JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 2005, p. 5639–5641 Vol. 43, No.

11
0095-1137/05/$08.00⫹0 doi:10.1128/JCM.43.11.5639–5641.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Evaluation of Techniques for Detection of Carbapenem-Resistant

Klebsiella pneumoniae in Stool Surveillance Cultures
D. Landman,* J. K. Salvani, S. Bratu, and J. Quale
Department of Medicine, SUNY-Downstate Medical Center, Brooklyn, New York
Received 5 July 2005/Returned for modification 15 August 2005/Accepted 2 September 2005

Screening for gastrointestinal colonization with multidrug-resistant nosocomial pathogens is an important

component of infection control protocols. In the New York City region, carbapenem-resistant Klebsiella pneu-
moniae strains, which harbor the KPC carbapenem-hydrolyzing ␤-lactamase, have rapidly emerged. The
potential utility of screening medium, which involved using 10-␮g imipenem disks, was investigated. The
method of placing a sample from a fecal surveillance culture into broth containing an imipenem disk appeared
to have the greatest sensitivity for detecting KPC-producing K. pneumoniae. Gastrointestinal colonization with
two other carbapenem-resistant nosocomial pathogens, Pseudomonas aeruginosa and Acinetobacter baumannii,
was also detected using this method. Placing fecal surveillance specimens into broth containing an imipenem
disk is an easy method for screening samples for carbapenem-resistant nosocomial pathogens.

Controlling the spread of antibiotic-resistant nosocomial MATERIALS AND METHODS

pathogens involves a combination of antibiotic control mea-
sures and effective infection control strategies. For example,
recommendations for the control of vancomycin-resistant en-
terococci include reduction of vancomycin use, strict adher-
ence to infection control practices, and early detection by
screening for gastrointestinal colonization (5). Implementation
of these measures has successfully controlled the spread of this
pathogen (14). The availability of commercially prepared me-
dia that can be used to screen rectal cultures for vancomycin-
resistant enterococci facilitates laboratory detection. Similarly,
effective control strategies for controlling members of the fam-
ily Enterobacteriaceae with extended-spectrum ␤-lactamases
(ESBLs) have involved aggressive infection control protocols
that included screening for gastrointestinal colonization (11).
Unfortunately, suitable media for screening fecal cultures for
resistant gram-negative pathogens are not commercially avail-
able. Studies examining rates of gastrointestinal carriage of
cephalosporin-resistant Enterobacteriaceae strains have used
locally prepared media, including agar supplemented with cefo-
taxime (7, 10, 11, 13).
The recent and rapid spread of carbapenem-resistant Kleb-
siella pneumoniae, due to the presence of the carbapenem-
hydrolyzing ␤-lactamase KPC, in the New York City region has
been alarming (1, 2, 4, 16). This pathogen presents a formida-
ble challenge because of its high degree of resistance to virtu-
ally all classes of antibiotics (4), and controlling their spread is
of utmost importance. In this report, we investigate potential
laboratory procedures that could be used by clinical microbi-
ology laboratories for the screening of fecal specimens for
these resistant bacteria.
* Corresponding author. Mailing address: Division of Infectious
Diseases, Box 77, SUNY Downstate, 450 Clarkson Ave., Brooklyn, NY
11203. Phone: (718) 270-3790. Fax: (718) 270-2468. E-mail: dlandman
@downstate.edu.
Preliminary studies. Six previously characterized clinical strains of carbap-
enem-resistant K. pneumoniae, known to possess blaKPC-2, were included in
experiments to assess the sensitivity of three potential screening methods (1, 4).
These six isolates represented the two major ribotypes (two isolates each) in-
volved in outbreaks in New York City, and two isolates from unique ribotypes.
Susceptibility testing using Etest, the disk diffusion technique, and the broth
microdilution method was performed according to established recommenda-
tions (6).
Starting with an initial inoculum of ⬃5 ⫻ 105 CFU/ml (range, 1 ⫻ 105 to 1.7 ⫻ 106
CFU/ml), serial 10-fold dilutions of the six isolates were made in normal saline.
To include the effect of possible interference from other carbapenem-resistant
organisms that can inhabit the gastrointestinal tract, the following organisms
were added to each dilution (⬃1 ⫻ 104 CFU/ml): a clinical isolate of imipenem-
resistant Pseudomonas aeruginosa, a clinical isolate of vancomycin-resistant En-
terococcus faecium, and Candida albicans ATCC 90028. The culture mixture of
each dilution was then processed in three ways. For method 1, 100 ␮l of the
culture mixture was placed in 5 ml of tryptic soy broth containing a 10-␮g disk of
imipenem. Following an overnight incubation at 37°C, 100 ␮l was streaked onto
MacConkey agar and incubated overnight; the presence or absence of lactose-
fermenting gram-negative rods was then recorded. For method 2, 100 ␮l of the
culture mixture was placed in 5 ml of MacConkey broth containing a 10-␮g disk
of imipenem. Following an overnight incubation at 37°C, 100 ␮l was streaked
onto MacConkey agar and incubated overnight; the presence or absence of
lactose-fermenting gram-negative rods was then recorded. For method 3, 100 ␮l
of the culture mixture was placed in 5 ml of tryptic soy broth. Following an
overnight incubation at 37°C, 100 ␮l was streaked onto MacConkey agar and an
imipenem disk was applied. Isolates of K. pneumoniae growing within the zone of
inhibition (within a zone diameter of ⱕ15 mm) were recorded. Prior studies
demonstrated comparable zones of inhibition for imipenem against P. aeruginosa
ATCC 28753 when Mueller-Hinton and MacConkey agars were used (data not
shown).
Stool surveillance studies. Surveillance rectal cultures are routinely obtained
from all patients in a 10-bed medical-surgical intensive care area as part of an
infection control program to identify vancomycin-resistant enterococci. The cul-
tures are performed upon admission to the intensive care area and once weekly
during the patients’ stay. Rectal swabs are placed in BBL CultureSwab Plus
culturettes (Becton Dickinson, Sparks, MD) containing 5 ml of Amies gel trans-
port medium. During the month of February 2005, an aliquot of transport
medium was collected from each culturette before it was discarded. Imipenem-
resistant K. pneumoniae strains were endemic in the intensive care area during
that period. To ensure an equal inoculum for each method, 100 ␮l of the
transport medium was placed into 5 ml of tryptic soy broth and processed
according to methods 1 and 3 described above. All gram-negative pathogens in
method 1, and all growing within a zone diameter of ⱕ15 mm in method 3,
underwent imipenem susceptibility testing using the Etest methodology. All
imipenem-resistant isolates were then identified using the API 20E system

5639
5640 LANDMAN ET AL. J. CLIN. MICROBIOL.

TABLE 1. Comparison of three methods for the detection of six In total, eight samples had imipenem-resistant K. pneu-
clinical isolates of carbapenem-resistant K. pneumoniae moniae (four by both methods and four by method 1 only). In
Imipenem MIC Lowest limit of detection (CFU/ml) addition, 23 samples had imipenem-resistant P. aeruginosa. Of
Ribo- (␮g/ml) as determined by in vitro studies these, 13 were positive by both methods, four were positive by
Isolate
type 5 ⫻ 105 5 ⫻ 104 method 1 only, and six were positive by method 3 only.
Method 1 Method 2 Method 3
CFU/ml CFU/ml Fourteen of the 33 patients had more than one surveillance
IC807 A ⬎32 4 3.3 ⫻ 104 3.3 ⫻ 104 ⬎3.3 ⫻ 105 sample taken during the study period. The cultures from these
CK850 A ⬎32 ⬎32 1.7 ⫻ 102 1.7 ⫻ 102 1.7 ⫻ 102 patients were either all positive or all negative in 7 of 14 cases.
OW822 B 8 1 1.7 ⫻ 106 1.7 ⫻ 106 ⬎1.7 ⫻ 106 For the remaining seven patients, the initial culture was neg-
OW555 B ⬎32 32 2.7 ⫻ 100 2.7 ⫻ 101 2.7 ⫻ 103 ative whereas the subsequent culture(s) grew carbapenem-re-
IC808 C ⬎32 4 ⬎1 ⫻ 105 1 ⫻ 105 ⬎1 ⫻ 105
sistant K. pneumoniae or P. aeruginosa. The latter patients may
IC855 D ⬎32 ⬎32 3.3 ⫻ 102 3.3 ⫻ 102 3.3 ⫻ 101
have become colonized after the initial culture was taken. No
patients with an initially positive culture had a subsequent
culture that was negative. Thirteen of the 33 patients had
(bioMérieux, France) and screened by PCR for the presence of blaKPC using cultures that were all negative. All 13 patients had other (non-
previously defined primers and conditions (1, 4). Amplified blaKPC products were surveillance) cultures taken by the medical team as part of
identified with bidirectional sequencing, as previously described (1, 4).
This study was approved by the Institutional Review Board at SUNY-Down-
their routine care. None of these cultures grew carbapenem-
state Medical Center. resistant K. pneumoniae or P. aeruginosa. However, 3 of the
13 patients had routine cultures that grew carbapenem-resis-
tant A. baumannii (two from respiratory cultures and one from
RESULTS
blood culture).
Preliminary studies. All six clinical isolates used in these
studies were intermediate or resistant to imipenem using the DISCUSSION
recommended inoculum for broth micodilution susceptibility
testing (Table 1). However, three isolates had MICs of imi- Detection of antibiotic-resistant nosocomial pathogens (e.g.,
penem that were highly dependent on the inoculum used, a vancomycin-resistant enterococci or Enterobacteriaceae strains
phenomenon frequently observed in KPC-possessing K. pneu- with ESBLs) that commonly colonize the gastrointestinal tract
moniae (4). The lowest limits of detection of six clinical isolates is an integral component of successful infection control proto-
of K. pneumoniae with blaKPC-2 by the three methods are re- cols (5, 11, 14). Commercially prepared media are available
ported in Table 1. Isolates that had MICs of imipenem that that can be used by clinical laboratories for the detection of
were highly inoculum dependent were more difficult to detect, vancomycin-resistant enterococci. However, because of the in-
regardless of the method, whereas isolates that had high MICs stability of many ␤-lactams in culture media, no such media
of imipenem regardless of inoculum were more easily detected. exist for the detection of resistant gram-negative pathogens.
Methods 1 and 2 yielded comparable results. Method 3 was The difficulty of screening fecal surveillance cultures for bac-
inferior in detecting three of the isolates (requiring at least 1 teria with ESBLs has undoubtedly been an impediment in
log10 CFU/ml more than the other two methods), superior in our efforts to control the spread of these pathogens in the
the detection of one isolate, and equivalent for the remaining New York City region (15).
two isolates. Because methods 1 and 2 yielded comparable Since 2001, our region has also witnessed the rapid spread of
results, evaluation of the clinical samples was performed by carbapenem-resistant K. pneumoniae; carbapenem resistance
only methods 1 and 3. in these isolates is due to the presence of KPC, an efficient
Stool surveillance studies. Fifty-one stool surveillance cul- class A carbapenem-hydrolyzing ␤-lactamase (2, 4, 16). Be-
tures from 33 patients underwent testing by methods 1 and 3. cause these pathogens are resistant to virtually all commonly
For method 1, 13 lactose-fermenting gram-negative bacilli used antibiotics, including ␤-lactams, fluoroquinolones, ami-
were recovered from 12 samples. Of these 13 isolates, five were noglycosides, and not infrequently polymyxins (4), controlling
imipenem susceptible (MIC range, 0.25 to 3 ␮g/ml) and eight their spread is of utmost importance. As with vancomycin-
were resistant to imipenem (MIC, ⬎32 ␮g/ml). All eight resis- resistant enterococci and ESBL-possessing Enterobacteriaceae,
tant isolates were identified as K. pneumoniae with blaKPC-2. In reliance on cultures taken from patients with clinically sus-
addition, 29 non-lactose-fermenting gram-negative bacilli were pected infection may fail to identify patients harboring KPC-
recovered. Ten were found to be imipenem susceptible (MIC possessing K. pneumoniae (2). In outbreak settings, screening
range, 0.25 to 4 ␮g/ml). Of the remaining 19 isolates, 17 were for asymptomatic colonization of the gastrointestinal tract will
P. aeruginosa and two were Acinetobacter baumannii, all with be necessary to identify patients with these pathogens, so that
imipenem MICs of ⬎32 ␮g/ml. proper infection control efforts can be instituted.
When the clinical samples were evaluated using method 3, In this report, the medium that consisted of 5 ml of broth
four carbapenem-resistant K. pneumoniae strains, all with containing a 10-␮g imipenem disk resulted in the highest re-
blaKPC-2, were identified. In addition, 19 imipenem-resistant covery of KPC-possessing K. pneumoniae from stool surveil-
isolates of P. aeruginosa were identified. Since only isolates lance cultures. This method can be easily performed by clinical
growing close to the imipenem disk were evaluated, none of laboratories. It was apparent that the targeted concentration of
the isolates evaluated were found to be imipenem susceptible. imipenem (2 ␮g/ml, assuming 100% diffusion of the antibiotic)
A. baumannii was not recovered from any of the samples using was not achieved with this method, as several of the pathogens
method 3. isolated by this method had imipenem MICs of between 0.25
VOL. 43, 2005
and 4 ␮g/ml. Confirming imipenem resistance by a disk diffu-
sion assay prior to identification of these isolates would mini-
mize unnecessary work, although it would delay reporting by
another 24 h. Increasing the number of imipenem disks placed
into the broth may reduce the likelihood of recovering imi-
penem-susceptible bacteria. The alternative method (plating
an overnight growth of the surveillance culture onto MacCon-
key agar with an imipenem disk and identifying only the iso-
lates near the disk) eliminated the problem of the recovery of
carbapenem-susceptible bacteria but appeared less sensitive
for the detection of KPC-possessing K. pneumoniae. blaKPC has
been found in other bacteria, including Salmonella enterica,
Klebsiella oxytoca, and Enterobacter spp. (3, 8, 12, 17). An
assessment of these methods for detecting other KPC-produc-
ing Enterobacteriaceae in stool surveillance cultures will have to
be performed.
Our region has also been beset with the clonal spread of
two other carbapenem-resistant pathogens, A. baumannii and
P. aeruginosa (9). An unexpected finding in our study was the
relatively large number of stool samples harboring these iso-
lates, since they are not commonly recognized as inhabitants of
the gastrointestinal tract. It appears that the laboratory meth-
ods outlined in this study will also be useful in detecting gas-
trointestinal colonization with these pathogens and assisting
infection control efforts.
REFERENCES
1. Bradford, P. A., S. Bratu, C. Urban, M. Visalli, N. Mariano, D. Landman,
J. J. Rahal, S. Brooks, S. Cebular, and J. Quale. 2004. Emergence of car-
bapenem-resistant Klebsiella spp. possessing the class A carbapenem-hydrolyz-
ing KPC-2 and inhibitor-resistant TEM-30 ␤-lactamases in New York City. Clin.
Infect. Dis. 39:55–60.
2. Bratu, S., D. Landman, R. Haag, R. Recco, A. Eramo, M. Alam, and J.
Quale. 2005. Rapid spread of carbapenem-resistant Klebsiella pneumoniae in
New York City: a new threat to our antibiotic armamentarium. Arch. Intern.
Med. 165:1430–1435.
3. Bratu, S., D. Landman, M. Alam, E. Tolentino, and J. Quale. 2005. Detec-
tion of KPC carbapenem-hydrolyzing enzymes in Enterobacter spp. from
Brooklyn, New York. Antimicrob. Agents Chemother. 49:776–778.
4. Bratu, S., M. Mooty, S. Nichani, D. Landman, C. Gullans, B. Pettinato, U.
Karumudi, P. Tolaney, and J. Quale. 2005. Emergence of KPC-possessing
Klebsiella pneumoniae in Brooklyn, New York: epidemiology and recommen-
dations for detection. Antimicrob. Agents Chemother. 49:3018–3020.
DETECTION OF RESISTANT K. PNEUMONIAE 5641
5. Centers for Disease Control and Prevention. 1995. Recommendations for
preventing the spread of vancomycin resistance. Recommendations of the
Hospital Infection Control Practices Advisory Committee. Morb. Mortal.
Wkly. Rep. 44:1–13.
6. Clinical and Laboratory Standards Institute. 2005. Performance standards
for antimicrobial susceptibility testing. Document M100-S15. Clinical and
Laboratory Standards Institute, Wayne, Pa.
7. Decre, D., B. Gachot, J. C. Lucet, G. Arlet, E. Bergogne-Berezin, and B.
Regnier. 1998. Clinical and bacteriologic epidemiology of extended-spec-
trum ␤-lactamase-producing strains of Klebsiella pneumoniae in a medical
intensive care unit. Clin. Infect. Dis. 27:834–844.
8. Hossain, A., M. J. Ferraro, R. M. Pino, R. B. Dew III, E. S. Moland, T. J.
Lockhart, K. S. Thomson, R. V. Goering, and N. D. Hanson. 2004. Plasmid-
mediated carbapenem-hydrolyzing enzyme KPC-2 in an Enterobacter sp.
Antimicrob. Agents Chemother. 48:4438–4440.
9. Landman, D., J. M. Quale, D. Mayorga, A. Adedeji, K. Vangala, J. Ravi-
shankar, C. Flores, and S. Brooks. 2002. Citywide clonal outbreak of mul-
tiresistant Acinetobacter baumannii and Pseudomonas aeruginosa in Brook-
lyn, N.Y. Arch. Intern. Med. 162:1515–1520.
10. Lucet, J.-C., S. Chevret, D. Decre, D. Vanjak, A. Macrez, J.-P. Bedos, M.
Wolff, and B. Regnier. 1996. Outbreak of multiply resistant Enterobacteri-
aceae in an intensive care unit: epidemiology and risk factors for acquisition.
Clin. Infect. Dis. 22:430–436.
11. Lucet, J.-C., D. Decre, A. Fichelle, M.-L. Joly-Guillou, M. Pernet, C. De-
blangy, M.-J. Kosmann, and B. Regnier. 1999. Control of a prolonged
outbreak of extended-spectrum ␤-lactamase-producing Enterobacteriaceae
in a university hospital. Clin. Infect. Dis. 29:1411–1418.
12. Miriagou, V., L. S. Tzouvelekis, S. Rossiter, E. Tzelepi, F. J. Angulo, and
J. M. Whichard. 2003. Imipenem resistance in a Salmonella clinical strain
due to plasmid-mediated class A carbapenemase KPC-2. Antimicrob. Agents
Chemother. 47:1297–1300.
13. Moustaoui, N., R. Bensghir, K. Mjahed, K. Hakim, R. Aimhand, M. Bou-
douma, L. Barrou, N. Elmdaghri, and M. Benbachir. 2000. Digestive tract
colonization with extended spectrum beta-lactamase producing Enterobac-
teriaceae in a surgical intensive care unit in Casablanca. J. Hosp. Infect.
46:238–240.
14. Ostrowsky, B. E., W. E. Trick, A. H. Sohn, S. B. Quirk, S. Holt, L. A. Carson,
B. C. Hill, M. J. Arduino, M. J. Kuehnert, and W. R. Jarvis. 2001. Control
of vancomycin-resistant enterococcus in health care facilities in a region.
N. Engl. J. Med. 344:1427–1433.
15. Quale, J. M., D. Landman, P. A. Bradford, M. Visalli, J. Ravishankar, C.
Flores, D. Mayorga, K. Vangala, and A. Adedeji. 2002. Molecular epidemi-
ology of a citywide outbreak of extended-spectrum ␤-lactamase-producing
Klebsiella pneumoniae infection. Clin. Infect. Dis. 35:834–841.
16. Woodford, N., P. M. Tierno, Jr., K. Young, L. Tysall, M.-F. I. Palepou, E.
Ward, R. E. Painter, D. F. Suber, D. Shungu, L. L. Silver, K. Inglima, J.
Kornblum, and D. M. Livermore. 2004. Outbreak of Klebsiella pneumoniae
producing a new carbapenem-hydrolyzing class A ␤-lactamase, KPC-3, in a
New York medical center. Antimicrob. Agents Chemother. 48:4793–4799.
17. Yigit, H., A. M. Queenan, J. K. Rasheed, J. W. Biddle, A. Domenech-
Sanchez, S. Alberti, K. Bush, and F. C. Tenover. 2003. Carbapenem-resistant
strain of Klebsiella oxytoca harboring carbapenem-hydrolyzing ␤-lactamase
KPC-2. Antimicrob. Agents Chemother. 47:3881–3889.