Great news from CDC! Help me to understand it...

Today our colleagues at the CDC’s Division of Healthcare Quality Promotion (DHQP) and the Tennessee and Iowa Departments of Public Health released an interesting MMWR Vital Signs report highlighting CDC’s efforts to control emerging antibiotic resistance (AR) phenotypes (main focus being carbapenem-resistance in Enterobacteriaceae (CRE) and Pseudomonas aeruginosa (CRPA)). The release is accompanied by a helpful summary of CDC’s efforts to better coordinate the detection and control of new and emerging resistance threats (including, of course, not only ESBL and CRE but Candida auris, colistin resistance, glycopeptide-resistance in S. aureus, etc.). Check it out!

My major takeaway: investments in laboratory detection of AR are beginning to pay off. By improving their diagnostic capabilities, more regional and state labs are now able to help health care facilities confirm, characterize and respond to pathogens that display unusual resistance phenotypes. This is unquestionably a good thing, though it still requires local facilities to be able to detect phenotypes of concern—for example, many hospitals don’t have ready access to MALDI-TOF or sequencing, and many do not identify Candida to species level routinely from non-sterile sources. How long might C. auris spread in such a facility prior to its detection? So although the AR Lab Network (ARLN) is a major step forward, we still need to emphasize that individual hospitals must invest in improved and timely AR diagnostics--and we ought to ensure they receive updated guidance about how best to detect new or emerging pathogens of concern.

Now for the data itself: the report presents NHSN CAUTI and CLABSI data from 2006-2015 on the percentage of all isolates of E. coli and K. pneumoniae with ESBL and CRE phenotypes over time. The ESBL phenotype was consistently detected in 16-19% of isolates over time, with no major change noted. However, the % CRE declined steadily, from a peak of 10.6% in 2007 down to 3.1% in 2015 (see Figure below). This is despite the fact that CLSI breakpoints for carbapenem susceptibility were lowered during the surveillance period, which should have caused more isolates to meet phenotypic criteria for resistance. And although the CDC published CRE-specific control guidance in 2009, the decline in %-resistant preceded this guidance (which would obviously not have been immediately implemented across the country, likely delaying any impact for 1-2 years). This also coincides with the spread of carbapenemase-producing Enterobacteriaceae (CPE) across the US. So I’m left wondering: why is the % CRE declining as a cause of CAUTI and CLABSI in acute care hospitals?

 

Is it related to the general improvements in infection prevention practices that have accompanied public reporting and pay-for-performance? Would such improvements differentially impact CRE over other pathogens (keep in mind this isn’t a rate, it’s a proportional decrease)? It’s clear that the epicenter of CRE/CPE seems not be in acute care but rather “post-acute care” as is also noted in the other part of the report (the results of CPE screening by ARLN labs for 9 months of 2017)….but I’d still expect increases in post-acute care or other health care settings to eventually be reflected in the NHSN CAUTI/CLABSI data. Granted, the NHSN data represents a relatively small number of organisms (for example, Clare Rock and colleagues demonstrated that CLABSI represent only 6% of all hospital-onset bacteremias).

I remain perplexed. And happy to hear any thoughts on this!

Pan-resistant NDM-containing K. pneumoniae lacking MCR-1

Hopefully, you've started your long weekend or begun planning your MLK-Jr Day of Service. But just in case, I wanted to let you know about a new case report in MMWR while simultaneously raising the MMWR Altmetric Score. Seriously, how can such a case only have one blog mention in 24 hours - Jon Otter must be just getting back from holiday.

CDC has reported a case of NDM-containing CRE K. pneumoniae isolated from a wound specimen of a woman in her 70s who had returned to the US after a log visit to India. She also reported prior visits with multiple hospitalizations in India. The isolate was resistant to the 26 antibiotics tested, including all aminoglycosides and polymyxins (despite being MCR-1 negative), while it was intermediately resistant to tigecycline. The fosfomycin MIC was 16 μg/mL. Unfortunately, the patient developed sepsis and died. Fortunately, the patient had been isolated and screening revealed no evidence of transmission.

CDC offered three take home points: (1) Most CRE remains susceptible to at least one aminoglycoside or tigecycline (2) When patients with this highly resistant organisms are identified, they should be placed under contact precautions* and (3) Facilities should obtain travel histories and consider screening for CRE if patients are from high-risk areas.


*Note: nice to see that somebody still loves contact precautions.

Excess Mortality in CRKp: A Non-Randomized (Fortunately) Trial

Understanding the burden of antimicrobial resistance is critically important if we are to appropriately target research and clinical resources. For years, lack of proper estimates of the morbidity, mortality and costs associated with multidrug-resistant bacteria greatly limited the attention paid to these pathogens. This changed with the 2013 CDC Antibiotic Resistance Threats Report which provided the public with the number 23,000.  In the report, carbapenem-resistant Klebsiella pneumoniae (CRKp) was estimated to cause 7,900 infections and 520 deaths per year. But questions remain: Is the 6.5% (520/7900) mortality estimate high or low and how can we estimate the burden of resistance since we can't (fortunately) perform randomized trials where we randomly infect patients?

To answer these important questions, a group of investigators formed The Consortium on Resistance against Carbapenems in K. pneumoniae (CRACKLE) and just published a cohort study in Clinical Microbiology and Infection. This group, of what appears to be 18 Great Lakes hospitals, prospectively collected CRKp BSI (N=90), pneumonia (N=49), and UTI (N=121) isolates along with a control group (N=223) of patients with CRKp urinary tract colonization. The use of patients colonized but not infected with the pathogen as controls is interesting. The authors explain that they chose these controls since "non-infection-associated contribution to overall mortality is relatively larger in patients colonized with CRKp compared with patients colonized with more susceptible organisms, since risk factors for mortality such as chronic and acute illness, overlap with risk factors for CRKp colonization. An estimate of this non-infection-related mortality may be approximated in patients who are colonized, but not infected with CRKp." This is another way of saying that they wanted to isolate the attributable mortality risk of infection, not underlying disease.

The primary outcome was time-to-hospital-mortality from the time of the first positive CRKp culture as calculated by an adjusted hazard ratio using Cox proportional hazard models. I've included the unadjusted outcomes below. The full paper includes separate models and Kaplan-Meier curves for each infection, which don't differ greatly from the unadjusted outcomes.

As you can see, 39% of both BSI and pneumonia patients died or were transferred to hospice care compared to 12% of controls giving an attributable mortality of 27% for CRKp infection. In the Cox models, the adjusted hazard ratio was 2.59 (1.52-4.50) for BSI and 3.44 (1.80-6.48) for pneumonia. In contrast CRKp UTI was protective in both the unadjusted (3% lower mortality) and adjusted (aHR=0.68, p=0.33) analyses. This is further evidence that we need to rethink our definitions and focus on UTI.

Overall, a very nice study that utilized a novel control group of patients colonized but not infected with the organism of interest. It is likely that this approach when coupled with multivariable analysis reduced the effects of measured and unmeasured confounders. And it looks like CDC should increase the attributable mortality from 6.5% in their 2013 report to something a bit higher - say 27%.

The Times They Are A-Changin’, and we need better data to keep up….

I recently pointed to the need for more studies comparing different durations of therapy for common infections. In an editorial in the September 2015 issue of Antimicrobial Agents and Chemotherapy, Jesus Rodriguez-Bano points to an equally important priority for comparative effectiveness studies: assessment of existing antimicrobials for organisms that are currently not considered good targets for those drugs. The problem he discusses is the rising rate of carbapenem use that has followed the global spread of extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae. 

Carbapenems are the drugs of choice for serious infections due to ESBL-producers, but several narrower spectrum agents (e.g. piperacillin-tazobactam, cephamycins (e.g. cefoxitin, cefotetan)) have in vitro activity against many of these organisms. What’s lacking are well-designed comparative trials examining whether these older agents might be similarly effective for selected serious infections (e.g. bacteremia, sepsis) due to ESBL-producers. The study by Matsumura and colleagues that the editorial accompanies provides some support for the effectiveness of cephamycins in the setting of ESBL-E. coli bacteremia, but is limited by its retrospective cohort design and its power.

Given the urgent threat of carbapenemase-producing Enterobacteriaceae (CRE), it would be nice to have more non-carbapenem options to turn to when confronted with these “garden-variety” ESBL-producers.

CRE decolonization: Can we? Should we?

There's a new paper in American Journal of Infection Control which poses an interesting question: can CRE (carbapenem-resistant Enterobacteriaceae) be eradicated from the gut? Given the increasing problems with these organisms, decolonization could play an important role in their control.

This study was performed at a large tertiary care hospital in Israel where active surveillance is performed on admission and weekly in targeted populations. Unfortunately the design of the study makes it difficult to interpret. There were 4 study arms: treatment with oral gentamicin, treatment with oral colistin, treatment with oral gentamicin + oral colistin, or no treatment. In part, the susceptibility of the colonizing strain dictated the study arm (e.g., patients with a colistin-resistant, gentamicin-susceptible strain were treated with gentamicin, unless they did not consent to treatment, in which case they were assigned to the control group). If the patient's isolate was susceptible to both drugs, they were randomized to either drug or to combination therapy. Treatment was given for 60 days or until eradication (defined as 3 consecutive negative rectal cultures with PCR performed on the 3rd negative sample), whichever came first. In the end, 26 patients received gentamicin alone, 16 colistin alone, 8 received both, and 102 were untreated. Eradication occurred in 42% for gentamicin, 50% for colistin and 37% for combined treatment. Only 7% of the untreated patients spontaneously decolonized.

What conclusions can we draw?

1. Spontaneous decolonization is a rare event.
2. It appears that some patients (one-half or less) can potentially be decolonized or at least suppressed while on treatment. Of note, 2 patients in the gentamicin arm, 4 in the colistin arm, and 2 in the combo arm relapsed. It would be interesting to know whether patients who had CRE eradicated remained culture negative for the long-term (e.g., 3 months or 6 months after treatment). 
3. Given that colistin remains one of the most important drugs for treatment of these infections, it worries me that widespread use of this drug for decolonization could result in resistance. Indeed, colistin resistance was reported in 1 of 16 treated patients, and 6 of 26 patients in the gentamicin arm developed gentamicin resistance.

In the end, how do we use the data from this study? I suppose if I practiced in a hospital with high rates of CRE colonization, I would be tempted to try decolonization for very high risk patients (e.g., neutropenic leukemic patients). I would favor use of gentamicin if the isolate were susceptible. But it would make me very nervous.

Photo: Klebsiella pneumoniae, CDC

NIGHTMARE BACTERIA coming to your PBS station October 22nd!

It's not quite the Zombie Apocalypse, but these carbepenem-resistant nightmare bacteria are clearly the next scariest thing. PBS's Frontline seems to think so. Producer/Writer/Director Rick Young has pulled together a 1-hour investigation into antibacterial resistant infections including (it appears) NDM-1 and the NIH CRE outbreak. From the promotional material it also seems that the program will touch on the lack of investment in drug discovery in addition to excess use, as causes of the epidemic. Remember, check your local listings.

NIH KPC Outbreak - The Final Word?

We've covered CRE extensively over the past couple of years.  Never so extensively as we did when the 2011 NIH KPC outbreak was first publicized last August following the whole-genome sequencing report in Science Translational Medicine. Almost a year has past since that report and kerfuffle, so it is nice to see that Tara Palmore and David Henderson have found the time to share their experiences controlling the outbreak and the media storm that followed the publication of the original manuscript. They decided to label the section on the public reaction the "Unintended Consequences of Publication." This title is very disturbing, as it highlights why many outbreaks like these are never reported - publication bias. I'm glad they weren't afraid to publish again, so that we can all learn for this difficult outbreak. The report is freely available in PDF over at CID. I'll stop writing and let you get on with your required reading.

"Nightmare" on Hospital Street

Not sure how we missed last week's JAMA medical news piece featuring our very own co-blogger Dan, but I suspect we might have been distracted by something pretty terrible. The article highlights the rise of carbepenem (CRE) and 3rd-generation cephalosporin-resistant Klebsiella pneumonia strains originally described in ICHE by Braykov et al. The CDC's Arjun Srinivasan emphasized that these strains are almost exclusively hospital-associated and now is the time for hospitals to implement the latest CRE recommendations from the CDC 2012 toolkit.

Dan "emphasized the importance of infection control basics such as ensuring a high rate of hand hygiene adherence among staff and making sure that surfaces and equipment are properly disinfected."  And said "If you don't shore up those things, screening [for CRE] isn't going to help."

I'll just paste in my favorite section: "An additional problem is the piecemeal approach to tracking these infections. Only 6 states require facilities to report CRE cases. “We need a more coordinated response,” Diekema said. He explained that the CDC is doing as much as it can with the resources it has, but underfunding of public health at the national and state levels makes it difficult to mount a more coordinated national effort to contain the spread of these infections. More research is also needed on the best strategies for environmental disinfection, ensuring adherence to hand hygiene and other measures that would prevent the spread of health care–acquired infections."

CRE isn't just a nightmare, it's a recurring nightmare. And just like the nine Nightmare on Elm Street films, it'll get worse over time. Especially if we continue with the current "piecemeal approach" prevention plan.

Source: Bridget Kuehn, JAMA 4/27/2013
Image source: wikipedia

Using QxMD's READ iOS? click here

Keeping "CRE" out

Last week was officially “CRE” (carbapenem-resistant Enterobacteriaceae) week, and I encourage anyone who hasn’t done so to read Eli’s excellent posts about the entirely predictable emergence and spread of these bad bugs. Before this crisis disappears down the media memory hole, it is worth mentioning one additional line of defense against CRE that is not up to the challenge: our clinical microbiology laboratories. This is particularly concerning when the new recommendations made by the CDC in response to CRE require additional laboratory support: (1) confirmatory testing and determination of the carbapenem resistance mechanism for CRE, and (2) rectal screening cultures of all patients admitted to U.S. hospitals who were hospitalized outside the U.S. in the prior 6 months.

The CDC provides a phenotypic method for CRE screening, and there are several other published methods (including commercial screening agars), but these approaches don’t provide clinically relevant turnaround times, nor do they tell us which mechanism is at work (e.g. KPC, NDM, VIM, IMP, etc.). Some referral centers use home-brew PCR tests for the most common carbapenemase in the U.S. (KPC), and there are commercial assays being developed that utilize real-time multiplex PCR to detect several enzymes at once, but these tests aren’t available to most labs. So if your hospital lab does find a bug that looks like a possible CRE (after 24-72 hours!), it can perform the Hodge Test (which isn’t great for emerging carbapenemases), send the bug out to a reference lab….and wait for the result.

I completely understand why the CDC has recommended new screening questions + rectal swabbing as part of CRE response. However, trying to keep these organisms from entering our hospitals and long-term care facilities using screening questions and rectal swabs is like depending upon a poorly built levee to hold back floodwaters. 

Dramatic advances in hand hygiene, environmental disinfection, and antimicrobial stewardship, on the other hand, would be the prevention equivalent of moving out of the floodplain to higher ground. Allow me to dream, please!

Further thoughts on CRE: "Once in A Lifetime" addition

"You may ask yourself, well, how did I get here?" - the Talking Heads

It's been a busy week in antibiotic resistance. From Tuesday's MMWR early release identifying terrifyingly large increases in CRE until today, there has been non-stop media coverage.  This morning, I had a long discussion with a colleague about CRE and it got me thinking about how we got to this point. As Maryn McKenna so astutely said this week: "it will be interesting to see whether the news sinks in this time." So, how did we get here?

1981: What we now call AIDS was recognized as a clinical syndrome. Thus began the cascade of  infectious diseases research towards a full-scale and massively-funded response to the AIDS crisis. This has been an incredibly successful, if not yet complete effort. At the 2011 IDSA meeting, Cornell's Trip Gulick said that we now have 10,000 possible combinations of antiviral therapy agents for HIV. Thirty years after the virus was discovered in 1983, the progress has just been incredible. Sure, we've not yet achieved a significant number of cures and an effective vaccine remains just out of reach, but if you'd asked most of us in 1990 if we'd wanted to switch places with 2013, we would have said yes in a heartbeat.

2013: Most ID research efforts still target HIV and other viral pathogens. As we published last year, 2009 NIAID funding for HIV was $1.3 billion, while funding for all ESCKAPE pathogens was around $49 million. When we reported this disparity, it seemed quite obvious that we needed to fund more research of antibacterial resistant pathogens. However, nothing is ever that easy. One of the peer-reviewers of our paper made the excellent point that the "historically poor funding for antimicrobial resistance over the years along with the exit of pharma from the field has led to a lack of significant infrastructure. There may not be enough productive labs to send more money to at this point." So, even if the US wanted to fund more research, it may take time to train or retrain investigators to undertake the relevant research.
___

Now that I've shared some brief thoughts on why we are where we are, I have some further thoughts on where we should go. First, as I said earlier this week, we need a national response to CRE and antimicrobial resistance in general. This response needs to be horizontal in approach, as Mike Edmond and Dick Wenzel recommended several years ago. For it is quite clear that if we follow the vertical approach recommended by the CDC and others, swab for CRE and isolate, that this will bankrupt hospitals and ultimately fail.

This surveillance approach will bankrupt us, because CRE isn't the only "nightmare" in our hospitals. Back in 2010, when Dan was discussing MRSA on NPR, he said so eloquently: "MRSA is not the only bad bug out there. It's just the most famous." And along with MRSA, we have VRE and ESBL and C. difficile and Acinetobacter and MDR-Pseudomonas. You see, we might not be able to take this single-hospital outbreak approach and extrapolate it to the entire country and 10+ pathogens. And if there is one lesson we should be taking away from the NIH CRE outbreak it's not that new-fangled whole genome sequencing stopped the outbreak (because it didn't), it's that the outbreak spread and killed many patients despite herculean efforts to detect and eliminate it. Furthermore, if we target MRSA like many hospitals are now doing with chlorhexidine (CHG) baths, this approach could select for Gram-negative bacteria like CRE.

My recommendations:
(1) Invest significantly in antimicrobial discovery. Apart from the need for new treatment options, optimal control of resistant pathogens may depend on availability of effective antibiotics.
(2) Invest in studies to improve compliance with hand hygiene – only 4 studies on this topic since 1980 per a recent Cochrane Review. Compliance is terrible, but currently the approach is to blame healthcare workers and not figure out how to help them easily clean their hands.
(3) Study universal gowning and gloving (several studies are ongoing). Dentists wear gloves with every patient, why not doctors?
(4) Undertake studies to further optimize environmental source control
(5) Actually study antimicrobial stewardship. Stop yelling at patients and clinicians to not use antimicrobials. Actually fund studies that use advertising and other other messaging techniques that have a chance to be effective.

Oh, and no more ridiculous hand hygiene song and dance videos...unless they include folks like the Talking Heads...

KPC, CRE, MDR-GNR: Call 'em what you will, but call for a national response!

There is an early release MMWR out today describing the recent trends in carbapenem-resistant Enterobacteriaceae (CRE) using data from the CDC's National Healthcare Safety Network (NHSN) and the Surveillance Network–USA (TSN). CRE infections, such as UTIs or intra-abdominal infections, cannot be treated with currently available antibiotics. The results of the MMWR study are not surprising - there are more CRE now - yet I find them rather chilling.

The proportion of Enterobacteriaceae that were CRE rose from 1.2% in 2001 to 4.2% in 2011 in NHSN hospitals and to 1.4% by 2010 in TSN facilities. In Klebsiella species, the situation is dire with 10.4% classified as CRE in 2011. By 2012, 4.6% of all facilities, 3.9% of short stay hospitals and 17.8% of long-term acute-care hospitals reported at least one CRE in their facility.

So things are pretty bad, at least 17.8% bad.

What can we do about this? This is what they say in the abstract: "Interventions exist that could slow the dissemination of CRE. Health departments are well positioned to play a leading role in prevention efforts by assisting with surveillance, situational awareness, and coordinating prevention efforts."

Which interventions? Certainly there are no evidence-based interventions, unless we call an uncontrolled quasi-experimental study evidence. The CDC authors made reference to VRE control efforts in the Siouxland, MRSA control efforts in The Netherlands and CRE control efforts in Israel. I think these are poor and overly optimistic comparisons for several reasons. For one, the first two efforts targeted Gram-positive bacteria and it's unclear if we can extrapolate these to control efforts targeting MDR-GNRs.  For example, we have effective antibiotics for VRE and MRSA (quinupristin/dalfopristin-approved in 1999; linezolid-approved in 2000, daptomycin-approved in 2003) but not for CRE.  Additionally, we can decolonize for MRSA but not CRE.

However, the most important aspect of the Netherlands and Israel examples were that they instituted NATIONAL RESPONSES aimed at MRSA and CRE. There are no such national efforts here in the US targeting CRE. This is what we have in the US per the MMWR report: "six states have made CRE reportable, and three additional states are actively pursuing this option." This is not a national response. This is a national tragedy. We can't expect CDC to say they can't handle the problem and we can't expect states to say they can't handle the problem. They are working their butts off with minimal support and no national will for a coordinated response. If we don't have a national response soon - it will be too late, but I'm not holding my breath.

I will leave you with an excerpt outlining the Israeli national CRE response from Mitch Schwaber's excellent report in CID. Do you think we have anything close to this going on here in the US? Me neither.

Update: Looks like NPR's "All Things Considered" will post an audio discussion of the CRE issue around 7pm ET here.

Stopping CRE: Where there's a will, there's a way?

Why is it that the DOD can ask for more money to fight their wars but CDC can't?  It seems to me that the fight against resistant bacteria is a war! - enp

I was interviewed by Peter Eisler of USA Today about a month ago concerning the recent NIH KPC Outbreak. His investigative report on carbapenem-resistant Enterobacteriaceae (CRE) was just published today and I found it a very good read.  The report centers around a CRE outbreak that began four years ago at the University of Virginia Medical Center and continues there to this day. The article is particularly interesting when you focus on the quotes from all of the other HAI-prevention folks in the article.  However, I also thought as I was reading it, what would Dan and Mike have highlighted if they'd posted on this same article?

I suspect that Dan would have wanted to highlight how difficult infection control becomes when gene/plasmids cross species during outbreaks and how this strongly impacts case definitions as you investigate. He might also point out how this limits the utility of whole genome sequencing, since the plasmids merrily skip from one species to another.  I think that Mike might have pointed out that the vertical control barriers, like nasal PCR, set up to prevent MRSA transmission are not very effective at preventing the spread of non-MRSA pathogens like CRE. He might suggest that a continuous and relentless focus on horizontal measures, like hand-hygiene, would be most effective.  Dan and Mike would both have been correct, but I have a different take...

My take is that CDC is on the wrong side of this debate and this has me very worried. Now, I don't remember exactly what was said about HIV during the 1980s, but I suspect there was a realization that something needed to be done and that significant investments would need to be made in its treatment and control. I'm not suggesting that CDC should flame fires of panic, but I wonder how long we can rely on the same old, largely ineffective, tools for outbreak control - patient isolation, subpar environmental control, inadequate hand-hygiene improvement bundles, lack of effective and novel antimicrobials - before CDC asks for help...and more funding.

So here are some select quotes from the article with my concerns bolded:

CDC:

"We're working with state health departments to try to figure out how big a problem this is," says the CDC's Srinivasan, noting that his agency can pool whatever incidence data states collect. "We're still at a point where we can stop this thing. You can never eradicate CRE, but we can prevent the spread. ... It's a matter of summoning the will."

The rest of us:

"In the Chicago area, where scores of CRE infections have been found since 2008, studies show that about 3% of hospital patients in intensive care carry the bacteria, says Mary Hayden, director of clinical microbiology and an infectious-disease doctor at Rush University Medical Center. Those same studies have found CREs being carried by about 30% of patients in long-term care facilities." "We have to think about a new approach, a regional approach, to controlling these organisms, because … no facility is an island," Hayden says.

"My concern is that there aren't a lot of methods in our tool kit that are significantly effective in curbing the spread of these infections," says Eli Perencevich, a professor and infectious-disease doctor at the University of Iowa's Carver College of Medicine. If unchecked, "these (bacteria) are going to greatly impact the kind of surgeries (and) treatments we can have," Perencevich says. "We're entering the post-antibiotic era; that's a very big problem."

"If you look at the current pipeline of antibiotics (in development) … none of them really is going to be active against these bacteria," says Gary Roselle, director of the Infectious Diseases Service for the Department of Veterans Affairs health care system. "The reality is, (CRE infections) are remarkably difficult to treat, they often have bad outcomes … and they're increasing nationally," adds Roselle. "I'm assuming this is going to get worse, and there likely won't be new antibiotics to treat it in the near future, so the focus has to be on prevention."

"So even if I had a perfect program to stop all patient-to-patient transmission in the hospital, the maximum impact I could have would be a 60% reduction in prevalence," says Brian Currie, the hospital's vice president for research and an assistant dean at the affiliated Albert Einstein College of Medicine.

"We have continued to have patients with CREs that are related to this (first) event," Costi Sifri, the hospital epidemiologist at UVA, says. "We haven't been able to close the door on this. ... I'm not sure you ever can."

References:
Peter Eisler, USA Today, 11/29/2012

Maryn McKenna, Wired, 8/24/2012