Skip Navigation


NDT Advance Access originally published online on March 27, 2006
Nephrology Dialysis Transplantation 2006 21(8):2191-2201; doi:10.1093/ndt/gfl068
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
21/8/2191    most recent
gfl068v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Unger, J. K.
Right arrow Articles by Rossaint, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Unger, J. K.
Right arrow Articles by Rossaint, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Original Articles: Dialysis and Transplantation

Relation of haemofilter type to venous catheter resistance is crucial for filtration performance and haemocompatibility in CVVH—an in vitro study

Juliane K. Unger1,2, Claudia Haltern2, Britta Portz2, Bernd Dohmen2, Axel Gressner3 and Rolf Rossaint2

1 Department of Comparative Medicine and Experimental Animal Sciences, Charité Campus Virchow Klinikum, Humboldt-University, Berlin, 2 Department of Anaesthesiology and 3 Department of Clinical Chemistry, Universitätsklinikum Aachen, Rheinisch-Westfälische Technische Hochschule Aachen, Aachen, Germany

Correspondence and offprint requests to: Juliane K. Unger, DVM, Department of Comparative Medicine and Experimental Animal Sciences, Charité Campus Virchow Klinikum, Augustenburger Platz 1, Forum 4, D-13353 Berlin, Germany. Email: juliane.unger{at}charite.de



   Abstract
Top
 Abstract
Introduction
 Subjects and methods
 Results
 Discussion
 Conclusions
 References
 
Background. Main factors for the overall performance of haemofilters (HF) are membrane features and filter durability without clogging/clotting of capillaries. However, the venous line resistance (Pv) is a powerful force for net filtrate flux resulting in haemoconcentration and thus is enhancing the phenomenon of membrane clogging. Therefore, we hypothesized that catheter type, as it is associated with Pv-levels, contributes to the extent in which filter longevity and filtration performance are restricted due to blocked hollow fibres.

Methods. Heparinized porcine blood (5 IU/ml) was circulated in an in vitro system for haemofiltration (FH6S®—filters were used, Ca. Gambro). Three different sizes of catheters for peripheral vein access (Vygonuele V®, Ca. Vygon) were alternately inserted into the circuit for blood return from the filter to the reservoir. To produce Pv-levels lower than commonly induced by Shaldon catheters, a 14G-vygonuele was used. Pv-levels standard for 11–12 French catheters were provided by using a 16G-vygonuele. To produce Pv-levels common for low-French or tri-lumen catheters, a shortened 18G-vygonuele was used. The respective Pv-levels attained were compared with respect to the overall filtration performance (system pressures, haemocompatibility and sieving coefficients).

Results. Catheters of 14 and 16G enabled transiently maximal blood flow (Qb)/filtration rates (Qf) of 300/60 ml/min and continuous filtration with Qb/Qf of 200/40 ml/min. The shortened 18G catheter reduced maximal flow rates down to Qb/Qf of 200/40 ml/min, and continuous flow rates down to Qb/Qf of 125/25 ml/min. At the end, median values for blocked hollow fibres were, 35% in the 14G-group, 40% in the 18G-group and 70% in the 16G-group. Haemocompatibility appeared to be higher in the 14G-group with respect to various parameters when compared with the other groups.

Conclusions. The flow resistance by the catheter chosen for haemofiltration clearly influenced the filtration performance. Thus, investigations focused on compatibility of catheter type as it related to Pv-levels with the particular method of renal replacement therapy that should be performed. This point could be crucial in reducing filter clogging and haemostasis during CVVH.

Keywords: haemocompatibility; haemofiltration; Shaldon catheters



   Introduction
Top
 Abstract
 Introduction
Subjects and methods
 Results
 Discussion
 Conclusions
 References
 
Apart from their established indication for dialysis, Shaldon catheters are routinely used for convection based renal replacement therapies (RRTs). Continuous research on how to improve catheter technology is focused on the avoidance of nosocomial, catheter related infections [1,2]; and engineering of double-lumen catheters is performed to minimize haemolysis and clot formation [3–5]. Also, the influence of catheter thrombogenecity on the filter's longevity was discussed [3]. Therefore, various aspects of the management of vascular access are already being considered, but the impact of the connection is still a matter of concern.

However, there is another very important point arising from former investigations. We found that membrane clogging as a result from red blood cell aggregation in convection-based filtration procedures is a dominant factor for reduced filter clearance, reduced haemocompatibility and filter clotting [6]. In a recent CVVH in vivo study performed by our group, membrane clogging was again the main reason for blocked hollow fibres, which confirmed our point of view concerning the aetiology of filter blockage [7,8]. Today, polymer membranes for RRT-filter types provide a very high spontaneous hydraulic capacity and net filtrate flux [9]. In addition to the membrane's specification for hydraulic capacity, the hydrostatic pressure inside the hollow fibres resulting from venous line resistance is another powerful driving force for the net filtrate flux. The venous line resistance induced by Shaldon catheters is high. This is due to their length and relatively small inner diameters. Both features, length and small diameter, are determined by the request for percutaneous puncture technique for central vein access (e.g. Seldinger technique) and body size/body mass of patients.

The problem of filter clogging due to haemoconcentration and cell aggregation in convection-based haemofilters led to a completely new aspect in vascular access management. Now, the question arises whether current principles in methods for vascular access, developed for diffusion based RRT, are compatible with the hydraulic capacity of new filter generations that are based on polymer membranes for convection-based ultrafiltration. Thus, we performed an in vitro study comparing the direct impact of the flow resistance of three different catheters paired with the same filter type, with respect to maximal flow rates, sieving coefficients, haemocompatibility and numbers of clogged hollow fibres.



   Subjects and methods
Top
 Abstract
 Introduction
 Subjects and methods
Results
 Discussion
 Conclusions
 References
 
Under in vitro conditions, three different pairs of haemofilter–catheter resistance were investigated.

Study groups
Since pressure levels, as they result from flow resistance rather than specific catheter engineering [5], were the point of interest, simple catheters/cannules for peripheral vein access (Vygonuele V®, Ca. Vygon, France) were used to mimic those Pv-levels that should be addressed with respect to the usual types of Shaldon catheters and an alternative providing lower Pv-levels. The groups were arranged as follows:

Group I (n = 6): 14 Gauge Vygonuele V®, length: 42 mm (alternative low Pv-level).

Group II (n = 6): 16 Gauge Vygonuele V®, length: 42 mm (~11–12 French, double lumen Shaldon catheter).

Group III (n = 6): 18 Gauge Vygonuele V® shortened to a length of 35 mm (Pv-levels as may be found with double lumen Shaldon catheter <11 French and >16 cm length).

Table 1 shows data for the Pv-levels of exemplary Shaldon catheter sizes measured in our whole blood in vitro system and/or in animal experiments; the corresponding sizes of the peripheral arm vein cannule, scheduled to represent the respective Pv-levels in this study, were also investigated under in vitro conditions and data are also shown in Table 1 for direct comparison.


View this table:
[in this window]
[in a new window]
  		Table 1. Pv-levels arising from venous line resistance of two Shaldon catheters

 
Parameters investigated
  1. System pressures: pressure above the filter (arterial pressure, Pa), pressure behind the filter arising from venous line and catheter (Pv), filtrate pressure (Pf), transmembrane pressure (TMP), pressure drop (Pa – Pv).
  2. Highest blood and filtration rates (Qb/Qf) which could be achieved.
  3. Sieving coefficients of creatinine, urea, glucose, sodium, potassium, calcium, albumin, LDH
  4. Haemocompatibility: platelet and leucocyte counts, concentrations of free plasma haemoglobin, antithrombin (ATIII) for methodological control, activated partial thromboplastin time (aPTT), prothrombin time (PT), fibrinogen concentrations, d-dimers (fibrinogen-cleavage product), thrombin-antithrombin-complex (TAT).
  5. Percentage of blocked capillaries at the end of experiment.

Thresholds for determination of maximal flow rates
We had two different aspects for highest flow rates possible: (i) highest flow rate that could be run under pressure control, and (ii) highest flow rate that could be performed without spontaneous haemolysis.

Materials and assays used for laboratory analysis
Laboratory analyses
Blood, plasma and filtrate samples were analysed using a blood counter Celltek M (Bayer Vital GmbH & Co KG, Munich, Germany); an ABL 510 and an EML 100 Radiometer (Copenhagen, Denmark); and Vettest 8008 (IDEXX GmbH, Wörrstadt, Germany). Spontaneous haemolysis was identified after centrifugation of blood samples by visual comparison of plasma supernatants with the plasma samples obtained at baseline. Thereafter, exact measurements to monitor haemolysis were performed by the determination of free plasma haemoglobin (fHb) (cyanide reaction, Diagnostica Merck, 9405, Darmstadt, Germany; UV photometer Ultraspec II, Biochrom LKB, Cambridge, UK) [10,11].

Plasma coagulation analysis
Extrinsic and intrinsic coagulation factors were analysed by measuring PT and aPTT. The aPTT was analysed with Dade® Actin®FS reagent, and PT with the Dade® Innovin® reagent (both reagents: Dade Behring, Marburg, Germany). For the determination of fibrinogen concentrations, the method of Clauss and the Fibrinogen a® reagent (Diagnostica Stago, Boehringer, Mannheim, Germany) were used. Polybrene® (Abbott Laboratories, Dallas, USA) was added to the original Actin®FS and Innovin® reagents in order to antagonize the high heparin dosage [10,11].

Blocked capillaries
At the end of the experiments, filters were rinsed with 3 l of normal saline using a flow rate of 200 ml/min, but the cannule inserted into the circuit for the time course of experimental protocol was removed to ensure similar basic flow resistance for the rinsing procedure. During the rinsing procedure, red blood cell aggregates were released from individual hollow fibres in all filters. Those aggregates were mainly dispensed in the normal saline used for filter rinsing. However, the use of similar volumes and flow rates for rinsing ensured that the blocked capillaries still actually contained the cell aggregates/clots. Thereafter, filters were cut open, the overall fibre bundle was visually examined and the percentage of blocked capillaries was estimated by two independent observations from two independent investigators. Blocked fibres appeared red due to the trapped erythrocytes. One of the two investigators was blinded for the study group. This procedure already proved to provide highly reproducible results for an estimation of the percentage of blocked fibres in former studies. Samples of blocked hollow fibres were stored in 4% buffered formalin solution until used in paraffin embedding and haematoxilin–eosin staining for microscopical examinations.

Blood
Porcine blood
Porcine blood was collected from healthy slaughterhouse pigs using a standardized sterile collecting system primed with heparin to achieve anticoagulation with 5 IU/ml of blood. The level of haematocrit (Hct) (35–40%) was chosen in accordance to the upper level of Hct that may be found during CVVH treatment of patients suffering from adult respiratory distress syndrome, acute liver failure, sepsis or in general at the end of treatment cycles. Furthermore, as in vitro systems based on whole blood require short protocols to avoid deteriorating blood quality, high Hct levels were also chosen to accelerate mechanisms hypothesized for this study. Duration of the experiment from priming to final value was 3 h.

Apheresis
Extracorporeal in vitro circuit (Figure 1)
A closed in vitro circuit was assembled [10,11] using a clinically proven plasma filtration pump (AK10, Gambro Hechingen, Germany). The haemofilter used was the FH 6S from Ca. Gambro (membrane: Polyamid STM, surface is 0.6 m2, inner diameter of fibres is 215 µm, effective length is 140 mm and wall thickness is 50 µm). Standard bloodlines were used. Arterial and venous lines were connected to a 2 l-blood reservoir that was stored in a heated water bath (37°C). The outflow of the filtration line was connected to the venous bubble trap allowing for a recirculation mode for both blood and filtrate. Using the commercially installed electronic-transducers of the apheresis device (AK10®) pressure values were continuously measured. The average pressure inside the filtrate compartment (Pf, filtration pressure) was measured with additional control-transducers (Transpac, Abbott GmbH, Wiesbaden, Germany) and a multichannel recorder outside the apheresis monitor (model S 66®, Hewlett Packard, Bad Homburg, Germany).


Figure 1
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. Principle of the in vitro circuit. The figure shows the principle of experimental circuit. Pa, Pv, TMP and Pf were measured as indicated according to the procedure of clinical application. Haemofiltration was performed for a total of 3 h for each experiment. Pa: arterial pressure, Pv: venous pressure, TMP: transmembrane pressure, Pf: average pressure inside the filtrate compartment.

 
Priming of the system
The in vitro circuit was rinsed with 2 l of normal saline. Exchange of the preload with normal saline against blood was performed in single pass mode wasting effluent as long as it still contained rinsing solution. Thereafter, blood was pumped from the collecting bag into the in vitro system via the filter. The recirculation mode was initiated when the scheduled system volume of 2 l of blood was achieved.

Protocol (Figure 2)
After the system was primed with blood (Qb 100 ml/min) for 20 min, the first flow rates were adjusted according to the manufacturer's recommendation for the filter type used. Blood flow was started with 125 ml/min (Qb) and filtrate pump (Qf) was started with 25 ml/min. These flow rates were maintained for 10–15 min, which ensured stable pressure values for 5–10 min and equilibration of fluid composition in both compartments of blood and filtrate. At this time, baseline (BS) measurements were performed. Thereafter, blood flow and filtration rates were increased to Qb/Qf of 200/40 ml/min and maintained. After further 20 min, measurements were repeated and again an increase of flow rates was performed (Qb/Qf 300/60 ml/min). Since Qb/Qf rates of 300/60 ml/min led to pressure levels exceeding the highest level measurable by the AK10 Monitor in the 18G-group, it was not possible to increase the "flow rate" in the study group with the highest Pv-level (18G-group). Thus, in this group the procedure was continued with 200/40 ml/min during the following 20 min. In the other two groups (14G, 16G) the 20 min period for the Qb/Qf setting of 300/60 ml/min could be completed and pressure recording and measurements over. Thereafter, in each experimental setting the highest flow rates, which had provided stable pressure values for 20 min, were reset and maintained for another 90 min after which final values (FV) were determined. In general, all measurements were initiated with the documentation of the pressure values and completed with a parallel sampling of blood and filtrate to obtain the required material for blood/plasma- and filtrate-analyses.


Figure 2
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 2. Time course of experimental protocol. The figure shows the principle of experimental protocol in context with the time spans needed and/or scheduled for the measurements or priming and changes in blood flow and filtration rates, respectively. Documentation of pressure levels for the respective flow rates and sampling procedures for laboratory analyses for sieving coefficients and haemocompatibility data needed 5 min during which the same flow rates were still maintained.

 
Statistical analyses
As a normal distribution could not be expected for n = 6/group, non-parametric tests were taken for further statistical analyses. Statistical inter-group comparison was performed using the Mann–Whitney U-test for unpaired samples. Intra-group analysis of haemocompatibility parameters to compare baseline (BS) and the final values (FV) at the end of the experiment was performed using the Wilcoxon rank sum test for paired samples. Friedman-MANOVA for repeated measures was used for intra-group comparison of the measurements performed at all different flow rate settings (pressure values and sieving coefficients). A probability value of P<0.05 was determined to indicate statistical significance. Analyses were performed using NCSS 2000. Sigma Stat 2.0 was used for correlation analyses. For normal distributed data the Pearson product moment correlation test was chosen to determine the correlation coefficient and P-levels; the Spearman rank order test was used for variables that failed normal distribution but with equal variance.



   Results
Top
 Abstract
 Introduction
 Subjects and methods
 Results
Discussion
 Conclusions
 References
 
Comparison of initial blood composition
Baseline analytical data from the laboratory did not differ between the groups with respect to blood composition (Table 2, BS-values).


View this table:
[in this window]
[in a new window]
  		Table 2. Blood cell counts and data of coagulation system for specification of haemocompatibility

 
Maximal flow rates and pressure values
In the 14G- and the 16G-groups pressure levels were stable for 125/25 and 200/40 ml/min (Figures 2 and 3). A further increase of the flow rates to 300/60 ml/min led to the continuously increasing TMP-values (data in the figures are showing values after 20 min). In consequence, the lower Qb/Qf-setting of 200/40 ml/min was chosen as highest flow rate possible for the filtration cycle with constant flow rates over 1.5 h scheduled for the second half of the experimental protocol. In the 18G-group, even 200/40 ml/min led to rapid increases in pressure levels exceeding a Pa-level of 420 mmHg, which was associated with an automated stop of the blood pump. Thus, concerning the 18G-group, only flow rates of 125/25 ml/min could be operated in the second period based on constant flow rates for 1.5 h (Figures 2 and 3).


Figure 3
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 3. Pressure values indicating flow resistance of catheter or haemofilter, respectively. The figure demonstrates the development of the Pv-levels (upper panel), the accompanying Pa-levels (middle panel) and the resulting pressure drop over the filter module (Pa – Pv) (lowest panel). Data are shown in box blots which represent the following values: median (bold line in the box), 25 and 75% quartiles (lower and upper end of the box), minimum and maximum (black circles). *Indicates significant differences, P<0.05.

 
Pa- and Pv-levels
Pa-values and thus flow resistance at the blood inflow port of the haemofilter, were dominated by differences in the Pv-levels given by the different types of venous catheters (Figure 2). While significant differences occurred for Pv- and Pa-levels, initially the pressure drop (Pa – Pv) demonstrated equalized values, which was reflected in the differences in Pf-values (Figure 4). However, at the end of the experiment even the pressure drop became significantly different between the groups, showing highest flow resistances by the filter module in the 16G-group.


Figure 4
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4. Pressure levels resulting from filter specifications and flow resistances. The figure shows the development of TMP and Pf-values. In this setting, a filtrate pump was used, thus Pf is an indirect parameter for net filtrate flux. Data are shown in box blots which represent the following values: median (bold line), 25 and 75% quartiles (lower and upper end of the box), minimum and maximum (black circles). *Indicates significant differences, P<0.05.

 
Filtration pressure (Pf)
As a filtrate pump was set with a given flow rate, the pressure development inside the filtrate compartment was an indirect parameter for net filtrate flux. The 18G-group demonstrated highest filtrate pressures (Figure 4) with respect to lower flow rates at an early point of filtration cycle. However, at the end of the experiment, which meant the end of an overall 3 h-filtration cycle, significance was lost. Interestingly, Pf-levels in the 14G-group were lower at the beginning of the experiment than those found in the other groups and remained more constant until the end of the experiment. In all groups, the Pf-values decreased in the second half of the experiment when flow rates were kept constant, but changes were not significant for intra-group comparison of time points. Nonetheless, the 16G-group, which provides Pv-levels comparable to an 11–12 French, double-lumen Shaldon catheter, demonstrated the highest range of Pf-levels and more filters within that group developed negative Pf-values than in the other groups. Interestingly, independent of the study group, negative Pf-levels were not associated with deterioration in haemocompatibility parameters when comparisons of individual results of each single experiment were performed within the groups (i.e. higher rates of haemolysis or decrease of platelets).

Transmembrane pressure (TMP)
The TMP is a resultant from all other pressure values, including the resulting developments in flow dynamic inside the hollow fibres of the haemofilter. While Pv- and Pa-levels (Figure 3) mirrored the differences in flow resistances of the different sizes of the venous catheters, the TMP (Figure 4) did not allow a clear differentiation of the different flow dynamics inside the filter. However, the TMP-values of the 14G-group were significantly lower at the end of experiment, when compared to the 16G-group.

Blocked hollow fibres
Though not significant, due to overall high ranges, a tendency for group-related ratios in the percentage of blocked hollow fibres at the end of the experiment was found with lowest levels in the 14G-group [median (min–max) 35 (5–70)%], followed by the 18G-group with 40 (30–80)% and in tendency the highest number of 70 (40–99)% in the 16G-group. Thus, the highest number of blocked hollow fibres was also associated with highest TMP- and Pa – Pv-levels at the end of the filtration cycle.

In Figure 5, correlation analysis is shown for the percentage of blocked hollow fibres at the end of the experiment and the final pressure levels. The 16G-group demonstrated the strongest correlation for the impact of blocked hollow fibres on the pressure drop over the filter module (Pa – Pv). With respect to TMP thresholds, which have to be considered during a clinical application, the 14G-group could be operated with higher flow rates than the 16G-group. Notably, analysis throughout all groups indicates a negative correlation between TMP and Pf (r = –0.862, P<0.0001), which is in context with negative Pf-values. The Pa – Pv-values are also negatively correlated with the Pf-values (r = –0.696, P = 0.00114).


Figure 5
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Linear regression of pressure levels vs percentage of blocked fibres. Values given for correlation coefficient and P-levels inside the scatter plot are related to n analysis based on grouped data irrespective of the allocation to a study group. The data points were allocated to the study group of origin in order to allow insight in the different impact of the individual groups on the overall regression analysis. The r-values for the respective group are given in the symbol-legend in the respective figure.

 
In Figure 6, an exemplary microscopic picture is shown to demonstrate the nature of hollow fibre blockage in that study. A dense red blood cell aggregate was formed. During standard processing of the sample, parts of the red blood cell aggregate were artificially removed from the intraluminal space of the fibre due to the lack of firmness that would be usual for a clot.


Figure 6
View larger version (157K):
[in this window]
[in a new window]
  		Fig. 6. Light microscopic picture of a blocked hollow fibre. Longitudinal cross section, paraffin embedding and haematoxilin–eosin staining, magnification: 400x. Gelatine: hollow fibre samples were flushed with 4% buffered formalin solution for fixation. Thereafter, gelatine was added to the samples to provide sample features allowing standard processing of the material. Parts of the erythrocyte aggregate were artificially removed from the hollow fibre lumen during standard processing of the sample.

 
Coagulation system
Since the removal of ultrafiltrate during filtrate sampling leads to a haemoconcentration, coagulation and fibrinolysis factors have to increase with time on a regular basis in case of inert haemofiltration procedure. In consequence of increasing concentrations of coagulations factors, clotting times should regularly decrease. In all study groups, ATIII-values and D-dimers were significantly increased at the end of the experiment. Results for fibrinogen and TAT differed between the groups.

A significant decrease in both measured standard coagulation tests (aPTT and PT) was exclusively seen in the 14G-group.

Blood cell compatibility
In a closed system with recirculating in vitro haemofiltration, two reasons for increasing concentrations of free haemoglobin are given related to the method: (i) haemoconcentration during ultrafiltrate sampling and (ii) haemolysis due to in vitro situation and the use of roller pumps. In this study, we did not find differences in the rate of haemolysis between the three study groups.

Neither group revealed significant decreases in platelet counts. However, the use of a 16G-cannule, demonstrated a significant decrease in white blood cell counts.

Sieving coefficients (SC)
Sieving coefficients (SC) (Table 3) for albumin and also for lactate dehydrogenase (LDH) with the molecular weight of ~150 000 Da were comparable in all groups.


View this table:
[in this window]
[in a new window]
  		Table 3. Sieving coefficients (SC)

 
Very low, but significant differences between the three groups occurred for the SC of potassium. However, increases in potassium-SC were paralleled by increases in TMP and a decrease in net filtrate flux. In spite of the negative correlation between Pf and TMP, as well as Pf and Pa – Pv, SC pattern remained relatively stable. This indicates that net filtrate flux depended on the number of blocked hollow fibres rather than on the grade of membrane fouling.



   Discussion
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
Conclusions
 References
 
We found three main results in our study: (i) Pa-values followed the differences in Pv-levels induced by the respective flow resistances of the different catheter sizes; (ii) the clinically used parameter TMP failed as a functional indicator for an influence of the given flow resistance by the venous catheter; and (iii) based on the use of the filter type chosen for this study, the respective combination of filter and the lowest Pv-levels investigated led to best results with respect to all points of question, i.e. filtration performance, haemocompatibility and percentage of blocked hollow fibres at the end of the experimental haemofiltration. In contrast, the peripheral vein catheter (16G) that was used to mimic Pv-levels comparable to the clinical situation, when 200–220 mm long 11–12 French Shaldon catheters are used at haematocrit values between 30–35%, led to a rapid deterioration of filter function and a high number of blocked hollow fibres.

Methods
In vitro data have to be proven for their relevance under in vivo conditions. However, this study is based on the use of an in vitro system [6,10–12], which could meanwhile be proven to provide suitable information and conclusions for the in vivo situation [7,8,13,14]. The in vivo study confirmed results for the impact of a respective colloid on the coagulation system during haemofiltration and the blockage of the hollow fibres. First results of this comparison were recently published in scientific meetings and are accepted for publication elsewhere [7,8,13,14]. Therefore, reliability can be assumed for a first assessment of the importance of the present in vitro study.

Interpretation of coagulation results
In vitro haemofiltration studies are complex to interpret with respect to the resulting coagulation pattern. This is based on two aspects. The first aspect is the high heparin dosage required for in vitro studies based on whole blood use [12]. This anticoagulation procedure reduces the possible range of clotting times in case of significant activation or impairment of coagulation. The latter clearly reduces the obvious relevance of changes even if significant. The second aspect is the removal of ultrafiltrate from the in vitro system for laboratory analysis. In the particular case of haemofiltration, this leads to haemoconcentration including all coagulation factors. High plasma concentrations of coagulation factors are associated with short clotting times of functional coagulation tests. Thus, in the case of an inert haemofiltration procedure, clotting times could be expected to moderately accelerate (decrease) due to the overall haemoconcentration. In consequence, optimal results were expected to show a slight decrease in aPTT and PT, an increase in fibrinogen values and stable TAT-values. We found unfavourable changes in coagulation parameters that occurred in the two groups based on as yet, clinically relevant Pv-levels (16G-, 18G-group). Most likely, these results indicate a consumption of coagulation factors and activation of coagulation pathways.

Transfer of results and methods to clinical situation
At this point of research, it is not possible to recommend a particular ‘one for all’ level of venous line and/or catheter resistance. Probably, this will remain a question of the individual combinations of haemofilter size, aimed/achieved blood delivery rates, flow resistance of the tubing and rheological properties such as haematocrit and plasma protein composition. Nonetheless, flow resistance lower than usual seems to be of advantage.

Context to research on filter's running times
Yet, research on optimization of filter's long life is associated with research on anticoagulation strategies [15–17]. Concentrating on anticoagulation means that clotting is assumed to be the main trigger for a total blockage of hollow fibres, although coagulation most often is severely impaired in critically ill patients. Interestingly, we found that cell aggregation, and thus membrane clogging is an important key in the vicious cycle of blocked hollow fibres [6,14]. Multiple organ failure and SIRS/sepsis do not exclusively lead to activation and malfunction of the coagulation system. The overall changes in homoeostasis of blood composition—such as increasing fibrinogen levels, reduced albumin/globulin ratio, decreased haematocrit or post-haemorrhage plasma dilution—also lead to deteriorated haemorheology [18,19]. In turn, red blood cell aggregability is increased and becomes an additional trigger for the rising risks of membrane clogging [6,10,14,19]. In this study, based on equal heparin dosage and rheological features of the blood in all study groups, deterioration of net filtrate flux, higher TMP values and pressure drops above the filter module, correlated with the percentage of clogged hollow fibres. Since the percentage of clogged hollow fibres clearly tends to differ between the study groups, Pv-levels given by the overall CVVH-system should be assumed to importantly influence the filtration performance and in turn filter running times. Therefore, another confirming result is given for the observation that there are important factors for filter's long life other than the anticoagulation regime. Baldwin and colleagues [20] described the reduction of real blood flow vs blood flow assumed to achieve by the blood pumps. They discussed the problem of intracapillary haemoconcentration due to the mismatch of the flow rate set for the filtration pump, which was chosen based on the intended blood flow rate and the blood flow rate that really results from the equipment. They found the filter long life negatively correlated with the percentage of blood flow reduction by the pumps.

Taken together, as a consequence of our results [6,10,14], the study of Baldwin et al. [20], and given by the ‘state of the art’ membrane technology, there is a triad of important factors contributing to the blockage of filter membranes: (i) increased red blood cell aggregability due to blood composition [6,10,14], (ii) low blood velocities inside the filter capillaries [20] and (iii) membrane type-related high net filtrate flux. In this context, it appears to be quite plausible that a particular intracapillary hydrostatic pressure, and in turn a particular Pv-level in the overall CVVH-system, required to enable a balance of suitable net filtrate flux and avoidance of membrane clogging.



   Conclusions
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Conclusions
References
 
With respect to our results and those of other studies in relation to short filter's running times and/or low filter clearance, more factors than coagulation are importantly involved in the blockage of the hollow fibres of haemofilters. Based on the use of high hydraulic capacity-membranes, the combination of low blood velocities and deteriorated haemorheology easily leads to membrane clogging. As shown in the present study, high Pv-levels, and in turn high intracapillary hydrostatic pressures, may contribute to a multifactorial aetiology of reduced filter running times. Reduction of Pv-levels via modifications in the venous part of the overall CVVH-system could become an interesting tool to escape from the problem of filter blockage without any relation to anticoagulation protocols. Therefore, Pv-levels should be involved in a more comprehensive concept for improvements in haemofilter performance and biocompatibility.



   Acknowledgments
 
This study was supported by ‘START-RWTH Aachen’, Gambro Dialysatoren, Hechingen, and Else Kröner-Fresenius-Foundation. Furthermore, we cordially thank our laboratory assistant Renate Nadenau and our students Wolf Siepen and Andrea Heuer for their committed and professional support throughout this study.

Conflict of interest statement. None declared.



   References
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Conclusions
 References
 

  1. Quarello F, Forneris G. Prevention of hemodialysis catheter – related bloodstream infection using antimicrobial lock. Blood Purif 2002; 20: 87–92[CrossRef][Web of Science][Medline]
  2. Wester JP, de Koenig EJ, Geers AB et al. Catheter replacement in continuous arteriovenous hemodiafiltration: the balance between infections and complications. Crit Care Med 2002; 30: 1261–1266[CrossRef][Web of Science][Medline]
  3. Davenport A. The coagulation system in the critically ill patient with acute renal failure and the effect of an extracorporeal circuit. Am J Kidney Dis 1997; 30: S20–S27[Web of Science][Medline]
  4. Walters GK, Jescowitch AJ, Jones HA, Jones CE. A comparison of the dual lumen and coaxial catheters for temporary hemodialysis access. Int J Artif Organs 1997; 20: 208–212[Web of Science][Medline]
  5. Tan HK, Bridge N, Baldwin I, Bellomo R. An ex-vivo evaluation of vascular catheters for continuous hemofiltration. Ren Fail 2002; 24: 755–762[CrossRef][Web of Science][Medline]
  6. Unger JK, Haltern C, Dohmen B et al. Albumin and hydroxyethyl starch 130kD/0.4 improve filter clearance and hemocompatibility in hemo- and plasmafiltration – an in vitro study. Nephrol Dial Transplant 2005; 20: 1922–1931[Abstract/Free Full Text]
  7. Kobela JB, Knopf A, Eger L, Grosse-Siestrup C, Unger JK. HES130/0.4 improves in vivo hemocompatibility during CVVH when compared to HES200/0.5. Int J Artif Organs 2005; 5: 521, presentation on the EWM 2005
  8. Eger L, Knopf A, Kobela JB, Haltern C, Grosse-Siestrup C, Unger JK. Albumin substitution may reduce heparin dosage for CVVH in pigs – confirmation of in vitro investigations. Int J Artif Organs 2005; 5: 521, presentation on the EWM 2005
  9. Uhlenbusch-Körwer I, Bonnie-Schorn E, Grassmann A, Vienken J. Understanding membranes and dialysers, In: Vienken J, ed. Lengerich Good Dialysis Practice (Vol 5), Papst Science Publishers: 2004: 103–172
  10. Unger JK, Horn NA, Kashefi A, Blumberg A, Klosterhalfen B, Rossaint R. The influence of hypoalbuminemia on maximal flow rates and transmembrane pressure during plasmapheresis – an in vitro study. Blood Purif 2001; 19: 408–416[CrossRef][Web of Science][Medline]
  11. Unger JK, Janssen VR, Kashefi A et al. Enhancing filtration rates by the use of blood flow around the capillaries of plasmafilters – an in vitro study. Int J Artif Organs 2001; 11: 821–831
  12. Unger JK, Haltern C, Dohmen B, Rossaint R. Influence of different heparin concentrations on the results of in vitro investigations in plasmaseparation technology using capillary membrane filters. Artif Organs 2003, 27: 649–657[CrossRef][Web of Science][Medline]
  13. Unger JK, Eger L, Haltern C, Kobela JB, Knopf A, Grosse-Siestrup C. Akute Hämokompatibilitätsinteraktion von Kolloiden und Hämofiltration: in vitro vs. in vivo. Journal für Anästhesie und Intensivbehandlung 2005; II: 190, Pabst Science Publishers, presentation on the HAI 2005
  14. Unger JK, Haltern C, Kobela JB, Francis R, Rossaint R, Grosse-Siestrup C. HES130kDa/0.4 and albumin improve CVVH-biocompatibility while gelatin and HES200/0.5 lead to adverse side effects of CVVH in anesthetized pigs. Shock 2006, in press
  15. Turney JH, Fewell MR, Williams LC, Parsons V, Weston MJ. Platelet protection and heparin sparing with prostacyclin during regular dialysis therapy. Lancet 1980; 2: 219–220[Web of Science][Medline]
  16. Schrader J, Stibbe W, Kandt M et al. Low molecular weight heparin versus standard heparin. A long-term study in hemodialysis and hemofiltration patients. ASAIO Trans 1990; 36: 28–32[Medline]
  17. De Pont AC, Oudemans-van Straaten HM, Roozendaal KJ, Zandstra DF. Nadroparin versus dalteparin anticoagulation in high volume, continuous venovenous hemofiltration: a double-blinded, randomized, crossover study. Crit Care Med 2000; 28: 421–425[CrossRef][Web of Science][Medline]
  18. Maeda N, Shiga T. Opposite effect of albumin on the erythrocyte aggregation induced by immunoglobulin G and fibrinogen. Biochem Biophys Acta 1986; 855: 127–135[Medline]
  19. Stefanidis I, Heintz B, Frank D, Mertens PR, Kierdorf HP. Influence of hematocrit on hemostasis in continuous venovenous hemofiltration during acute renal failure. Kidney Int Suppl 1999; 72: S51–S55[Medline]
  20. Baldwin I, Bellomo R, Koch B. Blood flow reductions during continuous renal replacement therapy and circuit life. Intensive Care Med 2004; 30: 2074–2079[CrossRef][Web of Science][Medline]
Received for publication: 4. 9.05
Accepted in revised form: 6. 2.06


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



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
21/8/2191    most recent
gfl068v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Unger, J. K.
Right arrow Articles by Rossaint, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Unger, J. K.
Right arrow Articles by Rossaint, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?