Relationship between shared patient care items and healthcare-associated infections: A systematic review

International Journal of Nursing Studies 52 (2015) 380–392

Contents lists available at ScienceDirect

International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns

Review

Relationship between shared patient care items and healthcare-associated infections: A systematic review Ilana Livshiz-Riven a,b,*, Abraham Borer b,c, Ronit Nativ b, Seada Eskira b, Elaine Larson d a

Department of Nursing, Recanati School for Community Health Professions, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel Infection Control and Hospital Epidemiology Unit, Soroka University Medical Center, Beer-Sheva, Israel c Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel d Center for Interdisciplinary Research to Prevent Infections, School of Nursing, Columbia University, New York, NY, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 14 December 2013 Received in revised form 31 May 2014 Accepted 4 June 2014

Background: Environmental surfaces may contribute to transmission of nosocomial pathogens. Noninvasive portable clinical items potentially shared among patients (NPIs) are part of the patient’s immediate surroundings and may pose a threat of pathogen transmission. Objective: To assess the body of literature describing the range of microorganisms found on NPIs and evaluate the evidence regarding the potential for cross-transmission of microorganisms between NPIs and hospitalized patients in non-outbreak conditions. Design: A comprehensive list of NPIs was developed, and a systematic review of these items combined with healthcare-associated infection related keywords was performed. Data sources: PubMed, Scopus, and Cochrane Library. Review methods: A systematic review was performed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist to identify and synthesize research reports published between January 1990 and July 2013 on studies regarding contamination of NPIs and association to infections in non-outbreak circumstances. Results: 1498 records were scanned for eligibility. Thirteen studies met inclusion criteria. Overall, rates of NPI contamination ranged from 23% to 100%. Normal skin or environmental flora were found on almost all positive cultures. Potential pathogens, e.g., Staphylococcus aureus, were present on up to 86%, and Pseudomonas spp. and/or Enterobacteriaceae in 38% of positive cultures. Multi-drug resistant organisms were isolated from up to 25% of items. Three studies explored association between NPIs contamination and patient colonization and infection. One study reported 5 patients with healthcare-associated infections with pathogens found concurrently on NPIs, one found cross-transmission between patient skin bacteria and NPI contamination, and a third did not find any cross-transmission. Conclusions: Potential pathogens and multiply resistant organisms present on NPIs in routine, non-outbreak conditions and in a variety of settings confirms the need to improve NPIs decontamination practices. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: Cross-infection Colonization Contamination Equipment and supplies Hospital Non-critical items

* Corresponding author at: Department of Nursing, Recanati School of Community Health Professions, Ben-Gurion University of the Negev, P.O.B. 653, Beer-Sheva 84105, Israel. Tel.: +972 8 6477572; fax: +972 8 6477683. E-mail address: [email protected] (I. Livshiz-Riven). http://dx.doi.org/10.1016/j.ijnurstu.2014.06.001 0020-7489/ß 2014 Elsevier Ltd. All rights reserved.

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

What is already known about the topic?  The hospital environment poses a risk of infection to susceptible patients.  Pathogenic microorganisms are frequently found on non-invasive clinical items in an outbreak condition.  Non-invasive portable clinical items are frequently shared among patients and often assumed to be clean unless visibly soiled. What this paper adds  This is the first systematic review to evaluate the evidence of the potential role of noninvasive portable clinical items potentially shared among patients (NPIs) in transmission of healthcare-associated infection in non-outbreak conditions.  NPIs harbored a substantial percentage of potential pathogens and multiple resistant organisms, e.g., Staphylococcus aureus, were present on up to 86%, and Gramnegative organisms (Pseudomonas spp. and/or Enterobacteriaceae) were present on up to 38% items. Multidrug resistant organisms were isolated from up to 25% of the items.  Reevaluation of the current recommendations and practices for cleaning and decontamination of NPIs is clearly needed.

381

2004). Yet, the environmental cleaning of noncritical items is far from optimal (Carling, 2013). The consequence of suboptimal cleaning or non-adequate decontamination of such ‘‘non-critical’’ equipment between patient contacts may still be unclear to many care providers. Furthermore, it is often unclear whose responsibility it is to clean and decontaminate the NPIs, given other priorities of the staff (Dancer, 2011). Unlike other non-critical items such as patient unit furniture, floors, and curtains, which are typically cleaned by environmental services/cleaning personnel or sent to a central location for cleaning, NPIs are typically expected to be cleaned at the point of care between patients (Rutala et al., 2008) by nursing aides or nurses. The objective of this systematic review was to identify the potential role of NPIs on risk of healthcare-associated infections, in patient care settings. The specific aims were to assess the body of literature describing the range of microorganisms found on NPIs and to evaluate the evidence regarding the potential for cross-transmission of microorganisms between NPIs and hospitalized patients. 2. Methods To accomplish these aims, it was first necessary to develop a comprehensive list of NPIs and then to conduct a systematic review which included these items.

1. Background 2.1. Development of a comprehensive list of NPIs The hospital environment poses a risk of infection to susceptible patients, as there are numerous reports of pathogens and multi-drug resistant organisms that contaminate patient care environmental surfaces (Boyce, 2007; Drees et al., 2008; Goodman et al., 2008; Rutala et al., 2006). In a recent literature review Otter et al. (2011) summarized the accumulating evidence that environmental surfaces contribute to the epidemic and endemic transmission of numerous nosocomial pathogens such as Clostridium difficile, vancomycin-resistant Enterococci, Pseudomonas aeruginosa, and Norovirus. There is agreement among infection control professionals, researchers, and policy makers that environmental factors play an important role in the risk of cross-infection (Weber and Rutala, 2013). Hence, environmental cleaning and decontamination of items used for noninvasive patient care (non-critical items) in the clinical setting are important measures to be taken in order to reduce the risk of healthcare-associated infections (Adams et al., 2008; Carling and Huang, 2013; Rutala et al., 2008; WHO, 2004). Noninvasive portable clinical items potentially shared among patients (NPIs) such as blood pressure cuffs, glucometers, and thermometers, are part of a larger category of clinical items that come in contact with patients’ intact skin or immediate surroundings and should be cleaned between patients (Rutala et al., 2008). Cleaning policies and procedures in various levels of detail are published in international, national, and local guidelines (Adams et al., 2008; Brady et al., 2011; Ministry of Health, 2006; National Patient Safety Agency, 2007; Rutala et al., 2008; Soroka University Medical Center, 2012; WHO,

To develop a comprehensive list of noninvasive portable items potentially shared among patients, the following inclusion criteria were used: (1) portable noninvasive clinical items that may contact patients or their immediate surroundings in the care setting; (2) cleaning process assumed to take place in the clinical unit/ ward; (3) item not assigned to one particular patient during their hospitalization; and (4) item belongs to the unit/ward and not personally owned by the health care provider. Exclusion criteria were: (1) items allocated to a specific patient for the duration of their hospitalization; (2) furniture and items firmly fixed/installed in patient units; and (3) items processed for cleaning outside the clinical setting. Nine healthcare professionals – eight registered nurses and infection preventionists, four of whom combined their work in infection control with part-time bedside nursing (medical ward, neonatal intensive care unit, and neurosurgery ward); and one physician hospital epidemiologist – were asked to respond to the following question: ‘‘What non-sterile portable items potentially shared among patients in a hospital setting can you identify?’’ Additionally, a literature search in PubMed for NPIs was performed with consultation of an information specialist. The following Medical Subject Headings (MeSH) were used: ‘‘Durable Medical Equipment’’ or ‘‘Electrical Equipment and Supplies’’ or ‘‘Equipment and Supplies, Hospital’’ under the ‘‘Equipment and Supplies’’ MeSH tree or ‘‘Fomites’’, combined with ‘‘Patient’’ and ‘‘Risk’’ and ‘‘infection’’ or ‘‘colonization’’, ‘‘noncritical item’’ or ‘‘non

382

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Table 1 List of noninvasive portable items potentially shared between patients (NPIs) and the number of records mentioning that item retrieved in the systematic literature review.a Part 1

Part 2

Category

Identified NPI names and synonyms

Number of records retrieved

Call button Feeding pump Glucometer Infusion stand/Pump Mobility aids

Portable call button/bell Portable Kangaroo Enteral Feeding Pump/Breast pumps Glucometer Portable IV stand/Portable IVAC/IV compression bag Walker/Crutches/Bathroom wheelchairs/wheelchairs/ Commodes Portable ECG/EKG/ECG wires Defibrillator/EEG/EEG wires Bi-level positive airway pressure (BPAP) Continuous positive airway pressure (CPAP; the machine). Water-sealed spirometer/pneumotachograph. Pulse oximeter Oxygen flow meter Pressure gauge cuff inflator for intubated patients Ophthalmoscopes/Otoscope Blood pressure meter/sphygmomanometer/Blood pressure cuff Digital thermometers PO/PR/PA/Tympanic Procedure tray Ultra Sound Transducers handle/Probe Tourniquet

40 17 83 82 51

Monitor/EKG/ECG Non-invasive respiration machine and spirometers Oximeter Oxygen flow meter Pressure gauge cuff inflator Scopes for external use Sphygmomanometer Thermometer Tray Ultrasound transducers Venipuncture tourniquets Total: 16

Number of studies found in systematic reviewb

1

525 25

1 1

142 24 14 139 133

1*

95 44 64 20 1498

7#* 2#* 2 1 8 item categories, 13 studies

a Part 1 – a list of noninvasive portable items potentially shared between patients (NPIs) identified by healthcare professionals (n = 9), and PubMed search (578 titles), arranged in categories. Part 2 – number of records mentioning that item retrieved from PubMed, Scopus, and Cochran Library. b *#Number of records included in this table exceeds the 13 articles included in the final systematic review since two studies evaluated more than one type of item: (Uneke and Ijeoma, 2011; Havill et al., 2011).

critical item’’ keywords, filtered for humans, and terms mentioned below. This review yielded 578 titles, all of which were screened for names of equipment that met the definition of NPIs, resulting in a list of commonly used NPIs in hospital settings (Table 1, part 1) aggregated in 16 categories. These 37 item names and/or synonyms and the 16 categories were the basis for the systematic review. 2.2. Systematic review 2.2.1. Search strategy We performed an electronic database literature search with the consultation of an information specialist. York University Center for reviews and dissemination guidance (CRD, 2009) and the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Statement (Moher et al., 2009) were used to ensure systematic, transparent, and complete reporting. The databases searched included PubMed, Scopus, and Cochran Library. Medical Subject Headings (MeSH) were used and additional terms added. Every identified NPI/ NPI category (Table 1, part 1) was explored as a search word, combined with the following terms in separate search protocols: ‘‘contamination’’ or ‘‘infection’’, ‘‘cross infection’’. Additional searches were performed with the same NPIs and terms ‘‘hospital’’ and ‘‘patient’’. The search terms were chosen with the focus on clinical studies exploring the potential role of NPIs in healthcareassociated infections.

2.2.2. Selection criteria Inclusion criteria were: (1) study reported evaluating NPI contamination and/or association to infections; (2) the item of interest was not a component of invasive equipment; (3) study published in a peer review journal, in English language, with title and abstract available online; (4) publication between January 1990 and July 2013. The decision to begin the systematic literature review in 1990 was made in light of the dramatic changes in health care systems since the end of the 1980s with respect to technology, the risks of transmission of multidrug resistant organisms, and the general approach to standard precautions since the emergence of acquired immunodeficiency syndrome (AIDS). Exclusion criteria were: (1) outbreak investigation, since the aim was to study endemic situations; (2) viral contamination; (3) studies to discover a specific pathogen, e.g., related to a previous outbreak; (4) outcome measure was a surrogate marker of contamination, e.g., ATPbioluminescence assay, invisible ink/dark light, blood stain detection; and (5) review articles. Studies were selected in three stages to minimize the risk of errors and to ensure that all relevant articles were included: (1) identification, the initial results of the literature search; (2) screening, by evaluating the titles and abstracts for inclusion/exclusion criteria; and (3) eligibility, evaluating the full text of articles that seemed appropriate on the basis of the screening. EndNote X6 software was used for bibliographic management. Two

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

383

Table 2 Quality assessment criteria for relevant studies. Criterion number

Criterion

Contenta

1

Outcome definition

2 3 4 5

Time unit Target items Number of items Description of screening procedure

6

Description of laboratory procedure

Was a valid definition given of the outcome for prevalence, contamination, or infection? Was the endpoint calculated for a standardized time unit (daily, monthly, yearly)? Were the items specified by inclusion or eligibility criteria? Was the number of items examined stated? Was the swabbing process elaborate, clear, and consistent? e.g., Who did it, how were they prepared for the swabbing? Was there a description of swabbing sites on the items?b Was the lab work to produce bacteria detailed, clear, and consistent? e.g., did it describe the media used, time and temperature of incubation?

Adapted from Dulon et al. (2011) and von Elm et al. (2008). a Answer choices and scoring: Yes, No, Not clear. 1 for Yes, 0 for not clear. b If explicit training of personnel collecting data was noted this item received 2 points.

reviewers (ILR, EL) assessed study eligibility. First, ILR independently screened abstract titles, which EL reviewed and confirmed eligibility. Differences in eligibility assessments were resolved by discussion and consensus between reviewers. When necessary, the complete article was reviewed to determine eligibility. 2.2.3. Primary outcomes The primary outcomes of the search were microbiologic results (types and quantity of bacterial contamination) and concomitant patient colonization or infection with the same organism. 2.2.4. Assessment of methodological and reporting quality The quality of included articles was assessed by using a checklist with possible scores of 0–7 (Table 2) adapted and modified from a systematic review on methicillin-resistant Staphylococcus aureus (MRSA) prevalence (Dulon et al., 2011) and strengthening the reporting of observational studies in epidemiology (STROBE) checklist (von Elm et al., 2008). Initially six articles were assessed independently by two reviewers (EL and ILR), with a 94% agreement on scores; one reviewer evaluated the remainder of the studies. 2.2.5. Data extraction, abstraction, and synthesis Data were extracted systematically using a standard tool consisting of: items evaluated, country where the study was conducted, hospital type (tertiary care, community, teaching, or other), setting/ward, patient population (if applicable and available), objective, study design, sample size of items, number of patients (if applicable), outcome measure, and results. Quality assessment scores were converted to percentages. The descriptive design of the studies selected, the heterogeneity in outcome measures, and variations in laboratory methods precluded the possibility of a meta-analysis; studies were therefore categorized and described in a narrative synthesis. 3. Results

titles were screened for inclusion criteria; of these, 1027 were excluded based on title screening and abstract review (Fig. 1). The reasons for exclusion were: articles unrelated to the research aims, non-English-language, non-research articles, product descriptions, use of surrogate markers, process assessment, outbreak investigations, and use of the item for invasive procedures. After applying these criteria, 63 articles remained. These articles were then evaluated for potential eligibility based on a more detailed full-text review. Fifty additional articles were excluded after full-text review because they met a priori exclusion criteria. Interventional studies were included only if baseline qualitative or quantitative bacteriological data were presented. The full search resulted in 13 remaining studies. 3.2. Characteristics of studies Of the 13 studies included in the current analysis (Table 3), 46% (n = 6) were conducted in the United States and 38% (n = 5) in European countries; one study was conducted in Nigeria and another in Australia. Ninety-two percent (n = 12) of the studies were single-centered and 46% (n = 6) used descriptive cross-sectional designs. A smaller proportion of the studies used prospective (31%, n = 4) or prospective sequential (15%, n = 2) designs. No published studies were found between 1990 and 1995. Three studies were published in 1996, followed by a hiatus of 10 years until 2006, then a single publication every subsequent year until 2011 when 5 (38%) studies were published. Features and main findings of the studies are reported in Table 3 together with quality assessment scores. The average quality score was 84% (range: 64–100%). Of the 16 NPI categories (Table 1), the published studies explored 8 (50%). The most frequently studied item was BP cuffs (n = 7 studies), followed by thermometers and ultrasound probes (2 studies each), and one study each of pulse-oximeters, tourniquets, ECG lead wires, patient controlled analgesia and epidural infusion pumps, and water-sealed spirometer and pneumotachograph.

3.1. Study selection

3.3. Laboratory and sampling processes

The electronic database searches recovered 1498 records. After removal of duplicates, 1090 abstracts and

Ten studies of the 13 (77%) utilized sterile swabs moistened with physiological saline or brain heart infusion

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Identification

384

Records identified through database search (PubMed, Scopus, Cochrane Library) N = 1,498

Records removed due to duplications N = 408

Titles and abstracts screened for eligibility N = 1,090 Screening

• • • • • •

Eligibility

• •

Records excluded based on title or abstract information N =1,027 Unrelated topics = 782 No Abstract/letter/expert opinion/nonresearch article = 77 Product description = 7 Involves invasive procedures = 19 Surrogate markers/process assessment = 28 Outbreak investigation/specific microorganisms = 24 Reviews of related topics = 9 Non-English = 81

Full-text articles assessed for eligibility N = 63 •

Included

• • Articles eligible for final inclusion in this review N = 13

•

Full-text articles excluded, based on inclusion/exclusion criteria N=50 Process assessment/Surrogate markers = 18 Involves invasive procedures/requires semi-critical or critical item care approach = 4 Targeting specific microorganisms or related to an outbreak = 27 Item assigned to a patient for the length of hospitalization (e.g., computer keyboard and mouse) = 1

Fig. 1. PRISMA flow diagram; study selection process (Moher et al., 2009).

broth to sample surfaces. One study (8%) sampled items by collecting each in a sterile bag and subsequent immersion in broth (Pinto et al., 2011); two studies (15%) used direct contact with agar plates, sterile neutralizing agar plates (Bhanot et al., 2011), or blood agar culture plates (Frazee et al., 2011). All the studies (100%) described non-selective media used for the initial inoculation. Eight studies (62%) described using blood agar plates and 6 studies (46%) used more than one type of medium for the initial inoculation. Of the 6 studies using only non-selective media for initial inoculation (46%, 6/13) 4 used sub-cultures to selective media (67%, 4/6). Two of the studies did not describe the selective media (Bhanot et al., 2011; Frazee et al., 2011), one (25%, 1/4) used oxoid chromogenic MRSA agar (Baruah et al., 2008), and one (25%, 1/4) sub-cultured as indicated onto: salt mannitol agar and MacConkey agar, Agglutination (for Streptococci), coagulase (for Staphylococci), or API (Analytical Profile Index), and identification system strip for Gram-negative bacteria (Patterson et al., 1996).

Six studies (46%, 6/13) used a quantitative method to describe the overall level of bacterial contamination load on NPIs. 3.4. Microbiologic results Overall NPI contamination rates ranged from 23% to 100%, but the different criteria used to define item contamination made it impossible to compare microbiologic results across studies. Similarly, laboratory methods to characterize microorganisms varied across studies. Some reports aggregated ‘‘less clinically important pathogens’’/low risk/normal skin flora and ‘‘more clinically important pathogens’’ (Albert et al., 2010; Frazee et al., 2011), and others reported specific microorganisms (Baruah et al., 2008). Therefore, we aggregated the microbial results into three groups (Table 3) and in each group the NPI categories are presented with the corresponding microorganisms’ growth rates.

Table 3 Selected studies’ characteristics, outcome measures, results and quality rating. Study, year, country, & item Objectivea

Bhanot et al. (2011),a USA BP cuffs

Study designb Setting

To assess the effect of a Intervention disposable barrier for the BP cuff

No. of cultures/ Outcome measures items (& patients)c

Teaching hospital; 101/5 (NA) 60-bed medical floor

1 cfu/cm2

Descriptive Teaching hospital, 36/36 thermometers cross-sectional all the hospital 58/58 BP cuffs units

3 cfu/plate

Frazee et al. (2011),a USA US probe

To assess item microbial contamination [and cleaning]

164/3 Descriptive Urban county cross-sectional teaching hospital; ED

Prevalence of microbial contamination and species

Havill et al. (2011),f USA BP units Thermometers, & Pulse-oximeter

To assess item cleanliness

Descriptive 500-bed, cross-sectional universityaffiliated hospital; med-surg wards

100/26

ACCs and potentially pathogenic bacteria

86

86

86

71

385

At baseline: 100% contamination. Mean (SD) 14.62 (11.76) cfu/cm2. Low risk: CoNS 93.06%. Clinically important: MSSA 3.96%; Strep. 2.97%; Enterococcus 1.98%; Acinetobacter 0.99%; Klebsiella 0.99%. MDOs: MRSA 1.98%. Thermometers and BP cuffs 62.1%, 82.1%, respectively, contaminated. Clinically important: S. aureus 73.9%, E. coli 8.7%, P. aeruginosa 4.4%, E. faecalis 13.0% (of 23 BP isolates); and S. aureus 86.1%, P. aeruginosa 8.3%, E. faecalis 5.6% (of 36 thermometer isolates). MDOs: E. coli with highest resistance. S. aureus was resistant to the beta-lactams. At baseline: Low risk: 67.68% (95% CI 60–74%). Clinically important: 1.2% (95% CI 0.3–4.3%); 1 grew 4 cfu of MSSA, and 1 grew 3 cfu of Acinetobacter lwoffii. Median ACCs ranged from 2 to 53, and total ACCs varied significantly from 0 to too numerous to count (p < .001). MROs: MRSA 6% (2 control buttons, 1 thermometer, 1 blood pressure cuff, 1 machine handle, and 1 pulse oximeter).

Quality assessment scoree

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

To assess item Uneke and Ijeoma contamination, (2011), Nigeria Thermometers and BP cuffs association to use, and microbial susceptibility

Resultsd

386

Table 3 (Continued ) Study designb Setting

No. of cultures/ Outcome measures items (& patients)c

Resultsd

Quality assessment scoree

Pinto et al. (2011), Australia To assess the Venesection tourniquets prevalence of MRO colonization on items

Prospective

100/100

Prevalence. Low risk, significant pathogenic potential and MRO

100

320/320

Item contamination Bacterial species

78% contaminated. Low risk: Only CoNS/Bacillus spp.17%; CoNS 13%; Bacillus spp. 54%. Clinically important: Gram-positive organisms (MSSA or Enterococcus spp.) 10%. Gram-negative organisms (Pseudomonas spp. and/or Enterobacteriaceae) 38%. MROs: 25% of items. MRSA 14%; VRE 19%; both MRSA and VRE from 9 items. IMP-4 MBL-positive Enterobacter cloacae and an ESBL-positive E. cloacae on 1 item each. 62.8% contaminated (49–80%). Clinically important: 37.8% of cultures (28.8–43.8%) including Acinetobacter lwoffii, S. haemolyticus, E. faecium, S. aureus, and E. faecalis. Two cultures were positive for the presence of fungi: Stemphylium and Penicillium. MROs: 9 species (non-S. aureus or Enterococci) were resistant to all penicillin or tetracycline antibiotics.

Albert et al. (2010), USA ECG lead wires

503-bed urban teaching hospital. General, ambulatory, and critical care areas

Descriptive 3 community/ To determine cross-sectional urban teaching bacterial, fungal, and hospitals, 1 non MRO contamination on teaching items and associated community sites hospital: ICU, Tele, ED, and OR

86

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Study, year, country, & item Objectivea

To assess bacterial load on items

Prospective

Hospital. 7 wards

112/13

Item contamination

Baruah et al. (2008), UK BP cuffs

To investigate item contamination

Prospective

University teaching 100/NA hospital; med-surg, pediatric, ICU

Item contamination cfu/plate

De Gialluly et al. (2006), France BP cuffs

Prospective To investigate the potential role of items in the spread of bacterial infections in hospitals

A university hospital. Units: 5 surgical, 7 medical, 2 ICUs, 2 pediatric, 1 ED, and 1 OR

203/NA (5)

Contaminated 100 cfu/25 cm2; highly contaminated 300 cfu/25 cm2. MRO. Concomitant nosocomial infection

100 45% of handsets and 46% of keypads contaminated. Low risk: All the positive cultures yielded commensal skin bacteria except for one (Aeromonas hydrophyla). 86 97% of inner sides and 89% of outer sides were contaminated. 29% of inner sides had >100 cfu compared to 8% of outer surface. Low risk: 67% CoNS; 25% Diphtheroids; Clinically important: 24% Coliforms; 30% S. aureus. MROs: 5% of S. aureus were MRSA. 86 45% (92/203) of inner sides and 23% (46/203) of outer sides were contaminated or highly contaminated. Low risk: Most bacterial colonization of BP cuffs was skin flora (CoNS and Coryneform bacteria). Clinically important: 27/203 (13%) BP cuffs yielded 30 types of pathogenic bacteria; S. aureus on 20/27 (74%). MROs: 45% of S. aureus were MRSA (9/20). Patient cross contamination/ infection: 5 patients were diagnosed with nosocomial infection due to bacterial species isolated from the BP cuff.

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Rothwell et al. (2009), UK Infusion pumps: PCA, epidural

387

388

Table 3 (Continued ) Study, year, country, & item Objectivea

Walker et al. (2006), UK BP cuffs

Burgos et al. (1996), Spain Water-sealed spirometer and pneumotacho-graph

No. of cultures/ Outcome measures items (& patients)c

Resultsd

Quality assessment scoree

TVC, cfu/100 cm2. Potential pathogens

86 100% viable organisms from samples. Range 2 1000 to >25,000 cfu/100 cm . Clinically important: 58% (14/24) of cuffs with 18 isolates. MSSA 33% (8/24). MROs: MRSA 8% (2/24), and C. difficile 33% (8/24) of cuffs. 64 Bacterial 81% (57/70) colonization. 707-bed, level-one 140/70 To determine bacterial Descriptive colonization Low risk: 100% of cross-sectional trauma center. OR, colonization, and colonized items grew MICU, SICU, BSICU, organic or inorganic CoNS, Corynebacterium CICU, ER, PACU, contamination (diphtheroids), Neisseria NSICU on items spp., Propionibacterium, Bacillus species, and nonhemolytic Streptococci. Clinically important: 1cuff (1.4%) was colonized with S. aureus. Prospective Respiratory 40/NA water-sealed Bacterial identification 90% (36/40) samples from To determine (1) the 86 water-sealed spirometers laboratory spirometer 30/NA and patient crossrate of colonization on sequential were colonized; 19 pneumotacho-graph transmission. items; and (2) the samples (48%) yielded (54) item’s role as reservoir counts 104 cfu/mL 1. and in transmission 13% (4/30) of samples from the pneumotachograph were colonized. Low risk: 62% isolates were Penicillium spp.; 32% Pseudomonas fluorescens. Clinically important: 48% Burkholderia cepacia. Patient crosscontamination/ infection: No transmission sequence of potentially pathogenic microorganisms could be demonstrated. To assess item bacterial contamination

Descriptive Medical and cross-sectional surgical wards

24/NA

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Base-Smith (1996), USA BP cuffs

Study designb Setting

General, urban hospital; perinatal testing center

NA (191)

Cross-transmission item-patient, skin flora

60% (105/175) of the US probes after abdominal examination of women with positive abdominal skin cultures were positive (before probes were wiped). 92% (175/191) of women had positive abdominal skin cultures: 18% (35/191) with potentially pathogenic organisms, and 73% with low virulence organisms (e.g., Corynebacterium spp. and CoNS). Patient cross contamination/ infection: 100% concordance between patients’ cultures and the isolates.

71

All abbreviations in alphabetic order: ACCs, aerobic colony counts; BP, blood pressure cuffs; BSICU, burn special intensive care unit; cfu, colony forming units; CICU, cardiac intensive care unit; ECG, electrocardiograph; ESBL, extended-spectrum beta-lactamase; ED, emergency department; ER, emergency room; ICUs, intensive care units; IMP-4 MBL, IMP-4 gene metallo-b lactamase; MICU, medical intensive care unit; MRO, multi-drug resistant organism; MSSA, methicillin-sensitive Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus aureus; NA, not available; NSICU, neurosurgical intensive care unit; OR, operating rooms; PACU, post-anesthesia care unit; PCA, patient-controlled analgesia; SICU, surgical intensive care unit; Tele, intermediate or telemetry care areas; TVC, total viable count; US, ultrasonographic probe; VRE, vancomycin-resistant Enterococcus. a Studies using interventional design to evaluate the effectiveness of new decontamination or other barrier methods to reduce contamination were included only if data on the bacteriological analysis were presented under the routine condition. b Objectives of studies aiming to explore new barrier/decontamination methods are in [] and these parts of the study results are not presented. c If applicable. d Results are presented in 3 parts: (1) overall level of contamination; (2) bacterial species by risk category, e.g., Low risk, Clinically important, and multi-drug resistant organisms (MROs); and (3) relationship to patients (if applicable). e Maximum 100% calculated of 7 possible points. f The Havill et al. (2011) study utilized a surrogate marker in addition to cultures; this was not presented here.

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Prospective Patterson et al. (1996),a USA To determine the sequential US probe/Transducer head (1) rate of bacterial study isolation from the abdomen of women having obstetric US; (2) the rate of bacterial spread to item; and (3) the eradication rate after routine wiping of item

389

390

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

 Normal skin or environmental flora/low risk microorganisms/commensal skin bacteria Growth of low risk microorganisms was found on the following NPI categories: (1) 100% of infusion stands and pumps (Rothwell et al., 2009); (2) 62.8% of ECG lead wires (Albert et al., 2010); (3) sphygmomanometers, thermometers, and oximeters comprised one NPI category. The results ranged from 100% of positive cultures yielding normal flora (Base-Smith, 1996), through 93.06% (Bhanot et al., 2011), 82.1%, 62.1% (Uneke and Ijeoma, 2011), to 45% and 23% (De Gialluly et al., 2006) of all collected cultures; (4) 67.7% (Frazee et al., 2011) and 60% (Patterson et al., 1996) of ultrasound transducers; and (5) 54% of venipuncture tourniquets (Pinto et al., 2011).  Potential pathogens/clinically important bacteria The recovery of potential pathogens arranged by NPI categories include the following: (1) infusion stands/ pumps did not yield potential pathogenic bacteria (Rothwell et al., 2009); (2) 37.8% of ECG lead wire cultures grew potential pathogens such as Acinetobacter lwoffii, Staphylococcus haemolyticus, Enterococcus faecium, S. aureus, and E. faecalis (Albert et al., 2010); (3) studies reporting on sphygmomanometer, thermometer, and oximeter cultures yielded a variety of microorganisms including: Methicillin-sensitive S. aureus in 33% (Walker et al., 2006) and 4% (Bhanot et al., 2011), Streptococcus in 3% and Enterococcus in 2%, Acinetobacter 1%, and Klebsiella 1% of cultures (Bhanot et al., 2011). Others reported as high as 86.1–73.9% S. aureus, 8.7% E. coli, 4.4–8.3% P. aeruginosa, and 5.6–13.0% E. faecalis (Uneke and Ijeoma, 2011). Baruah et al. (2008) found 24% coliforms and 30% S. aureus. De Gialluly et al. (2006) found 13% of items contaminated with 30 types of pathogenic bacteria and of those S. aureus was retrieved from 74%; (4) 1.2% Methicillin-sensitive S. aureus and A. lwoffii were found on ultrasound transducers (Frazee et al., 2011); (5) Methicillin-sensitive S. aureus or Enterococcus spp. were retrieved from 10% of the cultures, and Gram-negative organisms (Pseudomonas spp. and/or Enterobacteriaceae) from 38% of venipuncture tourniquets (Pinto et al., 2011).  Multi-drug resistant organisms (1) No multi-drug resistant organisms were reported on the infusion stands/pumps (Rothwell et al., 2009); (2) 9 species out of 226 isolates (4%) were resistant to all penicillin or tetracycline antibiotics (Albert et al., 2010) in cultures from ECG lead wires; (3) studies reporting on sphygmomanometer, thermometer, and oximeter cultures yielded Methicillin-resistant S. aureus in 1.98% (Bhanot et al., 2011) and 8% (Havill et al., 2011; Walker et al., 2006) of cultures. De Gialluly et al. (2006) reported that 45% of S. aureus were Methicillin-resistant. C. difficile was found on 33% of items in Walker et al.’s (2006) study; (4) no multi-drug resistant organisms were reported on ultrasound transducers; (5) 25% of the items were contaminated with multi-drug resistant organisms in the venipuncture tourniquets study. Methicillin-resistant S. aureus was retrieved from 14% of cultures, Vancomycin-resistant Enterococcus from

19%, Metallo-b lactamase IMP-4-positive Enterobacter cloacae, and extended-spectrum beta-lactamase-positive E. cloacae were each found in 1% of cultures (Pinto et al., 2011). Water-sealed spirometers and pneumotachographs were contaminated with a variety of bacterial species known to colonize water (Burgos et al., 1996). 3.5. Evidence of cross-contamination between NPIs and patients Three studies examined possible cross-contamination between NPIs and patients (Burgos et al., 1996; De Gialluly et al., 2006; Patterson et al., 1996). One assessed the transmission of microorganisms from patients to NPIs (Patterson et al., 1996), reporting that 60% (105/175) of ultrasound probes after abdominal examination of women with positive abdominal skin cultures were positive, all with the same organisms isolated from the patient. The study reported by Burgos et al. (1996) evaluated possible cross-transmission of microorganisms between watersealed spirometer and pneumotachograph and patients’ nasopharynx up to a week following exposure, and no transmission was demonstrated. De Gialluly et al. (2006) investigated the potential role of BP cuffs in cross-infection in a university hospital. Almost half of the cuffs (46%, 92/ 203) were contaminated with 30 strains of bacteria, and healthcare-associated infections among five patients were temporally associated with the same organisms. Although causality was impossible to assert, this study clearly demonstrated transmission of microorganisms between NPIs and patients. 4. Discussion This systematic review synthesizes more than 20 years of research regarding the role of NPIs in the risk of healthcare-associated infections in non-outbreak settings, and provides important data to generate hypotheses for future intervention studies. Guidelines exist for the cleaning of such items, but cleaning practices have been demonstrated to be suboptimal (Anderson et al., 2011; Dancer, 2011). The most common microorganisms found on NPIs in the current systematic review were environmental and low risk. Yet, the presence of potentially pathogenic organisms on more than 80% of items in one study (Uneke and Ijeoma, 2011) and ranging between 1% and 30% in others with similar results in multi-drug resistant organisms are data that cannot be overlooked. Moreover, evidence for transmission of pathogens from NPI to hospitalized patients was also detected. Although the data is limited, it demonstrates the possible association and stresses the need for further investigation. This is clearly a challenging area to study because of the multifactorial nature of healthcare-associated infections, making it difficult to identify a single causative exposure. Although interventional designs to demonstrate a causal role between NPIs and subsequent infections may not be feasible, quasi-experimental studies are likely to confirm that cross-transmission between patient and NPI or vice versa occurs and should be prevented whenever possible.

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

Although NPIs are ubiquitous and frequently shared in the clinical environment, they have been traditionally referred to as ‘‘non-critical items’’, a term used to describe instruments and items that do not require high-level disinfection (Rutala et al., 2008). Based on even this small body of literature, however, the prevalence of potential pathogens and multiply resistant microbial strains on noninvasive portable items is high. Hence, the concept of ‘‘noncritical item’’ is inappropriate and is an unfortunate term that needs to be changed; as long as clinical staff continue to perceive NPIs as ‘‘non-critical’’ the need for cleaning and decontamination is likely to be a low priority, and an item will be considered ‘‘clean’’ if no soiling is visible. Nursing interventions to reduce the risks of pathogen cross-transmission via NPIs should be designed and investigated. These may incorporate assessment of organizational needs or practices in specific institutions or clinical units in order to improve NPI decontamination. Furthermore, a rigorous research design is needed to discover the actual effect of such an intervention on healthcare-associated infection risks. 5. Limitations The current study used only three databases; PubMed, Scopus, and Cochran Library and focused on terms strictly describing NPIs and healthcare-associated infections. Searching other databases and using additional search terms may have identified additional publications. Furthermore, the review of only English-language publications, the accepted international scientific language, resulted in the omission of studies published in scientific journals in other languages. One of the findings and limitations in this review was the sparse number of studies addressing the potential association between the NPIs and patient cross-infection. Consequently, it is challenging to conclude the distinct role of NPIs in healthcare-associated infections. Moreover, variations in laboratory techniques used in the studies reviewed made it impossible to make comparisons across settings or by type of NPI. 6. Conclusions The findings emerging from the current systematic review demonstrate the existence of potential pathogens and multiply resistant organisms on NPIs in routine, nonoutbreak conditions and in a variety of hospital settings. Furthermore, the results suggest a possible relationship between contaminated non-invasive clinical items used for hospitalized patient care and healthcare-associated infections. To confirm a causal relationship, additional high quality research utilizing interventional or quasi-experimental designs is needed and more NPIs should be assessed to determine levels of microbial contamination. However, without an agreed upon cut-off point for contamination levels of non-invasive clinical items and standardized microbiological methods, it will not be possible to compare risks to the hospitalized patient. Additional studies are needed to clarify what might be acceptable levels and types of microbial growth (if any) on

391

NPIs using standardized laboratory protocols. Further, the term ‘‘non-critical item’’ should be revised to better communicate the potential risk associated with any patient care equipment that might be potentially shared. A term such as ‘‘noninvasive clinical item’’ may better describe items that are used in a clinical setting, touching patients’ intact skin and close surroundings, and may carry the potential of cross-transmitting potentially harmful pathogens. Conflict of interest: None declared. Ethical approval: No ethical permission was needed. References Adams, J., Bartram, J., Chartier, Y., 2008. Essential Environmental Health Standards in Health Care. World Health Organization, Geneva. Albert, N.M., Hancock, K., Murray, T., Karafa, M., Runner, J.C., Fowler, S.B., Nadeau, C.A., Rice, K.L., Krajewski, S., 2010. Cleaned, ready-to-use, reusable electrocardiographic lead wires as a source of pathogenic microorganisms. Am. J. Crit. Care 19 (6) e73–e80. Anderson, R.E., Young, V., Stewart, M., Robertson, C., Dancer, S.J., 2011. Cleanliness audit of clinical surfaces and equipment: who cleans what? J. Hosp. Infect. 78 (3) 178–181. Baruah, J., Kumar, S., Gratrix, A., Dibb, W., Madeo, M., 2008. Blood pressure cuffs as a potential fomite for transmission of pathogenic microorganisms: a prospective study in a university teaching hospital. Br. J. Infect. Cont. 9 (4) 19–21. Base-Smith, V., 1996. Nondisposable sphygmomanometer cuffs harbor frequent bacterial colonization and significant contamination by organic and inorganic matter. AANA J. 64 (2) 141–145. Bhanot, N., Rao, S., Sharma, S., Malka, E.S., Ghitan, M., Gupta, P., Sahud, A.G., McCaughey, B., Chapnick, E.K., 2011. Effectiveness and feasibility of using a physical barrier device in reducing rates of microbial contamination of sphygmomanometer cuffs. J. Infect. Prev. 12 (6) 241–245. Boyce, J.M., 2007. Environmental contamination makes an important contribution to hospital infection. J. Hosp. Infect. 65 (Suppl. 2) 50– 54. Brady, R.R., Hunt, A.C., Visvanathan, A., Rodrigues, M.A., Graham, C., Rae, C., Kalima, P., Paterson, H.M., Gibb, A.P., 2011. Mobile phone technology and hospitalized patients: a cross-sectional surveillance study of bacterial colonization, and patient opinions and behaviours. Clin. Microbiol. Infect. 17 (6) 830–835. Burgos, F., Torres, A., Gonza´lez, J., Puig de la Bellacasa, J., Rodriguez-Roisin, R., Roca, J., 1996. Bacterial colonization as a potential source of nosocomial respiratory infections in two types of spirometer. Eur. Respir. J. 9 (12) 2612–2617. Carling, P., 2013. Methods for assessing the adequacy of practice and improving room disinfection. Am. J. Infect. Control 41 (Suppl. 5) S20– S25. Carling, P.C., Huang, S.S., 2013. Improving healthcare environmental cleaning and disinfection: current and evolving issues. Infect. Control Hosp. Epidemiol. 34 (5) 507–513. CRD, 2009. Systematic Reviews. CRD’s Guidance for Undertaking Reviews in Health Care University of York. Dancer, S.J., 2011. Hospital cleaning in the 21st century. Eur. J. Clin. Microbiol. Infect. Dis. 30 (12) 1473–1481. De Gialluly, C., Morange, V., De Gialluly, E., Loulergue, J., Van Der Mee, N., Quentin, R., 2006. Blood pressure cuff as a potential vector of pathogenic microorganisms: a prospective study in a teaching hospital. Infect. Control Hosp. Epidemiol. 27 (9) 940–943. Drees, M., Snydman, D.R., Schmid, C.H., Barefoot, L., Hansjosten, K., Vue, P.M., Cronin, M., Nasraway, S.A., Golan, Y., 2008. Prior environmental contamination increases the risk of acquisition of vancomycin-resistant enterococci. Clin. Infect. Dis. 46 (5) 678–685. Dulon, M., Haamann, F., Peters, C., Schablon, A., Nienhaus, A., 2011. MRSA prevalence in European healthcare settings: a review. BMC Infect. Dis. 11, 138. Frazee, B.W., Fahimi, J., Lambert, L., Nagdev, A., 2011. Emergency department ultrasonographic probe contamination and experimental model of probe disinfection. Ann. Emerg. Med. 58 (1) 56–63. Goodman, E.R., Platt, R., Bass, R., Onderdonk, A.B., Yokoe, D.S., Huang, S.S., 2008. Impact of an environmental cleaning intervention on the presence of methicillin-resistant Staphylococcus aureus and

392

I. Livshiz-Riven et al. / International Journal of Nursing Studies 52 (2015) 380–392

vancomycin-resistant enterococci on surfaces in intensive care unit rooms. Infect. Control Hosp. Epidemiol. 29 (7) 593–599. Havill, N.L., Havill, H.L., Mangione, E., Dumigan, D.G., Boyce, J.M., 2011. Cleanliness of portable medical equipment disinfected by nursing staff. Am. J. Infect. Control 39 (7) 602–604. Ministry of Health, 2006. Guidelines for cleaning and disinfecting in medical institutions ( ). Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., 2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Br. Med. J. 339. National Patient Safety Agency, 2007. The National Specifications for Cleanliness in the NHS: A Framework for Setting and Measuring Performance Outcomes. National Health Service (NHS), UK. Otter, J.A., Yezli, S., French, G.L., 2011. The role played by contaminated surfaces in the transmission of nosocomial pathogens. Infect. Control Hosp. Epidemiol. 32 (7) 687–699. Patterson, S.L., Monga, M., Silva, J.B., Bishop, K.D., Blanco, J.D., 1996. Microbiologic assessment of the transabdominal ultrasound transducer head. South. Med. J. 89 (5) 503–504. Pinto, A.N., Phan, T., Sala, G., Cheong, E.Y., Siarakas, S., Gottlieb, T., 2011. Reusable venesection tourniquets: a potential source of hospital transmission of multiresistant organisms. Med. J. Aust. 195 (5) 276–279. Rutala, W.A., Weber, D.J., HICPAC, 2008. Guideline for Disinfection and Sterilization in Healthcare Facilities. Centers for Disease Control and prevention, US.

Rothwell, M., Pearson, D., Wright, K., Barlow, D., 2009. Bacterial contamination of PCA and epidural infusion devices. Anaesthesia 64 (7) 751– 753. Rutala, W.A., White, M.S., Gergen, M.F., Weber, D.J., 2006. Bacterial contamination of keyboards: efficacy and functional impact of disinfectants. Infect. Control Hosp. Epidemiol. 27 (4) 372–377. Soroka University Medical Center, 2012. Work guide for environmental and human cleaning in the hospital setting ( ). Clalit, Organization of Health Services & Soroca University Medical Center, Internal Guidelines, Beer Sheva. Uneke, C.J., Ijeoma, P.A., 2011. The potential for transmission of hospitalacquired infections by non-critical medical devices: the role of thermometers and blood pressure cuffs. World Health Popul. 12 (3) 5–12. von Elm, E., Altman, D.G., Egger, M., Pocock, S.J., Gotzsche, P.C., Vandenbroucke, J.P., 2008. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J. Clin. Epidemiol. 61 (4) 344–349. Walker, N., Gupta, R., Cheesbrough, J., 2006. Blood pressure cuffs: friend or foe? Journal of Hospital Infection 63 (2) 167–169. Weber, D.J., Rutala, W.A., 2013. Introduction: understanding and preventing transmission of healthcare-associated pathogens due to the contaminated hospital environment. Infect. Control Hosp. Epidemiol. 34 (5) 449–452. WHO, 2004. Environmental management practicesIn: Practical Guidelines for Infection Control in Health Care Facilities. .