Skip Navigation


NDT Advance Access originally published online on January 31, 2006
Nephrology Dialysis Transplantation 2006 21(4):1024-1031; doi:10.1093/ndt/gfi104
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
21/4/1024    most recent
gfi104v1
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 (18)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Mokrzycki, M. H.
Right arrow Articles by Rosenberg, S. O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mokrzycki, M. H.
Right arrow Articles by Rosenberg, S. O.
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

Tunnelled haemodialysis catheter bacteraemia: risk factors for bacteraemia recurrence, infectious complications and mortality

Michele H. Mokrzycki1, Meilin Zhang1, Hillel Cohen2, Ladan Golestaneh1, Jeffrey M. Laut3 and Stuart O. Rosenberg1

1 Department of Medicine, Division of Nephrology, Montefiore Medical Center and 2 Department of Epidemiology and Social Medicine, Albert Einstein College of Medicine, Bronx, NY and 3 Hartford Hospital, Hartford, CT, USA

Correspondence and offprint requests to: Michele H. Mokrzycki MD MS, 3332 Rochambeau Avenue, Centennial Building 423, Monteflore Medical Center, Bronx, NY 10467. Email: mokrzm{at}monteflore.org



   Abstract
Top
 Abstract
Introduction
 Subjects and methods
 Results
 Discussion
 References
 
Background. Infection is a serious complication of tunnelled cuffed catheter (TCC) use and is associated with high complication and mortality rates. Although attempts at TCC salvage after bacteraemia have been associated with high rates of recurrent bacteraemia, there have been no large studies in which multivariate analysis has been performed to control for confounding factors such as infecting organisms, diabetes, etc.

Methods. A prospective observational study was performed in chronic HD patients dialyzing with a TCC at seven outpatient HD centers. All patients diagnosed with TCC bacteraemia were observed for 3 months following initial presentation and outcomes were recorded.

Results. During the 2.5 year study period, 226 patients had an episode of TCC bacteraemia that met inclusion criteria, and 3 month follow-up data were available in 219 episodes. Treatment failure, defined as recurrent TCC bacteraemia with the same organism or death from sepsis, occurred in 26 patients (12%). Infectious complications (such as endocarditis, osteomyelitis, etc.) occurred in 16 patients (7%), bacteraemia with a different organism occurred in 19 patients (9%), and death from sepsis occurred in eight patients (4%). Significant predictors of treatment failure (by univariate analysis) were TCC salvage, and infection with Staphylococcus aureus, (OR = 4.2, P = 0.002; and OR = 3.3, P = 0.02, respectively). TCC salvage, when used in episodes of S. aureus bacteraemia, was associated with an 8-fold higher risk of treatment failure (P = 0.001). The presence of an abnormal TCC exit site was associated with a significantly higher rate of death from sepsis, (OR = 7, P = 0.001). Outcomes (treatment failure and infectious complications) did not differ among bacteraemic episodes where the TCC was exchanged over a guidewire compared to those in which the TCC was immediately removed followed by delayed reinsertion. In the multivariate analysis, adjusted for potential confounding covariates, the only significant predictors of treatment failure after an episode of TCC bacteraemia were TCC salvage (OR = 5.4, P = 0.003), and S. aureus (OR = 4.2, P = 0.002). In a multivariate analysis, controlling for TCC management, the only variable that was significantly associated with the development of an infectious complication was infection with S. aureus (OR = 3.5, P = 0.02).

Conclusions. We have shown, using multivariate analysis and adjusting for potential confounding factors, that the use of TCC salvage and S. aureus are independent risk factors for treatment failure after an episode of TCC bacteraemia, and that S. aureus is an independent risk factor for developing an infectious complication. An infected-appearing TCC exit site is associated with a higher mortality rate. Episodes of TCC bacteraemia treated using TCC salvage are associated with the highest treatment failure rates. TCC guidewire exchange can be an acceptable practice, unless severe exit site or tunnel infection is present.

Keywords: bacteraemia; catheter; haemodialysis; sepsis; tunnelled catheter



   Introduction
Top
 Abstract
 Introduction
Subjects and methods
 Results
 Discussion
 References
 
Permanent tunnelled, cuffed haemodialysis catheter (TCC) use has doubled in the period from 1996 to 2000 [1]. Infection remains a serious complication associated with TCC use, with a mean incidence of TCC bacteraemia of 3 per 1000 catheter days [2–7]. The relative risk of access-related bacteraemia is 9.2 for TCCs, and 0.48 for arteriovenous fistulas compared to arteriovenous grafts (reference category) [8]. Serious adverse events following an episode of TCC bacteraemia include recurrent bacteraemia, infectious complications and death. The rate of bacteraemia recurrence varies from 18–66% [4,5,9]. The catheter management strategy of ‘TCC salvage’, in which antibiotics are administered without catheter removal, has been associated with a higher rate of treatment failure [4,5,9]. However, many of these studies were relatively small series, and were not sufficiently powered to perform a multivariate analysis. It is possible that other clinical factors (such as age, diabetes mellitus, iron therapy, microbiologic isolate, immunocompromised states) influenced the outcome in these studies.

The purpose of this study was to evaluate patient outcomes after TCC bacteraemia, taking into account potential confounding clinical variables and TCC management strategies, using mulitvariate analysis. This type of analysis has not been performed in a large series of TCC associated bacteraemia, and there are no series comparing all possible modalities of TCC management in the same study.



   Subjects and methods
Top
 Abstract
 Introduction
 Subjects and methods
Results
 Discussion
 References
 
This is a prospective, observational study. The study was approved by the Institutional Review Boards at Montefiore Medical Center and Hartford Hospital. Written informed consent was not required. The study was performed in seven outpatient chronic HD centres; five centres were affiliated with Montefiore Medical Center, Bronx, New York (combined average census of approximately 600 patients) and two were located in Connecticut (Hartford Hospital and East Hartford Dialysis Center; combined average census of 334 patients). The prevalence of TCCs was 34% in the Bronx, and 26% in Hartford. The patient population included all chronic HD patients dialyzing using a TCC.

All TCC patients who had positive blood cultures were further evaluated to diagnose the primary source of the bacteraemia. A bacteraemic episode was determined to be TCC-related when blood cultures taken from the TCC were positive during a period in which an exit/tunnel infection was also present or when other primary sources of bacteraemia were absent (by physical examination, urinalysis, chest radiograph or other diagnostic tests). Designated research nurses and physician assistants, situated in each haemodialysis unit, identified patients who met criteria for TCC bacteraemia and followed them weekly for 3 months. The date of entry into the study (baseline period) was the date of the first positive blood culture. If a subsequent episode of bacteraemia with the same bacterial isolate occurred within the 3 month period, after having been determined initially cured by the presence of negative blood cultures and/or clinical evaluation, it was considered to be a recurrent bacteraemia.

The 3 month follow-up period was selected based on the time period used in all previously published series (45–90 day follow-up), and because we, and other investigators, determined 3 months to be a reasonable time period during which recurrent bacteraemia attributable to initial bacteraemia would manifest [4,5,9,10]. Microbiologic testing was performed in the clinical laboratory used for standard patient care in each dialysis unit. This included identification and antibiotic sensitivity of the isolate. If bacteraemia occurred after the 3 month follow-up period following the initial bacteraemic episode, it was considered to be a second, unrelated episode of bacteraemia and was excluded from analysis. Only the first bacteraemic episodes were included in the analysis. An individual patient's nephrologist was responsible for clinical decisions, including TCC management and antibiotic selection.

Baseline demographic and clinical data were obtained at the time of bacteraemia, or were the most recent available values drawn prior to bacteraemia using inpatient and outpatient charts. These included patient age, gender, race, cause of end-stage renal disease (ESRD), intravenous iron administration within 1 month of bacteraemia, and the presence of diabetes mellitus (DM). Data regarding the date of TCC insertion, TCC management, and TCC survival 3 months post-bacteraemia were recorded. In addition, laboratory data including hepatitis B and C viral serologies, human immunodeficiency viral status (when available), ferritin, transferrin saturation, complete blood count, serum albumin and blood cultures were accessed. We also collected data regarding TCC exit site appearance and temperature at the time of presentation, initial and subsequent antibiotic therapy, hospitalization, and documentation of secondary metastatic infection and/or death. An abnormal TCC exit site appearance was defined as the presence of erythema or exudate.

All catheters were tunnelled, cuffed, dual-lumen, silastic catheters used only for haemodialysis. Standard catheter care consisted of cleansing the catheter site with povidone–iodine solution alone or povidone-iodine followed by isopropyl alcohol. The TCC exit site was covered with a transparent; oxygen permeable dressing (OpSite TM Smith and Nephew Ltd., Largo, CA, USA) or with a gauze dressing. Catheter dressings were changed thrice weekly by HD staff at all sites except Hartford Hospital. At Hartford Hospital the TCC dressing was changed weekly using an antimicrobial dressing containing chlorhexidine gluconate (Biopatch TM, Johnson and Johnson, Somerville, NJ). Four strategies of managing a TCC-related bacteraemia were used: (1) TCC salvage (TCC remains in place during antibiotic therapy), [S]; (2) exchange of TCC over a guidewire for a new TCC (without waiting for negative blood cultures), [W]; (3) immediate removal of the TCC with delayed reinsertion after a minimum of 1-week after blood cultures turned negative, [DR]; (4) TCC removal with use of a functioning arteriovenous access (AVF, arteriovenous fistula or AVG, arteriovenous graft), [O]. The antibiotic lock technique (in the TCC lumen) was not used for prophylaxis or for treatment for TCC bacteraemia.

Statistics
Treatment failure (primary outcome) was defined as the composite outcome of recurrent bacteraemia with the same organism or death from sepsis. The secondary outcome was the development of infectious complications. Both univariate and multivariate analyses were performed for these outcomes. All continuous variables are reported as the mean±SEM values, and multiple comparisons were performed using one-way ANOVA, (P = 0.05). Categorical variables were evaluated using Chi-square analysis for comparisons between groups (Fisher's Exact Test, P = 0.05). Multivariate analyses were performed using logistic regression. Variables found to be statistically significant in the univariate analysis were entered into the initial model (model A), and were removed if P>0.08 (final model B). Survival analyses using Cox regression were performed. All statistics were performed using SPSS 10.0 for Windows®.



   Results
Top
 Abstract
 Introduction
 Subjects and methods
 Results
Discussion
 References
 
Patient demographics
During the 2.5 year study period (12/1/00–5/31/03), there were 294 episodes of bacteraemia. In 22 episodes, a non-TCC source of bacteraemia was identified and these episodes were excluded from the final study analysis, whereas 272 episodes of bacteraemia (occurring in 226 patients) were determined to be TCC-associated (criteria described in methods). During the study period, 186 patients had only one TCC bacteraemic episode, 34 patients had two episodes and six patients had three episodes. Six patients were lost to follow-up, and haemodialysis was withdrawn at the time of bacteraemia in one patient with metastatic malignancy. Complete 3 month follow-up data was available in 219 episodes, and were included in the final analysis.

The aetiology of ESRD for the total population was as follows: DM (46%), hypertension (31%), glomerulonephritis (5%), HIV-associated nephropathy (4%), idiopathic focal segmental glomerulosclerosis (3%) and other (11%). The site of TCC insertion was the internal jugular vein in 220 episodes, and translumbar in six episodes. Seventy patients (32%) had an abnormal TCC exit site appearance at the time of diagnosis of bacteraemia.

Data regarding patient demographics, TCC bacteraemia, and lab data are provided, according to TCC management strategy, in Tables 1 and 2. The variables found to be significantly different between the TCC management groups were as follows: diabetes mellitus, intravenous iron therapy, abnormal TCC exit site, days until TCC removal, hospitalization, duration of antibiotic therapy, microbiologic isolate type, hepatitis C antibody status.


View this table:
[in this window]
[in a new window]
  		Table 1. Patient demographic and TCC bacteraemia data

 

View this table:
[in this window]
[in a new window]
  		Table 2. Laboratory data

 
Microbiologic isolates
Thirty different organisms were isolated in 219 episodes of TCC-bacteraemia, and more than one organism was isolated in 13% of episodes. Gram-positive organisms accounted for 63% of isolates, and 43% of the Gram-positive isolates were S. aureus. Gram-negatives accounted for 36%, and yeast for 1% of isolates. Methicillin-resistant S. aureus (MRSA) was more prevalent in the hospital-based haemodialysis units (21% isolates) vs the free-standing haemodialysis units (12%), although this did not reach statistical significance (P = 0.07). The type of organism isolated was not associated with race, viral status (HIV, hepatitis C or B), diabetes mellitus or intravenous iron administration.

The initial antibiotic regimens were as follows: combination vancomycin 1 g and gentamicin 120 mg i.v. (55%), vancomycin 1 g i.v. (16%), combination cefazolin 1 g and tobramycin 120 mg i.v. (7%), combination vancomycin 1 g and tobramycin 120 mg i.v. (6%), cefazolin 1 g i.v. (3%), gentamicin 120 mg i.v. (1%), cefazolin 1 g and gentamicin 120 mg i.v. (1%), fluconazole (1%) i.v., linezolid and gentamicin 120 mg i.v. (1%), and the remainder of combinations occurred in less than 1% of cases.

There was a significant difference in the empiric antibiotic type between HD units: combination vancomycin and gentamicin, and vancomycin alone were the most common empiric antibiotic regimens in 6/7 HD units, whereas combination cefazolin and tobramycin, and vancomycin alone were the most common regimens in the remaining HD unit. In 87% of all episodes, the isolate was sensitive to the initial empiric antibiotics. Ninety-three percent of S. aureus isolates were sensitive to the initial antibiotic selection; whereas 86% of non-S. aureus Gram-positive and 85% of Gram-negative isolates were sensitive to the initial antibiotics, (P = 0.3).

The mean duration of antibiotics was 3.2±0.1 weeks. The most common antibiotic regimens prescribed after identification of the microbiologic isolate were as follows: vancomycin 500 mg i.v. q HD treatment (32%), cefazolin 1–2 gm i.v. q HD treatment (9%), vancomycin 500 mg and gentamicin 80 mg i.v. qd HD treatment (7%), vancomycin 1 gm i.v. q week (5%), vancomycin 500 mg i.v. q HD changed to cefazolin 1–2 gm i.v. q HD (4%), gentamicin 80 mg i.v. q HD (4%), gatifloxacin 200 mg po qd and gentamicin 80 mg i.v. q HD (4%). The remainder of antibiotic combinations were administered in less than 1% of cases. Antibiotic dosing was appropriate in 91.2% of episodes. The appropriate use of vancomycin and gentamicin (after the identification of the organism and antibiotic sensitivity were available) was applied in 88% of episodes. S. aureus was signficantly associated with a higher prevalence of abnormal exit site appearance compared to all other isolates (54 vs 23%, P = 0.006). The incidence of hospitalization was significantly higher when S. aureus was isolated compared to other isolates, (79 vs 50%, P<0.001). Diabetes was not significantly associated with treatment failure or infectious complications.

Primary and secondary outcomes
In the 3 month follow-up period, the composite primary outcome of recurrent bacteraemia with the same organism or septic death occurred in 26 of 219 episodes (12%). During this follow-up period, there were 19 episodes of bacteraemia occurring with a different organism than that isolated during the initial TCC bacteraemia (9% episodes). Eighteen patients (8%) expired and the causes of death are as follows: sepsis (8), urosepsis (2), post-transplant complications (1), unknown aetiology (6), complications of metastatic cancer (1). Table 3 provides data regarding the TCC management group and microbiologic isolate type for all bacteraemic episodes resulting in an infectious complication or death due to sepsis. Infectious complications occurred in 16 (7%) of bacteraemic episodes. These included endocarditis (7), septic pulmonary emboli (1), osteomyelitis (3), infection of an arteriovenous graft (2), and abscess (3).


View this table:
[in this window]
[in a new window]
  		Table 3. Univariate analysis of primary and secondary outcomes

 
Univariate analysis was performed to determine variables associated with both the primary and secondary outcomes (Table 3). The clinical variables included in the analyses were sex, race, age, diabetes mellitus, albumin, haemoglobin, intravenous iron administration, maximum temperature at presentation, duration of fever, TCC exit site appearance, TCC management strategy, organism isolated, hospitalization and length of stay, duration of antibiotics and haemodialysis unit. TCC salvage and infection with S. aureus were significant predictors of treatment failure (OR = 4.2, P = 0.002; and OR = 3.3, P = 0.02, respectively). The treatment failure rate associated with each TCC management strategy was as follows: S = 3.4/1000 TCC days, W = 0.4/1000 TCC days, DR = 1.5/1000 TCC days (Figure 1). Event rates were not performed for the TCC management group ‘O’ due to the fact that they were no longer using a TCC after the initial bacteraemia. If all TCCs were removed (either using W or DR) instead of using ‘S’, for the treatment of TCC bacteraemia, the hypothetical reduction in the treatment failure rate would be 1.5/1000 TCC days. The presence of an abnormal TCC exit site was significantly associated with a higher rate of death from sepsis, (OR = 7, P<0.001). The incidence of bacteraemia with a different organism occurring during the 3-month follow-up period was not associated with the presence of S. aureus infection or with TCC salvage. None of the other above variables was significantly associated with either the primary or secondary study outcomes. A series of subset analyses were performed to determine the outcomes associated with each TCC management strategy, according to the TCC exit site appearance and the microbiologic isolate (Table 4). Significant findings include: (1) a higher treatment failure rate (57% vs 17%) for episodes of S. aureus bacteraemia treated with TCC salvage P = 0.03), and (2) a higher infectious complication rate (14% vs 0) for episodes of S. aureus bacteraemia treated with TCC delayed reinsertion (P = 0.02). The TCC exit site appearance did not influence outcome in any of the therapeutic regimens. There was no statistical difference in TCC vintage (mean days±SEM) between those with treatment failure (217±71) and without treatment failure (163±18) [OR = 1, 95% 0.99–1.01, P = 0.5]. Data regarding TCC vintage was available in only 87 patients, and was not included in the multivariate analysis.


Figure 1
View larger version (8K):
[in this window]
[in a new window]
  		Fig. 1. Hazard plot of the incidence of treatment failure following initial TCC bacteremia, according to mode of TCC management. S salvage, DR delayed reinsertion, W exchange over wire.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Subset analysis of outcomes according to TCC management and microbiologic isolate

 
In a multivariate analysis, when controlling for the presence of diabetes mellitus, abnormal TCC exit site, hospitalization, hepatitis C, and intravenous iron administration, both S. aureus (OR = 4.2, P = 0.002), and TCC salvage (OR = 5.4, P = 0.003) were significant predictors of treatment failure (Table 5). A Cox regression analysis of time to treatment failure after the initial TCC bacteraemia, adjusted for S. aureus and TCC management (S vs all other) is provided in Figure 2. In a multivariate analysis, when controlling for TCC salvage, S. aureus was associated with a higher rate of developing an infectious complication (OR = 3.5, P = 0.02).


View this table:
[in this window]
[in a new window]
  		Table 5. Multivariate analysis using cox regression: treatment failure and infectious complications

 

Figure 2
View larger version (8K):
[in this window]
[in a new window]
  		Fig. 2. Hazard plot of the incidence of treatment failure comparing TCC salvage vs all other modes of TCC management, adjusted for the presence of S. aureus, S salvage, Allother: DR and W combined.

 
A series of subgroup analyses, including only the catheter management strategies of W vs DR were performed. Treatment failure (recurrent bacteraemia or death from sepsis) occurred in 11% of the DR group and 3% of the W group (OR = 4.2 [95% CI = 0.5–34], P = 0.15). Infectious complications occurred in 5% of the DR group and in 3% of the W group, (OR = 2 [95% CI = 0.2–18], P = 0.5.

Another set of subgroup analyses were performed including only episodes of S. aureus TCC bacteraemia. Of the 71 episodes of S. aureus bacteraemia, the catheter management strategy implemented was S in 15 episodes, W in seven episodes, DR in 36 episodes, and O in 13 episodes. Treatment failure (recurrent bacteraemia or death from sepsis) occurred in 15 episodes, and the incidence of treatment failure according to TCC management was as follows: S 53%, W 0%, DR 14%, O 15% (P = 0.006). There was a significantly higher incidence of treatment failure in episodes of S. aureus TCC bacteraemia when TCC salvage was performed compared to all other catheter management strategies, (OR = 8 [95%CI = 2.2–29.0], P = 0.001). Infectious complications occurred in nine episodes of S. aureus TCC bacteraemia, and the incidence of developing an infectious complication according to TCC management was as follows: S 20%, W 0%, DR 14%, O 8% (P = 0.56). Although the incidence of developing infectious complications in S. aureus was higher in the group in which TCC salvage was performed, it was not statistically significant (OR 2 [95% CI = 0.4–9.4], P = 0.35).



   Discussion
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
References
 
In the present study, using multivariate analysis controlling for potential confounding variables between all four modalities of TCC management, TCC salvage was found to be an important independent risk factor for treatment failure, as defined by recurrent bacteraemia or death due to sepsis. Other investigators have reported high treatment-failure rates with the TCC-salvage approach using univariate analysis [4,5]. Unfortunately, due to the relatively small number of cases in these studies multivariate analyses were either not performed or had insufficient power to control for more than one covariate. In addition, previous series compared one TCC-management strategy with one other. There are no previously published series evaluating all four possible TCC-management strategies in the same population. In the present study, another novel finding is the significant association between the presence of an abnormal TCC exit site with a higher rate of death from sepsis. Previous investigators did not report data regarding TCC exit site appearance in their series. This requires confirmation in future studies.

The reason why the TCC-salvage approach is associated with a high treatment-failure rate is probably due to the presence of a biofilm. In vivo electron micrograph studies of vascular catheters revealed bacteria covered by an extensive matrix composed of a hydrated matrix of polysaccharide and protein, the biofilm [11]. The incorporation of microorganisms into the intra- and extraluminal biofilm is important in the pathogenesis of bacterial resistance to systemic antibiotics, and biofilm infections usually persist until the colonized surface is removed [11]. Bacteria embedded in biofilm have been shown to survive after treatment with antibiotic concentrations substantially higher than the minimal inhibitory concentration [12]. The susceptibility of S. aureus to vancomycin and beta lactam antibiotics is markedly reduced when grown adherent to silicone catheter surfaces [13]. Possible mechanisms of resistance of bacteria grown within catheter biofilms include inadequate or slow penetration of antibiotic or inactivation of antibiotic in the biofilm, altered chemical microenvironment within the biofilm, and the development of a phenotypic alteration in the organism to a hibernated state characterized by slow bacterial growth [11,13].

Although not recommended in the Dialysis Outcomes Quality Initiative Guidelines for Vascular Access 2000, attempts at TCC salvage are still utilized by nephrologists because it is associated with the preservation of haemodialysis vascular access, uninterrupted haemodialysis scheduling, low immediate costs and convenience [14]. In a recent Infectious Disease review, Bouza et al. provide criteria (all five must be present) for cases in which catheter salvage be recommended. The criteria are as follows: difficult to replace catheters, blood sterile in 48–72 h, no sign of tunnel infection, no signs of metastatic infection, microorganisms medically treatable, and a haemodynamically stable patient [15]. We report a 5-fold higher risk of treatment failure when TCC salvage is attempted, and an 8-fold higher risk in cases of TCC salvage associated with S. aureus bacteraemia. One could argue that, due to the high risk of treatment failure associated with TCC salvage, even when adjusted for tunnel infection, and other potential variables that may effect outcome, salvage should be used only as a treatment of last resort. Preservation of vascular access can be maintained by exchanging the TCC over a guidewire, which is associated with a significantly lower treatment failure rate (3%).

S. aureus is an important independent predictor for adverse outcomes after TCC bacteraemia. Ours is the first study to report an association between S. aureus and the risk of higher treatment-failure rate, defined as recurrent bacteraemia or septic death. The high prevalence of S. aureus nasal carriage (35–62%) in the haemodialysis population may contribute to the higher rate of recurrent bacteraemia associated with this organism. Alternatively, colonization of S. aureus within the TCC biofilm, or tissue-trophism of S. aureus beyond the TCC are other possible sources of recurrent bacteraemia. Mortality rates of between 6–34% have been reported in association with S. aureus bacteraemia, and rates are higher when infectious complications are present [10,16]. In the present study, failure to remove or exchange the TCC when S. aureus was the infecting organism was associated with an 8-fold higher risk of treatment failure.

We are the first to find a statistically significant association between infection with S. aureus and the development of infectious complications using multivariate analysis, controlling for TCC management. Marr reported a higher complication rate in S. aureus TCC bacteraemia (41 vs 24% occurring with other gram positive organisms), although this did not reach statistical significance [10]. The virulence of S. aureus is, in part, due to its capacity to adhere and invade endovascular tissue, and its release of proteolytic enzymes which contribute to the development of metastatic invasive infections [16].

Study limitations
A limitation of the present study is the study's observational design. Although a randomized control trial is a more robust study design, we attempted to control for differences between patients in the four TCC-management strategies by adjusting for significant variables in the multivariate analysis. Our study was also limited by an unexpectedly lower event incidence than anticipated. The treatment failure rates for the previous series were between 47–65% for episodes of TCC bacteraemia in which catheters were left in, ‘TCC salvage’, whereas treatment failure occurred in only 26% TCC salvage in the present study.

In conclusion, this is the first large series to use multivariate analyses, controlling for confounding variables, to identify important predicators of adverse outcome after an initial episode of TCC-associated bacteraemia in haemodialysis patients. TCC salvage is a significant independent risk factor for treatment failure, even when controlling for potential confounding factors. This is the first report of an association of S. aureus with a significantly higher incidence of treatment failure following TCC bacteraemia. We confirm the association of S. aureus with the development of infectious complications. This is the first study to report that an infected TCC exit site appearance is an independent risk factor for death due to sepsis. Future studies, preferably randomized control trials, are needed to compare outcomes between TCC guidewire exchange versus immediate removal with delayed TCC reinsertion in clinically stable patients.



   Acknowledgments
 
The authors wish to thank Dulce Eleazar, RN, Ann Starinovich, RN and Peter Durkin, PA for their assistance in compiling the data. We also wish to thank our colleagues in the Bronx, NY (Maria Coco MD, MS, Manash Dasgupta MD, Michael Guccione MD, Robert Feingold MD, Vaughn Folkert MD, Joel Neugarten MD, Sushil Sagar MD) and Hartford, Connecticut (John D’Avella MD) for participating in this study. Dr M.H.M is funded with a grant from the Aetna Foundation.

Conflict of interest. None of the authors have had involvements that might raise the question of bias in the work reported or in the conclusions, implications, or opinions stated.



   References
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 References
 

  1. U.S. Renal Data System, USRDS 2002 Annual Data Report: Atlas of End-Stage Renal Disease in the United States, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2002, Figure 4.30
  2. Dryden MS, Samson A, Ludlam HA, Wing AJ, Phillips I. Infective complications associated with the use of Quinton Permcath for long-term central vascular access in hemodialysis. J Hosp Infect 1991; 19: 257–262[CrossRef][Web of Science][Medline]
  3. Moss AH, Vasilakis C, Holley JL, Foulks CJ, Pillai K, McDowell DE. Use of a silicone dual-lumen catheter with a Dacron cuff as a long-term vascular access for hemodialysis patients. Am J Kidney Dis 1990; 16: 211[Web of Science][Medline]
  4. Marr KA, Sexton DJ, Conlon PJ, Corey R, Schwab SJ, Kirkland KB. Catheter-related bacteremia and outcome of attempted catheter salvage in patients undergoing hemodialysis. Ann Intern Med 1997; 127: 275–280[Abstract/Free Full Text]
  5. Saad TF. Bacteremia associated with tunneled, cuffed haemodialysis catheters. Am J Kidney Dis 1999; 34: 1114–1124[Web of Science][Medline]
  6. Mokrzycki MH, Schroppel B, von Gersdorff G, Rush H, Zdunek M, Feingold R. Tunneled cuffed catheter associated infections in hemodialysis patients seropositive for the human immunodeficiency virus. J Am Soc Nephrol 2000; 11: 2122–2127[Abstract/Free Full Text]
  7. Mokrzycki MH, Jean-Jerome K, Rush H, Zdunek MP, Rosenberg SO. A randomized trial of minidose warfarin for the prevention of late malfunction in tunneled, cuffed hemodialysis catheters. Kidney Int 2001; 59: 1935–1542[CrossRef][Web of Science][Medline]
  8. Tokars JI, Miller ER, Stein G. New national surveillance system for hemodialysis-associated infections: initial results. Am J Infect Control 2002; 30: 288–295[CrossRef][Web of Science][Medline]
  9. Beathard GA. Management of bacteremia associated with tunneled-cuffed haemodialysis catheters. J Am Soc Nephrol 1999; 10: 1045–1049[Abstract/Free Full Text]
  10. Marr KA, Kong LK, Fowler VG et al. Incidence and outcome of bacteremia in hemodialysis patients. Kidney Int 1998; 54: 1684–1689[CrossRef][Web of Science][Medline]
  11. Stewart PS, Costerton JW. Antibiotic resistance of bacteria in biofilm. Lancet 2001; 358: 135–138[CrossRef][Web of Science][Medline]
  12. Ceri H, Olson ME, Stremick C et al. The Calgary biofilm device: new technology for rapid determination of antibiotic susceptibilities of bacterial biofilms. J Clin Microbiol 1999; 37: 1771–1776[Abstract/Free Full Text]
  13. Williams I, Venables WA, Lloyd D et al. The effects of adherence to silicone surfaces on antibiotic susceptibility in Staphylococcus aureus. Microbiology 1997; 143: 2407–2413[Abstract/Free Full Text]
  14. Mokrzycki MH, Singhal A. Cost-effectiveness of three strategies of managing tunneled-cuffed haemodialysis catheters (TCC) in clinically mild or asymptomatic bacteraemias. Nephrol Dial Transpl 2002; 17: 2196–2203[Abstract/Free Full Text]
  15. Bouza E, Burillo A, Munoz P. Catheter related infections: diagnosis and intravascular treatment. Clin Microbiol Inf 2002; 8: 265–274[CrossRef]
  16. Lowy FD. Staphylococcus aureus infections. N Engl J Med 1998; 339: 520–532[Free Full Text]
Received for publication: 21. 3.05
Accepted in revised form: 4. 8.05


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


This article has been cited by other articles:


Home page
 Am J Health Syst PharmHome page
B. S. Crawford, R. F. Largen, T. Walton, and J. J. Doran
Once-weekly vancomycin for patients receiving high-flux hemodialysis
Am. J. Health Syst. Pharm., July 1, 2008; 65(13): 1248 - 1253.
[Abstract] [Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
C. Ponticelli
New recommendations in the treatment of Gram-positive bacteraemia in dialysis patients
Nephrol. Dial. Transplant., January 1, 2008; 23(1): 27 - 32.
[Full Text] [PDF]

Home page
 Eur Heart JHome page
G. Nucifora, L. P. Badano, P. Viale, P. Gianfagna, G. Allocca, D. Montanaro, U. Livi, and P. M. Fioretti
Infective endocarditis in chronic haemodialysis patients: an increasing clinical challenge
Eur. Heart J., October 1, 2007; 28(19): 2307 - 2312.
[Abstract] [Full Text] [PDF]

Home page
 CJASNHome page
M. Allon
Current Management of Vascular Access
Clin. J. Am. Soc. Nephrol., July 1, 2007; 2(4): 786 - 800.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
21/4/1024    most recent
gfi104v1
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 (18)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Mokrzycki, M. H.
Right arrow Articles by Rosenberg, S. O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mokrzycki, M. H.
Right arrow Articles by Rosenberg, S. O.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?