1464 Crit Care Med 2012 Vol. 40, No. 5

Healthcare associated infec-tions (HAIs) are among the most common adverse events occurring during hospital

admissions. In 2002, 1.7 million pa-tients suffered from HAIs with an asso-ciated 155,668 fatalities, or nearly 10%. Furthermore, infection rates were the highest in the intensive care units (ICUs) (13%) (1). The attributable cost of HAIs

varies depending on the type of infection, and some estimate the cost to be $34,670 for surgical site infections and $29,156 for central line-associated bloodstream infections (CLABSIs) (2). Likewise, length of hospital stay increases by an average of 9.6 days among patients who develop HAIs (3).

CLABSIs constitute one of the most frequent hospital-acquired infections in the United States, with an associated mortality of up to 25% (4). A recent re-port by the Centers for Disease Control and Prevention showed that in 2009 the number of CLABSIs in American inpatient settings totaled 41,000 events (5). Interventions aimed at improving compliance with optimal insertion tech-niques have been proven to decrease the CLABSI rates by as much as 70% (6). Additionally, maintenance interven-tions to reduce the bacterial load on pa-tients’ skin have been demonstrated to be effective at preventing subsequent CLABSIs (7). However, the incremental effectiveness of several interventions applied in a sequential fashion is still unknown.

We sought to determine the effective-ness of a series of stepwise interventions,

which we hypothesized would incremen-tally lower CLABSI rates. These interven-tions targeted maintenance of central venous catheters and included a 15-sec “scrub-the-hub” process with chlorhexi-dine swabs, daily chlorhexidine baths, and daily ICU goal-oriented nursing rounds. We also assessed the effect of these inter-ventions on the type of causative organ-isms, the number of days from catheter insertion to first positive blood culture, and the rate of secondary bacteremia dur-ing the various phases of the study.

METHODS

This quasi-experimental study was per-formed at Jackson Memorial Hospital, a 1500-bed safety net teaching hospital affiliated with the University of Miami Miller School of Medicine. This facility has approximately 300 ICU beds (adult, pediat-rics, and neonatology) and serves a mixed patient population with a large, under-served and foreign national representation. This project was submitted to the Insti-tutional Review Board, which determined that it did not constitute human subject research.

The areas selected for this project con-sisted of adult ICUs with known initiation dates for each intervention. Interventions

Objective: Determine the impact of three stepwise interventions on the rate of central catheter-associated bloodstream infections.

Design: Quasi-experimental study.Setting: Three surgical intensive care units (general surgery,

trauma, and neurosurgery) at a 1500-bed county teaching hospital in the Miami metro area.

Patients: All consecutive central catheter-associated blood-stream infection cases as determined by the Infection Control Department.

Interventions: Three interventions aimed at catheter mainte-nance were implemented at different times in the units: chlorhexi-dine “scrub-the-hub,” chlorhexidine daily baths, and daily nursing rounds aimed at assuring compliance with an intensive care unit goal-oriented checklist.

Measurements and Main Results: The primary outcome was the monthly intensive care unit rate of central catheter-associated

bloodstream infections (infections per 1000 central catheter days). Over 33 months of follow-up (July 2008 to March 2011), we found decreased rates in each of the three intensive care units evaluated during the interventions, especially after implementation of chlo-rhexidine daily baths. rates in unit A decreased from a rate of 8.6 to 0.5, unit B from 6.9 to 1.6, and unit C from 7.8 to 0.6. secondary bloodstream infection rates remained unchanged throughout the observation period in units A and B; however, unit C had a decrease in its rates over time.

Conclusions: We report the progressive reduction of central catheter-associated bloodstream infection rates after the stepwise implementation of chlorhexidine “scrub-the-hub” and daily baths in surgical intensive care units, suggesting effectiveness of these interventions. (Crit Care Med 2012; 40: 1464–1469)

Key Words: central catheter-associated bloodstream infections; chlorhexidine; intensive care units

effectiveness of stepwise interventions targeted to decrease central catheter-associated bloodstream infections*

L. Silvia Munoz-Price, MD; Cameron Dezfulian, MD; Mary Wyckoff, PhD, ACNP, BC, CCNS, FNP-BC, NNP, FAANP; Joshua D. Lenchus, DO; Mara Rosalsky, RN; David J. Birnbach, MD, MPH; Kristopher L. Arheart, EdD

*see also p. 1659.From the Departments of Medicine (LSMP,

JDL),Public Health and Epidemiology (LSMP, DJB), Surgery (MW), and Anesthesiology (JDL, DJB), Division of Biostatistics (KLA), University of Miami Miller School of Medicine, Jackson Memorial Hospital (LSMP, MR), Miami, FL; and the Department of Critical Care Medicine (CD), University of Pittsburgh, Pittsburgh, PA.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (http://journals.lww.com/ccmjournal).

The authors have not disclosed any potential con-flicts of interest.

For information regarding this article, E-mail: smunozprice@med.miami.edu

Copyright © 2012 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e31823e9f5b

Crit Care Med 2012 Vol. 40, No. 5 1465

had to be fully implemented in the unit within a 4-wk timeframe to satisfy inclusion crite-ria. This requisite specifically excluded those units in which more gradual implementation occurred.

Unit A was the surgical ICU consisting of two adjacent 20-bed units totaling 40 beds. These two units shared nursing and ancillary staff, and were managed by the same nursing and medical administration. Unit A accepted patients that required in-tensive care management after surgeries, such as solid organ transplants, cardiotho-racic, or intra-abdominal procedures. The bed configuration was a mixture of single-bed rooms and three to four patient open pods.

Unit B was a trauma ICU consisting of a 20-bed unit with an open floor plan, with only two single-bed rooms. Most patients admitted to this unit suffered from trauma or burn injuries. Unit C (neurosurgical ICU) was a 40-bed unit with single rooms. Patients admitted to this unit consisted of traumatic and nontraumatic neurosurgical patients as well as non-neurosurgical patients, such as those who had suffered a stroke. These two units had separate nursing and medical leadership.

CLABSI. The Infection Control Depart-ment has had an active prospective CLABSI surveillance process for the past decade, which begins with the daily report of all posi-tive blood cultures within each adult ICU. The infection preventionist assigned to the particular unit then screens each positive culture using a paper-screening tool based on the National Healthcare Safety Network’s definitions (8). These definitions include a positive blood culture with a known patho-gen without an obvious source of bacteremia or the presence of an organism considered to be a skin contaminant (i.e., coagulase-negative staphylococci) in two separate blood cultures within a 72-hr period with signs or symptoms of infection (i.e., fever, hypoten-sion). Positive catheter-tip cultures are not part of these definitions. All paper records containing CLABSIs and bacteremias are stored in the Infection Control Department for a minimum of 5 yrs. For the purpose of this study and to standardize the observa-tions across time, units, and infection con-trol practitioners, a single reviewer (LSMP) performed a retrospective evaluation of all archived paper-screening tools within the Infection Control Department; the National Healthcare Safety Network’s definitions (8) were also used for this retrospective evalua-tion. Corrections in the classification of blood cultures (CLABSIs vs. secondary bacteremia) were made if necessary.

Secondary bacteremias were used as con-trols. Monthly central catheter days and ven-tilator days were obtained from a preexisting Infection Control electronic database (updated monthly by each infection preventionist). Furthermore, monthly case-mix indexes (CMIs)

for individual units were provided by the hos-pital’s Quality and Patient Safety Division.

Time Courses

Time courses were divided in each of the three ICUs into contiguous segments based on the time of initiation of the inter-ventions. Due to its quasi-experimental na-ture, times of initiation varied in each of the ICUs, and not all ICUs implemented all three interventions. Subsequent phases always incorporated all interventions previously applied.

Baseline (Phase 1)

This phase was comprised of all the initial months of observation before the implemen-tation of “scrub-the-hub.” No new interven-tions targeting CLABSIs were started during this phase. During the baseline phase, the following interventions were already in place and standard across the institution: the use of chlorhexidine-impregnated dressings (Biopatch, Ethicon), chlorhexidine-impreg-nated (ARROWg+ard, Teleflex; 90% of all catheters used) or minocycline/rifampin-impregnated (Cook Spectrum, Cook Medical) central venous catheters, chlorhexidine skin disinfection at the time of catheter inser-tion, and Micro-CLAVE neutral intravenous connectors (Hospira). Additionally, an in-sertion checklist was available in the nurs-ing computerized data entry, but adherence was not quantified during this study. This checklist included full sterile barrier pre-cautions and hand hygiene before catheter insertion, among other infection prevention items.

“Scrub-the-Hub” (Phase 2)

This intervention consisted of a 15-sec scrub of the intravenous ports using a single Chloraprep swabstick (2% w/v chlorhexidine gluconate and 70% v/v isopropyl alcohol, CareFusion, TX) before any intravenous ac-cess. Unit A’s nursing staff came up with the idea and implementation of scrub-the-hub in the adult ICUs. The roll out of this intervention included an educational campaign initially tar-geting nurse educators and selected nursing staff (“train-the-trainers”) followed by educa-tion of all staff nurses. Additionally, education-al emails regarding technique and rationale for “scrub-the-hub” were distributed among the staff. The supply manager removed all alcohol wipes from storage supplies, omnicells, patient rooms, and nursing stations at the initiation of this intervention and replaced them with packets of single-use chlorhexidine swabs. Thereafter, any procedure previously requir-ing alcohol wipes, such as insulin injections, were performed using chlorhexidine swabs. This intervention was implemented in all three ICUs. Exact months of implementation were

confirmed using purchasing records from the Purchasing Department (Fig. 1).

Daily Chlorhexidine Baths (Phase 3)

Daily body baths were performed using impregnated 2% chlorhexidine cloths (Sage, Cary, IL) rather than the traditional soap-and-water bath. This intervention was ini-tially implemented in unit A in August 2009, and was then launched across all other adult ICUs between the months of September and November 2009. The Purchasing Department was able to provide hospital-wide utilization records of chlorhexidine cloths. The specific months of implementation in units B and C were obtained from electronic communica-tions detailing education, observations, and problems encountered by Infection Control in each of the participating units. As report-ed elsewhere (9), chlorhexidine cloths were used to scrub the body from the neckline to the toes. Once applied, chlorhexidine was not rinsed off. All products that were po-tentially noncompatible with chlorhexidine, including soap used previously for patients’ baths, were removed from the nursing sta-tions and storage units. Furthermore, during the first few months after implementation, adequate application technique was directly observed by infection preventionists during night shifts (when most patient baths took place).

Daily Nursing Rounds (Phase 4)

The Associate Nurse Manager or charge nurses conducted daily rounds with bed-side nurses for each patient. These rounds covered items from an ICU goal-oriented checklist, which incorporated several infec-tion control measures tailored to decrease hospital-acquired infections. Items targeting CLABSI reduction included the catheter lo-cation (limiting the use of femoral sites) and assessment of the necessity for continued central venous catheterization. The Associate Nurse Manager or charge nurses ensured that any goal not being met was brought to the attention of the attending physician or immediately corrected. Additionally, nurs-ing hand-offs were modified to include all the items of the checklist. This intervention was only applied in unit A.

Statistics. CLABSI rates were calculated on a monthly basis (cases per month/cen-tral catheter days) and expressed as num-ber of infections/1000 central catheter days. Data were entered on an Excel spreadsheet (Microsoft, Seattle, WA) and included unit, date of admission to the hospital and unit, date of central venous catheter insertion, date of first positive blood culture, and type of organism identified. Monthly CMI, a standardized cal-culation based on the monthly average weight (as per the Centers of Medicare and Medicaid

1466 Crit Care Med 2012 Vol. 40, No. 5

Services) of all diagnosis-related groups with-in each unit, were provided by the hospital ad-ministration (Division of Quality and Patient Safety).

The number of monthly CLABSI infection cases was analyzed separately for each unit. SAS PROC GENMOD was used to perform a general-ized linear model for an outcome variable with a Poisson distribution. The model included the number of cases as the dependent variable. Independent variables were phase, month nest-ed within phase, and CMI nested within phase. A natural log link function was used, and the natu-ral log of ventilator days was used as an offset to control for variations in observation times. A de-viance scale parameter was also used to control for overdispersion. The months were centered within each phase. CLABSI rates and 95% con-fidence intervals for each phase were estimated from the model parameters and transformed by an inverse link. The CLABSI rates were adjusted for the effect of monthly CMIs. Contrasts were also performed to compare the CLABSI rates be-tween adjacent phases (e.g., phase 1 vs. 2).

A generalized linear model similar to the one used to analyze CLABSI rates was used to explore secondary bacteremias. The outcome variable was secondary infections with inde-pendent variables of month (centered) and CMIs. A natural log link was used, and an offset of ventilator days was used. A deviance scale parameter was also used to control for overdis-persion. Adjusted rates for CMIs were comput-ed as described above and shown in Figure 1.

Time-to-positive blood cultures were evalu-ated using SAS PROC LIFETEST to perform a Kaplan-Meier analysis. The data were combined from all units, and the analysis was stratified by phase. Log-rank statistics were used to deter-mine significant differences in times. p values .05 were defined as statistically significant. All analyses were performed using SAS, ver-sion 9.3 (SAS Institute, Cary, NC).

RESULTS

While CLABSI information from units A and B were available from July 2008

onward, information from unit C was available beginning in September 2008. During the 33 months of follow-up, a to-tal of 186 CLABSIs were detected (unit A: 98, unit B: 47, unit C: 41) among 42,430 central catheter days (4.38/1,000 cen-tral catheter days). We noted the follow-ing interventions: four phases in unit A (1 through 4), three phases in unit B (1 through 3), and two phases in unit C (1 and 3). CLABSI rates based on unit and phase are shown in Table 1. All CLABSI rates showed a statistically significant decrease from one phase to the next, except for unit B’s phases one to two (Fig. 1 and Table 1). Monthly CMIs were used to control for the effect of the interventions on the CLABSI rates. The median monthly cen-tral catheter days and their interquartile ranges in units A, B, and C were 696 (624- 736), 335 (290-372), and 317 (291-370), respectively.

figure 1. Central catheter-associated bloodstream infection rates. Rates are expressed as number of central catheter bloodstream infections per 1000 central catheter days. Unit A, surgical intensive care unit; Unit B, trauma intensive care unit; Unit C, neurosurgical intensive care unit. Solid line, observed rates; circle-solid line, adjusted rates; dotted line, chlorhexidine swabs monthly utilization (×1000). Phase 1, baseline; Phase 2, chlorhexidine “scrub-the-hub”; Phase 3, chlorhexidine “scrub-the-hub” plus daily chlorhexidine baths; Phase 4, chlorhexidine “scrub-the-hub” plus daily chlorhexidine baths plus daily intensive care unit goal sheet nursing rounds.

Crit Care Med 2012 Vol. 40, No. 5 1467

Secondary bloodstream infections were also determined (Supplemental Digital Content 1, http://links.lww.com/CCM/A393). A total of 119 secondary bac-teremias were identified (unit A: 63, unit B: 36, unit C: 20). Secondary bacteremia rates did not change over time in unit A (p = .096) and unit B (p = .988). However, there was a statistically significant de-crease in the secondary bloodstream in-fection rates in unit C (p = .028).

Discordance between the infection pre-ventionists and the reviewer were found in 24 cases (unit A: 12, unit B: 10, unit C: 2). Twenty-three of these discrepancies cor-responded to CLABSIs (as per reviewer) that were originally labeled by infection preventionists as secondary bacteremias. Interobserver agreement indexes () were calculated for all three units (unit A: 0.85, unit B: 0.76, unit C: 0.9).

CLABSIs were also analyzed based on the type of organism isolated in blood cultures (Fig. 2 and Supplemental Digital Content 2 [ http://links.lww.com/CCM/A394]). Out of 195 organisms identified during the 33 months of follow-up, 93 (48%) were Gram-positive organisms, 72 (37%) were Gram-negative rods, and 30 (15%) were yeasts. The organisms most frequently identified were Enterococcus species (22%), followed by coagulase-neg-ative staphylococci (15%) and Candida species (15%). The change in proportion of isolates of each type of organism is shown in Supplemental Digital Content 2 (http://links.lww.com/CCM/A394); en-terococcus, yeast, and coagulase-nega-tive staphylococci decreased throughout the phases in all three units. Similarly, the number of polymicrobial CLABSIs decreased after phase 1 across all units (Supplemental Digital Content 2, http://links.lww.com/CCM/A394).

The causative organisms of second-ary bloodstream infections are shown in Supplemental Digital Content 1 (http://links.lww.com/CCM/A393). Out of 119 or-ganisms, the most frequent isolate iden-tified was Klebsiella pneumoniae (18%), followed by Staphylococcus aureus

(17%). The respiratory tract was the most frequent source of secondary bacte-remia (54%), followed by intra-abdominal infections (17%).

Time-to-positive blood cultures (number of days from central catheter insertion to first positive blood culture) was calculated for all CLABSIs. Kaplan-Meier analysis failed to show a significant difference in time-to-positivity through-out the phases (log-rank p value = .614), although a trend toward an increase in median time was noted: phase 1, 10 days (6–19) (interquartile range); phase 2, 9 days (4–24); phase 3, 14.5 days (6.5-20); and phase 4, 16.5 days (2.5–31.5).

Additionally, 34 observations of com-pliance with “scrub-the-hub” were per-formed across the units over 3 months. We observed 100% compliance with us-age of chlorhexidine swabs for “scrub-the-hub” before accessing any central catheter port. Alcohol swabs were not used by healthcare providers in any of the obser-vations. Even though the recommended scrubbing time was 15 secs, the median duration of scrubbing during our obser-vations was 9 secs per access observed (in-terquartile range 5–10). In regards to side effects, one patient in unit C experienced skin irritation and redness thought to be secondary to chlorhexidine baths, which warranted their discontinuation in this individual patient.

DISCUSSION

We report the sequential implementa-tion of interventions aimed at decreas-ing CLABSIs at three ICUs from a single institution. Chlorhexidine “scrub-the-hub” was associated with a reduction of CLABSI rates in one (unit A) of two units (A and B) in which it was indepen-dently applied whereas following daily chlorhexidine baths there was a further reduction in CLABSIs in both units. The combination of the two interventions was also associated with a significant CLABSI reduction in all three units. The addition

of daily nursing rounds (unit A) was asso-ciated with a further decrease of CLABSI rates. Thus, chlorhexidine-based “main-tenance” disinfection using “scrub-the-hub,” skin decontamination via patient baths, and daily ICU goal rounds appeared to be associated with additive benefits re-ducing CLABSIs. Interestingly, contrary to unit A, unit B did not experience a de-crease in CLABSI rates during the scrub-the-hub phase. As previously mentioned, the nurses in unit A came up with the idea and drove the implementation of scrub-the-hub in their unit. This could have certainly driven a higher compli-ance with scrubbing in unit A than in unit B, where implementation was decid-ed by the Infection Control Department.

Secondary bacteremias remained con-stant during the study in two (units A and B) of the three units assessed. Additionally, controlling for CMIs did not impact the CLABSI rates throughout the study. The rates of other HAIs throughout the study were also generally constant. The rate of ventilator-associated pneumonias in unit A decreased during the study, but this co-incided with the institution of a pneumo-nia-prevention bundle and thus was most likely unrelated to the CLABSI interven-tions. The rate of catheter-associated uri-nary tract infections did not change over time (data not shown).

Our data suggests that chlorhexi-dine “scrub-the-hub” and daily patient chlorhexidine baths could have inde-pendent and additive benefits in re-ducing CLABSI rates. Our findings are important for a number of rea-sons. First, this is the first report of additive effectiveness using these two inter ventions and may represent the incremental benefit of targeting in-traluminal infections from the hub and extraluminal infections from the skin surface. Further, similar to the use of chlor hexidine antisepsis at the catheter insertion site (10), these interventions are aimed at reducing CLABSI rates in the “maintenance” phase and there-fore supplement the proven insertion

table 1. Estimates of central catheter-associated bloodstream infection rates by phases

Unit Phase 1 Phase 2 Phase 1 vs.2 Phase 3 Phase 2 vs. 3 Phase 4 Phase 3 vs. 4

A 8.6 (7.2–10.1) 4.9 (3.8–6.4) .001 2.5 (1.8–3.3) .001 0.5 (0.2-1.5) .005B 6.9 (4.3–11.0) 6.8 (3.7–12.7) .996 1.6 (0.8–3.2) .002 Not applicableC 7.8 (5.5–11.0) 0.6 (0.2–1.9) .001

Phase 1, baseline; Phase 2, chlorhexidine “scrub-the-hub”; Phase 3, chlorhexidine “scrub-the-hub” plus daily chlorhexidine baths; Phase 4, chlorhexidine “scrub-the-hub” plus daily chlorhexidine baths plus daily intensive care unit goal sheet nursing rounds. Central catheter-associated bloodstream rates per 1000 central catheter days (95% confidence interval).

1468 Crit Care Med 2012 Vol. 40, No. 5

techniques for reducing CLABSI (6). Finally, the repeated association of CLABSI rate reductions after imple-mentation of these interventions in multiple, independent ICUs within our institution strengthens the likelihood that this result is indeed a result of the intervention.

Contamination of the access port (hub) is a known risk factor for the devel-opment of CLABSIs (11). In a study done by Mahieu et al (12), the incidence of CLABSIs among colonized catheter hubs was higher than among those uncolonized (50% vs. 3%). Furthermore, Safdar and Maki (11) reported, in a prospective study, that contamination of the hubs (intralu-minal source) was associated with 26% of CLABSIs (nine out of 35), and skin coloni-zation (extraluminal sources) caused 45% of cases. The authors also determined that the treatment group (skin decon-tamination at the catheter insertion site) resulted in a decreased number and pro-portion of extraluminal infections while the control group had a relatively higher proportion of intraluminal CLABSIs (11). Despite the recommendation for disinfec-tion of intravenous hubs by the Society for Hospital Epidemiology of America and Infectious Diseases Society of America (13), we believe our report is the first one to show a clinical impact on CLABSI rates by implementing such an intervention, namely, decontaminating the hubs with

chlorhexidine swabs before each intrave-nous access.

Simmons et al (14) have previously evaluated the impact of the duration of “scrub-the-hub” on the bacterial contam-ination of the hubs. According to their study, scrubbing 3, 10, or 15 secs did not show statistically significant differences in the bacterial loads of the hubs, how-ever, the data demonstrated significantly lower bacterial loads when compared to those hubs not scrubbed.

Chlorhexidine is a long-acting anti-septic aimed at decreasing the bacterial load on patients’ skin and thus dimin-ishing the degree of extraluminal con-tamination. Several previously published studies showed the effectiveness of this intervention in other patient populations (7, 15, 16, 18, 19). Our positive results in this surgical ICU population are aligned with those other findings (18, 19) and differ from that of Popovich et al (17), who failed to see a benefit in a surgical ICU.

The limitations of the study include its quasi-experimental nature and con-sequently its lack of standardization. The ICUs started interventions at different times, mainly due to their independent nursing and medical leaderships; in the past, such variables have determined the different adoption times of Infection Control interventions at our institution. Nevertheless, data were gathered from

three different ICUs, totaling 80 beds, and the interventions were performed at slightly different times of the year, which might partially control for confounding factors. Furthermore, we were unable to find reliable data regarding the source of blood culture sticks (new stick vs. cen-tral catheter hubs), which could have influenced the yield of isolation of skin contaminants. Additionally, there were four different infection preventionists throughout the study (two in unit A, one in unit B, and one in unit C) that evalu-ated the positive cultures in these three units. Nonetheless, a single reviewer re-examined all positive blood culture as-sessments performed by the Infection Control Department; however, the lat-ter reviewer was not blinded and thus was subjected to misclassification bias. Survival analyses were unable to find sta-tistically significant differences in time to CLABSI, most likely due to the loss of power throughout the phases (decreased number of CLABSIs over time) and the lack of individualized data on central catheter days among patients who did not develop CLABSIs during their unit admissions.

These limitations notwithstanding, several features of our study strengthen the argument that the association between the interventions reported and reductions in CLABSI rates is indeed due to the ef-fectiveness of these measures in reduc-ing infections in the maintenance phase. First, the reductions in CLABSI rates con-sistently followed the introduction of the interventions described – despite differen-tial timing of these interventions in dif-ferent units – arguing against nonspecific seasonal or Hawthorne effects with time. Second, this association is noted in three independent units, which makes chance less likely. Finally, these interventions did not produce nonspecific reductions in other hospital-acquired infections, such as secondary bacteremias (units A and B) or catheter-associated urinary tract infections (data not shown). Thus, our association is suggestive that interven-tions aimed at reducing CLABSIs during the maintenance phase of central venous catheters can be effective, a result that awaits definitive randomized testing for proof.

American healthcare institutions have entered an era of pay-for-performance, with marked emphasis on prevention of hospital-acquired infections. To reach the goal of zero infections, interven-tions aimed at infection prevention from

figure 2. Proportion of type of organisms by unit and study phase. CLABSIs, central line-associated bloodstream infections.

Crit Care Med 2012 Vol. 40, No. 5 1469

insertion through timely removal of the catheter are required. It stands to reason that bundled interventions encompass-ing best practices for catheter insertion (checklists, chlorhexidine skin disinfec-tion, insertion cart, full barrier precau-tions, hand hygiene, location selection) as well as those addressing maintenance (daily chlorhexidine baths and catheter hub disinfection), are necessary to de-crease CLABSIs. A future study will evalu-ate the impact of the implementation of a more accountable catheter insertion checklist on our current CLABSI rates; another intervention in our quest for zero infections.

REFERENCES

1. Klevens RM, Edwards JR, Richards CL Jr, et al: Estimating health care-associated infections and deaths in U.S. hospitals, 2002. Public Health Rep 2007; 122:160–166

2. Roberts RR, Hota B, Ahmad I, et al: Hospital and societal costs of antimicrobial-resistant infections in a Chicago teaching hospital: Implications for antibiotic stewardship. Clin Infect Dis 2009; 49:1175–1184

3. Roberts RR, Scott RD 2nd, Hota B, et al: Costs attributable to healthcare-acquired infection in hospitalized adults and a comparison of economic methods. Med Care 2010; 48:1026–1035

4. O’Grady NP, Alexander M, Dellinger MP, et al: Guidelines for the prevention of intravascular catheter-related infections. Centers for

Disease Control and Prevention. MMWR Recomm Rep 2002; 51:1–29

5. Centers for Disease Control and Prevention (CDC): Vital signs: Central line-associated blood stream infections–United States, 2001, 2008, and 2009. MMWR 2011; 60:243–248

6. Pronovost P, Needham D, Berenholtz S, et al: An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med 2006; 355:2725–2732

7. Bleasdale SC, Trick WE, Gonzalez IM, et al: Effectiveness of chlorhexidine bathing to re-duce catheter-associated bloodstream infec-tions in medical intensive care unit patients. Arch Intern Med 2007; 167:2073–2079

8. Horan TC, Andrus M, Dudeck MA: CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008; 36:309–332

9. Vernon MO, Hayden MK, Trick WE, et al: Chlorhexidine gluconate to cleanse patients in a medical intensive care unit: The effectiveness of source control to reduce the bioburden of vancomycin-resistant enterococci. Arch Intern Med 2006; 166:306–312

10. Timsit JF, Schwebel C, Bouadma L, et al: Chlorhexidine-impregnated sponges and less frequent dressing changes for prevention of catheter-related infections in critically ill adults: A randomized controlled trial. JAMA 2009; 301:1231–1241

11. Safdar N, Maki DG: The pathogenesis of catheter-related bloodstream infection with noncuffed short-term central venous catheters. Intensive Care Med 2004; 30:62–67

12. Mahieu LM, De Dooy JJ, Lenaerts AE, et al: Catheter manipulations and the risk of

catheter-associated bloodstream infection in neonatal intensive care unit patients. J Hosp Infect 2001; 48:20–26

13. Marschall J, Mermel LA, Classen D, et al: Strategies to prevent central line-associated bloodstream infections in acute care hospitals. Infect Control Hosp Epidemiol 2008; 29:S22–S30

14. Simmons S, Bryson C, Porter S: “Scrub the hub”: Cleaning duration and reduction in bacterial load on central venous catheters. Crit Care Nurs Q 2011; 34:31–35

15. Popovich KJ, Hota B, Hayes R, et al: Effectiveness of routine patient cleansing with chlorhexidine gluconate for infection prevention in the medical intensive care unit. Infect Control Hosp Epidemiol 2009; 30:959–963

16. Munoz-Price LS, Hota B, Stemer A, et al: Prevention of bloodstream infections by use of daily chlorhexidine baths for patients at a long-term acute care hospital. Infect Control Hosp Epidemiol 2009; 30:1031–1035

17. Popovich KJ, Hota B, Hayes R, et al: Daily skin cleansing with chlorhexidine did not reduce the rate of central-line associated bloodstream infection in a surgical intensive care unit. Intensive Care Med 2010; 36: 854–858

18. Evans HL, Dellit TH, Chan J, et al: Effect of chlorhexidine whole-body bathing on hospital-acquired infections among trauma patients. Arch Surg 2010; 145:240–246

19. Dixon JM, Carver RL: Daily chlorohexidine gluconate bathing with impregnated cloths results in statistically significant reduction in central line-associated bloodstream infections. Am J Infect Control 2010; 38:817–821