| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Corresponding author: Carol Hatler, RN, PhD, Director, Nursing Research, 350 W. Thomas Road, Phoenix, AZ 85013 (e-mail: carol.hatler{at}chw.edu).
 |
Abstract |
|---|
 Top Abstract MethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
Objectives Few studies that directly address care of central venous catheters for children in intensive care units have been reported. This evaluation was designed to describe the extent of evidence-based practices for care of insertion sites of central venous catheters in the pediatric intensive care unit of an urban tertiary care center. Another goal was to determine the influence of 2 different regimens for dressing changes on rates of catheter-related bloodstream infections and costs.
Methods A convenience sample and an exploratory design were used to collect data in 2 phases, including 30 days to establish baseline information and 30 days each during which patients received dressing care for a central venous catheter with a transparent dressing alone and with a transparent dressing plus a chlorhexidine-impregnated dressing. Nurses also participated in a survey of knowledge about infection control practices related to central catheters.
Results Few differences were found between the transparent dressing alone and a chlorhexidine-impregnated dressing plus the transparent dressing. A serendipitous finding was the number of times that central catheters were accessed daily.
Conclusions The results of this project suggest that infection control efforts may be most appropriately focused on processes rather than on products.
Notice to CE enrollees:A closed-book, multiple-choice examination following this article tests your understanding of the following objectives:
|
Although few reported studies directly address CVC care for PICU patients, studies of acutely ill adults have yielded findings that may apply. Among the effective strategies identified is use of chlorhexidine skin preparation solution.5,6 Materials and processes used in dressing changes also have been implicated in CR-BSI, with studies done to compare gauze versus transparent dressing,7 transparent hydro-colloid versus polyurethane dressing,8 and povidone-iodine ointment, gauze, and transparent dressing versus transparent dressing alone.9 In one of the few studies of pediatric patients, Garland et al10 conducted a multicenter, randomized trial of a chlorhexidine-impregnated dressing (Biopatch, Johnson & Johnson, Somerville, NJ) for neonates who needed a CVC for at least 48 hours. The investigators determined that use of the chlorhexidine-impregnated dressing, replaced weekly, was equal in effect to disinfection with povidone-iodine solution and replacement of a transparent dressing every 3 to 7 days. Although support exists for some prevention strategies, more information is needed.
Because available studies used various approaches and evaluated a variety of dressing types, little consistent information exists to guide nursing care. This inconsistency predisposes to errors and increases costs because of the need to maintain equipment and supplies for a variety of methods of site care. The project described in this article was designed to (1) document the extent of evidence-based practices for CVC site care in the PICU of an urban tertiary care center and (2) determine the influence of 2 different dressing change regimens on CR-BSI rates and costs. By developing and following guidelines for CVC site care, we intended to reduce the number of CR-BSIs and thus reduce hospital costs.
|
Methods |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
| Catheter-related bloodstream infections result in 90 000 deaths per year.
|
Setting
The site for this project was a PICU at St Joseph’s Hospital and Medical Center in Phoenix, Arizona. The 24-bed level I PICU provides critical care services for children from newborn through 21 years old. With an emphasis on the special needs of children and their families, services include preoperative and postoperative care for neurosurgery, neurology, trauma, orthopedics, ear-nose-throat, craniofacial reconstruction, metabolic disorders, respiratory conditions, and some chronic care. Serving as a tertiary referral center for the state, a mean of 1800 patients are admitted annually. Pediatric specialty physicians and nurses work side by side with child life specialists, respiratory therapists, dietitians, pharmacists, chaplains, and social workers to provide compassionate care. The 24-bed unit occupies 17 000 square feet of space. In August 2007, services were expanded to include a new 24-bed cardiothoracic PICU, thus doubling the capacity and size of the pediatric intensive care facility.
| Little consistent information exists to guide central venous catheter care.
|
Care in the PICU is provided by 73 registered nurses who are assisted by an unlicensed caregiver and a health unit coordinator each 12-hour shift. PICU staff members have a mean of 6.84 (SD, 7.88) years of experience. Of these staff members, 40% have 2 years or fewer of experience as registered nurses. A primary care nursing structure is used, and the philosophy of care includes a commitment to continuity of care. The nurse to patient ratio depends on the acuity level of the patient, with critically ill patients requiring 1-to-1 care, but less severely ill patients receiving care in a 2-to-1 ratio. Registered nurses who provided care on a regular basis (
1 shift per week for 3 months or more) were eligible to take part in the study.
| Infection rates compared favorably to the 6 months prior to data collection.
|
Current infection control practices associated with CVC placement include maximal barrier precautions during catheter placement, with most central catheters containing some sort of antibiotic/anti-infective coating. Nursing care when accessing central catheters consists of hand washing and cleaning the access port with an alcohol-based wipe. Central catheters (single or multiple lumen) not in continuous use are flushed twice a day with saline, followed by a flush with 10 U/mL heparin. Dressings for central catheters are changed routinely every 7 days or more often if the dressing is loose or soiled. Nurses use a sterile dressing change kit that contains chlorhexidine preparation sticks for skin cleansing and face masks to prevent airborne contamination. Skin cleansing consists of a 30-second scrub, followed by application of a transparent dressing. In many instances, nurses secure intravenous tubing to decrease tugging at the entry site.
Data Collection
After approval was obtained from the institutional review board, the investigation was focused on gathering information from nurses and patients in 2 study phases. Data were collected for an 11-month period, beginning in June 2007 and ending in April 2008. The study phases included the baseline phase and the dressing-focused phase. The dressing-focused phase encompassed 30 days in which a transparent dressing alone was used for CVC dressings and 30 days in which catheter dressing changes involved a Biopatch plus a transparent dressing. Two data collection instruments were used, including a survey of knowledge for nurses and a data collection form to capture patients’ characteristics.
Nurse Knowledge Survey
To determine nurses’ knowledge of and beliefs about catheter-related infection control practices, staff nurses were asked to complete a self-administered survey before patient data collection began and again after baseline data collection ended. The survey was adapted from the literature11; however, no validity or reliability has been reported for the instrument. The original survey contained 18 items formatted in 2 parts, with 13 items arranged with Likert-type responses (always, often, sometimes, never) and a second section with multiple-choice options. An additional item related to hub care was added to the instrument for the second administration after an ongoing review of the literature indicated hub access as an important component of prevention of CVC infection. Sample items include the following:
Item responses were assigned numeric values, depending on the degree of accuracy of the response selected, with values ranging from 0 to 100 and higher scores indicating a greater degree of knowledge.
Staff members were informed of the survey in staff meetings and voluntary participation was emphasized. Thirty nurses returned surveys the first time it was administered, and 32 responses were obtained at the second administration, yielding a 41% to 43% response rate. After the knowledge instrument was administered the first time, infection control information was provided in verbal, written, and poster form during a 3-week period. Because we wanted to ensure anonymity of responses, we did not request identifying information and, as a result, could not match survey responses at the 2 administration times. However, 14 of the 32 nurses who answered the survey the second time indicated that they had also completed the instrument at the start of the study. The mean knowledge score when the survey was first administered was 86.67%, and the mean score for the second administration was 84.65%.
Patients’ Characteristics at Baseline
Patient-related factors were drawn from the literature and included admission diagnosis, acuity of illness, and comorbid/preexisting conditions.12,13 Additional information noted involved catheter site, number of days catheter was used, number of catheters and lumens used, and specific types of therapies delivered via CVC.14 Information about insertion and maintenance procedures was also gathered, as were data on PICU length of stay. For this project, we included CVCs placed in the subclavian, internal jugular, and femoral veins, but we did not include arterial catheters.
Because the literature indicated that patients’ severity of illness influences susceptibility to infection, we wanted to gather a nursing acuity score for study patients so we could control for this variable. However, the literature on nursing acuity in children is sparse. For our study, a nursing acuity tool was designed from observations of experts and reflected activities most frequently occurring in the PICU. The Pediatric Nursing Acuity Tool included physiological events and care activities such as procedures, medication administration, and patient/family teaching or support needs. Items were assigned a point value depending on the urgency or number of events that are present in a 12-hour shift, with activities that occurred more frequently or that required more frequent intervention receiving more points. Point values ranged from 0 to 170, and point levels are associated with anticipated staffing needs. Scores were categorized as high, medium, and low. Each patient’s score on the Pediatric Nursing Acuity Tool was determined by data collectors at the time of study enrollment, thus yielding an assessment of severity of illness.
In order to ensure consistent acquisition of data, before the observation period began, 3 patients with a CVC were identified and informed consent was obtained from them. Each data collector evaluated the patients within a 24-hour period and the evaluations were compared for consistency. Reliability among data collectors was acceptable, with 86% agreement for baseline data collection and 83% for the comparison data collection phase.
A total of 30 patients were enrolled in the baseline phase of this project. Patients were enrolled if they required a CVC for 48 hours or longer, weighed more than 1500 g, had a gestational age greater than 28 weeks, were free from skin conditions that would interfere with dressing adherence, and had no documented sensitivity to the anti-infective agent (chlorhexidine). Only those patients whose parent or guardian provided informed consent were included. Patients in the baseline group had a mean age of 3.14 years (SD, 4.66; range, 5 days to 15 years) with 73% requiring high-acuity care (Table 1
). The most common diagnosis for the baseline patients (85%) was congenitial heart abnormality. The mean length of stay in the PICU was 10.24 (SD, 8.01) days. Central catheters were used for a mean of 7.28 (SD, 4.9) days. The location of central venous catheters varied: 33% were in the subclavian vein, 37% were in the internal jugular vein, and 27% were in the femoral vein (3% not recorded). The use of the CVCs also varied, with 97% being used to give sedatives or paralytic medications, 60% being used for administration of total parenteral nutrition, and 20% being used to obtain blood samples.
|
| Number of days with elevated white blood cell counts was lower in those using the transparent dressing alone.
|
Patients’ Characteristics During the Site Care Comparison Phase
We enrolled a total of 69 patients in this study; 30 patients were enrolled in the baseline portion (phase 1) of data collection and 39 were enrolled in the comparison phase (phase 2) of the study. The 39 patients involved with phase 2 included 18 patients who received a transparent dressing alone and 21 patients who received a Biopatch plus a transparent dressing. Table 2 summarizes the demographic characteristics of the patients. Patients in phase 2 of the project were a mean of 4.4 (SD, 5.94) years old. For the total sample, 22 patients (56%) were male and 17 (44%) were female. Admitting diagnoses for patients included 28% (n = 11) with cardiac anomalies, 28% (n = 11) with respiratory disorders, 23% (n = 9) with head trauma or brain surgery, and 5% each (n = 2) with drug overdose or systemic infection. Most of the patients enrolled (51%) were below the 50th percentile on the growth chart with a mean growth percentile of 38% (SD, 31%). Most patients required high-acuity care (46%). A small percentage (18%) of the patients enrolled in the project demonstrated low acuity. This factor relates to the number of patients (70%) who were covered by the state’s Medicaid agency. Because the payor does not reimburse for step-down care, patients often remained in the PICU until their condition was stable. Length of stay for patients ranged from 3 to 90 days, with a mean length of stay of 10.72 days (SD, 14.3).
|
| Number of medications delivered within 24 hours was predictive of days with fever.
|
Most (87%) of the CVCs placed were coated with antibiotic, with locations in the internal jugular (49%), subclavian (26%), and femoral (26%) arteries. Most patients used single (46%) or double-lumen (38%) catheters, and a few (18%) had multiple central catheters. Length of catheter use ranged from 3 to 12 days with a mean use time of 6.15 (SD, 2.96) days.
|
Outcomes |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
In order to evaluate the 2 dressing types, proxy outcome variables were used, including number of days with recorded fever and number of days with elevated white blood cell (WBC) count. Study patients had a mean of 1.20 (SD, 1.59; range, 0–7) days with fever and a mean of 0.6 (SD, 1.19; range, 0–5) days with an elevated WBC count. To examine patients’ outcomes related to dressing type, we determined that the groups did not differ significantly in number of days with fever (P = .18) or in length of time the CVC was in place (P = .14). A difference was detected, however, in number of days with an elevated WBC count (P < .01); patients with a transparent dressing alone experienced a mean of 0.31 days (SD, 0.79) and patients with a Biopatch plus a transparent dressing had a mean of 0.82 days (SD, 1.4) with an elevated WBC count.
Of interest were the number of medications given per day and the number of blood samples obtained via the CVC per day. Long and colleagues16 reported that the number of catheter interruptions, whether accidental or for obtaining blood samples or administering medication, influenced infection rates for infants and toddlers, so we included these events in our monitoring. A mean of 1.17 (SD, 1.09) blood samples were obtained per day, and a mean of 16.31 (SD, 12.77; range, 1–64) intravenous medications were delivered per day. Because the number of medications delivered via the CVC was a surprising finding, we examined this factor more closely. Number of medications delivered within 24 hours was predictive of days with fever (P < .001), yet not significantly related to elevation in WBC count. This finding suggests that a greater number of catheter interruptions may predispose PICU patients to an infection developing.
|
Discussion |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
Because few investigations of CVC care in children have been published, this study adds to the body of nursing knowledge for care of that population. Pediatric patients do not simply respond as small adults, but demonstrate unique reactions to illness and treatment. Because our investigation was focused on pediatric patients, the results of this project may be useful in developing specialized care. This study also contributes to nursing knowledge by describing the number of times central catheters are accessed. We found only 1 published study that included catheter access times as a factor, yet our study demonstrated a link between catheter interruptions and days with fever. By identifying the extent of central catheter interruptions, our results may substantiate more aggressive nursing practices.
Nurses’ knowledge scores did not substantially improve over the course of the project. One explanation for this factor may relate to the small amount of overlap between those who responded the first time versus those who responded for the second administration. The results for nurses’ knowledge may reflect sampling from 2 different groups of practitioners.
Another possible explanation for these results may lie with nurses’ skepticism about the relevance and accuracy of practice guidelines based on research. For nearly 30 years, scholars and practitioners have recognized that a gap exists between generation and application of research findings. In an early study, Hunt18 suggested that research utilization is hampered because nurses are not aware of the existence of research results, do not understand them, do not believe them (emphasis added), do not know how to apply them, and/or are not allowed to use them. Two more recent studies suggest that low levels of research utilization persist in today’s health care settings. Parahoo19 surveyed 1368 practicing nurses in Ireland to establish the extent to which nurses used research findings in practice. Slightly more than 50% of the respondents reported "sometimes" using research findings in practice. Similarly, Valizadeh and Zamanzadeh20 surveyed nurses in Iranian teaching hospitals to determine attitudes about and use of nursing research in clinical practice. Although more than half (54.7%) of the respondents said they "sometimes" use research, nearly one-third (29.4%) of nurses reported that they never used research in their practice. Further, more than two-thirds agreed that "research is not relevant to the real day-to-day work in nursing." Barriers to using research in practice exist with other health care providers, including physicians. In a literature review examining barriers to using practice guidelines, Cabana and colleagues21 reported that physicians were hampered by lack of awareness, lack of agreement with recommendations, and difficulty overcoming the inertia of previous practice. The results of the knowledge portions of our project may reflect such attitudes.
|
Limitations |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
| A greater number of line interruptions may predispose PICU patients to infection.
|
A further confounding event occurred about midway during the comparison phase of the study. Marked skin irritation that alarmed all those involved in their care developed in 2 week-old patients. Study data collection was stopped until we could determine the source of the irritation. As a result of these events, we decided to limit use of Biopatch to patients more than 2 months old.
|
Implications for Research and Practice |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
| Nurses’ knowledge scores did not improve over the course of the project.
|
Implications for practice based on this examination relate to process standardization. Two strategies represent the next steps for our organization. First, we want to standardize interdisciplinary communication related to use of central catheters. Among the key elements coming from the Institute for Healthcare Improvement22 is daily review of the necessity for the catheter with prompt removal of intravenous catheters, when possible. By building this element into routine interdisciplinary communication, we may be able to reduce the number of days that CVCs remain in place and, subsequently, reduce the occurrence of bloodstream infection. In addition, we will pursue standardization of nursing practice for central catheter access. Akagi and colleagues23 suggest that entry into the intravenous tubing is the most common route of bacterial contamination. In a laboratory experiment, Kaler and Chinn24 demonstrated that at least 15 seconds of friction with alcohol-based pads successfully decontaminated a variety of intravenous access devices. Educating and monitoring nurses about practice that includes thorough hand-washing and using friction in a twisting motion (as if you were juicing an orange) to scrub the port with alcohol for 30 seconds before accessing may improve patients’ outcomes.
Accountability for one’s actions is integral to taking on the role of patient advocate. By ensuring the knowledge and skills necessary to prevent harm (eg, CR-BSI), nurses can more consistently actualize their roles in patient care and live up to the trust that patients and their families bestow on nurses.
|
ACKNOWLEDGMENTS |
|---|
This article is followed by an AJCC Patient Care Page on Page 522.
Support from St Joseph’s Foundation, Phoenix, Arizona, was instrumental in completing this project.
Now that you’ve read the article, create or contribute to an online discussion on this topic. Visit www.ajcconline.org and click "Respond to This Article" in either the full-text or PDF view of the article.
|
REFERENCES |
|---|
|
TopAbstractMethodsOutcomesDiscussionLimitationsImplications for Research and...References |
|---|
Related articles in AJCC:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
