A new cross-sectional study from Brazil has found that nearly two-thirds of symptomatic people living with HIV show immunological evidence of prior exposure to Bartonella species, a group of bacteria transmitted by fleas, ticks, and other arthropods, yet rigorous molecular testing could not confirm active infection with Bartonella, Borrelia, or piroplasms in any participant. The research, published in the journal Parasites & Vectors, was conducted by a transnational team spanning North Carolina State University in the United States and the University of São Paulo and São Paulo State University in Brazil, and it offers one of the most detailed looks to date at how these understudied vector-borne pathogens intersect with immune-compromised patients in a tropical setting.
The motivation for the study rests on a growing recognition that Bartonella, Borrelia, and piroplasmid parasites circulate widely among arthropod vectors and animal reservoirs in Brazil, and that people with weakened immune systems may be particularly vulnerable to their consequences. Bartonella species are fastidious, intracellular bacteria best known for causing cat scratch disease in immunocompetent individuals, but in patients with profound immunosuppression they have been linked to severe vascular lesions such as bacillary angiomatosis, a condition historically described in people with advanced AIDS. Borrelia species, the spirochetes responsible for relapsing fever and Lyme-like illness, and piroplasmids, the protozoan parasites that include Babesia, add further layers of diagnostic complexity because they can produce nonspecific febrile symptoms that overlap with many opportunistic infections.
To investigate these possibilities, the researchers enrolled 75 HIV-positive individuals who were symptomatic and living in Brazil, collecting detailed epidemiological histories and medical records for each participant. The team then took a two-pronged diagnostic approach. The first prong was serological: they used indirect immunofluorescence assays, or IFA, to detect IgG antibodies against six distinct Bartonella antigens, namely Bartonella henselae, Bartonella koehlerae, Bartonella quintana, and two genotypes of Bartonella vinsonii subsp. berkhoffii. All assays were performed in a blinded manner, meaning the laboratory staff did not know the clinical status of the samples they were testing, a design feature intended to reduce bias. A reciprocal antibody titer of 1:64 or greater was defined as seroreactivity, the threshold at which a sample was considered to show evidence of prior exposure.
The second prong was molecular. The researchers extracted DNA from blood samples and searched for genetic signatures of Bartonella, Borrelia, and piroplasmid organisms using both quantitative polymerase chain reaction, or qPCR, and digital PCR, or dPCR. Crucially, they did not rely on a single snapshot of the bloodstream. Because Bartonella bacteria are notorious for circulating at extremely low numbers and in a cyclic fashion, the team enriched blood cultures in Bartonella alpha-Proteobacteria Growth Medium, known as BAPGM, for up to 28 days, and then repeated the qPCR testing on the cultured material. This enrichment step is designed to amplify any sparse organisms that might be present below the detection limit of direct blood testing, making it one of the most sensitive protocols available for hunting these bacteria in human blood.
The serological results were striking. Of the 75 participants, 47, or 62.7 percent, were seroreactive to at least one Bartonella antigen. The most frequently detected antibodies were directed against B. henselae, found in 57.3 percent of the cohort, followed by B. vinsonii subsp. berkhoffii genotype II, detected in 44.0 percent. These figures suggest that exposure to Bartonella species is remarkably common among symptomatic HIV-infected individuals in this Brazilian population, a finding consistent with the documented circulation of these bacteria in fleas, ticks, and animal reservoirs across the country. Whether the antibodies reflect true past infection, cross-reactivity with other bacteria, or persistent low-level colonization that evaded molecular detection remains an open question that the authors themselves acknowledge.
Perhaps the most clinically provocative finding was the statistical association between Bartonella seroreactivity and systemic manifestations of disease. Among the 47 seroreactive patients, 11 exhibited systemic signs and symptoms, compared with only 1 of the 28 non-seroreactive patients. This difference translated into an odds ratio of 8.25, with a 95 percent confidence interval spanning 1.00 to 68.2 and a p value of 0.026. In practical terms, seroreactive patients were roughly eight times more likely to present with systemic manifestations than those without detectable antibodies, although the wide confidence interval reflects the modest sample size and signals considerable statistical uncertainty. The association was exploratory in nature, and the authors were careful to frame it as a hypothesis-generating observation rather than proof of causation.
On the molecular front, the picture was far more definitive, and far more negative. Every one of the 75 participants tested negative for Bartonella and Borrelia DNA by both qPCR and dPCR, both before and after the extended BAPGM enrichment culture period. Two participants did show a low positive partition number for piroplasmid DNA by dPCR, a result that initially raised the possibility of Babesia-like infection, but when the samples were retested the signal could not be reproduced. The authors interpreted these non-repeatable findings as most likely representing technical artifacts or contamination rather than true infection, underscoring a central principle of molecular diagnostics: a single low-level positive result, especially near the limit of detection, demands confirmation before it can be considered evidence of active disease.
The disconnect between high seroprevalence and absent molecular detection is the intellectual heart of the study, and it speaks to a broader challenge in the field of vector-borne pathogen research. Antibodies indicate that the immune system has encountered an organism at some point, but they do not distinguish between a resolved infection years ago and a persistent, sequestered infection hiding in tissue reservoirs such as the vascular endothelium, where Bartonella is known to reside. Conversely, negative blood PCR does not exclude infection in tissues, because these bacteria may only intermittently spill into the bloodstream. The BAPGM enrichment culture was designed specifically to address this intermittency, and its failure to yield any PCR-positive samples after nearly a month of incubation provides reasonably strong evidence that active bloodstream infection with these organisms was absent in this cohort at the time of testing.
The clinical implications are nuanced. On one hand, the study provides reassurance that in this population of symptomatic HIV-infected Brazilians, active Bartonella, Borrelia, and piroplasmid infections were not detectable, suggesting that these pathogens are unlikely to be a common hidden driver of unexplained symptoms in similar patients. On the other hand, the strong association between seroreactivity and systemic manifestations hints that prior Bartonella exposure might somehow shape the clinical course of HIV disease, or that both are correlated with shared risk factors such as living conditions, animal contact, or arthropod exposure that were not fully disentangled in this exploratory analysis. The authors conclude that the clinical relevance of Bartonella seroreactivity in this setting remains to be clarified, a candid acknowledgment that serology alone cannot settle the question.
Looking forward, the study lays groundwork for several important next steps. Larger cohorts with longitudinal follow-up would help determine whether seroreactive patients experience different disease trajectories over time. Tissue-based sampling, rather than blood alone, could reveal whether Bartonella organisms persist in vascular niches that blood PCR cannot reach. And the combination of dPCR technology with extended enrichment culture, as deployed here, offers a template for future investigations of other fastidious vector-borne pathogens in immunocompromised populations worldwide. For now, the message for clinicians is measured: exposure to Bartonella appears frequent among symptomatic HIV patients in Brazil, but confirmed active infection with these three pathogen groups was not found, and the meaning of a positive antibody test in this population remains an open scientific question worthy of continued investigation.
Subject of Research: Prevalence of Bartonella, Borrelia, and piroplasmid exposure and infection in symptomatic HIV-infected individuals in Brazil
Article Title: Prevalence of Bartonella spp., Borrelia spp., and piroplasmid in HIV-infected symptomatic individuals in Brazil: a cross-sectional study
Article References: Robveille, C., Sarris, A. B., Breitschwerdt, E. B., André, M. R., do Amaral, R. B., Maggi, R. G., Mussi-Pinhata, M. M., Dias, C. M., & de Carvalho Santana, R. (2026). Prevalence of Bartonella spp., Borrelia spp., and piroplasmid in HIV-infected symptomatic individuals in Brazil: a cross-sectional study. Parasites & Vectors. https://doi.org/10.1186/s13071-026-07584-x
Image Credits: AI Generated
DOI: 10.1186/s13071-026-07584-x
Keywords: Bartonella, Borrelia, piroplasmid, HIV, vector-borne disease, serology, immunofluorescence assay, qPCR, digital PCR, BAPGM culture, Brazil, opportunistic infection
News Source: Kristina Jarvis. (October 5, 2026). Hidden Tick- and Flea-Borne Probes in HIV Patients Reveal High Bartonella Exposure but No Active Infection. Scienmag.
