Five biomarkers separated a more invasive Klebsiella bloodstream infection subgroup in Japan
A five-marker PCR profile carved out a clinically distinct form of hypervirulent Klebsiella pneumoniae bloodstream infection in a Japanese tertiary hospital, where abscess complications were far more common and treatment courses stretched much longer than in classical infections. The genomic signature consisted of rmpA, rmpA2, iucA, iroB, and peg-344 — a combination meant to capture the pathotype’s linked machinery for capsule regulation, iron acquisition, and invasive spread.
The difference was not subtle at the bedside. Among 207 bloodstream infection episodes, 28 met the 5-biomarker definition, and 61% of those episodes were accompanied by abscess complications compared with 13% of classical K. pneumoniae episodes.
Abscess formation tracked with the hypervirulent genotype
The study drew from consecutive bloodstream infection episodes at Kameda Medical Center in Kamogawa, Japan, between January 1, 2022, and December 31, 2024. Investigators excluded species that proved to be K. variicola or K. quasipneumoniae on PCR-based species-level differentiation, leaving 207 true K. pneumoniae episodes for analysis. The median patient age was similar in both groups, with 76 years in the hypervirulent cohort and 78 years in the classical cohort, which makes the divergence in complications harder to dismiss as a simple age effect.
By design, abscesses were counted only when imaging confirmed them. That stricter definition strengthens the signal: the excess was not built on clinical suspicion alone, but on lesions visible enough to be documented radiographically. In practical terms, the hypervirulent phenotype behaved like a pathogen with a built-in ability to seed deeper tissue, sustain metastatic infection, and force source-control decisions that go well beyond routine bacteremia management.
Longer hospitalization followed the same pattern. Median post-onset length of stay reached 28 days for hvKp episodes versus 14 days for classical infections, and total antimicrobial treatment lasted 43 days versus 14 days. Those longer courses fit the biology of invasive hypervirulent K. pneumoniae: once abscesses form, antimicrobial therapy alone rarely closes the case, because drainage, repeated imaging, and prolonged treatment become part of the price of microbial escape from the bloodstream.
Sequence type 23 carried most of the abscess burden
The lineage data pointed toward a well-known invasive clone. Nine of 10 sequence type 23 hypervirulent episodes involved abscesses, linking this genetic background to the complication profile seen in the cohort. ST23 has long been associated with invasive liver abscess syndromes, and the current dataset fits that clinical pattern with a direct bloodstream-infection readout.
The study also showed that virulence and resistance can coexist in the same clinical space, even if not yet on a large scale in this hospital series. Short-read whole-genome sequencing found CTX-M–type extended-spectrum β-lactamase genes in 3 of 28 hypervirulent isolates, representing 11% of the hvKp group and 1.4% of all K. pneumoniae episodes. For clinicians, that matters because the older mental model of hypervirulence as a purely susceptible organism is no longer safe. The biological threat is no longer just invasion; it is invasion paired with narrowing treatment options.
Why the 5-biomarker definition matters for surveillance
The pathotype question has been unsettled for years because definitions of hypervirulence have varied across studies. Some use string tests or phenotypic proxies, while others rely on subsets of virulence genes. The five-biomarker definition used here takes a stricter approach, requiring all five markers together rather than treating any one virulence feature as sufficient. In a surveillance setting, that kind of threshold reduces noise and helps identify the subgroup most likely to carry the clinical burden of metastatic infection.
Investigators classified isolates retrospectively, so clinicians caring for the patients did not have the hvKp label available in real time. Even so, the outcome pattern emerged clearly. The 5-biomarker definition separated a group with more abscesses, longer hospitalization, and longer therapy, which supports its use as a genomic surveillance tool for emerging hypervirulent K. pneumoniae.
The work also fits a broader infection-control concern: hypervirulent lineages are increasingly discussed alongside antimicrobial resistance genes. When a strain’s virulence plasmid and β-lactam resistance determinants coexist, the result is a pathogen that can both spread aggressively within the body and erode standard empiric treatment choices. That combination has clear implications for hospital microbiology labs and for clinicians deciding when to image aggressively, pursue source control, and anticipate prolonged therapy.
A clinically distinct subgroup, not just a molecular label
The study’s design helped anchor the conclusions in routine care rather than in abstract genomics. Blood cultures were processed through standard laboratory platforms, susceptibility testing followed Clinical and Laboratory Standards Institute criteria, and the analysis compared abscess complications, length of stay, antimicrobial duration, and 30-day mortality risk across prespecified models. The primary signal centered on abscess burden and treatment intensity, not on a laboratory artifact.
Even the exploratory stratification by biomarker count pointed in the same direction: the full five-marker profile defined the clearest clinical phenotype. Episodes with only one to four markers were handled as part of the classical group for the main analysis, leaving the complete biomarker set to serve as the threshold for hypervirulence. That approach makes the resulting definition more conservative, and likely more useful for surveillance programs trying to avoid overcalling hypervirulence from incomplete virulence signatures.
A standout finding from the cohort is how quickly the microbiologic category translated into a care burden. When abscesses are present, imaging, procedural consultation, and extended antimicrobial exposure are not optional extras; they become the core of management. The genotype therefore maps onto a real and costly clinical course, which is exactly why the five-marker approach has value beyond taxonomy.
The cohort was single-center and retrospective, and the sequencing budget limited whole-genome analysis to a subset of isolates, even though all hypervirulent isolates were included. Those constraints matter for generalizability, especially across hospitals with different case mixes and resistance pressures. Still, the clinical separation was strong enough to argue that the 5-biomarker definition captures a meaningful bloodstream infection subgroup in Japan — one that deserves close genomic surveillance as hypervirulent K. pneumoniae and resistance determinants continue to converge.