Association between the malnutrition-inflammation score and post-transplant anaemia

Transcription

1 Nephrol Dial Transplant (2011) 26: doi: /ndt/gfq690 Advance Access publication 29 November 2010 Original Articles Association between the malnutrition-inflammation score and post-transplant anaemia Miklos Z. Molnar 1,2,3, Maria E. Czira 2, Anna Rudas 1, Akos Ujszaszi 1, Bela Haromszeki 1, Janos P. Kosa 4, Peter Lakatos 4, Gabriella Beko 5, Eniko Sarvary 6, Marina Varga 6, Katalin Fornadi 2,7, Marta Novak 2, Laszlo Rosivall 1,8, Istvan Kiss 9, Adam Remport 9, David J. Goldsmith 10, Csaba P. Kovesdy 11,12 and Istvan Mucsi 2,13 1 Institute of Pathophysiology, 2 Institute of Behavioral Sciences, Semmelweis University, Budapest, Hungary, 3 Harold Simmons Center for Chronic Disease Research & Epidemiology, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA, USA, 4 1st Department of Internal Medicine, 5 Central Laboratory, 6 Department of Transplantation and Surgery, 7 Department of Neurology, Semmelweis University, Budapest, Hungary, 8 Hungarian Academy of Sciences and Semmelweis University Research Group for Pediatrics and Nephrology, Budapest, Hungary, 9 Department of Nephrology, Szent Imre Hospital, Budapest, Hungary, 10 Renal Unit, Guy s Hospital, London, UK, 11 Division of Nephrology, Salem VA Medical Center, Salem, VA, USA, 12 Division of Nephrology, University of Virginia, Charlottesville, VA, USA and 13 Department of Medicine, Division of Nephrology, McGill University Health Centre, Montreal, QC, Canada Correspondence and offprint requests to: Miklos Zsolt Molnar; molnar@medformol.hu Abstract Background. Post-transplant anaemia (PTA) is common and is associated with adverse consequences. The proteinenergy wasting (PEW) syndrome is associated with erythropoietin resistance in patients on maintenance dialysis. We assessed the association between PEW and PTA in a large prevalent cohort of stable kidney-transplanted patients. Methods. Data from 942 prevalent kidney-transplanted patients were analysed. Socio-demographic parameters, laboratory results, transplantation-related data and medication were obtained from the charts. Biomarkers reflecting nutritional status and inflammation [serum leptin, albumin, interleukin-6 (IL-6), tumour necrosis factor-α (TNF-α) and C-reactive protein] were measured. Anthropometric measures and the malnutrition-inflammation score (MIS) were also tabulated. Anaemia was defined according to the guidelines of the American Society of Transplantation. Results. Mean age was 51 ± 13 years, 57% were males and 22% had diabetes. The prevalence of PTA was 33%. The haemoglobin (Hb) level significantly and negatively correlated with the MIS (rho = 0.316), marginally with serum TNF-α (rho = 0.079) and serum IL-6 (rho = 0.075) and positively with serum transferrin (r = 0.298), serum albumin (r = 0.274), abdominal circumference (r = 0.254) and serum leptin (rho = 0.152), P < 0.05 for all. In a multivariable linear regression model, MIS was independently associated with Hb (beta = 0.118, P = 0.004) in patients with estimated glomerular filtration rate (egfr) lower than or equal to 60 ml/min/1.73 m 2, but not in patients with higher egfr. Conclusions. The MIS is independently associated with PTA in the kidney-transplanted population with egfr lower than or equal to 60 ml/min/1.73 m 2. Keywords: inflammation; kidney transplantation; post-transplant anaemia; protein-energy wasting Introduction Post-transplant anaemia (PTA) is common [1 4] and is associated with impaired quality of life, and also worse patient and graft survival [5 7]. PTA is also associated with posttransplant cardiovascular risk [8], left ventricular hypertrophy [9] and congestive heart failure [10]. Conflicting results have been published about the association between anaemia and outcomes, but most studies seem to support the association between PTA and mortality and/or graft failure [5 7,11]. The presence of the protein-energy wasting (PEW) syndrome is associated with negative clinical outcomes [12 15] and atherosclerosis [16] in diverse patient populations. PEW has repeatedly been reported to be associated with erythropoietin hyporesponsiveness [17 19]. The malnutrition-inflammation score (MIS) is a simple semiquantitative instrument for the evaluation of PEW [14]. MIS was associated with anaemia in patients on maintenance dialysis [14]. We have shown that the score reliably measures PEW also in kidney transplant recipient (Tx) patients [20]. Previously, we reported a strong association between haemoglobin and serum transferrin [3]. The direction of the association, however, suggested that that serum transferrin may be a marker of inflammation and not of iron status in this setting [3,21]. Consequently, we hypothesize that the PEW is an important predictor of PTA. In this The Author Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Inflammation and anaemia in transplanted patients 2001 cross-sectional analysis, we tested this hypothesis by analysing the association between PTA and haemoglobin versus markers of inflammation and nutritional parameters, including the MIS in kidney-transplanted patients. Materials and methods Sample of patients and data collection All prevalent kidney transplant recipients 18 years of age or older (n = 1214) who were followed at the outpatient transplant clinic of the Department of Transplantation and Surgery at the Semmelweis University, Budapest on 31 December 2006 were invited to participate. Exclusion criteria were acute rejection within the last 4 weeks, current hospitalization, transplantation within the previous 3 months, acute infection or bleeding. The baseline assessment was conducted between February 2007 and August 2007 [Malnutrition-Inflammation in Transplant Hungary Study (MINIT-HU Study)]. Of the 1214 potentially eligible patients, 205 (17%) refused to participate and 16 (1%) patients were excluded. The MINIT-HU study sample, therefore, included 993 patients. To focus on the association between late PTA and MIS, we further excluded 51 patients who had been transplanted within the 12-month period before enrolment. The final study sample, therefore, included 942 patients. After enrolment, demographic data and medical history [age, gender, menopause status, co-morbidity, aetiology and history of chronic kidney disease (CKD)] were collected; weight, height and abdominal circumference (AC) were measured. Estimated glomerular filtration rate (egfr) was calculated using the abbreviated Modification of Diet in Renal Disease study formula [22]. The study was approved by the ethics committee of the Semmelweis University (49/2006). The study was conforming to ICP Good Clinical Practices Guidelines and the Declaration of Helsinki. Before enrolment, patients received detailed written and verbal information regarding the aims and protocol of the study and gave written consent to participate. MIS The MIS [14] has 10 components, each with four levels of severity, from 0 (normal) to 3 (severely abnormal). The sum of all 10 MIS components ranges from 0 (normal) to 30 (severely malnourished); a higher score reflects a more severe degree of malnutrition and inflammation status. In contrast to the original MIS, we did not include dialysis vintage in the score in this analysis; however, co-morbidity was computed as follows: 0 no medical illnesses except CKD; 1 mild co-morbidity, excluding major co-morbid conditions (MCCs) such as congestive heart failure classes III IV, severe coronary artery disease, clinically evident acquired immunodeficiency syndrome, moderate to severe chronic obstructive pulmonary disease and metastatic malignancies; score 2 moderate co-morbidity (including one of the diseases listed under MCCs); and score 3 two or more MCCs. All subjective global assessment (SGA) was performed by a physician (M.E. Czira) according to conventional SGA guidelines [23,24]. Co-morbidity was assessed using the modified Charlson co-morbidity index [25]. Laboratory data Laboratory data were extracted from the medical records. The following laboratory parameters were tabulated: haemoglobin (Hb), serum C-reactive protein (CRP), total cholesterol, triglyceride, ferritin, transferrin, albumin, serum intact parathormone (ipth), creatinine and blood urea nitrogen; for iron metabolism: serum iron, serum ferritin, serum transferrin and percentage of hypochrome reticulocytes. Additional serum samples collected at the time of the baseline assessment were stored at 70 C for future use. From these samples, high sensitivity interleukin-6 (IL-6), tumour necrosis factor-α (TNF-α) and leptin levels were measured using immunoassay kits based on solid-phase sandwich enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN). Definition of anaemia and iron deficiency In this manuscript, we focus on late post-transplant anaemia. The prevalence and predictors of PTA are different in the first year after transplantation and thereafter [26,27]. It is common to divide early PTA (1 year after transplantation) and late (more than 1 year) PTA. Only patients with late PTA were included in this analysis. Similar to previous publications, anaemia was defined according to the guideline of the American Society of Transplantation (AST): Hb under 130 g/l in adult males and 120 g/l in adult females [28]. In this analysis, we used only the baseline blood Hb but did not consider any previous changes in Hb levels due to treatment [iron supplementation or erythropoietin-stimulating agent (ESA) therapy] or spontaneous variations. In this manuscript, we used this definition unless stated otherwise. Anaemic patients were divided into three subcategories to assess the severity of anaemia: (i) Mild: males: Hb 130 g/l and >120 g/l; females: Hb 120 g/l and >110 g/l (ii) Moderate: males: Hb 120 g/l and >110 g/l; females: Hb 110 g/l and >100 g/l (iii) Severe: males: Hb 110 g/l; females: Hb 100 g/l. Transplantation- and donor-related data Transplantation-related data tabulated included current medications (including current immunosuppressive treatment, ESAs and iron supplementation), transplant vintage, i.e. time elapsed since the time of the transplantation, pre-transplant dialysis duration, donor source (deceased or living donor), history of acute rejection, number of HLA mismatch, panel reactive antibodies titer (PRA), cold ischaemic time (CIT), donor age and gender and history of delayed graft function. Acute rejection was defined as the documented administration of anti-rejection therapy at any time after transplantation. Delayed graft function was defined as the need for one or more haemodialysis sessions after transplantation [29]. Immunosuppressive therapy Standard immunosuppressive therapy generally consisted of prednisolone, either cyclosporine A (Neoral) (CsA) or tacrolimus, combined with mycophenolate mofetil (MMF), azathioprine or sirolimus. Statistical analysis We used the SPSS 15.0 and STATA 11.1 software. Descriptive statistics used include proportions, means (± standard deviation, SD) or median (interquartile range, IQR) as appropriate. To compare anaemic versus non-anaemic patients, we used the Student s t-test or the Mann Whitney U-test for continuous variables and the chi-square test for categorical variables. To compare characteristics between groups formed according to the severity of their anaemia, ANOVA or the Kruskal Wallis test was used. To identify factors which are independently associated with Hb or anaemia, linear or binary logistic regression models, respectively, were built. Independent variables were entered into the final model if they were significantly associated with Hb in the bivariate analyses or had been reported to be associated with Hb or anaemia by others. Independent variables were entered in two blocks and all variables were retained in the final, fully adjusted model. Block 1 included age, gender, transplant vintage, Charlson comorbidity score, egfr, use of ACEI/ARB, use of MMF, use of sirolimus, HLA mismatches, presence of delayed graft function, history of acute rejection and AC. Block 2 included serum CRP, percentage of hypochromic reticulocyte, serum ferritin, serum ipth, serum IL-6 and serum TNF-α. We did not adjust for the presence of diabetes, serum transferrin, serum albumin and BMI because these variables were included in the Charlson comorbidity index score or MIS, respectively. Non-linear relationships of all the continuous independent predictors in the multivariable models with the outcome variables were modelled using cubic regression splines by specifying a maximum acceptable complexity for each continuous function and applying a closed-test approach to each continuous predictor to simplify the model if possible [30,31]. The multivariable-adjusted means of Hb and of the percentage of patients with anaemia by subgroups of interest were estimated from the final multivariable regression models by using STATA post-estimation commands. To determine if MIS is associated with anaemia in patients with both preserved and impaired kidney function, these models were tested both in the total sample and in sub-samples defined by egfr

3 2002 M.Z. Molnar et al. (egfr: >60 ml/min/1.73 m 2 or 60 ml/min/1.73 m 2 ). Interactions with kidney function were also tested with the inclusion of multiplicative interaction terms in the multivariable models. For all analysis, twosided s are reported and results have been considered statistically significant if P < Results Correlates of anaemia Characteristics of patients with versus without anaemia are compared in Table 1. Anaemic patients were more likely females, had higher MIS, lower egfr, longer transplant vintage, lower serum transferrin, higher serum ferritin and higher proportion of hypochromic reticulocytes (Table 1). The MIS was significantly associated with increasing proportion of anaemia (Figure 1). Furthermore, several variables reflecting nutritional status and inflammation demonstrated significant associations with the severity of anaemia (Table 2). Body mass index (BMI), serum albumin and AC showed a significant decreasing trend with increasing severity of anaemia (Table 2). Serum IL-6, on the other hand, showed a significant increasing trend with increasing severity of anaemia (Table 2). Demographics and baseline characteristics of the sample Baseline patient characteristics of the 942 participants are presented in Table 1. The proportion of men was lower among participants versus non-participants (57% vs 67%, respectively; P < 0.01), but the mean age of the two groups was similar (data not shown). Overall, 315 patients (33%) were anaemic as defined by the AST criteria. Multivariable analysis To analyse the independent association between Hb or the presence of anaemia and the factors discussed above, multivariate linear and binary logistic regression analyses were used. A significant (P = 0.02) interaction was detected between MIS and egfr (egfr: >60 ml/min/1.73 m 2 or 60 ml/min/1.73 m 2 ) when predicting anaemia; the interaction was also nearly significant (P = 0.07) when predicting Hb level. Consequently, we tested the multivariate models both in the total study sample and in sub-samples defined by egfr (egfr 60 ml/min/1.73 m 2 or egfr >60 ml/min/1.73 m 2, respectively) (Table 3). In the unadjusted model, the MIS was significantly associated with Hb in all three analyses but the association lost significance in the group with egfr >60 ml/min/1.73 m 2 after adjusting for covariables. The association between MIS and Hb remained significant in both the total sample and in the group with egfr 60 ml/min/1.73 m 2 even in the fully Table 1. Characteristics of the study sample. Anaemia was defined according to the anaemia guideline of the American Society of Transplantation: haemoglobin under 130 g/l in adult males and 120 g/l in adult females Not anaemic (n = 627) Anaemic (n = 315) MIS (median; IQR) 3; 3 4; 5 <0.001 Male (%) <0.001 Age (years) (mean ± SD) 51 ± ± Diabetes mellitus (%) Transplant vintage (months) (median; IQR) 71; 70 86; Charlson co-morbidity index score [median; IQR, mean (95% of CI)] 2; 1; 2.98 ( ) 2; 2; 3.29 ( ) 0.02 Estimated GFR (ml/min/1.73 m 2 ) (mean ± SD) 57 ± ± 19 <0.001 ipth (pg/ml) (median; IQR) 66; 50 72; Serum transferrin (g/l) (mean ± SD) 2.43 ± ± 0.47 <0.001 Percentage of hypochrome reticulocytes (%) (median; IQR) 0.2; ; 1.0 <0.001 Ferritin (ng/ml) (median; IQR) 134; ; 542 <0.001 Transferrin saturation (TSAT) (%) (median; IQR) 29; 16 29; Immunosuppressant (%) Steroids Cyclosporine A Neoral Tacrolimus Mycophenolate mofetil Azathioprine Sirolimus Iron supplementation (%) <0.001 Erythropoietin-stimulating agent use (%) 4 24 <0.001 Use of ACEIs/ARBs (%) Number of HLA mismatches (n; (%)) 0.9 0, 1 36 (6) 17 (5) 2, (69) 210 (69) (20) 66 (22) 5, 6 31 (5) 13 (4) CIT (min) (mean ± SD) 1239 ± ± History of acute rejection (%) <0.001 History of DGF (%) Abbreviation: MIS, malnutrition-inflammation score; egfr, estimated GFR; BMI, body mass index; CRP, C-reactive protein; ipth, intact PTH; PRA, panel reactive antibody; CIT, cold ischaemic time; DGF, delayed graft function. showed difference between anaemic and non-anaemic patients. Note: Conversion factors for units: ipth in pg/ml to ng/l, 1; serum transferrin in mg/dl to g/l, 0.01, serum ferritin in ng/ml to μg/l, 1.

4 Inflammation and anaemia in transplanted patients 2003 % of patients adjusted model. Similar results were seen in the logistic regression analysis with anaemia as the dependent variable, except the association between the MIS and the presence of anaemia was not significant in patients with egfr >60 ml/ min/1.73 m 2 even in the unadjusted model (Table 3). We also repeated these models in males and females separately and we found similar qualitative results in both genders. Figures 2 and 3 demonstrate the association of the MIS and egfr with Hb or anaemia. The estimated multivariable-adjusted Hb levels of patient groups divided according to mutually exclusive levels of MIS and egfr are shown in Figure 2. Figure 3 shows the estimated multivariableadjusted proportion of patients with anaemia in the same subgroups of patients. Both figures demonstrate a linear association between MIS and Hb or the proportion of anaemia in patients with CKD 3 5, but not in patients with egfr >60 ml/min/1.73 m 2. Discussion : No anaemia : Mild anaemia : Moderate anaemia : Severe anaemia First Second Third Fourth Quartiles of MIS Fig. 1. The proportion of anaemic patients in groups defined by quartiles of MIS (linear-by-linear association; P < 0.001). We demonstrated here that the MIS is an independent predictor of late PTA in kidney-transplanted patients with CKD 3 5, but not in patients with CKD 1 2 in a prevalent cohort of stable kidney transplant recipients. Additionally, we found a moderate association between Hb concentration and markers of nutritional status and inflammation in stable kidney transplant recipients with CKD 3 5. The association between MIS and Hb remained significant after controlling for several covariables, including nutritional and inflammational markers. These results suggest that the MIS, a marker of the PEW, has an impact on Hb or anaemia similar in magnitude to that of egfr. Our analyses also demonstrated that these effects are additive in patients with CKD 3 5. To our knowledge, our study is the first to demonstrate the link between PEW and PTA in kidney transplant recipients. Previous studies considered residual renal function as the most important predictor of PTA. Importantly, in a recent study [32], Chadban et al. demonstrated substantial differences between serum Hb among Tx versus CKD patients with similar egfr suggesting that post-transplantation anaemia cannot be attributed solely to impaired kidney function. In an editorial, Weir suggested that PEW likely contributes to PTA [21]. To assess PEW and its association with PTA, we used the MIS [14,17,20] and additional markers, such as serum IL-6, TNF-α, CRP, leptin and albumin levels, BMI and AC. Both Hb concentration and the presence of anaemia were significantly correlated with MIS in patients with CKD 3 5. This consistent association between MIS and PTA remained significant in this subcohort after adjusting for many known risk factors and predictors of PTA. Furthermore, we found weak negative correlations between inflammatory cytokines (IL-6 and TNF-α) and Hb. A similar association has been reported between erythropoietin sensitivity and IL-6 in end-stage renal disease (ESRD) patients [33]. Markers of nutrition (leptin and AC) were also associated with PTA and Hb in our analysis. The association between MIS and Hb was significant only in patients with egfr corresponding to CKD 3 5, suggesting that the PEW assessed by the score is part of the CKD syndrome, i.e. it is related to chronic kidney failure. Accordingly, the MIS may be an important marker of the severity of kidney failure and the associated co-morbidity in this patient population, therefore the MIS could be a Table 2. Inflammation and nutritional markers and severity of anaemia. Anaemia was defined according to the anaemia guideline of the American Society of Transplantation as haemoglobin under 130 g/l in adult males and 120 g/l in adult females. Mild anaemia: males: Hb >120 g/l and 130 g/l; females: Hb >110 g/l and Hb 120 g/l; moderate anaemia (as anaemia defined by the NKF-K/DOQI): males: Hb >110 g/l and 120 g/l; females: Hb > 100 g/l and Hb 110 g/l; severe anaemia: males: Hb 110 g/l; females: Hb 100 g/l Total population (n = 942) Not anaemic n = 627 Mild anaemia n = 158 Moderate anaemia n = 106 Severe anaemia n = 51 MIS (median; IQR) 3; 3 3; 3 4; 3 4; 4 6; 4.5 <0.001 Estimated GFR (ml/min/1.73 m 2 ) (mean ± SD) 51 ± ± ± ± ± 16 <0.001 BMI (kg/m 2 ) (mean ± SD) 27 ± 5 27 ± 5 26 ± 5 26 ± 5 25 ± 5 <0.001 Abdominal circumference (cm) (mean ± SD) 99 ± ± ± ± ± 15 <0.001 Serum albumin (g/l) (mean ± SD) 40 ± 4 41 ± 4 39 ± 4 38 ± 4 37 ± 4 <0.001 Serum leptin (ng/ml) (median; IQR) 15; 25 14; 21 19; 37 21; 43 12; Serum IL-6 (pg/ml) (median; IQR) 2.1; ; ; ; ; Serum TNF-α (pg/ml) (median; IQR) 2.1; ; ; ; ; Serum CRP (mg/l) (median; IQR) 3.2; ; ; ; ; Abbreviation: MIS, malnutrition-inflammation score; BMI, body mass index; CRP, C-reactive protein. Note: Conversion factors for units: serum albumin in g/dl to g/l, 10.

5 2004 M.Z. Molnar et al. Table 3. Multivariate models to determine the independent association between MIS versus haemoglobin or the presence of anaemia Linear regression (dependent variable = Hb) Total sample egfr>60ml/min/1.73m 2 egfr 60mL/min/1.73 m 2 Beta Beta Beta Unadjusted < <0.001 Adjusted Model < <0.001 Adjusted Model < Binary logistic regression (dependent variable = presence of anaemia) Total sample egfr>60ml/min/1.73 m 2 egfr 60mL/min/1.73 m 2 Odds ratio (95% of CI) Odds ratio (95% of CI) Odds ratio (95% of CI) Unadjusted ( ) < ( ) ( ) <0.001 Adjusted Model ( ) < ( ) ( ) <0.001 Adjusted Model ( ) ( ) ( ) Model 1: adjusting for age, gender, transplant vintage, Charlson co-morbidity score, egfr, use of ACEI/ARB, use of MMF, use of sirolimus, HLA mismatches, presence of delayed graft function, history of acute rejection and abdominal circumference. Model 2: adjusting for Model 1 and serum CRP, percentage of hypochrom reticulocyte, serum ferritin, serum ipth, serum IL-6 and serum TNF-α. useful clinical tool to monitor patients with CKD, including kidney transplant recipients. Several potential mechanisms may contribute to the observed associations. Erythropoietin resistance and PEW are associated in dialysed patients [17,18,34,35] and also in patients who started dialysis due to end-stage chronic allograft nephropathy [19]. It is possible that micro-inflammation due to the presence of the allograft might contribute to reduced erythropoietin effectiveness in kidney-transplanted patients. Further studies are needed to prove this hypothesis. Micro-inflammation and PEW might also affect erythropoietin production and sequestration of iron [36], which are obligatory steps of erythropoiesis and their impairment may lead to anaemia. PEW may also induce or increase oxidative stress [37], which could also contribute to anaemia [38]. Additionally, severe malnutrition will result in deficiency of mineral elements and vitamins, which are essential to red blood cells production. Theoretically, PEW may also disturb the endocrine homeostasis further reducing erythropoiesis [39]. Although we used gender-specific definition of anaemia of the AST, we found significantly higher prevalence of anaemia in women versus men. A similar result was found in some [2,40], but not all [3,4], previous studies. Several potential mechanisms may contribute this gender difference. The prevalence of unrecognized iron deficiency may be higher in women compared to men [2,3], which may contribute to the higher prevalence of anaemia seen in women. Another potential explanation of this difference may lay in the fact that women in our cohort have significantly lower egfr than men (47 ± 21 vs 53 ± 20; P < 0.001). Fig. 2. Predicted Hb levels from the final, fully adjusted multivariate linear regression model (Model 2) in subgroups defined by MIS and egfr. Fig. 3. Predicted proportion of anaemic patients from the final, fully adjusted multivariate binary logistic regression model (Model 2) in subgroups defined by MIS and egfr.

6 Inflammation and anaemia in transplanted patients 2005 The prevalence and severity of PTA remained practically unchanged even after years from the publication of the results of several large studies [2 4,40]. One third of the patients in this study were anaemic according to the anaemia guidelines published by the AST [28]. Our analysis was not designed to determine the pathogenesis and origin of anaemia in these patients. Since we focused on late post-transplant anaemia (i.e. anaemia in patients who survived at least 12 months after transplantation), this could have originated in the early post-transplant period and even in the dialysis period preceding transplantation. It also could, however, be linked to factors associated with the post-transplant period such as PEW. Although most observational studies repeatedly found that anaemia predicted adverse clinical outcome in patients with various degrees of CKD, including Tx patients [5 7,11], recent reports of prospective randomized studies testing the effect of ESAs in patients with CKD (the CHOIR, CREATE and TREAT studies) [41 43] have suggested that in patients with renal anaemia using ESAs to achieve normal or close normal Hb levels does not confer survival benefit or even substantial improvement in health-related quality of life [44]. Furthermore, the potential risk of such treatment strategies has been raised. In a recent retrospective analysis of more than 1700 kidney transplant recipients, Heinze et al. showed significantly increased mortality in patients who received ESA and had a Hb above 125 g/l [7]. These results may raise concern about the effectiveness and safety of ESA therapy of renal anaemia. Importantly, however, the lowest mortality risk in patients receiving ESA in the study of Heinze et al. was at 125 g/l. Furthermore, patients who did not receive ESA but had Hb above 125 g/l had even lower mortality risk. Based on these results, it is difficult to conclude what the optimal target for Hb should be. The current guidelines, which suggest to initiate ESA treatment if Hb is consistently <110 g/l and maintain Hb between 110 and 120 g/l, seem to be safe. Appropriate large scale prospective studies to define the role and targets for ESA therapy in Tx patients are needed. Irrespective of the specific questions discussed above, we suggest that PTA still needs special attention. PTA has been repeatedly shown to be associated with increased risk of mortality and it may be an important marker of underlying processes and conditions (likely involving PEW and others), eventually leading to poor clinical outcome. The PEW has repeatedly been associated with erythropoietin hyporesponsiveness in ESRD patients [17 19]. Our results suggest that the PEW is also associated with late PTA. Hypothetically, the PEW might be a confounder in the complex associations between anaemia [5,6], the use of ESA [7] and negative outcome. This is supported by previous observations where the MIS predicted mortality in dialysed [14] and also in Tx patients [45]. Further studies to test this hypothesis in various CKD populations are needed. To our knowledge, this was the first study to assess the relationship between the PEW and PTA in kidneytransplanted patients. Our study is notable for enrolling a large sample. We used a reliable tool to assess PEW in kidney-transplanted patients [20]. An additional strength of the study was that we measured several cytokines associated with PEW and we presented both inflammation and nutrition markers also associated with anaemia. Finally, many important variables, which might have an effect on PTA, were included in our multivariate models, which increases the biological reliability of our models. Limitations should also be considered when interpreting our results. Patients were enrolled from a single centre; therefore, our results are not to be generalized without further considerations. Patients who were not participating in the study were different from participants, but it is unlikely that this would have any major impact on our results. Unobserved gastrointestinal or gynaecological bleeding, severe haematological disorders, haemolysis, vitamin B12 or folate deficiency, malignancy and haemoglobinopathies could potentially have contributed to the observed prevalence of anaemia in this study. Due to the ethnic background of the study population (all Caucasians), haemoglobinopathies are very rare. As the population consisted of stable outpatients, unobserved gastrointestinal or gynaecological bleeding and severe haematological disorders are also improbable causes of anaemia, although chronic occult bleeding might be a potential confounder. For the same reason, we suggest that ongoing haemolysis or bone marrow suppression is also likely to be quite rare in this specific population. We cannot rule out the possibility that including all these parameters, or other, yet unknown factors, would have altered our results. The anaemia definition used in this analysis is based only on the actual blood Hb value independent of any preceding spontaneous changes or treatment effects (due to iron supplementation or ESA therapy). This cohort was a prevalent cohort which is subject to the problem of incidenceprevalence bias. In summary, we have shown that the MIS, which assesses the PEW syndrome, is an independent predictor of late PTA in kidney-transplanted patients with CKD stages 3 5. Further studies are needed to assess whether treating PEW effectively would have an impact on the prevalence or the severity of anaemia in kidney-transplanted patients. Acknowledgements. The authors thank the patients and the staff in the Department of Transplantation and Surgery, Semmelweis University Budapest. This study was supported by grants from the National Research Fund (OTKA) (F-68841; HUMAN-MB08-A-81231), ETT (206/09), the Hungarian Kidney Foundation, Hungarian Society of Hypertension, Hungarian Society of Nephrology and the Foundation for Prevention in Medicine. M.Z.M. received grants from the National Research Fund (NKTH-OTKA-EU 7KP-HUMAN-MB08-A-81231), was also supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences ( ), and is recipient of the Hungarian Eötvös Scholarship (MÖB/66-2/2010). The research of M.N. has been supported by an unrestricted research grant from Canadian Home Healthcare Inc. and the Center for Integrative Mood Research, Toronto, Canada and she also received grants from János Bolyai Research Scholarship of the Hungarian Academy of Sciences. Conflict of interest statement. D.J.G. disclosures are Advisory Boards and Speaking for Roche, Amgen, J+J and Sandoz. References 1. Vanrenterghem Y. Anemia after kidney transplantation. Transplantation 2009; 87: Lorenz M, Kletzmayr J, Perschl A et al. Anemia and iron deficiencies among long-term renal transplant recipients. J Am Soc Nephrol 2002; 13:

7 2006 M.Z. Molnar et al. 3. Molnar MZ, Novak M, Ambrus C et al. Anemia in kidney transplanted patients. Clin Transplant 2005; 19: Vanrenterghem Y, Ponticelli C, Morales JM et al. Prevalence and management of anemia in renal transplant recipients: a European survey. Am J Transplant 2003; 3: Molnar MZ, Czira M, Ambrus C et al. Anemia is associated with mortality in kidney-transplanted patients a prospective cohort study. Am J Transplant 2007; 7: Chhabra D, Grafals M, Skaro AI et al. Impact of anemia after renal transplantation on patient and graft survival and on rate of acute rejection. Clin J Am Soc Nephrol 2008; 3: Heinze G, Kainz A, Horl WH et al. Mortality in renal transplant recipients given erythropoietins to increase haemoglobin concentration: cohort study. BMJ 2009; 339: b Djamali A, Becker YT, Simmons WD et al. Increasing hematocrit reduces early posttransplant cardiovascular risk in diabetic transplant recipients. Transplantation 2003; 76: Rigatto C, Foley R, Jeffery J et al. Electrocardiographic left ventricular hypertrophy in renal transplant recipients: prognostic value and impact of blood pressure and anemia. J Am Soc Nephrol 2003; 14: Rigatto C, Parfrey P, Foley R et al. Congestive heart failure in renal transplant recipients: risk factors, outcomes, and relationship with ischemic heart disease. J Am Soc Nephrol 2002; 13: Winkelmayer WC, Lorenz M, Kramar R et al. Percentage of hypochromic red blood cells is an independent risk factor for mortality in kidney transplant recipients. Am J Transplant 2004; 4: Kalantar-Zadeh K, Ikizler TA, Block G et al. Malnutritioninflammation complex syndrome in dialysis patients: causes and consequences. Am J Kidney Dis 2003; 42: Kalantar-Zadeh K, Kopple JD. Relative contributions of nutrition and inflammation to clinical outcome in dialysis patients. Am J Kidney Dis 2001; 38: Kalantar-Zadeh K, Kopple JD, Block G et al. A malnutritioninflammation score is correlated with morbidity and mortality in maintenance hemodialysis patients. AmJKidneyDis2001; 38: Lopes AA, Elder SJ, Ginsberg N et al. Lack of appetite in haemodialysis patients associations with patient characteristics, indicators of nutritional status and outcomes in the international DOPPS. Nephrol Dial Transplant 2007; 22: Stenvinkel P, Heimburger O, Lindholm B et al. Are there two types of malnutrition in chronic renal failure? Evidence for relationships between malnutrition, inflammation and atherosclerosis (MIA syndrome). Nephrol Dial Transplant 2000; 15: Kalantar-Zadeh K, McAllister CJ, Lehn RS et al. Effect of malnutrition-inflammation complex syndrome on EPO hyporesponsiveness in maintenance hemodialysis patients. AmJKidneyDis 2003; 42: Locatelli F, Andrulli S, Memoli B et al. Nutritional-inflammation status and resistance to erythropoietin therapy in haemodialysis patients. Nephrol Dial Transplant 2006; 21: Lopez-Gomez JM, Perez-Flores I, Jofre R et al. Presence of a failed kidney transplant in patients who are on hemodialysis is associated with chronic inflammatory state and erythropoietin resistance. JAm Soc Nephrol 2004; 15: Molnar MZ, Keszei A, Czira ME et al. Evaluation of the malnutrition-inflammation score in kidney transplant recipients. Am J Kidney Dis 2010; 56: Weir MR. Is anemia a predictor for mortality and loss of graft function in kidney transplant recipients? Am J Transplant 2007; 7: Levey A, Greene T, Kusek J et al. A simplified equation to predict glomerular filtration rate from serum creatinine {Abstract}. J Am Soc Nephrol 2000; 11: 155A 23. Kalantar-Zadeh K, Kleiner M, Dunne E et al. Total iron-binding capacity-estimated transferrin correlates with the nutritional subjective global assessment in hemodialysis patients. Am J Kidney Dis 1998; 31: Enia G, Sicuso C, Alati G et al. Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 1993; 8: Jassal SV, Schaubel DE, Fenton SS. Baseline comorbidity in kidney transplant recipients: a comparison of comorbidity indices. Am J Kidney Dis 2005; 46: Yorgin PD, Belson A, Sanchez J et al. Unexpectedly high prevalence of posttransplant anemia in pediatric and young adult renal transplant recipients. Am J Kidney Dis 2002; 40: Yorgin PD, Scandling JD, Belson A et al. Late post-transplant anemia in adult renal transplant recipients. An under-recognized problem? Am J Transplant 2002; 2: Kasiske BL, Vazquez MA, Harmon WE et al. Recommendations for the outpatient surveillance of renal transplant recipients. American Society of Transplantation. J Am Soc Nephrol 2000; 11: S1 S Yarlagadda SG, Coca SG, Garg AX et al. Marked variation in the definition and diagnosis of delayed graft function: a systematic review. Nephrol Dial Transplant 2008; 23: Royston P, Ambler G, Sauerbrei W. The use of fractional polynomials to model continuous risk variables in epidemiology. Int J Epidemiol 1999; 28: Royston P, Sauerbrei W. Building multivariable regression models with continuous covariates in clinical epidemiology with an emphasis on fractional polynomials. Methods Inf Med 2005; 44: Chadban SJ, Baines L, Polkinghorne K et al. Anemia after kidney transplantation is not completely explained by reduced kidney function. Am J Kidney Dis 2007; 49: Axelsson J, Qureshi AR, Heimburger O et al. Body fat mass and serum leptin levels influence epoetin sensitivity in patients with ESRD. Am J Kidney Dis 2005; 46: Agarwal R, Davis JL, Smith L. Serum albumin is strongly associated with erythropoietin sensitivity in hemodialysis patients. Clin J Am Soc Nephrol 2008; 3: Akgul A, Bilgic A, Sezer S et al. Effect of protein-energy malnutrition on erythropoietin requirement in maintenance hemodialysis patients. Hemodial Int 2007; 11: Bistrian BR, Carey LA. Impact of inflammation on nutrition, iron status, and erythropoietin responsiveness in ESRD patients. Nephrol Nurs J 2000; 27: Brunini TM, Mendes-Ribeiro AC, Ellory JC et al. Platelet nitric oxide synthesis in uremia and malnutrition: a role for L-arginine supplementation in vascular protection? Cardiovasc Res 2007; 73: Stenvinkel P, Barany P. Anaemia, rhuepo resistance, and cardiovascular disease in end-stage renal failure; links to inflammation and oxidative stress. Nephrol Dial Transplant 2002; 17: Carrero JJ, Qureshi AR, Parini P et al. Low serum testosterone increases mortality risk among male dialysis patients. J Am Soc Nephrol 2009; 20: Winkelmayer WC, Kewalramani R, Rutstein M et al. Pharmacoepidemiology of anemia in kidney transplant recipients. J Am Soc Nephrol 2004; 15: Singh AK, Szczech L, Tang KL et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 2006; 355: Drueke TB, Locatelli F, Clyne N et al. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med 2006; 355: Pfeffer MA, Burdmann EA, Chen CY et al. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med 2009; 361: Rebollo P, Baltar JM, Campistol JM et al. Quality of life of patients with chronic renal allograft rejection and anemia. J Nephrol 2004; 17: Remport A, Czira ME, Rudas A et al. Association between the malnutrition and inflammation complex syndrome score and mortality in kidney transplanted patients. Transpl Int 2009; 22: 163 Received for publication: ; Accepted in revised form: