- Email: [email protected]
Falsely elevated cyclosporin and tacrolimus concentrations over prolonged periods of time due to reversible adsorption to central venous catheters
Clinica Chimica Acta 433 (2014) 62–68
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Falsely elevated cyclosporin and tacrolimus concentrations over prolonged periods of time due to reversible adsorption to central venous catheters Charlotte Hacker a, Mareike Verbeek b, Heike Schneider a, Werner Steimer a,⁎ a b
Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Germany III. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar, Technische Universität München, Germany
a r t i c l e
i n f o
Article history: Received 19 December 2013 Received in revised form 12 February 2014 Accepted 27 February 2014 Available online 11 March 2014 Keywords: Therapeutic drug monitoring Cyclosporine A Tacrolimus Catheters Adsorption Blood sampling
a b s t r a c t Falsely elevated concentrations of immunosuppressants can be caused by reversible adsorption to central venous catheter (CVC) systems. If undetected, this may lead to dose reduction resulting in underdosage which may even entail graft-versus-host disease or organ rejection. We analyzed the adsorption and release for cyclosporine A (CsA) and tacrolimus (Tac) in vitro and in vivo. Four types of CVCs were examined in vitro: two made from polyurethane (PU), one from silicone and one from PU with an incorporated silver ion-based antimicrobial agent. All 26 CVCs analyzed in vitro showed significant reversible adsorption of CsA (n = 13; p = 0.001) and Tac (n = 13; p = 0.001, Wilcoxon signed rank test). Immediately after infusing the drugs, the mean concentrations of 6420 ng/mL of CsA and 250 ng/mL of Tac were measured. Flushing with NaCl lowered the drug release. Besides, blood samples of fifteen patients were taken simultaneously from all lumina of the CVC and via venipuncture. The samples from contaminated lumina showed the mean elevations by a factor of 11 for CsA (n = 12) and 89 for Tac (n = 3). Blood sampling for immunosuppressant monitoring should thus never be performed from lumina previously used for infusing the drug even after prolonged periods of time and extensive rinsing. © 2014 Elsevier B.V. All rights reserved.
1. Introduction For over 30 years, cyclosporine A (CsA) and tacrolimus (Tac; FK506) have been cornerstones in immunosuppressive therapy after solid organ as well as stem cell transplantation [1,2]. Both CsA, a lipophilic cyclic peptide, and Tac, a macrolide, inhibit calcineurin resulting in a reduction of lymphocytes [3]. For patients receiving these drugs, it is crucial to maintain the correct drug concentration within the narrow therapeutic range. Inappropriate levels can result in renal toxicity, hypertension, severe infections, transplant rejection or graft-versushost disease [4–6].
Abbreviations: CsA, cyclosporine A; CVC, central venous catheter; FFP, Fresh Frozen Plasma; P0, first 2 mL portion of the mimicked in vitro blood sampling (discarded); P1– P5, consecutive 2 mL portions of the mimicked in vitro blood sampling; PU, polyurethane; PU-AH-CVC, PU-In-CVC, silver-Vy-CVC, silicone-Vy-CVC, cf. Table 1; Tac, tacrolimus; TTP/ HUS, thrombotic thrombocytopenic purpura/hemolytic uremic syndrome. ⁎ Corresponding author at: Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Ismaningerstr. 22, D-81675 Munich, Germany. Tel.: +49 89 41404754; fax: +49 89 41404875. E-mail address: [email protected] (W. Steimer).
http://dx.doi.org/10.1016/j.cca.2014.02.031 0009-8981/© 2014 Elsevier B.V. All rights reserved.
Erroneously high immunosuppressant concentrations have been reported repeatedly to be caused by reversible adsorption of the drugs to different intravenous catheter systems [7–20]. This can be especially dangerous since falsely elevated levels may fall within the therapeutic range whereas in reality, the patient is in risky underdosage. Graft-versus-host disease or organ rejection could be a life-threatening consequence [4,5]. A case in our house (cf. 2.2.2, patient A) turned our special attention to this pitfall: two days after being switched to oral Tac administration, a blood sample, accidentally taken from the lumen of the central venous catheter (CVC) previously used for infusing Tac, yielded a toxic concentration of 86 μg/L of Tac. A control sample obtained by venipuncture, though, showed a value of 2.9 μg/L of Tac. This remarkable discrepancy prompted us to perform a thorough in vitro and in vivo study on this effect. To systematically examine this adsorption and release effect for CsA and Tac in vitro, we studied some of the CVC systems most widely used in Europe and the USA. The 26 analyzed CVCs consisted of different materials, namely polyurethane (PU), silicone and PU with an incorporated silver ion-based antimicrobial agent. To substantiate the findings from our extensive in vitro experiments, a prospective in vivo study was conducted with fifteen patients receiving CsA or Tac via a CVC after stem cell transplantation.
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
63
Table 1 Central venous catheters (CVCs) used in in vitro and in vivo experiments. Abbreviation
Name
Material
Brand
Number of lumina
Inner diameter of contaminated lumen (mm)
Total length of contaminated umen (cm)
PU-AH-CVC
Polyurethane
Arrow-Howes™, (Kernen, Germany) Intra, (Rehlingen-Siersburg, Germany) Vygon, (Aachen, Germany)
4
1.52
46.6
PU-In-CVC Silver-Vy-CVC
Quad-Lumen Central Venous Catheterization Set Trilucath Multicath 4 Expert
3 4
0.80 1.25
36.5 35.8
Silicone-Vy-CVC
Lifecath Apheresis Plus
Vygon (Aachen, Germany)
3
1.9
50.0
Polyurethane Polyurethane incorporated with silver ion-based antimicrobial agent Silicone
2. Material and methods 2.1. In vitro experiments 2.1.1. Basic study Four types of catheters, made of three different materials, were examined in vitro: Arrow-Howes™ “Quad-Lumen Central Venous Catheterization Set” (n = 6, PU-AH-CVC), Intra “Trilucath” (n = 2, PU-INCVC), Vygon “Lifecath Apheresis Plus” (n = 2, silicone-Vy-CVC) and Vygon “Multicath 4 Expert” (n = 2, PU with an incorporated silver ion-based antimicrobial agent, silver-Vy-CVC; Table 1). One of the following doses was infused into one lumen of a CVC to mimic a therapeutic in vivo application of the drug: 250 mg of CsA (n = 10; Sandimmun, Novartis Pharma) in 100 mL NaCl solution (0.9%) (2.5 g/L of CsA) over 6 h or 2 mg of Tac (n = 10; Prograf, Astellas) in 50 mL NaCl (40 mg/L of Tac) over 22 h. Right after infusing the drug, blood sampling was mimicked in all four lumina with 4 mL of Fresh Frozen Plasma (FFP) each, always discarding the first 2 mL. When blood sampling was simulated, special attention was given to avoid cross contamination between the different lumina, particularly at the exits of the individual ports at the distal end of the catheter. Therefore, FFP was infused retrogradely with a syringe and collected at the distal end without getting in contact with the exits of the other lumina. The direction of sampling FFP had no significant effect on the released drug concentrations. The CVC then was recurrently rinsed with NaCl (0.9%), the first volume always being 10 mL, followed by multiple flushes of 1 L of NaCl. Each rinsing was followed by mimicking a blood withdrawal with 12 mL of FFP. To thoroughly analyze the withdrawal, the 12 mL was collected in six separate 2 mL doses, always discarding the first 2 mL. The lumina of the CVCs were not blocked with any fluid at any time. 2.1.2. Water bath In eight additional cases, infusing the drug was performed in a water bath at 37 °C to create a more realistic setting. All four types of catheters were once infused with CsA (n = 4) and once with Tac (n = 4), each time using a new catheter. 2.1.3. EDTA blood In addition to the 20 experiments conducted with FFP only, six PU-AH-CVCs were used for further investigations. This time, some mimicked blood sampling was performed with uncontaminated whole blood taken into EDTA containing tubes in order to study whether the whole blood and FFP behave similarly: Four CVCs were investigated in the usual manner, except for mimicking the blood collection alternating with FFP and EDTA blood. These experiments were performed with both drugs once at room temperature and once with the drug being infused in a water bath at 37 °C. To study the possible adsorption effects with the drugs being in the therapeutic range, two PU-AH-CVCs were put into EDTA blood for 24 h with a concentration of 125 μg/L of CsA (n = 1) and 10 μg/L of Tac (n = 1) respectively. Blood sampling was mimicked with EDTA blood after rinsing the CVCs with 10 mL of NaCl.
2.1.4. Fat emulsion In order to examine the effect of oily fluid on the release of the drugs, two PU-AH-CVCs were infused with CsA and Tac respectively. After flushing them with 1.01 L of NaCl, 100 mL of fat emulsion (Clinoleic 20%, Baxter) was infused. Blood sampling was then mimicked with 3 × 2 mL of EDTA blood, discarding the first 2 mL. Another 1.01 L of NaCl and 50 mL of fat emulsion were infused into the CVC previously contaminated with Tac. Afterwards, blood sampling was performed in the same manner as just described. 2.2. Patients We conducted a monocentric, prospective pilot study, approved by the institution's responsible committee and in accordance with the current revision of the Helsinki Declaration. Fifteen in-patients after allogenic stem cell transplantation aged 22 to 68 were included after giving informed consent (patients 1–15; Table 2). Twelve of them were receiving CsA and three were treated with Tac. In all cases, blood withdrawal was performed at all lumina of the CVC, each time discarding the first 5 mL of blood as it is a clinical practice in our house. Additionally, a blood sample was taken simultaneously via peripheral venipuncture each time. All samples were collected into EDTA containing tubes of 2.7 mL. 2.2.1. Main study In thirteen cases (patients 1–13, Table 2), the drug had been administered via a PU-AH-CVC (cf. 2.1.1, in vitro experiments). Doses of 75 mg to 750 mg of CsA and 1.0 mg to 2.5 mg of Tac had been administered over 7 to 22 days. The drug was received each day for 22 h, followed by flushing with 50 mL of NaCl for 2 h. Trough blood samples for our study were taken after 2 h of rinsing. 2.2.2. Special cases The only distinction in patient 14 was the fact that drug administration had been switched to oral medication three days prior to our blood sampling. Patient 15 had received only a single dose of Tac via a silver-Vy-CVC (cf. 2.1.1, in vitro experiments). Due to high Tac concentrations in the blood, the infusion with a concentration of 60 mg/L of Tac was stopped after 40 mL had been administered. For the following three days, Tac administration was stopped completely. On the third day after withdrawing drug administration, blood samples were taken for our study. Patient A had received Tac for 27 days via a three-lumen CVC made from polyurethane by Arrow-Howes™. Blood samples were taken on days 7, 8 and 18 after stopping Tac administration through the CVC. 2.3. Measuring All in vivo and in vitro samples were measured in duplicate by a LC/MS/MS immunosuppressant method modified from Koal et al. [21] and Ceglarek et al. [22] using a two dimensional HPLC system coupled to an ABI 3000 tandem mass spectrometer. Mobile phase chemicals
64
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
were purchased from Sigma-Aldrich, controls from Chromsystems Instruments & Chemicals and standards from Sigma-Aldrich, Chromsystems Instruments & Chemicals and Recipe Chemicals + Instruments. Samples were precipitated by a ZnSO4/methanol mix containing the internal standards. The analytes were captured by a Cyclone® HTLC column and separated by a phenyl-hexyl column before being analyzed in the mass spectrometer. Accuracy was ensured by continuously participating in proficiency testing schemes. If not measured within the first 6 h after it was obtained, each sample was kept at 8 °C and analyzed within a maximum of 72 h. For longtime storage, all samples from the in vitro and in vivo experiments were frozen at −25 °C. 2.4. Calculations All in vitro data was normalized with regard to the total inner surface of the analyzed lumen using PU-AH-CVC as a reference. Statistical analyses of the in vitro experiments were performed using the Wilcoxon signed rank test. A two sided level of significance of 5% was used. 3. Results 3.1. In vitro experiments Regardless of the material, every CVC which had been in contact with high concentrations of the drug used in the infusion showed significant evidence of reversible adsorption for CsA (n = 13, p = 0.001) and Tac (n = 13, p = 0.001, Wilcoxon signed rank test). 3.1.1. Main study Immediately after infusing the drugs and after discarding only the first 2 mL of FFP, the mean concentrations of 6420 μg/L (range: 260–17,700 μg/L) of CsA and 250 μg/L (range: 93–395 μg/L) of Tac were measured. The highest concentrations obtained from the drug release in the in vitro experiments were about 100 times lower than a usual therapeutic concentration (cf. 2.1.1) of the drug solution infused into the patient. After the first mimicked blood sampling, the CVC was rinsed with 10 mL of NaCl followed by sampling 6 portions of 2 mL of FFP, discarding the first 2 mL. The mean concentrations measured in the first portion yielded 132 μg/L for CsA and 55 μg/L for Tac and were lowered to 115 μg/L for CsA and 36 μg/L for Tac in the last portion. Rinsing the lumen with five more liters of NaCl resulted in an average reduction of the mean value by 26% for CsA and 32% for Tac per liter NaCl after each time discarding the first 6 mL of FFP. However, in some cases, even extensive rinsing with volumes as much as 24.01 L of NaCl still produced concentrations of 4.4 μg/L of Tac after discarding the first 6 mL. The five consecutive portions obtained at each mimicked blood sampling showed a mean decline of 12% for CsA and 31% for Tac from portion 1 to portion 5. 3.1.2. Different materials All three materials did adsorb and release the drugs. This effect was strongest in CVCs made from silicone compared to PU and silver throughout the entire study (Fig. 1). 3.1.3. Water bath Performing the infusion of the drug in a water bath at 37 °C led to higher drug levels as opposed to conducting it at room temperature (Fig. 2). 3.1.4. EDTA blood Performing the blood sampling with EDTA blood instead of FFP showed similar drug concentrations (Fig. 3).
The CVCs placed in EDTA blood with comparably low concentrations within the therapeutic range of CsA and Tac did not show any relevant reversible drug adsorption when blood sampling was mimicked. 3.1.5. Fat emulsion After rinsing the PU-AH-CVCs with 1.01 L of NaCl and 100 mL of fat emulsion, the first 2 mL of EDTA blood samples resulted in concentrations of 17 μg/L of CsA and 16.8 μg/L of Tac. The following 2 mL yielded 11 μg/L of CsA and 7.7 μg/L of Tac. After another 1.01 L of NaCl and 50 mL of fat emulsion, 4.9 μg/L and 2.7 μg/L of Tac were measured. 3.1.6. Cross contamination For all CVCs, regardless of the material, the phenomenon of adsorption and release was only observed for the lumen utilized for immunosuppressant infusion. All samples obtained in vitro from the uncontaminated lumina did not show any detectable drug concentration. Thus, no cross contamination was observed in vitro. 3.2. Patients For every patient, remarkably elevated concentrations of CsA or Tac were found in the lumen previously used for infusing the drug. 3.2.1. Main study In the eleven cases of CsA (patients 1–11; Table 2), the drug level obtained from the contaminated lumen was increased by a mean factor of 11.1 (range 6.7–22.0) compared to the one acquired via venipuncture. The CsA concentrations from the samples obtained by venipuncture yielded a mean level of 191 μg/L (range: 96–249 μg/L) whereas the drug levels in the samples from the contaminated lumen resulted in a mean CsA concentration of 2120 μg/L (range 1460–4410 μg/L). The blood samples from the lumina not used for infusing CsA showed comparable results (±10%) to the ones obtained by venipuncture. Patients 12 and 13 who had received Tac showed drug concentrations in the lumen previously used for infusion raised by a factor of 150 and 116 respectively (Table 2). The Tac level measured in the blood sample obtained by venipuncture resulted in 5.1 μg/L and 6.1 μg/L while the drug concentration in the sample from the contaminated lumen yielded 767 μg/L and 705 μg/L. In contrast to the results from all patients treated with CsA as well as patients 13 and 15, the blood sample of patient 12, from a lumen reportedly not used for infusing Tac previously, contained a 1.5 times elevated concentration of 7.9 μg/L of Tac compared to 5.1 μg/L obtained via venipuncture. 3.2.2. Special cases Patient 14, who had been switched to oral medication three days prior to blood sampling, still showed an increase of the CsA concentration from the contaminated lumen (895 μg/L) by the factor of 3.0 compared to venipuncture (296 μg/L; Table 2). In patient 15, the concentration from the lumen used only once for Tac infusion was elevated by a factor of 2.9. The sample received by venipuncture came to 12.9 μg/L and the one from the contaminated lumen resulted in 37 μg/L. Patient A was administered with Tac through a CVC for 26 days. At some point between days 22 and 25, the patient had most likely received an accidental overdose of Tac resulting in toxic concentrations of up to 111 μg/L measured on day 26. Tac medication was thus stopped completely, restarting with oral Tac five days after the withdrawal. Two days later, another toxic concentration of 86 μg/L of Tac was determined. However, it turned out that this time the sample had accidentally been taken from the lumen previously used for infusing Tac, since a control sample obtained via venipuncture yielded only 2.9 μg/L. Two more blood withdrawals taken 8 and 18 days after the last i.v. administration showed elevated blood concentrations in the contaminated lumen both by a factor of 37 compared to venipuncture. Similar to patient 12, the
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
65
Table 2 All data from patients.
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 b Patient 15 b Patient A b
a b c
Drug
Type of central venous catheter (CVC)
Venipuncture (μg/L)
Contaminated lumen (μg/L)
Lumen 2 (μg/L)
Lumen 3 (μg/L)
Lumen 4 (μg/L)
Duration of i.v. drug application (days)
Current i.v. drug dosage (mg)
Time since last i.v. administration (days)
CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA Tac Tac CsA Tac Tac
PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC Silver-Vy-CVC PU-AH-CVC c
235 96 249 144 172 138 192 248 199 198 229 5.1 6.1 296 12.9 2.9 4.6 2.6
2490 2110 4410 2255 1615 1460 1755 1655 1990 1635 1930 767 705 895 37 86 171 96
231 97 257 138 172 147 194 254 194 201 233 4.9 6.0 292 13.5
a
277 102
8 7 22 10 9 13 7 16 14 15 16 21 11 22 1 26
300 480 200 750 450 75 200 400 300 500 200 2.0 1.75 – – – – –
– – – – – – – – – – – – – 3 3 7 8 18
a
99 275 143 171 a
a
188 265 195 201
190 249 193 200 238 5.5 6.2 286 13.6 – – –
a
7.6 6.4 295 13.4 a
5.9 6.3
a
147 172
3.3 4.4
Blood sampling not performed. Patient had been switched to oral medication. Arrow-Howes™ polyurethane CVC with three lumina.
Fig. 1. Drug concentrations from 12 CVCs: one set of 1 silicone-Vy-CVC, 1 silver-Vy-CVC and 3 PU CVCs (2 PU-AH-CVCs, 1 PU-In-CVC) infused with CsA and one set with Tac. P3 of blood sampling mimicked with FFP after rinsing multiple times with NaCl. Average line for PU enclosed. Part of the figure zoomed in and is displayed in the upper right corner.
66
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
Fig. 2. Drug concentrations from 20 CVCs: for CsA and Tac, 1 silicone-Vy-CVC, 1 silver-Vy-CVC, 1 PU-AH-CVC and 1 PU-In-CVC were infused in a water bath at 37 °C. 1 silicone-Vy-CVC, 1 silver-Vy-CVC, 3 PU-AH-CVCs and 1 PU-In-CVC per drug were infused at room temperature. P3 of blood sampling mimicked with FFP after rinsing multiple times with NaCl. Average lines enclosed. Part of the figure zoomed in and is displayed in the upper right corner.
samples for patient A taken from the lumina previously not used for Tac administration did not fall within the range of ±10% of the values determined by venipuncture (Table 2). 4. Discussion Our results from a systematic and comprehensive in vitro and in vivo study are in concordance with various reported cases [7–13] as well as smaller in vivo [14,15] and in vitro [16–18] experiments for CsA of spuriously elevated drug concentrations due to reversible adsorption to catheter materials. Mimicking blood sampling with whole EDTA blood showed similar results as compared to FFP (Fig. 3). Moreover, performing the experiments in a water bath only raised the drug concentrations to a small degree. This verifies that the setup of the main in vitro experiments was close to the in vivo conditions. Especially the first blood sample of each in vitro experiment as well as the concentration from the contaminated lumen of the patients yielded a multiple of the concentration of the reference range (i.e. 100–400 μg/L for CsA and 5–20 μg/L for Tac [23,24]). In clinical practice, many blood samples accidentally taken from the contaminated lumen lead to alarmingly high measurements resulting in further investigation. Thus, blood sampling usually is repeated correctly and no harm is done to the patient. The in vitro results show that rinsing the catheter with NaCl leads to a reduction of the drug concentration. However, volumes of up to 24.01 L of NaCl still resulted in detectable drug release. Infusing fat emulsion,
as being done quite often in critically ill patients with CVCs, did not put an end to releasing the drugs, either. Moreover, patients 14, 15 and A reveal that even a long time after switching to oral administration of CsA or Tac, the lumen previously used for administering the drug remains contaminated. This emphasizes how blood sampling from a lumen ever used for infusing the immunosuppressant needs to be avoided at all times, even if the patient is already switched to oral medication and different fluids have been infused through the contaminated lumen. Especially if a long time has passed since the last i.v. dose was administered, it is absolutely critical to still be aware of this phenomenon. In this case, the spurious elevation could be of minor degree resulting in concentrations within or only slightly above the therapeutic range over an extended period of time. Therefore, the falsely elevated values may remain unrecognized which may lead to erroneously reduced immunosuppressant dosage resulting in undiscovered underdosage. This would put patients after stem cell transplantation at risk for graft-versus-host disease [5] and patients after solid organ transplantation could even experience organ rejection [4]. Analyzing four patients receiving CsA via CVCs made from silicone led Leson et al. to conclude that a correction factor for the elevation from the contaminated lumen could be used [10]. In contrast to their data, both our in vitro and in vivo experiments show that the adsorption and release of the drugs are of varying degree and cannot be predicted precisely. Our data suggests that there is no fix correlation between the extent of the elevation and the dosage as well as the duration of the drug application.
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
67
Fig. 3. Drug concentrations from 4 PU-AH-CVCs, 2 of them infused in a water bath at 37 °C, one with CsA and one with Tac. P3 of blood sampling mimicked with alternating FFP and EDTA blood after rinsing multiple times with NaCl. Exponential fit line enclosed.
In vitro experiments by Carreras et al. show that CVCs made from silicone adsorb more CsA than CVCs made from PU [18]. In our experiments, we were able to demonstrate that CVCs made from silicone also release greater amounts of CsA and Tac than the ones made from PU or PU incorporated with silver. Patient 15 still showed an elevation by a factor of 2.9 although Tac had only been administered once through that lumen three days prior to sampling. This is the first reported case of a falsely elevated Tac concentration due to reversible adsorption of the drug at a CVC with an incorporated silver ion-based antimicrobial agent. To confirm this finding, further in vivo studies should be conducted. The question whether or not blood withdrawal for therapeutic drug monitoring of CsA and Tac should only be performed peripherally has been discussed in several publications [12–14,25]. Senner et al. analyzed 71 paired CsA blood samples taken peripherally and from the CsAuncontaminated lumen of a double-lumen silastic CVC [25]. They did not find a significant difference between the two sampling sites (paired Student t test; p = 0.13) and concluded that blood sampling for CsA monitoring can be performed from the uncontaminated lumen of the CVC. Claviez et al. came to the same recommendation regarding CsA [12]. Grouzmann et al., on the other hand, suggest that the blood samples for Tac monitoring should be performed via venipuncture [13]. Our in vivo studies showed no cross contamination for CsA. Therefore, we fall into line with the authors concluding that blood sampling for therapeutic drug monitoring from uncontaminated lumina is acceptable for CsA.
However, patient 13 and patient A showed alarming elevations by factors of 1.3 to 2.4 of the Tac concentration in lumina that had reportedly not been used for immunosuppressive infusion. Apart from incorrect sampling, possible reasons for these elevations could be substance properties of Tac or cross contamination at the exits of the individual ports at the distal end. Further experiments should be conducted in order to examine these correlations. For now, we strongly recommend using peripheral venipuncture to acquire blood samples for Tac monitoring. Thrombotic thrombocytopenic purpura/hemolytic uremic syndrome (TTP/HUS) is a rare but very severe complication after hematopoietic stem cell or solid organ transplantation which can be induced by CsA or Tac [26]. Therefore, discontinuation or sometimes dose reduction of the calcineurin inhibitors is the most important treatment against this condition [27,28]. Our findings suggest that the adsorption of CsA or Tac to the CVC is reversible not only when blood is withdrawn but also when fluid is infused through the contaminated lumen. This could possibly lead to small amounts of the drug still being washed into the body with other fluids long after the drug was discontinued. If 1 L of fluid was to be infused through the contaminated lumen of the patient and assuming that the drug was released with the same mean concentration (cf. 3.1.1) measured in vitro after discarding only 2 mL of FFP, a total amount of 6.42 mg of CsA or 0.25 mg of Tac could unintentionally be infused into the patient's body. Depending on the patient's course of treatment, it should be left to the physician's discretion whether or not to tolerate these comparable small amounts of the drug. However, there clearly is a need to further investigate this possible phenomenon. If the
68
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
patient is suffering from TTP/HUS and CsA or Tac is to be withdrawn completely, the lumen previously used for infusing CsA or Tac should not be utilized any longer or the CVC should be changed. Many CVC practice manuals and guidelines still lack these special adsorption and release aspects [29–31] and the package inserts of Tac and CsA only mention PVC incompatibility with potential extraction of phthalates regarding drug preparation for infusion [32–35] but not in connection with blood sampling for therapeutic drug monitoring. It is crucial to make sure that the medical staff performing blood withdrawal on patients with CVCs is aware of the reversible adsorption of drugs to CVCs and of the importance of the sampling technique. 5. Conclusion Catheter lumina used for immunosuppressant infusion should never be utilized for blood sampling even after rinsing with high volumes of other fluids and discarding a limited volume prior to blood sampling. As cross contamination between different lumina may occur for Tac, venipuncture certainly is the safest way to obtain correct blood samples for therapeutic drug monitoring. Acknowledgments We are indebted to C. Mueller and the members of the drug monitoring laboratory for the technical support as well as B. Haller for the statistical advice. We also thank the staff of the bone marrow transplant unit for the assistance during the blood sampling as well as C. Voelkl for performing the drug concentration analyses of the in vivo samples. References [1] Storb R, Deeg HJ, Whitehead J, et al. Methotrexate and cyclosporine compared with cyclosporine alone for prophylaxis of acute graft versus host-disease after marrow transplantation for leukemia. N Engl J Med 1986;314:729–35. [2] Starzl TE, Todo S, Fung J, Demetris AJ, Venkataramman R, Jain A. Fk-506 for liver, kidney, and pancreas transplantation. Lancet 1989;2:1000–4. [3] Wiederrecht G, Lam E, Hung S, Martin M, Sigal N. The mechanism of action of Fk-506 and cyclosporine-A. In: Allison AC, Lafferty KJ, Fliri H, editors. Immunosuppressive and antiinflammatory drugs. New York: New York Acad Sciences; 1993. p. 9–19. [4] Geissler EK, Schlitt HJ. Immunosuppression for liver transplantation. Gut 2009;58:452–63. [5] Ram R, Storer B, Mielcarek M, et al. Association between calcineurin inhibitor blood concentrations and outcomes after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2012;18:414–22. [6] Randomised trial comparing tacrolimus (Fk506) and cyclosporin in prevention of liver allograft rejection. European Fk506 Multicentre Liver Study Group. Lancet 1994;344:423–8. [7] Shaefer MS, Collier DS, Haven MC, et al. Falsely elevated Fk-506 levels caused by sampling through central venous catheters. Transplantation 1993;56:475–6. [8] Duffner U, Bergstraesser E, Sauter S, Bertz H, Niemeyer C. Spuriously raised cyclosporin concentrations drawn through polyurethane central venous catheter. Lancet 1998;352:1442. [9] Lorenz RG, Garrett N, Turk JW, Scott MG. Problems with therapeutic monitoring of cyclosporine using silicone central venous line samples. Transplantation 1991;52:1109–10. [10] Leson CL, Bryson SM, Giesbrecht EE, Saunders EF. Therapeutic monitoring of cyclosporine following pediatric bone-marrow transplantation — problems with sampling from silicone central venous lines. DICP Ann Pharmacol 1989;23:300–3. [11] Blifeld C, Ettenger RB. Measurement of cyclosporine levels in samples obtained from peripheral sites and indwelling lines. N Engl J Med 1987;317:509.
[12] Claviez A, Glass B, Dreger P, Suttorp M. Elevated blood drug levels obtained from indwelling silicon catheters during oral cyclosporine A administration. Bone Marrow Transplant 2002;29:535–6. [13] Grouzmann E, Buclin T, Biollaz J. Misleading tacrolimus concentration value in blood taken from a catheter used for tacrolimus administration. Am J Health Syst Pharm 2008;65:226–8. [14] Busca A, Miniero R, Vassallo E, Leone L, Oddenino O, Madon E. Monitoring of cyclosporine blood-levels from central venous lines — a misleading assay. Ther Drug Monit 1994;16:71–4. [15] de Witte T, Hoitsma A, Manoiu MM, Janssen J. Monitoring of cyclosporine during continuous intravenous administration. Bone Marrow Transplant 1986;1:147–50. [16] Fitzsimmons WE, Ponzillo JJ. Cyclosporine concentration sampling from an in vitro central venous catheter system. Transplantation 1988;45:1158–9. [17] Shibata N, Ikuno Y, Tsubakimoto Y, et al. Adsorption and pharmacokinetics of cyclosporin A in relation to mode of infusion in bone marrow transplant patients. Bone Marrow Transplant 2000;25:633–8. [18] Carreras E, Lozano M, Deulofeu R, Roman S, Granena A, Rozman C. Influence of different indwelling lines on the measurement of blood cyclosporine-A levels. Bone Marrow Transplant 1988;3:637–9. [19] Hacker C, Menzel H, Steimer W. Spuriously high levels of immunosuppressants due to reversible adsorption to central venous catheters [abstract]. Ther Drug Monit 2011;33:543–4. [20] Bleyzac N. On the importance of blood sampling for ciclosporin pharmacokinetic studies. Br J Clin Pharmacol 2013;75:869–70. [21] Koal T, Burhenne H, Romling R, Svoboda M, Resch K, Kaever V. Quantification of antiretroviral drugs in dried blood spot samples by means of liquid chromatography/ tandem mass spectrometry. Rapid Commun Mass Spectrom 2005;19:2995–3001. [22] Ceglarek U, Lembcke J, Fiedler GM, et al. Rapid simultaneous quantification of immunosuppressants in transplant patients by turbulent flow chromatography combined with tandem mass spectrometry. Clin Chim Acta 2004;346:181–90. [23] Uptodate: cyclosporine (systemic): drug information. http://www.uptodate.com/ contents/cyclosporine-systemic-drug-information?source=search_result&search= ciclosporin+eference+range&selectedTitle=4~150#F8012570 [accessed on December 14, 2013]. [24] Uptodate: tacrolimus (systemic): drug information. http://www.uptodate.com/ contents/tacrolimus-systemic-drug-information?source=search_result&search= tacrolimus+reference+range&selectedTitle=2~150#F9564239 [accessed on December 14, 2013]. [25] Senner AM, Johnston K, McLachlan AJ. A comparison of peripheral and centrally collected cyclosporine a blood levels in pediatric patients undergoing stem cell transplant. Oncol Nurs Forum 2005;32:73–7. [26] Dlott JS, Danielson CF, Blue-Hnidy DE, McCarthy LJ. Drug-induced thrombotic thrombocytopenic purpura/hemolytic uremic syndrome: a concise review. Ther Apher Dial 2004;8:102–11. [27] Ho VT, Cutler C, Carter S, et al. Blood and marrow transplant clinical trials network toxicity committee consensus summary: thrombotic microangiopathy after hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:571–5. [28] Agarwal A, Mauer SM, Matas AJ, Nath KA. Recurrent hemolytic-uremic syndrome in an adult renal-allograft recipient — current concepts and management. J Am Soc Nephrol 1995;6:1160–9. [29] Bishop L, Dougherty L, Bodenham A, et al. Guidelines on the insertion and management of central venous access devices in adults. Int J Lab Hematol 2007;29:261–78. [30] Moureau NL. Drawing blood through a central venous catheter. Nurs (Lond) 2004;34:28. [31] Johns Hopkins Medicine. Interdisciplinary Clinical Practice Manual, Infection Control, Vascular Access Device (Vad) Policy, Adult, Ifc035, Appendix F. http://www. hopkinsmedicine.org/armstrong_institute/_files/clabsi_toolkit/vad_appx/HF_Short_ Term_Central_Venous_Catheter.pdf [accessed on December 14, 2013]. [32] Astellas Prograf U.S. physician prescribing information. https://www.astellas.us/ docs/prograf.pdf [accessed on December 14, 2013]. [33] Novartis Pharma US sandimmune prescribing information. http://www.pharma.us. novartis.com/cs/www.pharma.us.novartis.com/product/pi/pdf/sandimmune.pdf [accessed on December 14, 2013]. [34] Fachinformation Sandimmun® 50 Mg/Ml Konzentrat Zur Herstellung Einer Infusionslösung Novartis Pharma. Revised September 2011. Novartis Pharma Gmbh 90327 Nürnberg. License Number 3123.00.00. [35] Fachinformation Prograf 5 Mg/Ml Konzentrat Astellas. Revised September 2013. Astellas Pharma Gmbh 80971 München. License Number 41954.00.01.
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Falsely elevated cyclosporin and tacrolimus concentrations over prolonged periods of time due to reversible adsorption to central venous catheters Charlotte Hacker a, Mareike Verbeek b, Heike Schneider a, Werner Steimer a,⁎ a b
Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Germany III. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar, Technische Universität München, Germany
a r t i c l e
i n f o
Article history: Received 19 December 2013 Received in revised form 12 February 2014 Accepted 27 February 2014 Available online 11 March 2014 Keywords: Therapeutic drug monitoring Cyclosporine A Tacrolimus Catheters Adsorption Blood sampling
a b s t r a c t Falsely elevated concentrations of immunosuppressants can be caused by reversible adsorption to central venous catheter (CVC) systems. If undetected, this may lead to dose reduction resulting in underdosage which may even entail graft-versus-host disease or organ rejection. We analyzed the adsorption and release for cyclosporine A (CsA) and tacrolimus (Tac) in vitro and in vivo. Four types of CVCs were examined in vitro: two made from polyurethane (PU), one from silicone and one from PU with an incorporated silver ion-based antimicrobial agent. All 26 CVCs analyzed in vitro showed significant reversible adsorption of CsA (n = 13; p = 0.001) and Tac (n = 13; p = 0.001, Wilcoxon signed rank test). Immediately after infusing the drugs, the mean concentrations of 6420 ng/mL of CsA and 250 ng/mL of Tac were measured. Flushing with NaCl lowered the drug release. Besides, blood samples of fifteen patients were taken simultaneously from all lumina of the CVC and via venipuncture. The samples from contaminated lumina showed the mean elevations by a factor of 11 for CsA (n = 12) and 89 for Tac (n = 3). Blood sampling for immunosuppressant monitoring should thus never be performed from lumina previously used for infusing the drug even after prolonged periods of time and extensive rinsing. © 2014 Elsevier B.V. All rights reserved.
1. Introduction For over 30 years, cyclosporine A (CsA) and tacrolimus (Tac; FK506) have been cornerstones in immunosuppressive therapy after solid organ as well as stem cell transplantation [1,2]. Both CsA, a lipophilic cyclic peptide, and Tac, a macrolide, inhibit calcineurin resulting in a reduction of lymphocytes [3]. For patients receiving these drugs, it is crucial to maintain the correct drug concentration within the narrow therapeutic range. Inappropriate levels can result in renal toxicity, hypertension, severe infections, transplant rejection or graft-versushost disease [4–6].
Abbreviations: CsA, cyclosporine A; CVC, central venous catheter; FFP, Fresh Frozen Plasma; P0, first 2 mL portion of the mimicked in vitro blood sampling (discarded); P1– P5, consecutive 2 mL portions of the mimicked in vitro blood sampling; PU, polyurethane; PU-AH-CVC, PU-In-CVC, silver-Vy-CVC, silicone-Vy-CVC, cf. Table 1; Tac, tacrolimus; TTP/ HUS, thrombotic thrombocytopenic purpura/hemolytic uremic syndrome. ⁎ Corresponding author at: Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Ismaningerstr. 22, D-81675 Munich, Germany. Tel.: +49 89 41404754; fax: +49 89 41404875. E-mail address: [email protected] (W. Steimer).
http://dx.doi.org/10.1016/j.cca.2014.02.031 0009-8981/© 2014 Elsevier B.V. All rights reserved.
Erroneously high immunosuppressant concentrations have been reported repeatedly to be caused by reversible adsorption of the drugs to different intravenous catheter systems [7–20]. This can be especially dangerous since falsely elevated levels may fall within the therapeutic range whereas in reality, the patient is in risky underdosage. Graft-versus-host disease or organ rejection could be a life-threatening consequence [4,5]. A case in our house (cf. 2.2.2, patient A) turned our special attention to this pitfall: two days after being switched to oral Tac administration, a blood sample, accidentally taken from the lumen of the central venous catheter (CVC) previously used for infusing Tac, yielded a toxic concentration of 86 μg/L of Tac. A control sample obtained by venipuncture, though, showed a value of 2.9 μg/L of Tac. This remarkable discrepancy prompted us to perform a thorough in vitro and in vivo study on this effect. To systematically examine this adsorption and release effect for CsA and Tac in vitro, we studied some of the CVC systems most widely used in Europe and the USA. The 26 analyzed CVCs consisted of different materials, namely polyurethane (PU), silicone and PU with an incorporated silver ion-based antimicrobial agent. To substantiate the findings from our extensive in vitro experiments, a prospective in vivo study was conducted with fifteen patients receiving CsA or Tac via a CVC after stem cell transplantation.
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
63
Table 1 Central venous catheters (CVCs) used in in vitro and in vivo experiments. Abbreviation
Name
Material
Brand
Number of lumina
Inner diameter of contaminated lumen (mm)
Total length of contaminated umen (cm)
PU-AH-CVC
Polyurethane
Arrow-Howes™, (Kernen, Germany) Intra, (Rehlingen-Siersburg, Germany) Vygon, (Aachen, Germany)
4
1.52
46.6
PU-In-CVC Silver-Vy-CVC
Quad-Lumen Central Venous Catheterization Set Trilucath Multicath 4 Expert
3 4
0.80 1.25
36.5 35.8
Silicone-Vy-CVC
Lifecath Apheresis Plus
Vygon (Aachen, Germany)
3
1.9
50.0
Polyurethane Polyurethane incorporated with silver ion-based antimicrobial agent Silicone
2. Material and methods 2.1. In vitro experiments 2.1.1. Basic study Four types of catheters, made of three different materials, were examined in vitro: Arrow-Howes™ “Quad-Lumen Central Venous Catheterization Set” (n = 6, PU-AH-CVC), Intra “Trilucath” (n = 2, PU-INCVC), Vygon “Lifecath Apheresis Plus” (n = 2, silicone-Vy-CVC) and Vygon “Multicath 4 Expert” (n = 2, PU with an incorporated silver ion-based antimicrobial agent, silver-Vy-CVC; Table 1). One of the following doses was infused into one lumen of a CVC to mimic a therapeutic in vivo application of the drug: 250 mg of CsA (n = 10; Sandimmun, Novartis Pharma) in 100 mL NaCl solution (0.9%) (2.5 g/L of CsA) over 6 h or 2 mg of Tac (n = 10; Prograf, Astellas) in 50 mL NaCl (40 mg/L of Tac) over 22 h. Right after infusing the drug, blood sampling was mimicked in all four lumina with 4 mL of Fresh Frozen Plasma (FFP) each, always discarding the first 2 mL. When blood sampling was simulated, special attention was given to avoid cross contamination between the different lumina, particularly at the exits of the individual ports at the distal end of the catheter. Therefore, FFP was infused retrogradely with a syringe and collected at the distal end without getting in contact with the exits of the other lumina. The direction of sampling FFP had no significant effect on the released drug concentrations. The CVC then was recurrently rinsed with NaCl (0.9%), the first volume always being 10 mL, followed by multiple flushes of 1 L of NaCl. Each rinsing was followed by mimicking a blood withdrawal with 12 mL of FFP. To thoroughly analyze the withdrawal, the 12 mL was collected in six separate 2 mL doses, always discarding the first 2 mL. The lumina of the CVCs were not blocked with any fluid at any time. 2.1.2. Water bath In eight additional cases, infusing the drug was performed in a water bath at 37 °C to create a more realistic setting. All four types of catheters were once infused with CsA (n = 4) and once with Tac (n = 4), each time using a new catheter. 2.1.3. EDTA blood In addition to the 20 experiments conducted with FFP only, six PU-AH-CVCs were used for further investigations. This time, some mimicked blood sampling was performed with uncontaminated whole blood taken into EDTA containing tubes in order to study whether the whole blood and FFP behave similarly: Four CVCs were investigated in the usual manner, except for mimicking the blood collection alternating with FFP and EDTA blood. These experiments were performed with both drugs once at room temperature and once with the drug being infused in a water bath at 37 °C. To study the possible adsorption effects with the drugs being in the therapeutic range, two PU-AH-CVCs were put into EDTA blood for 24 h with a concentration of 125 μg/L of CsA (n = 1) and 10 μg/L of Tac (n = 1) respectively. Blood sampling was mimicked with EDTA blood after rinsing the CVCs with 10 mL of NaCl.
2.1.4. Fat emulsion In order to examine the effect of oily fluid on the release of the drugs, two PU-AH-CVCs were infused with CsA and Tac respectively. After flushing them with 1.01 L of NaCl, 100 mL of fat emulsion (Clinoleic 20%, Baxter) was infused. Blood sampling was then mimicked with 3 × 2 mL of EDTA blood, discarding the first 2 mL. Another 1.01 L of NaCl and 50 mL of fat emulsion were infused into the CVC previously contaminated with Tac. Afterwards, blood sampling was performed in the same manner as just described. 2.2. Patients We conducted a monocentric, prospective pilot study, approved by the institution's responsible committee and in accordance with the current revision of the Helsinki Declaration. Fifteen in-patients after allogenic stem cell transplantation aged 22 to 68 were included after giving informed consent (patients 1–15; Table 2). Twelve of them were receiving CsA and three were treated with Tac. In all cases, blood withdrawal was performed at all lumina of the CVC, each time discarding the first 5 mL of blood as it is a clinical practice in our house. Additionally, a blood sample was taken simultaneously via peripheral venipuncture each time. All samples were collected into EDTA containing tubes of 2.7 mL. 2.2.1. Main study In thirteen cases (patients 1–13, Table 2), the drug had been administered via a PU-AH-CVC (cf. 2.1.1, in vitro experiments). Doses of 75 mg to 750 mg of CsA and 1.0 mg to 2.5 mg of Tac had been administered over 7 to 22 days. The drug was received each day for 22 h, followed by flushing with 50 mL of NaCl for 2 h. Trough blood samples for our study were taken after 2 h of rinsing. 2.2.2. Special cases The only distinction in patient 14 was the fact that drug administration had been switched to oral medication three days prior to our blood sampling. Patient 15 had received only a single dose of Tac via a silver-Vy-CVC (cf. 2.1.1, in vitro experiments). Due to high Tac concentrations in the blood, the infusion with a concentration of 60 mg/L of Tac was stopped after 40 mL had been administered. For the following three days, Tac administration was stopped completely. On the third day after withdrawing drug administration, blood samples were taken for our study. Patient A had received Tac for 27 days via a three-lumen CVC made from polyurethane by Arrow-Howes™. Blood samples were taken on days 7, 8 and 18 after stopping Tac administration through the CVC. 2.3. Measuring All in vivo and in vitro samples were measured in duplicate by a LC/MS/MS immunosuppressant method modified from Koal et al. [21] and Ceglarek et al. [22] using a two dimensional HPLC system coupled to an ABI 3000 tandem mass spectrometer. Mobile phase chemicals
64
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
were purchased from Sigma-Aldrich, controls from Chromsystems Instruments & Chemicals and standards from Sigma-Aldrich, Chromsystems Instruments & Chemicals and Recipe Chemicals + Instruments. Samples were precipitated by a ZnSO4/methanol mix containing the internal standards. The analytes were captured by a Cyclone® HTLC column and separated by a phenyl-hexyl column before being analyzed in the mass spectrometer. Accuracy was ensured by continuously participating in proficiency testing schemes. If not measured within the first 6 h after it was obtained, each sample was kept at 8 °C and analyzed within a maximum of 72 h. For longtime storage, all samples from the in vitro and in vivo experiments were frozen at −25 °C. 2.4. Calculations All in vitro data was normalized with regard to the total inner surface of the analyzed lumen using PU-AH-CVC as a reference. Statistical analyses of the in vitro experiments were performed using the Wilcoxon signed rank test. A two sided level of significance of 5% was used. 3. Results 3.1. In vitro experiments Regardless of the material, every CVC which had been in contact with high concentrations of the drug used in the infusion showed significant evidence of reversible adsorption for CsA (n = 13, p = 0.001) and Tac (n = 13, p = 0.001, Wilcoxon signed rank test). 3.1.1. Main study Immediately after infusing the drugs and after discarding only the first 2 mL of FFP, the mean concentrations of 6420 μg/L (range: 260–17,700 μg/L) of CsA and 250 μg/L (range: 93–395 μg/L) of Tac were measured. The highest concentrations obtained from the drug release in the in vitro experiments were about 100 times lower than a usual therapeutic concentration (cf. 2.1.1) of the drug solution infused into the patient. After the first mimicked blood sampling, the CVC was rinsed with 10 mL of NaCl followed by sampling 6 portions of 2 mL of FFP, discarding the first 2 mL. The mean concentrations measured in the first portion yielded 132 μg/L for CsA and 55 μg/L for Tac and were lowered to 115 μg/L for CsA and 36 μg/L for Tac in the last portion. Rinsing the lumen with five more liters of NaCl resulted in an average reduction of the mean value by 26% for CsA and 32% for Tac per liter NaCl after each time discarding the first 6 mL of FFP. However, in some cases, even extensive rinsing with volumes as much as 24.01 L of NaCl still produced concentrations of 4.4 μg/L of Tac after discarding the first 6 mL. The five consecutive portions obtained at each mimicked blood sampling showed a mean decline of 12% for CsA and 31% for Tac from portion 1 to portion 5. 3.1.2. Different materials All three materials did adsorb and release the drugs. This effect was strongest in CVCs made from silicone compared to PU and silver throughout the entire study (Fig. 1). 3.1.3. Water bath Performing the infusion of the drug in a water bath at 37 °C led to higher drug levels as opposed to conducting it at room temperature (Fig. 2). 3.1.4. EDTA blood Performing the blood sampling with EDTA blood instead of FFP showed similar drug concentrations (Fig. 3).
The CVCs placed in EDTA blood with comparably low concentrations within the therapeutic range of CsA and Tac did not show any relevant reversible drug adsorption when blood sampling was mimicked. 3.1.5. Fat emulsion After rinsing the PU-AH-CVCs with 1.01 L of NaCl and 100 mL of fat emulsion, the first 2 mL of EDTA blood samples resulted in concentrations of 17 μg/L of CsA and 16.8 μg/L of Tac. The following 2 mL yielded 11 μg/L of CsA and 7.7 μg/L of Tac. After another 1.01 L of NaCl and 50 mL of fat emulsion, 4.9 μg/L and 2.7 μg/L of Tac were measured. 3.1.6. Cross contamination For all CVCs, regardless of the material, the phenomenon of adsorption and release was only observed for the lumen utilized for immunosuppressant infusion. All samples obtained in vitro from the uncontaminated lumina did not show any detectable drug concentration. Thus, no cross contamination was observed in vitro. 3.2. Patients For every patient, remarkably elevated concentrations of CsA or Tac were found in the lumen previously used for infusing the drug. 3.2.1. Main study In the eleven cases of CsA (patients 1–11; Table 2), the drug level obtained from the contaminated lumen was increased by a mean factor of 11.1 (range 6.7–22.0) compared to the one acquired via venipuncture. The CsA concentrations from the samples obtained by venipuncture yielded a mean level of 191 μg/L (range: 96–249 μg/L) whereas the drug levels in the samples from the contaminated lumen resulted in a mean CsA concentration of 2120 μg/L (range 1460–4410 μg/L). The blood samples from the lumina not used for infusing CsA showed comparable results (±10%) to the ones obtained by venipuncture. Patients 12 and 13 who had received Tac showed drug concentrations in the lumen previously used for infusion raised by a factor of 150 and 116 respectively (Table 2). The Tac level measured in the blood sample obtained by venipuncture resulted in 5.1 μg/L and 6.1 μg/L while the drug concentration in the sample from the contaminated lumen yielded 767 μg/L and 705 μg/L. In contrast to the results from all patients treated with CsA as well as patients 13 and 15, the blood sample of patient 12, from a lumen reportedly not used for infusing Tac previously, contained a 1.5 times elevated concentration of 7.9 μg/L of Tac compared to 5.1 μg/L obtained via venipuncture. 3.2.2. Special cases Patient 14, who had been switched to oral medication three days prior to blood sampling, still showed an increase of the CsA concentration from the contaminated lumen (895 μg/L) by the factor of 3.0 compared to venipuncture (296 μg/L; Table 2). In patient 15, the concentration from the lumen used only once for Tac infusion was elevated by a factor of 2.9. The sample received by venipuncture came to 12.9 μg/L and the one from the contaminated lumen resulted in 37 μg/L. Patient A was administered with Tac through a CVC for 26 days. At some point between days 22 and 25, the patient had most likely received an accidental overdose of Tac resulting in toxic concentrations of up to 111 μg/L measured on day 26. Tac medication was thus stopped completely, restarting with oral Tac five days after the withdrawal. Two days later, another toxic concentration of 86 μg/L of Tac was determined. However, it turned out that this time the sample had accidentally been taken from the lumen previously used for infusing Tac, since a control sample obtained via venipuncture yielded only 2.9 μg/L. Two more blood withdrawals taken 8 and 18 days after the last i.v. administration showed elevated blood concentrations in the contaminated lumen both by a factor of 37 compared to venipuncture. Similar to patient 12, the
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
65
Table 2 All data from patients.
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 b Patient 15 b Patient A b
a b c
Drug
Type of central venous catheter (CVC)
Venipuncture (μg/L)
Contaminated lumen (μg/L)
Lumen 2 (μg/L)
Lumen 3 (μg/L)
Lumen 4 (μg/L)
Duration of i.v. drug application (days)
Current i.v. drug dosage (mg)
Time since last i.v. administration (days)
CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA Tac Tac CsA Tac Tac
PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC PU-AH-CVC Silver-Vy-CVC PU-AH-CVC c
235 96 249 144 172 138 192 248 199 198 229 5.1 6.1 296 12.9 2.9 4.6 2.6
2490 2110 4410 2255 1615 1460 1755 1655 1990 1635 1930 767 705 895 37 86 171 96
231 97 257 138 172 147 194 254 194 201 233 4.9 6.0 292 13.5
a
277 102
8 7 22 10 9 13 7 16 14 15 16 21 11 22 1 26
300 480 200 750 450 75 200 400 300 500 200 2.0 1.75 – – – – –
– – – – – – – – – – – – – 3 3 7 8 18
a
99 275 143 171 a
a
188 265 195 201
190 249 193 200 238 5.5 6.2 286 13.6 – – –
a
7.6 6.4 295 13.4 a
5.9 6.3
a
147 172
3.3 4.4
Blood sampling not performed. Patient had been switched to oral medication. Arrow-Howes™ polyurethane CVC with three lumina.
Fig. 1. Drug concentrations from 12 CVCs: one set of 1 silicone-Vy-CVC, 1 silver-Vy-CVC and 3 PU CVCs (2 PU-AH-CVCs, 1 PU-In-CVC) infused with CsA and one set with Tac. P3 of blood sampling mimicked with FFP after rinsing multiple times with NaCl. Average line for PU enclosed. Part of the figure zoomed in and is displayed in the upper right corner.
66
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
Fig. 2. Drug concentrations from 20 CVCs: for CsA and Tac, 1 silicone-Vy-CVC, 1 silver-Vy-CVC, 1 PU-AH-CVC and 1 PU-In-CVC were infused in a water bath at 37 °C. 1 silicone-Vy-CVC, 1 silver-Vy-CVC, 3 PU-AH-CVCs and 1 PU-In-CVC per drug were infused at room temperature. P3 of blood sampling mimicked with FFP after rinsing multiple times with NaCl. Average lines enclosed. Part of the figure zoomed in and is displayed in the upper right corner.
samples for patient A taken from the lumina previously not used for Tac administration did not fall within the range of ±10% of the values determined by venipuncture (Table 2). 4. Discussion Our results from a systematic and comprehensive in vitro and in vivo study are in concordance with various reported cases [7–13] as well as smaller in vivo [14,15] and in vitro [16–18] experiments for CsA of spuriously elevated drug concentrations due to reversible adsorption to catheter materials. Mimicking blood sampling with whole EDTA blood showed similar results as compared to FFP (Fig. 3). Moreover, performing the experiments in a water bath only raised the drug concentrations to a small degree. This verifies that the setup of the main in vitro experiments was close to the in vivo conditions. Especially the first blood sample of each in vitro experiment as well as the concentration from the contaminated lumen of the patients yielded a multiple of the concentration of the reference range (i.e. 100–400 μg/L for CsA and 5–20 μg/L for Tac [23,24]). In clinical practice, many blood samples accidentally taken from the contaminated lumen lead to alarmingly high measurements resulting in further investigation. Thus, blood sampling usually is repeated correctly and no harm is done to the patient. The in vitro results show that rinsing the catheter with NaCl leads to a reduction of the drug concentration. However, volumes of up to 24.01 L of NaCl still resulted in detectable drug release. Infusing fat emulsion,
as being done quite often in critically ill patients with CVCs, did not put an end to releasing the drugs, either. Moreover, patients 14, 15 and A reveal that even a long time after switching to oral administration of CsA or Tac, the lumen previously used for administering the drug remains contaminated. This emphasizes how blood sampling from a lumen ever used for infusing the immunosuppressant needs to be avoided at all times, even if the patient is already switched to oral medication and different fluids have been infused through the contaminated lumen. Especially if a long time has passed since the last i.v. dose was administered, it is absolutely critical to still be aware of this phenomenon. In this case, the spurious elevation could be of minor degree resulting in concentrations within or only slightly above the therapeutic range over an extended period of time. Therefore, the falsely elevated values may remain unrecognized which may lead to erroneously reduced immunosuppressant dosage resulting in undiscovered underdosage. This would put patients after stem cell transplantation at risk for graft-versus-host disease [5] and patients after solid organ transplantation could even experience organ rejection [4]. Analyzing four patients receiving CsA via CVCs made from silicone led Leson et al. to conclude that a correction factor for the elevation from the contaminated lumen could be used [10]. In contrast to their data, both our in vitro and in vivo experiments show that the adsorption and release of the drugs are of varying degree and cannot be predicted precisely. Our data suggests that there is no fix correlation between the extent of the elevation and the dosage as well as the duration of the drug application.
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
67
Fig. 3. Drug concentrations from 4 PU-AH-CVCs, 2 of them infused in a water bath at 37 °C, one with CsA and one with Tac. P3 of blood sampling mimicked with alternating FFP and EDTA blood after rinsing multiple times with NaCl. Exponential fit line enclosed.
In vitro experiments by Carreras et al. show that CVCs made from silicone adsorb more CsA than CVCs made from PU [18]. In our experiments, we were able to demonstrate that CVCs made from silicone also release greater amounts of CsA and Tac than the ones made from PU or PU incorporated with silver. Patient 15 still showed an elevation by a factor of 2.9 although Tac had only been administered once through that lumen three days prior to sampling. This is the first reported case of a falsely elevated Tac concentration due to reversible adsorption of the drug at a CVC with an incorporated silver ion-based antimicrobial agent. To confirm this finding, further in vivo studies should be conducted. The question whether or not blood withdrawal for therapeutic drug monitoring of CsA and Tac should only be performed peripherally has been discussed in several publications [12–14,25]. Senner et al. analyzed 71 paired CsA blood samples taken peripherally and from the CsAuncontaminated lumen of a double-lumen silastic CVC [25]. They did not find a significant difference between the two sampling sites (paired Student t test; p = 0.13) and concluded that blood sampling for CsA monitoring can be performed from the uncontaminated lumen of the CVC. Claviez et al. came to the same recommendation regarding CsA [12]. Grouzmann et al., on the other hand, suggest that the blood samples for Tac monitoring should be performed via venipuncture [13]. Our in vivo studies showed no cross contamination for CsA. Therefore, we fall into line with the authors concluding that blood sampling for therapeutic drug monitoring from uncontaminated lumina is acceptable for CsA.
However, patient 13 and patient A showed alarming elevations by factors of 1.3 to 2.4 of the Tac concentration in lumina that had reportedly not been used for immunosuppressive infusion. Apart from incorrect sampling, possible reasons for these elevations could be substance properties of Tac or cross contamination at the exits of the individual ports at the distal end. Further experiments should be conducted in order to examine these correlations. For now, we strongly recommend using peripheral venipuncture to acquire blood samples for Tac monitoring. Thrombotic thrombocytopenic purpura/hemolytic uremic syndrome (TTP/HUS) is a rare but very severe complication after hematopoietic stem cell or solid organ transplantation which can be induced by CsA or Tac [26]. Therefore, discontinuation or sometimes dose reduction of the calcineurin inhibitors is the most important treatment against this condition [27,28]. Our findings suggest that the adsorption of CsA or Tac to the CVC is reversible not only when blood is withdrawn but also when fluid is infused through the contaminated lumen. This could possibly lead to small amounts of the drug still being washed into the body with other fluids long after the drug was discontinued. If 1 L of fluid was to be infused through the contaminated lumen of the patient and assuming that the drug was released with the same mean concentration (cf. 3.1.1) measured in vitro after discarding only 2 mL of FFP, a total amount of 6.42 mg of CsA or 0.25 mg of Tac could unintentionally be infused into the patient's body. Depending on the patient's course of treatment, it should be left to the physician's discretion whether or not to tolerate these comparable small amounts of the drug. However, there clearly is a need to further investigate this possible phenomenon. If the
68
C. Hacker et al. / Clinica Chimica Acta 433 (2014) 62–68
patient is suffering from TTP/HUS and CsA or Tac is to be withdrawn completely, the lumen previously used for infusing CsA or Tac should not be utilized any longer or the CVC should be changed. Many CVC practice manuals and guidelines still lack these special adsorption and release aspects [29–31] and the package inserts of Tac and CsA only mention PVC incompatibility with potential extraction of phthalates regarding drug preparation for infusion [32–35] but not in connection with blood sampling for therapeutic drug monitoring. It is crucial to make sure that the medical staff performing blood withdrawal on patients with CVCs is aware of the reversible adsorption of drugs to CVCs and of the importance of the sampling technique. 5. Conclusion Catheter lumina used for immunosuppressant infusion should never be utilized for blood sampling even after rinsing with high volumes of other fluids and discarding a limited volume prior to blood sampling. As cross contamination between different lumina may occur for Tac, venipuncture certainly is the safest way to obtain correct blood samples for therapeutic drug monitoring. Acknowledgments We are indebted to C. Mueller and the members of the drug monitoring laboratory for the technical support as well as B. Haller for the statistical advice. We also thank the staff of the bone marrow transplant unit for the assistance during the blood sampling as well as C. Voelkl for performing the drug concentration analyses of the in vivo samples. References [1] Storb R, Deeg HJ, Whitehead J, et al. Methotrexate and cyclosporine compared with cyclosporine alone for prophylaxis of acute graft versus host-disease after marrow transplantation for leukemia. N Engl J Med 1986;314:729–35. [2] Starzl TE, Todo S, Fung J, Demetris AJ, Venkataramman R, Jain A. Fk-506 for liver, kidney, and pancreas transplantation. Lancet 1989;2:1000–4. [3] Wiederrecht G, Lam E, Hung S, Martin M, Sigal N. The mechanism of action of Fk-506 and cyclosporine-A. In: Allison AC, Lafferty KJ, Fliri H, editors. Immunosuppressive and antiinflammatory drugs. New York: New York Acad Sciences; 1993. p. 9–19. [4] Geissler EK, Schlitt HJ. Immunosuppression for liver transplantation. Gut 2009;58:452–63. [5] Ram R, Storer B, Mielcarek M, et al. Association between calcineurin inhibitor blood concentrations and outcomes after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2012;18:414–22. [6] Randomised trial comparing tacrolimus (Fk506) and cyclosporin in prevention of liver allograft rejection. European Fk506 Multicentre Liver Study Group. Lancet 1994;344:423–8. [7] Shaefer MS, Collier DS, Haven MC, et al. Falsely elevated Fk-506 levels caused by sampling through central venous catheters. Transplantation 1993;56:475–6. [8] Duffner U, Bergstraesser E, Sauter S, Bertz H, Niemeyer C. Spuriously raised cyclosporin concentrations drawn through polyurethane central venous catheter. Lancet 1998;352:1442. [9] Lorenz RG, Garrett N, Turk JW, Scott MG. Problems with therapeutic monitoring of cyclosporine using silicone central venous line samples. Transplantation 1991;52:1109–10. [10] Leson CL, Bryson SM, Giesbrecht EE, Saunders EF. Therapeutic monitoring of cyclosporine following pediatric bone-marrow transplantation — problems with sampling from silicone central venous lines. DICP Ann Pharmacol 1989;23:300–3. [11] Blifeld C, Ettenger RB. Measurement of cyclosporine levels in samples obtained from peripheral sites and indwelling lines. N Engl J Med 1987;317:509.
[12] Claviez A, Glass B, Dreger P, Suttorp M. Elevated blood drug levels obtained from indwelling silicon catheters during oral cyclosporine A administration. Bone Marrow Transplant 2002;29:535–6. [13] Grouzmann E, Buclin T, Biollaz J. Misleading tacrolimus concentration value in blood taken from a catheter used for tacrolimus administration. Am J Health Syst Pharm 2008;65:226–8. [14] Busca A, Miniero R, Vassallo E, Leone L, Oddenino O, Madon E. Monitoring of cyclosporine blood-levels from central venous lines — a misleading assay. Ther Drug Monit 1994;16:71–4. [15] de Witte T, Hoitsma A, Manoiu MM, Janssen J. Monitoring of cyclosporine during continuous intravenous administration. Bone Marrow Transplant 1986;1:147–50. [16] Fitzsimmons WE, Ponzillo JJ. Cyclosporine concentration sampling from an in vitro central venous catheter system. Transplantation 1988;45:1158–9. [17] Shibata N, Ikuno Y, Tsubakimoto Y, et al. Adsorption and pharmacokinetics of cyclosporin A in relation to mode of infusion in bone marrow transplant patients. Bone Marrow Transplant 2000;25:633–8. [18] Carreras E, Lozano M, Deulofeu R, Roman S, Granena A, Rozman C. Influence of different indwelling lines on the measurement of blood cyclosporine-A levels. Bone Marrow Transplant 1988;3:637–9. [19] Hacker C, Menzel H, Steimer W. Spuriously high levels of immunosuppressants due to reversible adsorption to central venous catheters [abstract]. Ther Drug Monit 2011;33:543–4. [20] Bleyzac N. On the importance of blood sampling for ciclosporin pharmacokinetic studies. Br J Clin Pharmacol 2013;75:869–70. [21] Koal T, Burhenne H, Romling R, Svoboda M, Resch K, Kaever V. Quantification of antiretroviral drugs in dried blood spot samples by means of liquid chromatography/ tandem mass spectrometry. Rapid Commun Mass Spectrom 2005;19:2995–3001. [22] Ceglarek U, Lembcke J, Fiedler GM, et al. Rapid simultaneous quantification of immunosuppressants in transplant patients by turbulent flow chromatography combined with tandem mass spectrometry. Clin Chim Acta 2004;346:181–90. [23] Uptodate: cyclosporine (systemic): drug information. http://www.uptodate.com/ contents/cyclosporine-systemic-drug-information?source=search_result&search= ciclosporin+eference+range&selectedTitle=4~150#F8012570 [accessed on December 14, 2013]. [24] Uptodate: tacrolimus (systemic): drug information. http://www.uptodate.com/ contents/tacrolimus-systemic-drug-information?source=search_result&search= tacrolimus+reference+range&selectedTitle=2~150#F9564239 [accessed on December 14, 2013]. [25] Senner AM, Johnston K, McLachlan AJ. A comparison of peripheral and centrally collected cyclosporine a blood levels in pediatric patients undergoing stem cell transplant. Oncol Nurs Forum 2005;32:73–7. [26] Dlott JS, Danielson CF, Blue-Hnidy DE, McCarthy LJ. Drug-induced thrombotic thrombocytopenic purpura/hemolytic uremic syndrome: a concise review. Ther Apher Dial 2004;8:102–11. [27] Ho VT, Cutler C, Carter S, et al. Blood and marrow transplant clinical trials network toxicity committee consensus summary: thrombotic microangiopathy after hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:571–5. [28] Agarwal A, Mauer SM, Matas AJ, Nath KA. Recurrent hemolytic-uremic syndrome in an adult renal-allograft recipient — current concepts and management. J Am Soc Nephrol 1995;6:1160–9. [29] Bishop L, Dougherty L, Bodenham A, et al. Guidelines on the insertion and management of central venous access devices in adults. Int J Lab Hematol 2007;29:261–78. [30] Moureau NL. Drawing blood through a central venous catheter. Nurs (Lond) 2004;34:28. [31] Johns Hopkins Medicine. Interdisciplinary Clinical Practice Manual, Infection Control, Vascular Access Device (Vad) Policy, Adult, Ifc035, Appendix F. http://www. hopkinsmedicine.org/armstrong_institute/_files/clabsi_toolkit/vad_appx/HF_Short_ Term_Central_Venous_Catheter.pdf [accessed on December 14, 2013]. [32] Astellas Prograf U.S. physician prescribing information. https://www.astellas.us/ docs/prograf.pdf [accessed on December 14, 2013]. [33] Novartis Pharma US sandimmune prescribing information. http://www.pharma.us. novartis.com/cs/www.pharma.us.novartis.com/product/pi/pdf/sandimmune.pdf [accessed on December 14, 2013]. [34] Fachinformation Sandimmun® 50 Mg/Ml Konzentrat Zur Herstellung Einer Infusionslösung Novartis Pharma. Revised September 2011. Novartis Pharma Gmbh 90327 Nürnberg. License Number 3123.00.00. [35] Fachinformation Prograf 5 Mg/Ml Konzentrat Astellas. Revised September 2013. Astellas Pharma Gmbh 80971 München. License Number 41954.00.01.