Nosocomial outbreak of Enteroccocus gallinarum: untaming of rare species of enterococci.

An unusual increase in infections caused by vancomycin-resistant Enterococcus gallinarum (VREG) was identified in May 2004, in a Colombian tertiary care teaching hospital. A case-control study was subsequently designed to identify risk factors associated with the development of infections due to these organisms. All VREG isolates were subjected to antimicrobial susceptibility testing, vancomycin resistance gene detection and pulsed-field gel electrophoresis (PFGE) typing. Additionally, the presence of genes associated with an acquired pathogenicity island of E. faecalis and a hyl-like gene of E. faecium was assessed by hybridisation assays. Eleven cases of VREG were identified between May through June 2004. VREG was isolated from blood (N=4), surgical secretions (N=4), paranasal sinus secretion (N=1), lung abscess (N=1) and urine (N=1). Infections with VREG were associated with mucositis, hospitalisation in the haematology ward and surgical unit, length of hospital stay prior to culture and invasive procedures within 30 days prior to the culture. Logistic regression found that female sex and hospitalisation in the surgical unit were independent factors for VREG infection. All isolates were identified as E. gallinarum, harboured the vanC1 gene and exhibited indistinguishable restriction patterns by PFGE. Virulence-associated genes were not detected. This is the first documented hospital-wide outbreak of VREG and highlights the fact that uncommon species of enterococci are capable of nosocomial dissemination.

Problems and perils of vancomycin resistant enterococci.

Enterococci have been a therapeutic challenge for haIfa century; first in the management of endocarditis, then associated with the emergence of resistance to streptomycin and later to all aminoglycosides, and now with the increasing levels of resistance to penicillins. A major leap in the problem of antimicrobial resistance occurred more than a decade ago when vancomycin resistant enterococci (VRE) were first identified. This resulted from the acquisition by Enterococcus faecium of vancomycin resistant genes. Five types of vancomycin resistance have since been described (VanA-VanE) and others also appear to exist. VanA and VanB are caused by complex gene clusters that may be plasmid and/or transposon encoded. As a result of the gene cluster, cell wall precursors in the bacteria are formed that do not allow effective vancomycin binding, thus the action of vancomycin to inhibit cell wall synthesis is prevented. Therapy of infections caused by VRE is difficult, but a number of potentially effective antibiotics are now being tested in humans, including quinupristin/dalfopristin, linezolid, evernimomycin, daptomycin and LY333328. Combinations of antibiotics such as ampicillin with quinupristin/dalfopristin or with imipenem, and newer fluoroquinolones are also being evaluated. Until the time when these drugs become available, we must rely on careful monitoring of microbial transmission in hospitals, and we must utilize multi-faceted approaches to prevent the increase in the number and spread of VRE.

Recurrent group B streptococcal infections in infants: clinical and microbiologic aspects.

OBJECTIVE: To describe the potential for recurrence of group B streptococcal (GBS) infection in infants, using pulsed-field gel electrophoresis as an epidemiologic tool. DESIGN: Retrospective review of cases identified by laboratory records and review of the literature. SETTING: Neonatal nurseries of a county hospital system. METHODS: Retrospective review of infants with second episodes of GBS bacteremia or meningitis. Digestion of chromosomal DNA with the restriction enzyme Sma I and separation of fragments by use of contour-clamped homogeneous electric field. RESULTS: Nine cases of recurrent GBS infection were identified during a 14-year period. Eight of the nine infants were born at 25 to 36 weeks of gestation, and one was born at term. The first episode of invasive GBS infection occurred at a mean age of 10.4 days (median, 3 days; range, 1 to 27 days). Parenteral antibiotic therapy was administered for a mean of 13.9 days (median, 14 days; range, 10 to 21 days). Recurrence occurred at a mean age of 42.3 days (median, 48 days; range, 23 to 68 days). One patient died during the second episode; eight infants survived to discharge home. Of seven sets of isolates analyzed from first and second GBS episodes, five were confirmed to be the same genotypically. CONCLUSION: Recurrence of GBS disease in infants may be associated with the original infecting strain or a second acquired strain.

Ceftibuten and trimethoprim-sulfamethoxazole for treatment of Shigella and enteroinvasive Escherichia coli disease.

In a prospective randomized study at two clinical sites, ceftibuten was compared with trimethoprim-sulfamethoxazole (TMP-SMX), both given orally for a period of 5 days, for the treatment of dysentery. Twenty-two children were found to have bacillary dysentery caused by Shigella and/or enteroinvasive Escherichia coli. All organisms isolated were susceptible to ceftibuten; 6 of 20 Shigella strains and 4 of 5 enteroinvasive E. coli were resistant to TMP-SMX. The diarrhea persisted for a mean (+/- SD) period of 2.4 +/- 1.4 days in the ceftibuten-treated patients vs. 3.4 +/- 1.7 days in the TMP-SMX-treated patients. The duration of fever was similar for both treatment groups. Patients treated with ceftibuten or TMP-SMX had equivalent clinical responses unless the pathogen was found to be TMP-SMX-resistant. Those who were randomized to receive TMP-SMX but who were eventually found to have TMP-SMX-resistant organisms had significantly more stools at days 3, 4 and 5 (P less than 0.02 to less than 0.00006) with more watery consistency for these days (P less than 0.02 to less than 0.005) compared to patients treated with ceftibuten. No clinical relapses were reported and no drug-related side effects were observed. We conclude that ceftibuten is at least as effective as TMP-SMX in the treatment of diarrhea caused by Shigella and enteroinvasive E. coli in children.

Intrahospital spread of a single gentamicin-resistant, beta-lactamase-producing strain of Enterococcus faecalis in Argentina.

Six beta-lactamase-producing (Bla+) isolates of Enterococcus faecalis recovered over a 17-month period from an Argentinian pediatric hospital were found to have identical or almost identical chromosomal restriction patterns by pulsed-field gel electrophoresis, although the plasmid patterns were different. These isolates, like Bla+ enterococci in the United States, hybridized to a staphylococcal Bla gene probe. The presence of a single strain was somewhat surprising, since all isolates transferred Bla by conjugation.

Evidence for clonal spread of a single strain of beta-lactamase-producing Enterococcus (Streptococcus) faecalis to six hospitals in five states.

Beta-lactamase-producing (Bla+) enterococci have been reported in three state and two countries. Pulsed-field gel electrophoresis was used to compare 14 Bla+ Enterococcus (Streptococcus) faecalis isolated from hospitalized patients in seven states and three continents. The restriction endonuclease digestion patterns of isolates from Connecticut, Massachusetts, Lebanon, and Argentina were all markedly different, indicating that these were different strains. However, isolates from Delaware, Texas, Pennsylvania (Philadelphia and Pittsburgh), Florida, and Virginia were similar, indicating that these isolates were derivatives of a single strain. This conclusion was supported by hybridization using individual fragments as probes. Spread of Bla+ enterococci within the hospital setting was also demonstrated. These findings illustrate the value of pulsed-field gel electrophoresis for epidemiologic analyses and support the importance of identifying and containing organisms with new resistance properties in an effort to decrease their transmission to and from, as well as within, hospitals.

Evidence of infection with organisms producing Shiga-like toxins in household contacts of children with the hemolytic uremic syndrome.

We conducted a prospective study in 87 household contacts of 51 children with hemolytic uremic syndrome to determine the frequency of infection with Shiga-like toxin-producing bacteria. Gastrointestinal tract symptoms occurred in only 1 of 87 contacts. Free fecal toxin was detected in 25 of 64 (39%) of the household members. Neutralization with specific antisera to Shiga-like toxins I and II (SLT-I, SLT-II) revealed that in 6 of these household contacts only SLT-I was present in stool, in 10 only SLT-II was present and in 9 both toxins were found. Thirty-three percent of the hemolytic uremic syndrome families in which 2 or more members were studied had more than 1 household member with free fecal toxin in stool. None of the household contacts was found to have E. coli O157:H7 in feces. Serum samples were available in 77 household contacts; 75% (58 of 77) had serum neutralizing titers of greater than or equal to 1:4 to 1 or both toxins. In those contacts for whom paired sera were available, seroconversion was found in 10 of 24 (42%). These data show that household contacts of children with hemolytic uremic syndrome are commonly colonized with Shiga-like toxin-producing E. coli and seroconversion to Shiga-like toxins occurs frequently in family members of children with hemolytic uremic syndrome.

Emergence of resistant fecal Escherichia coli in travelers not taking prophylactic antimicrobial agents.

Fecal specimens from individuals traveling to Mexico were examined before, during, and after travel for the presence of Escherichia coli resistant to ampicillin, chloramphenicol, gentamicin, kanamycin, streptomycin, sulfonamides, trimethoprim (TMP), and TMP-sulfamethoxazole (TMP-SMX). None of these individuals took prophylactic antibiotics, although 4 of 13 took short courses of an antimicrobial agent for therapy of traveler's diarrhea. With an average of 9.3 E. coli per sample, resistance to all agents tested except gentamicin was shown to increase during the time in Mexico (P less than 0.001 to P less than 0.05). For example, no TMP-resistant (Tmpr) E. coli isolates were found by this method before travel, whereas 57% of the individuals had Tmpr and Tmpr-Smxr E. coli by the final week in Mexico. This increase in resistance occurred regardless of whether an individual took a short course of antimicrobial therapy. This study shows that travel itself, even without the use of prophylactic or therapeutic antimicrobial agents, is associated with the acquisition of resistant E. coli. Travel to developing nations may rival other sources of resistant organisms.

Hemolytic uremic syndrome and diarrhea in Argentine children: the role of Shiga-like toxins.

Shiga-like toxin-producing Escherichia coli have been associated with hemorrhagic colitis and the hemolytic uremic syndrome (HUS). Because Argentina has the highest reported frequency of HUS in the world, Argentine children were prospectively studied during the HUS seasons for evidence of Shiga-like toxin-related diseases. On the basis of serology, fecal cytotoxin neutralization, stool cultures, and DNA hybridization of colony lysates, most children with HUS had evidence of infection with Shiga-like toxin-producing organisms. Children with spring-summer diarrhea also commonly (32%, confidence interval 18%-46%) had clear-cut evidence of such infection. No controls (children without gastrointestinal, renal, or hemolytic disease) had free fecal cytotoxin, positive cultures for E. coli O157:H7, or DNA probe-positive organisms; 20% of them had low serum titers of antibodies to Shiga-like toxins. E. coli O157:H7 was not common in either HUS or diarrhea patients. The high frequency of Shiga-like toxin-induced diarrhea in young children in Argentina probably explains the high incidence of HUS in this country and suggests that HUS is a relatively uncommon complication of Shiga-like toxin-related disease.

Increasing resistance to trimethoprim-sulfamethoxazole among isolates of Escherichia coli in developing countries.

Resistance of Escherichia coli to trimethoprim (TMP)-sulfamethoxazole remains at 3%-8% at many medical centers within the United States. In this study a 44% resistance rate was observed among E. coli isolated at a pediatric hospital in Santiago, Chile, and a 40% resistance rate at a general teaching hospital in Bangkok, Thailand. Most isolates were from urinary tract infections and showed high-level resistance (minimal inhibitory concentration of TMP greater than 1,000 micrograms/ml). Nineteen of 35 isolates tested transferred resistance to TMP; most cotransferred resistance to streptomycin and sulfonamides. Dihydrofolate reductase type I was detected by gene probing in 14 of 35 strains. Subsequent investigations in Brazil, Honduras, and Costa Rica revealed that this high rate of resistance was not an isolated phenomenon.

Evidence for an epidemic trimethoprim-resistance plasmid in fecal isolates of Escherichia coli from citizens of the United States studying in Mexico.

A previous study revealed the emergence of high-level resistance to trimethoprim-sulfamethoxazole in strains of Escherichia coli isolated from 95% of students from the United States who were taking either trimethoprim alone or trimethoprim-sulfamethoxazole for prophylaxis of travelers' diarrhea while in Guadalajara, Mexico. Many of these strains were subsequently demonstrated to cotransfer resistance to trimethoprim along with that to streptomycin and ampicillin. The present study demonstrated that at least 12 (60%) of 20 transferable (tra+) trimethoprim-resistance plasmids studied possessed both an identical Hind III restriction pattern and type I dihydrofolate reductase genes. Donor strains were shown to be distinct; this finding suggested that a common tra+ trimethoprim-resistance plasmid was widely disseminated among E. coli strains in Guadalajara. These results may explain in part the surprising degree of resistance encountered in trimethoprim-consuming persons in that region.

Transfer of trimethoprim resistance from fecal Escherichia coli isolated during a prophylaxis study in Mexico.

Forty-one of 100 trimethoprim-resistant fecal isolates of Escherichia coli isolated during a diarrhea prophylaxis study transferred trimethoprim resistance to an E. coli K12 derivative at frequencies ranging from 1 x 10(-7) to 2 x 10(-2). Analysis of these transconjugants, after attempting to select those that had received only one plasmid, revealed that 34 (83%) were resistant to trimethoprim plus only one or two of seven other antimicrobial agents tested, most often ampicillin and/or streptomycin. In contrast, 91% of the donor strains were resistant to four or more antimicrobial agents. Plasmid analysis demonstrated that a 35-megadalton plasmid was present in 35 transconjugants and was usually associated with resistance to trimethoprim, streptomycin, and ampicillin; 19 transconjugants had only a 35-megadalton plasmid and were resistant to only these three antimicrobial agents, raising the possibility that a common plasmid was present in many strains. Although resistance to multiple antimicrobial agents could be transferred to the K12 recipient, resistance to more than three agents was seldom mediated by a single plasmid. The high degree of transferability of trimethoprim resistance from these fecal isolates of E. coli emphasizes the potential for the dissemination of genes carrying trimethoprim resistance.