PEDIATRICS Vol. 108 No. 5 November 2001, pp. 1143-1148

Molecular Epidemiology of Antibiotic-Resistant Gram-Negative Bacilli in a Neonatal Intensive Care Unit During a Nonoutbreak Period

Philip Toltzis, MD, Michael J. Dul, PhD, Claudia Hoyen, MD, Ann Salvator, MS, Michele Walsh, MD, Laura Zetts, RN, and Hasida Toltzis, MS

From the Department of Pediatrics, Case Western Reserve University School of Medicine, Rainbow Babies and Children's Hospital of the University Hospitals of Cleveland, Cleveland, Ohio.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  Gram-negative organisms that are resistant to parenteral antibiotics are a growing threat to hospitalized patients. This study was conducted to define the epidemiologic characteristics of these organisms during a nonoutbreak period in a neonatal intensive care unit (NICU).

Methods.  Nasopharyngeal and rectal swab specimens were obtained 3 times a week from every infant in a tertiary care NICU during a 12-month period. Specimens were processed to identify aerobic Gram-negative species resistant to gentamicin, piperacillin-tazobactam, or ceftazidime. Selected clinical parameters were tested for their association with colonization with a resistant organism. Restriction endonuclease digests of genomic DNA were derived from isolates of the most frequently occurring species. The fragments were analyzed by pulsed-field gel electrophoresis (PFGE) to determine the genetic relatedness of the various isolates and thereby determine the length of colonization, the frequency of horizontal transmission, and the size and duration of clusters.

Results.  A total of 101 infants (8.6%) of 1180 admissions were colonized with at least 1 antibiotic-resistant bacillus before NICU discharge. Multiple parameters indicating a prolonged, complicated NICU course were associated with resistant colonization, including gestational age, length of stay, and exposure to several classes of antibiotics. Colonization with resistant bacilli occurred as early as the first NICU day, but acquisition continued throughout the infants' stay. A total of 436 isolates were analyzed by PFGE. On the basis of this molecular analysis, it was determined that duration of colonization was usually very short; the median for all species tested was <1 week. In addition, cross-colonization occurred in only 12% of all PFGE-analyzed isolates. Most clusters of cross-colonized infants were small, with the majority involving only 2 patients.

Conclusions.  During endemic periods, acquisition of antibiotic-resistant Gram-negative bacilli in the NICU may occur very soon after admission, but colonization continues over many weeks of NICU stay. The duration of colonization with resistant bacilli is short, and horizontal transmission is unusual. These characteristics suggest a gradual but temporary incorporation of these organisms from the NICU environment into the nascent newborn microflora over time with little cross-colonization. These observations may aid the rational development of infection-control strategies to contain the reservoir of resistant Gram-negative organisms in the NICU.antibiotic resistance, Gram-negative bacilli, neonatal intensive care, antibiotic utilization, colonization, pulsed-field gel electrophoresis.

Infections with Gram-negative bacteria that are resistant to commonly used parenteral antibiotics are a growing threat to hospitalized patients.^1-8 These organisms are most frequently detected in the intensive care setting. Antibiotic-resistant bacilli have been the source of numerous outbreaks in the neonatal intensive care unit (NICU).⁹ However, during endemic periods, even their most basic characteristics, such as frequency of colonization, duration of colonization, and the principal modes of acquisition, have only recently been evaluated in the critically ill newborn.¹⁰ Extrapolation from studies of older ICU populations^11-19 may be invalid, because the principal sites of colonization of these organisms, namely, the gut and pharynx, are initially sterile in the neonate. Therefore, importation of resistant bacilli into the unit by the patients themselves, a major component of their complex epidemiology among older ICU patients,^12,15,18,20 is absent, and the dynamics of ecological competition with other gut and pharyngeal colonizers possibly is very different.

We posited 3 scenarios to model colonization with antibiotic-resistant Gram-negative bacilli among critically ill newborns. In the first scenario, antibiotic-susceptible bacteria are acquired from the NICU environment, and subsequent exposure of the infant to broad-spectrum antibiotics results in the induction of resistance or the selection of resistant subpopulations. In the second scenario, selected infants, with no preexisting competing flora, become heavily colonized early in life with resistant bacteria and then serve as the principal reservoirs for dissemination of these organisms through patient-to-patient spread. The third scenario posits gradual and regular acquisition of these resistant organisms from the NICU environment with weak or no association with previous antibiotic exposure and little cross-colonization from one infant to the other. Although these scenarios are not mutually exclusive, they suggest very different strategies for interrupting acquisition. Therefore, identification of the relative contribution to colonization by these different mechanisms has important consequences for infection-control strategies. A year-long prospective survey of antibiotic-resistant Gram-negative bacteria in a large referral NICU enabled us to evaluate these competing models.

	METHODS

Top Abstract Methods Results Discussion References

Patient Setting and Clinical Data Collection

The study was conducted in a 38-bed tertiary care NICU that serves approximately 1200 admissions per year. During the period of the study, 77.1% of infants were born at a contiguous women's hospital and the remainder were transported from referring area facilities. Regardless of the place of birth, 93% of infants were admitted to the NICU by the third day of life. The physical plant of the NICU is composed of 6 rooms that house 3 to 6 infants each. In addition, a seventh room is used as a transition unit and 2 additional rooms containing a single incubator each are reserved for infants who require isolation. Incubators are surrounded by 100 sq feet of care space; each room contains 1 to 2 sinks. Patient care in the unit is divided between 2 teams of physicians, nurse practitioners, and nurses who are geographically separated, although some cross-coverage occurs during nighttime hours. A hand scrub with chlorhexidine for all staff and visitors is strictly enforced before entry through a locked door, controlled by the division secretaries, is permitted. Handwashing between each patient contact is encouraged but not enforced. Protective isolation with gloves is applied to all infants who weigh <1000 g. Gowns or gloves are not routinely required for other patient contact.

The current study was conducted as part of a larger project to test whether an antibiotic rotation schedule could decrease colonization with antibiotic-resistant Gram-negative organisms in a tertiary care NICU. Empiric and therapeutic antibiotic choices on one of the care teams were left to the discretion of the attending neonatologist and most frequently included ampicillin or vancomycin plus gentamicin. On the other team, the choice of antibiotic for Gram-negative organisms rotated monthly: gentamicin followed by piperacillin-tazobactam followed by ceftazidime. Use of vancomycin and ampicillin was not regulated. Because antibiotic rotation did not significantly affect the incidence of colonization with a resistant Gram-negative bacillus (manuscript in preparation, Toltzis et al), patients included in the current study were pooled from both teams.

The following parameters were recorded for every infant admitted to the NICU: sex, birth weight, gestational age, admission and discharge date, bed space assignment, receipt by the mother of peripartum antibiotics, and receipt of parenteral antibiotics by the patient.

Microbiologic Data Collection

For the purpose of this study, any Gram-negative bacillus that was resistant to gentamicin, piperacillin-tazobactam, or ceftazidime was termed an "antibiotic-resistant" Gram-negative organism. To determine the incidence of colonization with an antibiotic-resistant organism, we obtained rectal and pharyngeal samples for all infants in the NICU every Monday, Wednesday, and Friday with rayon-tip applicators, stored at 4°C until the next week day, and plated on screening MacConkey agar plates containing gentamicin at 8 µg/mL, piperacillin-tazobactam at 128 µg/mL piperacillin and 4 µg/mL tazobactam, or ceftazidime at 16 µg/mL. Organisms that grew after overnight incubation on the screening plates were identified by biochemical profile using an automated system (Microscan, Sacramento, CA). This system determines species using 24 biochemical tests to yield an 8-digit "octal code." Organisms were subjected to formal antibiotic susceptibility testing using a microdilution technique with the same system.

Analysis of Risk Factors for Colonization With an Antibiotic-Resistant Gram-Negative Organism

Patients were assigned a positive study outcome if they became colonized with at least 1 antibiotic-resistant Gram-negative organism; they were assigned a negative outcome if they were discharged from the NICU without detection of resistant colonization. The association between selected potential clinical risk factors and colonization with at least 1 resistant bacillus was tested by univariate analysis using ² for categorical variables and parametric and nonparametric tests, as appropriate, for continuous variables. For time-dependent variables, specifically, length of NICU stay and number of days of antibiotic exposure, the risk factor was measured until the first day of colonization with a resistant organism for those with a positive outcome or NICU discharge for those with a negative outcome.

Pulsed-Field Gel Electrophoresis

To evaluate the duration of colonization and the frequency of cross-colonization with an antibiotic-resistant Gram-negative bacterium, we subjected samples of selected organisms from the 10 most frequently isolated species to pulsed-field gel electrophoresis (PFGE). For organisms of the same species repeatedly cultured from the same patient, the following selection guidelines were applied: 1) for resistant organisms with the same octal code (for Enterobacteriaceae) or with an octal code that differed by only 1 reaction (for non-Enterobacteriaceae), isolates identified every seventh day were selected for PFGE analysis; 2) organisms with different octal codes (1 reaction difference for Enterobacteriaceae or 2 reaction difference for non-Enterobacteriaceae) compared with the first isolate were selected for PFGE analysis, regardless of when they were first cultured; 3) organisms with susceptibility patterns (to gentamicin, piperacillin-tazobactam, and ceftazidime) that differed from the first isolate also were selected for PFGE analysis, regardless of when they were first cultured.

PGFE was performed using previously described techniques.²¹ In short, organisms grown overnight to plateau phase were embedded in low-melt agarose and digested for 24 hours each first with lysozyme and then with proteinase K, both adjusted to 1 mg/mL. The resultant exposed bacterial DNA then was digested with selected endonuclease restriction enzymes designed to yield 10 to 25 large fragments: XbaI for Enterobacter, Klebsiella, Citrobacter, and Stenotrophomonas; NotI for Escherichia coli; SpeI for Pseudomonas and Serratia; and SmaI for Acinetobacter. DNA fragments then were separated on 1% agarose gels using preprogrammed protocols as recommended by the manufacturer of the apparatus (Biorad CHEF GenePath System, Hercules, CA). Organisms were defined as concordant when they differed by 3 or fewer bands after visual inspection of the restriction endonuclease pattern, as suggested by Tenover et al.²¹ Assignment of concordance was confirmed by creating digital images of each lane and clustering the patterns by the Dice coefficient using a commercially available computer program (Molecular Analysis Fingerprinting Plus Software, Biorad). All assignments of concordance or discordance were established by 2 of the investigators (P.T. and C.H.) while blinded to the identity of the patients from whom they were derived.

	RESULTS

Top Abstract Methods Results Discussion References

Descriptive Epidemiology

During a 12-month period (December 1, 1998, through November 30, 1999), a total of 1180 consecutive infants were evaluated prospectively in the NICU for colonization with an antibiotic-resistant Gram-negative organism. The study population had a mean gestational age of 35.5 weeks (standard deviation [SD]: 4.7 weeks) and a mean birth weight of 2512 g (SD: 1006 g) at the time of NICU admission. Slightly more than half (57.2%) were male. The average length of stay among these patients was 11.30 days (SD: 19.81 days).

A total of 101 infants (8.6%) were colonized with at least 1 Gram-negative organism that was resistant to gentamicin, piperacillin-tazobactam, or ceftazidime. Counting each species only once in any given patient, a total of 201 resistant Gram-negative bacteria were isolated among these patients. Thirty-one different species were identified. Resistance was detected most commonly to piperacillin-tazobactam (70.1% of all resistant organisms), followed by ceftazidime (57.2%) and gentamicin (53.2%). However, co-resistance was common. Fewer than half (82 [40.7%] of 201) of the resistant organisms were resistant to only 1 of the 3 antibiotics; 76 (37.8%) were resistant to 2, and 43 (21.4%) were resistant to all 3.

Risk Factors for Colonization With an Antibiotic-Resistant Gram-Negative Organism

Selected clinical parameters were tested by univariate analysis for their association with colonization with 1 or more antibiotic-resistant Gram-negative bacteria. All indicators of severity of illness and prolonged length of NICU stay were strongly associated with colonization with a resistant organism, including previous exposure to antibiotics (Table 1). When antecedent antibiotic exposure was analyzed further by class of antibiotic, all associations remained statistically significant (Table 1). An association with resistant colonization was noted with all 3 classes of antibiotics possessing broad activity against Gram-negative bacilli but was also noted with vancomycin, whose activity against Gram-negative organisms is nil. To test whether this association with vancomycin was attributable to its co-administration with an antibiotic with Gram-negative activity, we calculated correlation coefficients for vancomycin-days and ceftazidime-days (r = 0.23), piperacillin-tazobactam-days (r = 0.29), and gentamicin-days (r = 0.62). When vancomycin-days were correlated to the administration of any antibiotic possessing Gram-negative bacillary activity, the coefficient was 0.66. 

Duration of Colonization With an Antibiotic-Resistant Organism and the Frequency of Cross-Colonization

A total of 1214 pharyngeal and rectal specimens were positive for a resistant Gram-negative bacillus over the course of the study, 87.9% of which were from 1 of 10 species (see list in Table 2). Using the criteria stated in the Methods section, we selected 476 samples for analysis by PFGE to determine the duration of colonization and incidence of cross-colonization among the study patients. However, the DNA from all but one A baumanii isolate and the DNA from all A lwoffi isolates autodigested during preparation for PFGE and were not analyzed further (total = 22 isolates). Among the 454 isolates from the remaining eight species, 18 could not be processed for PFGE because of technical reasons (frozen stored bacteria could not be regrown, the organism autodigested, or digestion with the selected endonuclease restriction enzyme did not yield 10-25 bands), resulting in 436 samples (91.5% of all selected isolates) available for PFGE analysis. Electrophoresis of the restriction enzyme products of these 436 isolates documented 154 genetically discordant organisms.

The day of hospitalization on which the resistant organism was first isolated varied from species to species, but the mean for most organisms was between 25 and 45 days after NICU admission (Table 2). However, cumulative acquisition of an antibiotic-resistant Gram-negative rod occurred from the first NICU day onward, with the incidence gradually slowing over time (Fig 1). The duration of colonization once it was established was short (Table 2). The mean days of colonization ranged from between 4.22 days for E coli to 12.48 days for S maltophilia, with an average of 8.59 days across all tested species. The longest recorded duration of colonization was 92 days (in a patient with S maltophilia). However, the marked majority of genetically distinct isolates were found on only a single occasion. Indeed, the median number of positive cultures of a given genotype in a given patient was 1 (Table 2). Because isolates were selected for PFGE analysis every seventh day, this finding indicates a median duration of colonization of 6 days or less.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   The cumulative incidence of colonization with genetically discordant antibiotic-resistant bacilli, as judged by PFGE, is plotted against the day of NICU stay on which each organism was first detected.

Remarkably few infants were cross-colonized with genetically concordant organisms. Of the 154 genetically distinct organisms, only 19 (12.3%) were isolated from more than 1 patient (Table 3). In 13 of the 19 instances, the cluster involved only 2 patients; the largest cluster was a single instance involving 5 infants who were colonized with S marcescens (Table 3). Cross-colonization was not associated with prolonged colonization; 37.5% of infants who were colonized with a cross-colonizing organism were colonized for 7 days, versus 29.9% of infants who harbored an organism in which no cross-colonization was detected (P < .329).

The mean number of days that a given organism could be cultured during a cluster was 62 days. In 71.4% of instances in which cross-colonization occurred, the infants included in the cluster were housed in the same room. In nearly every instance, the concordant organisms were isolated from infants whose NICU hospitalizations overlapped in time. In 3 circumstances, however, 2 clusters involving S maltophilia and 1 involving S marcescens, genetically concordant organisms were isolated several weeks after the index case had been discharged.

	DISCUSSION

Top Abstract Methods Results Discussion References

Several features of the epidemiology of endemic antibiotic-resistant bacillary colonization in the NICU were revealed by the current study. Approximately 8% of infants were colonized in the pharynx or rectum with organisms resistant to broad-spectrum parenteral antibiotics before their NICU discharge. The range of organisms identified was large, composed primarily of 10 predominant enteric and nonenteric species but including several isolates each of genera as varied as Cedecea, Burkholderia, Providencia, Shigella, Yersinia, and Centers for Disease Control and Prevention class IV, among others. Moreover, resistance against more than 1 frequently used broad-spectrum Gram-negative antibiotic was common in this population of bacteria.

The risk for acquisition of these resistant bacilli began on the first day of admission and continued throughout the NICU stay. However, once acquired, there was little dissemination of organisms from patient to patient. A low incidence of cross-transmission of Gram-negative organisms during endemic periods has been noted in ICUs that care for older patients.^14,18,19,22 However, this observation was striking in NICU patients, where a developing microflora and close quartering should have provided ideal circumstances for cross-colonization; indeed, rapid horizontal spread of resistant bacilli during NICU epidemics is relatively common. At a superficial level, these data seem to be in contrast with those recently reported by Almuneef et al¹⁰ that emphasized cross-colonization of Gram-negative rods during an endemic period in the NICU at Yale-New Haven Hospital. Our lower incidence of patient-to-patient spread may reflect different hygienic practices in our unit (universal glove use for contact with low-birth-weight infants, a very strictly enforced hand scrub before NICU entry) or a different range of organisms (we frequently isolated Stenotrophomonas, a nonenteric bacillus that was not commonly transmitted from infant to infant). Even in the Yale study, however, most genetically discordant Gram-negative bacilli were not shared among their patients,¹⁰ a finding not very different from our own. Certainly in our unit, total elimination of cross-colonization would leave the marked majority of resistant Gram-negative bacillary colonization unaddressed.

Whether acquired from another infant or not, colonization in the current study usually was very short lived, frequently detected only on 1 occasion. This finding suggests a very fluid ecology of resistant (and, presumably, susceptible) aerobic Gram-negative bacilli in the critically ill neonate during endemic periods, in which relatively few organisms establish a prolonged presence. Taken together, these observations are most supportive of a model in which a heterogeneous population of antibiotic-resistant Gram-negative bacteria are gradually and cumulatively acquired from the newborn's surroundings, presumably via the hands of caregivers,^1023-25 with the patient most frequently serving as a transient reservoir and dead-end host.

Colonization by resistant organisms was associated with multiple markers of severity of illness, as has been seen in other intensive care settings.^1726-28 The relationship of acquisition of a resistant organism in the NICU and exposure to antibiotics requires particular consideration. The effects of antibiotics on the acquisition of resistant colonization could not be addressed with certainty in the current study. Although a strong association between antibiotic exposure and resistant Gram-negative colonization was detected by univariate analysis, this association was noted across all antibiotic groups, even with vancomycin. Some of this association with vancomycin may have been attributable to its moderate correlation with administration of an antibiotic with Gram-negative activity. Alternatively, it is possible that this association was simply reflective of a longer stay. Indeed, some infants in the current study acquired resistant colonization very early in their NICU course, before one would expect induction or selection of resistance attributable to antibiotic exposure. In other ICU settings, antecedent antibiotic exposure has been variably associated with development of colonization or infection with an autologously resistant organism.^2628-34 Presumably all resistant phenotypes initially develop under antibiotic pressure. However, as the epidemiology of these organisms matures in a given unit, they may be acquired under a variety of different circumstances,³⁵ and the strength of the association with antecedent antibiotic exposure thereby may be weakened.

NICU-specific studies examining the effects of antibiotic exposure on the appearance of resistance during nonoutbreak periods have revealed a complex pattern and variable results.^1036-40 Recently, de Man et al³⁶ compared 2 empiric antibiotic regimensamoxicillin plus cefotaxime versus penicillin plus tobramycinregarding their propensity to promote antibiotic-resistant organism-colonization in 2 contiguous NICUs. The amoxicillin plus cefotaxime regimen was significantly more likely to result in resistant colonization than the alternative regimen. However, their amoxicillin/cefotaxime group had a significantly longer mean length of stay than those who received penicillin/tobramycin, which may have accounted in part for their higher incidence of colonization with a resistant bacterium. It should be emphasized that antibiotic exposure in the ill newborn may have significant indirect effects. Several studies suggest that antibiotic use in the NICU results in the expansion and contraction of many elements of the newborn's nascent microflora. It is conceivable that the principal effect of administering broad-spectrum agents to the neonate is to reduce the density of the ecological competitors of aerobic bacilli in the pharynx and the gut.^41,42 This phenomenon itself may promote colonization with aerobic Gram-negative rods, antibiotic susceptible or not, once the infant is inoculated from the environment (perhaps accounting for our association between colonization and vancomycin use).

Unfortunately, a model in which there is gradual, regular, cumulative acquisition of disparate resistant organisms, such as supported by the current study, does not lend itself easily to intervention. Assuming that these organisms are introduced to the infant through the hands of caregivers, this mode of acquisition reemphasizes the importance of hand cleansing, and the recent availability of portable alcohol-based gels that can be placed next to each incubator and in the pockets of caregivers⁴³ may improve compliance and reduce resistant colonization. Universal barrier isolation likely would decrease the introduction of these bacilli into the infant's developing microflora from the NICU environment, but its imposition with every infant contact probably would prove to be overwhelming to permanent staff. Ultimately, some patient contact in the NICU with antibiotic-resistant aerobic Gram-negative bacilli may be inevitable regardless of these or any other measures.

	ACKNOWLEDGMENTS

This study was supported by Grant HD-31323-05 from the National Institutes of Health for the Pediatric Pharmacology Research Unit.

	FOOTNOTES

Received for publication Apr 4, 2001; accepted Jun 5, 2001.

This work was presented, in part, at the annual meeting of the Society for Pediatric Research; Boston, MA; May 12-16, 2000.

Reprint requests to (P.T.) Division of Pharmacology and Critical Care, Rainbow Babies and Children's Hospital, 11100 Euclid Ave, Cleveland, OH 44106. E-mail: pxt2{at}po.cwru.edu.

	ABBREVIATIONS

NICU, neonatal intensive care unit; PFGE, pulsed field gel electrophoresis; SD, standard deviation.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Aksaray S, Dokuzoguz B, Guvener E, Surveillance of antimicrobial resistance among gram-negative isolates from intensive care units in eight hospitals in Turkey. J Antimicrob Chemother 2000; 46:649 [Free Full Text]
2. Burwen DR, Banerjee SN, Gaynes RP Ceftazidime resistance among selected nosocomial gram-negative bacilli in the United States. National Nosocomial Infections Surveillance System. J Infect Dis 1994; 170:1622-1625 [Medline]
3. Bonfiglio G, Carciotto V, Russo G, Antibiotic resistance in Pseudomonas aeruginosa: an Italian survey. J Antimicrob Chemother. 1998; 41:307-310 [Abstract/Free Full Text]
4. Flournoy DJ, Reinert RL, Bell-Dixon C, Gentry CA Increasing antimicrobial resistance in gram-negative bacilli isolated from patients in intensive care units. Am J Infect Control 2000; 28:244-250 [CrossRef][Medline]
5. Hanberger H, Garcia-Rodriguez JA, Gobernado M, Goossens H, Nilsson LE, Struelens MJ Antibiotic susceptibility among aerobic gram-negative bacilli in intensive care units in 5 European countries. French and Portuguese ICU Study Groups. JAMA 1999; 281:67-71 [Abstract/Free Full Text]
6. Harbarth S, Rohner P, Auckenthaler R, Safran E, Sudre P, Pittet D Impact and pattern of gram-negative bacteraemia during 6 y at a large university hospital. Scand J Infect Dis 1999; 31:163-168 [CrossRef][Medline]
7. Itokazu GS, Quinn JP, Bell-Dixon C, Kahan FM, Weinstein RA Antimicrobial resistance rates among aerobic gram-negative bacilli recovered from patients in intensive care units: evaluation of a national postmarketing surveillance program. Clin Infect Dis 1996; 23:779-784 [Medline]
8. Musoke RN, Revathi G Emergence of multidrug-resistant gram-negative organisms in a neonatal unit and the therapeutic implications. J Trop Pediatr 2000; 46:86-91 [Abstract/Free Full Text]
9. Toltzis P, Blumer JL Antibiotic-resistant gram-negative bacteria in the critical care setting. Pediatr Clin North Am 1995; 42:687-702 [Medline]
10. Almuneef MA, Baltimore RS, Farrel PA, Reagan-Cirincione P, Dembry LM Molecular typing demonstrating transmission of gram-negative rods in a neonatal intensive care unit in the absence of a recognized epidemic. Clin Infect Dis 2001; 32:220-227 [CrossRef][Medline]
11. Flaherty JP, Weinstein RA Nosocomial infection caused by antibiotic-resistant organisms in the intensive-care unit. Infect Control Hosp Epidemiol 1996; 17:236-248 [Medline]
12. Flynn DM, Weinstein RA, Nathan C, Gaston MA, Kabins SA Patients' endogenous flora as the source of "nosocomial" Enterobacter in cardiac surgery. J Infect Dis 1987; 156:363-368 [Medline]
13. Davin-Regli A, Monnet D, Saux P, Molecular epidemiology of Enterobacter aerogenes acquisition: one-year prospective study in two intensive care units. J Clin Microbiol 1996; 34:1474-1480 [Abstract]
14. D'Agata E, Venkataraman L, DeGirolami P, Samore M Molecular epidemiology of acquisition of ceftazidime-resistant gram-negative bacilli in a nonoutbreak setting. J Clin Microbiol 1997; 35:2602-2605 [Abstract]
15. D'Agata EM, Venkataraman L, DeGirolami P, Colonization with broad-spectrum cephalosporin-resistant gram-negative bacilli in intensive care units during a nonoutbreak period: prevalence, risk factors, and rate of infection. Crit Care Med 1999; 27:1090-1095 [CrossRef][Medline]
16. D'Agata EM, Venkataraman L, DeGirolami P, Samore M Molecular epidemiology of ceftazidime-resistant gram-negative bacilli on inanimate surfaces and their role in cross-transmission during nonoutbreak periods. J Clin Microbiol 1999; 37:3065-3067 [Abstract/Free Full Text]
17. Toltzis P, Yamashita T, Vilt L, Blumer JL Colonization with antibiotic-resistant gram-negative organisms in a pediatric intensive care unit. Crit Care Med 1997; 25:538-544 [CrossRef][Medline]
18. Toltzis P, Hoyen C, Spinner-Block S, Salvator AE, Rice LB Factors that predict preexisting colonization with antibiotic-resistant gram-negative bacilli in patients admitted to a pediatric intensive care unit. Pediatrics 1999; 103:719-723 [Abstract/Free Full Text]
19. Chetchotisakd P, Phelps CL, Hartstein AI Assessment of bacterial cross-transmission as a cause of infections in patients in intensive care units. Clin Infect Dis 1994; 18:929-937 [Medline]
20. Olson B, Weinstein RA, Nathan C, Chamberlin W, Kabins SA Epidemiology of endemic Pseudomonas aeruginosa: why infection control efforts have failed. J Infect Dis 1984; 150:808-816 [Medline]
21. Tenover FC, Arbeit RD, Goering RV, Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33:2233-2239 [Medline]
22. Bingen E, Denamur E, Lambert-Zechovsky N, Brahimi N, el Lakany M, Elion J Rapid genotyping shows the absence of cross-contamination in Enterobacter cloacae nosocomial infections. J Hosp Infect 1992; 21:95-101 [CrossRef][Medline]
23. Guenthner SH, Hendley JO, Wenzel RP Gram-negative bacilli as nontransient flora on the hands of hospital personnel. J Clin Microbiol 1987; 25:488-490 [Abstract/Free Full Text]
24. Knittle MA, Eitzman DV, Baer H Role of hand contamination of personnel in the epidemiology of gram-negative nosocomial infections. J Pediatr 1975; 86:433-437 [CrossRef][Medline]
25. Sanderson PJ, Weissler S Recovery of coliforms from the hands of nurses and patients: activities leading to contamination. J Hosp Infect 1992; 21:85-93 [CrossRef][Medline]
26. Carmeli Y, Samore MH Comparison of treatment with imipenem vs. ceftazidime as a predisposing factor for nosocomial acquisition of Stenotrophomonas maltophilia: a historical cohort study. Clin Infect Dis 1997; 24:1131-1134 [Medline]
27. Lucet JC, Chevret S, Decre D, Outbreak of multiply resistant Enterobacteriaceae in an intensive care unit: epidemiology and risk factors for acquisition. Clin Infect Dis 1996; 22:430-436 [Medline]
28. Talon D, Capellier G, Boillot A, Michel-Briand Y Use of pulsed-field gel electrophoresis as an epidemiologic tool during an outbreak of Pseudomonas aeruginosa lung infections in an intensive care unit. Intensive Care Med 1995; 21:996-1002 [CrossRef][Medline]
29. Chow JW, Fine MJ, Shlaes DM, Enterobacter bacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann Intern Med 1991; 115:585-590
30. Brown EH, Spencer RC, Brown JM. The emergence of bacterial resistance in hospitalsa need for continuous surveillance. J Hosp Infect. 1990;15(suppl A):35-39
31. Troillet N, Samore MH, Carmeli Y Imipenem-resistant Pseudomonas aeruginosa: risk factors and antibiotic susceptibility patterns. Clin Infect Dis 1997; 25:1094-1098 [Medline]
32. Toltzis P, Yamashita T, Vilt L, Antibiotic restriction does not alter endemic colonization with resistant gram-negative rods in a pediatric intensive care unit. Crit Care Med 1998; 26:1893-1899 [Medline]
33. Jessop AB, John JF Jr, Paul SM Risk factors associated with the acquisition of amikacin-resistant gram-negative bacilli in central New Jersey hospitals. Infect Control Hosp Epidemiol 1998; 19:186-188 [Medline]
34. Jacobson KL, Cohen SH, Inciardi JF, The relationship between antecedent antibiotic use and resistance to extended-spectrum cephalosporins in group I beta-lactamase-producing organisms. Clin Infect Dis 1995; 21:1107-1113 [Medline]
35. Tenover FC, McGowan JE Jr Reasons for the emergence of antibiotic resistance. Am J Med Sci 1996; 311:9-16 [CrossRef][Medline]
36. de Man P, Verhoeven BA, Verbrugh HA, Vos MC, van den Anker JN An antibiotic policy to prevent emergence of resistant bacilli. Lancet 2000; 355:973-978 [CrossRef][Medline]
37. Goldmann DA, Leclair J, Macone A Bacterial colonization of neonates admitted to an intensive care environment. J Pediatr 1978; 93:288-293 [Medline]
38. Kalenic S, Francetic I, Polak J, Zele-Starcevic L, Bencic Z Impact of ampicillin and cefuroxime on bacterial colonization and infection in patients on a neonatal intensive care unit. J Hosp Infect 1993; 23:35-41 [CrossRef][Medline]
39. Tullus K, Burman LG Ecological impact of ampicillin and cefuroxime in neonatal units. Lancet 1989; 1:1405-1407 [Medline]
40. Tullus K, Berglund B, Fryklund B, Kuhn I, Burman LG Influence of antibiotic therapy on faecal carriage of P-fimbriated Escherichia coli and other gram-negative bacteria in neonates. J Antimicrob Chemother 1988; 22:563-568 [Abstract/Free Full Text]
41. Hokama T, Imamura T Members of the throat microflora among infants with different feeding methods. J Trop Pediatr 1998; 44:84-86 [Abstract/Free Full Text]
42. Lundequist B, Nord CE, Winberg J The composition of the faecal microflora in breastfed and bottle fed infants from birth to eight weeks. Acta Paediatr Scand 1985; 74:45-51 [Medline]
43. Maury E, Alzieu M, Baudel JL, Availability of an alcohol solution can improve hand disinfection compliance in an intensive care unit. Am J Respir Crit Care Med 2000; 162:324-327 [Abstract/Free Full Text]