Skip Navigation

High Rate of False-Negative Results of the Rectal Swab Culture Method in Detection of Gastrointestinal Colonization with Vancomycin-Resistant Enterococci

  1. Erika M. C. D Agata1,a,
  2. Shiva Gautam2,
  3. William K. Green1, and
  4. Yi-Wei Tang1
  1. 1Department of Medicine, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee
  2. 2Department of Preventive Medicine, Division of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee
  1. Reprints or correspondence: Dr. Erika D'Agata, Beth Israel Deaconess Medical Center, Div. of Infectious Diseases, 330 Brookline Ave., Mailstop 435G, Boston, MA 02215 (edagata{at}caregroup.harvard.edu).

  • ↵Present affiliation: Division of Infectious Diseases, Beth Israel Deaconess Medical Center, Boston.

 
Next Section

Abstract

The diagnostic accuracy of the rectal swab (RS) culture method in identifying gastrointestinal colonization with vancomycin-resistant enterococci (VRE) is not known. Serial quantitative stool cultures, skin cultures, and RS cultures were performed for patients with VRE infections to assess the false-negative rate of the RS and the prevalence of skin colonization, a prerequisite for cross-transmission, at varying VRE stool densities. A total of 35 stool samples were obtained from 13 patients. The sensitivity of the RS culture was 58%; it ranged from 100%, at VRE densities of ⩾7.5 log10 colony forming units (cfu) per gram of stool, to 0%, at densities of ⩽4.5 log10 cfu per gram of stool. Skin colonization was detected at these low VRE stool densities, but it was more common at higher VRE densities (P < .001). Antibiotic exposure was significantly associated with higher VRE stool densities (P < .001). The high false-negative rate of the RS may be contributing to the continued increase in the prevalence of VRE.

The prevalence of vancomycin-resistant enterococci (VRE) colonization and infection continues to increase throughout the world [1]. To prevent and control the dissemination of VRE throughout health care institutions, several infection-control measures have been recommended, including patient isolation, patient cohorting, and wearing of gloves and gowns during patient contact [2]. These VRE-related precautions attempt to prevent patient-to-patient transmission of VRE via the hands of health care workers who are in contact with the contaminated skin or surrounding inanimate surfaces of colonized patients [3–6]. Decisions regarding the implementation or removal of patients from these precautions are frequently made on the basis of the results of the rectal swab (RS) culture method, which is used to detect the presence or presumed eradication of gastrointestinal colonization with VRE [2].

Despite its widespread application and its major role in dictating infection control precautions, the sensitivity of the RS culture method is not known. A low density of VRE in the stool may not be detected by the RS culture, yet it may still result in skin contamination, a prerequisite for VRE dissemination. To test this hypothesis, serial quantitative stool cultures, skin cultures, and RS cultures were performed for patients with VRE infection. The aims of this study were to determine the sensitivity of the RS culture for the detection of VRE colonization of the gastrointestinal tract at varying VRE stool densities, to analyze the association between VRE stool density and skin colonization, and to study the effect of antibiotic exposure, which is a risk factor for VRE colonization [7], on VRE stool density.

Previous SectionNext Section

Methods

Patient population and data collection. The study was performed at Vanderbilt University Medical Center (VUMC; Nashville, Tennessee), a 663-bed tertiary-care hospital with an average of 28,000 patient discharges per year. From October 1999 through March 2000, patients with clinical cultures that yielded VRE were identified by daily review of microbiology records. After obtaining informed consent, demographic and clinical data were collected from medical charts. The Charlson score, which generates a composite value of preexisting medical conditions, was used as a measure of comorbidity [8]. Patients' ambulatory status and presence of diarrhea (⩾2 loose or watery stools in the previous 24 h) or VRE wound infection were documented at each specimen collection. Infections were classified according to the definitions of the Centers for Disease Control and Prevention [9]. Data regarding antibiotics that had been administered within the 7 days before stool sample collection were obtained from a computerized pharmacy database. Antibiotics with activity predominantly against anaerobes (metronidazole, imipenem, clindamycin, piperacillin-tazobactam, ampicillin-sulbactam, cefotetan) were grouped together.

Appropriate informed consent was obtained and clinical research was conducted in accordance with the guidelines for experimentation on humans, as specified by the US Department of Health and Human Services and the Institutional Review Board of VUMC.

Collection of specimens. Stool samples were obtained weekly from patients until discharge or death, then refrigerated at 4°C. Within 24 h, a weighed portion of feces was suspended in 15% glycerol and 0.9% saline, diluted to a final volume of 10 mL, and frozen at −70°C. Skin samples and RSs were obtained for culture within 24 h of stool sample collection with cotton swabs, which were moistened with sterile saline. RSs were required to be coated with stool. To detect skin colonization with VRE, 2 skin sites, frequently used for venipuncture, were chosen: a 10 × 5–cm area above the right antecubital fossa and an area above the right anterior triangle of the neck. The same swab was used for both sites.

VRE quantification. Stool samples were thawed at room temperature. Serial 10-fold dilutions of up to 10-12 g/mL were made with 0.9% saline. Fifty microliters of each dilution were streaked onto Enterococcosel media (Becton Dickson) supplemented with 6 µg/mL of vancomycin and incubated at 35°C for 48 h. Plates with 30–300 colonies were used for bacterial counts. Ten representative colonies from these plates were subcultured onto sheep blood agar.

Processing of skin and RS cultures. Within 2 h of collection, skin and RSs were immersed or streaked onto Enterococcosel broth or agar, respectively; supplemented with 6 µg/mL of vancomycin; and incubated at 35°C with 5% carbon dioxide. Agar was used for the RS cultures to simulate standard practice, and broth was used for skin cultures to increase the yield of VRE [10]. After 48 h, skin swabs were streaked onto sheep blood agar if the broth had black discoloration, and representative colonies producing black discoloration on the Enterococcosel agar were subcultured onto sheep's blood agar.

Identification of enterococci and molecular typing. Enterococci were identified by means of standard methods [11]. Speciation of Enterococcus faecium and Enterococcus faecalis isolates was performed by use of the API-Rapid Strep Strips (bioMérieux Vitex). Susceptibility to vancomycin was assessed by the E-test method (AB Biodisk) by use of the guidelines of the National Committee for Clinical Laboratory Standards and the manufacturer's instructions [12].

Isolates recovered from stool and skin specimens were typed by pulsed-field gel electrophoresis (PFGE), as described elsewhere [13]. PFGE patterns were compared by use of established criteria [14].

Statistical analysis. VRE density was transformed to the base 10 logarithm. The associations between VRE stool concentration with skin colonization, positive results of RS culture and antibiotic exposure, and risk factors for skin colonization were analyzed by repeated-measures analysis, because independence between samples could not be assumed.

Coefficients generated by logistic regression were used to model the probability of detecting VRE with an RS culture as a function of VRE stool density [15]. A Hosmer and Lemeshow goodness-of-fit test result of >0.05 was considered supportive of model fit [16]. To determine the sensitivity of the RS culture in detecting VRE in the stool at varying densities, VRE stool concentration was grouped into 6 categories; 2.5 log10 cfu per gram of stool (hereafter cfu/g stool), 3.5 log10 cfu/g stool, 4.5 log10 cfu/g stool, 5.5 log10 cfu/g stool, 6.5 log10 cfu/g stool, and 7.5 log10 cfu/g stool. A “true-positive result” was defined as isolation of VRE from an RS culture. The sensitivity of this test was calculated by dividing true-positive results by true-positive plus false-negative results above and below each category of VRE density [17]. The corresponding 95% CIs were generated by tabulated exact binomial limits for sample sizes of ⩽100 [18]. Specificity and predictive values were not calculated, because all patients had documented VRE colonization or infection, precluding false-positive results by RS cultures. Stata, version 7.0 (Stata Statistical Software), was used for all statistical analysis. Mean values are presented as ± SD.

Previous SectionNext Section

Results

Patient characteristics and stool samples. A total of 22 patients with VRE colonization or infection were identified during the 6-month study period. Four patients who refused to participate in the study and 5 patients who had undergone colostomies or who had rectal bags were excluded. Characteristics of the 13 enrolled patients are presented in table 1. A total of 35 stool samples were obtained from these patients. A mean of 3 stool samples (range, 1–8) were obtained per patient. The mean density (± SD for the number of organisms) of VRE was 6 ± 2 log10 cfu/g stool (range, 2.5–8.1 log10 cfu/g stool). Only vancomycin-resistant E. faecium was identified from stool, skin, and rectal specimens.

Correlation between fecal VRE concentration and VRE detection with an RS culture. A total of 26 sets of stool specimens and RSs were obtained from 12 patients (median, 2; range, 1–5). RSs were not obtained from 9 corresponding stool samples because of patients' hemodynamic instability, which precluded any movement, or discharge or death during the 24 h after stool collection.

A positive RS culture was associated with a mean VRE stool concentration of 7.1 ± 0.8 log10 cfu/g stool; a negative RS culture result was associated with a mean concentration of 3.5 ±1.5 log10 cfu/g stool (P = .003; figure 1). The probability of the RS culture to detect VRE in the gastrointestinal tract increased markedly at a VRE density of >4 log10 cfu/g stool, as shown in figure 2 (Hosmer and Lemeshow goodness-of-fit test, P = .73).

Figure 1
View larger version:
Figure 1

Findings regarding the detection of stool colonization with vancomycin-resistant Enterococcus faecium (VRE) by means of the rectal swab (RS) culture method and the presence of skin colonization with VRE at varying VRE stool densities. Skin, skin culture.

Figure 2
View larger version:
Figure 2

Relationship between the probability of detecting vancomycin-resistant Enterococcus faecium (VRE) by means of the rectal swab culture method and VRE stool density (Hosmer and Lemeshow goodness-of-fit test, P = .73).

The RS culture method did not identify VRE in 11 (42%) of 26 stool samples. The overall sensitivity of the RS culture was 58% (95% CI, 37–77); it ranged from 0% (at densities of ⩽4.5 log10 cfu/g stool) to 100% (at VRE densities of ⩾7.5 log10 cfu/g stool; tables 2 and 3).

Figure 3
View larger version:
Figure 3

Pulsed-field gel electrophoresis DNA band patterns of vancomycin-resistant Enterococcus faecium (VRE) strains recovered from stool specimens and skin culture sets from 7 patients. Lane M, bacteriophage λ ladder. R, VRE isolated by use of a rectal swab culture method; S, VRE isolated from skin cultures.

Table 1
View larger version:
Table 1

Demographic and clinical data for 13 patients who were infected with vancomycin-resistant Enterococcus faecium (VRE) at study enrollment.

Table 2
View larger version:
Table 2

Sensitivity of the rectal swab culture method for the detection of vancomycin-resistant Enterococcus faecium (VRE) at decreasing stool densities.

Table 3
View larger version:
Table 3

Sensitivity of the rectal swab culture method for the detection of vancomycin-resistant Enterococcus faecium (VRE) at increasing stool densities.

Correlation between fecal VRE concentration and skin colonization. A total of 28 sets of stool and skin specimens were obtained from 12 patients for culture (median, 1; range, 1–8). Corresponding skin samples were not obtained for culture within 24 h of collection of 7 stool specimens, either because of patient discharge or death, or because the patient was undergoing a procedure for most of the day. Corresponding skin cultures were positive for VRE among 15 (54%) of the 28 stool cultures. The VRE density in stool was significantly higher among patients in whom skin colonization with VRE was detected (6.4 ± 1.5 log10 cfu/g stool), as compared with patients in whom skin colonization was not detected (4.5 ± 1.8 log10 cfu/g stool; P < .001; figure 1). Skin colonization was detected even among patients whose stool contained low VRE stool densities (table 2). There were no significant differences between VRE stool density and factors associated with an increased risk of skin contamination, including the presence of diarrhea (P = .2), inability to walk (P = .2), Charlson score of ⩾2 (P = .5), age of ⩾50 years (P = .3), and the presence of a wound sample that yielded VRE on culture (P = .8).

PFGE DNA band patterns of VRE strains recovered from stool specimens and corresponding skin cultures were indistinguishable among the 28 sets (figure 3).

Correlation between fecal VRE concentration and antibiotic exposure. The association between antibiotic exposure during the week before sample collection and VRE stool density was analyzed for 35 stool samples obtained from 13 patients (table 4). A total of 25 stool samples (71%) were exposed to ⩾1 antibiotic. Types of antibiotics tested are as follows: vancomycin (for 15 samples), aminoglycosides (for 10), imipenem (for 9), quinolones (for 7), piperacillin-tazobactam (for 5), metronidazole (for 3), quinupristin-dalfopristin (for 3), rifampin (for 3), chloramphenicol (for 2), minocycline (for 2), clindamycin (for 2), ampicillin-sulbactam (for 2), cefotetan (for 1), cefepime (for 1), erythromycin (for 1), and trimethoprim-sulfamethoxazole (for 1). Every antibiotic regimen included either vancomycin or an antianaerobic agent.

Table 4
View larger version:
Table 4

Stool density of vancomycin-resistant Enterococcus faecium (VRE) and antibiotic exposure during the 7 days before stool sample collection among hospitalized patients who had previously been infected with vancomycin-resistant enterococci.

Samples exposed to antibiotics had a significantly higher mean VRE stool density (6.7 ± 1.6 log10 cfu/g stool) than did samples with no antibiotic exposure (4.2 ± 2.5 log10 cfu/g stool; P = .02). Individual antibiotics or the antianaerobic group were not analyzed, because the majority of samples were exposed to combinations of antibiotic regimens (table 4).

Antibiotics were administered for a mean duration of 4 ± 3 days. Duration of antibiotic exposure in the preceding 7 days was significantly associated with a higher VRE stool density (P ⩽ .001). When duration of exposure was dichotomized at the mean, the VRE stool density was 4.7 ± 2.1 log10 cfu/g stool, if antibiotics were received for <4 days, and 6.8 ± 1.5 log10 cfu/g stool, if antibiotics were received for ⩾4 days (P < .001).

Previous SectionNext Section

Discussion

The results of this study suggest that the RS culture method fails to detect a large proportion of patients who had gastrointestinal colonization with VRE. The overall sensitivity of this method was 58% and ranged from 0% (at VRE stool densities of <4.5 log10 cfu/g stool) to 100% (at densities of ⩾7.5 log10 cfu/g stool). Skin colonization and recent antibiotic exposure were significantly associated with higher VRE stool densities.

The high false-negative rate associated with the RS culture method among patients with low VRE stool densities is of concern if these patients pose an infection control risk. In such patients, VRE precautions may not be implemented or may be erroneously discontinued if presumed eradication is documented by a false-negative result of RS culture. Although the transmission dynamics of VRE are complex, the main mechanism of patient-to-patient spread is through health care workers, whose hands become contaminated after contact with colonized patients [19]. Therefore, skin colonization is a prerequisite for cross-transmission, either directly (by contaminating the health care workers' hands) or indirectly (by contaminating the surrounding environment). In this study, skin colonization was detected even among patients with low VRE stool densities. For example, among the stool samples with a VRE stool density of <7.5 log10 cfu/g stool, skin colonization was detected among 43% of corresponding samples. In this range of VRE density, the RS culture method did not detect VRE in 55% of the stool samples. These findings imply that the RS culture method failed to identify a substantial proportion of patients who pose an infection control risk.

To reduce the costs of screening for VRE, the Hospital Infection Control Practices Advisory Committee recommends plating RSs onto selective media containing vancomycin [2]. In this study, Enterococcosel media supplemented with 6 µg/mL of vancomycin was used to isolate VRE. Broth enrichment methods would significantly increase the recovery of VRE, because they can detect as few as 1–9 cfu or concentrations of <103 cfu/g stool of VRE in stool samples [11, 20]. However, although the sensitivity of the RS culture would increase, the labor, cost, and, possibly, number of false-positive results would also increase [11, 20]. These factors need to be considered when the optimal screening method is chosen.

Significantly higher densities of VRE in stool were associated with recent antibiotic exposure. These findings support the theory of “colonization resistance,” in which indigenous gastrointestinal flora—in particular, anaerobic bacteria—prevent other bacteria from colonizing the gastrointestinal tract [21]. Thus, by eradicating the normal flora with antibiotics, establishment and overgrowth of VRE is facilitated [22, 23]. Donskey et al. [24] demonstrated that exposure to antianaerobic regimens, as opposed to those with minimal antianaerobic activity, increased VRE stool density. These findings emphasize the importance of maintaining a normal anaerobic component of the gastrointestinal flora to prevent VRE colonization. As shown in this study, the presence of skin colonization is directly related to higher levels of fecal VRE. Environmental contamination has also been associated with higher VRE stool densities, even in the absence of diarrhea [24]. Therefore, limiting antibiotic exposure may decrease the rate of VRE cross-transmission by decreasing the probability of skin colonization and environmental contamination.

This study has several limitations. First, the rate of skin colonization may have been underestimated, because only 2 sites were sampled for culture. Second, although stool specimens were stored at 4°C, it is unlikely that the VRE fecal density was altered, because recovery of VRE is not affected by refrigeration [24, 25]. Third, single RSs were obtained per stool sample. Increasing the number of swabs obtained at each collection might have increased the yield of VRE. However, the objective of this study was to replicate the standard practice of surveillance; therefore, only 1 RS was obtained.

The reasons for the ongoing increase in rates of VRE are multifactorial; they include poor compliance with contact precautions, antibiotic exposure, and a large reservoir of patients with unrecognized VRE colonization who continue to disseminate VRE to other patients [3–6]. Surveillance studies that used the RS culture method have shown that, for every patient with a VRE infection, there are at least 10 other colonized patients [7, 26]. This ratio may be even higher, given the low sensitivity of the RS culture method demonstrated in this study.

To limit the spread of VRE, infection control efforts have focused on improving compliance with contact precautions and limiting antibiotic use [2–5]. With the documentation of a high rate of false-negative results associated with the RS culture method, efforts will now also have to focus on developing more sensitive (yet still cost-effective) tests to detect VRE. By increasing the sensitivity of these tests, the large reservoir of patients with unrecognized VRE colonization might be decreased and might ultimately curtail the persistent increase in the prevalence of VRE throughout hospitals. Future studies will need to quantify the risk of VRE dissemination from patients with low VRE stool densities and determine whether their contribution to the spread of VRE justifies the widespread use of screening techniques that are more sensitive yet potentially more costly and labor intensive.

Previous SectionNext Section

Acknowledgments

We thank Stephen Dummer, for his thoughtful review of the manuscript, and Haiging Li, for performing PFGE.

  • Received May 29, 2001.
  • Revision received August 22, 2001.
Previous Section
 

References

  1. ↵
    1. National Nosocomial Infection Surveillance (NNIS) system report
    . data summary from January 1990-May 1999, issued June 1999. Am J Infect Control 1999;27:520-32.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Hospital Infection Control Practices Advisory Committee
    . Recommendations for preventing the spread of vancomycin resistance. Infect Control Hosp Epidemiol 1995;16:105-13.
    MedlineWeb of Science
  3. ↵
    1. Boyce JM,
    2. Mermet LA,
    3. Zervos MJU,
    4. et al
    . Controlling vancomycin-resistant enterococci. Infect Control Hosp Epidemiol 1995;16:634-7.
    MedlineWeb of Science
    1. Boyce JM,
    2. Opal SM,
    3. Chow JW,
    4. et al
    . Outbreak of multidrug-resistant Enterococcus faecium with transferable VanB class vancomycin resistance. J Clin Microbiol 1994;32:1148-53.
    Abstract/FREE Full Text
  5. ↵
    1. Slaughter SM,
    2. Hayden MK,
    3. Nathan C,
    4. et al
    . A comparison of the effect of universal use of gloves and gowns with that of glove use alone on acquisition of vancomycin-resistant enterococci in a medical intensive care unit. Ann Intern Med 1996;125:448-56.
    Abstract/FREE Full Text
  6. ↵
    1. Montecalvo MA,
    2. Shya DK,
    3. Gedris C,
    4. et al
    . A semiquantitative analysis of the fecal flora of patients with vancomycin-resistant enterococci: colonized patients pose an infection control risk. Clin Infect Dis 1997;25:929-30.
    FREE Full Text
  7. ↵
    1. Ostrowski B,
    2. Venkataraman L,
    3. D'Agata E,
    4. et al
    . Vancomycin-resistant enterococci in intensive care units: high frequency of stool carriage during non-outbreak periods. Arch Intern Med 1999;159:1467-72.
    Abstract/FREE Full Text
  8. ↵
    1. Charlson M,
    2. Pompei P,
    3. Ales K,
    4. et al
    . A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-83.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Garner JS,
    2. Jarvis WR,
    3. Emori TG,
    4. et al
    . CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988;16:128-40.
    CrossRefMedlineWeb of Science
  10. ↵
    1. Van Horn KG,
    2. Gedris CA,
    3. Rodney KM
    . Selective isolation of vancomycin-resistant enterococci. J Clin Microbiol 1996;34:924-7.
    Abstract/FREE Full Text
  11. ↵
    1. Murray PR
    1. Facklam RR,
    2. Sahm DF
    . Enterococcus. In: Murray PR, editor. Manual of clinical microbiology. 6th ed. Washington, DC: ASM Press; 1995. p. 308-14.
  12. ↵
    1. National Committee for Clinical Laboratory Standards
    . 9th informational supplement. NCCLS document M100-S9. Wayne, Pennsylvania: National Committee for Clinical Laboratory Standards; 1999. Performance standards for antimicrobial susceptibility testing.
  13. ↵
    1. D'Agata EMC,
    2. Green K,
    3. Schulman G,
    4. et al
    . Vancomycin-resistant enterococci among chronic hemodialysis patients. Clin Infect Dis 2001;32:23-9.
    Abstract/FREE Full Text
  14. ↵
    1. Tenover FC,
    2. Arbeit RD,
    3. Goering RV,
    4. et al
    . Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995;33:2233-9.
    FREE Full Text
  15. ↵
    1. Collet D
    . Modeling binary data. London: Chapman and Hall; 1991.
  16. ↵
    1. Hosmer DW,
    2. Lemeshow S
    . Applied logistic regression. 2nd ed. New York: Wiley; 2000.
  17. ↵
    1. Fletcher RH,
    2. Fletcher SW,
    3. Wagner EG
    . Clinical epidemiology: the essentials. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 1996. p. 48-50.
  18. ↵
    1. Lentner C,
    2. Lentner C,
    3. Wink A
    , editors. Geigy scientific tables. Vol 2. West Caldwell, New Jersey: CIBA-Geigy; 1982. Exact confidence limits for P; p. 89-102.
  19. ↵
    1. Zervos MJ,
    2. Kauffman CA,
    3. Therasse PM,
    4. Bergman AG,
    5. Mikesell TS,
    6. Schaberg DR
    . Nosocomial infection by gentamicin-resistant Streptococcus faecalis: an epidemiologic study. Ann Intern Med 1987;106:687-91.
    Abstract/FREE Full Text
  20. ↵
    1. Ieven M,
    2. Vercauteren E,
    3. Descheemaeker P,
    4. et al
    . Comparison of direct plating and broth enrichment culture for the detection of intestinal colonization by glycopeptide-resistant enterococci among hospitalized patients. J Clin Microbiol 1999;37:1436-40.
    Abstract/FREE Full Text
  21. ↵
    1. Vollaard EJ,
    2. Clasener HAL
    . Colonization resistance. Antimicrob Agents Chemother 1994;38:409-14.
    FREE Full Text
  22. ↵
    1. Barza MGM,
    2. Jacobus NV,
    3. Gorback SL
    . Effect of broad-spectrum parenteral antibiotics on composition of intestinal microflora of humans. Antimicrob Agents Chemother 1987;31:202-6.
    Abstract/FREE Full Text
  23. ↵
    1. Donskey CJ,
    2. Hanrahan JA,
    3. Hutton RA,
    4. et al
    . Effect of parenteral antibiotic administration on persistence of vancomycin-resistant Enterococcus faecium in the mouse gastrointestinal tract. J Infect Dis 1999;180:384-90.
    Abstract/FREE Full Text
  24. ↵
    1. Donskey CJ,
    2. Chowdhry TK,
    3. Hecker MT,
    4. et al
    . Effect of antibiotic therapy on the density of vancomycin-resistant enterococci in the stool of colonized patients. N Engl J Med 2000;343:1925-32.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Bonten MJ,
    2. Nathan C,
    3. Weinstein RA
    . Recovery of nosocomial fecal flora from frozen stool specimens and rectal swabs. Diagn Microbiol Infect Dis 1997;27:103-6.
    CrossRefMedlineWeb of Science
  26. ↵
    1. Slaughter S,
    2. Hayden M,
    3. Nathan C,
    4. et al
    . A comparison of the effect of universal use of gloves and gowns with that of glove use alone on acquisition of vancomycin-resistant enterococci in a medical intensive care unit. Ann Intern Med 1996;125:448-56.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2002) 34 (2): 167-172. doi: 10.1086/338234
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most