The trusted source for
healthcare information and
MDR gram negatives moving under the radar and across the continuum
"Our pa tients, wherever they are going and receiving care, are at tremendous risk."
By Gary Evans, Executive Editor
Emerging multidrug resistant gram negative bacteria are spreading across the health care continuum, becoming entrenched in non-acute and long term care settings and threatening vulnerable hospital patients with untreatable infections, epidemiologists reported recently in Dallas at the annual conference of the Society for Healthcare Epidemiology of America (SHEA).
Of course established heavyweights like Clostridium difficile and MRSA are still very much a concern, but multidrug resistant (MDR) gram negative rods like carbapenem-resistant Klebsiella pneumonia (CRKP) and Acinetobacter baumannii are a whole new threat. These pathogens are capable of pan-resistance, there are few drug options in any case, resistance-empowering plasmids can transfer among species, and infection mortality rates in the 40% range have been reported in vulnerable patient populations.1,2
Not exactly what you want moving through the health care continuum, but that appears to be what's happening in a situation that could get considerably worse before it gets better. In the lessons unlearned category, we are reminded of the vancomycin resistant enterococi (VRE) outbreaks in the 1990s, when communication breakdowns between hospitals, nursing homes and other non-acute settings descended into accusations of who gave what to whom.
"We spend a lot of time blaming each other," said Jon Furuno, PhD, an assistant professor of epidemiology and public health at the University of Maryland School of Medicine in Baltimore. "If you talk to any nursing home administrator they will say that their infection control problems are primarily related to transferring residents to acute care. [Residents] come back colonized or infected with some resistant organism."
On the other hand, he told SHEA attendees, "If you talk to hospital medical directors and infection control, they tell you these nursing home patients are `cesspools' of resistance and they spread it all over the hospital. The bad thing is that they are both right. We have to figure out how we are going to work together." Doing just that are facilities in Maryland and Vermont, two states that have established collaboratives between hospitals and non-acute settings to open the lines of communication and bring the infection control fight to the full continuum. For example, the collaborative in Vermont has linked up 14 hospitals and approximately 40 long term care sites. Sustaining such collaboratives will be the key, Furono emphasized.
"We are starting to see these collaboratives form and everyone goes into these things with a lot of enthusiasmeveryone is really psyched," he said. "But can they be sustained? Can we continue to work together? Can we see real change in the culture of infection control in these facilitiesand stop pointing fingers at each other?"
In another major development, long term care settings will be the top priority in the next phase of the Department for Health and Human Services Action Plan to Prevent Healthcare-Associated Infections, said Nimalie Stone, MD, MS, the medical epidemiologist for long term care in the Centers for Disease Control and Prevention's division of healthcare quality promotion.
"Infection prevention has to really be [involved] in every part of the health care continuum," she said. "There's no setting that can't be aware [of this] and not have the resources to dedicate to infection control. Our patientswherever they are going and receiving careare at tremendous risk."
Lost in transition
It is a risk that may frequently go unrecognized. Infection prevention is too often deemphasized in patient transitions between facilities in most states, with the colonization status with a multidrug resistant organism (MDRO) almost treated like a "don't ask, don't tell" policy as patients move through the health care continuum, said Furuno, who is researching infection control during health care transitions.
Studies of discharge plans between various types of facilities point to a disturbing lack of documentation, he noted, citing an analysis that included 73 studies of discharge data.3
"In 65% of these [discharges] the pending test results were not included on the discharge summary," he told SHEA attendees. "This happens all the time. People will get cultures but the patient will be discharged before their culture results come back, and then the receiving facility has no idea how to handle that information. Shockingly, 88% of the discharge summaries were not sent to the outpatient provider before they had a follow-up visit, and 25% of the time they actually never received a discharge summary. A lot of you are nodding your headswe see this all the time. It's pretty amazing."
Regardless, many infection preventionists argue that knowing the patient's colonization status may be less important than practicing rigorous standard precautions with all patients. Those who favor active surveillance for certain organisms beg to differ, saying if you detect and isolate such patients you can reduce nosocomial transmission. Interestingly, the question Furuno raised was a little different: Does knowledge of colonization benefit the patient carrying the organism?
"With infection control we are concerned about the rest of the facility and often our infection efforts aren't directed right at the individual who may be colonized," he said. "We certainly can try to improve our empiric therapy if we know they are colonized."
In addition, colonization is a known precursor to infection, so one could make both an ethical and medical argument in favor of determining MDRO status and telling the patient.
"In our effort to try to improve infection control in these transitions we are going to need to benefit the individual and the receiving facility," he said. "That's whyand we don't want to drown people in paperworkwe need to consider some of these things better. We need to improve and clarify our colonization and infection test results on our discharge summaries."
Though electronic medical records and other system improvements may be part of the answer, the bottom line is that far too many patients are moving across the continuum with undocumented MDRO colonization.
'In the last six months of life, people that have at least one nursing home staythat's importantwill have more than three transitions [to other settings] on average," he said. "There are pretty staggering numbers about how often people are moving through these different segments of the health care system."
Colonized for years
About one in five of those coming in from long-term care facilities will be colonized with a gram negative organism, according to a study by Erika D'Agata, MD, MPH, an infectious disease physician at Beth Israel Deaconess Medical Center in Boston.4
"There are high acquisition rates39% of our study patients acquired a new MDR gram negative [in long term care],"she said. "There is prolonged duration of colonizationon average 144 daysand that could probably apply to hospitalized patients as well. There is frequent co-colonization. One in five patients who have MDR gram negatives are harboring two or more different bugs."
This is not an insignificant population, as the number of people older than 65 years increased by 12.5% from 1998 to 2008, she said. "Currently, there are 3.2 million long term care residents [and] at any one time there are more residents in the long term care setting than in the hospital," D'Agata said.
The average resident is in long term care for about three years, and some may be colonized the entire time. "Our end-point was 349 days because we only followed them for a year, but some of these residents have been followed now for two or three years and they remain colonized with the same strain of multi-drug resistant gram negatives," she said.
There is no clearly established decolonization protocol for such patients, but D'Agata and colleagues found that about 9% will "lose" their MDR gram negative bacteria over time.
"It was a very rare event," she says. "The infrequent loss of MDR-gram negatives has implications for infection control interventions because if they are colonized for such long periodsand rarely lose itwhen can we stop [isolation] precautions? Should we even stop precautions? Should we be doing active surveillance cultures?"
Colonization is so persistent with many gram negatives that some infection preventionists grimly describe the fate of these patients as "isolated for life" upon admission to a hospital. In any case, D'Agata was less equivocal than Furono in naming the source of the problem.
"Residents of long term care facilities are major contributors to the influx of MDR gram negatives in the hospital setting," she said. "If the patient resides in a long term care facility he or she will have 3.5-fold higher risk of harboring MDR gram negatives compared to those who do not reside in long term care facilities."
CRKP goes West
In addition to moving across the continuum, MDR gram negatives are moving across the country. Already endemic in areas in the East Coastparticularly New York CityCKRP is now widespread through health care facilities in Los Angeles County. The pathogen was thought to be rare in LA County, but since it was not a reportable disease actual numbers were unknown, said Dawn Terashita, MD, MPH, a medical epidemiologist at the LA County Department of Public Health. Concerned that CRKP could give rise to pan-resistant K. pneumoniae, Terashita and colleagues initiated a community-wide lab-based surveillance system in 2010.
"We were especially surprised to discover how high rates of CRKP were in long-term acute care hospitals (LTACs)," she said.
During the study period of June-December 2010, 356 cases of CRKP were reported from 52 of the 102 hospitals in Los Angeles county. In addition, 145 cases of CRKP (41% of the total) were reported from all 8 LTACs in the county. Another 6% of CRKP cases were found in patients residing in skilled nursing facilities (SNFs). The mean age of patients with CRKP was 73 years, she reported at SHEA.5
There were some particularly bad characters among the isolates. One patient who had received care in Pakistan was positive for the New Delhi metallo-beta-lactamase 1, the much ballyhooed "superbug" which is thought to be rampant in hospitals in India. In addition, three isolates were both carbapenem-resistant and extended-spectrum β-lactamase (ESBL) positive, a combination that narrows treatment options considerably. These ESBL-CRKP infections are thought to be linked to overuse of carbapenmens, which is becoming increasingly common. Another SHEA study of Veterans' Administration's acute care facilities revealed dramatically increased use of carbapenems, which are often considered the last treatment option for these infections. Using barcode medication administration data for antibiotics administered in 110 VA acute care health facilities from 2005-2009, Makoto Jones, MD, and colleagues identified an increasing trend in the use of broad spectrum antibiotics. In particular, over the study's five year period there was a striking 102% increase in the use of carbapenems.6
"The more these drugs are used, the more resistance we see," said Jones, a medical epidemiologist at the University of Utah. "Use of these antibiotics helps the patient receiving the treatment, but has future consequences for innocent bystanders."
Nursing sites, LTC implicated
And what of those bystanders, patients exposed to such pathogens as they moved through the LA county health care continuum? The question goes beyond the scope of the lab surveillance study, but there were indications of movement from the skilled nursing facilities to hospitals, Terashita told Hospital Infection Control & Prevention.
"Based on the address on admission of the patient, we did get a piece of information," she said. "When we cross referenced that with our SNF data base we found that 124 of the 320 cases [for which] we had an admission addressthat's 39% of our caseswere from SNFs. So even if they were reported as a CRKP from a hospital they were admitted from a SNF. We can hypothesize that perhaps the patient acquired it in a SNF and it's just a kind of back-and-forth thing with the patients going between facilities. You never know for sure where they acquired it, but there is this evidence of the [transmission along the] continuity of care."
Dekata wondered about the same thing, and though warning that her findings have statistical limitations, said it appeared that long term care residency and antibiotic administration were the prime risk factors for MDR gram negative colonization.
"Not surprisingly antibiotic exposure was associated with a six-fold higher risk of acquiring multi-drug resistant gram negatives," she said. "There was no difference in the number of hospitalizations between those who acquired [an MDR gram negative] and those who didn't. It raises that important question: Are long term care residents acquiring these resistant pathogens in the hospital or in the long term care setting? Our study strongly suggests that it is the long term care settingnot the hospital."
(Editor's note: For more information on the infection prevention collaborative in Vermont go to: http://bit.ly/g0gMbX)