Investigators examined documentation of dental disease around bacteremia episodes and found higher recording in patients with Staphylococcus aureus bacteremia (SAB) than in those with Escherichia coli bacteremia. Dental disease occurred in 528 of 16,435 SAB patients (3.21%) and in 179 of 24,942 E. coli bacteremia patients (0.72%). After adjustment for age, sex, race, and documentation intensity, the adjusted odds ratio for documented dental disease in SAB versus E. coli bacteremia was 3.86 (95% CI, 3.23 to 4.63). The study team reported these comparisons using the Protege database covering 2015–2023.
The authors defined documented dental disease by either a qualifying dental procedure or disease-specific note text. They searched a restricted window: from 7 days before the bacteremia index through 30 days after the index date. This operational definition was applied consistently across cohorts; the source did not report additional diagnostic criteria or dental examination protocols. The Protege data and derived code lists, protocol, and analysis plan are available from the corresponding author on request, but the underlying Protege data are not publicly available and access requires data-owner approval.
Data indicated a substantially greater likelihood of documented dental disease in SAB compared with E. coli bacteremia after multivariable adjustment. The adjusted odds ratio was 3.86 (95% CI, 3.23–4.63). The authors also used a pathology-only definition for disease and compared SAB with a general population cohort: by that definition, 517 SAB patients (3.15%) had documented dental disease vs 399 general population patients (0.49%), yielding an adjusted odds ratio of 3.71 (95% CI, 3.08–4.46). The study team adjusted analyses for demographic variables and a measure they termed documentation intensity; beyond that, the source text did not report additional model covariates or the full variable construction.
Within the SAB cohort, prevalence of documented dental disease varied by clinical and social-risk subgroups. The highest rates were in patients with prior infective endocarditis (9.8%), followed by opioid use disorder (6.2%), injection drug use (5.3%), and presence of a prosthetic valve (5.0%). The authors reported these subgroup prevalences without further stratified inferential statistics in the abstract. The source did not provide detailed breakdowns by age bands, comorbidity indices, or antimicrobial exposures in the abstract.
Among 437 SAB patients with an assignable first finding date for dental disease, 333 (76.2%) had the dental finding recorded after the bacteremia index date. The study team observed that most dental findings were recorded during the evaluation for SAB rather than preceding the bacteremia. The authors stressed that timing of documentation in electronic records does not establish causality; the observational data cannot determine whether dental disease preceded, followed, or contributed causally to the bloodstream infection beyond the sequence recorded in notes and procedures.
The analysis used the Protege database and included adults with bacteremia from January 1, 2015, through December 31, 2023. The analytic cohorts totaled 48,975 adults with bacteremia and 81,945 general population adults. Authors declare that before delivery the data were de-identified through expert determination. An external reviewer, Bluebonnet Ethical Review, classified the activity as Not Human Subjects Research (NHSR) and determined IRB review was not required. The Protege data are not publicly available; access is governed by the data owner. The study protocol, statistical analysis plan, code lists and aggregate results are available from the corresponding author on reasonable request, according to the source.
The authors explicitly stated that these data do not demonstrate that recorded dental disease caused the bacteremia. The analysis is retrospective and based on documentation (procedures and note text) rather than standardized dental examinations; the source did not report prospective dental assessments or confirmatory microbiologic linkage between oral pathogens and bloodstream isolates. The report is a preprint and has not undergone peer review. The Protege dataset is not publicly accessible, which may limit independent validation unless data access is granted by the owner. Those are the specific uncertainties and limitations noted in the source; the authors did not provide additional methodological details or downstream clinical recommendations in the abstract.