Skip Navigation


NDT Advance Access originally published online on June 7, 2007
Nephrology Dialysis Transplantation 2007 22(10):2970-2977; doi:10.1093/ndt/gfm352
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
22/10/2970    most recent
gfm352v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Churchwell, M. D.
Right arrow Articles by Mueller, B. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Churchwell, M. D.
Right arrow Articles by Mueller, B. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org


Trace element removal during in vitro and in vivo continuous haemodialysis

Mariann D. Churchwell1, Deborah A. Pasko2, Imad F. Btaiche2, Jinesh C. Jain3 and Bruce A. Mueller2

1Department of Pharmacy Practice, College of Pharmacy, University of Toledo, Toledo, OH, 2Department of Clinical Sciences, College of Pharmacy, University of Michigan, Ann Arbor, MI, USA,3Senior Scientist Hartford Waste Treatment and Immobilization Plant, Richland, WA, USA from the Renal Replacement Therapy Kinetics Study Group

Correspondence and offprint requests to: Mariann D. Churchwell, Department of Pharmacy Practice, University of Toledo College of Pharmacy, 2801 W. Bancroft MS 609, Toledo, OH 43606, USA Email: mariann.churchwell{at}utoledo.edu.



   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Background. Continuous renal replacement therapy (CRRT) increasingly is being used to treat critically ill patients with renal disease. CRRT removes waste products but also nutrients. Our understanding of trace element CRRT clearance has been limited by poor assay sensitivity. The development of inductively coupled plasma mass spectrometry (ICP–MS) allows for the measurement of CRRT trace element removal.

Methods. Continuous venovenous haemodialysis (CVVHD) transmembrane clearances of trace elements and urea were assessed using a bovine blood-based in vitro model using two different haemodialyser types. These findings were validated in 10 critically ill adult patients receiving continuous venovenous haemodiafiltration (CVVHDF). Calculated daily trace element loss was compared with a typical dose of daily trace element supplementation.

Results. The mean ± SD in vitro CVVHD transmembrane clearances (ml/min) for the polysulfone haemodialyser were chromium 0.97 ± 0.23, copper 0.47 ± 0.18, manganese 4.6 ± 3.6, selenium 1.2 ± 0.63 and zinc 2.3 ± 0.32 and for the cellulose diacetate haemodialyser chromium 1.54 ± 0.91, copper 0.21 ± 0.07, manganese 7.8 ± 4.1, selenium 0.76 ± 0.39 and zinc 2.7 ± 0.37. The in vivo CVVHDF transmembrane clearances (ml/min) were chromium 5.4 ± 2.4, copper 0.45 ± 0.33, manganese 1.9 ± 4.6, selenium 1.6 ± 1.2, and zinc 4.0 ± 1.3.

Conclusion. ICP–MS assays detected the five trace elements in the effluent of CVVHDF patients. Trace element CVVHD transmembrane clearance estimates for our in vitro model were supported by the in vivo CVVHDF findings. Calculated daily trace element loss attributed to CVVHD and CVVHDF with dialysate flow rates of 33.3 ml/min is less than what is provided in a daily dose of a trace element supplementation product.

Keywords: continuous dialysis; continuous renal replacement therapy; critically ill; haemodiafiltration; nutrition; trace elements



   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Acute renal failure (ARF) occurs in approximately 2–5% of hospitalized patients [1] and in 5–6% of critically ill patients [2]. Critically ill ARF patients are often treated with continuous renal replacement therapy (CRRT). Although CRRT removes waste products, it also removes beneficial medications and nutrients. Limited information is available on CRRT drug removal [3] but even less is known about nutrient clearance during CRRT, particularly the CRRT clearance of trace elements.

Available data on trace element abnormalities in renal failure patients have been generally derived from chronic renal failure patients. Extrapolation of these data to ARF patients is difficult especially for patients treated with CRRT. Information on optimal trace element requirements in patients receiving CRRT is scarce [4]. Bellomo [5] proposed that the recommended dietary allowances (RDA) of all trace elements be provided and Druml [6], in a CRRT nutrition review suggested providing trace element supplementation twice weekly. Currently, there is little evidence to support either suggestion. Story et al. [7] reviewed the potential loss of chromium, copper, manganese, selenium and zinc in critically ill patients receiving continuous venovenous haemofiltration (CVVH). Chromium and copper were found in the ultrafiltrate of CVVH patients. Critically ill patients had lower blood concentrations of selenium and zinc whether or not they were receiving CVVH when compared with normal healthy volunteers. Klein et al. [8] detected a small amount of zinc in the effluent of both their CVVH and continuous venovenous haemodialysis (CVVHD) patients. The authors concluded that zinc losses were not large enough to merit additional zinc supplementation in patients treated with CRRT who are receiving zinc in their parenteral nutrition. Berger et al. [9] reported that copper, selenium and zinc were found in the effluent during treatment with continuous venovenous haemodiafiltration (CVVHDF). Their most significant finding was that selenium was removed by CVVHDF at two times the daily supplementation.

Since Story et al. [7] published their findings, more sensitive trace element assay techniques have been developed. Inductively coupled plasma mass spectrometry (ICP–MS) can detect 90% of the periodic table at the sub-nanogram per gram level [10]. Utilizing this assay, we conducted an in vitro CVVH trace element clearance study [11]. Our findings reported measurable quantities of chromium, copper, manganese, selenium, and zinc in the ultrafiltrate of this in vitro CRRT model. Trace element loss was estimated using study-derived sieving coefficients (SC) and normal trace element blood concentrations. Our findings suggested that the removal rate of selenium by CVVH was greater than typical supplementation.

These findings led us to further investigate the potential in vivo loss of trace elements during other forms of CRRT since the removal characteristics of CVVHD or CVVHDF differ from CVVH. Further, we wished to determine whether our in vitro CRRT models were representative of trace element removal in critically ill patients receiving CRRT. We began this investigation by determining the transmembrane clearance of selected trace elements with our in vitro CVVHD model. We then followed up the in vitro data; by determining trace element transmembrane clearance in 10 critically ill patients treated with CVVHDF and compared these findings to our in vitro transmembrane clearance data.



   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
In vitro CVVHD
The trace element transmembrane clearance by common CRRT techniques was assessed using an in vitro bovine blood CRRT model developed in our laboratory [11–14]. The CVVHD model used 500 ml of pH regulated, continuously stirred bovine blood anticoagulated with 3.8% sodium citrate. This blood was placed in a 1 l Erlenmeyer flask which was submerged in a 37°C water bath.

Urea (Sigma, St. Louis, MO, USA lot #30K0221) was added to the bovine blood as a control solute to yield initial blood urea nitrogen (BUN) concentration of 75 mg/dl. B. Braun Diapact® CRRT machines and B.Braun Diapact CRRTTM tubing kits for the CRRT circuits were used with two different haemodialysers. A high permeability polysulfone [Optiflux F-160NR, Fresenius, (Lexington, MA) surface area 1.5 m2, ultrafiltration coefficient (Kuf) 45 ml/h/mmHg, lot # 3BU341] and a high efficiency cellulose diacetate hollow fibre [CA-HP 170, Baxter (Deerfield, IL, USA) SA 1.7 m2 Kuf 7.9 ml/h/mmHg, lot # E03J12]. After the extracorporeal circuit was primed, the blood was recirculated through the CRRT system for 20 min to allow for coating of the tubing and haemodialyser by blood proteins [15,16].

Dialysate was prepared by adding 240 ml of NormocarbTM sterile bicarbonate renal dialysis concentrate (Dialysis Solutions, Inc. Richmond Hills, Ontario, Canada) to a 3 l bag of sterile water (Abbott Laboratories, Chicago, IL, USA) according to the manufacturer's directions. Once reconstituted, Normocarb dialysate contains sodium 140 mEq/l, magnesium 1.5 mEq/l, chloride 106.5 mEq/l and bicarbonate 35 mEq/l. CVVHD was run in single-pass mode with a dialysate flow rate (Qd) of 33.3 ml/min. The ultrafiltrate production rate (Quf) was set at zero. Blood flow rate (Qb) was 200 ml/min (Table 1) and no adjustments in blood flow rate were necessary to maintain adequate pressures within the CRRT circuit. This process was repeated six times each with a new haemodialyser and tubing circuit for each haemodialyser membrane type.


View this table:
[in this window]
[in a new window]

  		Table 1. Trace element content of study liquids

 
In Vivo CVVHDF
Ten adult critically ill patients receiving CVVHDF at the University of Michigan Hospital, Ann Arbor, MI, USA were enrolled from March–September 2004. The study patients were identified using the daily CRRT roster maintained in the dialysis unit of University Hospital. The study investigators obtained informed consent and complied with University of Michigan standards including patient confidentiality according to the Health Information Portability and Accountability Act (HIPAA). The study was approved by the Institutional Review Board of the University of Michigan Medical School (IRBMED) (approval # 2004–0193)

Investigators played no role in determining who received CRRT or what nutritional regimens they received. At the University of Michigan, CRRT is administered as CVVHDF with a Qd of 33.3 ml/min and sufficient ultrafiltrate production to achieve or maintain euvolaemia. Typically, adult patients receive CVVHDF using the same high permeability polysulfone haemodialyser (Optiflux F-160NR) used in our in vitro trial. Physicians could choose a smaller alternative polysulfone haemodialyser (HF1000, Gambro, Lakewood, CO, USA) with a surface area of 1.1 m2 and a Kuf of 37 ml/h/mmHg.

Sample collection
All samples in the in vitro and in vivo studies were obtained in the same manner. All samples were obtained using trace metal-free stainless steel needles (Becton Dickinson Laboratories, Franklin Lakes, NJ, USA). Blood samples were drawn at the pre-haemodialyser port (A) and the effluent/spent dialysate (EF) and dialysate (D) samples from the effluent/spent dialysate and dialysate sampling ports of the CRRT circuits, respectively. One set of samples was obtained from each study patient or in vitro circuit. In vivo samples were placed on ice immediately and transferred to the laboratory at the University of Michigan, College of Pharmacy. Each effluent and dialysate sample was split into two separate Falcon® tubes (Becton Dickinson Labware, Franklin Lakes, NJ, USA) which are free of trace metals. Blood samples were transferred to Falcon® tubes and centrifuged at 3000 r.p.m for 30 min. The plasma was transferred by metal-free pipette to two Falcon® tubes and stored at –80°C until time of assay.

Assay
For the in vitro study, BUN was assayed on a COBAS Integra 400 plus (Roche Diagnostics, Indianapolis, IN, USA). The lower limit of detection was 1.8 mg/dl and the coefficient of variability was < 10%. Bovine and human plasma samples (from the in vitro and in vivo experiments, respectively) were analysed for chromium, copper, manganese, selenium and zinc content at the University of Notre Dame in South Bend, Indiana using ICP-MS. The analytical technique was performed according to previously described procedures [17–19] with slight modifications using a Vacuum Generators elemental PlasmaQuad model PQ2 ICP-MS.

Frozen plasma samples (5 ml) were thawed and transferred in a pre-cleaned, pre-weighed Teflon bomb and the sample weight was recorded for each sample. Each sample was transferred from a vial to a Teflon bomb by either pouring or with the aid of a clean plastic transfer pipette. Samples were then digested in open beakers using doubly distilled concentrated nitric acid (5 ml) and hydrogen peroxide (2 ml). After the first digestion was completed a second nitric acid digestion was performed. The second digestion used beakers closed at 100°C for 12 h, to ensure complete breakdown of the organic matrix. Two ml of HNO3 were added to the dried mass followed by 40 drops of H2O2 in a drop-wise method to avoid foaming. The analytes were taken to 15 ml in 2% nitric acid. Nanopure 18 M{Omega} cm water was used for sample dilutions. Monitoring of any contaminants from reagents and laboratory environment was done with a procedural blank included with each batch.

Single element solutions (Inorganic Ventures, Lakewood, NJ, USA) were used to prepare calibration and internal standard solutions. Analyses were performed using an external calibration procedure. Scandium (Sc), cobalt (Co), gallium (Ga) and yttrium (Y) were used as internal standards for matrix and instrument drift corrections. Procedural blank was analysed to check for any contribution from the re-agents and laboratory environment. The reference materials SRM-8414 (bovine muscle) and SRM-1577b (bovine liver) were digested and analysed to be identical to the unknown for quality control purposes. The assay was able to detect each trace element at the sub-ng/g level.



   Calculations
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Trace element loss was calculated by determining the extraction coefficient (EC) and transmembrane clearance of each element. Trace element and urea, in vitro transmembrane clearance equations for continuous haemodialysis were as follows:


Formula

Trace element transmembrane clearance, in vivo involved ultrafiltrate production so the following equation was used:

Continuous haemodiafiltration


Formula

A = solute concentration in plasma obtained from the pre-haemodialyser port;

EF = solute concentration in effluent/spent dialysate obtained from the effluent/spent dialysate port;

Quf = ultrafiltrate production rate;

Qd = dialysate flow rate;

Trace element loss (mg/day) = EC x mean trace element plasma concentration x total dialysate volume/day.

Trace element transmembrane clearance was calculated using the above equations. This transmembrane clearance data were used to calculate the projected daily loss of each trace element. The transmembrane clearance (ml/min) multiplied by the arterial plasma concentration (ng/g) of each trace element by 24 h gave us a calculated CRRT daily loss of each trace element. This CRRT loss for each trace element was compared with the amount supplied by a daily dose of 1 ml a common parenteral trace element supplement MultitraceTM-5 (MTE-5) Concentrate (American Regent Laboratories, Inc. Shirley, NY, USA) (Table 2). The calculated daily CRRT loss of each trace element was subtracted from the amount provided for supplementation by the MTE-5 Concentrate to determine if loss from CVVHD or CVVHDF was greater than current supplementation provided by MTE-5 Concentrate.


View this table:
[in this window]
[in a new window]

  		Table 2. Trace element results of in vitro continuous venovenous haemodialysis (CVVHD) experiments

 


   Sample size and data analysis
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Power analysis for the in vitro experiments indicated that a sample size of six haemodialysers would detect a 25% difference in trace element transmembrane clearance between haemodialyser types assuming a 10% standard deviation with a power of 0.9 and a significance level of P < 0.05. The mean EC (with corresponding transmembrane clearance values) between the two types of haemodialysers for each trace element were compared using ANOVA. Power analysis for the in vivo portion of the study was not conducted, but sample size was consistent with previous CRRT trace element clearance studies [7–9].



   Results
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
Discussion
 Conclusion
 Acknowledgements
 References
 
Trace element content of the liquids (in vivo dialysate, normal saline, distilled water, Normocarb in vitro dialysate and tap water as a control) used in the in vitro experiment was assessed to determine a possible source of additional trace elements that could confound our findings (Table 1). The custom dialysate used in patients is made by a commercial vendor for the hospital. The in vivo dialysate had the highest concentration of zinc (43.1 ng/g) with much lower concentrations of the other elements present. The Normocarb dialysate used in the in vitro experiments had relatively higher concentrations of zinc (38.9 ng/g) than the other trace elements. Overall the trace element profiles were similar between the in vitro Normocarb dialysate and in vivo hospital dialysate (Table 1).

In vitro CVVHD model
All experiments were conducted as outlined in the methods and there were no circuit failures during the in vitro continuous haemodialysis experiments. The mean ± SD urea EC for the cellulose diacetate haemodialyser was 0.87 ± 0.21 and for the polysulfone haemodialyser 0.98 ± 0.07. Table 2 contains the EC data for each trace element. The mean in vitro EC from the cellulose diacetate haemodialyser tended to be higher for chromium, manganese and zinc than with the polysulfone haemodialyser but these differences were not statistically significant. The polysulfone haemodialyser had higher EC for copper and selenium but only copper's EC difference was statistically significant from the cellulose diacetate haemodialyser (P < 0.02).

Based on the mean EC derived from these in vitro data, we predicted trace element transmembrane clearance for each haemodialyser using the experimental conditions (dialysate flow rate of 33.3 ml/min and no ultrafiltrate production). The amount of predicted daily trace element CRRT loss, calculated from the in vitro mean plasma trace element blood concentration and in vitro CRRT transmembrane clearance values appears in Table 2. The amount of daily trace element supplementation provided by a common, commercially available trace element product (MTE-5, components shown in Table 2) was subtracted from the calculated daily trace element loss to determine if CVVHD at 33.3 ml/min would result in a net daily loss of any trace element even if conventional trace element supplementation is given.

In vivo CVVHDF study
Adults who received CVVHDF were a diverse group. Large differences in duration of CVVHDF, patient weight, age and type of nutrition received were apparent (Table 3). The mean ± SD Qb was 173 ± 25 ml/min. The mean ± SD Qd was 34.2 ± 2.7 ml/min. The mean ultrafiltrate production rate in these patients was 7.65 ± 3.05 ml/min to achieve or maintain euvolaemia. All patients received anticoagulation with sodium citrate according to the University of Michigan protocol [20]. Nine patients received CVVHDF using the Optiflux F160 NR hemodialyser and one patient (patient 5) received treatment with a Prisma HF1000 hemodialyser.


View this table:
[in this window]
[in a new window]

  		Table 3. CVVHDF patient demographics

 
The arterial plasma concentrations of the five trace elements varied widely in our patients (Table 4). Copper and zinc showed the most inter-patient variability. The in vivo EC and trace element CRRT transmembrane clearance data appear in Table 5. Daily CRRT loss calculations indicate that very little copper, manganese, selenium and zinc were removed by CVVHDF. Indeed, manganese was not detectable in the effluent of 8/10 patients. Selenium was not detectable in the effluent of 2/10 patients, consequently the calculated CVVHDF transmembrane clearance of manganese and selenium was considered to be zero in these patients. Substantially more chromium was removed than other trace elements; however the daily CVVHDF loss was less than half of what would be supplemented with a daily dose of MTE-5 (Table 5).


View this table:
[in this window]
[in a new window]

  		Table 4. Patient trace element (TE) plasma concentrations compared with in vitro experiment bovine plasma values

 

View this table:
[in this window]
[in a new window]

  		Table 5. In vivo experiments: trace element continuous venovenous haemodiafiltration (CVVHDF) calculations

 


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
Conclusion
 Acknowledgements
 References
 
We were able to quantify the transmembrane clearance of five trace elements in critically ill patients receiving CVVHDF. The emergence of ICP-MS assay technology has allowed for the measurement of trace elements to the sub-ng/g level. Most of the previous studies of trace element disposition in patients receiving CRRT were conducted with less sensitive assays [21,22]. A later study by Berger used ICP-MS, but these authors did not report manganese and chromium values [23]. These more sensitive assays allow us to better characterize trace element disposition in CRRT.

Our study patients came from many different intensive care units in our hospital and consequently were a heterogeneous population. Plasma serum concentrations of trace elements likely differ between different types of critically ill patients and otherwise healthy individuals. Our one-time sampling strategy provides a snapshot of the trace element transmembrane clearance in these ten patients, many of whom were receiving CVVHDF for more than a week. Our methodology used a single plasma sample, which, for the purposes of calculating daily losses, we assumed was representative of the plasma concentration for an entire day. However, it is likely that critically ill patients have variability in their trace element plasma concentrations depending on underlying illness and state of inflammation or infection.

The differences in patient characteristics may partially explain the wide variety in observed trace element profiles (Table 4). For instance, study patient 9 had the lowest plasma copper, selenium and zinc concentrations. This individual received CRRT for 6 months to treat ARF resulting from trauma and severe burns from a motor vehicle accident. This prolonged hospitalization likely influenced his trace element profile. Other study patients with disparate trace element concentrations include study patient 8, who was post-liver transplant from hepatitis and had received one day of CRRT at the time of sampling. This enterally fed individual had the lowest manganese and chromium plasma concentrations. Study patient 6, a 43-year-old male with congestive heart failure and cardiomyopathy awaiting a heart transplant, had the highest copper value at twice the population mean. This patient also had the shortest duration (14 h) of CVVHDF at the time of sampling. Lastly, study patient 5 was an 18-year-old female with a recent bone marrow transplant and acute renal failure attributed to sepsis. This patient had an elevated serum zinc concentration at approximately twice the population mean and was receiving daily trace element supplementation with MTE-5.

Comparison of in vitro and patient-derived trace element transmembrane clearance values in the present study indicates agreement between in vitro polysulfone and in vivo polysulfone EC and transmembrane clearance (Tables 2 and 5). The trace elements chromium and zinc exhibited a statistically significant difference in transmembrane clearance between in vitro and in vivo investigations (P < 0.01). The larger in vivo transmembrane clearances might be partially explained by the fact that the in vitro study utilized only 33.3 ml/min of dialysate flow, while the CVVHDF patients received 33.3 ml/min of dialysate flow plus an additional 7.7 ± 3.1 ml/min of ultrafiltrate production providing additional convective transmembrane clearance. Although statistically significant, the amount of chromium and zinc removed by CVVHDF was substantially below what is typically supplemented. These differences may also be explained by variations in protein binding and tissue distribution in study patients compared with the in vitro setting. In vitro experiments were run with blood obtained from healthy cows. Bovine blood could differ from critically ill human blood in terms of trace element concentrations and to the extent that these trace elements bind to plasma proteins (and consequently, how available they are to be removed by CRRT). Inflammation can influence the expression of trace element containing proteins in the plasma as well as free trace element concentrations. Consequently, the fact that our patients were critically ill but the bovine blood came from otherwise healthy cattle, may account for some differences between the in vivo and in vitro studies. Table 4 lists the mean bovine plasma trace element concentration values and the plasma trace element values from each of the patients in the in vivo part of the study. We observed a wide range of trace element concentrations in critically ill patients. Nonetheless, there are substantial trace element differences between the critically ill human and the bovine plasma obtained from otherwise healthy cattle. The extent of trace element protein binding can be inferred from EC values [24]. EC is an indirect measurement of plasma protein binding. The in vitro polysulfone experiment EC values were compared with the EC values obtained in patients who also were treated with polysulfone haemodialysers. The EC values for chromium and zinc differed significantly between bovine and human values. Calculations from our bovine blood-based in vitro study suggested that only small amounts of trace elements would be removed by CVVHD and our in vivo study corroborated those projections. However, there were large differences between trace element plasma concentrations and EC values limiting the predictive value of the in vitro bovine blood-based model.

The in vitro portion of the study allowed us to compare the performance of two common CRRT haemodialysers F-160NR (polysulfone) and the CA-HP 170 (diacetate). The surface area between these two haemodialysers is similar but the ultrafiltration coefficient differences indicate the polysulfone haemodialyser is more permeable. For small solutes like urea, there are unlikely to be differences in EC. Indeed, the urea EC approached one for both haemodialyser types and is consistent with previously published data [21,22]. CRRT transmembrane clearance differences between haemodialysers are commonly observed in most drug studies [21,23], as most drugs have larger molecular weights than urea. However, the present study did not find many trace element EC differences between haemodialyser types. This is likely due to the fact that trace elements are highly protein bound and were minimally removed by either haemodialyser type. Typically, if large transmembrane clearance differences are observed between CRRT haemodialyser types, they are observed with larger molecular weight substances [14,21].

Our in vitro calculations were substantiated by the in vivo study, namely that supplementation with standard doses of MTE-5 results in more than enough trace elements to overcome the amount lost by CVVHD. However, there were limitations to this study. For the in vivo investigation of this study, we did not quantify the loss of trace elements in the faeces, urine or wounds of our patients. Table 3 indicates that nine of the 10 study patients were receiving nutrition, but not all study patients were receiving the same standardized nutrition. Two patients received parenteral nutrition and the remaining seven patients were fed a variety of enteral nutrition formulations during CRRT therapies. Another potential limitation is that our in vitro study employed CVVHD while our study subjects received CVVHDF. This additional convective clearance could influence EC. Nonetheless, the agreement between the reported patient derived and in vitro transmembrane clearance demonstrate that this in vitro bovine blood model can be used to generate initial CRRT transmembrane clearance estimates.

While we did not directly measure non-CRRT trace element losses, it appears that MTE-5 standard doses are sufficient to replace CVVHDF losses. However, whether the MTE-5 daily dose is the optimal dose has been recently called into question. Angstwurm et al. [25] recently reported that higher doses of selenium resulted in improved mortality rates in critically ill patients with septic shock.



   Conclusion
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
Acknowledgements
 References
 
Measuring five trace elements with a very sensitive ICP–MS assay, our in vitro, bovine blood-based CVVHD model demonstrated that these trace elements could be cleared by CVVHD with high or low permeability haemodialysers. Despite this transmembrane clearance, we calculate that standard trace element supplementation such as MTE-5 that is used in parenteral nutrition would more than replace the trace element CVVHD losses. Our follow-up clinical study conducted in 10 critically ill patients receiving CVVHDF validated this finding. Daily trace element supplementation exceeds trace element losses attributed to CVVHDF with a dialysate flow rate of 33.3 ml/min.



   Acknowledgements
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
References
 
This study was funded in part by a grant from the University of Michigan College of Pharmacy Clinical Research Resources Fund. The authors wish to acknowledge the contributions of Jason Blythe and Candace Chu for assistance with the in vitro dialysis portion of the study. The in vitro investigation was performed in the Mueller Laboratory at the University of Michigan, College of Pharmacy, Ann Arbor, MI, USA. The in vivo investigation was conducted at the University of Michigan Health System‘s University Hospital, Ann Arbor Michigan. Trace element analysis was conducted at the University of Notre Dame, Department of Civil Engineering and Geological Sciences, South Bend, IN, USA. Data analysis and manuscript preparation was carried out at the University of Michigan and University of Toledo, Toledo, OH, USA.

Conflict of interest statement. None declared.



   References
Top
 Abstract
 Introduction
 Materials and methods
 Calculations
 Sample size and data...
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 

  1. Shusterman N, Strom BL, Murray TG, Morrison G, West SL, Maislin G. Risk factors and outcome of hospital-acquired acute renal failure. Clinical epidemiologic study. Am J Med (1987) 83:65–71.[Web of Science][Medline]
  2. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients. JAMA (2005) 294:813–818.[Abstract/Free Full Text]
  3. Mueller BA, Pasko DA, Sowinski KM. Higher renal replacement therapy dose delivery influences on drug therapy. Artif Organs (2003) 27:808–814.[CrossRef][Web of Science][Medline]
  4. Frankenfield DC, Reynolds HN. Nutritional effect of continuous hemodiafiltration. Nutrition (1995) 11:38893.
  5. Bellomo R. How to feed patients will renal dysfunction? Blood Purif (2002) 20:296–303.[CrossRef][Web of Science][Medline]
  6. Druml W. CRRT in the critically ill: Metabolic aspects of continuous renal replacement therapies. Kidney Int (1999) 56([Suppl 72]):S56–S61.[CrossRef][Web of Science]
  7. Story DA, Ronco C, Bellomo R. Trace element and vitamin concentrations and losses in critically ill patients treated with continuous venovenous hemofiltration. Crit Care Med (1999) 27:220–223.[CrossRef][Web of Science][Medline]
  8. Klein CJ, Moser-Veillon PB, Schweitzer A, et al. Magnesium, calcium, zinc, and nitrogen loss in trauma patients during continuous renal replacement therapy. J Parenter Enteral Nutr (2002) 26:77–93.[Abstract/Free Full Text]
  9. Berger MM, Shenkin A, Revelly JP, Roberts E, Cayeux MC. Copper, selenium, zinc, and thiamine balances during continuous venovenous hemodiafiltration in critically ill patients. Am J Clin Nutr (2004) 80:410–416.[Abstract/Free Full Text]
  10. University of Notre Dame ICP-MS Analytical Research Facility Department Civil Eng. & Geological Sciences, University of Notre Dame, Notre Dame, IN 46556–0767 www.nd.edu~icpmslab.
  11. Nakamura AT, Btaiche IF, Pasko DA, et al. In-vitro clearance of trace elements via continuous venovenous hemofiltration. J Ren Nutr (2004) 14:214–219.[CrossRef][Web of Science][Medline]
  12. Pasko DA, Mottes TA, Mueller BA. Pre dialysis of blood prime in continuous haemodialysis normalizes pH and electrolytes. Ped Nephrol (2003) 18:1177–1183.[CrossRef][Web of Science][Medline]
  13. Pasko DA, Churchwell MD, Mueller BA. Development of an in vitro model of continuous renal replacement therapy (CRRT) solute removal. Blood Purif (2004) 22:229.[CrossRef]
  14. Churchwell MD, Pasko DA, Mueller BA. Daptomycin Clearance during Modeled Continuous Renal Replacement Therapy. Blood Purif (2006) 24:548–554.[Web of Science][Medline]
  15. Clark WR, Macias WL, Molitoris BA, Wang NH. Membrane adsorption. Kidney Int (1994) 46:1140–1146.[Web of Science][Medline]
  16. Clark WR, Macias WL, Molitoris BA, Wang NH. Plasma protein adsorption. Kidney Int (1995) 48:481–488.[Web of Science][Medline]
  17. Jain JC, Neal CR, Seidler JA. Assessing the accuracy of analysis for low level elemental composition of blood by ICP-MS. (2000) Proceedings of the 46th International Conference on Analytical Sciences and Spectroscopy. 26 (abstract).
  18. Jain JC, Neal CR, Seidler JA. Trace elements analysis of human blood samples by ICP-MS. (2000) Proc Pittsburg Conference 188ZP. (abstract).
  19. Mishra S, Sabbah HN, Jain JC. Reduced Ca2+-calmodulin-dependent protein kinase activity and expression in IV myocardium of dogs with heart failure. Am J Physiol Heart Circ Physiol (2003) 248:876–883.
  20. Swartz R, Pasko D, O’Toole J, Starmann B. Improving the delivery of continuous renal replacement therapy using regional citrate anticoagulation. Clin Nephrol (2004) 61:134–143.[Web of Science][Medline]
  21. Brunet S, Leblanc M, Geadah D, Parent D, Courteau S, Cardinal J. Diffusive and convective solute clearances during continuous renal replacement therapy at various dialysate and ultrafiltration flow rates. Am J Kidney Dis (1999) 34:886–892.
  22. Troyanov S, Cardinal J, Geadah D, et al. Solute clearances during continuous venovenous haemofiltration at various ultrafiltration flow rates using Multiflow-100 and HF1000 filters. Nephrol Dial Transplant (2003) 18:961–966.[Abstract/Free Full Text]
  23. Joy MS, Matzke GR, Frye RF, Palevsky PM. Determinants of vancomycin clearance by continuous venovenous hemofiltration and continuous venovenous hemodialysis. Am J Kidney Dis (1998) 31:1019–1027.[Web of Science][Medline]
  24. Rumpf KW, Rieger J, Doht B, Asorg R, Scheler F. Drug elimination by hemofiltration. J Dial (1977) 1:67–68.[Web of Science]
  25. Angstwurm MW, Engelmann L, Zimmermann T, et al. Selenium in Intensive Care (SIC): results of a prospective randomized, placebo-controlled, multiple-center study in patients with severe systemic inflammatory response syndrome, sepsis, and septic shock. Crit Care Med (2007) 35:118–126.[CrossRef][Web of Science][Medline]
Received for publication: 22.12.06
Accepted in revised form: 8. 5.07


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nephrol Dial TransplantHome page
M. D. Churchwell, D. A. Pasko, W. E. Smoyer, and B. A. Mueller
Enhanced clearance of highly protein-bound drugs by albumin-supplemented dialysate during modeled continuous hemodialysis
Nephrol. Dial. Transplant., January 1, 2009; 24(1): 231 - 238.
[Abstract] [Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
C. J. Diskin
An understanding of trace metal physiology as an aid to interpretation of clearance studies by artificial organs
Nephrol. Dial. Transplant., April 1, 2008; 23(4): 1462 - 1462.
[Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
M. D. Churchwell, I. F. Btaiche, D. A. Pasko, and B. A. Mueller
Reply
Nephrol. Dial. Transplant., April 1, 2008; 23(4): 1462 - 1463.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
22/10/2970    most recent
gfm352v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Churchwell, M. D.
Right arrow Articles by Mueller, B. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Churchwell, M. D.
Right arrow Articles by Mueller, B. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?