The story began in 2010, when four infants on the ward developed bacteremia — bacteria circulating in their bloodstream — caused by a strain of S. aureus that had become resistant to tobramycin, an aminoglycoside antibiotic widely used against serious gram-positive and gram-negative infections. The isolates all carried the spa type t084, a genetic fingerprint based on the sequence of the protein A gene that allowed microbiologists to confirm the infants were infected by a single circulating clone rather than unrelated strains. This clonal identity was the smoking gun of an outbreak. In response, the clinical team launched point prevalence surveys, systematically swabbing both infants and staff members to map who was carrying the organism, and then instituted active screening of all admitted infants so that any new colonization would be detected in real time. Screening of this kind transforms an invisible problem into a measurable one, giving infection control teams a surveillance signal they can track week by week.

What happened next illustrates how deceptive outbreak timelines can be. When the investigators later went back and retrospectively analyzed samples that had been collected before the outbreak was recognized, they found traces of the resistant strain as early as 2009 — meaning the organism had been circulating on the ward for months, perhaps longer, before anyone knew it was there. This kind of silent spread is a hallmark of S. aureus in neonatal settings. The bacterium colonizes the skin, nostrils and gut of infants without causing symptoms, hitchhiking on hands, equipment and surfaces, and only occasionally flares into invasive disease. By the time four babies had septicemia, the resistant clone was likely already woven into the microbial ecology of the unit, present on the skin of colonized infants and, crucially, in the environment itself.

The first line of attack was classical infection control. In 2010, and again in 2014, the ward implemented strengthened hygiene interventions, drilling staff on hand hygiene routines and contact precautions designed to interrupt the chain of transmission from colonized infants to susceptible ones. At the same time, the unit introduced pre-pasteurization of donor human milk, heat-treating donated breast milk before it reached the infants to eliminate bacteria that might otherwise serve as an inoculum. These measures produced a genuine, measurable effect: the overall prevalence of S. aureus carriage among the infants fell significantly, from 71 percent during 2010 and 2011 to 52 percent by 2013. For a vulnerable population, that reduction mattered — fewer colonized infants means fewer potential reservoirs for invasive infection. Yet the researchers observed something troubling. While the total burden of S. aureus declined, the proportion of isolates that were tobramycin-resistant did not budge. Hygiene alone, it seemed, was shrinking the pond without draining the resistant fish from it.

Between 2012 and 2013, the NICU underwent physical reconstruction, a drastic but not unprecedented step in stubborn outbreak situations. Rebuilding a ward allows teams to strip out reservoirs that routine cleaning cannot reach: contaminated sinks, porous surfaces, ventilation dead zones and hidden niches where bacteria persist for years. Consistent with the idea that the built environment was harboring the clone, environmental screenings stopped detecting the resistant strain after 2012. But the infants themselves continued to carry it, and the proportion of resistant isolates among their S. aureus populations remained stubbornly high. The outbreak, in other words, had two intertwined reservoirs — the physical ward and the colonized babies — and eliminating only one of them was not enough.

The decisive intervention came in 2014, when the unit substituted amikacin for tobramycin in its antibiotic regimen. The logic is elegant and rooted in the ecology of resistance. Tobramycin use creates a powerful selective pressure: any S. aureus strain carrying tobramycin resistance enjoys a survival advantage in an environment saturated with that drug, allowing it to outcompete susceptible strains and dominate the ward’s microbial landscape. Removing tobramycin and replacing it with amikacin — a different aminoglycoside to which the outbreak strain had no such advantage — reversed the selection. The resistant clone lost its ecological edge, and susceptible organisms could once again hold their ground. This is the same principle that underlies antibiotic stewardship programs worldwide: the drugs we use shape the resistance profiles of the bacteria around us, sometimes more powerfully than any amount of cleaning.

The results, however, tested everyone’s patience. Only after the ward reconstruction and the change in antibiotic regimen did the proportion of tobramycin-resistant S. aureus among the infants begin to decrease significantly. And even then, the decline was gradual. The outbreak was not declared apparently resolved until 2019 — five full years after the antibiotic substitution. Encouragingly, no cases of TRSA septicemia occurred at all after the regimen change, meaning the most dangerous manifestation of the outbreak was eliminated almost immediately once the selective pressure was removed. The slow fade of colonization in the following years reflects the natural turnover of the neonatal population: infants are discharged within weeks and new, uncolonized babies are admitted, so once transmission is truly broken and selection is lifted, the resistant clone is progressively diluted out of the ward.

The decade-long arc of this outbreak carries sobering lessons for hospitals everywhere. First, retrospective testing matters: the strain was present in 2009, a year before the outbreak was recognized, and earlier detection might have shortened the entire saga. Second, hygiene interventions, while essential and demonstrably effective at reducing overall S. aureus prevalence, could not on their own eliminate a clone that was being actively favored by the antibiotic in use. Third, environmental and patient reservoirs must be tackled simultaneously — reconstruction cleaned the ward, but only the antibiotic change broke the resistance advantage. The authors conclude that successful outbreak control requires a combination of adherence to hygiene routines, pasteurization of donor human milk, and a correct antibiotic regimen, and their data show each element contributing a distinct piece of the solution. The continued distribution of summarized screening results to staff, serving as a standing reminder to adhere to hygiene routines, also formed part of the sustained response.

For the field of neonatology and infection control, the study is a striking demonstration of the interplay between the ward environment and neonatal colonization. It also underscores the value of long-term surveillance: without years of active screening data, the researchers could never have disentangled which interventions worked, when they took effect, and how long the tail of an outbreak stretches after the last severe case. As antimicrobial resistance continues to climb globally, neonatal units — where treatment failure carries the highest imaginable stakes — are watching closely. The Swedish experience suggests that when a resistant clone digs in, the answer may lie not in any single dramatic measure but in patient, layered ecological warfare: clean the environment, protect the milk supply, screen relentlessly, and above all, choose antibiotics that do not hand the enemy an advantage.

Subject of Research: A decade-long tobramycin-resistant Staphylococcus aureus outbreak in a Swedish neonatal intensive care unit and the hygiene and antibiotic interventions that resolved it

Article Title: Combatting a Tobramycin-resistant Staphylococcus aureus outbreak in a neonatal intensive care unit: the impact of hygiene intervention and antibiotic substitution

Article References: Combatting a Tobramycin-resistant Staphylococcus aureus outbreak in a neonatal intensive care unit: the impact of hygiene intervention and antibiotic substitution. (n.d.). https://doi.org/10.1186/s12879-026-14481-2

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14481-2

Keywords: Staphylococcus aureus, neonatal intensive care unit, antimicrobial resistance, tobramycin, amikacin, outbreak control, infection control, bacteremia, donor human milk pasteurization, active screening, hospital hygiene, antibiotic stewardship