NDT Advance Access originally published online on December 29, 2005
Nephrology Dialysis Transplantation 2006 21(4):991-998; doi:10.1093/ndt/gfk011
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author [2005]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Original Articles: Dialysis and Transplantation
Nutritional-inflammation status and resistance to erythropoietin therapy in haemodialysis patients
1 Department of Nephrology and Dialysis, Ospedale A. Manzoni, Lecco, 2 Department of Nephrology, Università di Napoli Federico II, Naples, 3 Department of Nephrology and Dialysis, ‘Spedali Civili’, Brescia, 4 Department of Nephrology and Dialysis, Ospedale ‘Misericordia e Dolce’, Prato, 5 Department of Nephrology, Azienda Ospedaliera ‘G.Moscati’, Avellino and 6 U.O. Nefrologia, CNR - I.B.I.M., Ospedali Riuniti Reggio Calabria, Italy
Correspondence and offprint requests to: Prof. Francesco Locatelli Department of Nephrology and Dialysis, Ospedale A. Manzoni, Via dell’Eremo 9, 23900 Lecco, Italy. Email: nefrologia{at}ospedale.lecco.it
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionReferences |
|---|
Background. Chronic kidney disease patients who are resistant to erythropoietin (EPO) treatment may suffer from malnutrition and/or inflammation.
Methods. In a cross-sectional study of haemodialysis patients, we investigated the relationship between the natural logarithm of the weekly EPO dose normalized for post-dialysis body weight and outcome measures of nutrition and/or inflammation [BMI, albumin and C reactive protein (CRP)] by means of multiple linear regression analysis. On the basis of the decile distribution of weekly EPO doses, we also evaluated four groups of patients: untreated, hyper-responders, normo-responders and hypo-responders.
Results. Six hundred and seventy-seven adult haemodialysis patients were recruited from five Italian centres. BMI and albumin were lower in the hypo-responders than in the other groups (21.3±3.8 vs 24.4±4.7 kg/m2, P<0.001; and 3.8±0.6 vs 4.1±0.4 g/dl, P<0.001), whereas the median CRP level was higher (1.9 vs 0.8 mg/dl, P = 0.004). The median weekly EPO dose ranged from 30 IU/kg/week in the hyper-responsive group to 263 IU/kg/week in the hypo-responsive group. Transferrin saturation linearly decreased from the hyper- to hypo-responsive group (37±15 to 25±10%, P = 0.003), without any differences in transferrin levels. Ferritin levels were lower in the hypo-responsive than in the other patients (median 318 vs 445 ng/ml, P = 0.01). At multiple linear regression analysis, haemoglobin, BMI, albumin, CRP and serum iron levels were independently associated with the natural logarithm of the weekly EPO dose (R2 = 0.22).
Conclusions. Our findings support a clear association between EPO responsiveness and nutritional and inflammation variables in haemodialysis patients; iron deficiency is still a major cause of hypo-responsiveness.
Keywords: albumin; C reactive protein; EPO resistance; haemodialysis; inflammation; malnutrition
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Over the last 15 years, the availability of recombinant human erythropoietin (EPO) has led to the almost complete disappearance of severe anaemia requiring blood transfusions in patients with chronic kidney disease (CKD); however, despite an increase in its use and average dose, a substantial percentage of patients still fail to achieve the haemoglobin (Hb) targets recommended by the international guidelines [1,2]. In this setting, the term EPO resistance has been introduced to define the patients who fail to attain the target despite a higher than usual dose of erythropoiesis stimulating agent (ESA) or who continuously need this higher dose in order to maintain it. Nearly 10% of the considered populations would fulfil this definition, but the true incidence of resistance to ESA is not known.
Although iron deficiency is probably the most important factor affecting the response to ESA in most patients, occult blood loss, infection, hyperparathyroidism and inadequate dialysis are also important.
Chronic inflammation can also be a cause of anaemia in CKD patients as a result of the suppression of bone marrow erythropoiesis by some cytokines [3–5]. A number of studies have highlighted the association between increased inflammatory indices and a reduced response to ESA; in particular, high C reactive protein (CRP) levels are often found in haemodialysis patients requiring higher ESA doses [6–8], and increased CRP levels in CKD patients positively correlate with other inflammatory cytokines, such as interleukin 6 (IL-6) [9], whose levels are 8–10 times higher in haemodialysis patients and have been directly related to the EPO dose [10].
A large proportion of CKD patients also have protein energy malnutrition and wasting [11], low serum levels of albumin [7] and other more specific nutritional markers which are predictors of the response to EPO [8]. It is therefore possible that a diminished nutritional status could be a feature of patients who are resistant to ESA treatment, with malnutrition probably being due to a chronic inflammatory state [10,12].
The consequent anaemia could be one of the links between malnutrition–inflammation–arteriosclerosis (MIA) syndrome, and the higher rates of hospitalization and mortality particularly due to cardiovascular disease [13]. However, the possible interactions between inflammatory and nutritional markers and their impact on refractory anaemia and EPO resistance, as potentially modifiable risk factors of dialysis-associated morbidity and mortality, are still unclear.
Starting from the hypothesis that anaemia (partially due to a reduced response to ESA) could underlie the inflammation, malnutrition and poor outcome of CKD patients, we planned a multicentre, cross-sectional study aimed at evaluating the impact and possible causes of resistance to ESA in a large sample of haemodialysis patients.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Six hundred and seventy-seven patients aged
18 years who had been undergoing haemodialysis for at least 6 months were recruited by five dialysis Centres in Italy (Brescia, Lecco, Naples, Prato and Reggio Calabria). Patients with haemoglobinopathies, sickle cell anaemia, thalassaemia or malignancies were excluded, as were those who had experienced infection, vascular access thrombosis, stroke, myocardial infarction, heavy blood loss, or who had undergone major surgery or blood transfusion in the previous three months. The laboratory parameters (except for post-dialysis serum urea nitrogen, which was used to calculate the equilibrated Kt/V) were measured before the start of the dialysis session. Equilibrated Kt/V, calculated according to the second-generation equation of Daugirdas [14], was used to estimate the dialysis dose for low molecular weight uraemic toxins; the appearance of urea nitrogen was calculated in order to estimate daily protein intake. All of the routine laboratory measurements were made using automated methods. Serum CRP was measured in all centres by means of a turbidimetric immunoassay (normal range <0.3 mg/dl).
Statistical analysis
The descriptive analysis was based on arithmetical means and SDs for normally distributed continuous variables, and percentiles for non-normally distributed continuous variables. In order to normalize the amount of EPO required depending on the severity of anaemia, we calculated an EPO resistance index (ERI), defined as the weekly EPO dose divided by Hb level (g/dl). Both the EPO dose and ERI were divided by end-dialysis body weight to indicate the required EPO dose per kilogram of body weight.
Four patient groups were defined on the basis of the distribution of the weekly dose of EPO or darbepoetin alpha: group A (no EPO treatment), group B (hyper-responders; 1st decile of weekly EPO dose), group C (normo-responders; 5th and 6th deciles of weekly EPO dose) and group D (hypo-responders treated with the highest decile of weekly EPO dose). A ratio of 1:200 was used to convert darbepoetin alpha to the EPO equivalent dose (1 µg of darbepoetin alpha = 200 IU of epoetin alpha or beta).
The distribution of the three main outcome measures of nutrition and/or inflammation (body mass index, serum albumin and CRP) in the four groups was investigated by means of one-way analysis of variance (ANOVA), with group B and D being compared with group A (patients not treated with EPO) and C (patients treated with the usual standard EPO dose) in order to characterize the features of hyper- and hypo-responders to EPO. The chi-square test was used to investigate the distribution of categorical variables in the groups.
In the population as a whole, univariate and multivariate linear regression analyses were performed to obtain the predictive power of the nutritional and inflammatory outcome measures on the natural logarithm of the weekly EPO dose; multiple binary logistic regression analysis was used to calculate the odds ratios and their 95% confidence intervals (CI) for the highest EPO decile against the other patients. Non-normally distributed continuous variables such as weekly EPO dose, CRP and ferritin levels were logarithmically converted. All analyses were made using SPSS statistical software for Windows, release 11.01.
Informed consent and ethical surveillance
Before starting the study, the patients were informed about the aims, the expected benefits to them and/or others, the risks and inconveniences involved, and their right to refuse to participate or to withdraw from the study at any time without sanction. Their written consent was obtained. The study was carried out in accordance with the Declaration of Helsinki and its subsequent modifications, and was approved by the Ethics Committees of the participating Centres.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
One-hundred and twenty-seven of the 677 patients (18.8%) were not treated with EPO whereas 550 patients (81.2%) were receiving EPO (73.0%) or darbepoetin (8.2%). Only 16 of the 677 patients (2.4%) were hypo-responsive according to the definition currently suggested by international guidelines (an EPO dose of
300 IU/kg/week) [1,2]. On the basis of the percentile distribution of the weekly EPO doses, 347 of the 677 patients (212 men, 61%), aged 64.4±14.0 years (mean±SD) receiving thrice-weekly haemodialysis lasting 3.7±0.3 hours, with an equilibrated Kt/V of 1.27±0.22, were divided into four groups. Group A consisted of 127 patients (18.8%) not receiving EPO; group B of 55 hyper-responsive patients (8.1%) treated with a low EPO dose (1st decile: median 30.4 IU/kg/week, interquartile range 26.1–35.2); group C of 110 normo-responsive patients (16.2%) treated with a usual EPO dose (5th and 6th decile: median 111.7 IU/kg/week, interquartile range 98.1–123.1); and group D of 55 hypo-responsive patients (8.1%) treated with a high EPO dose (last decile: median 262.9 IU/kg/week, interquartile range 240.0–319.2).
Group characteristics
The clinical and laboratory characteristics of the groups are summarized in Table 1. There were no significant between-group differences in terms of age, diabetes as the primary cause of ESRD (10.4% overall), equilibrated Kt/V, urea nitrogen appearance or the type of vascular access. Group A had a higher proportion of males (72%), and groups A and B a higher percentage of patients with polycystic kidney disease (21 and 20%) than groups C and D (6 and 7%).
|
Serum iron levels and the percent saturation of transferrin linearly decreased from the hyper-responders of group B to the hypo-responders of group D (from 77±30 to 53±20 µg/dl, P<0.001; and from 37±15% to 25±10%, P = 0.003), without any differences in transferrin levels. Median ferritin levels were lower in group D than in the other groups (318 vs 445 ng/ml, P = 0.01), but there was no difference in intravenous iron levels in the four groups (median 250 mg/month). Among the hypo-responders of group D, CRP levels were higher in the patients whose ferritin levels were equal to or less than the median group value of 318 ng/ml (median CRP 3.2 mg/dl, IQR 0.8–12.0 mg/dl) than in those with ferritin levels of >318 ng/ml (median CRP 1.1 mg/dl, IQR 0.4–3.5 mg/dl).
Group D patients had the lowest post-dialysis body weight, BMI and albumin levels, but the highest median CRP level.
Hb levels linearly decreased from group A to group D throughout the intermediate deciles, whereas the weekly EPO dose and ERI increased exponentially (Figure 1); on the contrary, the decrease in BMI and albumin levels were only evident in group D (Figure 2). The distribution of CRP values was skewed to the right in all groups (Figure 3, panel A), and became more symmetrical and normal after natural logarithmic transformation (Figure 3, panel B). The median weekly EPO dose ranged from 30 IU/kg/week in group B to 263 IU/kg/week in group D (P<0.001), with a concomitant 10-fold increase in the ERI (from 2.6 to 26.9 IU/kg/week, P<0.001).
|
|
|
Multiple linear regression analysis
A number of nutritional and inflammatory variables significantly correlated with the natural logarithm of the weekly EPO dose. Higher levels of BMI, serum albumin, Hb, serum iron, transferrin saturation and ferritin were significantly associated with a lower natural logarithm of the weekly EPO dose, whereas higher CRP levels and intravenous iron dose per month were significantly associated with a higher natural logarithm of the weekly EPO dose (Table 2). However, these linear associations were weak; in fact, the absolute Pearson correlation coefficient was at maximum 0.26 (for serum iron, P<0.001) after exclusion of the correlation coefficient with Hb (–0.38, P<0.001). In the multiple linear regression analysis with the natural logarithm of the weekly EPO dose as dependent variable, the model included Hb, CRP, albumin, BMI and serum iron levels (Table 3). Among them, only CRP levels were directly related with the natural logarithm of the weekly EPO dose (positive B coefficient 0.051, P = 0.004). The other four variables were inversely associated (see negative values of their B coefficient). This means that higher levels of Hb, albumin, BMI and serum iron were associated with lower values of the natural logarithm of the weekly EPO dose. The predictive power of this model was not very high (the adjusted R2 was 0.216), but the analysis of residuals plotted against the predicted values was satisfactory given that no particular patterns emerged and that their distribution was normal (data not shown). It is important to underline that transferrin saturation, ferritin and the monthly intravenous iron dose were related to the natural logarithm of the weekly EPO dose only at univariate analysis (Table 2) whereas transferrin, gender, polycystic kidney disease and diabetes were not significantly related to the natural logarithm of the weekly EPO dose both at univariate (Table 2) and multivariate analyses (Table 3). The results of the multiple linear regression analysis did not change when the patients receiving darbopoietin alfa and those with depleted iron stores (transferrin saturation
20% or ferritin levels 100 ng/ml) were excluded.
|
|
Logistic regression analysis
Multiple logistic regression analysis using the group D hypo-responsive patients as cases and the patients in the other groups as controls also showed that Hb, BMI and serum albumin and CRP levels independently predicted the response to EPO (Table 4).
|
The results were consistent with those of the multiple linear regression analysis (Table 3); the natural logarithm of the weekly EPO dose increased with decreasing albumin and BMI values (the negative beta coefficient indicates an inverse relationship), and increasing CRP levels (the positive beta coefficient indicates a direct relationship). Only the serum iron level, which was significantly related at multiple linear regression analysis, was not selected by this model (P = 0.855). The results of the multiple logistic regression analysis did not change when the patients receiving darbopoietin alfa and those with depleted iron stores were excluded.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
One of the main findings of this study is that only 2.4% of a large sample of haemodialysis patients attending the five centres were hypo-responders to EPO therapy, as determined by the usual dose cut-off values of the EBPG and K-DOQI guidelines (
300 IU/kg/week) [1,2]. We also found that responsiveness to EPO is a continuous exponential function, rather than a categorical variable. Hypo-responsive patients are more likely to have lower ferritin and percentage transferrin saturation levels. Comparison of the mean and SD values of percentage transferrin saturation in group D (25±10%) with its normal distribution allowed us to estimate that 16% of the patients had a saturation of less than 15%, which is considerably below that recommended in the EBPG and K-DOQI guidelines, and underlines the fact that this aspect is less than optimally implemented in everyday clinical practice [15].
Our results also confirm the association between EPO resistance and increased CRP levels previously found in a number of mainly small-scale studies. Bárány et al. [6] studied 30 haemodialysis patients and showed that the weekly EPO dose was 80% higher in those whose CRP levels were 20 mg/l or more. Gunnell et al. [7] performed a cross-sectional study of 92 haemodialysis patients and found that CRP, serum albumin and age were the best predictors of EPO responsiveness, and also suggested that EPO resistance occurred in the context of high ferritin and low transferrin levels, a pattern that is expected in the acute-phase response and not in iron deficiency. However, in our study, CRP levels were lower in hypo-responsive patients whose ferritin levels were higher. In a cross-sectional analysis of 339 haemodialysis patients, Kalantar-Zadeh et al. [8] similarly found a positive correlation between CRP levels and EPO dose at linear regression analysis (r = 0.2, P<0.001), but did not find any relationship between CRP and hypo-responsiveness at logistic regression analysis using the highest EPO dose tertile as cases and the lowest EPO dose tertile as controls (odds ratio of hypo-responsiveness = 1.25, P = 0.21). The weakness of this relationship, in contrast with our findings, may have been due to the fact that their tertile classification is less precise than our decile classification, which better isolates the more extreme cases.
The relationship of the response to EPO with the various aspects of iron status and iron therapy is perhaps the most complex one to elucidate and thus major discrepancies emerge on this topic among available studies. This occurs at least for three reasons: (a) there is a high collinearity between the certain covariates (i.e. transferrin saturation and ferritin), (b) some iron variables have a mixed nutritional and inflammatory significance (i.e. transferrin) or are markers of both iron stores and inflammation (i.e. ferritin), and (c) different iron protocols were followed in different centres and in different countries, making comparisons among studies difficult. This can explain why some studies indicate that haemodialysis patients with refractory anaemia have increased serum ferritin levels [7], whereas in our and other studies [8] this finding was not confirmed and serum ferritin levels were not related to EPO response. Furthermore, in our study transferrin levels were not related to the natural logarithm of weekly EPO dose. By contrast, serum iron levels decreased linearly with increasing values of the weekly EPO dose, both at the univariate ANOVA and at the multivariate linear regression analysis. However, when considering as cases the patients with the highest decile of weekly EPO dose at the multiple logistic regression analysis, serum iron levels lose their relationship with hypo-responsiveness to EPO. The reason of this discrepancy remains unclear.
We also found that EPO resistance was associated with low albumin concentrations and a low BMI. This was expected as the cross-sectional study by Gunnell et al. [7] identified low serum albumin levels as the strongest predictor of the EPO/haematocrit ratio in both haemodialysis and peritoneal dialysis patients; however, Kalantar-Zadeh et al. [8] failed to confirm any significant association between low serum albumin levels and the EPO dose or ERI. This inconsistency may be partially explained by the fact that their patients were probably better dialyzed and better nourished than those of Gunnell et al. [7], who had lower mean Kt/V and urea nitrogen appearance values. However, it is interesting that Kalantar-Zadeh et al. [8] found significant associations between ERI and other more specific nutritional markers, such as prealbumin, serum TIBC, total cholesterol, BMI and the percentage of blood lymphocytes, because this is in line with our finding of a relationship between EPO responsiveness and BMI, possibly supporting the MIA syndrome hypothesis. It is also worth noting that an inverse correlation between the EPO dose and appetite had been reported [16].
Another important point is that the linear associations of the relevant covariates with EPO dose were weak, and the predictive power of our final model was not very high (the adjusted R2 was 0.216), possibly for at least two reasons: (a) important factors associated with hypo-responsiveness are missing or still have to be identified in our model, and (b) given the complexity of the subject, the linear models used may have been inadequate as is suggested by the fact that the BMI and albumin levels decreased in a stepwise fashion only in group D (Figure 2). Furthermore, the analysis excluded some possible hypo-responders to EPO, such as the patients who had been transfused in the previous 3 months.
The limitations of our study also include the fact that its cross-sectional design means that it can only estimate the presence and strength of the associations between the explored variables, but cannot investigate the direction of the link between the cause (malnutrition–inflammation?) and EPO resistance (the effect?). Secondly, given that we used the percentile distribution of EPO-treated patients to identify hypo-responders, it is likely that some of them had depleted iron stores, as suggested by their lower levels of serum iron and percentage transferrin saturation in comparison with the other EPO-treated groups. However, the results of the multiple linear and logistic regression analyses did not change when the patients with iron store depletion were excluded. This once again underlines the need to monitor the iron status of haemodialysis patients closely because absolute or relative iron deficiency is still an important clinical problem in everyday clinical practice, even in selected and highly motivated centres. Finally, although widely used in the literature [4,6–8], the terms hypo-responsiveness and/or EPO resistance are more problematic than expected because they imply a poor response of erythron to EPO; as red cell formation is not easily measurable, like other authors [8], we used Hb levels as a surrogate.
In conclusion, our results should contribute towards a better understanding of the factors involved in the response to EPO therapy. Elucidating these factors not only has an intrinsic relevance per se (as a means of better correcting the anaemic state), but may also be useful in helping to identify patients at high risk of morbidity and mortality.
Conflict of interest statement. All of the authors declare that they have no conflict of interest. This study was carried out without any financial support from companies.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
- Locatelli F, Aljama P, Barany P et al. European best practice guidelines working group: revised European best practice guidelines for the management of anaemia in patients with chronic renal failure. Nephrol Dial Transplant 2004; 19 [Suppl 2]: ii1–ii47
- National Kidney Foundation I, Kidney–Dialysis Outcomes Quality Initiative: K/DOQI Clinical practice guidelines: anemia of chronic kidney disease. Am J Kidney Dis 2001; 37 [Suppl]: 182–238
- Schooley JC, Kullgren B, Allison AC. Inhibition by interleukin-1 of the action of erythropoietin on erythroid precursors and its possible role in the pathogenesis of hypoplastic anaemias. Br J Haematol 1987; 67: 11–17[ISI][Medline]
- Goicoechea M, Martin J, de Sequera P et al. Role of cytokines in the response to erythropoietin in hemodialysis patients. Kidney Int 1998; 54: 1337–1343[CrossRef][ISI][Medline]
- Allen DA, Breen C, Yaqoob MM, Macdougall IC. Inhibition of CFU-E colony formation in uremic patients with inflammatory disease: role of IFN-gamma and TNF-alpha. J Investig Med 1999; 47: 204–211[ISI][Medline]
- Bárány P, Divino Filho JC, Bergström J. High C-reactive protein is a strong predictor of resistance to erythropoietin in hemodialysis patients. Am J Kidney Dis 1997; 29: 565–568[ISI][Medline]
- Gunnell J, Yeun JY, Depner TA, Kaysen GA. Acute-phase response predicts erythropoietin resistance in hemodialysis and peritoneal dialysis patients. Am J Kidney Dis 1999; 33: 63–72[ISI][Medline]
- Kalantar-Zadeh K, McAllister CJ, Lehn RS, Lee GH, Nissenson AR, Kopple JD. Effect of malnutrition-inflammation complex syndrome on EPO hyporesponsiveness in maintenance haemodialysis patients. Am J Kidney Dis 2003; 42: 761–773[CrossRef][ISI][Medline]
- Oberg BP, McMenamin E, Lucas FL et al. Increased prevalence of oxidant stress and inflammation in patients with moderate to severe chronic kidney disease. Kidney Int 2004; 65: 1009–1016[CrossRef][ISI][Medline]
- Stenvinkel P, Heimburger O, Paultre F et al. Strong association between malnutrition, inflammation and atherosclerosis in chronic renal failure. Kidney Int 1999; 55: 1899–1911[CrossRef][ISI][Medline]
- Rocco MV, Paranandi L, Burrowes JD et al. Nutritional status in the HEMO Study cohort at baseline. Hemodialysis. Am J Kidney Dis 2002; 39: 245–256[Medline]
- 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: 32–37[Abstract]
- Locatelli F, Pisoni RL, Combe C et al. Anaemia in haemodialysis patients of five European countries: association with morbidity and mortality in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 2004; 19: 121–132
[Abstract/Free Full Text] - Daugirdas J. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error. J Am Soc Nephrol 1993; 4: 1205–1213[Abstract]
- Locatelli F, Andrulli S, Del Vecchio L. Difficulties of implementing clinical guidelines in medical practice. Nephrol Dial Transplant 2000; 15: 1284–1287
[Free Full Text] - Kalantar-Zadeh K, Block G, McAllister CJ et al. Appetite and inflammation, nutrition, anemia, and clinical outcome in hemodialysis patients. Am J Clin Nutr 2004; 80: 299–307
[Abstract/Free Full Text]
Accepted in revised form: 24.11.05
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Agarwal, J. L. Davis, and L. Smith Serum Albumin Is Strongly Associated with Erythropoietin Sensitivity in Hemodialysis Patients Clin. J. Am. Soc. Nephrol., January 1, 2008; 3(1): 98 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Girndt, P. Stenvinkel, C. Ulrich, J. Axelsson, L. Nordfors, P. Barany, J. J. Carrero, G. H. Heine, H. Kaul, and H. Kohler Influence of cytokine gene polymorphisms on erythropoetin dose requirements in chronic haemodialysis patients Nephrol. Dial. Transplant., December 1, 2007; 22(12): 3586 - 3592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M.C. Brunini, A. C. Mendes-Ribeiro, J. C. Ellory, and G. E. Mann Platelet nitric oxide synthesis in uremia and malnutrition: A role for L-arginine supplementation in vascular protection? Cardiovasc Res, January 15, 2007; 73(2): 359 - 367. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||