Nontyphoidal Salmonella species mainly cause disease in children younger than 5 years of age, including acute gastroenteritis and extraintestinal disease such as bacteremia, arthritis, meningitis and pneumonia.1Salmonella species are a common cause of pediatric bacteremia in tropical Africa, and associations have been noted with malaria and human immunodeficiency virus (HIV) infection.1,2 Although the fecal-oral route of transmission of salmonellosis in hospitalized newborns and infants has been implicated, little is known about the source and transmission of infection in Africa. Person-to-person spread is thought to be the probable mode of spread in many outbreaks.
Increasing resistance in some Salmonella serotypes has been described in community-acquired infections and nosocomial outbreaks.3 A nosocomial outbreak of Salmonellaenteritidis in West Africa involved a strain that was resistant to multiple antibiotics and possibly of increased virulence.4
We describe a nosocomial outbreak of salmonella infection caused by an unusual serotype, Salmonella enterica serotype Isangi, that also produces an extended spectrum β-lactamase (ESBL). This is the first report of Salmonella spp. ESBL resistance in our hospital.
Laboratory and hospital records for pediatric patients with positive cultures of Salmonella spp. ESBL from the beginning of 2000 through 2001 at Chris Hani Baragwanath Hospital, Johannesburg, South Africa, were reviewed. This hospital serves a population of about 1.2 million black urban South Africans, including an estimated 45,000 children younger than 2 years of age.
All isolates were identified by Gram stain, biochemical characterization with the API 20E system (Biomérieux, Marcy L'Ėtoile, France), and by serologic identification of somatic (O) and flagellar (H) antigens with commercial antisera (Sanofi Pasteur Diagnostics, Paris, France) according to the Kauffman-White serotyping scheme, as previously described.5
Antibiotic susceptibility testing was performed by disc diffusion method and interpreted in accordance with criteria of the Clinical and Laboratory Standards Institute (formerly National Committee of Clinical Laboratory Standards).6 Minimum inhibitory concentrations (MICs) of ceftazidime, cefotaxime, and cefepime were determined. Two ESBL screening methods were used: (a) Double disc synergy/disc approximation method using multiple target discs with amoxicillin-clavulanic acid, ceftazidime, cefotaxime and cefepime on Mueller-Hinton agar; (b) ESBL production was defined as a ≥4-fold decrease in MIC for either ceftazidime or cefotaxime when tested with clavulanic acid versus its MIC when tested alone. Escherichia coli ATCC 25922 was used as a control strain.6 Environmental cultures were not performed.
Strain relatedness was analyzed using pulsed-field gel electrophoresis (PFGE) according to the protocol used by Pulse-Net, Centers for Disease Control and Prevention, Atlanta, GA.7
Forty-six specimens from 41 patients were reviewed including 22 blood cultures, 19 stools, 3 rectal swabs, 1 urine and 1 tracheal aspirate. The first identified case was a 4-month-old HIV-infected female, admitted twice in a period of 3 months at the beginning of 2000. She initially presented with bronchopneumonia, was hospitalized for 8 days, and was readmitted with gastroenteritis 10 days later. Five weeks into her second admission, Salmonella isangi (SI) was isolated from blood and stool specimens.
The ages of infected children ranged from 3 weeks to 4 years (median, 5 months); 23 patients were male. Clinical features included gastroenteritis (n = 25), dysentery (n = 10), fever (n = 23) or bronchopneumonia (n = 6). Seventy-five percent (26/41) of the patients were HIV-seropositive. Additional underlying conditions noted were pulmonary tuberculosis, Pneumocystis jiroveci pneumonia, Down's syndrome, myasthenia gravis, ventricular septal defect, kwashiorkor and prematurity. There were 9 (22%) fatalities, of which 7 (78%) had underlying HIV infection. Twenty-three patients received ampicillin, trimethoprim-sulfamethoxazole, and/or gentamicin before a positive culture, and 9 had received a third-generation cephalosporin.
The SI-ESBL isolates were susceptible to imipenem, meropenem, amikacin, nalidixic acid, ciprofloxacin and cefoxitin. Resistance was noted to ampicillin, amoxicillin-clavulanic acid, piperacillin-tazobactam, ceftriaxone, ceftazidime, cefotaxime, cefepime, gentamicin, chloramphenicol, trimethoprim-sulfamethoxazole and tetracycline. One resistance phenotype was observed in all the SI-ESBLs. The MIC values for cefotaxime, ceftazidime, cefepime, and ciprofloxacin were >64, >128, >64 and <0.12 μg/ml, respectively.
On double-disc synergy testing for the demonstration of ESBL production, enhanced activity was more pronounced between cefepime and amoxicillin-clavulanic acid than between either ceftazidime or ceftriaxone. The enhancement between ceftazidime and amoxicillin-clavulanic acid was only visible when these antibiotic discs were moved closer to each other (20 mm).
An outbreak investigation clearly documented spread of the organism to infants sharing beds and cubicles with the index case.
PFGE confirmed clonality of the isolates, and clusters of the same strain were noted in other hospitals in Gauteng. Eleven out of 14 strains studied had ≤3 band differences, 2 of which were from 2 different hospitals.
Control of the outbreak included isolating colonized or infected patients and emphasizing appropriate glove use and handwashing. Patients were treated initially with either imipenem or ciprofloxacin for 7–10 days, but later treatment was changed to ciprofloxacin for 4–6 weeks in an attempt to eradicate long-term carriage.
The genus Salmonella is not common hospital flora, and ESBL production with multiple antibiotic resistance is rarely associated with this organism. Constant antibiotic pressure can select multidrug-resistance and ESBL-producing bacteria, enabling their transmission among hospitalized patients.8 We cannot exclude the possibility that this mechanism of resistance was due to selective pressure, despite the evidence that very few patients had been exposed to the third-generation cephalosporins before isolation of the SI-ESBL.
In Africa, the source and mode of transmission of Salmonella species is unclear and includes animals, animal products, water and infected humans. Young children may excrete the organisms for up to 4 months, and outbreaks in hospitals and day-care centers support the importance of person-to-person spread.1,8 Person-to-person spread in hospital units is associated with poor infection control and is often exacerbated by staff shortages. Transmission from clinical and nursing staff carriers to patients is not thought to play a major role.9,10
Horizontal transmission was considered the most likely route of transmission. This is supported by the finding that some patients were microbiologically negative on stool and blood culture for SI-ESBL before their clinical presentation. The clustering of patients in wards and the timing of presentation also support the spread of the organism from patient to patient. In the outbreak investigation conducted, we documented spread of the organism to adjacent infants. Clear epidemiologic links between patients were documented among some cases; however, this information was not always possible to ascertain because of the numbers of patients in the wards and the retrospective nature of the data. Staff members were not screened during this outbreak.
It is not common to see ESBL expression between cefepime and amoxicillin-clavulanic acid. In this study, cefepime was a useful marker for ESBL detection, as we encountered hidden resistance with the third-generation cephalosporins. The absence of zones of inhibition around ceftriaxone and ceftazidime may have affected the potentiation of ESBL production. Our laboratories do not routinely perform MICs on stool isolates. This study should serve to reinforce the importance of template distance when screening for ESBL production and the fact that we should no longer confine ourselves to determining this expression only with third-generation cephalosporins.
ESBL genotyping of isolates from this outbreak have been described elsewhere, and many ESBL-producing Salmonella isolates in South Africa, including SI, harbor TEM 131 or TEM 63.5 The presence of TEM 131 in an organism characteristically manifests with increased in vitro resistance to cefepime.
Clinical features in Malawian children presenting with nontyphoidal salmonellosis included fever, cough and diarrhea1,2; overall mortality was 38%; and young age and clinical HIV infection were risk factors for death.2 Studies have reported that children infected with HIV in Africa have a higher risk of acquiring salmonellosis.1,3
This report should serve to increase the awareness and importance regarding the potential for outbreaks of ESBL-producing Salmonella spp. in immunosuppressed children.
The actual number of patients infected with ESBL strains is likely greater than that which we detected. Asymptomatic carriers or mild cases, which were missed, could be an explanation for the persistence and dissemination of this clone in the pediatric wards.
Special thanks to Mr J. Kolobe, Ms T. Kruger, Mrs A. Sooka, and Mrs. T. Capper for technical assistance, Professor H. K. Koornhof for reviewing the manuscript, NHLS/SAIMR for grant support, and to the patients, laboratory and nursing staff at CHB Hospital for their contribution.
1. Graham SM, Molyneux EM, Walsh AL, Cheesbrough JS, Molyneux ME, Hart CA. Nontyphoidal Salmonella
infections of children in tropical Africa. Pediatr Infect Dis J
2. Walsh AL, Phiri AJ, Graham SM, Molyneux EM, Molyneux ME. Bacteremia in febrile Malawian children: clinical and microbiologic features. Pediatr Infect Dis J
3. Nathoo KJ, Chigonde S, Nhembe M, Ali MH, Mason PR. Community-acquired bacteremia in human immunodeficiency virus-infected children in Harare, Zimbabwe. Pediatr Infect Dis J
4. Hadfield TL, Monson MH, Wachsmuth IK. An outbreak of antibiotic-resistant Salmonella
enteritidis in Liberia, West Africa. J Infect Dis
5. Kruger T, Szabo D, Keddy KH, et al. Infections with nontyphoidal Salmonella
species producing TEM-63 or a novel TEM enzyme, TEM-131, in South Africa. Antimicrob Agents Chemother
6. National Committee of Clinical Laboratory Standards. Performance Standards for Antimicrobial Susceptibility Testing: Fourteenth Informational Supplement: NCCLS document M100-S14
. Wayne, PA: NCCLS; 2004.
7. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol
8. Hammami A, Arlet G, Ben RS, et al. Nosocomial outbreak
of acute gastroenteritis in a neonatal intensive care unit in Tunisia caused by multiply drug resistant Salmonella
wien producing SHV-2 beta-lactamase. Eur J Clin Microbiol Infect Dis
9. Joseph CA, Palmer SR. Outbreaks of salmonella infection in hospitals in England and Wales 1978–87. BMJ
10. Tauxe RV, Hassan LF, Findeisen KO, Sharrar RG, Blake PA. Salmonellosis
in nurses: lack of transmission to patients. J Infect Dis