NAN200045_212-218

7/2/09

7:28 PM

Page 212

The Art and Science of Infusion Nursing

Alice Mathew, MSN, RN, CRNI Tracey Gaslin, PhD, RN, CPNP, CRNI Kari Dunning, PhD, PT Jun Ying, PhD

Central Catheter Blood Sampling

The Impact of Changing the Needleless Caps Prior to Collection

ABSTRACT Little is known about the association between central catheter needleless connectors and bacteremia. In a cohort study on 91 patients, central catheter blood samples were collected using 3 methodsold cap (the existing cap), new cap (after replacing the old cap with a new sterile cap), and peripheral methodsfrom each patient and their correlation was examined. The old cap method identified 36 positive bacteremia cases. However, only 17 cases were verified by the new cap method, yielding a positive predictive value of 47.2% (17/36). The 19 false-positive cases indicated old cap contamination. This study recommends that changing the needleless cap before drawing blood samples would be an ideal practice for obtaining more specific and reliable results in diagnosing bacteremia.

C
212

entral catheters provide a reliable intravenous access that complies with todays treatment regimens in acute care and longterm acute care (LTAC) settings. Central catheters enable infusion of intravenous fluids, pharmacotherapies, parenteral nutrition, and blood products. Central catheters are also used for
Author Affiliations: Drake Center (Ms Mathew and Dr Dunning) and Department of Public Health, University of Cincinnati (Dr Ying), Cincinnati, Ohio; and Indiana Wesleyan University, Marion, Indiana (Dr Gaslin). Corresponding Author: Alice Mathew, MSN, RN, CRNI, Drake Center, 151 W Galbraith Rd, Cincinnati, OH 45216 (alicejoysm@yahoo. com).

blood samplings and hemodynamic monitoring. Central catheters remain an integral part of critical care medicine.1,2 Although central catheters serve many purposes and their need is inevitable, frequent accessing of these catheters by healthcare professionals creates an opportunity for developing central catheter infections in patients.1 Because of recent Medicaid and Medicare reforms, more acutely ill patients are discharged from acute care settings to LTAC facilities, causing an increasing number of patients to be treated with central catheters in LTAC settings.3 While central catheters contribute positively to patient care, as many as a quarter of a million bloodstream infections (BSIs) are attributed to central catheters, which increase costs, length of hospitalization, and mortality and morbidity rates.2,4,5 Needleless catheter connectors, otherwise known as catheter caps, are an integral component of an infusion system. Catheter caps were initially designed to reduce needlestick injuries among clinicians and to provide a safer workplace environment. The use of needleless caps on central catheters has reduced needlestick injuries among healthcare clinicians.1,6 However, infusion specialists are concerned about the use of needleless central catheter caps in association with BSIs.7 This becomes an even larger concern due to the new regulations of the Centers for Medicare & Medicaid Services that threaten nonreimbursement for hospital-acquired infections, including catheter-related bloodstream infections (CR-BSIs). Therefore, clinicians are investigating new interventions and evidence to decrease nosocomial (hospital-acquired) infections such as CR-BSIs in all healthcare settings.8,9 BSIs are serious infections in hospitalized patients requiring intensive care. Intravenous CR-BSIs are widely considered the most preventable cause of infections occurring in a hospital. It is estimated that central catheter-related BSIs develop in more than 250,000 patients

Journal of Infusion Nursing

NAN200045_212-218

7/2/09

7:28 PM

Page 213

in US hospitals each year.4,10 Among central catheters, the mortality risk has been reported to be as high as 30% in intensive care units, costing from $4000 to $56,000 per episode.4

SIGNIFICANCE OF THE PROBLEM

In the past 2 decades, central catheters have become more widely used in all hospital settings.1 CR-BSIs are considered nosocomial (hospital acquired) in origin. Nosocomial BSIs are the sixth leading cause of death in the United States.11 Nosocomial infections are a threat to patient safety because microorganisms that are responsible for nosocomial infections often add to functional disability and emotional stress and may be resistant to many antibiotics used in hospitals.12 It is documented that central catheters are the most common source of infection for hospital-acquired bacteremia (HAB).1 Previous studies and anecdotal experiences showed discordances in centrally drawn and peripherally drawn blood culture results.5,13 Common catheter blood-drawing procedures involve cleaning the catheter needleless cap with isopropyl alcohol before drawing the blood sample to avoid contamination.14 In the authors anecdotal experience, rates of positive catheter-drawn blood cultures were higher than the rates of samples drawn from peripheral venipuncture. The authors have also observed a decrease in positive catheter-drawn cultures and an increase in concordance between catheter samples. One possible reason for a catheter-positive and peripheralnegative discordant result could be a contaminated central catheter needleless cap. Cleaning the catheter with isopropyl alcohol prior to a blood draw might not be adequate. Therefore, the purpose of this study was to examine the correlation between blood culture collection before changing a central catheter cap (with the existing cap), blood culture collection after changing and replacing with a new cap, and blood culture collection from a peripheral venipuncture. An analysis of the 3 venues of blood culture collection could provide information necessary to determine whether collection site and collection procedures had an impact on the blood culture results. Since no study comparing blood culture collection methods has been conducted, this study would be filling the gap in the current literature. The goal of this study was to assess diagnostic accuracy of the old cap method and the peripheral method, using the new cap method as the gold standard and comparing diagnostic agreements between blood drawing methods.

LITERATURE REVIEW
When patients have central catheters, these catheters provide ready access for blood sample collection. However,

the use of catheter-drawn samples is controversial due to concerns of contamination.12,15,16 Other studies suggest that the use of catheter samples is useful.5,17 In a hospital-wide surveillance on HAB, using a standard protocol created by the Nosocomial Infection National Surveillance Scheme, the researchers found that the overall incidence of HAB was higher in teaching than in nonteaching hospitals. The study found that 5.39 and 2.83 HABs per 1000 patients were at risk; the devicerelated sources were responsible for 52.4% and 43.2% of all HABs, and central catheters were the common source of HABs, causing 38.3% and 22.3% of all HABs in teaching and nonteaching hospitals, respectively.18 A study performed on central venous catheter-related infections regarding central catheter insertion method, purpose and duration of catheterization, infection rate, and complication rate showed that 7.1% of BSIs among all central catheter insertions were catheter-related.19 Many studies showed discordance in blood culture results between centrally drawn and peripherally drawn samples.5,13,17,20,21 Beutz et al conducted a prospective cohort study in determining the sensitivity, specificity, positive predictive values (PPVs) and negative predictive values (NPVs) of blood cultures obtained through a central vascular catheter compared to peripheral venipuncture. Three hundred paired blood culture specimens were collected from 119 patients. The researchers found that 34 paired culture results (11.3%) were accepted as true-positive bacteremia. The sensitivity of catheter-drawn and peripheral venipuncture samples was 82.4% and 64.7%, respectively, and the specificity was 92.5% and 95.9%, respectively.13 Garland et al20 performed a prospective nested cohort study to examine the pathogenesis of CR-BSIs in neonates in a level III neonatal intensive care unit in a community hospital where 23 of the 82 neonates developed nosocomial infections. Fifteen of the BSIs were considered definite or probable CR-BSIs. Among the 15 cases, 10 (67%) were acquired intraluminally, 3 (20%) were acquired extraluminally, and 2 (13%) were acquired by both means. Researchers concluded that most CR-BSIs in neonates with peripherally inserted central catheters (PICCs) were caused by coagulasenegative Staphylococcus and derived from intraluminal contamination.20 Do et al7 conducted a cohort study involving 53 patients with 65 BSI episodes; the highest BSI rate was seen when the end cap was changed only weekly. The BSI rate steadily decreased as the frequency of changing the catheter caps increased from once a week to every 2 days, which suggested that contamination from the cap might be the cause of BSIs.7 The findings suggested that the mechanism for BSI might have involved colonization of the end cap with microorganisms that eventually reached the intravascular segment

VOLUME 32

NUMBER 4

JULY/AUGUST 2009

213

NAN200045_212-218

7/2/09

7:28 PM

Page 214

of the catheter. Changing the end cap more frequently might have served to reduce the load of potential pathogens that could enter the bloodstream whenever the intravenous catheter was accessed.

SETTINGS AND RESEARCH DESIGN

This study was conducted at the Drake Center, which is a specialty center for rehabilitation that consists of 110 LTAC beds and 80 skilled-care beds. The study took place in the LTAC units. The demographic data, diagnosis, prior antibiotic administration, and culture results of each subject were accessed from medical records. The data were collected over a 10-month period. The types of central catheters used for this study were PICCs, triple-lumen catheters (TLCs), and port-a-caths (PACs). Patients who had a central catheter in place and received a physicians order for central catheter blood culture were included in this study. Patients whose blood hemoglobin level was less than 6 g were excluded from this study. Each subject was screened for other signs and symptoms of septicemia, such as the evidence of a body temperature greater than 100.5F, systolic blood pressure less than 90 mm Hg, and the presence of rigor. Patients were also screened for having prior infections and antibiotic therapy. Institutional Review Board approval was obtained from the University of Cincinnati prior to this study. The registered nurses trained for the study obtained informed consent from the patient or from the durable power of attorney. Registered nurses received detailed training on how to obtain an informed consent, education of the study purpose and procedure, collection and processing of blood samples, and maintenance of subject confidentiality. All efforts were made to ensure consistency among all individuals collecting data and performing the procedure, subsequently minimizing any variance in results. Blood samples were drawn from each subject using 3 methods: (1) with the old or the existing central catheter cap (old cap), (2) with a new central catheter cap (new cap), (3) by peripheral venipuncture (peripheral). All draws were obtained on the same day with a specific procedure as follows. The 2 central catheter draws were obtained by trained registered nurses. The first central catheter draw was conducted by disinfecting the port with a chlorhexidine swab stick, and 5 mL of blood was aspirated and discarded. Using another sterile syringe, 20 mL of blood was aspirated and used for old cap culture. Then, the needleless cap was changed per protocol. Then, 5 mL of blood was aspirated and discarded. Using another sterile syringe, 20 mL of blood was aspirated and used for new cap culture. For both catheter draws, the 20 mL of blood was divided equally by injecting 10 mL of blood each

into 2 blood culture bottles (resulting in 2 old cap and 2 new cap cultures) and sent to the laboratory for processing. The peripheral blood draw was obtained by a laboratory technician within 2 hours of the above central draws. The peripheral site was disinfected with a chlorhexidine swab by scrubbing the site in an up-anddown motion and allowing it to dry for 30 seconds. The vascular puncture was performed with a sterile syringe and needle. All samples were inoculated into aerobic and anaerobic media and processed using the Bactec blood culture system (BD, Franklin Lakes, New Jersey).

STATISTICAL ANALYSIS
Numerical and categorical variables were summarized in median (range) and frequency (percentage), respectively. Nonparametric Wilcoxon rank sum tests and Fishers exact tests were used to compare medians of numerical variables and frequencies of categorical variables, respectively, between patients with and without organisms detected by the new cap method. Using the new cap method as the gold standard, the diagnostic accuracy of old cap and peripheral methods was assessed using sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). For each of the sensitivities, specificities, and predictive values, its 95% confidence interval (CI) was estimated using a bootstrap resampling method to generate a total of 2000 replicated samples. Comparisons of sensitivities and specificities between old cap and peripheral methods were carried out using McNemar tests, and comparisons of PPVs and NPVs between the methods were carried out using logistical regression models with repeated measures. Agreements between blood-drawing methods were assessed using kappa statistics. The correlating agreements for almost perfect, substantial, moderate, fair, slight, and poor in a kappa statistic are 0.81-1.00, 0.61-0.80, 0.41-0.60, 0.21-0.40, 0-0.20, and less than 0, respectively. Comparisons of agreements were assessed using standard normal tests. P values less than .05 were considered statistically significant.

RESULTS
A total of 91 patients with a median age of 61 (22-86) and male-to-female ratio of 61:30 were studied. Of these 91 patients, 83 had PICCs and 8 had TLCs (no subjects had a PAC). Of the 91 patients, the results of 54 (59.3%) were negative by all 3 methods and the results of 9 (9.9%) were positive by all 3 methods. Thirty-six (39.6%) had a positive culture result by 1 of the 3 methods, and 8 (8.8%) had a positive culture result by 2 of the 3 methods. Positive culture results were

214

Journal of Infusion Nursing

NAN200045_212-218

7/2/09

7:28 PM

Page 215

found in 18 (19.8%) patients using the new cap, the gold standard method. There were no significant differences in clinical characteristics between patients with positive and negative culture results (Table 1). Among the total 18 cases that were identified to be positive by the new cap method, the old cap method was correct in 17 cases, yielding a sensitivity (95% CI) of 94.4% (63.9%-99.2%). Among the 73 negative cases, the old cap method was correct in 54 cases, yielding a specificity (95% CI) of 74% (62.8%82.7%). On the other hand, the peripheral method had a sensitivity (95% CI) of 55.6% (33.1%-76%), lower than that of the old cap method (P .02), while at the same time, it showed a specificity of 100%, higher than that of the old cap method (P .001). The PPV (95% CI) of the old cap method was 47.2% (31.7%-63.2%), lower than that of 100% from the peripheral method (P .001). NPVs (95% CI) were 98.20% (88.2%-99.7%) and 90.10% (81.4%-95%) using the old cap method and the peripheral method, respectively, and were not significantly different (P .09) (Table 2). As shown in Table 3, the peripheral method agreed substantially and moderately to the new and old cap

methods with kappa standard errors of 0.67 0.11 and 0.5 0.09, respectively. The difference of agreements was, however, not significant (P .089). The new cap and the old cap methods agreed only fairly, with a kappa standard error of 0.26 0.08, and the level of agreement was much lower than the previous 2 agreements (P .001) (Table 3). Among the 18 true positives, as identified by the new cap method, a total of 5 different types of organisms were found: (1) gram positive (n 8), (2) gram negative (n 8), (3) Klebsiella (n 3), (4) diptheriods (n 2), and (5) methicillin-resistant Staphylococcus aureus (n 1) (Table 4). Among the 17 true-positive cases identified using the old cap method, the findings of organisms matched those of the new cap method. Among the 19 false-positive cases identified using the old cap method, 68.4% (or 13/19) of the cases were gram positive. There was 1 case of Pseudomonas identified using the old cap method that was not found in the new cap method. The peripheral method agreed with the new cap method on most of organisms among the true-positive cases, but in 1 case it missed the grampositive organism. Among the 8 false-negative cases, the organisms failed to identify by the new cap were

TABLE 1

Summary of Demographics and Clinical Characteristicsa

Variables
Demographic Age, y Male Clinical characteristics Prior infection (yes) Temperature 100.5 F 90 mm Hg 98.9% (90) 95.6% (87) 8.8% (8) 1.1% (1) 79.1% (72) 8.8% (8) 9.1 (6.4-15) 11.4 (3.9-32.1) 100.0% (18) 100.0% (18) 11.1% (2) 0.0% (0) 66.7% (12) 5.6% (1) 9.2 (6.6-10.9) 8.5 (4.2-26.3) 98.6% (72) 94.5% (69) 8.2% (6) 1.4% (1) 82.2% (60) 9.6% (7) 9.1 (6.4-15) 11.5 (3.9-32.1) NS NS NS NS NS NS NS NS 61 (22-86) 67.0% (61) 68 (47-86) 72.2% (13) 60 (22-80) 65.8% (48) NS NS

(N

All 91)

New cap ( ) (n 18)

New cap ( ) (n 73)

P valueb

Systolic blood pressure Rigor (yes)

Received prior antibiotic (yes) Type of line (triple-lumen catheter) Hemoglobin, g White blood cell

Abbreviations: NS, not significant between T and T groups with a P .05; , negative; , positive. a Values in cells are median (range) if numerical variables and frequency in percentage (count) if dichotomous variables, respectively. b P values are from Wilcoxon rank sum tests if numerical variables and Fishers exact tests if dichotomous variables, respectively.

VOLUME 32

NUMBER 4

JULY/AUGUST 2009

215

NAN200045_212-218

7/2/09

7:28 PM

Page 216

Summary of Diagnostic Accuracya

New cap ( ) (n Blood draw methods
Old cap Peripheral

TABLE 2

18)

New cap ( ) (n True negative

54 73

73) Positive predictive value (95% CI)

47.2% (31.7%-63.2%) 100% .001

True positive
17 10

Sensitivity (95% CI)

94.4% (69.3%-99.2%) 55.6% (33.1%-76%) .02

Specificity (95% CI)

74% (62.8%-82.7%) 100.00% .001

Negative predictive value (95% CI)

98.2% (88.2%-99.7%) 90.1% (81.4%-95%) .09

P valueb

Abbreviations: CI, confidence interval; , negative; , positive. a Old cap method resulted in 36 positives and 55 negatives. Peripheral method resulted in 10 positives and 81 negatives. b P values are from McNemar's tests to compare sensitivities and specificities and from logistic models with repeated measure to compare positive predictive values and negative predictive values between blood drawing methods using new cap as gold standard.

(a) gram positive (n 3), (b) gram negative (n (c) Klebsiella (n 2), and (d) diptheriods (n 1).

3),

DISCUSSION
This study compared 3 methods of blood draws in diagnosing bacteremia among 91 subjects in an LTAC facility. Agreement between the methods ranged from kappa 0.26 (old cap and peripheral) to 0.67 (new cap and peripheral). Using the new cap as a gold standard, predictive values were best for the peripheral method (PPV 100% and NPV 90.1%). In this study, 54 (59.3%) patients had all 3 negative cultures and 9 (10%) had all 3 positive cultures. Thirty-

TABLE 3

Summary of Kappa Agreement Between Blood-Drawing Methods a

Blood draw methods
Old cap Peripheral
a

New cap
0.50 0.67 0.09 0.11

Old cap

0.26

0.08

Values in cells are mean standard error of kappa statistics.

six (40%) patients had a positive result, either from a central or a peripheral sample. Among the 36 positive cases, 17 cases were true positives while the other 19 cases were false positives identified using the old cap method. The high percentage (53%) of false-positive cases suggests that the old caps could have been contaminated, putting those patients at higher risk for immediate infection, causing them to be treated unnecessarily if infections were unlikely, and resulting in unnecessary discontinuation of the central catheter. Therefore, it would be a better practice to change the cap before drawing culture samples from a central catheter. The most concerning results were the 53% (19/36) of false positives identified using the old cap method. However, among the 54 negative cases identified using the old cap method, 98.2% were correctly identified free of bacteremia (NPV, 98.2%). The old cap method did not provide accurate blood culture results in comparison with the other methods of blood collection used in this study. This might have been due to a contaminated old cap and supports the need to change the cap prior to sample collection in ruling out bacteremia. Among the 9 patients who had positive culture results by all 3 methods, organisms identified from the different cultures were the same. Further, among the 8 patients who were positive by 2 methods, organisms identified from the different cultures were the same. This supports that the sample collections were performed without contaminations. Therefore, it is expected that the sample collections were also done appropriately with the other patients including the false positives. Once a positive result is obtained from either central catheter or peripheral venipuncture, the patient is considered as having bacteremia, and treatments are initiated. The culture results of old cap and new cap methods were

216

Journal of Infusion Nursing

NAN200045_212-218

7/2/09

7:28 PM

Page 217

TABLE 4

Summary of Type of Organisms Found in Positive Cases

New cap Type of organisms
Pseudomonas Klebsiella
Gram negative Gram positive

Old cap TP (n 36) FP (n

1 0 4 13 2 0 0

Peripheral FN (n
0 0 1 0 0 0 0

TP (n

18)

19)

1)

TP (n

10)

FP (n
0 0 0 0 0 0 0

0)

FN (n = 8)
0 2 3 3 0 0 1

0 (0%) 3 (17%) 8 (44%) 8 (44%) 0 (0%) 1 (6%) 2 (11%)

1 (3%) 3 (8%) 11 (31%) 21 (58%) 2 (6%) 1 (3%) 2 (6%)

0 (0%) 1 (10%) 5 (50%) 4 (40%) 0 (0%) 1 (10%) 1 (10%)

Enterococcus species
MRSA Diptheriods

Abbreviations: FN, false negative; FP, false positive; MRSA, methicillin-resistant Staphylococcus aureus; TP, true positive.

blinded so that the physicians were not able to identify false-positive cases. Blinding the results of old cap and new cap methods enabled us to observe how physicians treated the positive results. As a result, all 36 patients, including the 19 false-positive cases, received systemic administration of antibiotics for 10 to 14 days, which increased the hospital cost. Central catheters were removed in all 36 patients, and all 36 patients received another central catheter placement because of the extended use of systemic administration of antibiotics for 10 to 14 days. Patients central catheters are often accessed numerous times each day, which increases the patient risk for having a contaminated needleless cap and thereby acquiring CR-BSIs.7 In this study, the centrally drawn blood cultures showed discordance between the old cap method and the new cap method, which could be due to a contaminated cap. Therefore, changing the needleless caps before drawing blood samples would decrease this discordance and provide more reliable results in ruling out bacteremia. Further, it prevents the introduction of any bacteria into the bloodstream from the old cap. Therefore, changing the needleless cap prior to blood sample collection for cultures would be a better practice to improve patient outcomes and reduce healthcare costs.

teremia is a stressor for the client because all positive results are treated and taken seriously. When a positive result was obtained from a central catheter sample, that catheter was discontinued and replaced with another central catheter, and the patient started on systemic administration of antibiotics for at least 10 days. This treatment regimen increases hospital costs. Administration of antibiotics can cause adverse drug effects in patients, which can also cause additional costs and lengthy hospital stays. Unwanted replacement of central catheters is costly and sometimes difficult in patients with poor vascular access, causing additional trauma to the patient.

LIMITATIONS AND RECOMMENDATIONS

Limitations of this study included the fact that the study was conducted in only 1 LTAC facility. The patients in this facility had multiple medical complications and may have had previous infections that made them more susceptible to septicemia. Therefore, further research is recommended in other acute care and LTAC facilities and in different populations to obtain additional findings. A high false-positive result (53%) obtained (old cap method) in this study recommends that changing the needleless caps before drawing blood samples from a central catheter would be a better practice to rule out bacteremia. The strength of this study was that each patient served as his or her own control, which enabled the study to identify the old cap contamination instantly.
217

IMPLICATIONS
Neumans nursing theory suggests a holistic approach to prevent illness by identifying stressors and providing strategies for removing those stressors in maintaining wellness in clients.22,23 A false-positive result for bacJULY/AUGUST 2009

VOLUME 32

NUMBER 4

NAN200045_212-218

7/2/09

7:28 PM

Page 218

CONCLUSION
Intravenous therapy is among the most widely used invasive procedures in all healthcare settings. The use of needleless caps contributes to providing a safer workplace from needlestick injuries. However, the use of needleless caps raises the concern of developing BSIs. This study showed discordance in blood culture results obtained from central catheter-drawn and peripheral venipuncture. Furthermore, this study showed a difference in culture results with old cap and new cap methods when samples were collected from the central catheter, which resulted in 19 false-positives. This could have been due to a contaminated old cap. In all 19 cases, the central catheters were replaced and all patients were given systemic antibiotics for 10 to 14 days. Unneeded replacing of the central catheters caused added trauma to the patient, and the administration of additional medications increased the cost of healthcare services for each patient as well as the chances of causing adverse drug effects. The evidence of having a high false-positive rate (53%) should be considered seriously. Therefore, the current study supports changing the needleless caps on central catheters before obtaining blood samples for cultures. This may result in reduced costs and more accurate diagnoses of bacteremia. Changing the needleless cap on central catheters before obtaining blood samples is a simple procedure and may benefit patients physically, emotionally, and financially. Providing accurate and timely care will increase patient satisfaction and help improve overall health outcomes for all patients requiring central catheter access and infusion therapy care.
REFERENCES
1. Lorente L, Henry C, Martin MM, Jimnez A, Mora M. Central venous catheter-related infection in a prospective and observational study of 2,595 catheters. J Crit Care. 2005;9(6):631635. 2. Mermel AL, Faar MB, Radd I, et al. Guidelines for the management of intravascular catheter-related infections. Clin Infect Dis. 2001;32:1249-1272. 3. Hugonnet S, Sax H, Eggimann P, Chevrolet J-C, Pittet D. Nosocomial bloodstream infection and clinical sepsis. Emerg Infect Dis. 2004;10(1):76-81. 4. Murphy J. (September 22, 2006). All intravascular devices pose risk of bloodstream infection to patients, study finds. Available at: http://findarticles.com/p/articles/mi_hb5554/is_200609/ai_ n21881405/?tag=content:coll. 5. DesJardin JA, Falagas ME, Ruthazer R, et al. Clinical utility of blood cultures drawn from indwelling central venous catheters in 6.

7.

8. 9.

10.

11. 12.

13.

14.

15. 16. 17.

18.

19.

20.

21.

22. 23.

hospitalized patients with cancer. Ann Intern Med. 1999;131(9): 641-647. Hanchett M. (November 1, 2005). Needleless connectors and bacteremia: is there a relationship? Available at: http://www. infectioncontroltoday.com/articles/410/410_5b1feat2.html Do AN, Ray BJ, Banerjee SN, et al. Bloodstream infection associated with needleless device use and the importance of infectioncontrol practices in the home health care setting. J Infect Dis. 1999;179(2):442-448. Hadaway LC. Best-practice interventions: keeping central line infections at bay. Nursing. 2006;36(4):58-63. Posa PJ, Harrison D, Vollman KM. Elimination of central lineassociated bloodstream infections: application of the evidence. AACN Adv Crit Care. 2006;17(4):446-454. Maki DG, Stolz SM, Wheeler S, Mermel LA. Prevention of central venous catheter-related bloodstream infection by use of an antiseptic-impregnated catheter. Ann Intern Med. 1997;127(4):257-266. Wenzel PR, Edmond BM. The impact of hospital-acquired bloodstream infections. Emerg Infect Dis. 2001;7(2):174-177. Bates DW, Goldman L, Lee TH. Contaminated blood cultures and resource utilization: the true consequences of false-positive results. JAMA. 1991;265:365-369. Beutz M, Sherman G, Mayfield J, Fraser V, Kollef M. Clinical utility of blood cultures drawn from central vein catheters and peripheral venipuncture in critically ill medical patients. Chest. 2003;123:854-861. Menyhay ZS, Maki DG. Disinfection of needleless catheter connectors and access ports with alcohol may not prevent microbial entry: the promise of a novel antiseptic-barrier cap. Infect Control Hosp Epidemiol. 2006;23:23-27. Aronson MD, Bor DH. Blood cultures. Ann Intern Med. 1987;106:246-253. Weinstein MD. Current blood culture methods and systems. Clin Infect Dis. 1996;23(1):40-46. Martinez JA, DesJardin JA, Aronoff M, et al. Clinical utility of blood cultures drawn from central venous or arterial catheters in critically ill surgical patients. J Crit Care Med. 2002;30(1):7-13. Coello R, Charlett A, Ward V, et al. Device-related sources of bacteremia in English hospitals: opportunities for the prevention of hospital-acquired bacteremia. J Hosp Infect. 2003;53(1):46-57. Nagashima G, Kikuchi T, Tsuyuzaki H, et al. To reduce catheterrelated bloodstream infections: is the subclavian route better than the jugular route for central venous catheterization? J Infect Chemother. 2006;12(6):363-365. Garland JS, Alex CP, Sevallius JM, et al. Cohort study of the pathogenesis and molecular epidemiology of catheter-related bloodstream infections in neonates with peripherally inserted central venous catheters. Infect Control Hosp Epidemiol. 2008; 29(3):243-249. Shah SS, Kagen J, Lautenbach E, et al. Bloodstream infections after median sternotomy at a childrens hospital. J Thorac Cardiovasc Surg. 2007;133(2):435-440. McEwen M, Wills EM. Theoretical Basis for Nursing. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. Neuman B, Newman LD, Holder P. Leadership-scholarship integration: using the Neuman systems model for 21st-century professional nursing practice. Nurs Sci Q. 2000;13:60-63.

218

Journal of Infusion Nursing