DOI: 10.1542/peds.2010-3675; originally published online October 24, 2011; 2011;128;e1077Pediatrics

Infection Quality Transformation TeamsAssociated Bloodstream−Institutions Pediatric Intensive Care Unit Central Line

Richard J. Brilli and the National Association of Children's Hospitals and RelatedGayane Yenokyan, Michele Moss, Tom B. Rice, Debra Ridling, Deborah Campbell, Marlene R. Miller, Matthew F. Niedner, W. Charles Huskins, Elizabeth Colantuoni,

Associated Bloodstream Infections: 3-Year Results−Reducing PICU Central Line  

  http://pediatrics.aappublications.org/content/128/5/e1077.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2011 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

Reducing PICU Central Line–Associated BloodstreamInfections: 3-Year Results

WHAT’S KNOWN ON THIS SUBJECT: Pediatric centralline–associated bloodstream infections (CLA-BSIs) are common,are costly (up to $45 000 per event), cause considerablemorbidity, and are the focus of significant research and qualityimprovement efforts.

WHAT THIS STUDY ADDS: This is the first study to showsustained and continually decreasing PICU CLA-BSI rates over anextended period of 3 years and across multiple PICUs.

abstractOBJECTIVES: To evaluate the long-term impact of pediatric central linecare practices in reducing PICU central line–associated bloodstreaminfection (CLA-BSI) rates and to evaluate the added impact of chlorhexi-dine scrub and chlorhexidine-impregnated sponges.

METHODS: A 3-year, multi-institutional, interrupted time-series design(October 2006 to September 2009), with historical control data, wasused. A nested, 18-month, nonrandomized, factorial design was used toevaluate 2 additional interventions. Twenty-nine PICUs were included.Two central line care bundles (insertion and maintenance bundles)and 2 additional interventions (chlorhexidine scrub and chlorhexidine-impregnated sponges) were used. CLA-BSI rates (January 2004 to Sep-tember 2009), insertion and maintenance bundle compliance rates(October 2006 to September 2009), and chlorhexidine scrub andchlorhexidine-impregnated sponge compliance rates (January 2008 toJune 2009) were assessed.

RESULTS: The average aggregate baseline PICU CLA-BSI rate de-creased 56% over 36 months from 5.2 CLA-BSIs per 1000 line-days (95%confidence interval [CI]: 4.4–6.2 CLA-BSIs per 1000 line-days) to 2.3CLA-BSIs per 1000 line-days (95% CI: 1.9–2.9 CLA-BSIs per 1000 line-days) (rate ratio: 0.44 [95% CI: 0.37–0.53]; P � .0001). No statisticallysignificant differences in CLA-BSI rate decreases between PICUs usingor not using either of the 2 additional interventions were found.

CONCLUSIONS: Focused attention on consistent adherence to the useof pediatrics-specific central line insertion and maintenance bundlesproduced sustained, continually decreasing PICU CLA-BSI rates. Addi-tional use of either chlorhexidine for central line entry scrub orchlorhexidine-impregnated sponges did not produce any statisticallysignificant additional reduction in PICU CLA-BSI rates. Pediatrics 2011;128:e1077–e1083

AUTHORS: Marlene R. Miller, MD, MSc,a,b,c Matthew F.Niedner, MD,d W. Charles Huskins, MD, MSc,e ElizabethColantuoni, PhD,f,g Gayane Yenokyan, MD, PhD,g MicheleMoss, MD,h Tom B. Rice, MD,i Debra Ridling, RN, MS,CCRN,j Deborah Campbell, RNC, MSN, CCRN,k Richard J.Brilli, MD,l and the National Association of Children’sHospitals and Related Institutions Pediatric IntensiveCare Unit Central Line–Associated Bloodstream InfectionQuality Transformation Teams

Departments of aPediatrics and fAnesthesia and Critical CareMedicine, School of Medicine, and Departments of bHealth Policyand Management and gBiostatistics, Bloomberg School of PublicHealth, Johns Hopkins University, Baltimore, Maryland; cNationalAssociation of Children’s Hospitals and Related Institutions,Alexandria, Virginia; dDepartment of Pediatrics andCommunicable Diseases, School of Medicine, University ofMichigan, Ann Arbor, Michigan; eDivision of Pediatric InfectiousDiseases, Mayo Clinic, Rochester, Minnesota; hDepartment ofCardiology, Arkansas Children’s Hospital, Little Rock, Arkansas;iDepartment of Pediatrics, Children’s Hospital of Wisconsin,Milwaukee, Wisconsin; jDepartment of Pediatrics, Children’sHospital and Regional Medical Center, Seattle, Washington;kDepartment of Pediatrics, Kosair Children’s Hospital, Louisville,Kentucky; and lDepartment of Pediatrics, Nationwide Children’sHospital, Columbus, Ohio

KEY WORDSbloodstream infections, children, nosocomial infections,pediatric intensive care unit

ABBREVIATIONSCLA-BSI—central line-associated bloodstream infectionCDC—Centers for Disease Control and PreventionNHSN—National Healthcare Safety NetworkNACHRI—National Association of Children’s Hospitals andRelated InstitutionsCI—confidence interval

www.pediatrics.org/cgi/doi/10.1542/peds.2010-3675

doi:10.1542/peds.2010-3675

Accepted for publication Jul 28, 2011

Address correspondence to Marlene R. Miller, MD, MSc,Department of Pediatrics, Johns Hopkins University MedicalSchool, 200 N Wolfe St, Room 2094, Baltimore, MD 21287. E-mail:mmille21@jhmi.edu

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2011 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose.

ARTICLES

PEDIATRICS Volume 128, Number 5, November 2011 e1077 at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

Central line–associated bloodstreaminfections (CLA-BSIs) are commonamong children, are costly, cause con-siderable morbidity, and are the focusof significant research and quality im-provement efforts.1–14 In 2006–2008,the pooled mean rate of CLA-BSIsamong all ICU types in the Centers forDisease Control and Prevention (CDC)National Healthcare Safety Network(NHSN) system ranged from 1.3 to 5.5CLA-BSIs per 1000 catheter days.15–18

With pooled mean rates of 3.3 CLA-BSIsper 1000 catheter days for pediatriccardiothoracic ICUs and 3.0 CLA-BSIsper 1000 catheter days for pediatricmedical/surgical ICUs, PICUs have thethird and fourth highest rates among16 unit types.16 The higher rates in pe-diatric settings likely derive from dif-ferences in how central lines are han-dled for adult and pediatric patients,and they make pediatric units an idealenvironment for learning and processimprovement.

Since October 2006, the National As-sociation of Children’s Hospitals andRelated Institutions (NACHRI) hassupported an ongoing quality trans-formation effort, involving multiplePICUs, to identify and to test the impactof pediatrics-specific catheter carepractices in reducing CLA-BSIs. In 2010,we reported on a new, effective, cen-tral line maintenance bundle that wasable to reduce the rate of pediatricCLA-BSIs by nearly one-half in 29 PICUsduring the first 12 months after im-plementation.19 We now report onlonger-term, 3-year outcomes ofthese efforts through September2009. In addition, we report on a non-randomized, nested, factorial designstudy conducted for 18 consecutivemonths during these 3 years to eval-uate the effects of either chlorhexi-dine scrub on central line caps/ac-cess points before all line entries orchlorhexidine-impregnated sponge

(Biopatch [Ethicon, Somerville, NewJersey]) application on CLA-BSI rates.

METHODS

We described previously the study de-sign, setting, participants, objectives,interventions, and measures.19 Briefly,since October 2006, the NACHRI hassupported collaborative work among�65 PICUs, focused on reducing CLA-BSI rates. The initial cohort included 29PICUs and is the focus of this 3-yearpostintervention report encompassingOctober 2006 to September 2009. Thestudy design is a multi-institutional, in-terrupted time-series with historicalcontrol data from January 2004 to Sep-tember 2006. The goal of this ongoingeffort is to eliminate CLA-BSIs by reli-ably implementing pediatrics-specificinsertion and catheter care mainte-nance bundles (Table 1).

Data on 2 process measures, that is,insertion bundle compliance andmaintenance bundle compliance, werecollected. The audit process was de-scribed previously.19 Our outcomemeasure was the monthly PICU CLA-BSI

rate, defined as the number of CLA-BSIcases per 1000 line-days, and datawere collected by trained, hospital-based, infection control practitioners,in accordancewith CDC definitions.20 In2008, the NHSN definition for CLA-BSIwas revised, most notably to require 2positive blood culture results with or-ganisms considered to be commonskin contaminants.21 Our efforts incor-porated this definitional change bothin practice and in analyses.

In addition to core efforts in reliablyimplementing and sustaining insertionand maintenance bundle care prac-tices to reduce CLA-BSI rates, we con-ducted a nested, 18-month (January2008 to June 2009), nonrandomized,factorial design study to examine theimpact of 2 other interventions. Theteams were voluntarily divided into4 groups, that is, a control group (8PICUs), a group using chlorhexidinescrub on central line caps/accesspoints with each central line entry (11PICUs), a group using chlorhexidine-impregnated sponges with all central

TABLE 1 Central Line Care Bundles

Insertion bundleHand washing before procedureChlorhexidine scrub at insertion site (30-s scrub [2 min for groin] and 30–60-s air dry for all children

�2 mo of age)No iodine skin preparation; no iodine ointment at insertion sitePrepackaged or filled insertion cart, tray, or boxInsertion checklist (with staff empowerment to stop nonemergency procedure if sterile insertionpractice not being followed)

Full sterile barrier for providers and patientInsertion training for all providers (eg, slides and video)Maintenance bundleDaily assessment of whether catheter is neededCatheter site careNo iodine ointmentChlorhexidine scrub to site with dressing changes (30-s scrub and 30-s air dry)Change gauze dressings every 2 d unless soiled, dampened, or loosened (CDC recommended)Change clear dressing every 7 d unless soiled, dampened, or loosened (CDC recommended)Prepackaged dressing change kit (each unit to define package contents)Catheter hub/cap/tubing careReplace administration sets, including add-on devices, no more frequently than every 72 h unlesssoiled or suspected to be infectedReplace tubing used to administer blood, blood products, or lipids within 24 h after initiation ofinfusion (CDC recommended)Change caps no more often than 72 h (or according to manufacturers’ recommendations), but capsshould be replaced when administration set is changed (CDC recommended)Prepackaged cap change kit/cart/central location (elements designated by local institution)

e1078 MILLER et al at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

line site care (5 PICUs), and a groupusing both chlorhexidine scrub andchlorhexidine-impregnated sponges(5 PICUs). Furthermore, given thestrong motivation by many PICUs to re-duce CLA-BSIs before our nested 18-month study, some of the PICU teamshad implemented 1 or both of theseadditional interventions but had notdone so universally for all patientswith a variable degree of standardiza-tion and with no adherence monitor-ing. Once the nested factorial studywas added to the collaborative work,all teams audited compliance with useof the core insertion and maintenancebundles and the 2 supplemental inter-ventions according to standardizedprotocols. The chlorhexidine scrub in-tervention was used for all central lineaccess via cap-type access points (ex-cept for emergency access) and con-sisted of a 30-second scrub followed bya 30-second air drying. Compliancewith this intervention was audited bycollecting data for all PICU patients for48 hours per month (typically dividedinto 8- to 12-hour continuous periodsof data collection) and recordingwhether chlorhexidine was used foreach line entry. For chlorhexidine-impregnated sponges, the protocol in-volved applying a chlorhexidine-impregnated sponge at the central lineentry site at the time of line insertionand applying a new sponge with eachdressing change, according to themanufacturer’s recommendations.Compliance was assessed throughweekly examinations of all patients inthe PICU. Monthly data were analyzedfrom January 2004 through Septem-ber 2009 and subdivided into 3 timeperiods, namely, a preinterventionbaseline period (January 2004 to Sep-tember 2006; months 1–33), a ramp-upperiod of intervention deployment (theinitial 3 months, from October 2006 toDecember 2006; months 34–36), and apostintervention period (January 2007to September 2009; months 37–69).

We used marginal generalized linearmodels assuming a negative binomialdistributionwith logarithmic link to es-timate the average CLA-BSI rate as afunction of time within each period(different temporal slopes within thebaseline, ramp-up, and stable peri-ods), and adjustments were made forPICU-level characteristics (geographicregion, average length of stay, and bedcapacity). To quantify the effect of therevised NHSN definition of CLA-BSIs, weincluded a main effect of time formonths January 2008 through Septem-ber 2009; this main effect allowed theaverage CLA-BSI rate to change be-cause of the revised NHSN definitionstarting in January 2008. To accountfor the clustering effect of CLA-BSIswithin individual PICUs, the modelswere fit by using generalized estimat-ing equations assuming an autore-gressive working correlation struc-ture.22 Robust variance estimateswereobtained to account for any misspeci-fication of the working correlationstructure.

To explore the effects of the facto-rial design nested study (controlgroup, chlorhexidine scrub group,chlorhexidine-impregnated spongegroup, and dual intervention group),we modeled the average CLA-BSI ratesover time separately for both of the ad-ditional interventions. The PICU facto-rial design group membership waspermitted to change before January2008, reflecting the historical nonpro-tocolized use of these 2 additional in-terventions by some PICUs. BetweenJanuary 2008 and June 2009, the addi-tional intervention group membershipwas not changed and audited compli-ance data were collected regularly. Av-erage CLA-BSI rates were modeledover time separately for each group byusing a marginal generalized linearmodel incorporating 3 time periodsand addressing 2 phases of use ofthese 2 additional interventions (non-

protocolized use in the ramp-up periodfrom October 2006 to December 2006and in postintervention year 2007 andprotocolized formal factorial design inJanuary 2008 to June 2009). The mod-els included indicators for factorial de-sign group (with the control group asreference) and the interaction of timeand each additional intervention. Anautoregressive working correlationstructure was specified, and SEs wereestimated with robust variance esti-mation. All analyses were conductedby using SAS 9.2 (SAS Institute, Cary,NC).

RESULTS

Characteristics of the 29 PICUs in theNACHRI PICU CLA-BSI collaborative arepresented in Table 2. Most PICUs weremixed PICUs and cardiac ICUs; 2 ofthemwere solely cardiac ICUs. Thema-jority of facilities were level 1 traumacenters and performed solid-organtransplants, bone marrow trans-

TABLE 2 Characteristics of 29 ParticipatingICUs

Characteristic n (%)

Type of unitPICU 9 (31)PICU/cardiac ICU 18 (62)Cardiac ICU 2 (7)No. of beds10–16 12 (41)17–27 13 (45)28–36 4 (14)Annual mean PICU length of stay2.7–4.5 d 14 (48)4.6–6.3 d 12 (41)6.4–9.6 d 3 (10)Yearly total of PICU patient-days2100–3700 10 (34)3701–6300 10 (34)6301–8700 9 (31)No. of ICU admissions in 2005300–900 10 (34)901–1800 15 (52)1801–2400 4 (14)Institution is level 1 trauma center 19 (66)Institution performs solid-organtransplants

25 (86)

Institution performs bone marrowtransplants

21 (72)

ICU performs extracorporeal membraneoxygenation

26 (90)

ARTICLES

PEDIATRICS Volume 128, Number 5, November 2011 e1079 at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

plants, and extracorporeal membraneoxygenation.

A total of 1759 CLA-BSI events were re-ported by the 29 PICUs during the 36-month postintervention study period(501 911 total line-days). Figure 1 pres-ents the average monthly PICU CLA-BSIrates during the study period. The av-erage aggregate baseline PICU CLA-BSIrate for the 29 ICUs was 5.2 CLA-BSIsper 1000 line-days (95% confidence in-terval [CI]: 4.4–6.2 CLA-BSIs per 1000line-days) for the period from January2004 to September 2006. After initia-tion of the interventions, the rate be-

gan to decrease in the ramp-up period(4.3 CLA-BSIs per 1000 line-days [95%CI: 3.2–5.7 CLA-BSIs per 1000 line-days]) and decreased to an averagepostintervention rate of 2.3 CLA-BSIsper 1000 line-days (95% CI: 1.9 –2.9CLA-BSIs per 1000 line-days) for theperiod from January 2007 to Septem-ber 2009. This indicates a significantdifference between the baseline pe-riod CLA-BSI rate and the postinter-vention period CLA-BSI rate (rate ra-tio: 0.44 [95% CI: 0.37– 0.53]; P �.0001); there was a decrease in theCLA-BSI rate of 56%.

The marginal generalized linear modelwas used to estimate the monthlychanges in CLA-BSI rates during thebaseline, ramp-up, and postinterven-tion periods. The CLA-BSI rate de-creased by an estimated 1% permonthin the baseline period; however, thisdecrease was not statistically signifi-cant (P � .307). During the ramp-upperiod, the estimated decrease was�11% per month (95% CI: 3%–18% de-crease per month; P � .006). Duringthe postintervention period, the infec-tion rate decreased by 2% per month(95% CI: 1% increase to 4% decreasepermonth; P� .088); however, this de-crease reached only marginal statisti-cal significance. The infection rate de-creased by �15% in January 2008because of the change in the NHSN def-inition for CLA-BSIs; however this de-crease was not statistically significant(P� .448). The estimates were similarafter adjustment for geographical re-gion, average length of stay, and bedcapacity (Table 3).

The results of the factorial design eval-uation of the 2 additional interventions(chlorhexidine scrub of central linecaps/access points and chlorhexidine-impregnated sponges) indicated thatthe CLA-BSI rates decreased for all 4groups (Fig 2). There were no statisti-cally significant differences in therates of decrease among the 4 groups,during the nonprotocolized use ofthese additional interventions from

FIGURE 1Plot of CLA-BSI rates and insertion and maintenance compliance rates (and 95% CIs) in the preinter-vention baseline and intervention periods for the 29 PICUs.

TABLE 3 Results of Generalized Linear Marginal Models Examining Temporal Trends and Effects of PICU-Related Characteristics on CLA-BSI Rates

Variables Unadjusted Models Adjusted Models

Rate Ratio (95% CI) P Rate Ratio (95% CI) P

Time periods (per month)Baseline decrease (January 2004 to September 2006) 0.99 (0.98–1.01) .307 0.99 (0.98–1.01) .339Ramp-up period decrease (October 2006 to December 2006) 0.89 (0.82–0.97) .006 0.89 (0.81–0.97) .006Stable decrease (January 2007 to September 2009) 0.98 (0.96–1.01) .088 0.98 (0.96–1.01) .084NHSN definition change (January 2008 to September 2009) 0.86 (0.57–1.28) .448 0.85 (0.58–1.26) .416Length of stay (per day) 0.99 (0.90–1.10) .867 1.00 (0.88–1.13) .962No. of beds (per 100) 2.04 (0.18–23.51) .569 2.44 (0.23–25.60) .456RegionNortheast vs West 0.83 (0.48–1.42) .494 0.84 (0.5–1.4) .497Midwest vs West 0.85 (0.54–1.36) .507 0.82 (0.53–1.27) .385South vs West 0.95 (0.59–1.53) .824 0.94 (0.59–1.5) .802

e1080 MILLER et al at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

October 2006 to December 2007 or dur-ing the protocolized use from January2007 to June 2009.

DISCUSSION

Results from the 3-year NACHRI qualitytransformation effort focused on re-ducing CLA-BSI rates in 29 PICUs showthat such targeted efforts can achievesignificant decreases in CLA-BSI rates,that the rates continue to decreaseover time with ongoing focused ef-forts, and that the overall decreasecan be sustained for an extended pe-riod. These results attest to the sus-tainability of efforts to improve centralline maintenance care, the main driverfor CLA-BSI rate reduction in pediat-rics.19 Maintenance care practices arecomplex and multidimensional fornurses to master reliably. Further-more, they need to be performed con-sistently and with precision manytimes each day for each patient. Thisfocused, 3-year, collaborative effortwith repetitive teaching on how toachieve reliable consistent use of thebundles resulted in statistically signif-icant decreases in CLA-BSI rates, build-ing on the reduction achieved during

the first year. Sustaining care practicechanges that achieve measureablereductions in preventable harm(hospital-acquired CLA-BSIs) often ismore difficult that the initial workneeded to achieve improvement, andthis report demonstrates that an ex-tended period of improvement can beachieved. Using literature-based adultestimates of CLA-BSI–attributablemor-tality rates and pediatrics-specificdata on CLA-BSI morbidity, such as acost of $45 000 per CLA-BSI episode, weestimate (recognizing the potential im-precision of these estimates and thelimitations of using them to quantifythe impact of this work) that the 29PICUs participating in this 3-year col-laborative have prevented �900 CLA-BSIs, saved �100 children’s lives byavoiding CLA-BSIs, and saved more than$31 million in CLA-BSI–attributablehealth care costs, on the basis of 2.5years of historic baseline perfor-mance data for each team.4,5,12,14,22 Interms of the costs of this effort to the29 institutions, the estimates we canprovide are crude, because of signifi-cant variations in institutional contex-

tual issues such as the size of the PICU,the number of PICUs, the baseline rateof CLA-BSIs, baseline adherence tobundle practices, local influences ofpublic reporting and regulatory re-quirements, institutional quality re-sources, institutional financial status,local safety culture, local availability ofinstitutional quality thought leadersand quality improvement consultants,institutional training resources, andinstitutional quality incentives.23

Teams participating in our effort, overthe course of the 3-year period, paid anestimated annual “pay to participate”fee of $9000 (fees were adjusted annu-ally to reflect operating costs, includ-ing staff members, faculty members,data systems, and meeting facilities),paid approximately $5000 per year topermit 2 or 3 team members to travelto 2 face-to-face learning sessionmeetings each year (although someteams voluntarily sent many morestaff members to learning sessionmeetings), and needed �0.4 full-timeequivalents (as a crude average), typi-cally registered nurse–level, to sup-port data collection, data entry, andlocal quality improvement efforts. Col-lectively, the “return on investment”from this effort was substantial; 29 in-stitutions paid approximately $75 000in each of 3 years in expenses to sup-port this work, which resulted in $31million in savings to the health caresystem.

To our knowledge, this is the first studyto show sustained and continually de-creasing PICU CLA-BSI rates over an ex-tended period of 3 years and acrossmultiple PICUs, correlated directlywith implementation of a standardizedapproach to central line care. For chil-dren in ICUs, however, the main driverfor CLA-BSI rate reductions is the dailycentral line maintenance care deliv-ered primarily by nurses.19 Given themanyfold greater numbers of nursingstaff members employed in PICUs,

FIGURE 2Factorial design evaluation of chlorhexidine scrub of central line caps/access points andchlorhexidine-impregnated sponges using 4 groupings of PICUs, showing CLA-BSI rates (and 95% CIs)during nonprotocolized and protocolized periods of evaluation. CHG indicates chlorhexidinegluconate.

ARTICLES

PEDIATRICS Volume 128, Number 5, November 2011 e1081 at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

compared with physician staff mem-bers, and given the fact that any 1 cen-tral line has 1 episode of insertion buthundreds of episodes of maintenancecare, the fact that PICU CLA-BSI ratesdecrease more gradually than seenwith adult ICU efforts is not surpris-ing.13 Similar gradual decreases in pe-diatric CLA-BSI rates were reported fora single cardiac PICU and a large co-hort of NICUs.24,25 Focused attentiononly on central line insertion, as is oc-curring in the adult ICU community, isnot likely to achieve significant suc-cess in the PICU community.

Although these PICUs were focused onconsistently adhering to the standardbest practice bundles for central lineinsertion and maintenance care, addi-tion of either a chlorhexidine scrubat central line access points orchlorhexidine-impregnated sponges atinsertion sites did not provide signifi-cantly greater reductions in CLA-BSIrates, compared with those achievedthrough reliable adherence to use ofthe core bundles. These negative re-sults are most appropriately consid-ered in the context that PICU teamswere focused on consistently usingbest practices for both insertion andcentral line maintenance care in addi-tion to evaluating these 2 additionalstrategies. The literature containsinconclusive evidence that either of

these 2 interventions is linked toCLA-BSI rate reductions amongchildren.26–29

There are several limitations to thisstudy. First, the CLA-BSI definition usedby the CDC does not take into accountthe number of central lines and/or lu-mens per patient-day. Furthermore,the CDC definition does not necessarilyidentify truly preventable CLA-BSIs. Be-cause the CDC CLA-BSI definition iswidely accepted and is commonly usedby most hospitals, however, it offeredthe best method to limit variations indefinition interpretation. Variability inthe application of this definition, asdocumented in other settings, mighthave influenced our findings.30,31 Sec-ond, the factorial design evaluationof both chlorhexidine scrub for cen-tral line entry and chlorhexidine-impregnated sponges was not ran-domized and was influenced by teamsusing these tools in a nonprotocolizedmanner before the beginning of the 18-month evaluation. Despite these fac-tors, it is compelling that, in the con-text of reliable use of the core bundlepractices, neither of these supple-mental interventions provided anyadditional benefit in CLA-BSI rate re-ductions. Overall, our 3-year effortshows that focus on reliable adher-ence to core practices for insertionand maintenance care may be the best

way to further CLA-BSI rate reductions,rather than adding new bundle compo-nents (eg, chlorhexidine scrub for en-try of all central line caps/accesspoints and chlorhexidine-impregnatedsponges).

CONCLUSIONS

CLA-BSIs are a preventable cause of pa-tient harm for critically ill children. Our3-year quality transformation effort,conducted in 29 PICUs, shows thatfocused consistent adherence to bothinsertion and maintenance bundlesproduces sustained and continuallydecreasing CLA-BSI rates. The key is-sues for pediatrics are as follows: (1)understanding that themain driver forfurther reducing pediatric CLA-BSIrates is daily maintenance care forcentral lines and (2) accepting that re-liably performing tasks that occurmultiple times per day for each PICUpatient is a challenging but crucialbody of work for all pediatric provid-ers. In the context of our bundles ofcare, supplemental use of either chlo-rhexidine for central line entry scrubor chlorhexidine-impregnated spongesdid not produce any appreciable addi-tional reduction in CLA-BSI rates. Morestudy is needed todetermine theoptimalmaintenance care bundle that will facili-tateeliminationofpediatricCLA-BSIsandbe sustainable.

REFERENCES

1. Mermel L. Prevention of intravascularcatheter-related infections. Ann InternMed.2000;132(5):391–402

2. Burke JP. Infection control: a problem forpatient safety. N Engl J Med. 2003;348(7):651–656

3. Richards MJ, Edwards JR, Culver DH,Gaynes RP; National Nosocomial InfectionsSurveillance System. Nosocomial infectionsin pediatric intensive care units in theUnited States. Pediatrics. 1999;103(4). Avail-able at: www.pediatrics.org/cgi/content/full/103/4/e39

4. Pittet D, Tarara D, Wenzel RP. Nosocomialbloodstream infection in critically illpatients: excess length of stay, extra costs,

and attributable mortality. JAMA. 1994;271(20):1598–1601

5. Pittet D, Wenzel RP. Nosocomial bloodstreaminfections: secular trends in rates, mortality,andcontribution to total hospital deaths.ArchIntern Med. 1995;155(11):1177–1184

6. Urrea M, Pons M, Serra M, Latorre C, Pal-omeque A. Prospective incidence study ofnosocomial infections in a pediatric inten-sive care unit. Pediatr Infect Dis J. 2003;22(6):490–494

7. Yogaraj JS, Elward AM, Fraser VJ. Rate, riskfactors, and outcomes of nosocomial pri-mary bloodstream infection in pediatric in-tensive care unit patients. Pediatrics. 2002;110(3):481–485

8. Grohskopf LA, Sinkowitz-Cochran RL, Gar-rett DO, et al. A national point-prevalencesurvey of pediatric intensive care unit-acquired infections in the United States. JPediatr. 2002;140(4):432–438

9. Odetola FO, Moler FW, Dechert RE, Van-DerElzen K, Chenoweth C. Nosocomialcatheter-related bloodstream infections ina pediatric intensive care unit: risk andrates associatedwith various intravasculartechnologies. Pediatr Crit Care Med. 2003;4(4):432–436

10. Wisplinghoff H, Seifert H, Tallent SM,Bischoff T, Wenzel RP, Edmond MB. Nosoco-mial bloodstream infections in pediatric pa-tients in United States hospitals: epidemiol-

e1082 MILLER et al at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

ogy, clinical features and susceptibilities.Pediatr Infect Dis J. 2003;22(8):686–691

11. Aly H, Herson V, Duncan A, et al. Is blood-stream infection preventable among pre-mature infants? A tale of two cities. Pediat-rics. 2005;115(6):1513–1518

12. Elward AM, Hollenbeak CS, Warren DK,Fraser VJ. Attributable cost of nosocomialprimary bloodstream infection in pediatricintensive care unit patients. Pediatrics.2005;115(4):868–872

13. Pronovost P, Needham D, Berenholtz S, et al.An intervention to decrease catheter-related bloodstream infections in the ICU. NEngl J Med. 2006;355(26):2725–2732

14. Nowak JE, Brilli RJ, Lake MR, et al. Reducingcatheter-associated bloodstream infec-tions in the pediatric intensive care unit:business case for quality improvement. Pe-diatr Crit Care Med. 2010;11(5):579–587

15. Edwards JR, Peterson KD, Andrus ML, et al.National Healthcare Safety Network (NHSN)report, data summary for 2006, issued June2007. Am J Infect Control. 2007;35(5):290–301

16. O’Grady N, Alexander M, Dellinger E, et al.Guidelines for the prevention of intravascu-lar catheter-related infections. MMWR Re-comm Rep. 2002;55(RR-10):1–26

17. Centers for Disease Control and Prevention.Available at: http://www.cdc.gov/nhsn/psc_da.html. Accessed October 3, 2011

18. Edwards JR, Peterson KD, Mu Y, et al. Na-tional Healthcare Safety Network (NHSN)report: data summary for 2006 through

2008, issued December 2009. Am J InfectControl. 2009;37(10):783–805

19. Miller MR, Griswold M, Harris JM II, et al.Decreasing PICU catheter-associated blood-stream infections: NACHRI’s quality trans-formation efforts. Pediatrics. 2010;125(2):206–213

20. Centers for Disease Control and Prevention.CDC/NHSN surveillance definition of healthcare–associated infection and criteria forspecific types of infections in the acute caresetting. Available at: www.cdc.gov/ncidod/dhqp/pdf/nnis/NosInfDefinitions.pdf. Ac-cessed July 29, 2010

21. Centers for Disease Control and Prevention.Central line-associated bloodstream infec-tion (CLABSI) event. Available at: www.cdc.gov /nhsn /PDFs /pscManua l / 4PSC_CLABScurrent.pdf. Accessed October 3, 2011

22. Slonim AD, Kurtines HC, Sprague B, et al. Thecosts associated with nosocomial blood-stream infections in the pediatric intensivecare unit. Pediatr Crit Care Med. 2001;2(2):170–174

23. Shekelle PG, Pronovost PJ, Wachter RM, etal. Advancing the science of patient safety.Ann Intern Med. 2011;154(10):693–696

24. Costello JM, Morrow DF, Graham DA, Potter-Bynoe G, Sandora TJ, Laussen PC. System-atic intervention to reduce central line as-sociated bloodstream infection rates in apediatric cardiac intensive care unit. Pedi-atrics. 2008;121(5):915–923

25. Wirtshafter DD, Pettit J, Kurtin P, et al. Astatewide quality improvement collabora-t i ve to reduce neonata l centra l

line–associated blood stream infections. JPerinatol. 2010;30(3):170–181

26. Soothill JS, Bravery K, Ho A, Macqueen S,Collins J, Lock P. A fall in bloodstream infec-tions followed a change to 2% chlorhexidinein 70% isopropanol for catheter connectionantisepsis: a pediatric single center before/after study on a hemopoietic stem celltransplant ward. Am J Infect Control. 2009;37(8):626–630

27. Ho KM, Litton E. Use of chlorhexidine-impregnated dressing to prevent vascularand epidural catheter colonization and in-fection: a meta-analysis. J Antimicrob Che-mother. 2006;58(2):281–287

28. Timsit JF, Schwebel C, Bouadma L, et al.Chlorhexidine-impregnated sponges andless frequent dressing changes for preven-tion of catheter-related infections in criti-cally ill adults: a randomized controlledtrial. JAMA. 2009;301(12):1231–1241

29. Onder AM, Chandar J, Coakley S, FrancoeurD, Abitbol C, Zilleruelo G. Controlling exit siteinfections: does it decrease the incidence ofcatheter-related bacteremia in children onchronic hemodialysis? Hemodial Int. 2009;13(1):11–18

30. Lin MY, Hota B, Khan YM, et al. Quality oftraditional surveillance for public reportingof nosocomial bloodstream infection rates.JAMA. 2010;304(18):2035–2041

31. Niedner MF. The harder you look, the moreyou find: catheter associated bloodstreaminfection surveillance variability. Am J In-fect Control. 2010;38(8):585–595

ARTICLES

PEDIATRICS Volume 128, Number 5, November 2011 e1083 at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2010-3675; originally published online October 24, 2011; 2011;128;e1077Pediatrics

Infection Quality Transformation TeamsAssociated Bloodstream−Institutions Pediatric Intensive Care Unit Central Line

Richard J. Brilli and the National Association of Children's Hospitals and RelatedGayane Yenokyan, Michele Moss, Tom B. Rice, Debra Ridling, Deborah Campbell, Marlene R. Miller, Matthew F. Niedner, W. Charles Huskins, Elizabeth Colantuoni,

Associated Bloodstream Infections: 3-Year Results−Reducing PICU Central Line  

ServicesUpdated Information &

htmlhttp://pediatrics.aappublications.org/content/128/5/e1077.full.including high resolution figures, can be found at:

References

html#ref-list-1http://pediatrics.aappublications.org/content/128/5/e1077.full.at:This article cites 27 articles, 12 of which can be accessed free

Subspecialty Collections

diseasehttp://pediatrics.aappublications.org/cgi/collection/infectious_Infectious Disease & Immunitythe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

mlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhttables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2011 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Pontificia Universidad Catolica de Chile on November 24, 2011pediatrics.aappublications.orgDownloaded from