The Likelihood of Developing a Carbapenem-Resistant Enterobacteriaceae Infection during a Hospital Stay

Of 1,455 unique patients in U.S. intensive care units (ICUs), 4% were rectally colonized with CRE on admission. A total of 297 patients were initially negative for carbapenem-resistant Enterobacteriaceae (CRE) and remained in the ICU long enough to contribute additional swabs; 22% of these patients had a subsequent CRE-positive swab, with a median time to CRE colonization of 13 days (interquartile range, 7 to 21 days).

ABSTRACT

Of 1,455 unique patients in U.S. intensive care units (ICUs), 4% were rectally colonized with CRE on admission. A total of 297 patients were initially negative for carbapenem-resistant Enterobacteriaceae (CRE) and remained in the ICU long enough to contribute additional swabs; 22% of these patients had a subsequent CRE-positive swab, with a median time to CRE colonization of 13 days (interquartile range, 7 to 21 days). Patients colonized with carbapenemase-producing CRE were more likely than those colonized with non-carbapenemase-producing CRE to develop CRE infections during their hospitalizations (36% versus 3%; P < 0.05).

INTRODUCTION

Rates of carbapenem-resistant Enterobacteriaceae (CRE) infections continue to rise in the United States, with almost half of CRE infections consisting of carbapenemase producers (CP-CRE) (1–3). Patients hospitalized in the intensive care unit (ICU) setting are at particularly high risk of newly acquiring CRE during their hospital stay because the multiple preexisting medical conditions, compromised immune systems, lengthy unit stays, and significant rates of device and antibiotic utilization establish an ideal milieu for antibiotic resistance (4, 5).

Although several studies have identified CRE colonization as a risk factor for subsequent CRE infection (6, 7), available investigations do not explore whether this risk differs based on the production of carbapenemases by the colonizing strain. This information is important because it can guide whether the additional step of performing phenotypic or genotypic testing on rectal isolates positive for CRE provides added value. CRE infections are associated with deleterious patient outcomes, including lengthy hospital stays and significant morbidity and mortality, but this risk is particularly high when carbapenem resistance is due to carbapenemase production. CP-CRE infections are associated with 4 times the risk of 14-day patient mortality compared with non-CP-CRE infections (8). Researchers are increasingly exploring strategies for interrupting pathways between CRE colonization and subsequent infection (e.g., selective decontamination of the digestive tract, fecal microbiota transplantations) (9, 10). A better understanding of whether the risk of subsequent infection varies based on carbapenemase production status for colonized patients is important because this information can highlight the subpopulation that would be most likely to benefit from these promising but costly and invasive interventions. We determined (a) the prevalence of CRE colonization on ICU admission, (b) incident CRE colonization during the ICU stay, and (c) colonized patients’ likelihood of developing CRE infections during the same hospital encounter, with differentiation based on carbapenemase production status.

The Johns Hopkins Hospital is a 1,194-bed academic medical center in Baltimore, Maryland. Adult patients admitted to any medical or surgical ICU were eligible for inclusion. ICU patients are routinely screened for vancomycin-resistant Enterococcus (VRE) rectal colonization on ICU admission and weekly thereafter until ICU discharge using rectal ESwabs (COPAN Diagnostics, Inc., Murrieta, California).

From September 2017 to February 2018, after routine VRE testing, 30 μl of residual Amies broth from ESwabs was plated in real time on MacConkey plates using the WASP DT (Copan, Murrieta, CA) automated specimen processor. Ertapenem and meropenem disks were added manually to plates and incubated overnight at 37°C (11). Colonies growing within ≤27 mm of ertapenem and ≤32 mm of meropenem were further identified by matrix-assisted laser desorption ionization–time of flight mass spectrometry (Bruker Daltonics, Inc., Billerica, MA), as previously described (11–13). For Enterobacteriaceae isolates growing within the aforementioned zone diameters (including both discrete colonies and confluent growth), carbapenem (ertapenem, meropenem, and imipenem) antimicrobial susceptibility testing was performed to identify CRE by the disk diffusion method (14). Enterobacteriaceae isolates resistant to any carbapenem were classified as CRE (14). For all confirmed CRE isolates, the modified carbapenem inactivation method was conducted to identify carbapenemase production (15). All CP-CREs were further tested by the Xpert Carba-R assay (Cepheid, Sunnyvale, CA) to identify bla_KPC, bla_NDM, bla_IMP, bla_VIM, and bla_OXA-48 genes. Proportions were compared using the chi-square or Fisher’s exact test, as appropriate. This study was approved by the Johns Hopkins University School of Medicine Institutional Review Board, with a waiver of informed consent.

During the 6-month study period, 1,455 unique ICU patients were screened for CRE rectal colonization. In total, 119 patients (8.2%) tested positive for CRE colonization, either on ICU admission or during their ICU stay. Fifty-four patients (3.7%) were positive for CRE colonization on admission. Two hundred ninety-seven unique patients (20.4%) were negative on their admission screens and remained in the ICU long enough to contribute additional swabs; 65 of these patients (21.8%) had a subsequent CRE-positive swab, with a median time to CRE colonization of 13 days (interquartile range [IQR], 7 to 21 days). Four (6.2%) of the 65 incident colonization swabs were positive for carbapenemase producers. CP-CRE organisms and carbapenemase genes acquired during the ICU stay were comprised of Enterobacter cloacae (bla_NDM plus bla_KPC), E. cloacae (bla_KPC), Klebsiella pneumoniae (bla_KPC), and Escherichia coli (bla_NDM).

Overall, of the 119 patients colonized with CRE on admission or during ICU stay, 25 (21%) were colonized with CP-CRE (Table 1). Twenty-eight carbapenemase genes were identified from these 25 patients, including bla_KPC (25 [89.2%]), bla_NDM (1 [3.6%]), bla_OXA-48 (1 [3.6%]), and bla_KPC plus bla_NDM (1 [3.6%]). On unit admission, 27 ICU patients needed to be screened to detect 1 CRE-colonized patient, and 71 patients needed to be screened to detect 1 CP-CRE-positive patient.

TABLE 1.

Carbapenem-resistant Enterobacteriaceae (CRE) isolated from rectal swabs, comparing carbapenemase-producing and non-carbapenemase-producing CRE

Organism (no. of isolates)	Carbapenemase-producing CRE				Non-carbapenemase-producing CRE
	n	%	Carbapenemase genes identified	No. (%) that progressed to infection	n	%	No. (%) that progressed to infection
Citrobacter freundii (5)	1	20	blaKPC (1)	1 (100)	4	80
Enterobacter cloacae (33)	4	12	blaKPC (3), blaKPC plus blaNDM (1)		29	88	1 (3)
Escherichia coli (22)	3	14	blaKPC (2), blaNDM (1)	1 (33)	19	86
Klebsiella (Enterobacter) aerogenes (8)	1	12	blaKPC (1)		7	88
Klebsiella oxytoca (2)	1	50	blaKPC (1)	1 (100)	1	50
Klebsiella pneumoniae (39)	16	41	blaKPC (15), blaOXA-48 (1)	5 (31)	23	59	2 (9)
Morganella morganii (2)	0	0			2	100
Proteus mirabilis (9)	0	0			9	100
Proteus vulgaris (11)	2	18	blaKPC (2)	1 (50)	9	82
Providencia rettgeri (1)	0	0			1	100
Providencia stuartii (1)	0	0			1	100
Serratia marcescens (1)	0	0			1	100

Of the 119 patients rectally colonized with CRE, 12 (10%) developed a subsequent CRE clinical infection at any time during their hospital stay. The median duration of hospital stay (day 1 was the date the initial rectal swab was obtained) was 6 days (IQR, 3 to 11 days). The probability of developing a carbapenem-resistant infection with the colonizing organism differed based on carbapenemase status: 9 patients (36%) colonized with CP-CRE developed a CP-CRE infection during the same hospitalization versus 3 (3.2%) patients colonized with a non-CP-CRE (P < 0.05). Sites of infection (adjudicated to be an infection and not colonization by an infectious diseases physician [P.D.T.]) included respiratory (67%), urine (14%), intra-abdominal (10%), wound (7%), and blood (2%).

Researchers are increasingly exploring strategies for interrupting pathways between CRE colonization and subsequent infection (e.g., selective decontamination of the digestive tract, fecal microbiota transplantations) (9, 10). Identifying the subpopulation of CRE-colonized patients at greatest risk for subsequent infection may facilitate targeted use of these interventions. Awareness of a patient’s CP-CRE colonization status may limit patient-to-patient transmission, ensuring that appropriate infection control measures are in place. Moreover, newer β-lactam agents that exert differential activity based on the specific carbapenemases produced have recently gained U.S. Food and Drug Administration approval or are in late-stage clinical testing (16). Early identification of CP-CRE-colonized patients may help guide empirical treatment decisions with these agents when subsequent infections develop.

This was a single-center study, and the prevalence and molecular epidemiology of CRE colonization may not be generalizable to other ICUs. CRE carriage was only evaluated in the rectal region, where the largest burden of Enterobacteriaceae exists; however, this may have underestimated the true CRE burden. In addition, sequencing analysis was not undertaken to confirm that the colonizing and clinical infection CRE isolates were the same. However, a subsequent CRE infection was required to match the colonizing isolate on all available microbiological parameters, including genus and species, phenotypic carbapenem resistance profile, carbapenemase production status, and carbapenemase genes (as relevant). Our findings suggest that introduction and acquisition of CRE in ICUs is not insignificant, and CP-CRE colonized patients may be at particularly high risk of subsequent CRE infections during their hospital stay.