Elimination of teicoplanin by adsorption to the filter membrane during haemodiafiltration: screening experiments for linezolid, teicoplanin and vancomycin followed by in vitro haemodiafiltration models for teicoplanin.
It has been assumed that free drugs (unbound to albumin) are eliminated during HDF by diffusion and/or ultrafiltration3,4. The effects of hypoalbuminemia on the pharmacokinetics and pharmacodynamics of highly protein-bound antibiotics were reported and reviewed5. However, the various anti-MRSA agents have different affinities for plasma albumin. For instance, teicoplanin shows the highest degree of binding to plasma albumin (~90%) compared with other drugs in this class (6), with concomitant implications for clearance. Thus, the effects of HDF or CHDF on the pharmacokinetics of teicoplanin are complicated by the presence or absence of albumin.
Accumulating evidence indicates that some drugs may be eliminated during HDF or CHDF by adsorption onto filter membranes (7-9), as well as by diffusion and ultrafiltration. The anti-MRSA agents are among these drugs. The issue of antibiotic adsorption onto haemofilters is a largely neglected but important area of research; to this point, the in vitro interaction of drugs and albumin alike with filter membranes has been reported (10). Thus the effects of HDF or CHDF on the pharmacokinetic parameters of anti-MRSA drugs are suspected to depend partially on the filter media and partially on the presence of plasma albumin. The aim of the present study was therefore to investigate the affinity of anti-MRSA agents for different kinds of haemofilters and the pharmacokinetic parameters of teicoplanin during HDF. The affinities of linezolid, teicoplanin and vancomycin for various filter membranes were first compared in a simple screening experiment. Next, an in vitro HDF circulation model was employed to investigate the effects of the filter material and the presence/absence of plasma albumin on teicoplanin elimination via adsorption onto filter membranes. This model was chosen given the considerable technical and ethical challenges associated with the measurement of adsorption in the clinical setting and/or with in vivo animal experiments.
MATERIALS AND METHODS
Screening for an anti-MRSA agent with high affinity for filter membranes
The affinity of linezolid, teicoplanin and vancomycin was compared for three different kinds of haemofilters: polyacrylonitrile (PAN) (APF-06S, Asahi Kasei Kuraray Medical Co., Ltd, Tokyo, Japan), polysulfone (PS) (AEF-07, Asahi Kasei Kuraray Medical Co. Ltd, Tokyo, Japan) and polymethylmethacrylate (PMMA) (CH-0.6N, Toray Medical, Tokyo, Japan). Since albumin was expected to attenuate the adsorption of anti-MRSA agents and since small changes in concentration of the agent are difficult to detect, we applied free anti-MRSA agents in this experiment. As shown in Figure 1, the anti-MRSA agents were dissolved in 50 ml of Krebs-Ringer-Bicarbonate (KRB) solution in an Erlenmeyer flask. The pH of the KRB solution was adjusted to 7.4 and maintained with a 5% carbon dioxide gas mixture throughout the course of the experiment. Taking into consideration the maximum anti-MRSA agent concentration in the plasma (Cmax) that results from typical clinical dosages, the initial concentrations of linezolid, teicoplanin and vancomycin were set at 20, 50 and 50 [micro]g/ml, respectively. The filter membranes were first primed with KRB solution at an appropriate transmembrane pressure to obtain sufficient filtration fluid. The membranes were then cut into 5 [mm.sup.2] pieces. The membrane pieces were added to the KRB solutions containing each anti-MRSA agent, and the flasks containing the anti-MRSA agent/KRB solution and the membrane pieces were incubated for 60 minutes in a 37[degrees]C water bath. The membrane pieces were eliminated to provide blank controls for each drug. Next, the concentrations of each anti-MRSA agent before and after the incubation were determined in collected KRB/anti-MRSA samples using a high performance liquid chromatography (HPLC) system (described in detail below). The adsorption-dependent elimination rate of each drug was calculated using the following equation:
Adsorption-dependent elimination rate (%/hour)
= [[[C.sub.0] - [C.sub.60]]/[C.sub.60]] x 100
where [C.sub.0] is the concentration of the drug at 0 minutes, and [C.sub.60] is the concentration of the drug at 60 minutes.
Because the volume (50 ml) of the anti-MRSA agent/KRB solutions in this experiment represented 1/240 of the volume of the total extracellular fluid (i.e. 12 litres in the adult patients), the haemofilter surface area was adjusted to 1/240 of the filter column area (0.6 or 0.7 m2) used clinically in Japan.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
In vitro HDF circulation model
The in vitro HDF model consisted of a beaker one litre in size containing KRB, with or without the addition of human albumin (Figure 2). HDF was performed using an ACH-10 dialysis/ultrafiltration system (Asahi Kasei Kuraray Medical Co. Ltd, Tokyo, Japan). The albumin concentration was adjusted to 3 g/dl using 25% human serum albumin (CSL Behring, Japan). KRB solution containing teicoplanin at an initial concentration of 50 [micro]g/ml was circulated at a flow rate of 100 ml/minute through each of three filter membranes, i.e., PS, PAN and PMMA, and circuits (CHD-400N, Asahi Kasei Kuraray Medical Co. Ltd, Tokyo, Japan). This flow rate was identical to the practical rate that is covered by Japanese social health insurance. The pH of the KRB solution was adjusted to 7.4 and then continuously equilibrated with a 5% carbon dioxide gas mixture. KRB (200 ml) was discarded to wash the circuit. The flow rate of the dialysate was 500 ml/hour and the flow rates of replacement fluid and ultrafiltrate were equal and set at 500 ml/hour; these rates were again identical to the common and practical rates that are covered by Japanese social health insurance. After the KRB/teicoplanin solution was added to the beaker, a five-minute period was allowed for priming of the system at a flow rate of 100 ml/minute. Following priming, the in vitro dialysis and ultrafiltration was initiated (time point=0 minutes).
The KRB/teicoplanin samples for assay were taken at the inlet and outlet of the haemofilter con currently with the filtrate samples at time points of 0, 15, 30, 45, 60, 90 and 120 minutes. KRB/teicoplanin samples were also collected before priming (baseline). The samples were stored at -80[degrees]C and the concentrations of teicoplanin were later determined by HPLC. The adsorption-dependent elimination ratio (the amount of teicoplanin adsorbed/total amount of teicoplanin in the KRB solution at baseline) at 120 minutes was obtained according to the following equation:
Adsorption-dependent elimination ratio at 120 minutes
= [[[C.sub.BL] x 1.2 - [C.sub.200ml] x 0.2 - [C.sub.F Total] x [V.sub.F Total] - [C.sub.120]]/[[C.sub.BL] x 1.2]] x 100 (%)
where [C.sub.BL] is the concentration of teicoplanin at baseline, [C.sub.200 ml] is the concentration of teicoplanin in the discarded 200 ml of KRB, [C.sub.F Total] is the concentration of teicoplanin in total filtrate at 120 minutes, [V.sub.F Total], is the volume of total filtrate at 120 minutes, [C.sub.120] is the concentration of teicoplanin at the inlet of the filter membranes at 120 minutes and V is the total volume of KRB.
The HDF clearance ([CL.sub.HDF]), which represents drug elimination by adsorption as well as by diffusion and ultrafiltration, was calculated by the following equation:
[CL.sub.HDF] = [k.sub.e] x [V.sub.d] (ml/minute) where k is the elimination rate constant, and [V.sub.d] is the volume of distribution for teicoplanin, i.e. the volume of KRB used in the model system (=1000 ml). The [k.sub.e] was estimated from the initial slope of the concentration versus time curve in the semi-logarithmic plot.
Simple HDF circulation model without albumin, dialysis or ultrafiltration
To confirm the adsorption of teicoplanin onto the membrane filters and to eliminate the effect of albumin, dialysis and ultrafiltration, a simple circulation experiment was next performed. The simple circulation model consisted of a one litre beaker containing KRB without human albumin. KRB solution containing teicoplanin at an initial concentration of 50 g/ml was circulated at a flow rate of 100 ml/minute through each of the three filter membranes, without dialysis or ultrafiltration. As noted above, the flow rate of 100 ml/minute was identical to the rate that is covered by Japanese social health insurance.
The KRB/teicoplanin samples for assay were simultaneously collected at the inlet and outlet of the filter membranes at time points of 0, 15, 30, 45, 60, 90 and 120 minutes. KRB/teicoplanin samples were also collected before equilibration was reached. The samples were stored at -80[degrees]C and the concentrations of teicoplanin were later determined by HPLC. To evaluate the extent of teicoplanin absorption onto the circuit, similar experiments were performed in which the KRB solution containing teicoplanin was circulated through the blank HDF circuit, excluding the filter membranes.
Determination of the concentrations of anti-MRSA agents using HPLC
The concentrations of linezolid, teicoplanin and vancomycin were determined in the various samples collected as described above, using the HPLC method. The HPLC system was composed of an LC-10AD pump, a Shim-pack CLC-ODS ([C.sub.18], 150x6.0 mm) column, an SIL-10A auto injector, a CTO-10AC column oven, an SPD-6A UV spectrometric detector and a C-R8A chromatopac integrator. All components of the HPLC system were from Shimadzu Corporation (Kyoto, Japan).
The concentration of linezolid was determined by a modified HPLC technique (11). Briefly, during the mobile phase, a mixture of acetonitrile and 50 mM sodium acetate buffer (25:75, v/v) adjusted to pH 4.0 was pumped through the column at a flow rate of 1.0 ml/minute. The UV absorbance of the eluent was monitored at a wavelength of 253 nm. The temperature of the column was maintained at 40[degrees]C.
The concentration of teicoplanin was measured by HPLC, with slight modifications to the methods previously described (12). The mobile phase consisted of an acetonitrile/sodium dihydrogen phosphate (50 mM) aqueous solution (28:72, v/v) pumped through the column at a flow rate of 1.5 ml/minute. Teicoplanin was detected at a wavelength of 218 nm. The temperature of the column was maintained at 40[degrees]C.
Finally, the concentration of vancomycin was determined by HPLC, using a modified method of Luksa et al (13). The mobile phase was prepared by premixing acetonitrile and 50 mM sodium dihydrogen phosphate buffer (pH 2.5) in a 10:90 (v/v) ratio. The mixture was pumped through the column at a flow rate of 1 ml/minute. Vancomycin was detected at a wavelength of 230 nm and the temperature of the column was maintained at 40[degrees]C.
Data are expressed as the mean [+ or -] SE. One-way or two-way analysis of variance (ANOVA) was applied for non-repeated measurements. Repeated measures ANOVA followed by the Tukey-Kramer test was applied for repeated or serial determinations. A P value of <0.05 was considered statistically significant.
Screening for an anti-MRSA agent with high filter membrane-binding activity
Figure 3 shows the adsorption-dependent elimination rates associated with the adsorption of each anti-MRSA agent onto the three different filter membranes in the screening experiment. When the adsorption-dependent elimination rates were compared with the similar values calculated in the absence of filter membrane (blank control), significant adsorption of teicoplanin onto PS and PMMA membranes, but not PAN membranes, was observed. Linezolid and vancomycin were not significantly adsorbed onto any filter membrane. Therefore, the in vitro HDF experiments focused on teicoplanin.
In vitro HDF model using teicoplanin with or without albumin
To confirm the adsorption of teicoplanin onto the filter membrane under the conditions of HDF, an in vitro HDF circulation model using teicoplanin was constructed (Figure 2). In this model, both with (Figure 4: Panel B) and without (Figure 4: Panel A) albumin, the concentration of teicoplanin in the circulation fluid decreased with time for each filter membrane employed. However, the extent of the decrease was significantly different among the three membranes. The extent of the decline was largest for the PMMA membrane and smallest for the PAN membrane (two-way repeated measures ANOVA followed by the Tukey-Kramer test, P <0.05).
[FIGURE 3 OMITTED]
The HDF clearance (CLHDF) for teicoplanin was calculated from the circulation fluid volume and the slope of the concentration versus time curve, as described in the Materials and Methods section. The [CL.sub.HDF] with and without albumin are summarised in Table 1 for each kind of filter membrane. When the data were analysed by two-way ANOVA, the [CL.sub.HDF] was significantly affected by both the filter membrane (P <0.01) and the presence of albumin (P <0.01). In addition, there was a significant interaction between the membrane and albumin (P <0.01). The [CL.sub.HDF] was largest for the experimental condition that employed the PMMA membrane in the absence of albumin and smallest for the PAN membrane in the presence or absence of albumin (Tukey-Kramer test, P <0.01). Addition of albumin into the KRB solution significantly decreased the CLHDF for all the filter membranes (Tukey-Kramer test, P <0.05).
[FIGURE 4 OMITTED]
As shown in Table 1, independent of the existence of albumin, the adsorption-dependent eliminations of teicoplanin by the PS and PMMA filter membranes were significantly higher than the elimination by the PAN membrane. The PMMA membrane displayed the highest binding capacity for teicoplanin, while the PAN membrane had a negligible binding capacity. As for the CLHDF, the adsorption-dependent elimination of teicoplanin was significantly influenced by both the filter membrane (P <0.01) and the presence of albumin (P <0.01). There was a significant interaction between the membrane and albumin (P <0.01). Addition of albumin into the KRB solution slightly but significantly decreased the adsorption-dependent eliminations of teicoplanin by PS and PMMA membranes (Tukey-Kramer test, P <0.05).
Simple circulation model without albumin, dialysis or ultrafiltration
To confirm the adsorption of teicoplanin onto membrane filters and eliminate the effect of albumin, dialysis and ultrafiltration, a simple circulation HDF experiment was performed in the absence of human albumin. A blank control experiment was performed that excluded the filter membrane so as to evaluate the extent of teicoplanin absorption onto the circuit. Figure 5 shows the time course of the changes in the teicoplanin concentration in the KRB solution during a simple circulation. In the control experiment, the concentration of teicoplanin was unchanged throughout the experiment, indicating that the adsorption of teicoplanin onto the circuit was minimal. Compared with the control experiment, circulation through each of the three filter membranes significantly decreased the teicoplanin concentration over time (repeated measures ANOVA followed by the Tukey-Kramer test, P <0.01). The extent of the decrease was largest for the PMMA membrane and smallest for the PAN membrane. Thus, the adsorption of teicoplanin onto the filter membranes was confirmed in the simple circulation model without albumin, dialysis or ultrafiltration.
[FIGURE 5 OMITTED]
In the present study, the first goal was to screen for an anti-MRSA agent with high affinity for haemodialysis filter membranes in a straightforward binding experiment that incorporated small pieces of filter membrane suspended in a KRB solution. Because the binding of drug to filter may occur within the wall and the pores of the membrane as well as on the membrane surface, the entire filter was carefully primed with KRB prior to sectioning it into pieces. The results showed that teicoplanin was significantly and predominantly adsorbed by PS and PMMA membranes compared with PAN membranes. Linezolid and vancomycin were not significantly adsorbed onto any filter membrane. Contrary to these results, other studies recently showed that vancomycin can be adsorbed by haemodialysis membranes. For example, Qi and colleagues used a closed circuit model containing an unbuffered blood-containing solution and reported a small but significant adsorption of vancomycin by PAN, PS and polyamide membranes (14). Furthermore, Quale and colleagues used both in vivo and in vitro techniques and reported a possible binding of vancomycin to a newer, more permeable PAN membrane (15). The discrepancy between these reports and the current study may potentially be explained by differences in experiment methods. For example, the current screening model was quite simple and excluded the effect of perfusion and transmembrane pressure on the adsorption of the anti-MRSA agents.
Given that teicoplanin demonstrated the highest affinity for the PMMA and PS filter membranes in the initial screening experiment, an in vitro HDF circulation model was next employed to explore the influence of filter material on teicoplanin elimination by adsorption onto the membrane. The in vitro HDF experiment demonstrated that the high affinity of teicoplanin for PS and PMMA membranes may attribute to a large [CL.sub.HDF] when teicoplanin is administered to human patients during HDF. Finally, a simple HDF model that eliminated both dialysis and ultrafiltration was employed to show that mere passage through the PMMA and PS membranes in and of itself caused a significant decrease in teicoplanin concentration in the circulating KRB solution. In this model, adsorption of teicoplanin onto the circuit was minimal.
Because the plasma albumin-bound fraction of teicoplanin is ~90%6, this study investigated the effect of albumin on the pharmacokinetic parameters of teicoplanin in the original in vitro HDF model (with dialysis and filtration). When albumin was added to the KRB solution, the CLHDF for teicoplanin was greatly reduced for all membranes tested. The addition of albumin exerted a small but significant downward effect on the adsorption-dependent elimination of teicoplanin by PS and PMMA membranes. A similar observation was reported by Oborne and colleagues, who noted that the presence of albumin decreased the amount of fluconazole adsorbed by PS and polyamide membranes (10). Drugs that are highly albumin-bound are eliminated ineffectively by diffusion and ultrafiltration (3,4), which suggests that albumin attenuated teicoplanin clearance in this study primarily by affecting diffusion and ultrafiltration through the membranes, as opposed to adsorption onto the membranes.
Serum albumin levels of critically ill patients are often very low. The apparent volume of distribution volume ([V.sub.d]) and clearance of highly protein-bound drugs such as teicoplanin may therefore be increased in these patients compared with patients that have a normal serum albumin level. Although dose adjustments of teicoplanin in critically ill patients should thus be optimised, there is currently very limited information available regarding the clearance of substances during HDF or CHDF to guide the physician (16,17). The results obtained in this study provide an important step toward com-pensating for the paucity of knowledge in this area.
In accordance with the results of the present study, Menth et al reported that teicoplanin may be eliminated by adsorption to several dialysis membranes, including PAN, PS and PMMA (18). Moreover, accumulating clinical evidence indicates that elimination of teicoplanin by haemodialysis and/or haemofiltration may be dependent on the type of filter membrane employed (2,7,19,20). As such, the therapeutic drug monitoring-guided dosage regimen of teicoplanin administration is essential for patients treated with extracorpeal blood purification techniques (16,20). A high affinity to some filter membranes is also reported for other antibiotics (7-9,21-23).
The results of the present study underscore the importance of adjusting the dosage and timing of teicoplanin administration during HDF or CHDF using PS or PMMA membranes; in particular, a much higher dose than that currently recommended may be required because of increased drug clearance in serum albumin-deficient patients and drug loss resulting from adsorption to the filter membrane. Furthermore, it may be advantageous to select PAN membranes instead of PS or PMMA membranes when teicoplanin administration is necessitated during HDF.
The electrostatic coupling of drug with membrane is a proposed mechanism by which some drugs bind to filter membranes. Electrostatic coupling is in turn dependent upon the electric charge of each component. Teicoplanin is characterised by carboxyl and amino terminals with pharmacokinetic values of 3.1 and 7.1, respectively and is negatively charged at a physiological pH of 7.4. PS membranes have no net charge, while PMMA and PAN membranes have a negative charge (24,25). Therefore, electrostatic coupling does not explain why teicoplanin is preferentially adsorbed by PMMA and PS membranes. Teicoplanin is a glycopeptide with a structure that is reminiscent of a protein. Because various proteins and polypeptides (e.g. alpha2-microglobulin (26) and many cytokines27) bind to PMMA membranes, a non-specific mechanism may be involved in the teicoplanin/PMMA association.
The present study was conducted using an in vitro model and therefore the obtained results must be clinically confirmed. Critically ill patients generally receive numerous drugs and their plasma concentrations of many biologically active substances are elevated due to renal failure necessitating HDF. However, teicoplanin and other active substances may interact with plasma albumin and the haemodialysis filter membrane during HDF, with effects on pharmacokinetic parameters. Notably, teicoplanin was eliminated mainly by adsorption onto PS and PMMA haemofilters in the in vitro HDF model described herein. The PS and PMMA membranes eliminated teicoplanin more rapidly than the PAN membrane. The presence of albumin significantly reduced both HDF clearance and the adsorption-dependent elimination ratio, although there were complex but significant interactions between albumin and filter membranes.
In this study, the initial concentration of teicoplanin was set at 50 [micro]g/ml, based on the Cmax value obtained by a common clinical dosage. However, the Cmax of teicoplanin may be higher in the clinic when the loading dose is given at the beginning of treatment, and the higher drug concentration may influence the adsorption-dependent elimination and hence, the clearance. Furthermore, the concentration of albumin in the plasma can alter the degree of antibacterial agent-albumin binding, presumably producing significant variations in the pharmacokinetics of many highly protein-bound drugs (5). This was not taken into consideration in the present study. In addition, the parameters of HDF, including the flow rates of blood, dialysate, ultrafiltrate and replacement fluid, may vary among institutes and countries. While the flow rates applied herein are identical to the rates that are covered by Japanese social health insurance, they are likely not applicable to all HDF cases. Despite these limitations, the present results suggest that clinical studies will be imperative to confirm the in vitro observations, and to re-evaluate the current recommendations of teicoplanin dosing for patients managed with HDF or CHDF.
(1.) Pea F, Viale P, Pavan F, Furlanut M. Pharmacokinetic considerations for antimicrobial therapy in patients receiving renal replacement therapy. Clin Pharmacokinet 2007; 46:997-1038.
(2.) Choi G, Gomersall CD, Tian Q, Joynt GM, Freebairn RC, Lipman J. Principles of antibiotic dosing in continuous renal replacement therapy. Crit Care Med 2009; 37:2268-2282.
(3.) Bressolle F, Kinowski JM, de la Coussaye JE, Wynn N, Eledjam JJ, Galtier M. Clinical pharmacokinetics during continuous haemofiltration. Clin Pharmacokinet 1994; 26:457-471.
(4.) Reetze-Bonorden P, Bohler J, Keller E. Drug dosage in patients during continuous renal replacement therapy. Pharmacokinetic and therapeutic considerations. Clin Pharmacokinet 1993; 24:362-379.
(5.) Ulldemolins M, Roberts, JA, Rello J, Paterson DL, Lipman J. The effects of hypoalbuminaemia on optimizing antibacterial dosing in critically ill patients. Clin Pharmacokinet 2011; 50:99-110.
(6.) Wilson AP. Clinical pharmacokinetics of teicoplanin. Clin Pharmacokinet 2000; 39:167-183.
(7.) Kroh UF, Dinges GK, Lukasewitz P, Steinhausser WU, Holl TJ, Lennartz H. Elimination of drugs by the new polyamide hemofilter FH77H during various in vitro conditions. Blood Purif 1998; 16:49-56.
(8.) Tian Q, Gomersall CD, Ip M, Tan PE, Joynt GM, Choi GY. Adsorption of amikacin, a significant mechanism of elimination by hemofiltration. Antimicrob Agents Chemother 2008; 52:1009-1013.
(9.) Tian Q, Gomersall CD, Wong A, Leung P, Choi G, Joynt GM et al. Effect of drug concentration on adsorption of levofloxacin by polyacrylonitrile haemofilters. Int J Antimicrob Agents 2006; 28:147-150.
(10.) Oborne CA, Buckton G, Barber N. The interaction of albumin and drugs with two haemofiltration membranes. J Clin Pharm Ther 1994; 19:119-125.
(11.) Boak LM, Li J, Nation RL, Rayner CR. High-performance liquid chromatographic method for simple and rapid determination of linezolid in human plasma. Biomed Chromatogr 2006; 20:782-786.
(12.) Hanada K, Kobayashi A, Okamori Y, Kimura T, Ogata H. Improved quantitative determination of total and unbound concentrations of six teicoplanin components in human plasma by high performance liquid chromatography. Biol Pharm Bull 2005; 28:2023-2025.
(13.) Luksa J, Marusic A. Rapid high-performance liquid chromatographic determination of vancomycin in human plasma. J Chromatogr B Biomed Appl 1995; 667:277-281.
(14.) Tian Q, Gomersall CD, Leung PP, Choi GY, Joynt GM, Tan PE et al. The adsorption of vancomycin by polyacrylonitrile, polyamide and polysulfone hemofilters. Artificial Organs 2008; 32:81-84.
(15.) Quale JM, O'Halloran JJ, DeVincenzo N, Barth RH. Removal of vancomycin by high-flux hemodialysis membranes. Antimicrob Agents Chemother 1992; 36:1424-1426.
(16.) Meyer B, Traunmuller F, Hamwi A, Hollenstein UM, Locker GJ, Sperr WR et al. Pharmacokinetics of teicoplanin during continuous hemofiltration with a new and a 24-h used highly permeable membrane: rationale for therapeutic drug monitoring-guided dosage. Int J Clin Pharmacol Ther 2004; 42:556-560.
(17.) Wolter K, Claus M, Wagner K, Fritschka E. Teicoplanin pharmacokinetics and dosage recommendations in chronic hemodialysis patients and in patients undergoing continuous veno-venous hemodialysis. Clin Nephrol 1994; 42:389-397.
(18.) Menth M, Fiegel P. Elimination of teicoplanin via different dialysis membrane: in vitro investigation. Int J Antimicrob Chemother 1992; 1 (Suppl 1):45-48.
(19.) Thalhammer F, Rosenkranz AR, Burgmann H, Graninger W, Hollenstein U, Schenk P et al. Single-dose pharmacokinetics of teicoplanin during hemodialysis therapy using high-flux polysulfone membranes. Wien Klin Wochenschr 1997; 109:362-365.
(20.) Yagasaki K, Gando S, Matsuda N, Kameue T, Ishitani T, Hirano T et al. Pharmacokinetics of teicoplanin in critically ill patients undergoing continuous hemodiafiltration. Intensive Care Med 2003; 29:2094-2095.
(21.) Choi G, Gomersall CD, Lipman J, Wong A, Joynt GM, Leung P et al. The effect of adsorption, filter material and point of dilution on antibiotic elimination by haemofiltration an in vitro study of levofloxacin. Int J Antimicrob Agents 2004; 24:468472.
(22.) Inagaki O, Nishian Y, Iwaki R, Nakagawa K, Takamitsu Y, Fujita Y. Adsorption of nafamostat mesilate by hemodialysis membranes. Artif Organs 1992; 16:553-558.
(23.) Kronfol NO, Lau AH, Barakat MM. Aminoglycoside binding to polyacrylonitrile hemofilter membranes during continuous hemofiltration. ASAIO Trans 1987; 33:300-303.
(24.) Renaux JL, Thomas M, Crost T, Loughraieb N, Vantard G. Activation of the kallikrein-kinin system in hemodialysis: role of membrane electronegativity, blood dilution, and pH. Kidney Int 1999; 55:1097-1103.
(25.) Valette P, Thomas M, Dejardin P. Adsorption of low molecular weight proteins to hemodialysis membranes: experimental results and simulations. Biomaterials 1999; 20:1621-1634.
(26.) Aoike I. Clinical significance of protein adsorbable membranes--long-term clinical effects and analysis using a proteomic technique. Nephrol Dial Transplant 2007; 22 (Suppl 5):13-19.
(27.) Nakada TA, Oda S, Matsuda K, Sadahiro T, Nakamura M, Abe R et al. Continuous hemodiafiltration with PMMA Hemofilter in the treatment of patients with septic shock. Mol Med 2008; 14:257-263.
Y. SHIRAISHI *, M. OKAJIMA ([dagger]), Y. SAI ([double dagger]), K. MIYAMOTO ([section]), H. INABA **
Department of Emergency Medical Science, Kanazawa University Graduate School of Medicine, Kanazawa, Japan
* MS, Graduate student, Department of Clinical Pharmacy, Kanazawa University Graduate School of Natural Science and Technology.
([dagger]) MD, PhD, Assistant Professor, Department of Emergency and Critical Care Medicine.
([double dagger]) PhD, Associate Professor and Vice Director, Department of Hospital Pharmacy, Kanazawa University Hospital.
([section]) PhD, Professor and Director, Department of Hospital Pharmacy, Kanazawa University Hospital.
** MD, PhD, Professor and Chairman, Department of Emergency and Critical Care Medicine.
Address for correspondence: Dr H. Inaba, email: hidinaba@med. kanazawa-u.ac.jp
Accepted for publication on January 19, 2012.
Table 1 Effects of the filter membrane and albumin on the [CL.sub.HDF] and adsorption-dependent elimination ratio of teicoplanin in the in vitro HDF model Albumin = 0 g/dl Filter [CL.sub.HDF], Adsorption-dependent membrane ml/min elimination ratio at 120 min, % PAN 12.6 [+ or -] 1.0 0.4 [+ or -] 0.3 PS 50.6 [+ or -] 2.6 69.8 [+ or -] 0.9 PMMA 60.8 [+ or -] 4.7 89.4 [+ or -] 2.4 Albumin = 3 g/dl Filter [CL.sub.HDF], Adsorption-dependent membrane ml/min elimination ratio at 120 min, % PAN 6.7 [+ or -] 0.9 3.7 [+ or -] 2.3 PS 27.8 [+ or -] 0.7 61.4 [+ or -] 4.8 PMMA 26.7 [+ or -] 1.0 75.6 [+ or -] 1.9 Each value is the mean [+ or -] SD (n = 3). The CLHDF and adsorption -dependent elimination ratio at 120 minutes was estimated as described in the materials and methods section. [CL.sub.HDF] was significantly affected by both the filter membrane and albumin (P <0.01, two-way analysis of variance [ANOVA]). There was a significant interaction between the filter membrane and albumin (P <0.01, two-way ANOVA). The adsorption-dependent elimination ratio of teicoplanin significantly differed among the three filter membranes (PMMA >PS >PAN, two-way ANOVA followed by the Tukey-Kramer test, P <0.01). There was a significant interaction between the filter membrane and albumin (P <0.01, two-way ANOVA). Addition of albumin significantly decreased the adsorption-dependent eliminations by PS and PMMA membranes (P <0.05). [CL.sub.HDF] = clearance haemodiafiltration, PAN = polyacrylonitrile, PS = polysulfone, PMMA = polymethylmethacrylate.
|Printer friendly Cite/link Email Feedback|
|Author:||Shiraishi, Y.; Okajima, M.; Sai, Y.; Miyamoto, K.; Inaba, H.|
|Publication:||Anaesthesia and Intensive Care|
|Date:||May 1, 2012|
|Previous Article:||Assessing the performance of a continuous infusion for potassium supplementation in the critically ill.|
|Next Article:||Enhanced recovery after surgery program for elective abdominal surgery at three Victorian hospitals.|