Evonne T Curran

Epidemiology and clinical characteristics of parainfluenza virus 3 outbreak in a Haemato-oncology unit

OBJECTIVES We describe molecular investigations of a large hospital outbreak of parainfluenza virus type 3 (PIV3), in which 32 patients became infected. We outline infection control measures that successfully limited further spread of PIV3 in a Haemato-oncology unit. METHODS Clinical retrospective review of infected haemato-oncology patients was undertaken. PIV3 haemagglutinin sequences from each case (n = 32) and local epidemiologically unlinked controls (n = 53) were compared to identify potential linkage. RESULTS PIV3-infected patients presented with upper (n = 18) and lower (n = 11) respiratory tract infections, 3 showed pyrexia only and one was asymptomatic. All symptomatic patients received antibiotics; bacterial co-infection was confirmed in eleven patients. PIV3 infections were associated with lower mortality than documented previously; three of the PIV3-infected patients died (3/32; 9%). All deaths were associated with relapsed malignancies, and PIV3 was not believed to be the primary cause of death in any of these patients. Sequences from 27 cases clustered closely together, consistent with nosocomial infections from PIV3 circulating within the ward. Factors favouring transmission were high patient turnaround between the day treatment unit and in-patient ward, and limited isolation facilities for immunocompromised and infected patients, especially within the day treatment unit. New infections reduced to baseline levels three days after enhanced infection control interventions were introduced. CONCLUSIONS Molecular epidemiological analysis provided evidence for nosocomial transmission of PIV3 infection that facilitated effective implementation of infection control measures. These were instrumental in restricting further spread of the virus among high-risk patients.

Using data effectively to prevent and control infection

Data is an essential tool for convincing healthcare workers to accept that problems exist, inspire them to better performance, or demonstrate that their performance is improving. Infection control professionals have access to a wealth of data on healthcare associated infections. Using simple graphical examples this paper illustrates how data can be analysed and presented in accessible ways that will help infection control practitioners to better understand infection problems and to use the information to influence practice.

Aiming to reduce catheter associated urinary tract Infections (CAUTI) by adopting a checklist and bundle to achieve sustained system improvements

Introduction evice related infections such as those associated with catheters, e.g. catheter associated urinary tract infections (CAUTI) and peripheral vascular catheter (PVC) insertion site infections can seem trivial to healthcare workers (HCWs) in that they are often considered easy to treat, i.e. all that has to be done is to remove the catheter and start antibiotics if necessary. However, these infections are so numerous they must be considered indicators of signifi cant suboptimal device care. Furthermore they are possibly the fi rst step en route to more complicated secondary infections such as bloodstream infections, complications of antibiotic therapy such as Clostridium diffi cile associated disease (CDAD) and colonisation/ infection with antibiotic resistant organisms (AROs). It follows therefore that efforts should be targeted primarily at reducing device related infections by optimising device care and that if these efforts are successful then it would seem reasonable to assume that there will be a positive concurrent effect in reducing the pressure on CDAD and cross-transmission by AROs. This paper demonstrates how quality improvement tools have been devised in Scotland using Human Error Theory to assist HCWs to reduce CAUTI.

Pseudo outbreaks and no-infection outbreaks (part 2)

Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/1757177413484546 108 Journal of Infection Prevention May 2013 VOL. 14 NO. 3 O pi ni on /C om m en t n the first column on pseudo-outbreaks I argued for the use of two categories for both outbreaks and pseudo-outbreaks based on whether the infections were real, the patients were exposed to micro-organisms, and the cases were related (Curran, 2013). There was also detailed discusion on pseudo-outbreaks which were unrelated clusters of real infections. In this column the focus will be on the second category of pseudo-outbreaks, related clusters of false infections where the results themselves are false but related to a common systematic error. There will also be further discussion of no-infection outbreaks where patients have been exposed to micro-organisms and are colonised but no infections are present. Finally, from the review of all the pseudo-outbreak papers I could muster – which was almost 100 – I will present an algorithm on the decision-making needed to classify alert signals into one of the four categories: an infection outbreak, a no-infection outbreak, an unrelated cluster of real infections or a related cluster of false infections.

The Where is Norovirus Control Lost (WINCL) Study: an enhanced surveillance project to identify norovirus index cases in care settings in the UK and Ireland

Background: Norovirus outbreaks have a significant impact on all care settings; little is known about the index cases from whom these outbreaks initiate. Aim: To identify and categorise norovirus outbreak index cases in care settings. Methods: A mixed-methods, multi-centre, prospective, enhanced surveillance study identified and categorised index cases in acute and non-acute care settings. Results: From 54 participating centres, 537 outbreaks were reported (November 2013 to April 2014): 383 (71.3%) in acute care facilities (ACF); 115 (21.4%) in residential or care homes (RCH) and 39 (7.3%) in other care settings (OCS). Index cases were identified in 424 (79%) outbreaks. Of the 245 index cases who were asymptomatic on admission and not transferred within/into the care setting, 123 (50%) had been an inpatient/resident for 4 days. Four themes emerged: missing the diagnosis, care service under pressure, delay in outbreak control measures and patient/resident location and proximity. Conclusion: The true index case is commonly not identified as the cause of a norovirus outbreak with at least 50% of index cases being misclassified. Unrecognised norovirus cross-transmission occurs frequently suggesting that either Standard Infection Control Precautions (SICPs) are being insufficiently well applied, and or SICPs are themselves are insufficient to prevent outbreaks.

Standard precautions: what is meant and what is not

Recent guidance for controlling carbapenemase-producing Enterobacteriaceae has brought a critical issue for infection prevention and control teams (IPCTs) to a head: there is an erroneous and dangerous assumption that IPCTs and healthcare workers (HCWs) have a common understanding of what is meant by, and included in, the term ‘standard precautions’. Exactly what is meant by this term differs between recent guidance from the World Health Organization (WHO), the US Centers for Disease Control and Prevention (CDC), Health Protection Scotland and epic3. This situation is unsafe because these differences can lead to misunderstandings and, potentially, to actual harm. The time has come for professional consensus on a single definition of the term ‘standard precautions’.

Outbreak Column 12: Nosocomial Staphylococcus aureus outbreaks (part 1)

Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/1757177413513817 36 Journal of Infection Prevention January 2014 VOL. 15 NO. 1 O pi ni on /C om m en t aving decided that the topic for Outbreak Column 12 would be Staphylococcus aureus the problem came in deciding where to start, what to include and what to omit. It soon became clear that one column would not suffice. Therefore this is the first column which looks back at some early work; the second will discuss outbreaks, guidance evolution and the current situation. This has been a difficult challenge. Trying to produce useful Outbreak Columns on nosocomial outbreaks of S. aureus is like trying to describe the world’s weather in a very short synopsis. Having got the defeatist part over, I will attempt to highlight valuable information for those who work in infection prevention and control (IPC) in clinical settings where this commensal and sometime pathogen has been and remains a constant companion and threat. I have left the title bland so as to include any strain of S. aureus regardless of resistance pattern and virulence. The column was going to do the usual, extract useful information from outbreak reports and relate this to implications for practice. However, I have been drawn to explore what I shall call the ‘Wyllie conundrum’. In my search for useful reports to include I found a paper which suggests that reductions in meticillin resistant Staphylococcus aureus (MRSA) began before efforts at control were introduced. Obviously, this needed to be explored before any advice on the control of outbreaks of S. aureus could be offered.

Outbreak Column 7: Pseudo-outbreaks (part 1)

Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/1757177413477462 VOL. 14 NO. 2 March 2013 Journal of Infection Prevention 69 O pi ni on /C om m en t riting the JIP Outbreak Column has necessitated that I read more outbreak reports, and read them in greater depth than I did before the column started. In doing so, what has surprised me most is that before I started on the column I thought I could easily and clearly define both an outbreak and a pseudo-outbreak – now I am not quite so sure. Indeed, in writing for this column I have found a series of what the authors called ‘pseudo-outbreaks’ that are not outbreaks, in that patients did not develop infections; however, they were exposed to and colonised with a common pathogenic microorganism within a care environment. Consequently, I will argue in this column for a new category to be considered. The main topic for this outbreak column is pseudo-outbreaks, that is, alert signals that on investigation are not outbreaks – this of course includes the definition. The purpose of choosing pseudo-outbreaks for a column was to learn lessons on alert signal analysis that would help Infection Prevention and Control Teams (IPCTs). What I have discovered is that there are many lessons to be learned from pseudo-outbreaks and not just about alert signals. Indeed, such are the variety of pseudo-outbreaks that this is the first of two columns on the topic. Every outbreak starts with an ‘alert signal’ from clinical, microbiological or surveillance data. The alert signal suggests there are more patients with a particular type of infection or specific alert organism than would be expected within a given time period. This is clearly illustrated in the Health Protection Scotland Hospital Outbreak Algorithm and Checklist (Health Protection Scotland, 2012). The first task of the IPCT is to decide – is the alert signal a real or a pseudooutbreak. To aid the decision-making process, the IPCT will review the patients’ clinical conditions, collect various time, place and person data and consider from all the available information whether the alert signal is more likely than not to be an outbreak. At the time of the assessment there may be insufficient evidence to be certain; when this is the case, the IPCT should still act as though it is an outbreak until there is sufficient evidence to prove otherwise. The second column will deal with pseudo-outbreaks that can be attributed to failures in specimen collection, processing, result interpretation and clinical diagnoses. This first column will focus on the categorisation of pseudo-outbreaks and discuss those pseudo-outbreaks that arise from surveillance, testing and chance. The newly proposed category, which for now I will call ‘no-infection outbreak’, will also be outlined in this column with more extensive discussion given in the second column. To start, let us be clear what is meant by the term pseudo-outbreak.

Outbreak column 3: outbreaks of Pseudomonas spp from hospital water

Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/1757177412447531 VOL. 13 NO. 4 July 2012 Journal of Infection Prevention 125 O pi ni on /C om m en t he recent outbreak of Pseudomonas aeruginosa in a neonatal unit in Northern Ireland has highlighted once again the imperative of a safe patient environment (Burns and Feeley, 2012). If water is not safe, then how can we have safe hospitals? This Outbreak column reviews some of the recent reports of outbreaks of P. aeruginosa and other Gram negative waterborne organisms, which highlight that it is not just the water coming in through the tap that is a cause for concern. The problem extends to the water outlets (wash-hand basins, sinks, showers and taps) themselves that can become a nidus for ongoing dissemination, and finally, to the clinical practices that complete the transmission pathway from sink organisms to patient infections. In addition, these recent outbreak reports demonstrate that our current opinions on what an optimally designed infection prevention healthcare environment should look like may be erroneous. The many outbreaks of P. aeruginosa reported in the literature from exogenous sources are not discussed here. Pseudomonas aeruginosa lives in water and soil but it can also be a human commensal. The tap water coming into hospitals is not, and never can be, sterile. What is crucial is that the journey of tap water through hospitals’ water pipework systems does not encourage the growth of opportunistic pathogens in biofilms. It is the intermittent sloughing off of organisms in biofilm that presents undetectable spikes or continuous infection risks where opportunistic pathogens are concerned. As P. aeruginosa is an opportunistic pathogen, for infection to happen there needs to be an accompanying transmission pathway from the source (water/tap/drain) to the patient; there must also be a susceptible host and a ‘way in’ for the organism. Although we can recognise our susceptible hosts, we currently have limited methods to detect the presence of biofilm and the pathogens therein. The first outbreak report by Hota et al (2009) details the difficulties in managing such a situation of unknowns. Hota et al (2009) describe an outbreak of multidrug-resistant P. aeruginosa in three areas of a hospital (intensive care unit, transplant step-down and the transplant ward). The outbreak took place over 15 months between December 2004 and March 2006. It involved 36 patients, 12 of whom died – five deaths were directly attributable to the outbreak organism. Microbiological culture from water samples did not identify any positive cultures of the outbreak organism; however, two external plumbing fixtures, a shower head and tap spout were positive. Contamination with the outbreak organism was identified in 26 of 213 sink drains. In addition, the outbreak team confirmed the presence of confluent biofilms, which contained the outbreak organism in the wash-hand basin drains. The key investigation performed by this team was a fluorescent splash test of the wash-hand basins. This experiment showed that splash contamination from the washhand basin reached an area one metre beyond the sink. Within the one metre subject to splash contamination was the preparation area and patient equipment. The control team also considered that microparticles (aerosols) containing the outbreak organism could travel further than the visibly splashed one metre area. It would be interesting to investigate this using air samples that fractionate the particle sizes to determine if the smaller aerosols, which are respirable, could be shown to contain the outbreak organism. This extended area of contamination would, of course, include the patient. What this report showed was that healthcare workers who were assiduously undertaking care procedures – including hand hygiene – could have inadvertently colonised their patients. The room design and fittings were all considered to have played a part in the outbreak. In particular, the flow of water from the tap landed directly on the sink drain, which maximised the contaminated splash-back and potential for aerosol production from the colonised sink drain. The authors reiterate that biofilm will form in drains over time. Therefore ensuring that sinks near patients have minimal aerosol generation potential and that the water does not directly flow from the tap into the drain is essential. The outbreak team managed the problem rather than eradicated it. The organism was still present in their drains more than a year after the outbreak – but with just one further case identified (Hota et al, 2009). Cervia et al (2009) provided comment on the Hota et al (2009) study. Their opinion was that tap water was still (despite the negative microbiology) the most likely source for initial colonising of the washhand basin drains. They suggested that biofilm colonising the pipework, which only intermittently shed organisms could explain the negative findings. They suggest that successful culturing of P. aeruginosa requires special growth media and conditions (Cervia et al, 2009) (it also requires appropriate sampling). This assessment provides a warning to infection prevention and control teams (IPCTs) that local negative culture results from water outlets can, if not properly sampled, be false negatives and that a wide range of practices using water should also be considered as routes of transmission. Durojaiye et al (2011) identify characteristics that are emerging as common factors in these recent nosocomial outbreaks of P. aeruginosa. The outbreak setting was a refurbished area, the unit (a 15-bedded ITU) cared for vulnerable patients and the identified reservoirs were new sensor mixer taps that had been introduced as part of the upgrade. Their outbreak, which involved 10 patients, took place over five months between November 2009 and March 2010 in an Outbreak column 3: outbreaks of Pseudomonas spp from hospital water

Outbreak Column 15: Carbapenemase-producing Enterobacteriaceae

Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav 10.1177/1757177414546707 VOL. 15 NO. 5 September 2014 Journal of Infection Prevention 193 Pe er re vi ew ed o pi ni on /c om m en t utbreak Column 15 covers the often confusing ‘carbapenemase producing Enterobacteriaceae’ (CPE), the epidemic curve which has yet to reach base camp. Although there is no doubt that this family of organisms presents a formidable public health challenge, there is some debate about what needs to be done to prevent and control outbreaks. Although the challenge from CPE has only been recognised since the mid-1990s (Queenan and Bush, 2007), the task for the Outbreak Column appears equally as daunting as that presented by Staphylococcus aureus – which has been a nosocomial challenge for considerably longer (Curran, 2014a). The high-reliability characteristic ‘deference to expertise’ (Weick et al, 1999), is being demonstrated in this column in that it has been coauthored by Dr Jon Otter, whose personal expertise has enabled the topic to be tackled from a more learned perspective. In this review, we provide: