Abstract
This study evaluated the impact of preenrichment on the detection of extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBL-E) in clinical stool samples. ESBL-E were detected in 41 of 343 patients (12.0%). As 31.7% of the ESBL-E carriers were identified by preenrichment, only this additional diagnostic step significantly improved the detection of ESBL-E.
TEXT
Extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBL-E) are increasingly reported worldwide. Treatment options for infections due to these organisms are limited, and initial empirical therapy is often inappropriate and is associated with increased morbidity and mortality (1). ESBL-E carriers are known to develop bacteremia with the isolates colonizing the patients' intestines (2). Therefore, in many hospitals, risk-adapted screening for ESBL-E carriage is performed, but limited data exist on the optimal screening strategy. Currently, laboratory detection of ESBL-E in fecal specimens mostly relies on culture methods using selective agar medium. This method has a modest sensitivity (3). Preenrichment of swabs is used by many laboratories for methicillin-resistant Staphylococcus aureus (MRSA) screening and has demonstrated significantly improved sensitivity (4). To our knowledge, the impact of preenrichment on ESBL-E detection in stool samples has not been systematically investigated. In our study, we compared three different selective and nonselective broths for the detection of ESBL-E in clinical stool samples to the standard procedure—direct plating on ESBL agar.
(This study was presented in part as a poster at the 25th European Congress of Clinical Microbiology and Infectious Diseases, Copenhagen, Denmark, 25 to 28 April 2015.)
From September 2013 to March 2014, 496 stool samples submitted for ESBL-E screening from 343 patients at the University Hospital Cologne, a 1,400-bed tertiary care facility, were included in the study. All samples were stored at 4°C and were processed within 24 h of receipt. A pea-sized portion of solid stool or 250 μl of liquid stool was suspended in 1 ml 0.9% NaCl to reach a standardized inoculum. One hundred microliters of the stool suspension was inoculated directly onto chromID ESBL agar (bioMérieux, Marcy l'Etoile, France) and into 5 ml of nonselective tryptic soy broth (TSB), MacConkey (MC) broth (Roth, Germany, Karlsruhe) selecting only Gram-negative rods, and MacConkey broth supplemented with cefuroxime (32 mg/liter) and vancomycin (64 mg/liter) (MC-CV) additionally inhibiting the nonresistant Gram-negative flora and further inhibiting Gram-positive bacteria. Each medium was stored and processed according to the manufacturers' instructions.
Agar plates and broths were incubated at 36°C ± 1°C in ambient air for 18 h to 24 h. Using calibrated inoculation loops, 10 μl of the respective enrichment broth was inoculated onto ESBL agar, which was incubated at 36°C ± 1°C in ambient air for another 18 h to 24 h. Plates were interpreted according to the manufacturer's instructions. Uncolored colonies growing on ESBL agar were tested for oxidase production using an oxidase test strip. Oxidase-positive colonies were not further analyzed. All other colonies growing on the ESBL agar were identified by matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry (Bruker Daltonik, Bremen, Germany). ESBL production was tested phenotypically by the combination disk diffusion test using cefotaxime (30 μg), ceftazidime (30 μg), and cefepime (30 μg) disks alone and in combination with the beta-lactamase inhibitor clavulanic acid (10 μg) (Mast Diagnostics, Merseyside, United Kingdom). The combination disk diffusion test has been shown to have a high sensitivity (5) and is recommended by the CLSI (6) and EUCAST (7). The susceptibility of all phenotypically ESBL-producing Enterobacteriaceae was determined using the Vitek 2 system (bioMérieux, Marcy-l'Etoile, France). MICs were interpreted according to EUCAST breakpoints. Detection of ESBL resistance genes blaTEM, blaSHV, and the blaCTX-M-1, blaCTX-M-2, and blaCTX-M-9 groups was performed by multiplex PCR assay as described previously (8). Identification of the TEM and SHV types was done by sequencing PCR products.
Performance characteristics of the four different protocols were calculated on the total number of ESBL-E-positive samples. A sample was considered ESBL-E positive when the ESBL phenotype displayed in the combination disk diffusion test was confirmed by molecular characterization. As the number of samples per patient was variable (between 1 and 7 stool samples per patient), the modified Obuchowski test for clustered paired binary data (9) was used to analyze the difference between the four protocols. The significance level of 5% was corrected for multiple testing using the Bonferroni method (i.e., α = 0.05/5 = 0.01).
During the 5-month study period, 496 stool samples from 343 patients were analyzed. In total, 48 ESBL-E isolates from 45 positive stool samples of 41 patients were detected by either direct culture or culture with one of the preenrichment broths (i.e., three stool samples were found to have two different ESBL-E isolates). Thirteen patients (31.7%) were identified by preenrichment only.
Using direct culture on ESBL agar, ESBL-E isolates were only detected in 32 (71.1%) of the 45 positive stool samples. Using preenrichment in TSB, 40 positive stool samples (88.9%) were identified compared to 41 (91.1%) positive stool samples identified by preenrichment in MC broth and in MC-CV broth. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the four different screening protocols on a sample level are shown in Table 1. Escherichia coli was the predominant ESBL-positive species (n = 73%); most isolates belonged to the CTX-M-1 group (Table 2). All preenrichment protocols significantly improved sensitivity compared to that of direct culture on ESBL agar (TSB versus direct culture, P = 0.00017; MC versus direct culture, P = 0.00144; MC-CV versus direct culture, P = 3.9 × 10−6). There was no significant difference in the performance of the three different preenrichment broths (P > 0.01).
TABLE 1.
Performance characteristics of the four different screening protocols for detection of ESBL-E
| Screening protocol and result | No. of stool samples with result: |
Performance characteristics |
||||
|---|---|---|---|---|---|---|
| ESBL positive | ESBL negativea | Sensitivity, % (95% CI) | Specificity, % (95% CI) | PPV, % (95% CI) | NPV, % (95% CI) | |
| chromID ESBL (no enrichment) | 71.1 (56.6–82.3) | 83.8 (80.1–86.9) | 30.5 (22.5–39.8) | 96.7 (94.4–98.0) | ||
| Growth (n = 105) | 32 | 73 | ||||
| No growth (n = 391) | 13 | 378 | ||||
| Total (n = 496) | 45 | 451 | ||||
| Pre-enrichment in TSB | 88.9 (76.5–95.2) | 80.0 (76.1–83.5) | 30.8 (23.5–39.2) | 98.6 (96.8–99.4) | ||
| Growth (n = 130) | 40 | 90 | ||||
| No growth (n = 366) | 5 | 361 | ||||
| Total (n = 496) | 45 | 451 | ||||
| Preenrichment in MC broth | 91.1 (79.3–96.5) | 80.9 (77.1–84.3) | 32.3 (24.8–40.8) | 98.9 (97.2–99.6) | ||
| Growth (n = 127) | 41 | 86 | ||||
| No growth (n = 369) | 4 | 365 | ||||
| Total (n = 496) | 45 | 451 | ||||
| Preenrichment in MC-CV broth | 91.1 (79.3–96.5) | 78.7 (74.7–82.2) | 29.9 (22.9–38.1) | 98.9 (97.2–99.6) | ||
| Growth (n = 137) | 41 | 96 | ||||
| No growth (n = 359) | 4 | 355 | ||||
| Total (n = 496) | 45 | 451 | ||||
Enterobacteriaceae resistant to third-generation cephalosporins (as grown on selective agar) but not ESBL (i.e., AmpC or others); growth of oxidase-positive bacteria was not included in the calculation.
TABLE 2.
Distribution of ESBL resistance genes blaTEM, blaSHV, and the blaCTX-M-1 and blaCTX-M-9 groups in the different species among the 48 ESBL isolates derived from 45 clinical stool samples
| Species | CTX-M-1 |
CTX-M-9 |
SHV |
CTX-M-1 + CTX-M-9 |
TEM + CTX-M-1 |
Total |
||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | |
| Escherichia coli | 24 | 68.5 | 6 | 17 | 1 | 3 | 3 | 8.5 | 1 | 3 | 35 | 73 |
| Klebsiella pneumoniae | 5 | 50 | 5 | 50 | 10 | 21 | ||||||
| Enterobacter cloacae | 2 | 100 | 2 | 4 | ||||||||
| Citrobacter freundii | 1 | 100 | 1 | 2 | ||||||||
| Total (no.) | 30 | 63 | 8 | 17 | 6 | 12 | 3 | 6 | 1 | 2 | 48 | 100 |
The group of isolates that was detected by preenrichment only was compared to the group of isolates recovered by direct culture. There were no substantial differences between the two groups (Tables 3 to 5) regarding species, susceptibility, or patient characteristics.
TABLE 3.
Strain characteristics of isolates recovered by preenrichment only compared to those of isolates recovered by both methods
| Strain characteristic | No. recovered by both methods (%) (n = 34) | No. recovered by preenrichment only (%) (n = 14) |
|---|---|---|
| Species | ||
| Escherichia coli | 24 (70.7) | 11 (78.6) |
| Klebsiella pneumoniae | 8 (23.5) | 2 (14.3) |
| Enterobacter cloacae | 1 (2.9) | 1 (7.1) |
| Citrobacter freundii | 1 (2.9) | 0 |
| ESBL resistance gene(s) | ||
| CTX-M-1 group | 20 (58.8) | 10 (71.5) |
| CTX-M-9 group | 6 (17.6) | 2 (14.3) |
| SHV | 5 (14.7) | 1 (7.1) |
| CTX-M-1 group + CTX-M-9 group | 2 (5.9) | 1 (7.1) |
| TEM + CTX-M-1 group | 1 (2.9) | 0 |
TABLE 5.
Patient characteristics for isolates recovered by preenrichment only compared to those of isolates recovered by both methods
| Patients characteristics (patient levela) | Recovered by both methods (n = 28) | Recovered by preenrichment only (n = 13) |
|---|---|---|
| Age (yr) | 50 (±26) | 37 (±35) |
| No. male (%) | 16 (47) | 9 (64.2) |
| Clinical setting (no. [%]) | ||
| Internal medicine | 21 (75) | 4 (28.6) |
| Pediatrics | 3 (10.7) | 6 (46.2) |
| Intensive care unit | 1 (3.6) | 1 (7.7%) |
| Surgery | 3 (10.7) | 2 (15.4) |
Calculation based on a patient level (i.e., patients with more than one positive sample were only included in the calculation once).
TABLE 4.
Susceptibility characteristics of isolates recovered by preenrichment only compared to those of isolates recovered by both methods
| Drug susceptibility | Recovered by both methods (n = 34) |
Recovered by preenrichment only (n = 14) |
||||
|---|---|---|---|---|---|---|
| MIC50a | MIC90a | I/Ra (no., %) | MIC50 | MIC90 | I/R (no., %) | |
| Piperacillin-tazobactam | 8 | 64 | 13 (38.2) | <4 | >128 | 4 (28.6%) |
| Cefuroxime | >64 | >64 | 31 (91.2) | >64 | >64 | 14 (100%) |
| Cefpodoxime | >8 | >8 | 34 (100) | >8 | >8 | 14 (100%) |
| Cefotaxime | >64 | >64 | 32 (94.1) | >64 | >64 | 14 (100%) |
| Ceftazidime | 16 | 16 | 24 (70.6) | 8 | >64 | 9 (64.3%) |
| Meropenem | <0.25 | <0.25 | 0 (0) | <0.25 | <0.25 | 0 (0%) |
| Ciprofloxacin | >4 | >4 | 22 (64,7) | <0.25 | >4 | 5 (35.7%) |
| Trimethoprim-sulfamethoxazole | >320 | >320 | 23 (67.6) | >320 | >320 | 9 (64.3%) |
MIC50/MIC90 estimates the antibiotic concentration (mg/liter) that inhibits 50% (MIC50) and 90% (MIC90) of tested bacterial isolates; I/R, number of isolates considered to be intermediate (I) or resistant (R) to the indicated antimicrobial agent according to EUCAST clinical breakpoints.
Currently, the use of preenrichment broth for ESBL-E is not a common practice in clinical or research settings and has been discussed controversially in the past (10, 11). A more recent report from the Netherlands showed that >25% of the ESBL-E carriers may only be identified by preenrichment when a rectal swab was used (12). Rectal swabs are believed to be inadequate for the recovery and detection of various pathogens due to high variation in the quantity of fecal material on each swab and the low inoculum (13). Therefore, in our study, we used stool samples, the gold standard specimen, for the screening of gastrointestinal bacteria.
Our data clearly demonstrate the benefit of a preenrichment step in the sensitivity of clinical diagnostic screening for ESBL-E independent of the enrichment broth used. Thus, if only ESBL producers are in the field of interest, we suggest the use of TSB or MC broth, both of which are inexpensive, easy to prepare, and highly sensitive. The specificity of the ESBL agar in our study was comparable to former studies (14, 15) and was not significantly affected by the use of enrichment broth. A trend to lower specificity was observed for the MC-CV broth, most likely because cefuroxime is an inducer of the AmpC beta-lactamase and is therefore more ESBL-negative, but third-generation-cephalosporin-resistant Enterobacteriaceae can be recovered with the MC-CV broth (see Table 1). Thus, if Enterobacteriaceae resistant to third-generation cephalosporin with resistance mechanisms other than ESBL (e.g., AmpC producers) should be detected, the use of MC-CV broth is recommended.
In conclusion, a higher detection rate of ESBL can be achieved by using stool samples in combination with a preenrichment step before plating on selective ESBL agar. However, with this strategy, results are available 24 h later, which might be a disadvantage in some settings but can be overcome by direct culturing side-by-side. This approach is associated with higher costs, an increased work load (stool sample collection), and increased hands-on time. However, as by now there are no reasonable molecular solutions (16), this screening strategy might be the best option to maximize the sensitivity of the screening process and might therefore improve the empirical treatment of colonized patients and the prevention of the nosocomial spread of ESBL-E.
ACKNOWLEDGMENT
We declare no conflicts of interest.
REFERENCES
- 1.Rottier WC, Ammerlaan HS, Bonten MJ. 2012. Effects of confounders and intermediates on the association of bacteraemia caused by extended-spectrum beta-lactamase-producing Enterobacteriaceae and patient outcome: a meta-analysis. J Antimicrob Chemother 67:1311–1320. doi: 10.1093/jac/dks065. [DOI] [PubMed] [Google Scholar]
- 2.Cornejo-Juarez P, Suarez-Cuenca JA, Volkow-Fernandez P, Silva-Sanchez J, Barrios-Camacho H, Najera-Leon E, Velazquez-Acosta C, Vilar-Compte D. 27 May 2015. Fecal ESBL Escherichia coli carriage as a risk factor for bacteremia in patients with hematological malignancies. Support Care Cancer doi: 10.1007/s00520-015-2772-z. [DOI] [PubMed] [Google Scholar]
- 3.Overdevest IT, Willemsen I, Elberts S, Verhulst C, Kluytmans JA. 2011. Laboratory detection of extended-spectrum-beta-lactamase-producing Enterobacteriaceae: evaluation of two screening agar plates and two confirmation techniques. J Clin Microbiol 49:519–522. doi: 10.1128/JCM.01953-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bocher S, Smyth R, Kahlmeter G, Kerremans J, Vos MC, Skov R. 2008. Evaluation of four selective agars and two enrichment broths in screening for methicillin-resistant Staphylococcus aureus. J Clin Microbiol 46:3136–3138. doi: 10.1128/JCM.00478-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Drieux L, Brossier F, Sougakoff W, Jarlier V. 2008. Phenotypic detection of extended-spectrum beta-lactamase production in Enterobacteriaceae: review and bench guide. Clin Microbiol Infect 14(Suppl):90–103. doi: 10.1111/j.1469-0691.2007.01846.x. [DOI] [PubMed] [Google Scholar]
- 6.Clinical and Laboratory Standards Institute. 2012. Performance standards for antimicrobial susceptibility testing; 22nd informational supplement. CLSI M100-S22. Clinical and Laboratory Standards Institute, Wayne, PA. [Google Scholar]
- 7.EUCAST. 2013. EUCAST guideline for the detection of resistance mechanisms and specific resistances of clinical and/or epidemiological importance. http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Resistance_mechanisms/EUCAST_detection_of_resistance_mechanisms_v1.0_20131211.pdf.
- 8.Dallenne C, Da Costa A, Decre D, Favier C, Arlet G. 2010. Development of a set of multiplex PCR assays for the detection of genes encoding important beta-lactamases in Enterobacteriaceae. J Antimicrob Chemother 65:490–495. doi: 10.1093/jac/dkp498. [DOI] [PubMed] [Google Scholar]
- 9.Yang Z, Sun X, Hardin JW. 2010. A note on the tests for clustered matched-pair binary data. Biom J 52:638–652. doi: 10.1002/bimj.201000035. [DOI] [PubMed] [Google Scholar]
- 10.Diederen B, Chang C, Euser S, Stuart JC. 2012. Evaluation of four screening protocols for detection of extended-spectrum beta-lactamase-producing members of the Enterobacteriaceae. J Med Microbiol 61:452–453. doi: 10.1099/jmm.0.036467-0. [DOI] [PubMed] [Google Scholar]
- 11.Murk JL, Heddema ER, Hess DL, Bogaards JA, Vandenbroucke-Grauls CM, Debets-Ossenkopp YJ. 2009. Enrichment broth improved detection of extended-spectrum-beta-lactamase-producing bacteria in throat and rectal surveillance cultures of samples from patients in intensive care units. J Clin Microbiol 47:1885–1887. doi: 10.1128/JCM.01406-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kluytmans-van den Bergh MF, Verhulst C, Willemsen LE, Verkade E, Bonten MJ, Kluytmans JA. 2015. Rectal carriage of extended-spectrum-beta-lactamase-producing Enterobacteriaceae in hospitalized patients: selective preenrichment increases yield of screening. J Clin Microbiol 53:2709–2712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.D'Agata EM, Gautam S, Green WK, Tang YW. 2002. High rate of false-negative results of the rectal swab culture method in detection of gastrointestinal colonization with vancomycin-resistant enterococci. Clin Infect Dis 34:167–172. doi: 10.1086/338234. [DOI] [PubMed] [Google Scholar]
- 14.Reglier-Poupet H, Naas T, Carrer A, Cady A, Adam JM, Fortineau N, Poyart C, Nordmann P. 2008. Performance of chromID ESBL, a chromogenic medium for detection of Enterobacteriaceae producing extended-spectrum beta-lactamases. J Med Microbiol 57:310–315. doi: 10.1099/jmm.0.47625-0. [DOI] [PubMed] [Google Scholar]
- 15.Grohs P, Tillecovidin B, Caumont-Prim A, Carbonnelle E, Day N, Podglajen I, Gutmann L. 2013. Comparison of five media for detection of extended-spectrum beta-lactamase by use of the wasp instrument for automated specimen processing. J Clin Microbiol 51:2713–2716. doi: 10.1128/JCM.00077-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gazin M, Paasch F, Goossens H, Malhotra-Kumar S. 2012. Current trends in culture-based and molecular detection of extended-spectrum-beta-lactamase-harboring and carbapenem-resistant Enterobacteriaceae. J Clin Microbiol 50:1140–1146. doi: 10.1128/JCM.06852-11. [DOI] [PMC free article] [PubMed] [Google Scholar]