The research team, led by A. M. G. A. Laheij of the Academic Centre for Dentistry Amsterdam, followed 186 adult patients scheduled for autologous or allogeneic transplantation at five centers in Sweden, the United States, and the Netherlands. Recruitment took place between 2011 and 2018, and every participant underwent a comprehensive oral examination, complete with radiographs, one to eight weeks before the start of conditioning therapy. Investigators documented cavities reaching into the dentin or pulp, periapical lesions, periodontal pockets deeper than five millimeters, partially impacted teeth, plaque levels, and bleeding on probing. Urgent dental treatment was provided where needed. After the transplant, calibrated examiners visited patients at the bedside three times a week, scoring oral mucositis with both the World Health Organization toxicity scale and the modified Oral Mucositis Assessment Scale, which grades the cumulative ulcerated area across nine sites in the mouth. Body temperature was recorded daily, and fever was defined either as 38.0 degrees Celsius or, in a stricter analysis, as 38.5 degrees Celsius.

The headline numbers are striking. Nearly seven in ten patients, 69.4 percent, developed fever of at least 38.0 degrees Celsius during the pre-engraftment phase, and 55.9 percent crossed the higher 38.5-degree threshold. Meanwhile, 43.1 percent of patients developed an oral ulcerative or pseudomembranous lesion, and 38.1 percent experienced ulcerative oral mucositis of WHO grade 2 or worse, with ulceration typically peaking between days nine and eleven after the transplant. Blood cultures were obtained from 104 patients according to institutional protocols, and half of those cultures came back positive, a microbiological confirmation rate of 48 percent that is considerably higher than the 20 to 25 percent typically reported in comparable cohorts. The researchers attribute this elevated detection rate to the unusually frequent temperature monitoring and culture sampling in the study, which caught transient bacteremias that routine surveillance might miss. The organisms recovered were diverse, with Staphylococcus epidermidis, Escherichia coli, and Staphylococcus aureus the most common isolates, and twelve of the fifty positive cultures were polymicrobial.

When the team ran the statistical models, the pattern that emerged challenged decades of clinical dogma. Traditionally, the duration of neutropenia, the period during which the patient has virtually no infection-fighting neutrophils, has been considered the dominant risk factor for fever. In the full multivariate model, which adjusted for age, study site, conditioning regimen, diarrhea, vomiting, and the duration of leukopenia, patients with oral ulcerations of one square centimeter or larger had more than six times the odds of developing fever compared with those whose ulcers were smaller, an odds ratio of 6.36. Vomiting, a surrogate marker for gastrointestinal mucosal injury, carried an odds ratio of 3.72, and each additional day of leukopenia added a modest but significant increase in risk. Yet the presence of any pre-transplant oral pathology, whether untreated caries, periapical lesions, deep periodontal pockets, or impacted teeth, showed no significant relationship with fever whatsoever, with an odds ratio of 1.09 and a p-value of 0.82.

The choice of fever threshold turned out to matter enormously. When the analysis was repeated using the stricter 38.5-degree cutoff, only the size of the oral ulceration remained significantly associated with fever, while vomiting, diarrhea, and the duration of leukopenia all dropped out of significance. The authors interpret this as evidence that higher fever thresholds may better capture clinically meaningful inflammation driven by mucosal barrier injury. Ulcerated mucosa, they note, facilitates microbial translocation into the bloodstream and triggers the release of inflammatory cytokines, which can produce a pronounced febrile response even when no culturable organism is recovered. In other words, a large oral ulcer may itself behave like an infection, generating systemic inflammation that registers as fever, and the higher temperature cutoff may filter out the low-grade, noninfectious temperature elevations that muddy the clinical picture.

Equally revealing is what the study found about the scoring systems themselves. The WHO classification of oral mucositis, which weighs the functional impact of the condition, such as whether a patient can eat solid food, showed no significant association with fever. The OMAS score, which quantifies the actual size and extent of ulcerative lesions, did. The discrepancy makes anatomical sense: a patient can harbor a large ulcerated area that does not interfere with swallowing and therefore earns a low WHO grade, yet that same raw surface represents a wide-open gateway for bacteria and their toxins. The finding suggests that clinicians and trialists who rely solely on functional grading may be underestimating the infectious and inflammatory potential of treatment-induced mucosal wounds, and that lesion-based measurements deserve a more central place in risk assessment.

The results also carry practical implications for pre-transplant dental care. For years, guidelines have recommended aggressive treatment of periodontitis and other chronic oral infections before transplantation, on the theory that these reservoirs of bacteria could flare during neutropenia and seed the bloodstream. The Orastem data, together with several recent studies, paint a more nuanced picture. In this cohort, most chronic oral foci were actually eliminated before conditioning began, and exacerbations of dental or periodontal disease during the engraftment phase were rare. A large cohort study of 500 transplant recipients previously found that multiple oral foci were associated with longer hospital stays but not with early mortality, and a retrospective analysis of nearly 16,000 patients suggested that pretreatment oral management may reduce post-transplant infections without affecting severe mucositis or survival. The authors argue for personalized dental protocols rather than a one-size-fits-all mandate for extraction.

The study is not without limitations, and the authors are candid about them. Data from one participating center in Vancouver were incomplete and excluded, and thirty additional participants were dropped for insufficient data, leaving 186 of the original 239 patients for analysis. Because neutrophil-specific counts were not consistently available, the team used the duration of leukopenia, defined as a white blood cell count below 0.5 billion cells per liter, as a proxy for neutropenia. The dataset also lacked information on clinical infections originating outside the mouth, such as in the lungs, urinary tract, or skin, which prevented a direct comparison of oral versus extra-oral infection risks. Differences in prophylaxis regimens, catheter types, and blood-culture thresholds among the five centers introduced institutional variability, although the models adjusted for study site. Fever incidence itself differed between centers, likely reflecting variations in surveillance and supportive care rather than true biological differences.

Looking forward, the Orastem findings reinforce a concept that has been gaining traction in hematology: febrile mucositis, the idea that barrier injury in the mucosa, whether oral or gastrointestinal, is itself a driver of the febrile response, distinct from classical bloodstream infection. Interventions that reduce mucositis, such as oral cryotherapy during high-dose melphalan conditioning, photobiomodulation therapy, and keratinocyte growth factor in selected autologous regimens, may therefore do double duty, easing patient suffering while potentially lowering the burden of fever and empirical antibiotic use. The authors also highlight the oral microbiome as a frontier, noting that dysbiosis has been linked to mucositis severity and that prebiotic or microbial-based strategies warrant investigation. For now, the message for transplant teams is concrete: when a patient spikes a fever in the fragile days before engraftment, the mouth deserves a careful look, because the size of what is breaking down there may explain why the thermometer is climbing.

Subject of Research: The association between oral ulcerative lesions, oral infections, and fever during the pre-engraftment phase after hematopoietic cell transplantation

Article Title: Association between oral ulcerative lesions, oral infections, and fever during the pre-engraftment phase after hematopoietic cell transplantation: results from the Orastem study

Article References: Laheij, A. M. G. A., Raber-Durlacher, J. E., Hasséus, B., Huysmans, M. C., van Leeuwen, S. J. M., Legert, K. G., Johansson, J. E., Isom, S., Kline, D. M., Hazenberg, M. D., Rozema, F. R., Brennan, M. T., von Bültzingslöwen, I., & Blijlevens, N. M. A. (2026). Association between oral ulcerative lesions, oral infections, and fever during the pre-engraftment phase after hematopoietic cell transplantation: results from the Orastem study. Supportive Care in Cancer, 34(10), Article 1044. https://doi.org/10.1007/s00520-026-11267-9

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11267-9

Keywords: hematopoietic cell transplantation, oral mucositis, fever, febrile neutropenia, oral ulceration, dental infections, pre-engraftment phase, mucosal barrier injury, bloodstream infection, supportive care in cancer, Orastem study, oral microbiome