Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 47: 901-909, 2001;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by De Smet, R.
Right arrow Articles by Vanholder, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Smet, R.
Right arrow Articles by Vanholder, R.
Related Collections
Right arrow Drug Monitoring and Toxicology
(Clinical Chemistry. 2001;47:901-909.)
© 2001 American Association for Clinical Chemistry, Inc.

Articles

Heparin-induced Release of Protein-bound Solutes during Hemodialysis Is an in Vitro Artifact

Rita De Smet1a, Jacqueline Van Kaer1, Hartmut Liebich3, Gerrit Lesaffer1, Alain Verstraete2, Annemieke Dhondt1, Phillippe Duym1, Norbert Lameire1 and Raymond Vanholder1

University Hospital Gent, Departments of
1 Internal Medicine, Nephrology Division, and
2 Clinical Biology, B 9000 Gent, Belgium.
3 Medizinische Universitätsklinik, D 7400 Tübingen, Germany.

aAddress correspondence to this author at: University Hospital Gent, Department of Internal Medicine, Nephrology Division, De Pintelaan 185, B 9000 Gent, Belgium. Fax 32-09-240-4599; e-mail rita.desmet{at}rug.ac.be.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Several studies have pointed to a release of drugs or protein-bound solutes from their binding sites during heparinization. The effect is attributed to the metabolism of triglycerides to free fatty acids (FFAs), which compete with drugs for protein binding sites. This study evaluated the impact of intradialytic heparin on the free concentration of the uremic toxin p-cresol and on FFAs.

Methods: Blood samples from hemodialysis (HD) patients, before and during HD, were collected with selected anticoagulation strategies. We assessed the effects of standing time, temperature, pH, and the addition of a lipase inhibitor, tetrahydrolipstatin (THL) to blood samples on the free p-cresol concentration. p-Cresol was analyzed by HPLC with fluorescence detection. We measured FFAs by gas chromatography, and the free fractions of added valproic acid and phenytoin were evaluated by fluorescence polarization immunoassay.

Results: In blood samples (n = 22) not submitted to a specific treatment, free p-cresol increased from 9.9 ± 5.1 to 31.9 ± 22.3 µmol/L after 30 min of heparin HD (P <0.001) and correlated significantly with FFAs (r = 0.80; P = 0.002; n = 12). There was no increase in free p-cresol during heparin-free HD (n = 6) and trisodium citrate HD (n = 9). In addition, p-cresol in ultrafiltrates (n = 3) did not correspond to the free p-cresol in heparinized blood, suggesting that the increase in free p-cresol was artifactual. The release of p-cresol in the test tube was enhanced by standing time (n = 6), sample temperature (n = 6), and alkaline pH (n = 6). Inhibition of lipase activity with THL prevented the increase of FFAs (n = 6) and the release of free p-cresol during HD (n = 22). These results were corroborated by the study of the free fraction of valproic acid (n = 6) and phenytoin (n = 6).

Conclusions: The free concentrations of protein-bound solutes in plasma of heparinized patients are influenced by external factors that alter the lipase activity in the test tube. The free fraction does not increase during HD when lipase activity is neutralized at the time of blood sampling, so that previously reported increases are probably artifacts.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The free fractions of several protein-bound drugs, such as phenytoin, valproic acid, and free fatty acids (FFAs),1 increase during heparinization (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17), and therapeutic complications have been made based on these findings (1)(2)(8)(13)(17). The administration of heparin is believed to provoke the release of lipolytic enzymes from their sites of action at the capillary endothelium into the plasma, leading to hydrolysis of triglycerides followed by an increase in the FFA plasma concentration (10)(16)(18)(19)(20). These FFAs displace drugs or ligands from their protein binding sites (1)(3)(5)(8)(9)(14)(15)(17)(21)(22).

Studies during hemodialysis (HD) with heparin have concentrated on the protein binding of drugs (1)(7)(9)(15). One study on the behavior of protein-bound uremic retention solutes points to a correlation between FFAs and unbound uremic compounds (15). Analogous to most drugs, where the free fraction exerts biological action, the acute release of protein-bound compounds may have an impact on their toxicity (15)(23)(24)(25)(26).

Our group recently identified p-cresol as a uremic retention compound (27). It exhibits substantial protein binding, partially lipophilic properties, and strong biochemical inhibitory actions (27)(28)(29)(30)(31). The present study was undertaken to evaluate the effect of HD-related anticoagulation on free p-cresol and FFAs during HD and ultrafiltration. We asked whether changes in free ligands result from the enzymatic activity of lipase in the test tube after sample collection, rather than a systemic in vivo reaction. We assessed the influence of storage time, temperature, and pH and the effect of inhibition of lipase activity at the time of blood collection. We performed similar experiments with valproic acid and phenytoin because most studies on the effect of heparin on protein binding have been performed with drugs and because valproic acid and phenytoin are two drugs showing substantial changes in this setting.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
patients
We studied 50 chronic HD patients (32 women and 18 men; mean age, 65 ± 14 years) who had been on dialysis for 49 ± 46 months. The eligibility criteria were as follows: (a) patients with an arteriovenous fistula2 on chronic intermittent HD 3 times per week; (b) a stable clinical condition with a stable unfractionated heparin (UFH) or low-molecular weight heparin (LMWH) schedule for at least 2 months; and (c) age >18 years. Patients with previous thrombocytopenia induced by heparin, an allergy to heparin, or intradialytic hyperalimentation were excluded.

The local Ethics Committee approved the study, and all patients gave informed consent. HD was performed with low-flux polysulfone (F8; Fresenius AG). The target Kt/V,3 as determined by two-pool kinetics, was 1.3. The composition of the dialysate, unless stated otherwise, was 38.5 mmol/L bicarbonate, 138 mmol/L sodium, 104 mmol/L chloride, 4 mmol/L acetate, 1.25 mmol/L calcium, and 0.5 mmol/L magnesium. The dialysate potassium concentration was adapted to the needs of the patients and ranged from 1 to 3 mmol/L. Dialysate and dialyzer blood flows were ~500 and 250 mL/min, respectively, for a dialysis time of 3.5–4.5 h. Mean predialysis blood urea was 19.9 ± 5.6 mmol/L (55.9 ± 15.7 mg/100 mL of blood urea nitrogen), and creatinine was 911 ± 221 µmol/L (10.3 ± 2.5 mg/100 mL).

The following anticoagulants were used: (a) UFH (Heparin Leo, 5000 IU/mL; Leo Pharmaceutical Products) with an average molecular weight of 12 000–15 000; (b) LMWH (Innohep,® 10 000 anti-Xa4 IU/mL; Leo Pharmaceutical Products) with an average molecular weight of 2500–6000; and (c) trisodium citrate (1.2 mol/L; Sterop SA).

Blood samples were collected before anticoagulation of the patients, at the start of HD (0 min of HD), and at selected times during HD. The test tubes contained no anticoagulant. Some patients underwent pure ultrafiltration, and uremic ultrafiltrate and blood were collected simultaneously.

reagents
HPLC-grade water, HPLC-grade methanol, and isopropyl ether were purchased from Acros Organics. The ammonia solution (HiPerSol; specific gravity, ~0.88) for HPLC was from BDH Laboratory Supplies. The methanolic solution of p-cresol (46.2 mmol/L) and all other reagents were purchased from Sigma Chemical. Tetrahydrolipstatin (THL) was a gift of Hoffmann-La Roche (Basel, Switzerland). Amersham (Little Chalfont, United Kingdom) prepared the [14C]phenytoin (250 µCi).

determination of p-cresol and FFAs
p-Cresol was determined as described previously (27). In brief, total p-cresol was measured after deproteinization of serum or plasma with HCl and NaCl. Free p-cresol was determined in the filtrate obtained after ultracentrifugation through a molecular filter with a cutoff of Mr 30 000 (Centrifree® Micropartition Devices; Amicon). p-Cresol was extracted with isopropyl ether and analyzed by HPLC on a C18 reversed-phase column.

The concentrations of 20 FFAs (lauric acid, myristic acid, pentadecanoic acid, palmitic acid, palmitoleic acid, heptadecanoic acid, stearic acid, oleic acid, octadecenoic acid, linoleic acid, {gamma}-linolenic acid, {alpha}-linoleic acid, arachidic acid, eicosadienoic acid, eicosatrienoic acid, arachidonic acid, eicosapentaenoic acid, docosatetraenoic acid, docosapentaenoic acid, and docohexaenoic acid) were determined in uremic serum or plasma. In brief, the FFAs were reacted with a mixture of methanol (absolute) and acetyl chloride (50:1 by volume). The samples were analyzed by capillary gas chromatography on a fused-silica column with flame ionization detection. The quantification was based on an internal standard (13,16,19-docosatrienoic acid) and on calibrators (32). The sum of the molar concentrations of the determined FFAs and the individual concentration of the saturated FFAs palmitic acid (C16:0) and stearic acid (C18:0) and the unsaturated FFAs oleic acid (C18:1 n-9) and linoleic acid (C18:2 n-6) are reported.

concentrations of p-cresol and FFAs without specific sample treatment or modulation of lipase activity
p-Cresol and FFAs during HD with heparin anticoagulation.
Twenty-two patients (six determinations per patient) were anticoagulated with UFH or LMWH in random order. UFH was administered as a bolus injection of 1700 ± 470 IU at the inlet bloodline before the start of HD, followed by a continuous infusion of 1450 ± 610 IU/h during the HD session. For HD with LMWH, a single bolus of 4500 ± 1500 anti-Xa IU was administered into the inlet line before the start of HD, after which no further anticoagulant was given. Six blood samples were collected: at the start of the HD session before the administration of the anticoagulant, and at 30, 60, 120, 180, and 240 min (end of the HD session). The blood samples were processed at room temperature. No specific instructions were followed regarding the standing time of the samples. In addition, the FFA concentration was determined in 12 of these patients at the start of HD and at 30 and 240 min.

Effect of the in vitro addition of palmitic acid and heparin.
To evaluate the effect of FFAs on free p-cresol, increasing quantities of palmitic acid were added to six pre-HD uremic serum samples (four determinations per patient). From each sample four 1-mL aliquots of blood were transferred into test tubes. Palmitic acid was dissolved in ethanol at three different concentrations, and 5 µL was added to three serum samples to obtain final FFA concentrations of 100, 500, and 1000 µmol/L. To one sample, 5 µL of ethanol was added to exclude a possible bias induced by the ethanol. The test tubes were incubated at 37 °C for 15 min, and free p-cresol was determined.

To evaluate the role of heparin per se on free p-cresol, UFH or LMWH was added to pre-HD blood samples from six patients (eight determinations per patient) to obtain final concentrations of 0, 12.5, 62.5, and 125 IU/mL for UFH and 0, 6, 30, and 60 anti-Xa IU for LMWH. The blood was incubated at 37 °C for 15 min, and free p-cresol was determined.

p-Cresol during heparin-free HD.
In six patients (seven determinations per patient), the heparinization was postponed until 30 min of HD; a bolus containing 4150 ± 1100 anti-Xa IU LMWH was then given. Blood samples were collected at 0, 30, 60, 120, 180, and 240 min, with an additional sample at 35 min, 5 min after the start of the heparinization. The blood samples were processed at room temperature. No specific instructions were followed regarding the standing time of the samples.

p-Cresol during trisodium citrate HD.
Anticoagulation with UFH and trisodium citrate was alternatively administered in random order and in a crossover design to nine patients as described previously (33). During the last hour of the session, citrate anticoagulation was shifted to heparin. Blood samples were collected at 0, 30, 180, and 210 min. The blood samples were processed at room temperature. No specific instructions were followed regarding the standing time of the samples (eight determinations per patient).

p-Cresol in uremic ultrafiltrate.
Starting from the hypothesis that p-cresol in ultrafiltrate and plasma should equilibrate, we compared free p-cresol in pre-HD serum to ultrafiltrate and serum or plasma obtained simultaneously during an ultrafiltration session in three patients (five determinations per patient). A blood sample was taken before the start of HD. Ultrafiltration was performed for 5 min without heparinization; blood and ultrafiltrate were then collected simultaneously. Heparinization with a bolus containing 5100 ± 1900 anti-Xa IU LMWH was then started, and blood and ultrafiltrate were again collected simultaneously after 5 min. The blood samples were processed at room temperature. No specific instructions were followed regarding the standing time of the samples.

effects of sample manipulations with potential impact on the lipase activity
Effect of temperature.
Blood samples from six HD patients (four determinations per patient) anticoagulated with 4600 ± 1300 anti-Xa IU LMWH were obtained at 0 min and at 30 min after the start of HD. Each sample was immediately divided into two equal aliquots. One part was placed on ice and centrifuged at 4 °C; the second part was centrifuged at room temperature. The serum or plasma samples were then immediately ultracentrifuged to remove the protein-bound solutes and the lipase enzyme.

Effect of time.
The effect of time on free p-cresol was evaluated in two samples obtained at 0 min and after 30 min of HD (4800 ± 1500 anti-Xa IU LMWH) from six patients (eight determinations per patient). The blood was divided into four aliquots. Free p-cresol was immediately determined in one 0-min sample and in one sample collected after 30 min of HD. The other serum or plasma samples were left standing at room temperature for 1, 2, and 3 h before ultracentrifugation.

Effect of pH.
To 1-mL serum or plasma samples from six patients (four determinations per patient), obtained at 0 and 30 min of HD (4750 ± 1350 anti-Xa IU LMWH), 100 µL of KH2PO4–Na2HPO4 (pH 6) or 100 µL of isotonic saline was added. The samples were left at room temperature for 60 min. The free p-cresol concentration was compared in the buffered and nonbuffered samples.

comparison of p-cresol and FFAs in samples with and without thl during hd
Twenty-two patients (six determinations per patient) were anticoagulated with LMWH at a dose of 4750 ± 1400 anti-Xa IU. Blood was collected at the start of the session, at 30 min, and at the end of the HD session. Immediately after blood collection, each sample was divided into two test tubes, one without THL and one with THL (1 mg of THL powder was weighed into a test tube to which to 1 mL blood was added). The samples were left at room temperature for 60 min. The FFA concentration was also determined in samples from six patients.

free fraction of valproic acid and [14c]phenytoin with and without thl during hd
Blood from 12 patients (six determinations per patient) was obtained at the start, at 30 min, and at the end of the HD session (4950 ± 1800 anti-Xa IU LMWH). The same manipulation procedure, using test tubes with and without THL, as described above was applied for the evaluation of the free fraction of valproic acid and phenytoin.

To 1-mL plasma samples from six patients, 5 µL of unlabeled valproic acid was added to achieve a concentration of 80 µmol/L. Valproic acid was determined by fluorescence polarization immunoassay (Abbott Laboratories) on an AxSYM analyzer. For the other samples, 50 µL of radiolabeled [14C]phenytoin (dissolved in 20 mL/L methanol and diluted 1:30 with isotonic NaCl) was added to 250 µL of plasma. The samples were incubated at 37 °C for 15 min. The free fractions of those two drugs were determined after ultracentrifugation at ambient temperature (Centrifree). [14C]Phenytoin was measured by liquid scintillation counting as described previously (34).

statistical evaluation
Values are expressed as mean ± SD. Statistical nonparametric analyses were performed with the Wilcoxon test for paired data and with the Mann–Whitney U-test for unpaired data. Correlation analysis was performed by the Spearman test. P <0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
concentrations of p-cresol and FFAs without specific sample treatment or modulation of lipase activity
p-Cresol and FFAs during HD with heparin anticoagulation.
The evolution of total and free p-cresol in untreated blood samples during HD with heparin is illustrated in Table 1 . The increase of free p-cresol after UFH and LMWH anticoagulation was equal for both heparins until 120 min of HD; at the end of HD, the concentration of free p-cresol was slightly higher with UFH than with LMWH. However, because the intradialytic evolution was similar, the data were pooled. For most of the other experiments, LMWH was used because it currently is used as an anticoagulant in our unit.


View this table:
[in this window]
[in a new window]
  		Table 1. Evolution of total and free p-cresol and of protein binding during HD with UFH (n = 11) and LMWH (n = 11) in untreated samples.

The total p-cresol concentration decreased over the entire dialysis session. An increase of free p-cresol was observed 30 min after heparinization, with a gradual decline during further dialysis (at the end of HD, values returned almost to the predialysis value). Protein binding of p-cresol showed a decline at 30 min with only partial recovery at the end of HD (Table 1Up ).

In Table 2 , the evolution of the FFAs (sum of the molar concentration of all measured FFAs), as well as that of the individual compounds palmitic acid, stearic acid, oleic acid, and linoleic acid is reported. In parallel with free p-cresol, an increase in FFAs was observed at 30 min of HD. All individual FFAs behaved in a similar way.


View this table:
[in this window]
[in a new window]
  		Table 2. Evolution of FFAs (sum of molar concentration of all measured FFAs) and of palmitic acid, stearic acid, oleic acid, and linoleic acid during HD (n = 12).

To evaluate the possible relationship between free p-cresol and FFA concentration, the correlation between the two variables was calculated at different points during HD. A significant linear correlation was found at 30 and 240 min of HD, but not at the start of the session (Table 2Up ).

Effect of the in vitro addition of palmitic acid and heparin.
Free p-cresol in pre-HD samples increased from 6.8 ± 1.8 µmol/L to 7.6 ± 2.2, 9.2 ± 2.4, and 13.0 ± 2.9 µmol/L (P <0.05) when 0, 100, 500, and 1000 µmol/L palmitic acid, respectively, were added. Ethanol alone (5 µL/mL of serum) had no significant effect on free p-cresol. The ethanol concentration was the same in all samples, and free p-cresol was the lowest in the sample without added palmitic acid. It can be concluded that the observed increase of free p-cresol was caused by the increasing amount of the fatty acid. Both UFH and LMWH had no influence on free p-cresol (data not shown).

p-Cresol during heparin-free HD.
Fig. 1A shows the evolution of total and free p-cresol when heparinization was postponed until 30 min of dialysis. Free p-cresol decreased slightly, but not significantly, during the first 30 min and began to increase only after heparinization was started. Total p-cresol decreased progressively during the dialysis session, from 89.1 ± 50.8 µmol/L to 69.9 ± 36.8 µmol/L (P <0.05).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 1. Effect of anticoagulant administration during HD on p-cresol evolution.

(A), evolution of total ({blacktriangleup}) and free p-cresol (•) during heparin-free HD (up to min 30) and after heparinization. *, P <0.05 vs 0 min. H, heparinization period. (B), evolution of free p-cresol when heparin was administered during the entire dialysis session (dashed line) or during HD with trisodium citrate as anticoagulation until min 180, after which trisodium citrate was replaced by heparin (solid line). H, heparinization period; C, citrate anticoagulation period. {circ} and *, P <0.01 and P <0.05, respectively, vs 0 min; +, P <0.01 vs heparin HD.

p-Cresol during trisodium citrate HD.
The evolution of free p-cresol in patients who had been anticoagulated in random order and in a crossover design with citrate or heparin is given in Fig. 1BUp . Free p-cresol increased at 30 min of HD only during anticoagulation with heparin (Fig. 1BUp , dashed line). When patients were anticoagulated with citrate, free p-cresol did not change, whereas the increase was observed only when heparin was administered after 180 min of HD (Fig. 1BUp , solid line), again indicating that the increase in free p-cresol was directly related to heparinization.

p-Cresol in uremic ultrafiltrate.
The free p-cresol in serum before a dialysis session (5.4 ± 3.7 µmol/L) was identical to the values in uremic ultrafiltrate and serum obtained immediately before heparinization (5.6 ± 4.0 and 5.5 ± 4.1 µmol/L, respectively). After heparinization, plasma free p-cresol increased to 17.8 ± 19.5 µmol/L, whereas the p-cresol concentration in simultaneously collected ultrafiltrate was unchanged, 5.3 ± 3.4 µmol/L.

effects of sample manipulations with potential impact on the lipase activity
Effect of temperature.
In samples collected after 30 min of HD, free p-cresol was higher in samples processed immediately at 22 °C than in those processed at 4 °C (Fig. 2A , right-hand columns). The difference was absent in samples obtained at 0 min of HD (Fig. 2A , left-hand columns).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Effect of temperature (A), time (B), and phosphate buffer (C) on free p-cresol evolution.

(A), effect of temperature on free p-cresol in blood collected at 0 and 30 min of HD. *, P <0.05 vs 0 min at 22 °C; {circ}, P <0.05 vs 4 °C. (B), effect of time (0, 60, 120, and 180 min) on free p-cresol in serum or plasma allowed to stand at room temperature in samples obtained at 0 and 30 min of HD. *, P <0.05 vs immediately treated samples (open column). (C), influence of phosphate buffer on free p-cresol in serum or plasma collected at 0 and 30 min of HD. - B, without buffer; + B, with buffer). *, P <0.05 vs 0 min of HD; {circ}, P <0.05 vs buffered samples.

Effect of time.
In samples collected at 0 min of HD and allowed to stand at 22 °C, the free p-cresol remained constant (Fig. 2BUp , left-hand columns). When plasma collected at 30 min of HD was allowed to stand at 22 °C for 60, 120, and 180 min, respectively, the concentration of increased progressively compared with samples in which free p-cresol was measured immediately (P <0.05; Fig. 2BUp , right-hand columns).

Effect of pH.
When serum was allowed to stand at room temperature for 60 min, a gradual increase of pH was observed. In phosphate-buffered samples, the pH remained at 7.2 ± 0.1 after 60 min. In samples collected at 0 min of HD, with or without buffer, free p-cresol remained the same (Fig. 2CUp , left-hand columns). The increase in free p-cresol observed in samples obtained at 30 min of HD and allowed to stand for 60 min was markedly attenuated with the addition of phosphate buffer to the samples (P <0.05; Fig. 2CUp , right-hand columns).

comparison of p-cresol and FFAs in samples with and without thl during hd
THL had no effect on free p-cresol in samples collected before the start of HD (Fig. 3 ). In the samples obtained at 30 min, the free p-cresol concentration without and with THL was 14.4 ± 12.0 and 5.5 ± 4.2 µmol/L, respectively (P <0.001). At the end of HD, free p-cresol still remained higher in the non-THL-treated samples (P <0.05).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. Evolution of free p-cresol during HD in samples without THL (hatched columns) and with THL (filled columns).

* and {circ}, P <0.001 and P <0.05, respectively, with vs without THL; + and {wedge}, P <0.001 and P <0.01, respectively, vs 0 min of HD.

FFAs were determined in six samples. The addition of THL had no effect on the FFA concentrations before the start of HD. In samples obtained at 30 min and at the end of dialysis, the FFA concentration in samples without THL was significantly higher than the concentration in samples with THL (P <0.05; Table 3 ). Only in samples without THL was the FFA concentration higher at 30 min compared with samples before HD (P <0.05). The individual FFAs followed the same evolution.


View this table:
[in this window]
[in a new window]
  		Table 3. Concentration of the sum of the FFAs and of palmitic acid, stearic acid, oleic acid, and linoleic acid during HD (n = 6).

free fraction of valproic acid and [14c]phenytoin with and without thl during hd
THL had no effect on the free fraction of valproic acid and phenytoin in serum before the start of HD. At 30 min, the free fractions of both drugs were significantly higher in plasma without THL than in plasma with THL (P <0.05; Table 4 ).


View this table:
[in this window]
[in a new window]
  		Table 4. Free fraction of valproic acid (n = 6) and phenytoin (n = 6) added to blood samples collected during HD.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our results demonstrate that in blood from heparinized HD patients (a) heparin-related increases in free p-cresol and FFAs were observed, and (b) no such increases were seen when the lipase inhibitor, THL, was added to the test tube before blood sampling.

Several studies point to an increase in the free fractions of protein-bound solutes and FFAs during heparinization (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17). This effect is attributed to the heparin-induced activation of lipoprotein lipase, which hydrolyzes the triglycerides, releasing FFAs; this release currently is accepted as a classical epiphenomenon of heparinization (18)(19)(20). Increases in the unbound fractions of drugs or of FFAs have been reported not only during HD, but also in other therapeutic and/or diagnostic conditions where heparin is applied, such as cardiac catheterization and cardiac surgery and in healthy volunteers who received intravenous heparin (3)(5)(6)(11)(12)(13).

HD might be of specific interest for the study of protein binding in relation to heparinization because of (a) the intermittent and repetitive nature of the anticoagulation, (b) the presence of basic disturbances of solute protein binding in the context of the uremic syndrome, and (c) the potential impact on the biological activity of protein-bound retention solutes and protein-bound drugs (15)(23)(24)(25)(26).

Sakai et al. (15) reported a correlation of the free concentrations of indoxyl sulfate and indole acetate and FFAs during HD with heparin anticoagulation. Changes in protein binding might have an impact on biological functions because it is conceivable that for many ligands the free fraction exerts biological activity (15)(23)(24)(25)(26).

In the present study, an increase in free p-cresol in untreated blood samples collected from heparinized HD patients was demonstrated (Table 1Up ). The effects of UFH and LMWH were similar, except that LMWH released less p-cresol at the end of HD. This can probably be attributed to the lower lipolytic activity of LMWH (35). No changes in free p-cresol were observed when no heparin was applied (Fig. 1Up ). In view of the reported relationship between heparin-induced production of FFAs and release of protein-bound ligands, we found a correlation in heparinized blood of free p-cresol with FFAs. Heparin per se, added in vitro to pre-HD blood, had no influence on the release of p-cresol. Likewise, heparin per se had no direct effect on the protein binding of propranolol and clofibric acid (6)(7).

The increasing amounts of FFAs observed in samples collected during HD can be attributed to lipase activity induced by heparin (20). The in vitro addition of palmitic acid to pre-HD serum induced the release of p-cresol from its binding sites, pointing to the role of FFAs in ligand displacement. Similar effects have also been demonstrated for propranolol, warfarin, furosemide, phenytoin, and valproic acid after the addition of FFAs to serum (8)(14)(15)(16)(17)(21)(22).

In ultrafiltrates obtained during HD, after heparinization of the patients, no increase of free p-cresol was observed. According to the law of convection, the solute concentration in the ultrafiltrate should be in equilibrium with the free concentration in the blood. This finding suggested that the increase in free p-cresol observed after 30 min of dialysis was an in vitro artifact.

There was also another set of data that could not be explained if free p-cresol really increased. We previously demonstrated that p-cresol inhibited metabolic response of phagocytic cells (28), but on the other hand, after 15 min of dialysis with heparin, no change in phagocytic response was observed when non-complement-activating dialysis membranes were used (36). Preliminary data from our laboratory indicate that an increase in free p-cresol enhances the suppressive action of p-cresol on phagocytic NAPH-oxidase activity (R. Vanholder, unpublished data). If one accepts that free p-cresol exerts biological action and that free p-cresol increases as a result of heparinization, a consistent decrease in phagocytic activity should have been the consequence, but this was not the case.

To explain the discrepancy between the in vitro and in vivo observations, we hypothesized the presence of ongoing lipase activity in the samples after collection. In the first part of the study, the blood samples had been processed at room temperature, and no specific instructions were followed regarding the standing time of the samples. In the subsequent experiments, the standing time of the samples, as described in Materials and Methods, was defined exactly. Maintenance of the blood obtained at 30 min of HD at 4 °C decreased the release of p-cresol compared with samples processed at room temperature (Fig. 2AUp ). In addition, a progressive increase of free p-cresol was observed when plasma obtained at 30 min of HD was kept at room temperature for 3 h (Fig. 2BUp ). Because of dissipation of carbon dioxide, serum becomes alkaline. After neutralization of the pH by the addition of phosphate buffer, the increase in free p-cresol was also prevented (Fig. 2CUp ). The optimal lipase activity is at pH 8–9 (37). These data suggest that lipase activity continues in the test tube in the blood or plasma of heparinized patients.

A next step was to inactivate the lipase activity in the collected blood samples. THL, a lipase blocker that is used for the treatment of obesity (38)(39)(40), has been demonstrated in vitro to be a lipase blocker that induces an almost 100% inhibition (41). The addition of THL to the test tubes before blood collection prevented increases of free p-cresol and FFAs in intradialytic heparinized blood samples (Fig. 3Up and Table 3Up ).

Similar experiments were undertaken with valproic acid and phenytoin. Here also transient increases in the free fractions of the drugs in samples obtained at 30 min of HD were observed. This can be attributed to higher lipase activity in those samples. The addition of THL to the plasma had also an inhibitory effect on the increase of the free drug fraction, as for p-cresol (Table 4Up ).

These data suggest that most if not all of the increases in FFAs and free p-cresol during heparinization are artifacts, induced by continued lipase action in the test tube once collection of the sample has been accomplished. A few other studies have also suggested that ligand release during heparinization is at least in part artifactual (42)(43)(44)(45)(46). These results are in agreement with the study by Krebs et al. (47), who found, in parallel with the present data, that THL is a potent inhibitor of in vitro lipolysis, blocking lipase activity in blood samples drawn during triglyceride/heparin infusion.

In conclusion, the increases in free concentrations of p-cresol, FFAs, and drugs during HD with heparin are influenced by different external factors such as standing time, temperature, and pH of the samples, which are related to the continuing lipase activity once the sample has been collected. There are no increases in the free fractions when the lipase activity is inhibited by THL at the time of blood sampling. The observed increases of free drugs and of free uremic compounds that have repeatedly been reported during HD and other conditions necessitating heparinization need to be interpreted with caution because there are no increases in free fractions when lipase activity is blocked.


   Acknowledgments
 
This study was supported by a Baxter extramural grant.


   Footnotes
 
1 Nonstandard abbreviations: FFA, free fatty acid; HD, hemodialysis; UFH, unfractionated heparin; LMWH, low-molecular weight heparin; and THL, tetrahydrolipstatin.

2 An arteriovenous fistula is a surgical connection between an artery and a vein under the skin for the purpose of HD.

3 Kt/V is a parameter used to express the dialysis adequacy. K = dialyzer clearance (mL/min), t = duration of the dialysis (h), and V = volume of body water of a patient.

4 Anti-Xa, anticoagulant activity.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Storstein L, Janssen H. Studies on digitalis. VI. The effect of heparin on serum protein binding of digitoxin and digoxin. Clin Pharmacol Ther 1976;20:15-23.[ISI][Medline] [Order article via Infotrieve]
  2. Bergrem H, Leivestad T. Dialysis death and increased free fatty acids [Letter]. Lancet 1978;25:1160.
  3. Wood M, Shand DG, Wood AJ. Altered drug binding due to the use of indwelling heparinized cannulas (heparin lock) for sampling. Clin Pharmacol Ther 1979;25:103-107.[ISI][Medline] [Order article via Infotrieve]
  4. Wessel T, Christophersen B. Systemic heparinization of uremic patients and its effects on blood lipids on in vitro toxicity of the plasma. Clin Nephrol 1982;18:135-140.[ISI][Medline] [Order article via Infotrieve]
  5. Naranjo CA, Sellers EM, Khouw V, Alexander P, Fan T, Shaw J. Variability in heparin effect on serum drug binding. Clin Pharmacol Ther 1980;28:545-550.[ISI][Medline] [Order article via Infotrieve]
  6. Sager G, Hansteen V, Aakesson Jacobsen S. Effect of heparin on serum binding of propranolol in the acute phase of myocardial infarction. Br J Clin Pharmacol 1981;12:613-620.[ISI][Medline] [Order article via Infotrieve]
  7. Lacour B, Di Giulio S, Nicolaï A, Drücke T, Debray M, Boulu RG. Etude de la liaison de l’acide clofibrique aux protéines sériques chez l’urémique chronique: influence de la dialyse et de l’héparine. Nephrologie 1982;3:19-22.[Medline] [Order article via Infotrieve]
  8. Gronhoj Larsen F, Gronhoj Larsen C, Andersen S, Norgaard A, Hansen E, Brodersen R.. Warfarin binding to plasma albumin, measured in patients and related to fatty acid concentration. Eur J Clin Invest 1986;16:22-27.[ISI][Medline] [Order article via Infotrieve]
  9. Horiuchi T, Johno I, Kitazawa S, Goto M, Hata T. Plasma free fatty acids and protein binding of disopyramide during haemodialysis. Eur J Clin Pharmacol 1987;33:327-329.[ISI][Medline] [Order article via Infotrieve]
  10. Boobis AR, Burley D, Davies DM, Davies DS, Harrison PI, Park BK, Goldberg LI. Heparin sodium and calcium heparin. In: Dollery C, ed. Therapeutic drugs. Edinburgh: Churchill Livingstone, 1991:H15..
  11. Hynynen M, Hynninen M, Soini H, Neuvonen PJ, Heinonen J. Plasma concentration and protein binding of alfentanil during high-dose infusion for cardiac surgery. Br J Anaesth 1994;72:571-576.[Abstract/Free Full Text]
  12. Harder S, Thurmann P. Clinically important drug interactions with anticoagulants. An update. Clin Pharmacokinet 1996;30:416-444.[ISI][Medline] [Order article via Infotrieve]
  13. Stevenson HP, Archbold GP, Johnston P, Young IS, Sheridan B. Misleading serum free thyroxine results during low molecular weight heparin treatment. Clin Chem 1998;44:1002-1007.[Abstract/Free Full Text]
  14. Dasgupta A, Jacques M, Malhotra D. Diminished protein binding capacity of uremic sera for valproate following hemodialysis: role of free fatty acids and uremic compounds. Am J Nephrol 1996;16:327-333.[ISI][Medline] [Order article via Infotrieve]
  15. Sakai T, Maruyama T, Imamura H, Shimada H, Otagiri M. Mechanism of stereoselective serum binding of ketoprofen after hemodialysis. J Pharmacol Exp Ther 1996;278:786-792.[Abstract/Free Full Text]
  16. Orme CE, Harris RC. A comparison of the lipolytic and anticoagulative properties of heparin and pentosan polysulphate in the thoroughbred horse. Acta Physiol Scand 1997;159:179-185.[ISI][Medline] [Order article via Infotrieve]
  17. Dasgupta A, Crossey MJ. Elevated free fatty acid concentrations in lipemic sera reduce protein binding of valproic acid significantly more than phenytoin. Am J Med Sci 1997;313:75-79.[ISI][Medline] [Order article via Infotrieve]
  18. Williams SP, Johnson EA. Release of lipoprotein lipase and hepatic triglyceride lipase in rats by heparin and other sulphated polysaccharides. Thromb Res 1989;55:361-368.[ISI][Medline] [Order article via Infotrieve]
  19. Samama MM, Bara L, Gouin Thibault I. New data on the pharmacology of heparin and low molecular weight heparins. Drugs 1996;52(Suppl 7):8-14.
  20. Shibasaki T, Matsuda H, Ohno I, Gomi H, Nakano H, Misawa T, et al. Significance of serum lipase in patients undergoing hemodialysis. Am J Nephrol 1996;16:309-314.[ISI][Medline] [Order article via Infotrieve]
  21. Lim CF, Curtis AJ, Barlow JW, Topliss DJ, Stockigt JR. Interactions between oleic acid and drug competitors influence specific binding of thyroxine in serum. J Clin Endocrinol Metab 1991;73:1106-1110.[Abstract]
  22. Takamura N, Shinozawa S, Maruyama T, Suenaga A, Otagiri M. Effects of fatty acids on serum binding between furosemide and valproic acid. Biol Pharm Bull 1998;21:174-176.[ISI][Medline] [Order article via Infotrieve]
  23. Barre J, Didey F, Delion F, Tillement JP. Problems in therapeutic drug monitoring: free drug level monitoring. Ther Drug Monit 1988;10:133-143.[ISI][Medline] [Order article via Infotrieve]
  24. Nowak I, Shaw LM. Mycophenolic acid binding to human serum albumin: characterization and relation to pharmacodynamics. Clin Chem 1995;41:1011-1017.[Abstract/Free Full Text]
  25. Shagel L, Yu ABC. Drug distribution and protein binding. Shagel L Yu ABC eds. Applied biopharmaceutics and pharmacokinetics, 3rd ed 1992:87 Appleton & Lange Norwalk, CT. .
  26. De Smet R, Vogeleere P, Van Kaer J, Lameire N, Vanholder R. Study by means of high-performance liquid chromatography of solutes that decrease theophylline/protein binding in the serum of uremic patients. J Chromatogr A 1999;847:141-153.[ISI][Medline] [Order article via Infotrieve]
  27. De Smet R, David F, Sandra P, Van Kaer J, Lesaffer G, Dhondt A, et al. A sensitive method for the quantification of free and total p-cresol in patients with chronic renal failure. Clin Chim Acta 1998;278:1-21.[ISI][Medline] [Order article via Infotrieve]
  28. Vanholder R, De Smet R, Waterloos MA, Van Landschoot N, Vogeleere P, Hoste E, Ringoir S. Mechanisms of uremic inhibition of phagocyte reactive species production: characterization of the role of p-cresol. Kidney Int 1995;47:510-517.[ISI][Medline] [Order article via Infotrieve]
  29. Abreo K, Sella M, De Smet R, Vogeleere P, Ringoir S, Vanholder R. p-Cresol, a uremic compound, enhances the uptake of aluminum in hepatocytes. J Am Soc Nephrol 1997;8:935-942.[Abstract]
  30. Wratten ML, Tetta C, De Smet R, Neri R, Sereni L, Camussi G, Vanholder R. Uremic ultrafiltrate inhibits platelet activating factor synthesis. Blood Purif 1999;17:134-141.[ISI][Medline] [Order article via Infotrieve]
  31. Vanholder R, De Smet R, Lesaffer G. p-Cresol: a toxin revealing many neglected but relevant aspects of uraemic toxicity. Nephrol Dial Transplant 1999;14:2813-2815.[Free Full Text]
  32. Liebich HM, Wirth C, Jakober B. Analysis of polyunsaturated fatty acids in blood serum after fish oil administration. J Chromatogr 1991;572:1-9.[ISI][Medline] [Order article via Infotrieve]
  33. Dhondt A, Vanholder R, Waterloos MA, Glorieux G, De Smet R, Lameire N. Citrate anticoagulation does not correct cuprophane bioincompatibility as evaluated by the expression of leucocyte surface molecules. Nephrol Dial Transplant 1998;13:1752-1758.[Abstract/Free Full Text]
  34. Vanholder R, Van Landschoot N, De Smet R, Schoots A, Ringoir S. Drug protein binding in chronic renal failure: evaluation of nine drugs. Kidney Int 1988;33:996-1004.[ISI][Medline] [Order article via Infotrieve]
  35. Persson E, Nordenström J, Nilsson-Ehle P, Hagenfeldts L. Lipolytic and anticoagulant activities of low molecular weight fragment of heparin. Eur J of Clin Invest 1985;15:215-220.
  36. Vanholder R, Van Loo A, Dhondt AM, Ringoir S. The role of dialysis membranes in infection. Nephrol Dial Transplant 1996;11(Suppl 2):101-103.[Abstract/Free Full Text]
  37. Salvayre R, Nègre A, Radom J, Douste-Blazy L. Fluorometric assay for pancreatic lipase. Clin Chem 1986;32:1532-1536.[Abstract/Free Full Text]
  38. Hauptman JB, Jeunet FS, Hartmann D. Initial studies in humans with the novel gastrointestinal lipase inhibitor Ro 18-0647 (tetrahydrolipstatin). Am J Clin Nutr 1992;55:309S-313S.[Abstract/Free Full Text]
  39. Zhi J, Melia AT, Eggers H, Joly R, Patel IH. Review of limited systemic absorption of orlistat, a lipase inhibitor, in healthy human volunteers. J Clin Pharmacol 1995;35:1103-1108.[Abstract]
  40. Guerciolini R. Mode of action of orlistat. Int J Obesity 1997;21(Suppl 3):S12-S23.
  41. Zambon A, Schmidt I, Beisiegel U, Brunzell JD. Dimeric lipoprotein lipase is bound to triglyceride-rich plasma lipoproteins. J Lipid Res 1996;37:2394-2404.[Abstract]
  42. Dubé LM, Davies RF, Beanlands DS, Mousseau N, Beaudoin N, Chan B, et al. Dissociation of authentic and artifactual effect of circulation heparin on drug protein binding. Biopharm Drug Dispos 1989;10:55-68.[ISI][Medline] [Order article via Infotrieve]
  43. Brown JE, Kitchell BB, Bjornsson TD, Shand DG, Durham NC. The artifactual nature of heparin-induced drug protein-binding alterations. Clin Pharmacol Ther 1981;30:636-642.[ISI][Medline] [Order article via Infotrieve]
  44. Lohman J, Hooymans P, Koten M, Verhey M, Merkus F. Effect of heparin on digitoxin protein binding. Clin Pharmacol Ther 1985;37:55-60.[ISI][Medline] [Order article via Infotrieve]
  45. Rierersma RA, Russell DC, Oliver MF. Heparin-induced lipolysis, an exaggerated risk [Letter]. Lancet 1981;29:71.
  46. Zambon A, Hashimoto SI, Brunzell JD. Analysis of techniques to obtain plasma for measurement of levels of free fatty acids. J Lipid Res 1993;34:1021-1028.[Abstract]
  47. Krebs M, Sting H, Nowotny P, Weghuber D, Bischof M, Waldhäusl W, Roden M. Prevention of in vivo lipolysis by tetrahydrolipstatin. Clin Chem 2000;46:950-958.[Abstract/Free Full Text]



The following articles in journals at HighWire Press have cited this article:


Home page
 Nephrol Dial TransplantHome page
T. Nishio, N. Takamura, R. Nishii, J. Tokunaga, M. Yoshimoto, and K. Kawai
Influences of haemodialysis on the binding sites of human serum albumin: possibility of an efficacious administration plan using binding inhibition
Nephrol. Dial. Transplant., July 1, 2008; 23(7): 2304 - 2310.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
R. De Smet, J. Van Kaer, B. Van Vlem, A. De Cubber, P. Brunet, N. Lameire, and R. Vanholder
Toxicity of Free p-Cresol: A Prospective and Cross-Sectional Analysis
Clin. Chem., March 1, 2003; 49(3): 470 - 478.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by De Smet, R.
Right arrow Articles by Vanholder, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Smet, R.
Right arrow Articles by Vanholder, R.
Related Collections
Right arrow Drug Monitoring and Toxicology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS