Because central venous catheters provide venous access to patients for the administration of fluids/medications, blood products, blood sampling, and hemodynamic monitoring, they disrupt skin integrity, making infection with bacteria and/or fungi possible. This infection may spread to the bloodstream, causing sepsis with hemodynamic changes, organ dysfunction, and ultimately death.
About half of all U.S. intensive care unit (ICU) patients have central venous catheters, which accounts for 15 million central-venous-catheter days per year in the ICUs. There are approximately 5.3 catheter-related bloodstream infections (CR-BSIs) per 1,000 catheter days, and the mortality associated with these infections is approximately 18%-or 14,000 deaths per year. 1 In addition, CR-BSIs prolong hospitalization 7 days on average and cost between $3,700 and $29,000 per infection. 2
The Institute for Healthcare Improvement (IHI) has described care bundles, or groups of best practices with respect to a disease process, that individually improve care, but when applied together, result in better outcomes than when implemented individually. The IHI central line bundle has five components: hand hygiene; maximal barrier precautions; chlorhexidine skin antisepsis; optimal catheter site selection (with the subclavian vein as the preferred site for nontunneled catheters); and daily review of the need for a central line, with prompt removal of unnecessary lines.
In many hospitals, application of the central line bundle yields impressive reduction in central line infection. Research demonstrates that ICUs which have implemented multiple interventions similar to the central line bundle have nearly eliminated CR-BSIs. 3
Description of the initiative
In 1998, the surgical intensive care unit (SICU) staff of Johns Hopkins Hospital (JHH) set out to decrease its CR-BSI rate. The JHH department of hospital epidemiology and infection control (HEIC) reported that the CR-BSI rate for the SICU was 14.56 infections per 1,000 catheter days-a rate well above the 50th percentile of the National Nosocomial Infections Surveillance System (NNISS). In response, the SICU staff, including nurses, attending physicians, HEIC staff, pharmacists, and respiratory therapists, met in March 1999 to discuss the infection rates and possible causes, then review evidence. The multidisciplinary team overwhelmingly agreed that its first initiative was to implement total barrier precautions for central line insertion and rewires. The initial goal was to decrease CR-BSIs in the SICU.
As total barrier precautions were implemented, the staff identified problems in maintaining compliance with the new requirement. For example, the nurses and physicians found the time and complexity of gathering necessary supplies challenging. To address this, the staff designed a cart that included all of the supplies necessary for total barrier precautions. The creation of the "line cart" brought forth the first key point of this initiative: the need to decrease complexity in carrying out total barrier precautions or any safety precaution.
In October 2000 the SICU team, in collaboration with HEIC, designed a Web-based program for teaching the practices related to central line insertion to physicians and other clinicians. All providers were required to complete this program prior to their first central line insertion in the SICU. The goal of the teaching tool was to educate providers and ensure practice continuity-a second key point of the initiative.
The implementation of total barrier precautions, the creation of a line cart, and the educational program impacted the CR-BSI rate. As the percentage fell, nurses became even more passionate about preventing the occurrence of the infections and sought additional factors that could improve results.
SICU nurses implemented a pilot study to evaluate the effectiveness of occlusive dressings and identified one that proved to be the most occlusive for the greatest number of days. At the same time, a hospital-wide committee was evaluating the vascular access device (VAD) protocol. It selected the same type of dressing as the SICU nurses and incorporated it into the VAD protocol. The committee also incorporated several guideline recommendations, supported by clinical trials and systematic reviews. These included the appropriate use of hand hygiene and chlorhexidine skin preparation swabs for insertion and cleaning around the central line site.
During this period, JHH's associate professor of anesthesiology and critical care medicine and the SICU team recognized the need to improve collaboration and communication as an imperative for better patient outcomes. They targeted the dual importance of patient rounds: to educate medical students, residents, and fellows, and to medically treat patients. After surveying residents and nurses following rounds, they found that less than 10% understood the daily tasks and therapies for their patients. With input from multidisciplinary ICU team members, they developed a daily goals form in a checklist format by patient systems that identify necessary work to progress the patient.
The form communicated the care plan for the shift to all team members. It also identified the patient's greatest safety risk and family communication plan. The daily goals sheet was first implemented in JHH ICUs July 2001. After 7 weeks of use, more than 95% of nurses and residents reported they understood the daily goals/plans for their patients.
Staff speculated that the goals sheet may have contributed to decreased ICU length of stay in the pilot ICU from 2.2 days to 1.1 days. 4 All care team members review the goals/plan for each patient during the course of the day; it's located at the patient bedside for easy access by all providers. And as goals of care change, so does the form, which prompts clinicians to ask themselves daily, "Is this catheter necessary or can it be removed?"
Even though the SICU team significantly decreased infection rates through practice changes, it still hadn't achieved 100% compliance with the total barrier precautions. In January 2002 the SICU nurses developed a care team checklist to empower nurses to stop the physician at any point in the central line insertion or rewire procedures if all infection control practices weren't followed. This checklist also provided an independent redundancy in the system as an additional check.
Several other practice changes were implemented that contributed to the decrease in CR-BSI rates. HEIC revised the hospital-wide VAD policy, including successful SICU practices for insertion, dressing change, catheter care, and maintenance. This policy was communicated throughout the institution in an interdisciplinary clinical practice manual.
Additionally, the SICU eliminated the use of stopcocks and switched to a different device that provided less of an opportunity for a source of infection. Finally, the SICU implemented an in-line blood drawing system to reduce the breaks in the system for arterial and central venous lines. The SICU nursing staff contributed to all of these practice recommendations and changes.
JHH's culture of safety begins at the top leadership, through the executive safety rounds program. Each member of the administrative team has adopted a hospital unit and conducts safety rounds at least monthly, meeting with frontline staff to discuss care barriers and job needs. During the last 5 years of implementing these practice changes, the SICU team watched a steady decline in CR-BSI rates, from the original rate of 14.56 infections per 1,000 central line days in 1998 to 0.93 per 1,000 central line days in 2004. (See "SICU CR-BSI yearly results.") The JHH CR-BSI rate also compares favorably to that of the NNISS. (See "SICU yearly CR-BSI rates compared to NNISS.") The SICU team continues to sustain these results. As of press date, the SICU's last CR-BSI was identified in November 2004.
Organizational implementation
JHH managers strive to keep patient safety at the forefront of daily work and emphasize that priority to the staff. (See "Summary of key interventions.")
The SICU team takes information dissemination seriously. For example, members have spread the CR-BSI toolkit throughout all JHH ICUs and are presently introducing it into other Johns Hopkins medical institutions. This work can also be seen on the IHI Web site, in VHA programs, and statewide ICU initiatives hosted by the Johns Hopkins Quality and Safety Research Group.
REFERENCES