Reduction in Central Line--Associated Bloodstream Infections Among Patients in Intensive Care Units --- Pennsylvania, April 2001--March 2005

Persons using assistive technology might not be able to fully access information in this file. For assistance, please send e-mail to: mmwrq@cdc.gov. Type 508 Accommodation and the title of the report in the subject line of e-mail.

Each year, an estimated 250,000 cases of central line--associated (i.e., central venous catheter--associated) bloodstream infections (BSIs) occur in hospitals in the United States, with an estimated attributable mortality of 12%--25% for each infection (1). The marginal cost to the health-care system is approximately $25,000 per episode (1). In 2001, CDC was invited by the Pittsburgh Regional Healthcare Initiative (PRHI)* (2) to provide technical assistance for a hospital-based intervention to prevent central line--associated BSIs among intensive care unit (ICU) patients in southwestern Pennsylvania. During a 4-year period, BSI rates among ICU patients declined 68%, from 4.31 to 1.36 per 1,000 central line days. The results suggest that a coordinated, multi-institutional infection-control initiative might be an effective approach to reducing health-care--associated infections.

In 2000, PRHI convened an advisory committee of regional infection-control experts to discuss strategies for prevention of health-care--associated infections. In April 2001, this group initiated a regional infection-control intervention with the goal of eliminating central line--associated BSIs in ICUs. The intervention was designed collaboratively and led by infection-control professionals and medical staff from the participating hospitals. Participation was voluntary. The intervention was multifaceted, consisting of five components: 1) promotion of targeted, evidence-based catheter insertion practices (i.e., use of maximum sterile barrier precautions during insertion, use of chlorhexidine for skin disinfection before catheter insertion, avoidance of the femoral insertion site, use of recommended insertion-site dressing care practices, and removal of catheters when no longer indicated) (1); 2) promotion of an educational module about central line--associated BSIs and strategies for their prevention; 3) promotion of standardized tools for recording adherence to recommended catheter insertion practices; 4) promotion of a standardized list of contents for catheter insertion kits that includes all supplies required to adhere to recommended insertion practices; and 5) measurement of central line--associated BSI rates and distribution of data to participating hospitals in confidential quarterly reports, allowing comparison of individual unit-specific rates with pooled mean rates from other participating ICUs in the region and pooled mean rates from all other U.S. hospitals participating in the National Nosocomial Infection Surveillance (NNIS) system, stratified by type of ICU.

To measure the effect of the intervention, participating hospitals prospectively collected and reported data on central line--associated BSIs, beginning in April 2001. Data were collected using standardized definitions and methods from the NNIS system, a voluntary, hospital-based reporting system established to monitor risk-adjusted health-care--associated infection rates (3). Trends in central line--associated BSI rates during April 2001--March 2005 were assessed using multivariable Poisson regression analyses that controlled for central line use.

Thirty-two hospitals in 10 southwestern Pennsylvania counties participated in the intervention, including 28 (72%) of the 39 acute care hospitals that provided intensive care services in the six-county Pittsburgh metropolitan statistical area. The median size of participating hospitals was 215 beds (range: 27--796 beds). Among the participating hospitals, 69 ICUs participated. However, three ICUs that submitted five or fewer quarters of data were excluded from the analysis. Of the 66 ICUs included in the analysis, 48% were medical/surgical, 11% cardiothoracic, 14% coronary, 9% surgical, 6% neurosurgical, 5% trauma, 3% medical, 3% burn, and 3% pediatric. The ICUs provided data for a median of 15 quarters (range: 6--16 quarters) during April 2001--March 2005.

Overall, the pooled mean rate of central line--associated BSIs per 1,000 central line days in participating ICUs decreased by 68%, from 4.31 to 1.36 (p<0.001) during April 2001--March 2005 (Figure). BSI rates among medical/surgical ICUs decreased by 67%, from 3.64 to 1.18 (p<0.001), and BSI rates among other ICU types decreased by 69%, from 4.72 to 1.47 (p<0.001). Similar decreases were observed when rates were analyzed for ICUs that reported data for all 16 study quarters.

Reported by: C Muto, MD, Univ of Pittsburgh Medical Center Presbyterian Hospital; C Herbert, West Penn Allegheny Health System, Allegheny General Hospital; E Harrison, Pittsburgh Regional Healthcare Initiative. JR Edwards, MS, T Horan, MPH, M Andrus, JA Jernigan, MD, Div of Healthcare Quality Promotion, National Center for Infectious Diseases; PK Kutty, MD, EIS Officer, CDC.

Editorial Note:

Health-care--associated infections in U.S. hospitals account for an estimated 2 million infections and 90,000 deaths annually (4). Central line--associated BSIs are the third most common health-care--associated infections (after ventilator-associated pneumonia and catheter-associated urinary tract infections) reported by medical/surgical ICUs participating in the NNIS system (5). CDC has identified catheter-associated adverse events, including BSIs, as one of its seven health-care safety challenges, with a goal to reduce such complications by 50% in 5 years (6). The 32 Pennsylvania hospitals that participated in this regional patient-safety intervention reduced BSI rates by 68% in 4 years, suggesting that coordinated infection-control initiatives among health-care facilities in a region might be an effective way to reduce catheter-associated events such as BSIs.

The prevention practices promoted during this intervention were not novel; since 1996, most have been included in the Healthcare Infection Control Practices Advisory Committee recommendations for the prevention of central line--associated BSIs (1,7). The results described in this report suggest that adhering to these evidence-based preventive practices can prevent BSIs. Nonetheless, previous reports suggest that adherence to these practices remains low (8,9).

Hospitalized patients, especially those in ICUs, are at increased risk for infection because of underlying illness, compromised immune systems, and the use of invasive devices; therefore, elimination of all health-care--associated infections is challenging. A review of 30 reports on programs to reduce nosocomial infections determined reductions of 10%--70% in the number of infections, with the greatest success among programs to reduce central line--associated BSIs (10). One study has reported nearly complete elimination of central line--associated BSIs in a surgical ICU (9). The 67% and 69% reductions observed in the regional initiative described in this report provide additional evidence that decreases in central line--associated BSI rates >50% can be achieved in hospital ICUs of varying types.

The findings in this report are subject to at least three limitations. First, participation in the initiative was voluntary, and ICUs did not report data every quarter. However, incomplete reporting did not appear to influence the results; the findings were unchanged when results for all ICUs were compared with a subset analysis that included only those units reporting data in all 16 quarters. Second, data from nonparticipating hospitals in the region were not available for comparison. Finally, data on implementation of and adherence to the promoted practices or other facility-specific interventions were not systematically reported; therefore, determining the relationship between adherence and the observed decrease in infection rate was not possible, nor was determining the relative contribution of the individual components of this intervention. However, no other infection-control interventions were observed in the participating ICUs that might have accounted for the reduction in rates.

This report describes a substantial reduction in central line--associated BSI rates after a coordinated intervention among hospitals in a region. Additional studies are needed to determine whether similar levels of success can be achieved by applying this strategy to other health-care--associated infections.

This report is based, in part, on contributions by SS Stephens, Butler Memorial Hospital; D Lauze, Canonsburg General Hospital; Children's Hospital of Pittsburgh; MJ Bellush, Excela Health Frick Hospital; R Volpe, Heritage Valley Health System, The Medical Center, Beaver; S Silvestri, Heritage Valley Health System, Sewickley Valley Hospital; S Krystofiak, MS, Mercy Hospital of Pittsburgh; K Liberatore, Monongahela Valley Hospital, Inc.; SL Jacobs, MS, St. Clair Hospital; J Shuck, Uniontown Hospital; B Hullihen, Univ of Pittsburgh Medical Center (UPMC) Bedford Memorial; CM Miller, MSN, UPMC Braddock; DC Carl, UPMC Horizon; UPMC Lee Regional; C Orbison, UPMC Magee; DM Inglot, UPMC McKeesport; S Carr, UPMC Northwest Medical Center; UPMC Passavant; UPMC Presbyterian; UPMC Shadyside; P Adomatis, UPMC South Side; SL Smith, MPM, UPMC St. Margaret; M Palfreyman, MS, The Washington Hospital; L Boody, West Penn Allegheny Health System, Alle-Kiski Medical Center, The Western Pennsylvania Hospital, The Western Pennsylvania Hospital-Forbes Regional Campus; SL Albright, Excela Health Westmoreland Regional Hospital; JA Grote, Excela Health Latrobe Area Hospital; M Dembinski, MPH, M Klevens, DDS, Div of Healthcare Quality Promotion, National Center for Infectious Diseases, CDC.

References

* A nonprofit consortium of regional health-care facilities, insurers, employers, health-care providers, corporate and civic leaders, and local health authorities.

Disclaimer All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.