In hemodialysis, catheter complications are a matter of substantial concern. Thrombus formation is a major source of clinical complications, in addition to bacterial adhesion, biofilm formation, and subsequent infections, and leads eventually to early treatment failure. Thus, short catheter survival also induces high costs. Several antimicrobial coating techniques have been developed to reduce bacterial adhesion and infection. 1,2 Although thrombus formation is acknowledged as the key problem for short catheter survival, only a few studies have investigated this entity, mainly on the basis of the impact of blood–material contact. Surface properties turned out to play a key role for activation of the coagulation system and adhesion of clot components such as fibrin net or platelet aggregates. 3–5
We hypothesized that the improvement of blood–surface interaction by a reactive polymer film coating used in vitro might reduce thrombogenic events in the vascular access device and subsequently lead to prolonged catheter survival in the clinical setting. We therefore compared, in a randomized observational study, the clinical application of conventional and surface modified catheters with identical geometry and flow design with respect to catheter survival as the primary endpoint.
A novel coating technology provides the surface with a micropatterned structure of hydrophilic-hydrophobic domains in a polymer system containing hydrophobic polydimethylsiloxane (PDMS) blocks in a polyurethane net. 6 Micropatterned surfaces have been shown to improve blood compatibility. 7,8 As depicted in Figure 1, atomic force microscopy analysis shows a substantial difference between the standard catheter surface and the domain structured reactive polymer film.
Methods and Patients
In Vitro Assessment of Thrombogenicity
Surface thrombogenicity of the standard and modified catheters was assessed simultaneously in an in vitro recirculation perfusion system at 37°C with freshly donated platelet rich human ACD plasma. The shear rates provided in the experimental setting were comparable with the clinical situation (200 ml/min). Calcium was added to trigger coagulation, and plasma samples were taken at different points of time up to 90 minutes and frozen at −70°C until analysis. The formation of thrombin-antithrombin (TAT) III complex was measured in once-thawed samples to assess the extent of surface induced coagulation activation using a commercially available enzyme linked immunoabsorbent assay kit (Behring, Marburg, Germany) according to the manufacturer’s instructions.
Twenty patients treated in Essen University Hospital who required intermittent hemodialysis therapy for acute (n = 9) or chronic (n = 11) renal failure were randomly assigned to application of a standard or a surface modified catheter. Five patients dropped out because of accidental catheter removal (n = 1) or transfer to another department (n = 4). Patient characteristics of the remaining 15 patients are summarized in Table 1.
Standard or low molecular weight heparin was used for anticoagulation according to body weight. Exclusion criteria were drug therapy affecting coagulation (e.g., vitamin K antagonists, fibrinolytic therapy), continuous hemodialysis modes, hemodynamic instability, and age less than 18 or greater than 80 years.
The following catheter types with a length of 12.5 or 15 cm were applied during the clinical observation: standard double lumen catheters (STD-DC) (GamCath GDK-1112.5 or GDK-1115, Gambro, Hechingen, Germany) and film coated domain structured double lumen catheters (FCDS-DC) (GamCath Dolphin GDK-1112.5/SMA or GDK-1115/SMA, Gambro, Hechingen, Germany). All catheters were was inserted into the right vena jugularis.
As depicted in Figure 2, modified catheters are standard double lumen catheters coated with a 5 to 50 μm layer of a novel reactive polyurethane copolymer. This coating provides a surface with micropatterned hydrophobic and hydrophilic domains, markedly reducing the interaction with proteins and cells. 7 The coating is applied on all surfaces of the catheter, both outer and inner.
Treatment Modes and Monitoring
Intermittent hemodialysis was performed using a volume controlled Fresenius 4002 machine (Fresenius, Bad Homburg, Germany) with either a high flux (F60, surface area 1.3 m2) or a low flux polysulfone hemodialyzer (F6, surface area 1.3 m2) (Fresenius, St. Wendel, Germany). Technical setup and type of blood lines were kept constant throughout the study period. Blood flow rates and blood pressures were recorded at start and end of each treatment. In addition, thrombocyte counts, fibrinogen, antithrombin III (AT-III), and activated clotting time (ACT) were determined on each day of treatment to assess the patients’ state of coagulation. Catheter thrombosis was defined as continuous blood flow drop below 200 ml/min. Patients routinely received volume expansion and repositioning of the catheter. If blood flow remained below 200 ml/min, catheters were removed for electron microscopic studies.
If not otherwise specified, data are given as means ± standard deviation. To assess statistically significant differences, t-tests and Fisher-Yates tests were performed. p < 0.05 was considered statistically significant.
In Vitro Assessment of Surface Thrombogenicity
As depicted in Figure 3, the formation of TAT complex is markedly reduced and delayed in the modified (FCDS-DC) compared with the standard catheter (STD-DC).
Twenty patients were included in the study. During the study period, five patients dropped out. Of these, four were transferred to other units, and one patient accidentally removed the catheter himself. Reasons for drop out were thus not catheter related.
Data were evaluated from seven patients during treatment with the standard catheter (STD-DC) and from eight patients treated with the surface modified catheter (FCDS-DC). Hemodialysis in patients with chronic renal failure was performed three times a week. In patients with acute renal failure, the frequency was modified. However, overall hemodialysis sessions per days of inserted catheter were comparable (STD-DC 0.38 vs. FCDS-DC 0.41; p > 0.05).
Catheter survival was significantly prolonged with FCDS-DC compared with STD-DC (14.5 ± 11.7 days, range 2–39 vs. 10.3 ± 11 days, range 3–30; p < 0.05) as assessed with the Fisher-Yates test (categories: catheter survival < 7 days and ≥ 7 days). Significantly more treatments could be performed with a single catheter when using FCDS-DC compared with STD-DC (7.1 ± 3.4, range 2–12 vs. 4.3 ± 4.6, range 1–12; p < 0.05) (Fisher-Yates test, categories: < 4 and ≥ 4 treatments).
Duration of treatments (4.5 ± 0.22 hours, range 4–5 hours vs. 4.3 ± 0.15 hours, range 4–5 hours; p > 0.1), venous pressures (start 103 ± 30 mm Hg vs. 93 ± 41 mm Hg, p > 0.1; end 95 ± 37 mm Hg vs. 81 ± 39 mm Hg, p > 0.1), and blood flow rates (start 209 ± 29 ml/min vs. 210 ± 43 ml/min, p > 0.1; end 203 ± 24 ml/min vs. 202 ± 36 ml/min, p > 0.1) were similar with FCDS-DC and STD-DC. Reasons for catheter removal were a representative cross section of usual clinical requirements.
Concerning anticoagulation, the heparin bolus at start of dialysis was not significantly different with FCDS-DC and STD-DC (1,147 ± 493 vs. 1,017 ± 383 IU; p > 0.2). The continuous heparin dose was significantly higher with FCDS-DC compared with STD-DC (1,000 ±398 IU/h vs. 793 ± 190 IU/h; p < 0.02). However, in relation to body weight, the difference was no longer significant (Table 1). Furthermore, there was a mismatch in the distribution of single values between the treatment groups (FCDS-DC 51 vs. STD-DC 29 single value). In the FCDS-DC group, more patients received more than three treatments as compared with those treated with STD-DC.
To assess the state of coagulation, we investigated AT III, fibrinogen, thrombocyte counts, and ACT. As depicted in Table 1, no significant differences between patients treated with FCDS-DC and STD-DC were observed for thrombocyte counts (204 ± 52 × 1,000/μl vs. 243 ± 130 × 1,000 /μl; p > 0.1) or ACT (146 ± 19 s vs. 144 ± 20 s; p > 0.6), whereas AT III and fibrinogen were significantly different in the two study groups. However, all means were in the normal range (AT III 80–120%, fibrinogen 200–400 mg/dl, thrombocyte count 150–350 × 103/μl, ACT 96–152 s). AT III and fibrinogen were lower with FCDS-DC compared with STD-DC (84.6 ± 12.7%vs. 92.7 ± 10.1%, p < 0.03; 381 ± 103 mg/dl vs. 436 ± 58 mg/dl, p < 0.04). Again, the distribution of single values between the groups was nonequal (e.g., fibrinogen: FCDS-DC 27 vs. STD-DC 21 single values) because more patients treated with FDCS-DC received more than three treatments as compared with those treated with STD-DC.
In the present prospective observational pilot study, we compared two different types of central venous double lumen catheters with respect to functional application characteristics (i.e., number of hemodialysis sessions and catheter survival). Patients included in the study had their first vascular access because of need for dialysis treatment. When analyzing the number of treatments performed with a single catheter, the surface modified catheter was superior to the standard catheter. In addition, the mean application period until catheter removal was significantly elevated with the surface modified catheter. Although the study was performed with a limited number of patients, the difference with respect to number of possible treatments as well as catheter lifetime were both statistically significant, which points to an obvious difference in clinical behavior of these two materials. It is known that many factors can affect catheter survival and thrombosis of the access site. Therefore, great care has been taken to limit the number of possibly interacting confounding factors (e.g., technical setup and anticoagulation regime were similar in both groups). To investigate patients with acute and chronic renal failure, we had a special interest in patients with similar characteristics, including altered dialysis strategy in acute and chronic renal failure, origin of renal failure, and hemodynamic stability.
To assess the state of coagulation in each patient and at each treatment, we investigated AT III, fibrinogen, thrombocyte counts, and ACT. No significant difference was noted for ACT and thrombocyte counts. Small but statistically significant differences were found for AT III and fibrinogen levels. These data were collected for each individual treatment, and, therefore, they contain a bias with respect to patients being treated for different periods of time. Distribution of single values between the groups was unequal because more patients received more than three treatments in the group with modified catheters compared with those treated with standard catheters. Furthermore, it is important to note that all means of the parameters assessing coagulation—AT III, fibrinogen, thrombocyte counts, and ACT—remained in the normal range.
The quality of central venous catheters as temporary vascular access is of incremental importance because of increasing incidences of renal replacement therapies in acute renal failure or of comorbid conditions such as bad vascular situations (e.g., in patients with diabetes). Catheter related clinical complications such as occlusive events, severe infections, and subsequent prolongation of hospital stay are a considerable strain for affected patients and an important expense factor in health care.
It is known from a number of publications that interaction of blood components (i.e., plasma proteins from coagulation and complement cascades) as well as thrombocytes or polymorphonuclear cells with synthetic materials trigger a sequence of events leading to proinflammatory and procoagulatory signals. 8–10 It has not been considered before what role catheter modifications might play in improving vascular access situations in the clinical setting. Vanholder and Ringoir 11 describe the dilemma of finding suitable polymer modifications for catheter materials. Polyurethane is known to be quite a biocompatible material, especially with respect to activation of the intrinsic and extrinsic pathways of the coagulation system. Over the past years, different modifications have been proposed to improve catheter materials. Impregnation of catheters with antimicrobial as well as with anticoagulatory substances such as heparin did not bring significant improvements in the clinical setting. 12 Canaud 13 recently called for actions to improve hemodialysis related complications, especially infections, and stressed the need for developing new technology for polymer and catheter manufacturing with thromboresistant and infection-resistant properties. In our protocol, great care is taken to handle catheters aseptically, and catheter handling and protocol of catheter care were identical in both groups.
It is not an easy task to balance the different sources of biologic responses to nonphysiologic surfaces in contact with blood. 14 The catheter modification tested in this clinical study for the first time consists of a domain patterned structure of hydrophilic-hydrophobic domains in a polymer system containing hydrophobic PDMS blocks in a polyurethane net. Atomic force microscopy analyses showed a substantial difference between the standard catheter surface and the domain structured reactive polymer film. In vitro studies revealed that the new polymer coating shows lower and delayed start of thrombin formation, indicating thrombosis resistance. Thrombogenicity in an extracorporeal circuit depends on many factors that may confound the final result of fibrin or platelet deposits. In contrast with bioactive modifications by heparin or other anticoagulatory substances, the microdomain approach used here is likely to prevent the deposition of protein and cells at the surface and, by this, clot formation in the catheter.
Twardowski and Moore 5 recently demonstrated that, because of manufacturing conditions, side holes and surfaces of industrially manufactured catheters show quite significant pyrogenic surfaces in the range of several micrometers because of mechanical shaping of the catheter design. 5 In the catheters tested here, a polymer film of approximately 5 to 50 μm is applied in a final step to the standard catheter geometry. Because of this procedure, the surface is more smooth and prevents the deposition or damage of blood cells that is observed in areas with sharp edges in uncoated devices. The initial clinical findings support the concept that improvement of surface roughness as well as introducing a nonadhesive biopassive surface may result in significant improvement of catheter characteristics in the clinical setting.
In 1985, Hecker 15 had already identified roughness as a leading cause of enhanced thrombogenic behavior of catheter surfaces, mainly caused by the use of radio-opaque particles (i.e., barium sulfate in catheters). The approach selected for catheter modification for this new type also reduces a possible release of particles by sealing the surface with a continuous and closed polymer film. Schwab and Beathard 16 recently reviewed the catheter situation in an article entitled “Hate Living with Them, but Can’t Live without Them.” Catheters for vascular access in hemodialysis are essential in the maintenance of acute and chronic renal disease. Thrombotic complications are difficult to treat, and their treatment should be the first measure in preventing thrombus deposition. Surface design and structuring of less thrombogenic surfaces to prevent intrinsic thrombosis is therefore of major concern. Although our study reports only experience with a limited number of patients at one dialysis unit with homogeneous catheter care protocols, it is important to facilitate the progress and clinical research toward improved access solutions.
Further prospective clinical studies with higher patient numbers at different study sites need to be performed to assess the relative importance of surface modification versus catheter care protocols to better control vascular access complications that may not be adequately addressed by clinical nephrology research.
We would like to thank the nurse team of the dialysis ward for their continuous support in making the protocols work.
1. Pai MP, Pendland SL, Danziger LH: Antimicrobial-coated/bonded and impregnated intravascular catheter. Ann Pharmacother 35: 1225–1263, 2001.
2. Schierholz JM, Fleck C, Beuth J, Pulverer G: The antimicrobial efficacy of a new central venous catheter with long-term broad-spectrum activity. J Antimicrob Chemother 46: 45–50, 2000.
3. Park S, Bearinger JP, Lautenschlager EP, Castner DG, Healy KE: Surface modification of poly(ethylene terephthalate) angioplasty balloons with a hydrophilic poly(acrylamide-co-ethylene glycol) interpenetrating polymer network coating. J Biomed Mater Res 53: 568–576, 2000.
4. Francois P, Vaudaux P, Nurdin N, et al: Physical and biological effects of a surface coating procedure on polyurethane catheters. Biomaterials 17: 667–678, 1996.
5. Twardowski ZJ, Moore HL: Side holes at the tip of chronic hemodialysis catheters are harmful. J Vasc Access 2: 8–16, 2001.
6. Tsai CC, Deppisch R, Forrestal LJ, et al: Surface modifying additives for improved device-blood compatibility. ASAIO J 40: M619–M624, 1994.
7. Göhl H, Bell CM, Buck R, et al: Visualization and measurement of microdomains in polyamide dialysis membranes: optimised size facilitates performance and bioincompatibility. Blood Purif 10: 86, 1992.
8. Deppisch R, Göhl H, Smeby L: Microdomain structure of polymeric surfaces: potential for improving blood treatment procedures. Nephrol Dial Transplant 13: 1354–1359, 1998.
9. Deppisch R, Haug U, Göhl H, Ritz E: Role of proteinase/antiproteinase inhibitor disequilibrium in the bioincompatibility induced by artificial surfaces. Nephrol Dial Transplant 9 (Suppl. 3): 17–24, 1994.
10. Deppisch R, Ritz E, Hänsch GM, Schöls M, Rauterberg EW: Bioincompatibility: perspectives in 1993. Kidney Int 45 (Suppl. 44): 77–84, 1994.
11. Vanholder R, Ringoir S: Vascular access for haemodialysis. Artif Organs 18: 263–265, 1994.
12. Mysliwiec M: Vascular access thrombosis: what are the possibilities of intervention? Nephrol Dial Transplant 12: 876–878, 1997.
13. Canaud B: Haemodialysis catheter-related infection: time for action. Nephrol Dial Transplant 14: 2288–2290, 1999.
14. Matsuda T: Biological responses at non-physiological interfaces and the molecular design of biocompatible surfaces. Nephrol Dial Transplant 7 (Suppl. 4): 60–66, 1992.
15. Hecker JF: Roughness and thrombogenicity of the outer surfaces of intravascular catheters. J Biomed Mat Res 19: 381–395, 1985.
16. Schwab SJ, Beathard G: The haemodialysis conundrum: hate living with them, but can’t live without them. Kidney Int 56: 1–17, 1999.