Fax: (713) 792-8233
Novel approach using antimicrobial catheters to improve the management of central line-associated bloodstream infections in cancer patients
Article first published online: 14 DEC 2010
Copyright © 2010 American Cancer Society
Volume 117, Issue 11, pages 2551–2558, 1 June 2011
How to Cite
Chaftari, A.-M., Kassis, C., El Issa, H., Al Wohoush, I., Jiang, Y., Rangaraj, G., Caillouet, B., Pravinkumar, S. E., Hachem, R. Y. and Raad, I. I. (2011), Novel approach using antimicrobial catheters to improve the management of central line-associated bloodstream infections in cancer patients. Cancer, 117: 2551–2558. doi: 10.1002/cncr.25807
- Issue published online: 19 MAY 2011
- Article first published online: 14 DEC 2010
- Manuscript Accepted: 25 OCT 2010
- Manuscript Revised: 9 SEP 2010
- Manuscript Received: 7 JUL 2010
- antimicrobial catheters;
- catheter exchange;
Central venous catheter (CVC) removal has often been recommended for the treatment of central line-associated bloodstream infections (CLABSIs). However, CVC removal is not always practical in patients with cancer, and changing CVCs with noncoated CVCs over guidewire may result in cross-infection of the new CVC. Therefore, the current matched retrospective cohort study was conducted to evaluate the effectiveness of exchanging infected CVCs for minocycline- and rifampin (MR)-coated CVCs in cancer patients with CLABSIs.
The authors identified all cancer patients with CLABSIs who had undergone either CVC exchange with MR-coated CVCs or CVC removal at the study institution. All patients were treated with appropriate systemic antibiotics. The exchange group was matched in a 1:2 ratio with the removal group by organism, underlying disease, and neutropenia. The demographics, clinical characteristics, and outcome were compared. Overall response was defined as the resolution of clinical signs and symptoms and eradication of bacteremia within 72 hours after CVC exchange or removal, without disease recurrence or infection-related death.
A total of 120 cancer patients were included (40 in the exchange group and 80 in the removal group). Overall response rates were 95% in the exchange group and 76% in the removal group (P = .011). No disease recurrences or infection-related deaths occurred in the exchange group; 8 disease recurrences or deaths (11%) occurred in the removal group (P = .05). Patients in the exchange group also experienced lower rates of mechanical failure (3% vs 15%; P = .049).
Exchanging CVCs for MR-coated CVCs in cancer patients with CLABSIs may improve the overall response rate and decrease the risk of mechanical failure, disease recurrence, and infection-related mortality. Cancer 2011. © 2010 American Cancer Society.
Central venous catheters (CVCs) are essential in the treatment of patients with cancer, chronic diseases and critically ill patients. However, they are the leading source of bloodstream infections.1 CVC removal has been recommended for the treatment of central line-associated bloodstream infections (CLABSIs).2 However, in critically ill patients and patients with cancer, CVC removal is not always possible, and changing CVCs over a guidewire with noncoated CVCs can result in cross-infection of the newly exchanged CVCs.3 The insertion of a new CVC at a different site may be associated with serious mechanical complications such as pneumothorax and bleeding, especially in cancer patients with thrombocytopenia.4-6 In a large prospective randomized trial, the insertion of catheters in the subclavian vein was associated with a 9.7% rate of mechanical complications that increased to 24% with >2 passes.6 Cobb et al indicated that exchanging catheters with the use of a guidewire increases the risk of bloodstream infections; however, insertion of a catheter at a new site was associated with more mechanical complications (5% in the insertion group vs 1% in the exchange group; P = .005).4
Multiple randomized controlled studies and meta-analyses have demonstrated that minocycline- and rifampin (MR)-coated catheters substantially decrease the risk of CLABSIs.7-11 To the best of our knowledge, none of the previous exchange studies used the MR-coated CVCs to replace the infected CVCs in patients with bacteremia. Hence, the risk of cross-infections with the newly exchanged antimicrobial MR-coated catheters is unknown. Therefore, in this matched retrospective cohort study, we evaluated the effectiveness of exchanging infected CVCs with MR-coated CVCs in cancer patients with CLABSIs.
MATERIALS AND METHODS
We searched the infection control surveillance database at The University of Texas M. D. Anderson Cancer Center (MDACC) in Houston between July 2005 and July 2008 to identify all cancer patients who had developed CLABSIs. This retrospective study was approved by the Institutional Review Board, and a waiver of informed consent was obtained.
We included all cancer patients aged ≥16 years in whom an infected CVC had been exchanged for an MR-coated catheter over a guidewire. However, we excluded all patients in whom MR-coated CVCs had been inserted within 4 weeks before the CLABSI occurred. We matched each exchange patient with 2 historical control cancer patients who had been diagnosed with CLABSIs between June 2004 and July 2008 and had undergone removal of the infected CVC. Control patients were also aged ≥16 years and were matched for the same organisms, underlying disease (hematologic or solid tumor), and neutropenic status as exchange cases. Data were collected from the institutional electronic medical records and included demographics, cancer type, bacteremia date, organism, blood culture source (CVC or peripheral vein), blood culture colony count, time to positivity, CVC type, CVC insertion date and site, number of lumens, CVC exchange or removal date, clinical signs and symptoms, hospital admission and duration, intensive care unit admission, fever resolution date, microbiologic eradication date, disease recurrence within 3 months, complications, and death. We also recorded the type and duration of antibiotic therapy. Each patient was followed for 3 months from the date of CVC exchange or removal or until death if it occurred during the follow-up period.
CLABSIs were defined as per the Centers for Disease Control and Prevention (CDC) criteria.12, 13 Patients with intravascular catheters were considered to have a CLABSI if they had at least 1 positive blood culture with a recognized pathogen or 2 positive blood cultures with a common skin contaminant in the presence of clinical manifestations of infection and no apparent source for the CLABSI except the catheter.13
Neutropenia was defined as an absolute neutrophil count of <1000/μL. Severe neutropenia was defined as a count of <500/μL.
Overall response was defined as the resolution of clinical signs and symptoms and the eradication of bacteremia within 72 hours of CVC exchange or removal and no evidence of 1) disease recurrence within the first 3 months; 2) related deep-seated infections occurring >1 week after exchange or removal and during follow-up; or 3) infection-related mortality.
Disease recurrence was defined as bacteremia recurrence with the same organism (similar species and antibiogram) after an initial clinical or microbiologic response to antibiotics and within 3 months of CVC exchange or removal. Mechanical failure was defined as the inability to insert a CVC at a different site, exchange a CVC over a guidewire, or use a new CVC after the CLABSI, requiring ultimate and definite removal of the CVC.
Infection-related complications were defined as the occurrence of either disease recurrence, related deep-seated infections occurring >1 week after exchange or removal and during follow-up, or infection-related death.
Appropriate antibiotic therapy represented the administration of antibiotics approved for the treatment of CLABSIs and active against the bacteria isolated as causing the CLABSI based on hospital susceptibility pattern.
The chi-square or Fisher exact tests were used to compare categorical variables, as appropriate. Continuous variables were compared using Wilcoxon rank sum tests because of the data's deviation from normal distribution. In addition, survival analysis of late infection-related complications was performed. The patients were followed for 3 months after the catheter was exchanged or removed. The probability of being free from late infection-related complications was estimated using the Kaplan-Meier method for those patients in whom the catheter was removed and those in whom the catheter was exchanged, respectively, and was compared by the log-rank test. Patients who died during the follow-up period were considered to be censored observations. All tests were 2-sided and statistical significance was set at P ≤ .05. The statistical analyses were performed using SAS statistical software (version 9.1; SAS Institute Inc, Cary, NC).
During the study period, we observed 40 cancer patients with CLABSIs who had their infected catheter exchanged over a guidewire for an MR-coated CVC. These patients were matched on the basis of organism, neutropenia, and cancer type with 80 patients with CLABSIs who had their CVC removed.
In the exchange group, there were 16 episodes of gram-negative CLABSIs and 24 episodes of gram-positive CLABSIs. The isolated micro-organisms causing the CLABSIs are listed in Table 1. There was no significant difference in the polymicrobial infection rate noted between the groups (15% in the exchange group vs 8% in the removal group; P = .21) (Table 2).
|Micro-Organisms Causing CLABSI||No. of Cases|
|Methicillin-sensitive Staphylococcus aureus||3|
|Methicillin-resistant S. aureus||2|
|Vancomycin-resistant E. faecium||1|
|Characteristics||Exchange Group (n = 40) No. (%)||Removal Group (n = 80) No. (%)||P|
|Hematologic||23 (58)||46 (58)|
|Solid tumor||17 (42)||34 (42)|
|Neutropenia, <500/μL||11 (28)||26 (33)||.58|
|Neutropenia, <1000/μL)||13 (33)||27 (34)||.89|
|Gram stain of first organism||>.99|
|Gram positive||24 (60)||48 (60)|
|Gram negative||16 (40)||32 (40)|
|Polymicrobial bacteremia||6 (15)||6 (8)||.21|
|CNS||15 (38)||30 (38)|
|Non-CNS||25 (62)||50 (62)|
The percentage of long-term and short-term catheters used was similar in both groups (Table 3). There were more peripherally inserted central catheters (PICC) (basilic and cephalic locations) in the removal group. The significant difference in catheter location is a by-product of the finding that PICC lines once inserted are difficult to exchange , although they could be easily removed and inserted in a different location. The majority of the catheters were double lumen in both groups; however, there were more single-lumen catheters in the removal group and more triple-lumen catheters in the exchange group (Table 3).
|Characteristics||Exchange Group (n = 40) No. (%)||Removal Group (n = 80) No. (%)||P|
|Median age (range), y||54 (17-76)||60 (16-80)||.2|
|Male sex||28 (70)||49 (61)||.35|
|History of bone marrow transplantation, no. (%)||9 (23)||6 (8)||.019|
|Temperature at presentation, no. (%)||.4|
|>38°C||29 (72)||66 (82)|
|36-38°C||9 (23)||12 (15)|
|<36°C||2 (5)||2 (3)|
|Sepsis syndrome at presentation, no. (%)||33 (83)||67 (84)||.86|
|Median no. of platelets at exchange or removal (range), ×103||81 (4-524)||96 (2-599)||.73|
|Type of catheter|
|Long term (silicone), no. (%)||31 (78)||63 (79)|
|Short term (polyurethane), no. (%)||9 (22)||17 (21)|
|Single, no. (%)||2 (5)||25 (31)|
|Double, no. (%)||31 (78)||52 (65)|
|Triple, no. (%)||7 (17)||3 (4)|
|Basilic, no. (%)||3 (8)||29 (36)|
|Cephalic, no. (%)||0||5 (6)|
|Femoral, no. (%)||0||1 (1)|
|Jugular, no. (%)||2 (5)||7 (9)|
|Subclavian, no. (%)||35 (87)||38 (48)|
|Catheter insertion side||.36|
|Left, no. (%)||15 (38)||37 (46)|
|Right, no. (%)||25 (62)||43 (54)|
|Signs of inflammation at insertion site, no. (%)||2 (5)||12 (15)||.14|
|Intensive care unit admission, no. (%)||4 (10)||18 (23)||.1|
|Intubation, no. (%)||1 (3)||10 (13)||.1|
|Median no. of d between bacteremia and exchange or removal (range)||3 (0-8)||2 (0-12)||.022|
The 2 groups were similar with regard to risk factors, including sepsis at the time of presentation (Table 3). Despite trends that suggest that the control group was more likely to be hospitalized in the intensive care unit (10% for the exchange group vs 23% for the removal group; P = .1) and to be intubated (3% for the exchange group vs 13% for the removal group; P = .1), significantly more patients in the exchange group had undergone hematopoietic stem cell transplantation compared with those in the control removal group (23% vs 8%; P = .019) (Table 3). All patients had been treated with appropriate systemic antibiotic therapy in addition to the exchange or removal of the infected CVC. The 2 groups were comparable for the duration of appropriate antibiotics, with a median duration of 14 days for both groups (Table 4).
|Outcome||Exchange Group||Removal Group||P|
|Median duration of antibiotic treatment (range), d||14 (5-40)||14 (1-47)||.81|
|Fever resolution within 3 d after exchange or removal, no./total (%)a||38/40 (95)||68/79 (86)||.21|
|Persistent bacteremia ≥3 d after exchange or removal, no./total (%)||2/40 (5)||3/80 (4)||>.99|
|Mean duration in d of hospitalization after bacteremia, no. (range)||6 (1-87)||7 (1-68)||.87|
|Deep-seated infection occurring >1 wk after exchange or removal, no./total (%)||0/40 (0)||1/80 (1)||>.99|
|Death during hospital stay, no./total (%)||0/40 (0)||5/80 (6)||.17|
|Infection-related death within 3 mo of infection, no./total (%)b||0/40 (0)||4/78 (5)||.3|
|Disease recurrence, no./total (%)c||0/38 (0)||5/75 (7)||.17|
|Recurrence or infection-related death, no./total (%)b||0/38 (0)||8/76 (11)||.05|
|Overall response, no./total (%)d||38/40 (95)||58/76 (76)||.011|
|Mechanical failure, no./total (%)e||1/40 (3)||10/66 (15)||.049|
We found a significant difference in the median number of days between the date of bacteremia and the exchange or removal of the infected CVC, with the exchange occurring 1 day later in the exchange group compared with the removal group (3 days vs 2 days; P = .022) (Table 3). There was no statistically significant difference noted with regard to the rate of bacteremia persistence >72 hours after CVC exchange or removal (5% in the exchange group vs 4% in the removal group; P > .99) (Table 4). At 72 hours, the fever resolution rates were similar in both groups (95% in the exchange group vs 86% in the removal group; P = .21).
No disease recurrences or infection-related deaths were reported to occur during hospital stays in the exchange group; however, in the removal control group, 8 patients experienced a recurrence or an infection-related death (0% vs 11%; P = .05) (Table 4). In these patients, the infected catheter was removed after a median of 3 days (range, 0 days-6 days) from the date of bacteremia.
A log-rank test indicated that patients in whom the catheter was removed were more likely to develop infection-related complications than those in whom the catheter was exchanged during the 3-month period of follow-up after catheters were removed or exchanged (P = .03) (Fig. 1). The overall response rates to antibiotic therapy were higher in the exchange group (95%) than in the removal group (76%) (P = .011) (Table 4).
All the patients, including those who had an unfavorable outcome (Table 5) (such as persistent of bacteremia, disease recurrence, deep-seated infection, or infection-related death), received appropriate antibiotic therapy that was active against the bacteria isolated as causing the CLABSI.
|Type of Complication||Patient (Catheter Exchanged) or Control (Catheter Removed)||Catheter Removal or Exchange||Pathogen||No. of Days to Removal or Exchange From Date of Bacteremia||No. of Days to Initiation of Antibiotic Therapy||Treatment Duration, Days|
|Persistent bacteremia||Patient 1||Exchange||Klebsiella pneumoniae||3||0||16|
|Disease recurrence||Control 1||Removal||VRE||6||1||36|
|Control 7||Removal||Escherichia coli||3||0||9|
|Deep-seated infection||Control 2||Removal||MSSA||6||2||19|
|Infection-related death||Control 1||Removal||VRE||6||1||36|
|Control 9||Removal||Pseudomonas aeruginosa||3||0||13a|
In the exchange group at the time of last follow-up, 13 patients retained their newly exchanged CVCs, 9 patients had undergone a second exchange that occurred after a median duration of 34 days (range, 0 days-123 days), and 18 patients had undergone CVC removal after a median duration of 52 days (range, 2 days-179 days). We performed a subset analysis that included only those patients who retained the exchanged CVCs for ≥7 days, unless their CVCs had been removed earlier for persistent bacteremia or disease recurrence. The overall response rate was significantly higher in the exchange group than in the control group (94% vs 76%, respectively; P = .025).
In the removal group, among the 66 patients (83%) who had undergone at least 1 attempt to insert the CVC at a different site within 3 months after CVC removal, 10 patients (15%) experienced a mechanical failure. However, in the exchange group, only 1 patient (3%) experienced a mechanical failure after an unsuccessful exchange that necessitated the ultimate removal of the exchanged CVC (P = .049) (Table 4) (Fig. 2).
The results of the current study indicate that, in cancer patients, exchanging infected CVCs in the setting of CLABSI over a guidewire for MR-coated CVCs is more effective at reducing the risk of mechanical complications when compared with CVC removal. Furthermore, antimicrobial MR-coated CVC exchange improves the overall response to systemic antibiotic treatment and decreases the risk of infection-related complications.
The results of several studies suggest that CVC exchange is useful in the treatment of patients with CLABSIs.14-20 However, most of these studies were open trials (without comparators such as CVC removal), included small numbers of CLABSI episodes (mostly coagulase-negative Staphylococcus CLABSIs), or used loose definitions of CLABSI. None of these studies used antimicrobial CVCs to replace infected CVCs.14-18 In 2 comparative nonrandomized studies, the CVC was exchanged only in a subset of patients with mild symptoms who experienced a response to antimicrobial therapy within 48 hours.19, 20 The lack of randomization, uniform entry criteria, and use of antimicrobial CVCs, and the inconsistent outcome definitions and follow-up procedures have made it difficult to evaluate the role of CVC exchange, particularly antimicrobial CVCs, in patients with documented CLABSIs.
In the current study, approximately 62% of the CLABSIs were caused by non-coagulase–negative Staphylococcus organisms, including gram-positive and gram-negative pathogens. Infected CVCs were exchanged for MR-coated antimicrobial CVCs at a mean of 3 days after the initial occurrence of bacteremia. Furthermore, patients had been followed for 3 months after the exchange.
Exchanging colonized or infected CVCs over a guidewire may lead to cross-colonization of the lumen of newly exchanged, uncoated CVCs, which may perpetuate the CLABSI. Thus, this practice is not advocated in the setting of CLABSI by the CDC in its guidelines regarding CLABSI prevention.2 Olson et al demonstrated that uncoated CVCs that were exchanged became colonized by bacteria within 48 hours.21 Pettigrew et al demonstrated that organisms can be transferred to the newly exchanged, uncoated CVC through the guidewire.3
Both in vitro and in vivo studies have shown that coating the internal and external surfaces of CVCs with MR prevents the surface adherence of resistant bacteria.22-25 These antimicrobial CVCs are also highly effective at preventing CLABSIs in high-risk patients.7-10 Hence, organisms transferred through the guidewire to MR-coated CVCs might be inhibited by antimicrobials and fail to attach to the catheter surface.
In the current study, CVC exchange with MR-coated CVCs led to a higher overall response to systemic antibiotic therapy as well as a significant decrease in infection-related complications, although there were more transplantation patients in the exchange group compared with the removal group (P = .019) (Table 3). Furthermore, the exchange group had another risk factor, namely the longer duration between the onset of bacteremia and intervention (3 days in the exchange group vs 2 days in the removal group; P = .022) (Table 3). This finding of improved outcome despite the risks of transplantation and delayed intervention further demonstrates the efficacy and validity of this approach. Although not statistically significant, more patients in the removal arm were admitted to the intensive care unit. This could suggest that the removal group could have been sicker and thus would have been more likely to have an unfavorable outcome.
The insertion of a new CVC at a different site is often associated with serious mechanical complications such as hemopneumothorax.4-6 Exchanging CVCs over a guidewire decreases the likelihood of mechanical complications in patients with cancer, especially those with thrombocytopenia and high-risk patients in the intensive care unit with hypocoagulation conditions such as disseminated intravascular coagulation. In the current study, more functional CVCs resulted from CVC exchange than from removal and reinsertion. There was a lower rate of mechanical failure in the exchange group (3%) compared with the removal group (15%) (P = .049). Exchanging the CVC over a guidewire may also result in less discomfort and pain for the patient, thus decreasing symptom burden and anxiety.
Not only is CVC exchange with MR-coated CVCs safer and more convenient than the insertion of a new CVC, it can also result in cost savings. At MDACC, CVC exchanges are performed by nurses, whereas CVC insertions at new sites are performed by physicians or physician extenders. No professional fee is incurred when a nurse performs an exchange; however, there is an approximately $1500 fee when a physician or physician extender inserts a new CVC. There is also a $750 fee for the sonography required for CVC insertion; no imaging is required for a CVC exchange. Thus, the estimated cost savings at MDACC between exchange and reinsertion procedures is $2250. The cost of reinserting tunneled and other surgically implantable CVCs (ports) would further increase the cost of removal and reinsertion because of the costs related to the operating room, anesthesia, recovery room, or interventional radiology. The total cost could also include the treatment of noninfectious mechanical complications such as pneumothorax and bleeding.
The current study is subject to several limitations. The first is its observational, retrospective, nonrandomized design, although it was a matched cohort study. Patients were not on defined prospective clinical protocols, and symptoms and signs were not consistently monitored. Serial blood cultures were not always available during or after infections, and bacteremia resolution was not always documented. As a result, we may have underestimated the duration of true bacteremia and the recurrence rate. In addition, a relatively small number of patients in the current study had undergone CVC exchange. Another limitation is that we did not standardize the types and durations of antibiotic therapy used or the timing of CVC exchange or removal from the initial date of bacteremia (0 days-8 days in the exchange group, with a median duration of 3 days, and 0 days-12 days in the removal group, with a median duration of 2 days). However, the 2 groups received appropriate antibiotic therapy and were comparable in terms of the duration of such antibiotic therapy.
In an attempt to salvage the vascular access, this novel approach of exchanging the catheter over a guidewire with an antimicrobial MR-coated catheter may change the practice paradigm in the management of patients with CLABSIs. It may improve patient care and outcome while decreasing the rate of mechanical complications and may be associated with cost savings. A well-designed, multicenter, prospective, randomized clinical trial is needed to validate the results of the current study and to determine the efficacy, safety, and cost-effectiveness of the salvage intervention when compared with the removal of the catheter and insertion of a new CVC at a different vascular site. Other salvage interventions such as the use of effective lock therapy in patients with nonexchangeable, surgical tunneled, or implantable catheters such as ports warrant further investigation through well-designed prospective, randomized studies.
Exchanging CVCs for MR-coated CVCs in cancer patients with CLABSIs may be associated with a better outcome compared with that of CVC removal and reinsertion. It decreases the rate of mechanical complications, and also may improve the overall response to antibiotic therapy as well as decrease the risks of late infection-related complications. Large, prospective, randomized clinical trials are needed to validate these findings and to evaluate the cost savings as well as the improvement in quality of life associated with the potential decrease in symptom burden.
CONFLICT OF INTEREST DISCLOSURES
Dr. Raad is co-inventor of technology related to minocycline and rifampin-coated catheters. This technology is licensed to Cook, Inc. Dr. Raad receives royalties related to this technology, which is owned by The University of Texas M. D. Anderson Cancer Center.
- 7Central venous catheters coated with minocycline and rifampin for the prevention of catheter-related colonization and bloodstream infections. A randomized, double-blind trial. The Texas Medical Center Catheter Study Group. Ann Intern Med. 1997; 127: 267-274., , , et al.
- 12Guidelines for the prevention of intravascular catheter-related infections. MMWR Morb Mortal Wkly Rep. In press., , .