A Triple-Disk Enrichment Method for Carbapenem-Resistant Enterobacteriaceae (CRE) Screening

LETTER

The global emergence of carbapenem-resistant Enterobacteriaceae (CRE) is a major concern for public health. The Centers for Disease Control and Prevention (CDC)-recommended procedure for surveillance is overnight incubation of a rectal/perianal swab in 5 ml of tryptic soy broth (TSB) supplemented with a 10-μg ertapenem (ETP) or 5-μg meropenem (MEM) disk, followed by subculture to MacConkey agar. Any lactose-fermenting Gram-negative rods are identified, and susceptibility and/or carbapenemase testing is performed (1). However, this method yields a high number of CRE-negative turbid broths. The development and validation of a TSB enrichment method with ertapenem, fluconazole, and linezolid (TSB-EFL) are presented as an alternative to the CDC enrichment method.

(The results of this study were partially presented at the 114th General Meeting of the American Society for Microbiology, Boston, MA, 17 to 20 May 2014 [poster 2527].)

Rectal swabs from 487 intensive care unit (ICU) patients were collected and placed into 10-ml TSB tubes (Hardy, Santa Maria, CA) and homogenized for 10 s. A 5-ml aliquot was immediately taken and placed into a separate tube with three TSB-EFL antimicrobial disks. To the remaining 5-ml aliquot (including the original patient swab), a single ETP disk was added. Both tubes were incubated aerobically for 24 h at 35°C and visually inspected (after inversion) for turbidity. Nonturbid tubes were considered negative, whereas all turbid tubes were further evaluated by both culture and molecular methods of detection. All turbid screening broths were tested by a laboratory-developed, New York State Department of Health Clinical Laboratory Evaluation Program-approved (2) PCR assay targeting the bla_KPC gene. In addition, a small subset of nonturbid broths were tested by PCR and culture to demonstrate that nonturbid broths are true negatives. PCR-positive broths were subcultured to chromogenic Klebsiella pneumoniae carbapenemase (KPC) agar (ChromeAgar, France) from which identification and sensitivity determination were performed for epidemiological purposes. Identification was performed by the matrix-assisted laser desorption ionization time of flight method (Biotyper LT; Bruker, Billerica, MA), and sensitivity was determined with the Sensititre System (Thermo Fisher, Waltham, MA).

For limit-of-detection (LOD) experiments, 90 CRE isolates with various ETP and MEM MICs (Table 1) were serially diluted 100-fold in saline solution starting with a 0.5 McFarland standard. A 100-μl volume was then inoculated into TSB-EFL, a small amount of CRE-negative stool was added, and the mixture was incubated for 24 h. Estimation of CFU counts for LOD experiments was performed in triplicate.

TABLE 1.

MICs for reference and previous positive strains used for performance evaluation of broth enrichment methods

Organism	No. of isolates	No. of isolatesa with:
		Ertapenem MIC (μg/ml) of:				Meropenem MIC (μg/ml) of:
		≤0.5 (S)	1.0 (I)	2.0–4.0 (R)	>4.0 (R)	≤1.0 (S)	2.0 (I)	4.0 (R)	>4.0 (R)
Aeromonas hydrophila	1	0	0	0	1	0	1	0	0
Citrobacter freundii	8	1	1	5	1	4	0	2	2
Citrobacter amalonaticus	1	0	0	1	0	0	0	1	0
Enterobacter aerogenes	1	0	0	0	1	1	0	0	0
Escherichia coli	7	2	0	2	3	4	1	1	1
Enterobacter cloacae	19	0	0	7	12	1	6	5	7
Klebsiella oxytoca	3	0	1	1	1	0	1	1	1
Klebsiella pneumoniae	48	2	1	20	25	9	7	9	23
Providencia rettgeri	1	0	0	1	0	0	0	0	1
Pantoea species	1	0	1	0	0	1	0	0	0
Total	90	5	4	37	44	20	16	19	35

^aClinical interpretation of MICs was done in accordance with CLSI document M100-S25. S, susceptible; I, intermediate; R, resistant.

During clinical parallel testing, 290 of 483 CRE-negative turbid broths were observed when using TSB-ETP. In contrast, only 47 CRE-negative TSB-EFL broths were turbid (Table 2). The causes of the false turbid TSB-EFL broths were that 19 broths grew Pseudomonas aeruginosa, 10 had Gram-positive flora, 7 were turbid because of excess specimen, 5 grew other non-Enterobacteriaceae species, and 6 grew various species of Enterobacteriaceae (all Enterobacteriaceae isolates were confirmed CRE negative by Check-Direct CPE assay).

TABLE 2.

Comparison of performance of enrichment methods with surveillance specimens

Method and result(s)	No. of samples
	KPC positive	KPC negative	Total
TSB-ETP
Turbidity positive	4	290	294
Turbidity negative	0	193	193
Both	4	483	487
TSB-EFL
Turbidity positive	4	47	51
Turbidity negative	0	436	436
Both	4	483	487

Our laboratory uses TSB-EFL enrichment in combination with molecular methods for the screening of any patient admitted to our ICU. Since its implementation in 2012, we have identified at least 93 new CRE carriers. Incidentally, the first Citrobacter freundii isolate producing NDM reported in the United States was recovered by the TSB-EFL method (confirmation by the New York State Department of Health Wadsworth Center, Albany, NY, May 2013).

We found that when a fluconazole and linezolid disk was added to the CDC TSB-ETP enrichment method, TSB-EFL decreased the number of turbid CRE-negative broths by 6-fold, saving a substantial amount of time and money. A study by Cheng et al. reported that the average level of colonization in a CRE carrier was as low as 500 CFU/g of stool sample (3). LOD testing revealed that >75% of positive results were recovered with 200 CFU/ml or less, and our low LOD was 2 CFU/ml. The implementation of the TSB-EFL enrichment broth is an inexpensive, readily available, and simple screening method that can be used to reduce the burden associated with the screening of CREs.