Bisphenol A in Chronic Kidney Disease

Bisphenol A in Chronic Kidney Disease
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Bisphenol A in Chronic Kidney Disease *Detlef H. Krieter, **Bernard Canaud, †Horst-Dieter Lemke, **Annie Rodriguez,‡Andreas Morgenroth, §Kai von Appen, ¶Gerd-Peter Dragoun, and *Christoph Wanner *Department of Medicine, Division of Nephrology, University of Würzburg, Würzburg; †Research and Development,Membrana GmbH, Wuppertal; ‡Dialysis Center, Elsenfeld; §Hamburger Dialyse Lohbrügge, Hamburg; ¶KfH Center, Aschaffenburg, Germany; and **Department of Nephrology Hôpital Lapeyronie, University of Montpellier, Montpellier,France Abstract:  The estrogenic endocrine-disrupting substancebisphenolA(BPA)isextensivelyusedasastartingmaterialforavarietyofconsumerplasticproductsincludingdialyzermaterials. The present study was performed to exploreplasma BPA levels in patients with impaired renal functionand to investigate if dialyzers differing in elutable BPAinfluenceplasmalevelsinpatientsonmaintenancehemodi-alysis. In vitro BPA was eluted from high-flux polyether-sulfone (PUREMA H, referred as PUR-H), high-fluxpolysulfone (referred as HF-PSu),and low-flux polysulfone(referred as LF-PSu) dialyzers by recirculation with waterfor 180 min. In a cross-sectional clinical study, plasma BPAlevels of outpatients with different stages of chronic kidneydisease (CKD) from four different centers were deter-mined. Furthermore, in a prospective, randomized, andcrossover setting, 18 maintenance dialysis patients weresubjected successively to 4 weeks of thrice-weekly hemodi-alysis with each LF-PSu,HF-PSu,and PUR-H.In addition,the fractions of protein-bound and free BPA were deter-mined in a subset of dialysis patients. The mass of BPAelutedfromthebloodcompartmentsinvitrounderaqueousconditions varied for the three dialyzers being very low forPUR-H (6.2  2.5 ng;  P   <  0.001), intermediate for HF-PSu(48.1  7.7 ng), and highest for LF-PSu (140.8  38.7 ng; P   <  0.01). In 152 prevalent patients with CKD enrolled inthe cross-sectional trial, plasma BPA started to rise afterstage 3. Maintenance hemodialysis patients had more thansix times higher BPA concentrations than patients withCKD stage 5 not yet on dialysis (10.0  6.6 vs.1.6  1.8 ng/mL;  P   <  0.001). The BPA concentrations highly andinversely correlated with renal function. In the rando-mized controlled study, the plasma BPA concentrationswere highly elevated compared with healthy controls(range 9.1  4.5–12.0  6.0 ng/mL vs.   0.2  0.1 ng/mL; P   <  0.001),but no change of the plasma levels was observedduring hemodialysis with any of the three dialyzers in thecourse of a single treatment and over a period of 4 weeks.The protein-bound fraction of plasma BPA in the dialysispatients was 74  5%. Renal function and, most likely, thetotal quantity of ingested BPA are essential parametersaffecting plasma BPA concentrations. Dialyzers are oneadditional source of BPA, but differences in the elutableBPA content are not associated with a significant effect onBPAplasmalevelsinWesternEuropeanmaintenancedialy-sis patients. Due to high protein binding, the removal of BPA by hemodialysis is limited.  Key Words:  BisphenolA—Toxicity—Protein binding—Chronic kidney disease—End-stage renal disease—Hemodialysis. BisphenolA (BPA) is used extensively as a startingmaterial for epoxy resins lining food and beveragecontainers and as a monomer in polycarbonate andpolysulfone-based plastics in a multitude of con-sumer products. Ninety-three percent of the USpopulation has detectable levels of BPA, providingevidence of widespread and continuous exposure toBPA primarily through food, drinking water, dentalfillings,dermal exposure,and inhalation of householddusts (1,2).BPA acts as an estrogenic endocrine-disruptingchemical. In rodents, it has been demonstrated toelicit neurochemical alterations and reproductiveand developmental abnormalities (3,4). Additionalactions include liver damage (5–7), disrupted thyroidhormone and pancreatic cell function (8,9), andobesity-promoting effects (10).Whether the findingsin rodents and other animals can be extrapolated tohumans is highly debated (11,12). Concerns havebeen raised particularly with regard to an implicationin the development of newborns and infants as well doi:10.1111/j.1525-1594.2012.01556.xReceived March 2012; revised June 2012.Address correspondence and reprint requests to Dr. Detlef H.Krieter,Department of Medicine,Division of Nephrology,Univer-sity Hospital Würzburg, Oberdürrbacher Str. 6, 97080 Würzburg,Germany. E-mail: krieter_d@medizin.uni-wuerzburg.de bs_bs_banner © 2012, Copyright the AuthorsArtificial Organs © 2012, International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.  Artificial Organs  2012, ••(••):••–••  as a possible interaction with breast and prostatecancers in adults (11–13). Recently, an association of higher urinary concentrations of BPA with anincreased prevalence of heart disease,type 2 diabetesmellitus, obesity, and liver enzyme abnormalities wasdemonstrated in cross-sectional studies (14–18),fueling a controversy about potential hazards of low-dose effects and currently recommended exposurethresholds in humans (14,19).After glucoronization in the liver, BPA is rapidlyeliminated by the kidneys with a half-life in blood of less than 2 h after oral ingestion that generally resultsin low blood levels (12). In contrast, in patients withimpaired renal function, the urinary excretion of BPA seems to decline permitting its accumulation(20,21). In addition to other environmental sources,patients with end-stage renal disease on hemodialysisare repeatedly exposed to BPA from components of the dialyzer, more specifically polycarbonate hous-ings and polysulfone dialysis membranes (20,22,23).Thus, it is not surprising that hemodialysis patientshave much higher BPA blood levels than healthysubjects (20,23). Nevertheless, only a very few small-scale studies, which exclusively focus on Japanesepatient populations, have addressed BPA in chronickidney disease (CKD).The purpose of the present study was to exploreplasma BPA levels in patients with different stages of CKD and to investigate if dialyzers with differentamounts of elutable BPA impact on the plasma con-centrations in hemodialysis patients. PATIENTS AND METHODSIn vitro elution of BPA from dialyzers BPA elution was performed in vitro on the threedifferent dialyzers: Nipro Pureflux 170 H (NiproCorp.,Osaka,Japan),Fresenius F 60 S,and FreseniusF 6 HPS (both Fresenius Medical Care AG, BadHomburg, Germany). All had polycarbonate hous-ings but different dialysis membranes, which wereselected because of assumed variations in BPAcontent.The dialysis membranes were high-flux poly-ethersulfone PUREMA H (referred as PUR-H;1.7 m 2 ; Membrana GmbH, Wuppertal, Germany),high-flux polysulfone (referred as HF-PSu; 1.3 m 2 ),and low-flux polysulfone (referred as LF-PSu;1.3 m 2 ). Compared with both polysulfone mem-branes, PUR-H had a considerably larger surfacearea. Initially, pilot in vitro experiments were per-formed comparing sterile water,a 50% water/ethanolmixture,and freshly donated human blood to identifythe most suitable elution fluid in a recirculationsetting. Although with the water/ethanol mixture,BPA concentration exceeded the upper test limit of 250 ng/mL, the level of BPA eluted with blood wasrather low (data not shown). Therefore, the in vitroexperiments ( n  =  6 for each dialyzer) were per-formed with sterile water because the differences ineluted BPA between dialyzers had the same trend asobserved with blood. The dialyzer blood and dialy-sate compartments were recirculated countercur-rently with 400 mL of verified BPA-free water eachat 250 mL/min and 37°C for 180 min (Fig. 1). Therecirculated fluids were lyophilized and reconstitutedin phosphate buffered saline buffer for BPA mea-surement, which was done separately for the bloodand the dialysate compartment. BPA was expressedas mass eluted per dialyzer. Clinical studies Two clinical trials were performed consecutivelystarting with a cross-sectional reference study fol-lowed by a prospective trial. Study approval wasgiven by the Freiburg Ethics Committee Interna-tional. Study enrollment proceeded after patients,and controls had given written informed consent. Cross-sectional clinical study Plasma BPA levels of outpatients with differentstages of CKD from four different nephrologicalcenters (three German centers and one Frenchcenter) were measured on a single occasion.In main-tenance dialysis patients,the blood sample was drawnprior to a midweek dialysis treatment. All dialysispatients were on polysulfone or polyethersulfonedialysis membranes for   3 months. Basic demo-graphic characteristics were documented. The esti-mated glomerular filtration rate (eGFR) of CKD FIG. 1.  Setup of the in vitro elution experiments. Dialyzer bloodand dialysate compartments were recirculated countercurrentlywith 400 mL of sterile water each at 250 mL/min and 37°C for180 min. D.H. KRIETER ET AL. 2  Artif Organs,Vol. ••, No. ••, 2012  patients not on dialysis was determined by applyingthe modification of diet in renal disease (MDRD)formula (24). If available, residual renal function inCKD 5 D patients was expressed as 24-h urinaryexcretion.Twenty-four healthy volunteers (mean age51.2  6.2 years) of both genders (three female and21 male) with no history of renal impairment, diabe-tes mellitus,or arterial hypertension participating in ablood donation program served as a control group. Prospective clinical study The study design was prospective,randomized,andcrossover. Inclusion criteria were age   18 years,maintenance extracorporeal renal replacementtherapy  3 months, hematocrit  > 30%, stable antico-agulation and erythropoietin regimen, no vascularaccess related problems, and the absence of malig-nant and ongoing inflammatory or infectious diseasesat the time of study enrollment. Exclusion criteriawere pregnancy, unstable clinical condition (e.g.,cardiac or vascular instability), estimated life expect-ancy  12 months, known coagulation problems, andparticipation in another study.The month before andduring the trial period and the patients’ concomitantmedications were continued in an unchanged mannerincluding anticoagulation under study conditions.The patients were randomly assigned to succes-sively receive each 4 weeks of hemodialysis treat-ment with LF-PSu (Fresenius F 8 HPS, 1.8 m 2 ),HF-PSu (Fresenius HF 80 S, 1.8 m 2 ), and PUR-H(PUREMA H, 1.7 m 2 ). Dialyses were performedthrice weekly, and each of the three periods startedwith a midweek session. Blood flow rate ( Q B ) andtreatment duration were set individually for eachpatient based upon the previous dialysis treatmentand were not changed during the study period. Thedialysate flow rate was set at 500 mL/min.The ultra-filtration flow rate of each session was set accordingto the individual patient’s interdialytic weight gain.Anticoagulation was performed by unchanged adop-tion of the previous routine heparinization.Blood samples were drawn from the arteriovenousfistula immediately after insertion of the arterialdialysis needle before the first session and aftercompletion of each 4-week study period. Sampleswere also obtained at the end of each first sessionfrom the arterial line of the extracorporeal circuitafter reduction of   Q B  to 50 mL/min for 120 s and thedialysate flow turned off. Postdialysis values werecorrected for hemoconcentration, and the reductionrate of BPA was calculated as described elsewhere(25). The respective dialysis doses for urea werequantified by means of eKt/V (26). Measurement of BPA BPA was measured in ethylenediaminetetraac-etate (EDTA) plasma with a commercially availableenzyme-linked immunosorbent assay (BPA ELISA“Eiken,” Eiken Chemical Co. Ltd., Tokyo, Japan)using the second antibody solid-phase competitivemethod (27). Protein binding of BPA In a subset of three maintenance dialysis patients,the fractions of protein-bound and free BPA weredetermined following a previous study on protein-bound uremic toxins (25).Total BPA was detected inan EDTA plasma sample. A second plasma samplewas filtered using a centrifugal filter unit (CentriprepYM 30, Millipore Corp., County Cork, Ireland) at1500 ×  g  for 30 min after prerinsing with saline toremove BPA from the filter housing. Unbound BPAwas determined in the plasma water filtrate. Thebound BPA fraction equaled total minus free BPA.Athird plasma sample was spiked with BPA to doublethe baseline concentration and incubated for 30 minat room temperature, and total and unbound BPAwere measured accordingly.The same procedure wasapplied on EDTA plasma from a healthy control towhich BPA was added in order to obtain approxi-mately twice the concentration observed in uremicplasma. Data analysis Descriptive analysis of the results was performedby calculating mean values  standard deviations ormedian and range.Within-subject between-treatmentdifferences were analyzed by analysis of variance anda Tukey post hoc test for normally distributedsamples and by Kruskal–Wallis test if normal distri-bution did not apply.Within-subject within-treatmentchanges from baseline were assessed using the two-sided paired  t  -test. For comparisons of patientsamples with the control group, a Mann–Whitney U  -test was used. Correlation coefficients were deter-mined for BPA, eGFR, residual renal function, andage according to Pearson and Spearman,respectively.A  P   value of   < 0.05 was considered as statisticallysignificant. The statistical analysis was performed bymeans of the Minitab 15 Statistical Software package(Minitab, Inc., State College, PA, USA). RESULTSBPA elution from dialyzers The mass of eluted BPA differed between the threedialyzers (Fig. 2). BPA in the blood compartment of PUR-H was only 4.4 and 12.9% ( P   <  0.001) of the BISPHENOL A IN CKD 3  Artif Organs,Vol. ••, No. ••, 2012  mass eluted from LF-PSu and HF-PSu, respectively.A significantly higher BPA amount was observedfor LF-PSu versus HF-PSu (140.8  38.7 vs. 48.1  7.7 ng;  P   <  0.01). The BPA masses detected in thedialysate compartments of the three dialyzersbehaved similarly (Fig. 2). No significant differencesbetween blood and dialysate compartments of thesame dialyzer were determined. Cross-sectional study In total, 152 prevalent patients with CKD wereanalyzed in the cross-sectional trial. The sample sizeof the different CKD stages varied considerably. Sixpatients with stage 1 (eGFR 121  14 mL/min),12 patients with stage 2 (eGFR 69  6 mL/min), 31patients with stage 3 (eGFR 44  7 mL/min), 40patients with stage 4 (eGFR 21  4 mL/min),10 patients with stage 5 (eGFR 10  2 mL/min), and53 patients with stage 5 D were enrolled. Thirty-sixpercent of the overall population were women (54female and 98 male). The mean age was 69.7  12.6years.The dialysis vintage of patients with CKD stage5 D was 84.0  97.1 months (median 52 months,range 8–419 months).Mean plasma BPA levels in patients with CKDstages 1 and 2 were almost exclusively below thedetection limit of 0.2 ng/mL and did not differ fromhealthy controls (Fig. 3). In CKD stages 3 and 4,BPA concentrations were similar (0.7  1.0 and0.7  0.7 ng/mL, respectively), and both were signifi-cantly elevated compared with the controls(  0.2  0.1 ng/mL;  P   <  0.01 and  P   <  0.05, respec-tively).CKD stage 5 patients presented a significantlyhigher BPA level (1.6  1.8 ng/mL;  P   <  0.05) com-pared with stages 1–4 and controls.In CKD stage 5 Dpatients on maintenance dialysis,BPA was very muchelevated in comparison with any other group(10.0  6.6 ng/mL;  P   <  0.001). No difference wasfound between dialysis patients on polysulfone orpolyethersulfone dialysis membranes (9.1  6.6 vs.11.9  6.0 ng/mL).The plasma BPA concentrations of controls andprevalent CKD patients not on renal replacementtherapy inversely correlated with eGFR ( r   = - 0.557, P   <  0.001) (Fig. 4). In maintenance dialysis patients, FIG. 2.  Eluted BPA per dialyzer ( n   =  6) separated for blood anddialysate compartment. Elution was performed by recirculation ofsterile water. The mass of eluted BPA differed significantlybetween the three dialyzers. LF-PSu, low-flux polysulfone,1.3 m 2 ; HF-PSu, high-flux polysulfone, 1.3 m 2 ; PUR-H, PUREMAH, 1.7 m 2 . * P   <  0.01 versus LF-PSu; ** P   <  0.001 versus LF-PSu; # P   <  0.01 versus HF-PSu;  ## P   <  0.001 versus HF-PSu. FIG. 3.  Plasma BPA of prevalent patients with CKD ( n   =  152)and healthy controls ( n   =  24). BPA levels started to rise in stages3 and 4. Plasma BPA concentrations of the patients on mainte-nance hemodialysis (stage 5 D) were more than six times highercompared with patients with CKD stage 5.  # P   <  0.01 versushealthy controls;  ## P   <  0.05 versus CKD stages 1–2 and controls;* P   <  0.05 versus CKD stages 1–4 and controls; ** P   <  0.001versus CKD stages 1–5 and controls. FIG. 4.  Association of plasma BPA with eGFR in healthy controlsand prevalent patients not on renal replacement therapy. Theplasma BPA concentrations inversely correlated with eGFR. D.H. KRIETER ET AL.4  Artif Organs,Vol. ••, No. ••, 2012  BPA also inversely correlated with 24-h urine excre-tion, but this association was not significant( r   = - 0.397, P   =  0.067) (Fig. 5).Despite a strong nega-tive association between eGFR and age ( r   = - 0.562, P   <  0.001),no correlation was noted for BPA and age( r   =  0.206,  P   =  0.151) or for BPA and dialysis vintage( r   = - 0.044,  P   =  0.778). Prospective study For the prospective study, 18 stable CKD stage5 D patients (mean age 60  14 years; three femaleand 15 male) on regular thrice-weekly maintenancehigh-flux hemodialysis with Polyamix dialyzers(Gambro Polyflux H series, Gambro DialysatorenGmbH, Hechingen, Germany) were enrolled.Underlying renal diseases were hypertensive neph-ropathy ( n  =  5), diabetic nephropathy ( n  =  3), auto-somal dominant polycystic kidney disease ( n  =  3),glomerulonephritis ( n  =  2), tubulointerstitial nephri-tis ( n  =  2), nephrectomy ( n  =  2), and postrenalfailure ( n  =  1). The mean dialysis vintage was46  29 months (range 11–114 months). The bodymass index averaged at 27.8  4.3 kg/m 2 . The mean24-h urinary excretion was 539  719 mL includingsix patients without residual renal function. Seven-teen patients had efficient native arteriovenousfistulas. One patient had a patent biflow dialysiscatheter for blood access. Medications includederythropoietin subcutaneously, intravenous iron,phosphate binders, oral vitamins, diuretics, beta-blockers, angiotensin-converting enzyme inhibitors,angiotensin receptor antagonists, digitalis, calciumchannel blockers, vasodilators, statins, and aspirin.Sixteen patients received standard heparin, and twopatients received fractionated heparin for antico-agulation during dialysis. For details of the delivereddialysis treatment, refer to Table 1.Baseline BPA concentrations before treatmentwith the three study dialyzers were not different (LF-PSu 12.0  6.0 ng/mL, HF-PSu 9.1  4.5 ng/mL,PUR-H 10.0  4.9 ng/mL). Furthermore, no signifi-cant change in plasma BPA levels was determinedduring a single hemodialysis session in any of thedialyzers (Fig. 6).However,reduction ratios were cal-culated being 12.3, 9.6, and 15.5% for LF-PSu,HF-PSu, and PUR-H, respectively. In addition, nochange of plasma BPA concentrations was notedafter 4 weeks of thrice-weekly hemodialysis with anyof the three filters (Fig. 7). Significant differences( P   <  0.001) were only found between healthy con-trols and dialysis patients at any given time pointduring each of the three 4-week periods (Fig. 7). Protein binding of BPA The fractions of protein-bound and free plasmaBPA in maintenance dialysis patients were 74  5and 26  5%, respectively (BPA spiked sample66  7 and 34  7%, respectively). The BPA-enriched healthy control demonstrated similar valuesof 70  3 and 30  3%, respectively. DISCUSSION The widespread use of BPA has caused an intenseexposure to this estrogenic endocrine-disrupting sub-stance and has led to a controversial debate aboutpotentially harmful consequences (11,12). After FIG. 5.  Association of plasma BPA with daily urinary excretion inprevalent patients with CKD stage 5 D. BPA inversely but notsignificantly correlated with 24-h urine excretion. Data on urinaryexcretion were available only for a limited number of patients inthis group. TABLE 1.  Characteristics of the delivered dialysis treatment. Data represent mean values   standard deviation LF-PSu HF-PSu PUR-HBlood flow rate (mL/min) 385  30 381  36 383  28Dialysate flow rate (mL/min) 500  0 500  0 500  0Ultrafiltration volume (L/session) 2.78  1.78 2.62  1.41 2.51  1.20Session duration (min) 268  15 268  15 268  15eKt/V 1.48  0.3 1.42  0.3 1.45  0.2 BISPHENOL A IN CKD 5  Artif Organs,Vol. ••, No. ••, 2012
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!