The efficacy of infection control interventions in reducing the incidence of extended-spectrum b-lactamase- producing Enterobacteriaceae in the nonoutbreak setting: A systematic review

F C T

A S O

C

0

C C r

d

The efficacy of infection control interventions in reducing the incidence of extended-spectrum b-lactamase- producing Enterobacteriaceae in the nonoutbreak setting: A systematic review

Shannon Goddard, MD, FRCPC,a and Matthew P. Muller, MD, PhD, FRCPCa,b 

Toronto, Canada

rom ana oron

ddr t. M N,

onfl

196

opy ont eser

oi:1

We conducted a systematic review to examine the efficacy of infection control interventions for the control of ESBL-producing Enterobacteriaceae in hospitals in the non-outbreak setting. Although 4 uncontrolled, retrospective studies were included in the review, no well designed prospective studies capable of informing infection control practice were identified, underscoring the ur- gent need for research in this area. Key Words: ESBL; infection control; systematic review; Enterobacteriaceae; multi-drug resistant gram negative bacteria.

Copyright ª 2011 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. (Am J Infect Control 2011;39:599-601.)

Despite the increasing global incidence of extended- spectrum b-lactamase (ESBL)-producing Enterobacteri- aceae (ESBL-E), there are no evidence-based guidelines specifying the infection control interventions that should be used to control ESBL-E transmission in hos- pitals in the absence of a defined outbreak.1 We con- ducted a systematic review to determine which, if any, infection control interventions are effective in re- ducing the incidence of ESBL-E among hospitalized patients in the nonoutbreak setting.

METHODS

We searched MEDLINE, CINAHL, EMBASE, and the Cochrane Library for studies published from January 1985 to April 2010. The search terms used included terms to identify the organism (eg, Enterobacteriaceae),

the Department of Medicine, University of Toronto, Toronto, daa; and Division of Infectious Diseases, St. Michael’s Hospital, to, Canada.b

ess correspondence to Matthew P. Muller, MD, PhD, FRCPC, ichael’s Hospital, 30 Bond St, 4CC wing, room 4-178, Toronto, M5B 1W8, Canada. E-mail: mullerm@smh.ca.

icts of interest: None to report.

-6553/$36.00

right ª 2011 by the Association for Professionals in Infection rol and Epidemiology, Inc. Published by Elsevier Inc. All rights ved.

0.1016/j.ajic.2010.09.018

the mechanism of resistance (eg, ESBL), and the inter- vention (eg, contact precautions). The complete search strategy is available from the authors on request.

Studies were included if the study was designed as a randomized controlled trial (RCT), time series analysis, or quasi-experimental study; if they included an infection control intervention designed to reduce the person-to-person transmission of ESBL-E; and if they reported total or nosocomial ESBL-E incidence as an outcome. Studies were excluded if third-generation cephalosporin resistance was not confirmed to be due to ESBL production; if the intervention was con- ducted during a defined clonal outbreak; if the inter- vention focused on antibiotic stewardship or digestive tract decolonization; or if the study was not published in a peer-reviewed journal or not published in English.

Articles were reviewed independently by both au- thors to determine eligibility; differences of opinion were resolved through consensus. Studies were not ex- cluded because of poor quality if the inclusion criteria were met. Instead, the strength of the study design was assessed using a guidelines for the evaluation of quasi- experimental studies of infection control interventions (no RCT were identified).2

RESULTS

Our search strategy identified 995 articles: 10 were considered eligible on initial review, but 6 were

599

 

 

Table 1. Recommendations for future trials of infection control interventions for ESBL-E

Study design Interventions should be evaluated using time series analysis or prospective controlled quasi-experimental studies with

repeated, removed, or crossover designs that will strengthen the internal validity of the results.

Interventions that are effective in well-designed quasi-experimental studies should be prioritized for testing using a cluster

randomized design, which minimizes bias because of unmeasured confounders.

Interventions Studies should focus on isolation and/or contact precautions with and without active surveillance for ESBL-E colonization

rather than on routine practices that are already required (eg, hand hygiene).

The intervention should be described in sufficient detail as to be reproducible.

Preintervention infection control strategies for ESBL-E should be clearly described.

Cointerventions Cointerventions with the potential to impact ESBL-E incidence (eg, initiation of an antibiotic stewardship program,

intervention to improve hand hygiene) should be avoided; if cointerventions occur, they should be described and their

impact assessed (eg, report change in antimicrobial usage patterns, change in hand hygiene compliance over study period).

Outcome measures Nosocomial ESBL-E incidence per patient day or per admission should be the preferred outcome.

Surveillance cultures on admission should be used to help distinguish patients with colonization at admission from those who

develop colonization following hospitalization because in-hospital interventions can only influence the latter group.

ESBL-E production should be confirmed using standard phenotypic methods; additional molecular data should be presented

to exclude both clonal and plasmid outbreaks (eg, PFGE, plasmid typing).

Confounding factors Potential confounding factors including changes in the incidence risk factors for ESBL-E acquisition should be monitored (eg,

antibiotic use, length of hospital or ICU stay, hand hygiene compliance, severity of illness).

ESBL-E, extended-spectrum b-lactamase (ESBL)-producing Enterobacteriaceae; ICU, intensive care unit; PFGE, pulsed-field gel electrophoresis.

600 Goddard and Muller American Journal of Infection Control September 2011

excluded following full review because ESBL-E inci- dence was not measured at baseline (n 5 2),3,4 no in- tervention was studied (n 5 2),5,6 the study was conducted during an ESBL-E outbreak (n 5 1),7 or the study intervention involved antibiotic stewardship (n 5 1).8 Therefore, 4 studies were included in our systematic review.9-12

All 4 studies were conducted retrospectively. One was an interrupted time series analysis of methicillin- resistant Staphylococcus aureus incidence that exam- ined ESBL incidence in a post hoc analysis.10 The other 3 studies were uncontrolled pretest-post-test studies.9-11,12 There were no RCT or controlled trials of any type identified. No study reported obtaining ap- proval from a Research Ethics Board.

The study by Conterno et al9 was an uncontrolled pretest-post-test study conducted at a tertiary care hos- pital. Patients identified as ESBL-E colonized/infected on clinical cultures were placed in isolation, and contact precautions were used for patients at risk of contaminating their environment. The nosocomial ESBL-E incidence rose from 0.28 per 1,000 admissions in 1999 to 0.67 per 1,000 admissions in 2005 despite the initiation of the intervention in 2001. The results are difficult to interpret because the study was not con- trolled, the regional incidence of ESBL-E increased 7-fold during the study period, and the lack of active surveillance may have led to the classification of patients with unrecognized colonization at the time of admission as nosocomial cases.

The study by Johnson et al10 was an uncontrolled time series analysis conducted at a tertiary care hospital. A program to reduce methicillin-resistant Staphylococcus aureus that included a hand hygiene

intervention was identified post hoc as resulting in a re- duction in the incidence of clinical cultures positive for ESBL-E, which had been rising for 28 months before the intervention (slope . 0; P , .05) but fell to zero over the 36-month postintervention period (slope , 1; P , .05). Interpretation is limited by the lack of a control group, the post hoc nature of the analysis, the limited data on ESBL-E presented, and the possibility that the rise and fall in ESBL-E incidence may be indic- ative of a prolonged ESBL-E outbreak.

The study by Souweine et al11 was an uncontrolled pretest-post-test study conducted in a 10-bed intensive care unit. The intervention included active surveillance for antibiotic resistant organisms including ESBL- producing Klebseilla pneumoniaewith isolation and con- tact precautions for colonized/infected patients and a hand hygiene promotion. The incidence of ESBL- producing Klebseilla pneumoniae fell from 1.3 per 1,000 patient-days preintervention to 0 per 1,000 patient- days postintervention. The study is difficult to interpret because of the lack of a control group, the small sample size (only 4 cases of ESBL-Ewere identified in the study), and the possibility that the 4 cases represented a cluster.

The study by Soulier et al12 was an uncontrolled pretest-post-test study conducted in an 11-bed inten- sive care unit. An educational intervention was used to improve hand hygiene and environmental cleaning. The proportion of ESBL-E colonized patients fell from 70% during a 3-month preintervention period to 40% during a 7-month postintervention period (P 5 .001). This study is also difficult to interpret because of the lack of a control group, and the high proportion of col- onized patients suggests the possibility of an extensive ESBL-E outbreak.

 

 

www.ajicjournal.org Vol. 39 No. 7

Goddard and Muller 601

DISCUSSION

Our systematic review did not identify any prospec- tive, controlled, or randomized trials addressing the ef- ficacy of infection control interventions for the control of ESBL-E incidence in the nonoutbreak setting. The in- cluded studies were all retrospective, uncontrolled, quasi-experimental studies with a variety of additional methodologic flaws (eg, no description of baseline ap- proach to ESBL-E control; confounders not measured; presence of cointerventions; limited sample size; post hoc analysis; failure to exclude an outbreak). No con- clusions regarding the efficacy of any of the interven- tions studied can therefore be made.

Our review demonstrates that, despite the recogni- tion that gram-negative resistance is a public health crisis requiring urgent action,13 there is little research activity addressing the control of ESBL-E transmission in hospitals. This is emphasized by the lack of pro- spective or Research Ethics Board approved studies identified. Whereas included studies represent the laudable efforts of infection control programs to retro- spectively evaluate their efforts to control ESBL-E transmission, this should not be viewed as a replace- ment for the conduct of high-quality, prospective re- search that addresses the optimal way for hospitals to reduce ESBL-E incidence (and the incidence of other multidrug-resistant, gram-negative organisms).

We therefore recommend that researchers and fund- ing agencies prioritize studies designed to identify ef- fective infection control strategies for the control of ESBL-E and other multidrug-resistant, gram-negative organisms. Such studies should be prospective studies using a cluster randomized, time series, or controlled quasi-experimental design. Additional recommenda- tions for future studies are presented in Table 1.

References

1. Paterson DL, Bonomo RA. Extended-spectrum b-lactamases: a clinical

update. Clin Microbiol Rev 2005;18:657-86.

2. Harris AD, Bradham DD, Baumgarten M, Zuckerman IH, Fink JC,

Perencevich EN. The use and interpretation of quasi-

experimental studies in infectious diseases. Clin Infect Dis 2004;

38:1586-91.

3. Eveillard M, Eb F, Tramier B, Schmit JL, Lescure FX, Biendo M, et al.

Evaluation of the contribution of isolation precautions in prevention

and control of multi-resistant bacteria in a teaching hospital. J Hosp

Infect 2001;47:116-24.

4. Kola A, Holst M, Chaberny IF, Ziesing S, Suerbaum S, Gastmeier P.

Surveillance of extended-spectrum b-lactamase-producing bacteria

and routine use of contact isolation: experience from a three-year

period. J Hosp Infect 2007;66:46-51.

5. Gardam MA, Burrows LL, Kus JV, Brunton J, Low DE, Conly JM, et al.

Is surveillance for multidrug-resistant enterobacteriaceae an effective

infection control strategy in the absence of an outbreak? J Infect Dis

2002;186:1754-60.

6. Thouverez M, Talon D, Bertrand X. Control of Enterobacteriaceae

producing extended-spectrum b-lactamase in intensive care units: rec-

tal screening may not be needed in non-epidemic situations. Infect

Control Hosp Epidemiol 2004;25:838-41.

7. Rogues AM, Boulard G, Allery A, Arpin C, Quesnel C, Quentin C,

et al. Thermometers as a vehicle for transmission of extended-

spectrum-b-lactamase producing Klebisiella pneumoniae. J Hosp

Infect 2000;45:76-7.

8. Demir S, Soysal A, Bakir M, Kaufmann ME, Yagci A. Extended-

spectrum b-lactamase-producing Klebsiella pneumoniae in paediatric

wards: a nested case-control study. J Paediatr Child Health 2008;44:

548-53.

9. Conterno LO, Shymanski J, Ramotar K, Toye B, Zvonar R, Roth V.

Impact and cost of infection control measures to reduce nosoco-

mial transmission of extended-spectrum b-lactamase-producing

organisms in a non-outbreak setting. J Hosp Infect 2007;65:

354-60.

10. Johnson PD, Martin R, Burrell LJ, Grabsch EA, Kirsa SW,

O’Keeffe J, et al. Efficacy of an alcohol/chlorhexidine hand hygiene

program in a hospital with high rates of nosocomial methicillin-

resistant Staphylococcus aureus (MRSA) infection. Med J Aust

2005;183:509-14.

11. Souweine B, Traore O, ublet-Cuvelier B, Bret L, Sirot J, Laveran H,

et al. Role of infection control measures in limiting morbidity

associated with multi-resistant organisms in critically ill patients.

J Hosp Infect 2000;45:107-16.

12. Soulier A, Barbut F, Ollivier JM, Petit JC, Lienhart A. Decreased trans-

mission of Enterobacteriaceae with extended-spectrum b-lactamases

in an intensive care unit by nursing reorganization. J Hosp Infect

1995;31:89-97.

13. BoucherHW,TalbotGH,Bradley JS, Edwards JE,GilbertD, Rice LB, et al.

Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases

Society of America. Clin Infect Dis 2009;48:1-12.