prealbumin measurements. This method was reported to exhibit the lower intercenter variations and to be the most reproducible method for measuring these plasma proteins.34 Dialysis adequacy was estimated by urea Kt/V.35,36 nPNA, a reflection of protein intake in stable conditions, was calculated from urea generation rate after measurements of pre- and postdialysis plasma urea.37 
and 
