Deadly ESBL Klebsiella Outbreak Swept a Bangui Neonatal Unit, Study Finds

An outbreak of bloodstream infections caused by extended-spectrum beta-lactamase-producing Klebsiella pneumoniae, or ESBL-KP, swept through a neonatal intensive care unit in Bangui, Central African Republic, infecting 44 newborns and killing more than a third of them, according to a retrospective case-control study published in Antimicrobial Resistance & Infection Control. The outbreak ran from 25 October 2023, when the first confirmed case was admitted, to 14 January 2024, when the last case was discharged, at the Centre Hospitalier Universitaire Communautaire, a Ministry of Health hospital whose neonatal services are supported by Médecins Sans Frontières. It is the first formally published analysis of a neonatal outbreak in the country, and its authors argue that the event exposes critical gaps in hospital management that low-income settings elsewhere cannot afford to ignore.

The numbers behind the outbreak are stark. Of 242 neonates admitted to the unit during the outbreak period, 44 became laboratory-confirmed cases and 153 served as controls in the case-control analysis. The incidence rate of confirmed ESBL-KP bloodstream infection was 29.7 cases per 1,000 patient-days, and cases accounted for 18.2 percent of all admissions. Mortality among cases reached 34.1 percent, compared with 9.8 percent among controls, a difference of 24.3 percentage points that the authors report as 3.5 times higher with a p-value below 0.0001. Hospitalisation was also roughly twice as long for cases, averaging 14.5 days against 6.8 days for controls, a statistically significant difference.

A particularly troubling feature was how early in life many infections appeared. Twenty-five of the 44 cases, or 56.8 percent, were classified as early sepsis, meaning suspected sepsis was diagnosed within three days of hospitalisation. Because ESBL-KP circulates endemically in the community, the study team deliberately included both early- and late-onset sepsis among cases to capture every potential source of infection, from vertical transmission during delivery by a colonised mother to transmission inside the hospital. Eleven pairs of twins were hospitalised during the outbreak, and in three of those pairs both twins became confirmed cases.

The case-control analysis pointed firmly toward the delivery room rather than the intensive care unit alone. After adjustment in a multivariate logistic regression model, the only factors independently associated with ESBL-KP bloodstream infection were the presence of maternal risk factors, which the study defines as maternal sepsis, meconium-stained amniotic fluid during labour, or premature rupture of membranes, and gestational age at birth of less than 31 weeks. In unadjusted analysis, resuscitation at birth more than doubled the odds of infection, with an odds ratio of 2.61, and a significant trend linked lower birth weight to higher risk. The authors note that published data consistently report maternal risk factors in neonatal multidrug-resistant organism outbreaks, although more recent evidence on maternal colonisation specifically has been conflicting.

The neonatology unit’s occupancy rate exceeded 100 percent in early October 2023, and infection prevention and control assessments using MSF’s stepwise SIPCA tool revealed poor adherence to IPC measures at the beginning of the outbreak. Only four ESBL-KP bloodstream infections had been recorded at the hospital in the first nine months of 2023; cases began rising at the end of October, and a probable outbreak was declared on 6 November after the first three cases were identified. The peak came on 6 December. The authors conclude that ESBL-KP probably spread due to the high occupancy rate and poor IPC standards, and possibly through multiple simultaneous sources, though they were unable to identify the source of transmission definitively.

The laboratory picture underscored how limited the treatment options were. Antimicrobial susceptibility testing of the 44 isolates, performed by the Kirby-Bauer method and read with the Antibiogo digital application, confirmed 100 percent resistance to gentamicin, a first-line treatment for neonatal sepsis in MSF protocols, and 97.3 percent resistance to ciprofloxacin. Imipenem, the only carbapenem tested, remained active against just over half of the isolates, with 23 cases, or 52.3 percent, showing susceptibility, 20 requiring increased exposure, and one resistant isolate. During the outbreak, clinicians replaced gentamicin with amikacin in the first-line regimen for suspected cases, based on consistent susceptibility to amikacin, and initiated meropenem once ESBL-KP was confirmed in blood cultures.

The response combined clinical, microbiological and environmental measures. An outbreak committee led by the medical director met weekly, suspected cases were immediately isolated, confirmed cases were cohorted, and admissions from external health posts were suspended. Hand hygiene with alcohol-based rub was reinforced with dispensers at every cot and room entrance, gowns became mandatory in the neonatal service, single-use equipment was introduced for case care, and incubators, rooms and the delivery room underwent deep cleaning. In mid-December, with cases still appearing, stricter rules added direct monitoring of hand hygiene and a physically delimited isolation area. Only mothers could stay with their neonates. The authors are candid that they could not establish a clear link between any single intervention and the outbreak’s resolution, but a series of IPC interventions in the first month successfully controlled further spread, and the outbreak was declared over in approximately two months.

The context in Bangui made this outbreak especially foreseeable. A 2021 study found that 71.2 percent of children in the city were colonised with ESBL-producing Enterobacterales, indicating community circulation of these pathogens. A retrospective analysis of neonatal bacteraemia at the same hospital found resistance to third-generation cephalosporins in 100 percent of K. pneumoniae isolates tested between 2018 and 2020. Previous ESBL-KP outbreaks had already struck the unit in 2017 and 2022; the 2017 outbreaks included 25 confirmed cases with a 28 percent case-fatality rate. Literature reviews of NICU outbreaks are dominated by high- and middle-income countries, with only one paper from a middle-income country among 39 in one recent review, leaving low-income settings like the Central African Republic largely undocumented.

The study has limitations the authors acknowledge plainly. Data on maternal colonisation were lacking, and no molecular typing was performed, so the true transmission dynamics remain uncertain; although the strains differed slightly in their phenotypic resistance profiles, clonality cannot be ruled out given the limits of disk diffusion methods. Matching by age was not possible because age data for controls were unavailable, leaving confounder control inadequate, and potential transmission routes such as contaminated incubators, maternal milk handling or infusion bags were not investigated. Delivery mode was missing for most controls. Including early-onset sepsis among cases was a deliberate, if controversial, choice intended to highlight prevention strategies during delivery and early neonatal care.

For the authors, the broader lesson is about access. The hospital’s co-management arrangement with MSF enabled access to a quality-assured microbiology laboratory, which allowed prompt identification of the outbreak, adaptation of antibiotic treatment and initiation of prevention activities; in settings without such laboratory access, spikes in neonatal mortality may go unrecognised as resistant bacterial outbreaks. They argue that basic IPC measures, which are less expensive than microbiology or carbapenem access, should be prioritised as a feasible first step, alongside a structured national response to antimicrobial resistance in line with World Health Organization recommendations. Without investment in staffing, laboratory capacity and consistently implemented infection control, they warn, vulnerable newborns in low-income countries will remain exposed to outbreaks that better-resourced systems can detect and contain.

Subject of Research: An ESBL-producing Klebsiella pneumoniae bloodstream infection outbreak in a neonatal intensive care unit in Bangui, Central African Republic

Article Title: Extended spectrum beta-lactamase Klebsiella pneumoniae (ESBL-KP) bloodstream infection outbreak in a Neonatal Intensive Care Unit (NICU) in Bangui, Central African Republic (CAR)

Article References: Comelli, A., Ebode Ela, M., Baluhe Muke, I., Danwesse, E., Manirarora, S., Nazita, S. N., Mafuko, J. M., Isango, A., Yasselet, G., Ndrouma, G., Tombola, B., Vaizi, B. T., F Ogundipe, O., Sepou, A., Ngbalé, R. N., Gordillo Gomez, F., Martino, C., Farra, A., Khalife, M., & Garcia-Vello, P. (2026). Extended spectrum beta-lactamase Klebsiella pneumoniae (ESBL-KP) bloodstream infection outbreak in a Neonatal Intensive Care Unit (NICU) in Bangui, Central African Republic (CAR). Antimicrobial Resistance & Infection Control, 15(1), Article 121. https://doi.org/10.1186/s13756-026-01784-x

Image Credits: AI Generated

DOI: 10.1186/s13756-026-01784-x

Keywords: ESBL, Klebsiella pneumoniae, neonatal intensive care unit, bloodstream infection, outbreak, Central African Republic, antimicrobial resistance, infection prevention and control, newborns, Médecins Sans Frontières, low-income countries, case-control study