Carbapenem-Resistant Enterobacteriaceae (CRE) Control and Prevention Toolkit

Section 4. Implementing Best Practices

The decision to choose a KPC detection methodology should be made in collaboration with your institution's clinical microbiology laboratory director and should include a careful review of existing staffing, equipment, financial resources, and laboratory expertise. Planning strategies should include the resources to handle the anticipated volume of samples and include a plan to notify appropriate staff when patients colonized with KPC are identified. Important questions to address include—

  • How and when will samples be collected and transmitted to the lab as part of the clinical workflow?
  • How will the lab handle this new and additional workflow?
  • What is the anticipated volume of samples?
  • How and when will clinicians be notified of screening results? What will happen then?
  • Are adequate supplies of personal protective equipment available on the unit(s) where screening occurs? Do all staff know when/how to use them in caring for KPC-positive patients?

4.1 Collecting Patient Specimens

Although patients can be colonized with KPC at any anatomic location (e.g., asymptomatic bacteruria, chronic wounds, etc.), previous studies have documented that a peri-anal swab sample will detect all KPC colonized patients.2 Cotton-tipped swabs in use at the institution will suffice. Do not assume that the collection of samples is intuitive to hospital staff—consider providing laminated sheets that document and illustrate the sampling process (go to Tool 4A for an example). While obtaining a peri-anal swab sample is a noninvasive procedure, it is a sensitive issue for many patients. Strong consideration should be given to having nursing staff obtain the samples. They can incorporate sampling seamlessly into their normal patient care activities, enhancing patient acceptance and compliance with sampling and preventing disruption caused by unnecessary patient maneuvering for the sole purpose of sampling. Patient samples should be identified with routine labels currently in hospital use. Place pickup containers on each patient care unit for pickup of specimens at designated times. Specimen pickup provides an opportunity to review with nursing staff whether any issues or problems associated with obtaining samples. Specimens do not require either incubation or refrigeration, as Enterobacteriaceae are robust organisms. The importance of complete sampling of all target patients should be stressed as an important determinant of intervention success.

Return to Contents

4.2 Surveillance Strategies

No single approach to KPC surveillance will optimally address all scenarios of KPC prevalence, and your approach should be designed and customized to address the situation at your institution. Active surveillance should be considered when KPC clinical cultures begin to appear at your institution, or at another health care facility in your geographic area (many public health departments are facilitating the sharing of this information about MDROs in recognition that the problem requires population-level monitoring). Surveillance approaches might include periodic prevalence surveys involving a representative sample of inpatients, including both geriatric and intensive care unit (ICU) patients, such that every third inpatient is sampled during a 1- to 2-week period. 

Given the importance of inter-institutional transfer of KPC-colonized patients to the spread of KPC, the survey should include a sampling of emergency department patients who are nursing home residents or recent discharges from other acute-care hospitals who have been admitted but not yet transferred to an inpatient unit. Also include a sample of patients directly transferred to your hospital's inpatient areas from similar facilities. Here rapid reporting of test results is not important, and you may elect to use a culture-based technique for KPC detection.

If results reveal a significant presence of KPC among your patients or in patients recently discharged or transferred from other institutions, escalate your KPC control program to include more aggressive approaches. If results indicate a low prevalence, repeat your survey at regular intervals. If the KPC prevalence rate is greater than 2 percent or KPC is present at a neighboring institution, immediately escalate your KPC control efforts. Even with KPC colonization rates of 5 to 7 percent (endemic to epidemic rates), enhanced KPC control programs have significantly reduced the prevalence of KPC colonization. If you and your neighboring institutions share a significant KPC colonization prevalence rate, consider collaborating on a joint KPC control program.

Infection control practitioners may find it valuable to contact their peers at other institutions in their region to periodically exchange information about the emergence and/or prevalence of KPC in nearby institutions. Both acute- and long-term–care facilities should be included in these information exchanges. In addition, the CDC recommends that public health authorities develop programs to monitor KPC activity within their jurisdictions. Some state and county health departments now require reporting of all KPC isolates. Local health authorities may provide valuable information about potential KPC activity in your area. See the recent CDC toolkit for more information (CDC CRE 2012 Toolkit, "Guidance for Control of Carbapenem-Resistant Enterobacteriaceae (CRE), 2012 CRE Toolkit" http://www.cdc.gov/hai/pdfs/cre/CRE-guidance-508.pdf [1.73 MB]). Escalation of your control program should include active surveillance coupled with rapid initiation of contact isolation precautions for every colonized patient. Here the emphasis is on testing all patients in a target group, rapidly identifying KPC-colonized patients, and initiating contact isolation. The process needs to be designed and executed in timeframes that will allow at least 85 percent of total patient KPC colonization days to be in contact isolation. Complete sampling of the target population is required, as well as daily sampling of patients transferred or admitted to the target population.

A typical active surveillance program would include weekly sampling of the total target population and daily testing of newly admitted/transferred patients. To measure your impact on KPC colonization prevalence rates, collect baseline data before you initiate your intervention strategy. This can be accomplished by weekly sampling of all target patients pre-intervention.

Results from a previous cross-sectional sample of your patients can provide important information to identify your target population. KPC colonization prevalence may cluster in certain locations in your facility, such as geriatric service, critical care units, or patients transferred from other institutions directly to inpatient areas or admitted through the emergency department. Alternatively, if you have not conducted exploratory cross-sectional sampling for KPC colonization or believe the prevalence of KPC patient colonization is already increased at your hospital, you may choose to strategize to limit the prevalence of KPC among vulnerable patient populations, such as ICU patients.

References from Section 3.1 provide road maps for successful KPC control interventions for a single clinical unit, all ICU patients in a multihospital network, and an entire acute-care or long-term–care facility.


2. Simkins J, Pokharel R Dogra S, et al. Prevalence of carbapenemase-producing Klebsiella pneumoniae colonization at an academic medical center in New York City. Abstract 1349 presented at Infectious Diseases Society of America annual meeting, Boston, October 22, 2011.


Return to Contents

Page last reviewed April 2014
Page originally created March 2014
Internet Citation: Section 4. Implementing Best Practices. Content last reviewed April 2014. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/professionals/quality-patient-safety/patient-safety-resources/resources/cretoolkit/cretoolkit4.html