Tunneled catheters are frequently used for vascular access in hemodialysis patients despite all efforts to promote timely creation of arteriovenous fistulas.1,2 In fact, about 20% to 40% of chronic hemodialysis patients require tunneled central venous catheters during their dialysis dependency.3 However, central venous catheters carry a risk of catheter-related bloodstream infections (CRBSI). These can affect the patients seriously. Catheter-related bloodstream infections and catheter dysfunction are known to cause morbidity, cause mortality, and increase health care cost in hemodialysis patients.1,2 These patients are particularly vulnerable to infections due to their immune-compromised status, blood exposure during hemodialysis treatments and the extracorporeal circuit, contact with other patients and health care workers, and frequent hospitalization and surgery.4 In fact, catheter infection is a leading cause of death in patients with end-stage renal disease,5 and the main cause of catheter removal and morbidity in dialysis patients.6 In addition, catheter malfunction necessitates interventions such as instillation of thrombolytic agents, fibrin sheath stripping, and finally replacement of catheters if adequate blood flow cannot be attained. The result is low dialysis quality, which can be life-threatening.1
Heparin has been used traditionally to block dialysis catheters at the end of dialysis sessions. However, it has biofilm-promoting activities7,8 and therefore current guidelines suggest using antimicrobial locking solutions rather than heparin to reduce the incidence of CRBSI.9,10,11 In addition, leakage of heparin- or citrate-containing catheter lock solutions into the circulation may cause significant side effects such as embolism, cardiac arrhythmia, paresthesia, and bleeding.9,10,12,13
On the other hand, routine use of antibiotic catheter lock solutions may promote microbial resistance development, and they are associated with additional side effects such as ototoxicity.14,15
To date, a perfect catheter lock solution that prevents CRBSIs, ensures catheter patency, and does not cause adverse effects has not been identified.16,17 An ideal solution should have antimicrobial properties to reduce biofilm formation, and prevent intraluminal thrombus formation leading to catheter occlusion. Furthermore, it should not promote bacterial resistance development and should cause as few as possible undesirable effects in hemodialysis patients.
Taurolidine seems to be a promising substance in a catheter lock solution: it is a potent antimicrobial agent with proven effectiveness against a broad spectrum of bacteria and fungi18 that has been shown to reduce biofilm formation.19 As an antimicrobial solution it is not susceptible to promoting the development of microbial resistance,20 even after prolonged use.21 In addition, taurolidine has antithrombotic properties and prevents intraluminal thrombus formation and catheter occlusion.22,23,24 Finally, it is not toxic to humans, even after injection of up to 25 g daily into the circulation, since it is rapidly metabolized into taurine, water, and carbon dioxide.25,26
A combination of 1.35% taurolidine with citrate and heparin/urokinase as a catheter lock in hemodialysis patients has already been investigated in a randomized controlled trial and showed positive results.3 Furthermore, a 2% taurolidine catheter lock solution without citrate and heparin was associated with significantly lower rates of CRBSI and catheter occlusion compared with a heparin solution in home parenteral nutrition patients.24 However, in the setting of hemodialysis via central venous catheter, a catheter lock solution containing only 2% taurolidine as ancillary medicinal substance has not yet been studied.
Therefore, this retrospective study was conducted to investigate the rate of CRBSI, catheter complications, and adverse events with a 2% taurolidine catheter lock solution in a Swiss cohort of patients receiving chronic hemodialysis via a tunneled central venous catheter. The results were compared with the existing literature in a review.
Methods
Study design
Routinely documented data from chronic hemodialysis patients were retrospectively analyzed.
Ethical approval
The responsible ethics committee (Ethikkommission Ostschweiz) approved this retrospective study under project 300382-001.
Patient population
All 21 adult patients on long-term chronic dialysis at the documenting center who used tunneled central venous catheters were included. Patients had received a tunneled catheter (mostly either EQUISTREAM® long-term hemodialysis catheter or GLIDEPATH® long-term hemodialysis catheter; C. R. Bard GmbH, Karlsruhe, Germany) between 25 months and 1 month before enrollment. Most of the patients had received heparin lock solutions. At enrollment, all of them received 2% taurolidine solution (TauroSept®, Geistlich Pharma AG, Wolhusen, Switzerland) as catheter lock and gave informed consent.
Data collection
Gender, year of birth, reason for hemodialysis, and concomitant medication were documented, and the Charlson Comorbidity Index (CCI) was calculated from comorbidities. In addition, date of catheter implantation, start and stop date of the application of 2% taurolidine solution, reason for stopping application, number of hemodialysis sessions per week, and duration of hemodialysis per session were recorded.
Endpoints
The primary endpoint was the rate of CRBSI. Secondary endpoints were adverse events; catheter dysfunction rate, treatment, and cost; catheter technical problem rate, resolution, and cost; overall catheter-related cost; rate of alteplase use (Actilyse®, Boehringer Ingelheim Deutschland GmbH, Ingelheim, Germany) to safeguard catheter patency; CRBSI treatment and cost; and pathogen spectrum.
Diagnosis of CRBSI
Patients were diagnosed with CRBSI according to the definition of verified bloodstream infection by KDOQI27 and CDC,28 that is, when the same organism was isolated in semiquantitative or quantitative cultures of blood drawn from both the catheter lumen and a peripheral vein of patients with clinical symptoms of bloodstream infection. The CRBSI rate was defined as the number of CRBSI episodes per 1,000 catheter days. If patients showed raised temperatures and an attributable cause was identified, for example, influenza, or if blood cultures became positive, this was documented in the medical charts as an adverse event.
Definition and treatment for catheter dysfunction
Manifest catheter dysfunction suggesting occlusion was determined when the flow rate was persistently below 200 mL/min despite measures such as switching lines, flushing the catheter, and repositioning the patient. Alteplase use was categorized in alteplase rescue, in case of catheter dysfunction, and alteplase prophylaxis, which was performed based on the judgment of 2 certified nurses, to prevent catheter dysfunction.
Cleaning protocol of catheter and insertion site
Catheter care standards were not changed during the study. Catheter care was performed according to recommended standards of care including dressing changes, exit site care, and hygiene. After each hemodialysis session, the exit site was inspected and covered with either sterile swabs (IV3000; Smith & Nephew GmbH, Hamburg, Germany; or Tegaderm I.V.; 3M Deutschland GmbH, Neuss, Germany) or chlorhexidine dressing (Tegaderm CHG; 3M Deutschland GmbH). When deemed necessary by 2 experienced nurses, Mupirocin (Bactroban Nasal Nasensalbe, GlaxoSmithKline AG, Münchenbuchsee, Switzerland) would be applied to the exit site.
Application of catheter lock solution
Central venous catheters were locked with a solution containing 2% taurolidine, sterile water, 5% polyvinylpyrrolidone as emulsifier, and traces of HCl or NaOH for pH adjustment to 7.3 (TauroSept; Geistlich Pharma AG) after each dialysis treatment. The solution was taken directly out of the ampule and was instilled into both catheter channels after flushing each with 10 mL 0.9% saline solution. Two separate syringes were used and filled with about 2.2 mL 2% taurolidine solution for catheter locking: 1.8 mL per channel + 0.2 mL per Tego connector + 0.1 to 0.2 mL extra to flush the tip of the catheter. The lock solution was removed with a sterile syringe in advance of each dialysis treatment.
Statistical analysis
For numerical results, the mean was reported with its standard deviation, and qualitative results were summarized as absolute and relative counts. Rates were computed as the total sum of occurrences divided by the total sum of catheter days, or the total sum of occurrences divided by the number of patients, respectively. The number of catheter days for 1 patient was the number of days between beginning and end of the catheter locking period with a 2% taurolidine solution. The confidence intervals for the incidence rates are exact confidence intervals based on the Poisson distribution.
Results
Baseline characteristics
The 21 consecutively included patients had an average age of 72.6±9.9 years and a mean CCI of 4.4±1.9. The total observation period was 5,639 catheter days, with a mean of 268.5±135.6 catheter days per patient. The most common causes for chronic kidney disease were vascular disease (12 patients; 57.1%) and diabetes mellitus (5 patients; 23.8%). The most common comorbidities were chronic pulmonary disease (9 patients; 42.9%) and coronary heart disease (8 patients, 38.1%). Detailed patient characteristics are shown in Table 1.
Table 1. Baseline characteristics of the study patients
| Variable | 2% taurolidine group, n=21 |
|---|---|
| Demographics | |
| Age, y ± SD | 72.6±9.9 |
| Gender, n (%) | |
| Male | 12 (57.1) |
| Female | 9 (42.9) |
| Cause of chronic kidney disease, n (%) | |
| Vascular disease including hypertension | 12 (57.1) |
| Diabetes mellitus | 5 (23.8) |
| (Poly-)cystic kidney disease | 2 (9.5) |
| Acute kidney failure | 1 (4.8) |
| Graft failure | 1 (4.8) |
| Comorbidities, n (%) | |
| Chronic pulmonary disease | 9 (42.9) |
| Coronary heart disease | 8 (38.1) |
| Peripheral vascular disease | 7 (33.3) |
| Diabetes mellitus type 2 | 6 (28.6) |
| Diabetes with end-organ damage | 6 (28.6) |
| Heart failure | 6 (28.6) |
| Cardiomyopathy | 5 (23.8) |
| Cerebrovascular disease | 5 (23.8) |
| Myocardial infarction | 2 (9.5) |
| Mild liver disease | 2 (9.5) |
| Dementia | 1 (4.8) |
| Diabetes mellitus type 1 | 1 (4.8) |
| Hepatitis C | 1 (4.8) |
| Tumor | 1 (4.8) |
| Charlson Comorbidity Index, score ± SD | 4.4±1.9 |
| Central venous catheter data | |
| Total catheter days | 5,639 |
| Catheter days per patient, mean ± SD | 268.5±135.6 |
| Number of catheters, n (%) | 26 (100) |
SD=standard deviation
Primary endpoint
No episode of CRBSI occurred over the period of 5,639 catheter days; thus the CRBSI rate was 0 per 1,000 catheter days (Table 2). A comparison of CRBSI rates with antimicrobial, citrate, and heparin catheter lock solutions reported in the literature is shown in Table 3.
Table 2. Key outcomes of the study
| Variable | Per patient during observation time, mean [95% CI] | Per 1,000 catheter days, mean [95% CI] |
|---|---|---|
| CRBSI | 0 [0.00; 0.18] | 0 [0.00; 0.65] |
| Catheter dysfunction | 0.19 [0.05; 0.49] | 0.71 [0.19; 1.82] |
| Need for alteplase rescue | 0 [0.00; 0.18] | 0 [0.00; 0.65] |
| Need for prophylactic alteplase | 7.52 [6.40; 8.79] | 28.02 [23.82; 32.74] |
| Technical problems | 0.10 [0.01; 0.34] | 0.35 [0.04; 1.28] |
| AEs related to 2% taurolidine solution | 0 [0.00; 0.18] | 0 [0.00; 0.65] |
| AEs not related to 2% taurolidine solution | 1.29 [0.85; 1.87] | 4.79 [3.16; 6.97] |
AE=adverse event; CI confidence interval; CRBSI=catheter-related bloodstream infection
Table 3. Comparison of catheter-related bloodstream infection (CRBSI) rates with different catheter lock solutions in patients receiving hemodialysis via central venous catheter in the literature
| First author, year | Intervention (observation time in catheter days) | Control (observation time in catheter days) | Antimicrobial (CRBSIs per 1,000 catheter days) | Citrate (CRBSIs per 1,000 catheter days) | Heparin (CRBSIs per 1,000 catheter days) |
|---|---|---|---|---|---|
| Neusser 2020 (current study) | 2% taurolidine (5,639) | - | 0 | ||
| Winnicki 20183 | 1.35% taurolidine + citrate + heparin | Citrate (6,708) | 0.67 | 2.7 | |
| 1.35% taurolidine + citrate + urokinase (8,982) | |||||
| Correa Barcellos 201729 | Citrate (6,052) | Heparin (6,927) | 5.2 | 3.4 | |
| Moghaddas, 201530 | Cotrimoxazole + heparin (11,932) | Heparin (12,559) | 0.58 | 4.4 | |
| Broom, 201231 | Ethanol + heparin (3,614) | Heparin (1,834) | 0.28 | 0.85 | |
| Moran, 201232 | Gentamicin (39,827) | Citrate (32,933) | 0.28 | 0.91 | |
| Oguzhan, 201233 | NaCl + heparin (3,368) | Heparin (3,099) | 1.1 | 0.96 | |
| Maki, 201134 | Citrate + methylene blue (25,274) | Heparin (24,395) | 0.24 | 0.82 | |
| Zhang, 200915 | Gentamicin + heparin (17,781) | Heparin (16,299) | 0.06 | 0.67 | |
| Kim, 200635 | Cefazolin + gentamicin + heparin (2,273) | Heparin (2,244) | 0.44 | 3.12 | |
| Saxena, 200536 | Cefotaxime + heparin (58,035) | Heparin (17,885) | 1.65 | 3.13 | |
| Weijmer, 200537 | Citrate (8,431) | Heparin (8,116) | 1.1 | 4.1 | |
| Betjes, 200438 | 1.35% taurolidine + citrate (1,519) | Heparin (1,885) | 0 | 2.1 | |
| Dogra, 200239 | Gentamicin + citrate (3,280) | Heparin (2,643) | 0.3 | 2.7 | |
| CRBSI rate range | 0–1.65 | 1.1–5.2 | 0.67–4.4 | ||
Catheter dysfunction
Four episodes of catheter dysfunction occurred during the observation period in 2 of 21 patients (9.5%), corresponding to 0.71 dysfunctions per 1,000 catheter days. Two of the 4 catheter dysfunctions were treated by catheter replacement and 2 by fibrin sheath stripping. There was no need for alteplase rescue therapy. Prophylactic alteplase given before flow rates dropped below 200 mL/min was used 158 times, corresponding to 28.02 episodes per 1,000 catheter days (Table 2). Of note, 44 of the 158 prophylactic alteplase applications were conducted in just 1 patient. A comparison of catheter dysfunction rates with antimicrobial, citrate- and heparin-containing catheter lock solutions reported in the literature is shown in Table 4. Total costs due to catheter dysfunction in this trial were 11,398.00 CHF (Swiss Franc), or 2,021.28 CHF per 1,000 catheter days.
Table 4. Comparison of catheter dysfunction (CD) rates with different catheter lock solutions in patients receiving hemodialysis via central venous catheter in the literature
| First author, year | Intervention (observation time in catheter days) | Control (observation time in catheter days) | Antimicrobial (CRBSIs per 1,000 catheter days) | Citrate (CRBSIs per 1,000 catheter days) | Heparin (CRBSIs per 1,000 catheter days) |
|---|---|---|---|---|---|
| Neusser 2020 (current study) | 2% taurolidine (5,639) | - | 0.71 | ||
| Winnicki 20183 | 1.35% taurolidine + citrate + heparin | Citrate (6,708) | 18.7 | 44.3 | |
| 1.35% taurolidine + citrate+ urokinase (8,982) | |||||
| Correa Barcellos 201729 | Citrate (6,052) | Heparin (6,927) | 11.4 | 9.4 | |
| Oguzhan 201233 | NaCl + heparin (3,368) | Heparin (3,099) | 1.2 | 0.97 | |
| Saxena 200536 | Cefotaxime + heparin (58,035) | Heparin (17,885) | 0.41 | 0.95 | |
| Weijmer 200537 | Citrate (8,431) | Heparin (8,116) | 3.2 | 3.6 | |
| Dogra 200239 | Gentamicin + citrate (3,280) | Heparin (2,643) | 3.96 | 4.16 | |
| CD rate range | 0.41–18.7 | 3.2–44.3 | 0.95–9.4 | ||
Catheter technical problems
None of the catheter technical problems were attributed to the use of the 2% taurolidine-containing catheter lock solution. Two catheter technical problems occurred in 1 patient (4.8%) during the observation period, corresponding to 0.35 catheter technical problems per 1,000 catheter days. Both cases were resolved by catheter replacement. Total costs due to catheter technical problems were 5,400.00 CHF, or 957.62 CHF per 1,000 catheter days. The overall catheter-related costs are shown in Table 5.
Table 5. Overall catheter-related costs
| Variable | Costs, CHF |
|---|---|
| Total costs due to catheter dysfunction | 11,398.00 |
| Costs due to catheter dysfunction per 1,000 catheter days | 2,021.28 |
| Total costs due to catheter technical problems | 5,400.00 |
| Costs due to catheter technical problems per 1,000 catheter days | 957.62 |
| Overall catheter-related costs | 16,798.00 |
| Overall catheter-related costs per 1,000 catheter days | 2,978.90 |
CHF =Swiss Franc
Adverse events
None of the adverse events were attributed to the 2% taurolidine solution. Twenty-seven adverse events were recorded in 12 patients, corresponding to 1.29 adverse events per patient or 4.79 per 1,000 catheter days. The most common adverse events were arterial disease (3 episodes) and atrial flutter (3 episodes).
Discussion
The results of this retrospective study and comparison with the existing literature in this field show that using a 2% taurolidine-containing catheter lock solution in chronic hemodialysis patients is feasible and safe. The observed CRBSI rate of 0 per 1,000 catheter days compared favorably with the literature on other antimicrobial lock solutions, where rates from 0 (1.35% taurolidine + citrate38) to 1.65 (cefotaxime + heparin36) per 1,000 catheter days were reported. The CRBSI rates with other locking solutions ranged from 1.1 to 5.2 per 1,000 catheter days for citrate solutions, and from 0.67 to 4.4 for heparin solutions (Table 3).
The standard of care in this study included prophylactic use of alteplase. It may be argued that this influenced CRBSI rates because fibrin sheaths and thrombi are known sources of infections and thus alteplase prophylaxis may reduce bacteremia rate.5 Yet alteplase itself is unlikely to have influenced infection rates directly, since the thrombolytic alone does not seem to exhibit antimicrobial activity.40 In fact, use of alteplase as an indicator of thrombus formation in catheters has been associated with higher infection rates in critically ill children.41
The observation time of 5,639 catheter days in this cohort is well within the range reported in other studies on catheter lock solutions for hemodialysis patients, which range from 1,51938 to 39,82732 per group. In fact, this study has the longest observation time in which no CRBSIs at all were detected, compared with other studies in the same setting (Table 3).
Studies investigating other taurolidine formulations such as 1.35% taurolidine with 4% sodium citrate showed a reduction in bacteremia episodes compared with heparin but a greater need for thrombolytic therapy with alteplase.42,43 In contrast to these findings, the catheter dysfunction rate of 0.71 per 1,000 catheter days in this cohort was within the lower range of the rates reported in the literature for antimicrobial solutions, which ranged from 0.41 with cefotaxime + heparin36 to 18.7 for 1.35% taurolidine with citrate and heparin and 1.35% taurolidine with citrate and urokinase. 3 Catheter dysfunction rates with other locking solutions were 3.2 to 44.3 per 1,000 catheter days for citrate solutions and 0.95 to 9.4 for heparin solutions (Table 4). Hence, the antithrombotic effect of 2% taurolidine-containing lock solution, in combination with prophylactic alteplase use on demand, was effective in preventing catheter dysfunction. Winnicki et al. in 20183 reported a dysfunction rate of 18.7 per 1,000 catheter days, and a weekly prophylactic urokinase use to prevent catheter dysfunction, corresponding to 143 applications of thrombolytic urokinase per 1,000 catheter days. In the cohort of this study, the rate of prophylactic alteplase use as thrombolytic agent was 5 times lower, at 27 applications per 1,000 catheter days. Therefore, there seems to be no need for constant use of thrombolytic agents in these patients. However, the varying definitions for catheter dysfunction used in previous studies impede a direct comparison between the relevant studies.
No adverse events attributed to the 2% taurolidine solution were observed in this cohort, which is in line with other reports25 and demonstrates the safety of its use as a catheter lock.
Importantly, in this study, catheters were overfilled by 0.1 to 0.2 mL to reach the tip of the catheter and guarantee complete filling as well as to ensure contact of the lock solution with the tip of the catheter. It is reported that flushing of the 2% taurolidine-containing catheter lock solution into the circulation, although not witnessed in this cohort, causes no harm.25,44 In contrast to these findings, citrate as catheter lock solution has reportedly caused severe side effects such as embolism,12 cardiac arrhythmia,9 and paresthesia,13 and heparin is associated with bleeding15 and alopecia.45 Other antimicrobial catheter lock solutions contain antibiotic components such as gentamicin, which is effective against a wide range of bacterial infections but is also ototoxic and may cause tinnitus.15 Furthermore, its routine use promotes the development of microbial resistance: in a study investigating gentamicin as a prophylactic catheter lock solution in hemodialysis patients, gentamicin-resistant CRBSI emerged within 6 months.14
The reduction of CRBSI and catheter dysfunction rates saves expenses by reducing hospitalization, catheter replacement, and antibiotic treatment. Furthermore, costs for weekly prophylactic recombinant tissue plasminogen activator locking are as high as 1,206 Canadian dollars per patient per 6 months,5 and regular weekly prophylaxis was not a cost-saving strategy in an economic evaluation.46
In addition, the requirement for prophylactic alteplase use varied from 0 to 44 applications between individual patients in this study. Hence, a prophylactic alteplase treatment based on individual patient assessment, rather than a rigid, regular weekly prophylaxis, may be a more efficient and economic approach to prevent catheter complications.
The strength of this retrospective cohort study is its pragmatic design. The collected data reflect a real-world setting, the daily routine of the dialysis unit at the investigated hospital. The inclusion of all adult hemodialysis patients using tunneled central dialysis catheters in this unit provided a representative patient cohort.
Due to the single-center design, all patients and their catheters were managed according to the same standard procedures such as locking catheters with 2% taurolidine-containing solution, aseptic handling, overfilling of the catheter, and the administration of thrombolytic agent after individual assessment.
Conclusions
This retrospective study presents comprehensive data from 21 hemodialysis patients with central venous catheters regarding the use of a 2% taurolidine-containing catheter lock solution without heparin or citrate. The study documented that this catheter lock solution effectively prevented and reduced the number of CRBSI, limited catheter dysfunction, and was cost-effective. No CRBSIs were detected over the longest observation time compared with other studies with the same setup. The study and a review of the literature confirmed safety of a 2% taurolidine-containing catheter lock solution since no adverse events were observed. The results warrant a larger, randomized controlled trial to ascertain the promising results observed in this cohort.
Limitations
This single-center retrospective study shares the limitations of other retrospective, uncontrolled studies, in which adverse events may be underestimated or confounders insufficiently documented. However, the risk seems to be limited since the hemodialysis patients in this study were closely monitored by the physicians, and any changes to their health status or any medical occurrences were routinely documented in their medical records.
Recommendations for practice
The results, especially if confirmed in a randomized trial, may have clinical relevance and may support physicians in optimizing hemodialysis catheter care. Importantly, this is the first clinical study in hemodialysis patients with a catheter lock solution containing 2% taurolidine but no citrate and no heparin.