Secondary Logo

Journal Logo

Renal/Extracorporeal Blood Treatment

Central Venous Line and Dialysis Catheter Position Affects Drug Clearance during Continuous Renal Replacement Therapy in an Animal Model

Frithiof, Robert*; Bandert, Anna*,†; Larsson, Anders; Lipcsey, Miklos§; Smekal, David*

Author Information
doi: 10.1097/MAT.0000000000000839
  • Free


Renal replacement therapy (RRT) has since long been a standard treatment in intensive care for patients with renal failure or fluid overload. The incidence of acute kidney injury in intensive care is between 39% and 57%.1,2 Approximately 13–23% of these patients receive RRT and its continuous form (CRRT) is commonly used.1,2 According to the Swedish ICU registry 11% of patients with intensive care length of stay over 48 hours required CRRT.3

Different aspects of CRRT have been thoroughly studied and losses of hormones and endogenous catecholamines have been supposed to be minimal.4 However, little is known about how and if clearance of drugs is affected by infusion site during CRRT.

Central venous catheters (CVC) are widely used in intensive care and if RRT is initiated a central venous dialysis catheter (CVDC) is also inserted. Depending on insertion site, the tips of the different catheters commonly end up with both catheter tips in the superior caval vein with the possibility of direct aspiration of infusions through the CVC into the CRRT system.5 Such phenomena have been reported, for example, during thermodilution for cardiac output measurement when cardiac index, global end-diastolic index, and extravascular lung water index were influenced by RRT, when the CVDC was close to the site of saline infusion.6–8

If the tips of the different catheters are close to each other, it is reasonable to believe that there is a risk that a given drug infused in the CVC will be directly aspirated through the CVDC. However, the literature addressing this topic is scarce. There are a few case reports where this phenomenon is highly suspected.9,10 In addition, Kam et al.11 investigated this hypothesis in a bench model and showed that drugs delivered through a CVC can be immediately aspirated in an adherent CVDC.

The aim of this study was to investigate if the relationship between the position of CVC and CVDC influences the elimination of infused drugs, during CRRT. We hypothesized that an infused dose of gentamicin would give a lower serum concentration if infused in a CVC with the tip close to the CVDC compared with if the CVC and CVDC were in different vessels. We also hypothesized that more noradrenaline is needed to reach a target pressure with the CVC and CVDC close together, compared with if the CVC and CVDC are separated.

Materials and Methods


The study was performed in 8–10 weeks old pigs of Norwegian landrace breed. All animals were handled according to the guidelines of the Swedish Board of Agriculture and the European Convention on Animal Care. The experiment was approved by the Animal Ethics Committee of Uppsala University, Sweden (C 155/14).

Anesthesia and Preparation

Anesthesia was induced by injecting a mixture of tilétamin-zolazepam 6 mg/kg and xylazin 2.2 mg/kg intramuscularly, and was maintained by an infusion of sodium pentobarbital 8 mg/kg/h and morphine 0.26 mg/kg/h dissolved in 2.5% glucose in a peripheral vein. Rocuronium bromide 1–3 mg/kg/h, was given as continuous infusion. Ringer’s Acetate solution was administered at 2 ml/kg/h, resulting in a total fluid administration rate of 10 ml/kg/h.

A tracheotomy was performed and the animals were mechanically ventilated throughout the experiment (Servo-I, Maquet, Stockholm, Sweden) with constant setting of: respiratory rate 25/min, inspired oxygen fraction 0.3, with the tidal volume that at baseline yielded a PaCO2 of 4.5 to 5.5 kPa. A cervical artery was catheterized for pressure monitoring and blood sampling.

A central venous line and a 13.5 Fr dialysis catheter (HemoCath SDL136e, MedComp, Harleyville, PA) were inserted through the right external jugular vein (EJV) into the superior caval vein. An additional 13.5 Fr dialysis catheter was inserted through the femoral vein into the inferior caval vein (Figure 1).

Figure 1.:
Cannulation: In addition to standard cannulation the pigs received five important accesses. One peripheral access in an auricular vein that was used for infusion of sodium nitroprusside. One arterial line in a cervical artery, used for blood sampling. One central venous catheter (CVC) in EJV, with tip in superior caval vein, used for infusion of noradrenaline and gentamicin. Two central venous dialysis catheters (CVDC), one in EJV with tip close to the CVC tip and one CVDC in FV with tip in the inferior caval vein.

A 7 Fr Swan-Ganz catheter was inserted in the pulmonary artery from the left EJV. A urinary catheter was introduced via a bladder incision. As the preparation was completed, a 30-minute stabilization period followed before baseline values were obtained.


CRRT, using continuous veno venous hemodialysis, was started (Multifiltrate, Fresenius medical, Stockholm, Sweden) with the following settings: blood flow: 60 ml/min, dialysate flow 1200 ml/h, citrate target 4 mmol/L, postfilter ionized calcium level 0.25–0.34 mmol/L, ultrafiltration 0 ml/min. The dialysis flow was chosen according to the body surface of the pigs with approximately 1000ml/h/m2.

Since pigs are hypercoagulable, compared with humans, a single dose of Heparin 5000IE was administrated intravenously at the start of CRRT.


The pigs were anesthetized and after 30 minutes of steady state the pigs were randomized to start CRRT in one of the CVDC, that is, in FV or EJV. CRRT was started followed by 30 minutes stabilization period. An infusion of sodium nitroprusside was started in a peripheral auricular vein and titrated with the goal of reaching a mean arterial pressure of 50 mm Hg. Next, an infusion of noradrenaline (20 µg/ml) was initiated in the CVC with the goal of keeping mean arterial pressure at 75 mm Hg during 30 minutes. At the same time, a gentamicin infusion was started and continued for 30 minutes in the CVC with a total dose of 3 mg/kg. After 30 minutes, the administered noradrenaline dose was registered and the plasma concentration of gentamicin was measured. The dialysis circuit was then moved to the other CVDC followed by a 30 minute wash out period during which the CRRT continued. The experiment was repeated; hence, the pigs acted as their own controls in a cross-over design (Figure 2).

Figure 2.:
The design of the study. The pigs were randomized to start continuous renal replacement therapy (CRRT) via a central venous dialysis catheter (CVDC) placed in the external jugular vein (EJV) or femoral vein (FV). The CVDC catheter in the EJV was close to the central venous catheter (CVC). CRRT was started. Hypotension to mean arterial pressure (MAP) 50 mm Hg was induced by infusion of sodium nitruprusside (SNP) given in a peripheral vein. Gentamicin (GM) infusion was started in the CVC. At the same time noradrenaline (NA) infusion was started in the CVC with aim of reaching a MAP of 75 mm Hg. The infusions were kept for 30 minutes. Plasma sample for GM concentration was taken and registration of NA dose was made before and after the infusion, marked with black arrow. The dialysis circuit was thereafter moved to the other CVDC. After a wash-out period the experiment was repeated.

Laboratory Analyses

Blood samples were obtained from an arterial line in a cervical artery. Plasma for analysis of gentamicin concentration was obtained at the start and at the end of the 30 minutes’ period of antibiotic infusion. Samples for blood gases were obtained at the start of CRRT, at beginning and at the end of both 30 minutes’ periods. Blood gas and point care analyses were performed on an ABL 800 Blood gas analyzer (Radiometer, Brønshøj, Denmark). Gentamicin was analyzed on an Architect ci16200 (Abbott Laboratories, Abbott Park, IL) with reagents (1P31) from the same manufacturer. The Gentamicin concentration before infusion period 1 (i.e., 0) was subtracted from the concentration after period 1. The gentamicin concentration before infusion period 2 was subtracted from the concentration after period 2.

Statistical Analysis Methods

Data were analyzed using SPSS, Version 22.0 (IBM Corporation, Armonk, NY). Tests for normal distribution were performed using the Shapiro-Wilk test. Normally distributed data were analyzed with independent students’ t-test. Non-normally distributed variables were analyzed with Mann–Whitney U test. Normally distributed data are presented as mean ± standard deviation (SD); data with a non-normal distribution are presented as median and interquartile range (IQR). The level of significance was set at p < 0.05.


For baseline characteristics see Table 1.

Table 1.:
Variables Before and After Experiment

Ten of 10 pigs were included in the experiment, and all of them followed the entire study protocol. Two pigs started with a lactate level of 3 mmol/L or above, but did not exhibit other physiological derangements, and their lactate levels were normalized before the start of the CRRT. There was no observed acidosis in any of pigs in any of the arterial blood samples.

Outflow and inflow pressures monitored in the CRRT circuit during CRRT were within the preset normal range for all pigs during the whole experiment.

Concentration of gentamicin in plasma was measured after 30 minutes of drug infusion given in a CVC through the EJV. The mean concentration was 5.66 (SD, ±1.23) mg/L with CRRT in EJV, that is, CVDC close to the infusion site, compared with 7.76 (SD, ±2.30) mg/L with CVDC in FV, that is, distant from the infusion site (p = 0.02; Figure 3).

Figure 3.:
Results gentamicin concentration: The mean plasma gentamicin (GM) concentration was 5.66 (SD, ±1.23) mg/L when infusion was close to the central venous dialysis catheter (CVDC), that is, CVDC and continuous renal replacement therapy (CRRT) in external jugular vein (EJV). The mean plasma concentration was 7.76 (SD, ±2.30) mg/L when infusion and CVDC was in different vessels, that is, CRRT in femoral vein (FV) (p = 0.02). The dot with number 7 indicate animal number 7.

The amount of given noradrenaline was obtained from the intravenous pumps after the 30 minutes’ periods in each pig. The mean infusion rate of noradrenaline given in a CVC in EJV during 30 minutes with CRRT in CVDC in EJV versus FV were 0.32 (SD, ±0.16) and 0.15 (SD, ±0.08) µg/kg/min, respectively (p = 0.006; Figure 4).

Figure 4.:
Results noradrenaline infusion rate µg/kg/min: The mean infusion rate of noradrenaline (NA) given to elevate mean arterial blood pressure from 50 to 75 mm Hg for 30 minutes in the pigs. The NA infusion rate was 0.32 (SD, ±0.16) µg/kg/min when it was close to the central venous dialysis catheter (CVDC), that is, continuous renal replacement therapy (CRRT) in external jugular vein (EJV). The NA infusion rate was 0.15 (SD, ±0.08) µg/kg/min when CVDC and infusion were in different vessels, that is, CRRT was in the femoral vein (FV). The dot with number 1 indicate animal number 1.

The 5 pigs with the highest cardiac output (CO), measured before each of the two infusion periods, were compared with the 5 pigs with the lowest CO. The median noradrenalin infusion rate was 0.19 (IQR, 0.21) versus 0.20 (IQR, 0.20) µg/kg/min in the high CO group compared with low CO group (p = 0.97). The mean gentamicin concentration in the group with high CO was 6.24 (SD, ±2.01) and with low CO 7.18 (SD, ±2.18) mg/L (p = 0.33).

Comparing the infusion rate and concentration between the first part of the experiment, that is, before cross-over with the second part of the experiment, that is, after cross-over showed no difference. The mean noradrenaline infusion rate in the first part of the experiment was 0.27 (SD, ±0.17) and the second 0.21 (SD, ±0.12) µg/kg/min (p = 0.34). The mean gentamicin concentration at the end of the first part of the experiment was 7.3 (SD, ±1.61) and at the end of the second part was 6.13 (SD, ±2.43) mg/L (p = 0.23). Adjustment for CO and dialysis dose did not alter the results. A dataset supporting the results can be seen in Table 2.

Table 2.:
Data for Each Animal During the Experiment


In this study, the plasma concentration of gentamicin was lower if given in a CVC close to the CVDC compared with CVC and CVDC in different vessels during CRRT.

Moreover, the infusion rate of noradrenaline needed to normalize the blood pressure, in hypotensive piglets during CRRT, was double if the infusion through the CVC was close to the CVDC compared with if CVC and CVDC were in different vessels.

This is, to our knowledge, the first in vivo study to systematically investigate the role of CVC and CVDC-placement on the removal of infused drugs. We hypothesized that, during CRRT, an infused drug in a CVC close to the CVDC would result in a lower blood concentration compared with if drug infusion and CVDC were positioned in different vessels. The hypothesis was dually tested, in an experimental randomized cross-over protocol, with both noradrenaline dose registration and plasma gentamicin concentration measurements, both indicating that there is a higher clearance of drugs in the dialysis system when the catheters are close compared with if they are apart. The plausible explanation for this observation could be that there is a direct aspiration of noradrenaline and gentamicin from the CVC into the CVDC.

Another potential contributing mechanism underlying the results obtained here is access recirculation (AR). It is described in intermittent hemodialysis and occurs when blood that has been dialyzed re-enters the dialysis circuit in the inflow-port, thereby diluting undialyzed blood from the system circulation. It is unclear if AR is of importance during CRRT. In intermittent hemodialysis, AR is higher in catheters placed in FV compared with internal jugular vein or subclavian vein (SV), and this is speculated to be due to a lower blood flow in the FV.12–14 AR was not measured in this study and thus we are unable to determine if it was present and to what extent. However, the length of the catheters and size of the pigs (leading to a probable tip position in the inferior caval vein), the dialysis method (CRRT), information from the manufacturer of the CVDCs (stating an AR of less than 0.5%) makes it unlikely that AR is a major cause of the differences in CRRT-site described here.

Drug dosing in critically ill patients, during CRRT, is based on the assumption that drugs are eliminated from a more or less equilibrated central compartment.15,16 Direct aspiration of a drug from CVC to CVDC leads to more complex and less predictable pharmacokinetic principles, in turn leading to the risk of drug underdosing. This could be particularly problematic for drugs with a high protein binding or large volume of distribution as they could enter the dialysis circuit, before equilibration in the central compartment has taken place, resulting in an increased clearance in the CRRT and further adding to drug underdosing.15–17

A bench model has suggested that the amount of aspirated drug could be higher and closer to the two tips of a CVC and CVDC are to each other, but this model did not account for variations in blood flow or catheter movements during cardiac and breathing cycles.11 There are some case reports published where the phenomenon of direct aspiration is highly feasible.9,10 In the cases described the suspected aspirated drugs were, among others, inotropes and vasopressors that had serious effects on the patients with circulatory failure. If this phenomenon would occur with less dramatic effects, it would be a risk of under-detection, which would explain the lack of literature in this field.

The potential harm of drug underdosing will depend on the given drug. It would seem that this risk is less imminent concerning drugs that are dosed bedside to predefined targets, such as noradrenaline and to some extent insulin, that is, a persistent low blood pressure or high blood glucose levels, would lead to a higher rate of drug infusion, but not at corresponding increase in drug concentration in the patient. On the other hand, giving an inadequately low dose of antibiotics may be of greater consequence. There is evidence supporting adequate, timely antibiotic administration in sepsis.18,19 This is important for mortality, length of stay, and health care costs.20,21 An insufficient antibiotic concentration may facilitate the development of antibiotic resistance of concern both for the patient but also for society.22,23

Infused noradrenaline has been shown to have high interindividual variation in critically ill patients and its clearance to be affected by severity of illness (SAPS-II).24,25 In this study, all animals were healthy and after randomization received CRRT in CVDCs in EJV or FV during the first part of the experiment. Thereafter as the circuit was shifted to the other CVDC for the second part of the experiment, in a cross over design, the animals served as their own controls. By doing so interindividual differences in the effect of noradrenalin was circumvented.

The limitations to our study are, firstly, all of those associated with animal experiments. The homogenous material of healthy pigs obviously differs from our complex intensive care patients in numerous ways, but considering the physiological and anatomical similarities between pig and human, this model seems applicable in this experiment. The use of an animal model also makes the observed result less clouded by interindividual differences, such as difference in severity of illness, between the experimental subjects. Second, in this study, we induced hypotension pharmacologically, which also differs from the mechanisms leading to hypotension during, for example, septic or cardiogenic shock, but there is no reason to believe that this would affect our findings on noradrenaline infusion rate or gentamicin concentration. Third, the observation time of 30 minutes is short compared with a clinical intensive care setting but considering the clear results of our study we do not believe that a longer observational time would alter the result in a different direction. Fourth, clearance of the infused substances by CRRT would have added a more exact estimation of the effect of catheter placement on drug removal by CRRT. Still, considering the robust differences in plasma concentration of gentamicin and effect of noradrenaline the conclusion of the study would likely have been the same.

Finally, no x-ray control of the catheter placement was made. CVC and CVDC catheters are known to be misplaced or migrate once in position.26,27 The length of the catheters used in our study in relation to the pig’s size makes it highly unlikely that the femoral CVDC would have reached above the diaphragm and by that become located close to the CVC.


This experimental study indicates that the removal of noradrenalin and gentamicin is substantially increased if these drugs are infused in close vicinity of the CVDC, during CRRT. The plausible explanation for this is direct aspiration of drug from the CVC to the CVDC. If direct aspiration is more likely to occur during specific intensive care conditions, such as sepsis, remains to be studied. The question whether there is a distance between the tips where no direct aspiration occurs is also unanswered.


The authors are most grateful for the excellent technical assistance provided by the staff at the Hedenstierna laboratory.


1. Nisula S, Kaukonen KM, Vaara ST, et al.; FINNAKI Study Group: Incidence, risk factors and 90-day mortality of patients with acute kidney injury in Finnish intensive care units: the FINNAKI study. Intensive Care Med 2013.39: 420–428.
2. Hoste EA, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive Care Med 2015.41: 1411–1423.
3. Swedish Intensive Care Registry: Utdataportalen: CRRT-antal registrerade åtgärder/vtf 2015-2016. Data aquisition: March 20th, 2017.
4. Bellomo R, McGrath B, Boyce N. Effect of continuous venovenous hemofiltration with dialysis on hormone and catecholamine clearance in critically ill patients with acute renal failure. Crit Care Med 1994.22: 833–837.
5. Frykholm P, Pikwer A, Hammarskjöld F, et al. Clinical guidelines on central venous catheterisation. Swedish Society of Anaesthesiology and Intensive Care Medicine. Acta Anaesthesiol Scand 2014.58: 508–524.
6. Sakka SG, Hanusch T, Thuemer O, Wegscheider K. The influence of venovenous renal replacement therapy on measurements by the transpulmonary thermodilution technique. Anesth Analg 2007.105: 1079–1082, table of contents.
7. Pathil A, Stremmel W, Schwenger V, Eisenbach C. The influence of haemodialysis on haemodynamic measurements using transpulmonary thermodilution in patients with septic shock: an observational study. Eur J Anaesthesiol 2013.30: 16–20.
8. Schmidt S, Westhoff TH, Hofmann C, et al. Effect of the venous catheter site on transpulmonary thermodilution measurement variables. Crit Care Med 2007.35: 783–786.
9. Stricker KH, Takala J, Hullin R, Ganter CC. When drugs disappear from the patient: elimination of intravenous medication by hemodiafiltration. Anesth Analg 2009.109: 1640–1643.
10. Mohammad A, Zafar N, Feerick A. Cardiac arrest in intensive care unit: Case report and future recommendations. Saudi J Anaesth 2010.4: 31–34.
11. Kam KY, Mari JM, Wigmore TJ. Adjacent central venous catheters can result in immediate aspiration of infused drugs during renal replacement therapy. Anaesthesia 2012.67: 115–121.
12. Leblanc M, Fedak S, Mokris G, Paganini EP. Blood recirculation in temporary central catheters for acute hemodialysis. Clin Nephrol 1996.45: 315–319.
13. Little MA, Conlon PJ, Walshe JJ. Access recirculation in temporary hemodialysis catheters as measured by the saline dilution technique. Am J Kidney Dis 2000.36: 1135–1139.
14. Tan J, Mohan S, Herbert L, Anderson H, Cheng JT. Identifying hemodialysis catheter recirculation using effective ionic dialysance. ASAIO J 2012.58: 522–525.
15. Schetz M. Drug dosing in continuous renal replacement therapy: general rules. Curr Opin Crit Care 2007.13: 645–651.
16. Bohler J, Donauer J, Keller F. Pharmacokinetic principles during continuous renal replacement therapy: drugs and dosage. Kidney Int Suppl 1999.72: S24–S28.
17. Clark WR, Ronco C. CRRT efficiency and efficacy in relation to solute size. Kidney Int Suppl 1999.S3–S7.
18. Ferrer R, Martin-Loeches I, Phillips G, et al. Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: Results from a guideline-based performance improvement program. Crit Care Med 2014.42: 1749–1755.
19. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006.34: 1589–1596.
20. Shorr AF, Micek ST, Welch EC, Doherty JA, Reichley RM, Kollef MH. Inappropriate antibiotic therapy in Gram-negative sepsis increases hospital length of stay. Crit Care Med 2011.39: 46–51.
21. Zhang D, Micek ST, Kollef MH. Time to appropriate antibiotic therapy is an independent determinant of postinfection ICU and hospital lengths of stay in patients with sepsis. Crit Care Med 2015.43: 2133–2140.
22. Abdul-Aziz MH, Lipman J, Mouton JW, Hope WW, Roberts JA. Applying pharmacokinetic/pharmacodynamic principles in critically ill patients: optimizing efficacy and reducing resistance development. Semin Respir Crit Care Med 2015.36: 136–153.
23. Baquero F, Negri MC. Strategies to minimize the development of antibiotic resistance. J Chemother 1997.9: 29–37.
24. Johnston AJ, Steiner LA, O’Connell M, Chatfield DA, Gupta AK, Menon DK. Pharmacokinetics and pharmacodynamics of dopamine and norepinephrine in critically ill head-injured patients. Intensive Care Med 2004.30: 45–50.
25. Beloeil H, Mazoit JX, Benhamou D, Duranteau J. Norepinephrine kinetics and dynamics in septic shock and trauma patients. Br J Anaesth 2005.95: 782–788.
26. Pikwer A, Bååth L, Davidson B, Perstoft I, Akeson J. The incidence and risk of central venous catheter malpositioning: a prospective cohort study in 1619 patients. Anaesth Intensive Care 2008.36: 30–37.
27. Kowalski CM, Kaufman JA, Rivitz SM, Geller SC, Waltman AC. Migration of central venous catheters: implications for initial catheter tip positioning. J Vasc Interv Radiol 1997.8: 443–447.

central venous catheter; continuous renal replacement therapy; intensive care; clearance

Copyright © 2018 by the ASAIO