Soluble transferrin receptor as a marker of erythropoiesis in patients undergoing high-flux hemodialysis

Anemia is a common complication in chronic kidney disease (CKD) patients receiving hemodialysis. The effect of high-flux dialysis (HFD) on anemia remains unclear. This prospective study aimed to evaluate the effect of HFD on anemia, and the potential of soluble transferrin receptor (sTfR) as a marker of iron status and erythropoiesis in CKD patients on hemodialysis. Forty patients, who switched from conventional low-flux dialysis to HFD for 12 months, were enrolled in this study. The levels of sTfR, hemoglobin (Hb), iron, and nutritional markers, as well as the dose of recombinant human erythropoietin (rhEPO) and use of chalybeate were determined at 0, 2, 6, and 12 months after starting HFD. HFD significantly increased the hemoglobin level and reduced sTfR level in CKD patients (p < 0.05). In addition, significant decreasing linear trends were observed for rhEPO dosage and chalybeate use (p < 0.05). The level of sTfR was positively correlated with the percentage of reticulocytes (RET%), rhEPO dose, and chalybeate use, while it was negatively correlated with Hb levels and total iron-binding capacity results (all p < 0.05). A univariate generalized estimating equation (GEE) model showed that the Hb level, RET%, rhEPO dose, and chalybeate use were the variables associated with sTfR levels. A multivariate GEE model showed that the time points when hemodialysis was performed were the variables associated significantly with sTfR levels. Overall, our findings suggest that HFD can effectively improve renal anemia in hemodialysis patients, and sTfR could be used as a marker of erythropoiesis in HFD patients.


INTRODUCTION
Anemia is a common complication of chronic kidney disease (CKD), especially in patients on hemodialysis for end-stage renal disease (ESRD); it results in reduced quality of life and is associated with considerable morbidity and mortality [1][2][3].The main cause of renal anemia is decreased production of erythropoietin (EPO), a glycoprotein hormone that is produced by the kidneys and is involved in the regulation of erythropoiesis [4].Deficiency or limited availability of iron for erythropoiesis is another important cause of renal anemia [5].Moreover, in hemodialysis patients, poor iron absorption, and the loss of blood and iron associated with hemodialysis also contribute to renal anemia [4,6].
High-flux dialysis (HFD) was developed to improve the efficiency of dialysis and to decrease the adverse effects.However, the effects of HFD on anemia remain unclear.
A European multicenter controlled study by Merello et al. [7] showed that, during 6 months of HFD, there was a significant increase in hemoglobin (Hb) levels in these patients.Similarly, Ayli et al. [8] reported that 6-month HFD treatment significantly increased Hb levels and reduced the dose of recombinant human EPO (rhEPO) used in patients with ESRD.Contradictory to these results, multicenter controlled studies by Locatelli et al. [9] and Schneider et al. [10] reported that HFD treatment lasting for 3 months and 1 year, respectively, had no superior effect on Hb levels.Thus, any advantages of HFD on renal anemia remain yet to be determined.
At present, rhEPO is frequently used to treat renal anemia.The clinical practice guidelines for anemia in CKD patients developed by The National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF KDOQI)™ [11] and European Best Practice Guidelines for the management of anemia in CKD [12], both indicate that iron status should be evaluated and iron deficiency should be corrected in patients with renal anemia receiving rhEPO therapy.This is due to the fact that optimal iron stores are necessary for rhEPO to be maximally effective [13].The standard biochemical markers of iron status, also widely used in hemodialysis patients, include serum ferritin (SF) and transferrin saturation (TSAT) [14].However, these standard tests may not always accurately reflect the iron status of patients due to interference from disease, nutritional status, or inflammatory conditions.
The soluble transferrin receptor (sTfR) is an extracellular portion of the transferrin receptor, a transmembrane protein responsible for iron transport.It is truncated from the cell surface and released into the serum [15].sTfR has been used as a marker for the iron status in long-term hemodialysis patients [16,17].Furthermore, low iron stores caused elevated levels of sTfR, and this was not affected by age, sex, tissue injury, or acute phase inflammation [18].In addition, sTfR is a marker of erythropoiesis [17][18][19][20].So far, no study has investigated sTfR levels in patients treated with HFD.
The purpose of this study was to prospectively investigate whether HFD improves renal anemia and to evaluate sTfR as a marker of iron status and erythropoiesis in patients with CKD treated with HFD.

Patients
A total of 40 patients undergoing maintenance hemodialysis at the Blood Purification Center of the First Affiliated Hospital of Chinese PLA General Hospital, from June 2012 to June 2013, were enrolled in this study.The inclusion criteria were 1) 18-75 years old; 2) treated with low-flux hemodialysis for at least 6 months; 3) receiving low-flux hemodialysis 3 times/week, 4 hours each session, and a single-pool Kt/V (spKt/V) index ≥1.2;4) had a permanent vascular access; 5) receiving rhEPO or iron therapy for at least 1 month.The exclusion criteria were 1) Vitamin B12 or B9 (folate) deficiency; 2) severe heart failure; 3) active or concealed bleeding; 4) received a blood transfusion or antibiotic therapy in the month prior to HFD; 5) malignancy, end-stage liver disease, or other serious comorbidities.The study was performed in accordance with the ethical principles of the Declaration of Helsinki and approved by the Institutional Review Board of Our Hospital.Written informed consent was obtained from each participant.

Hemodialysis
Hemodialysis was performed with a Fresenius FX-class dialyzer (Fresenius Medical Care, Germany) using a high-flux membrane (FX-60, surface area of 1.4 m 2 , ultrafiltration coefficient of 46 mL/h/mm Hg, Fresenius Medical Care) through an autologous arteriovenous fistula, for 4 hours and 3 times a week.The blood flow rate was 250-300 mL/min, and the dialysate flow rate was 500 mL/min.Low-molecular heparin was used for anticoagulation.

Anemia therapy
During the study period, the Hb level of patients was maintained within a range of 10-13 g/dL by EPO therapy.When the Hb level was outside of the desired range, rhEPO was administered according to the following criteria: 1) Hb <10 g/dL: rhEPO dose increased by 25%; 2) Hb >13 g/dL: rhEPO dose reduced by 25%; 3) Hb >14 g/dL: rhEPO administration stopped immediately.In the patients with iron deficiency (SF <100 µg/L or TSAT <20%), iron replacement therapy was conducted in accordance with the standard practice at our center.rhEPO and chalybeate doses were adjusted weekly based on the laboratory test results of each patient.

Data collection
At 0, 2, 6, and 12 months after beginning of HFD, fasting blood samples were collected at the first dialysis session of the week.The following commercial kits, reagents, or devices were used for biochemical analysis: enzyme-linked immunosorbent assay (sTfR ELISA kit, Santa Cruz, USA) for serum sTfR; COULTER LH 780 hematology analyzer (Beckman Coulter, USA) for Hb, hematocrit (HCT), and percentage of reticulocytes (RET%); Ferritin Reagent (Beckman Coulter) and a UniCel DxI 800 Immunoassay System (Beckman Coulter) for SF; Quick auto neo Fe 7170 kit (SHINO-TEST Co., Japan) for total iron-binding capacity (TIBC), TSAT, and serum iron (SI); Bromocresol green albumin assay kit (SLB010, Beijing Leadman Biochemistry, China) for albumin; an assay kit from Desai Diagnostic System Co. (China) for cholesterol; Triglycerides (TGs) glycerol phosphate oxidase (GPO)-peroxidase (POD) assay kit (Proline, Indonesia) for TG; Low-density lipoprotein (LDL) test kit (GS141Z, Beijing Strong Biotechnologies, China) for LDL; TIBC, TSAT, SI, albumin, cholesterol, TG, and LDL were analyzed on an automatic biochemistry analyzer (Hitachi 7600-120E, Tokyo, Japan).All assays were performed according to the respective manufacturer' s instructions.rhEPO and chalybeate doses were recorded weekly.The rate of chalybeate use was defined as the percentage of patients using chalybeate in the current month.

Statistical analysis
Continuous variables were reported as mean ± standard deviation (SD) and compared with one-way repeated measures analysis of variance (ANOVA) at different time points (i.e.0, 2, 6, and 12 months after beginning of HFD).Fisher' s Least Significant Difference test was used for post hoc comparisons.Categorical variables were presented as number and percentage (%) and compared using Chi-square test.The linear trend across the 4 time points was analyzed by the linearity test.The correlation between variables was determined by the Pearson correlation coefficient.Generalized estimating equations (GEE) models were applied to determine independent variables associated with sTfR, as there were four repeated measurements for all patients.The significance level for all tests was set at a two-tailed p < 0.05.Statistical analyses were performed using IBM SPSS Version 20 (SPSS Statistics V20, IBM Corporation, Somers, New York, USA).

Patients
A total of 40 patients treated with conventional low-flux hemodialysis were enrolled in the study, and switched to HFD for 12 months.The sample included 13 (32.5%)males and 27 (67.5%)females, with a mean age of 55.83 ± 14.54 years and a mean hemodialysis duration of 5 years (range: 2-7 years).The list of reasons why patients required hemodialysis is shown in Table 1, and no significant difference was found between the male and female patients (p = 0.284).

Biochemical parameters of patients after HFD
To evaluate the effect of HFD on the biochemical status of patients, markers of anemia, iron, and nutrition were assessed at 0, 2, 6, and 12 months after the onset of HFD.The changes across time points and linear trends were analyzed.As shown in Table 2, the sTfR, Hb, HCT, albumin, prealbumin levels, rhEPO dose, and rate of chalybeate use were significantly different across the 4 time points (all p < 0.05, one-way repeated measures ANOVA), with a significant linear trend (all p < 0.05, linearity test).The TIBC, SF, and LDL levels were also significantly different across the 4 time points (all p < 0.05), but no linear trend was observed (all p > 0.05).We did not observed significant differences in the SI, TSAT, RET%, Kt/V, cholesterol, and TG level across the 4 time points (all p > 0.05).

Independent variables associated with sTfR
To evaluate sTfR as a marker of iron status and erythropoiesis, we identified independent variables associated with sTfR, by correlation analysis and regression models.As shown in Table 3, the Hb, HCT, and TIBC were negatively correlated with sTfR (all p < 0.05), while the RET%, albumin, prealbumin, rhEPO dose, and rate of chalybeate use were positively correlated with sTfR (all p < 0.05).Next, univariate and multivariate GEE models were used to investigate the variables associated with sTfR.As shown in Table 4, the time points when hemodialysis was performed, HCT, Hb, RET%, Kt/V, albumin, prealbumin, rhEPO dose, and rate of chalybeate use were the variables significantly associated with sTfR in the univariate GEE model (all p < 0.05).However, only the time points when hemodialysis was performed were significantly associated with sTfR, in the multivariate model adjusted for

DISCUSSION
In this study, we investigated the effect of HFD on renal anemia and the potential of sTfR as a marker of iron status and erythropoiesis, in 40 patients who switched from conventional low-flux dialysis to HFD for 12 months.The results showed that HFD significantly increased the Hb and HCT, and reduced serum sTfR level, rhEPO dose, and rate of chalybeate use in the CKD patients, with significant linear trends.sTfR was positively correlated with the RET%, albumin, prealbumin, rhEPO dose, and rate of chalybeate use, while it was negatively correlated with the Hb, HCT, and TIBC.Furthermore, the univariate GEE model revealed that the time points when hemodialysis was conducted, HCT, Hb, RET%, rhEPO dose, and rate of chalybeate use were the variables associated with sTfR.However, in the multivariate GEE model, only the time points when hemodialysis was performed were significantly associated with sTfR.Taken together, these data suggest that HFD could improve renal anemia in hemodialysis patients, and that sTfR might be used as a marker of erythropoiesis in HFD patients.To the best of our knowledge, this is the first study to report changes of sTfR level in patients treated with HFD.
In our study, both Hb and HCT were significantly increased at 2 months after the onset of HFD, and this trend  continued during the next 10 months, indicating that HFD can reduce the severity of renal anemia.Furthermore, the significantly reduced rhEPO dose and rate of chalybeate use by HFD, also support the positive effects of HFD on renal anemia.In addition, our findings are consistent with previous reports [7,8].By contrast, changes in the four conventional markers of iron status (i.e.SF, TIBC, TSAT, and SI) did not show a linear trend during the 12 months of HFD treatment.Among these markers, measurement of SF is the most specific biochemical test for determining relative total body iron store [21].However, SF level can be affected by infectious or inflammatory conditions [21].Our results showed that SF levels were significantly different across the 4 time points, but a linear trend was not observed for these changes.For example, the SF levels were increased at 6 months, but markedly decreased at 12 months, which is inconsistent with the linear trends observed for the Hb, HCT, rhEPO dose, and rate of chalybeate use.These observations may indicate that SF level is easily affected by other factors and, thus, is not a suitable marker for iron status.However, we did not assess inflammatory markers in these patients; therefore, it is unclear if any inflammatory process contributed to the changes in SF levels.
In this study, sTfR level was significantly reduced after the onset of HFD.In addition, the multivariate GEE model confirmed that sTfR level had gradually declined over time.These changes of sTfR showed a significant decreasing linear trend, consistent with the trends observed for rhEPO dose and rate of chalybeate use.Contrary to that, Hb and HCT levels showed a significant increasing linear trend.In addition, sTfR was significantly correlated with the Hb, HCT, RET%, rhEPO dose, and rate of chalybeate use, indicating that sTfR could be used as a marker of erythropoiesis, which is also consistent with previous studies [17][18][19][20][22][23][24][25].Only one (TIBC) out of the four iron parameters was negatively correlated with sTfR, indicating that sTfR could not accurately reflect the iron status of HFD patients in our study.
Our study has several limitations.First, the small sample size was included.Second, we did not enrolled patients undergoing low-flux dialysis as a control group.Finally, we did not collect the data of the same patient before switching to HFD.A well-designed study with a large sample size is necessary to validate our findings.Furthermore, we did not assess inflammatory status nor hepcidin levels of the patients, during HFD.Nevertheless, a study on hemodialysis patients reported that sTfR was positively associated with rhEPO dose, and no correlation between sTfR and C-reactive protein (an inflammatory marker) was observed, suggesting that sTfR level is not affected by inflammatory processes [26].

CONCLUSION
In summary, our study demonstrated that HFD can improve renal anemia in hemodialysis patients, and that sTfR could be used as a potential marker of erythropoiesis in HFD patients.

TABLE 1 .
The main reasons for hemodialysis among chronic kidney disease patients