A randomized-crossover trial to decrease bacterial contamination on hospital scrubs

Abstract

Healthcare worker attire may become contaminated with pathogenic organisms during a normal shift. We performed a randomized-crossover study to assess whether treatment with an antimicrobial coating would decrease bacterial contamination on scrubs. Thirty percent of all scrubs were contaminated; there was no difference in the rate of contamination between intervention/control.

Introduction

Healthcare worker (HCW) attire is regularly contaminated during the course of a typical shift and has the potential to serve as an important bacterial reservoir, leading to healthcare-associated infections (HAIs). Up to 40% of HAIs can be attributed to cross-contamination from HCWs who have become contaminated in the course of their daily work.¹

We performed a blinded, randomized-crossover trial to assess the efficacy of a novel antimicrobial treatment, Chitosan/DMDM Hydantoin (Sanogiene, BioMed Protect, Earth City, MO), of hospital scrubs in decreasing bacterial contamination on scrubs at the end of a typical hospital shift. Chitosan, the primary active ingredient in the Sanogiene product, is a derivative of chitin, a major component of crustacean shells, and has demonstrated antimicrobial properties. Through a proprietary curing process, Sanogiene can be applied to various textiles containing at least 20% cotton fibers and has been used in commercial and industrial applications. For more information see www.http://www.biomedprotect.com/index.php/products/sanogiene/

Methods

Study Design and Randomization

We used a blinded, randomized-crossover study design in which participants served as their own control. Nurses and Patient Care Technicians from adult intensive and intermediate care units at the University of Maryland Medical Center were enrolled. Each participant was given four sets (top and bottom) of scrubs: two sets treated with the antibacterial agent using a proprietary curing process (intervention) and two sets left untreated (control). All scrubs were labeled 1–4 and each participant was given a randomized sequence of wear (e.g., participant A wears scrubs 2-4-3-1 in a repeating pattern). Randomization was performed using an online random sequence generator.² Participants were instructed to perform standard laundering at home. All scrubs looked and felt identical and participants were blinded to intervention/control.

Outcomes

The outcome of interest was the frequency of scrub contamination for overall rate of scrub contamination with pathogenic bacteria (any of Staphylococcus aureus, Enterococcus spp., or Gram-negative bacteria) after a hospital shift. The secondary outcome of interest was the total colony count of all aerobic bacteria.

Data Collection

To detect pathogenic bacteria, a sterile, premoistened double tipped swab (Remel; Lenexa, KS) was rubbed over the front of the scrub top in a large W pattern, including the waist and chest, and down the front of both thighs. For the secondary outcome of total aerobic colony counts, a RODAC (Becton Dickinson; Sparks, MD) agar plate was directly stamped on to the scrub top near the belly button. These sampling areas are the most likely to be touched by one’s own hands (especially on scrub-top styles with lower front pockets) and come into regular contact with patients and items in the environment (e.g. bedding) during the course of regular duties. Samples were taken during the last four hours of the 12-hour shift in order to allow ample time within a shift for potential scrub contamination. Redundancy was included in scheduling to ensure full data collection, i.e. if a participant was not wearing appropriate study scrubs during a sample collection visit, no cultures were collected and the visit was repeated.

Laboratory

Enrichment cultures plated to selective media were used to detect Gram-negative bacteria, S. aureus and Enterococcus spp. Bacterial isolates were identified by standard laboratory protocols. Susceptibilities were performed following CLSI guidelines.³ Gram-negative isolates were considered multi-drug resistant with resistance to three or more classes of antibiotics.⁴

For total colony counts, RODAC plates were incubated for 48 hours and colonies were counted using an eCount precise electronic counter (Heathrow Scientific, Vernon Hills, IL).

Analysis

Analysis was performed using Generalized estimating equations in SAS, version 9.2 (SAS Institute Inc., Cary, NC).

Results

A total of 110 HCWs were enrolled; 90 completed full data collection and were included in analysis, for a total of 720 samples. Excluded from the analysis were 20 participants with incomplete data due to changes in shift or employment status. Overall, 30% (217/720) of scrubs were contaminated with pathogenic bacteria; 30.0% (108/360) of treated scrubs and 30.3% (109/360) of non-treated scrubs (p=0.93). When stratified by type of bacteria (i.e. S. aureus, Enterococcus spp., or Gram-negative), the difference in contamination on treated versus non-treated scrubs remained insignificant in each category (Table 1). There was also no difference in contamination between newer scrubs (less than 16 days old, the median number of participation days) and scrubs older than 16 days and no significant difference in contamination with pathogenic bacteria by ICU vs. IMC or by Trauma vs. non-Trauma. The average colony count was 49 for treated and 52 for non-treated scrubs (p=0.67). 100% of participants self-reported scrubs where home laundered prior to each wear. No adverse events were reported.

Table 1.

Bacterial contamination on treated versus non-treated scrubs

	Bacteria Present		P-value
	Treated N= 360 (%)	Non-Treated N=360 (%)
Overall Pathogenic Bacteria	108 (30)	109 (30.3)	p= .93
Staphylococcus aureus	61 (16.9)	55 (15.3)	p= .50
Enterococcus spp.	7 (1.9)	14 (3.9)	p= .17
Gram-negative Rods	54 (15)	59 (16.4)	p= .55
Multidrug Resistant (MRSA, VRE, or Multi-drug Resistant gram-negatives)	16 (4.4)	28 (7.8)	p= .06

Discussion

Antimicrobial coating of scrubs was not effective in preventing bacterial contamination. Regardless of bacterial species or resistance profile, nearly equal amounts of treated and non-treated scrubs became contaminated by pathogenic bacteria.

Other recent studies testing antimicrobial scrubs but using different products have shown similar ineffectiveness. An organosilane-based quaternary ammonium antimicrobial agent was tested against control scrubs in 2012; the treated textiles were found to be slightly protective against contamination by MRSA, but, in concordance with our study, not against VRE or MDR Gram-negative bacteria, in the course of regular clinical duties among 30 participants. The investigators only sampled the opening of the pockets and although HCW hands likely lead to contamination in this area, these sites are unlikely to be contaminated directly from interactions with the patient and their environment, potentially lessening the opportunity to measure real-world efficacy.⁵ In a similar study from 2013, 105 HCWs participated in a three-armed randomized-control trial testing two different antimicrobial products (group A: proprietary anti-microbial chemical, group B: silver and two proprietary anti-microbial chemicals) against standard scrubs. Again, no difference in levels of contamination between either of the treated products and the control scrubs was found. It was also reported that several participants experienced dermatologic side effects related to wearing both types of treated scrubs.⁶

Although in vivo testing has not been positive, many studies assessing the antimicrobial effectiveness of materials including metal fibers, nanoparticles, and chemical coatings (including chitosan, one of the active ingredients in this study), have found success in the laboratory. Some bactericidal action is often detected in treated textiles in as few as three hours and nearly complete by 48 hours, while control textiles show essentially no reductions over time.^7–9

Laboratory achievement may still be translated into successful non-attire based healthcare applications. Recent research found that privacy curtains incorporating a proprietary metal alloy with antimicrobial properties in ICU patient rooms increased the time of initial bacterial contamination of the curtains seven fold (14 days versus two days until contamination).¹⁰ These findings may be translatable to other environmental textiles including bedding and other soft surfaces in the patient environment.

Breaking the link between bacterial contamination of HCW attire and HAI is an important step in the ongoing fight against these insidious infections. However, the lack of efficacy demonstrated in this study shows that more research and innovation, including in terms of agent used and method of impregnation, is necessary to determine the optimal mode of prevention for HCW attire contamination before hospitals or healthcare workers invest in antimicrobial scrubs.