Original Paper Received: July 2, 2013 Accepted: December 12, 2013 Published online: March 1, 2014
Blood Purif 2014;37:93–98 DOI: 10.1159/000357968
Comparison between On-Line High-Efficiency Hemodiafiltration and Conventional High-Flux Hemodialysis for Polyclonal Free Light Chain Removal Thomas Lamy a Patrick Henri a Thierry Lobbedez a Elisabeth Comby b Jean-Philippe Ryckelynck a Maxence Ficheux a
Departments of a Nephrology and b Immunology, CHU Clemenceau, Caen, France
Key Words Hemodialysis · Hemodiafiltration · Polyclonal free light chains
Abstract Background: Polyclonal free light chains (FLC) are considered as middle molecular weight uremic toxins in chronic kidney disease. In this study, we investigate polyclonal FLC removal by comparing conventional high‐flux hemodialysis (HD) and online high-efficiency hemodiafiltration (ol‐HDF) in end-stage renal disease patients. Methods: We analyzed 31 chronic dialysis patients who were treated by HD then by postdilution ol-HDF during a prospective study. All patients were anuric and without monoclonal gammopathy. Serum pre- and postdialysis FLC were collected during 4 sessions: 1 HD session and 3 ol-HDF sessions. We calculated the reduction ratio using kinetic modeling. Results: The κ reduction ratio was higher with ol-HDF than with HD (66 ± 14 vs. 52 ± 13%, p < 0.001). However, the λ reduction ratio was not significantly higher with ol-HDF (37 ± 20 vs. 37 ± 15%, p = 0.67). Furthermore, predialysis κ- and λ-FLC increased with ol-HDF compared with HD (κ 155 ± 82 vs. 87 ± 47 mg/l, p < 0.05; λ
© 2014 S. Karger AG, Basel 0000–0000/14/0372–0093$39.50/0 E-Mail [email protected] www.karger.com/bpu
101 ± 46 vs. 72 ± 41 mg/l, p < 0.05). Postdialysis FLC levels were raised only for λ-FLC with ol-HDF (74 ± 39 vs. 53 ± 31 mg/l, p < 0.05) and were not significantly different for κ. Conclusions: This study shows that κ-FLC removal is better in ol-HDF compared with HD, whereas there is no difference in λ-FLC removal. Surprisingly, predialysis κ and λ levels are both increased in ol-HDF, which is disturbing since polyclonal excess of λ-FLC is associated with mortality in chronic kidney disease. © 2014 S. Karger AG, Basel
Introduction
During the production of intact immunoglobulins, light chains are produced in excess of heavy chains at a rate of 500 mg/day. There are two types of free light chains (FLC) (κ and λ), which are produced by B lymphocytes. These FLC are filtered by the glomeruli and eliminated by the proximal renal tubule. Cubilin, a glycoprotein receptor located in the apical membrane of apical tubular cells, is involved in the tubular trafficking of FLC [1, 2]. Light chain serum half-lives are between 2 and 6 h in normal subjects [3]. Maxence Ficheux Department of Nephrology CHU Clemenceau FR–14033 Caen Cedex (France) E-Mail ficheux-m @ chu-caen.fr
It has been shown that, compared with healthy subjects, patients with chronic renal insufficiency have higher concentrations of serum polyclonal FLC [4, 5]. Renal impairment negatively affects FLC clearance; consequently, serum half-life of FLC increase in patients with renal insufficiency. It has been demonstrated that patients with end-stage renal disease (ESRD) treated by hemodialysis (HD) or peritoneal dialysis have higher serum concentrations of polyclonal FLC than their counterparts not under renal replacement therapy [4, 5]. It has been hypothesized that, in patients with renal impairment, serum FLC in excess act as uremic toxins [6–8]. Indeed, when in excess, FLC inhibit neutrophils functions, which may increase the risk of bacterial infections [9]. Furthermore, excess of polyclonal λ-FLC is associated with an increased risk of mortality, both in chronic kidney disease patients and in the general population [10–13]. FLC are classified as middle-sized molecular solutes since their molecular weight is 25 kDa. λ-FLC are present in the plasma of normal subjects in dimeric forms of 50 kDa. Several studies have shown that on-line hemodiafiltration (ol-HDF) with high substitution volume increases middle-sized molecule removal compared with conventional HD [14–18]. In an attempt to remove monoclonal FLC, plasma exchange, high-flux HD, and ol-HDF were used in patients suffering from multiple myeloma [19–22]. In addition, one recent study, which focused on the effect of dialysis membrane on FLC removal during renal replacement therapy, showed that FLC clearance increased when a high molecular weight cut-off membrane was used [23]. As excess FLC may play a role in the outcome of ESRD patients on dialysis, information about the clearance of FLC during the different dialysis modalities is required. This study was carried out to compare polyclonal FLC removal during conventional high-flux HD and ol-HDF in anuric ESRD patients on renal replacement therapy. The second point was to compare β2-microglobulin removal between HD and ol-HDF. Patients and Methods Patients This was a prospective, observational, before and after study which enrolled ESRD patients treated by high-flux HD at our center. We included anuric patients on dialysis for more than 3 months. To be included in the study, patients had to be treated with regular three times weekly HD sessions. Patients with monoclonal gammopathy were excluded from the study. Patients were switched from high-flux HD to postdilution olHDF. Blood samples were collected at inclusion (w1) and after 5 (w5), 9 (w9) and 13 (w13) weeks on ol-HDF.
94
Blood Purif 2014;37:93–98 DOI: 10.1159/000357968
Patients were dialyzed with an FX membrane, a high-flux polysulfone-based Helixone membrane (Fresenius Medical Care, Bad Homburg, Germany) for the entire duration of the study. FX-80 or FX-100, with effective surface areas of 1.8 and 2.2 m2 and ultrafiltration coefficients of 59 and 73 ml/h × mm Hg respectively, were used during the study period. HD and ol-HDF sessions were performed with a Fresenius 5008 monitor (Fresenius Medical Care). The dialysate flow rate was fixed at 500 ml/min; blood flow rate was targeted to be >300 ml/min. Acetate-free dialysis solution was used during the study. During ol-HDF sessions, the substitution fluid flow was automatically matched to the effective blood flow rate by the monitor (AutoSub; Fresenius Medical Care). Anticoagulation treatment was by identical form and dosage of the previous routine heparinization, using a low molecular weight heparin. The study was approved by the ethical board of our institution. Blood Sample Analysis Blood samples were taken before the dialysis session (Cpre) and at the end of the dialysis session (Cpost) from the arterial blood line using the slow-flow method. As recommended, blood samples were drawn at the mid-week dialysis session. κ- and λ-FLC concentrations were measured by nephelometry with a Siemens BNII analyzer, using the Freelite® immunoassay (The Binding Site Company, Birmingham, UK). β2-Microglobulin concentration was also assessed by a nephelometry method (Immage®; Beckman Coulter, USA). Calculations κ- and λ-FLC reduction ratios were calculated with the following formula: Reduction ratio = (Cpre – Cpost)/Cpre, where Cpre and Cpost were the pre- and posttreatment concentrations. Posttreatment FLC concentrations were corrected for hemoconcentration using a single-compartment kinetic model with the following formula according to Bergström and Wehle [24]: Cpost-corrected = Cpost/(1 + ΔBW)/(0.2 × BWpost) where ΔBW is the weight lost during the session and BWpost the body weight at the end of the session. Effective κ- and λ-FLC clearances (Kd) were calculated as follows, according to Leypoldt et al. [25]: Kd = (ΔBW/D) × [1 – log(Cpost/Cpre)/log(1 + ΔBW/0.2 × BWpost)] where D is the session length. Statistical Analysis Continuous variables were expressed as mean ± SD or median and interquartile ranges. Categorical variables were expressed as proportions. The univariate analysis was performed using Fischer’s exact test for the categorical variables and the paired t test or Wilcoxon’s test when required for the continuous variables. A linear mixed model with subject as a random effect was used for the multivariate analysis to estimate the FLC concentration over time. Normality of the distribution of the serum FLC concentrations was checked graphically by density curves. In an attempt to estimate the relationship between serum FLC concentration and HDF, time on HDF after dialysis modality switch was entered in the model as a fixed effect. Multivariate analysis was adjusted for the initial FLC concentration and HD duration. Statistical difference was considered to be significant for a p value
Blood Purif 2014;37:93–98 DOI: 10.1159/000357968
Comparison between On-Line High-Efficiency Hemodiafiltration and Conventional High-Flux Hemodialysis for Polyclonal Free Light Chain Removal Thomas Lamy a Patrick Henri a Thierry Lobbedez a Elisabeth Comby b Jean-Philippe Ryckelynck a Maxence Ficheux a
Departments of a Nephrology and b Immunology, CHU Clemenceau, Caen, France
Key Words Hemodialysis · Hemodiafiltration · Polyclonal free light chains
Abstract Background: Polyclonal free light chains (FLC) are considered as middle molecular weight uremic toxins in chronic kidney disease. In this study, we investigate polyclonal FLC removal by comparing conventional high‐flux hemodialysis (HD) and online high-efficiency hemodiafiltration (ol‐HDF) in end-stage renal disease patients. Methods: We analyzed 31 chronic dialysis patients who were treated by HD then by postdilution ol-HDF during a prospective study. All patients were anuric and without monoclonal gammopathy. Serum pre- and postdialysis FLC were collected during 4 sessions: 1 HD session and 3 ol-HDF sessions. We calculated the reduction ratio using kinetic modeling. Results: The κ reduction ratio was higher with ol-HDF than with HD (66 ± 14 vs. 52 ± 13%, p < 0.001). However, the λ reduction ratio was not significantly higher with ol-HDF (37 ± 20 vs. 37 ± 15%, p = 0.67). Furthermore, predialysis κ- and λ-FLC increased with ol-HDF compared with HD (κ 155 ± 82 vs. 87 ± 47 mg/l, p < 0.05; λ
© 2014 S. Karger AG, Basel 0000–0000/14/0372–0093$39.50/0 E-Mail [email protected] www.karger.com/bpu
101 ± 46 vs. 72 ± 41 mg/l, p < 0.05). Postdialysis FLC levels were raised only for λ-FLC with ol-HDF (74 ± 39 vs. 53 ± 31 mg/l, p < 0.05) and were not significantly different for κ. Conclusions: This study shows that κ-FLC removal is better in ol-HDF compared with HD, whereas there is no difference in λ-FLC removal. Surprisingly, predialysis κ and λ levels are both increased in ol-HDF, which is disturbing since polyclonal excess of λ-FLC is associated with mortality in chronic kidney disease. © 2014 S. Karger AG, Basel
Introduction
During the production of intact immunoglobulins, light chains are produced in excess of heavy chains at a rate of 500 mg/day. There are two types of free light chains (FLC) (κ and λ), which are produced by B lymphocytes. These FLC are filtered by the glomeruli and eliminated by the proximal renal tubule. Cubilin, a glycoprotein receptor located in the apical membrane of apical tubular cells, is involved in the tubular trafficking of FLC [1, 2]. Light chain serum half-lives are between 2 and 6 h in normal subjects [3]. Maxence Ficheux Department of Nephrology CHU Clemenceau FR–14033 Caen Cedex (France) E-Mail ficheux-m @ chu-caen.fr
It has been shown that, compared with healthy subjects, patients with chronic renal insufficiency have higher concentrations of serum polyclonal FLC [4, 5]. Renal impairment negatively affects FLC clearance; consequently, serum half-life of FLC increase in patients with renal insufficiency. It has been demonstrated that patients with end-stage renal disease (ESRD) treated by hemodialysis (HD) or peritoneal dialysis have higher serum concentrations of polyclonal FLC than their counterparts not under renal replacement therapy [4, 5]. It has been hypothesized that, in patients with renal impairment, serum FLC in excess act as uremic toxins [6–8]. Indeed, when in excess, FLC inhibit neutrophils functions, which may increase the risk of bacterial infections [9]. Furthermore, excess of polyclonal λ-FLC is associated with an increased risk of mortality, both in chronic kidney disease patients and in the general population [10–13]. FLC are classified as middle-sized molecular solutes since their molecular weight is 25 kDa. λ-FLC are present in the plasma of normal subjects in dimeric forms of 50 kDa. Several studies have shown that on-line hemodiafiltration (ol-HDF) with high substitution volume increases middle-sized molecule removal compared with conventional HD [14–18]. In an attempt to remove monoclonal FLC, plasma exchange, high-flux HD, and ol-HDF were used in patients suffering from multiple myeloma [19–22]. In addition, one recent study, which focused on the effect of dialysis membrane on FLC removal during renal replacement therapy, showed that FLC clearance increased when a high molecular weight cut-off membrane was used [23]. As excess FLC may play a role in the outcome of ESRD patients on dialysis, information about the clearance of FLC during the different dialysis modalities is required. This study was carried out to compare polyclonal FLC removal during conventional high-flux HD and ol-HDF in anuric ESRD patients on renal replacement therapy. The second point was to compare β2-microglobulin removal between HD and ol-HDF. Patients and Methods Patients This was a prospective, observational, before and after study which enrolled ESRD patients treated by high-flux HD at our center. We included anuric patients on dialysis for more than 3 months. To be included in the study, patients had to be treated with regular three times weekly HD sessions. Patients with monoclonal gammopathy were excluded from the study. Patients were switched from high-flux HD to postdilution olHDF. Blood samples were collected at inclusion (w1) and after 5 (w5), 9 (w9) and 13 (w13) weeks on ol-HDF.
94
Blood Purif 2014;37:93–98 DOI: 10.1159/000357968
Patients were dialyzed with an FX membrane, a high-flux polysulfone-based Helixone membrane (Fresenius Medical Care, Bad Homburg, Germany) for the entire duration of the study. FX-80 or FX-100, with effective surface areas of 1.8 and 2.2 m2 and ultrafiltration coefficients of 59 and 73 ml/h × mm Hg respectively, were used during the study period. HD and ol-HDF sessions were performed with a Fresenius 5008 monitor (Fresenius Medical Care). The dialysate flow rate was fixed at 500 ml/min; blood flow rate was targeted to be >300 ml/min. Acetate-free dialysis solution was used during the study. During ol-HDF sessions, the substitution fluid flow was automatically matched to the effective blood flow rate by the monitor (AutoSub; Fresenius Medical Care). Anticoagulation treatment was by identical form and dosage of the previous routine heparinization, using a low molecular weight heparin. The study was approved by the ethical board of our institution. Blood Sample Analysis Blood samples were taken before the dialysis session (Cpre) and at the end of the dialysis session (Cpost) from the arterial blood line using the slow-flow method. As recommended, blood samples were drawn at the mid-week dialysis session. κ- and λ-FLC concentrations were measured by nephelometry with a Siemens BNII analyzer, using the Freelite® immunoassay (The Binding Site Company, Birmingham, UK). β2-Microglobulin concentration was also assessed by a nephelometry method (Immage®; Beckman Coulter, USA). Calculations κ- and λ-FLC reduction ratios were calculated with the following formula: Reduction ratio = (Cpre – Cpost)/Cpre, where Cpre and Cpost were the pre- and posttreatment concentrations. Posttreatment FLC concentrations were corrected for hemoconcentration using a single-compartment kinetic model with the following formula according to Bergström and Wehle [24]: Cpost-corrected = Cpost/(1 + ΔBW)/(0.2 × BWpost) where ΔBW is the weight lost during the session and BWpost the body weight at the end of the session. Effective κ- and λ-FLC clearances (Kd) were calculated as follows, according to Leypoldt et al. [25]: Kd = (ΔBW/D) × [1 – log(Cpost/Cpre)/log(1 + ΔBW/0.2 × BWpost)] where D is the session length. Statistical Analysis Continuous variables were expressed as mean ± SD or median and interquartile ranges. Categorical variables were expressed as proportions. The univariate analysis was performed using Fischer’s exact test for the categorical variables and the paired t test or Wilcoxon’s test when required for the continuous variables. A linear mixed model with subject as a random effect was used for the multivariate analysis to estimate the FLC concentration over time. Normality of the distribution of the serum FLC concentrations was checked graphically by density curves. In an attempt to estimate the relationship between serum FLC concentration and HDF, time on HDF after dialysis modality switch was entered in the model as a fixed effect. Multivariate analysis was adjusted for the initial FLC concentration and HD duration. Statistical difference was considered to be significant for a p value
