Figures
Abstract
The Bartonella genus consists of neglected pathogens associated with potentially transfusional-transmitted and fatal human diseases. We aimed to evaluate Bartonella sp. prevalence in 500 blood donors and compare the results with the data already published about these samples. We used molecular diagnostic methods to detect Bartonella sp.-DNA from blood and liquid culture samples: (A) conventional PCR for two gene regions, the ITS targeting the genus Bartonella and the specific gltA Bartonella henselae; (B) nested PCR for the ftsZ gene and (C) qualitative real-time PCR for the gltA gene, both B. henselae specific. We obtained 30/500 (6%) DNA detections from the blood samples; 77/500 (15.4%) DNA detections from liquid culture samples and five (1%) samples had DNA detection from both. In total, we detected B. henselae DNA from 102/500 (20.4%) donors. The samples used in this study had already been submitted for Bartonella sp.-DNA detection using only a conventional PCR in liquid culture. Sixteen samples (3.2%) were positive previously, and from these 16 samples, 13 were negative in the new investigation. We concluded that the use of liquid culture combined with different molecular tests increases the possibility of detecting Bartonella sp.-DNA, but the tests do not avoid false-negative results. More than a fifth of blood donors had at least one PCR that detected Bartonella sp.-DNA among the eight molecular reactions performed now (four reactions in whole blood and four in liquid culture). Seven percent had B. henselae-DNA detection for two or more distinct regions. Considering the results obtained previously, the DNA of Bartonella spp. was detected or the agent isolated in 23% of analyzed blood donors. The results establish that the low bacteremia and the fastidious characteristics of the bacterium are challenges to laboratory diagnosis and can make it difficult to confirm the infection in patients with bartonelloses.
Author summary
Bartonella are bacteria that can infect humans and cause fatal diseases. They can cause chronic infection and can potentially be transmitted by transfusion since they infect red blood cells. They are difficult to cultivate in a laboratory, and they are present in low numbers in blood. There is no laboratory test that is sufficiently sensitive to detect them. To compare several laboratory diagnostic tests, we searched for these bacterial DNAs in blood and in microbiological liquid cultures of 500 blood donors. We used diverse molecular techniques and then compared the results with the previously published project. We concluded that the use of liquid culture combined with different molecular tests increases the possibility of detecting Bartonella sp.-DNA, but the tests do not avoid false-negative results. We found Bartonella henselae DNA in the blood of at least one in five donors. Hemovigilance programs are unlikely to contribute substantially to the identification of chronic posttransfusion infections since they are designed to identify well-defined acute outcomes, so it is urgent to review B. henselae transfusional risk of transmission and Bartonella sp. infection diagnosis in donors and in patients.
Citation: Drummond MR, dos Santos LS, de Almeida AR, Lins KdA, Barjas-Castro ML, Diniz PPVdP, et al. (2023) Comparison of molecular methods for Bartonella henselae detection in blood donors. PLoS Negl Trop Dis 17(6): e0011336. https://doi.org/10.1371/journal.pntd.0011336
Editor: Ivo Elliott, Laos-Oxford-Mahosot Hospital Wellcome Trust Research Unit, LAO PEOPLE’S DEMOCRATIC REPUBLIC
Received: August 12, 2022; Accepted: April 25, 2023; Published: June 1, 2023
Copyright: © 2023 Drummond et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The authors received funding from Doctoral scholarship by The Brazilian National Council for Scientific and Technological Development (CNPq) 159717/2013-2 (to MRD), The Brazilian National Council for Scientific and Technological Development (CNPq) productivity grant 301900 / 2015- 9 (to PENFV); The São Paulo Research Foundation (FAPESP) Regular Research Assistance 2013/14211-3 (to PENFV) and Master's Scholarship by The Brazilian National Council for Scientific and Technological Development (CNPq) 4570/2018 (to LSDS). https://www.gov.br/cnpq/pt-br, https://fapesp.br/en. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The genus Bartonella comprises small coccobacillary, gram-negative, facultative intracellular bacteria belonging to the alpha-2 subgroup of the phylum Proteobacteria [1]. One of their most important characteristics is fastidious growth [2–5]. They infect erythrocytes and endothelial cells, usually causing chronic and cyclic bacteremia in their hosts [4,6–8]. Bacteria of this genus have been linked with many diseases, one of them described from pre-Inca times [9,10].
The genus Bartonella has more than 47 species and subspecies, and at least 17 of them have already been related to clinical manifestations in humans [5,7]. Of these, three species are associated with the largest number of diseases: Bartonella bacilliformis, Bartonella quintana and Bartonella henselae [11], the latter being the most frequent species [12–14].
Several manifestations have already been related to infection by Bartonella spp. including Carrión disease (B. bacilliformis) [15]; trench fever, culture-negative endocarditis, bacillary angiomatosis, and chronic bacteremia (B. quintana) [14,16]; cat scratch disease (CSD), ocular manifestations such as Parinaud’s syndrome and neuroretinitis, bacillary angiomatosis, culture-negative endocarditis, bacillary peliosis, fever of undetermined origin, encephalopathy or osteomyelitis (B. henselae) [14,17,18]. Other manifestations, such as malaise, fatigue, insomnia, memory loss, splenomegaly, hepatitis, and meningitis, have also been described, and in some cases, the infection can be fatal [14,15,19–23].
Blood-sucking arthropods are the main mode of Bartonella transmission [6,24,25]. Other types of transmission, such as percutaneous accidents, transplantation of solid organs (kidney and liver), and vertical transmission, have been linked with infections by Bartonella spp. [22,26–29]. Transmission by blood transfusion is also possible. In a case reported by Pons et al., transmission of B. bacilliformis was described after platelet transfusion from an asymptomatic blood donor [30]. A study conducted at UNICAMP showed that B. henselae remained viable for 35 days in experimentally infected blood bags stored at 4°C [31]. This study was used by the American Association of Blood Banks (AABB) to include the bacterium as a pathogen that can be transmitted by blood transfusion. In another study, Ruiz et al. demonstrated that B. bacilliformis remained viable in samples collected from patients with symptoms of Carrión disease after 30 months of storage at 4°C [32]. A study by Silva et al. detected the DNA of B. henselae in the spleen of mice that received transfusion of blood from animals that had been experimentally infected, although molecular blood tests were negative in all transfused animals [33].
Asymptomatic blood donors infected by Bartonella spp. may carry this bacterium in his red blood cells, posing a real risk of infection to blood recipients. However, laboratory confirmation of Bartonella sp. infection remains a major challenge. The fastidious nature of this genus, even in a specific culture medium, limits the diagnostic use of blood or tissue cultures [4,8,14,34]. Molecular methods have expanded the detection of Bartonella sp.; however, no current diagnostic method is able to confirm infection by Bartonella spp. in all infected patients, since these bacteria have low bacteremia, which makes detection even more difficult [35]. In addition, different molecular methods may yield distinct results. A study conducted with cats observed that 27.7% (31/112) of the animals had B. henselae-DNA detected in conventional PCR tests performed in liquid cultures. The same material was examined with nested PCR, and 45.5% (51/112) of all cats had DNA of B. henselae detected. DNA extracted directly from blood was also tested by nested PCR, and B. henselae DNA was detected in 76.8% (86/112). If we consider all tests performed with blood and liquid culture samples, 90.2% (101/112) of cats had B. henselae-DNA detected [36]. Therefore, a combination of several PCRs from different regions and from different samples increases the chances of detecting the pathogen.
A published study investigated the prevalence of Bartonella species in a population of blood donors using just one conventional blood liquid culture PCR. Blood samples from 500 voluntary blood donors were incubated in BAPGM (Bartonella alpha-proteobacteria growth medium) liquid medium and cultured at 37°C in 5% CO2 for 14 days. Then, the samples were subinoculated on agar medium containing 30% sheep blood for another 42 days. DNA was extracted from the liquid culture and tested by Bartonella sp.-specific conventional PCR, which amplifies the ITS region. The amplified products were sequenced to identify the species. Gram-negative isolates obtained from solid culture were also tested by the same technique. Sixteen blood donors (3.2%) were positive for Bartonella spp. in PCR after culture in liquid and solid media. DNA sequencing confirmed the homology of 15 samples with B. henselae and one sample with Bartonella clarridgeiae [37,38].
The present study aimed to evaluate the presence of Bartonella-DNA in blood and liquid culture of 500 blood donor samples from the University of Campinas (UNICAMP) Blood Bank, Campinas, Sao Paulo, Brazil, using four different PCRs in whole blood and in liquid culture samples and compare the results obtained in the previous project and already published [37], since the samples were the same.
Methods
Ethics statement
This project was submitted to the University of Campinas Institutional Review Board (IRB) under n°122/2005, and formal written consent was obtained from donors who agreed to participate in the research. In 2015, the IRB reapproved it under n°1.135.941 for further tests.
Samples
This study analyzed two samples (whole blood and liquid culture) of each 500 blood donors from the University of Campinas (UNICAMP) Blood Bank, randomly collected from November 2009 to January 2010 during a blood donation procedure. These were the same samples used in a previous study that has already been published [37]. Epidemiological data about blood donors, such as gender; occupational animal exposure; contact with cats, dogs, other companion animals, bites from dogs, cats, and other animals; arthropod bites caused by ticks, fleas, or other insects; previous blood transfusion, etc., were already analyzed and published [39].
Whole blood samples collected in tubes with EDTA and liquid culture (another aliquot of whole blood incubated in BAPGM liquid medium and cultured at 37°C in 5% CO2 for 14 days) were stored at -20°C.
DNA extraction
DNA extraction from whole blood and from liquid culture was performed using a QIAamp DNA Mini Kit (Qiagen). Controls were added to each extraction following the protocol already described [37].
PCR
Controls were used in each reaction, and molecular techniques were performed carefully to avoid contamination following the procedures described in previous work [37].
All samples (both DNA extracted from the whole blood and liquid culture) were tested for all PCR techniques described below. Promega enzyme (GoTaq Flexi) was used in all reactions, except for qualitative real-time PCR. The PCR primers and conditions are described in Table A in S1 Appendix.
Quality control PCR
The quality of the extracted DNA and the absence of PCR inhibitors in DNA samples were tested by the amplification of a fragment of the GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene, which encodes a glycolysis enzyme expressed by all mammalian cells [40].
Conventional PCR
Two different reactions for the target gene were performed: the ITS region, or the 16S-23S rRNA intergenic region, for Bartonella spp. [41]; and the gltA gene, or the citrate synthase gene, for B. henselae [42].
Nested PCR
A species-specific nested PCR was used in this study for the target region that encodes protein FtsZ that plays a role in cell division of B. henselae [43].
All PCR products were submitted to electrophoresis in 1.5% agarose gel stained with GelRed and visualized in a photodocumenter with UV light.
Qualitative real-time PCR
The samples were tested by real-time PCR using the same primers used in conventional PCR for amplification of the citrate synthase gene (gltA) in the Sybr Green system using enzyme Fast SYBR Green Master Mix (ThermoFisher Scientific) [42]. In this study, real-time PCR results were used as qualitative PCR, considering results as positive or negative. In addition to the amplification curve, the melt curve was analyzed in comparison with the curve of the diluted Bartonella DNA used as a positive control. We considered positive samples with melt curves between 73.08°C and 73.41°C. We also performed electrophoresis, and the positive samples were confirmed by the presence of a band in the 1.5% agarose gel stained with GelRed.
Sequencing
Amplified DNA with enough concentration was sent for Sanger sequencing. The results were analyzed using Chromas 2.6.6 software and compared to the GenBank database using the BLAST tool from the National Center for Biotechnology Information (NCBI).
Statistical analysis
The McNemar-Bowker test was performed to decide which type of sample (blood or liquid culture) would be the most appropriate considering the results of any PCR method. This test is an extension of the McNemar test when there are nominal variables with more than two categories of nonindependent variables.
Bayesian latent class model (LCM) analysis was performed to find the best species-specific PCR method for B. henselae, regardless of the tested sample, and the best PCR method for blood or liquid culture samples. This test assumes that none of the tests is perfectly accurate; then, an ‘imperfect gold standard’ model is defined according to the results of multiple diagnostic tests with the same samples. The Bayesian approach can infer the prevalence of the studied agent and the test properties by adjusting the possibility of conditional dependence between the tests [44]. The limiting factor of Bayesian LCM analysis refers to the need for intense computer programming. In our study, an online tool was used (Modeling of Infectious Disease Centre–Imperfect Gold Standard Model) with results from three different PCR methods analyzing the same samples. The number of repetitions of the analysis was set at 25,000. Accuracy measures such as sensitivity, specificity, positive predictive value, and negative predictive value [45] were calculated using this method, as well as the 95% confidence interval. All p values were considered statistically significant if p < 0.05.
Additionally, the receiver operating characteristic (ROC) curve method was applied. This analysis uses a simple graphic method to study the variation in sensitivity and specificity for different cutoff values. The area under the ROC curve (AUC) is associated with the discriminatory power of a diagnostic test [46].
Results
PCR
All participants had whole blood and liquid culture analyzed, and all extracted samples had amplified in the quality control PCR (GAPDH), demonstrating the presence of DNA and the absence of PCR inhibitors.
For a better understanding, the results are presented in Venn diagrams, showing how many samples were amplified in each reaction and detected in more than one PCR (Fig 1).
A: Bartonella sp.-DNA was amplified in 30/500 (6%) blood samples. B: Bartonella sp.-DNA was detected in 77/500 (15.4%) liquid cultures. C: Bartonella sp.-DNA was detected in 102/500 (20.4%) from blood and liquid culture.
Fig 1A shows the results of the PCR tests performed with the DNA extracted from whole blood samples. Nested and real-time PCR were the most sensitive tests for this type of sample, and only one sample had simultaneous detection in all four tests. Qualitative real-time PCR was the most efficient test for DNA extracted from liquid culture, showing amplification of 60 samples. Only four samples had simultaneous detection in all four PCR tests (Fig 1B). Fig 1C shows the results of all PCR tests performed with the DNA extracted from blood and liquid culture samples. The most efficient reaction to detect Bartonella spp. was real-time PCR with amplification of 72 samples, 45 of which were only in this test. Note that Fig 1C is not a product of the sum of the two previous Venn graphs. An example is that a sample positive only in ITS in blood was also positive when its liquid culture was tested in real-time PCR. This condition occurred in other samples as well.
To test the detection limit of each PCR, tests with known Bartonella-DNA concentrations were performed as previously described [36]. Although one genome equivalent (GE) of B. henselae was amplified in at least one of several reactions performed in each PCR technique (analytical sensitivity), the detection limit of each test, which refers to the minimum GE that amplified in all reactions (diagnostic sensitivity), was 50 GE in conventional PCR, 20 GE in real-time PCR and 10 GE in nested PCR. These data allow us to hypothetically calculate the amount of GE required in the initial sample for detection by the methods described in this study, considering 1) the detection limit of each reaction, 2) the amount of initial sample used in each diagnosis stage (extraction: 1 mL of liquid culture and 200 μL of blood/PCR: 5 μL in conventional and real-time PCRs and 2.5 μL in nested PCR), and 3) the dilution effect (in case of liquid culture) (Table 1).
Sequencing
Seventy-two amplicons presented sufficient quality for Sanger sequencing, including the 11 ITS amplicons. It was possible to sequence samples from 66 of the 102 positive donors. In six donors, more than one region could be sequenced (Table 2). All samples presented 100% similarity to B. henselae (Access code in GenBank database for each region: gltA (KT945243.1- Bartonella henselae strain BR_LHR human 346 citrate synthase gene, partial cds), ftsZ (HG965802.1—Bartonella henselae strain BM1374163 complete genome, and ITS (BX897699.1-Bartonella henselae strain Houston1 complete genome).
Statistical analysis
The McNemar-Bowker test was performed to determine the best sample (blood or liquid culture) to be used regardless of the PCR technique. The result of ‘detectable in any PCR’ was considered the gold standard. The data show that the reactions conducted in liquid culture were significantly more effective in detecting DNA of Bartonella spp. than those performed in blood (p < 0.0001). Liquid culture was also better in relation to the negative predictive value and the negative odds ratio (Fig 2 and Table B in S1 Appendix).
Comparison between results obtained analyzing blood versus liquid culture samples (considering ‘detectable in any PCR’ as the gold standard) by McNemar-Bowker test. PPV: positive predictive value; NPV: negative predictive value.
Bayesian LCM analysis was performed to determine the best PCR technique for blood or liquid culture samples and then to combine the results of blood and liquid culture, all of them using ‘detectable in any PCR’ as the gold standard (Fig 3). The results of conventional PCR for the ITS region were not used, as it was the only reaction targeting the Bartonella genus; all others targeted B. henselae.
A: Blood samples; B: Liquid culture; C: Combining blood and liquid culture. PPV: positive predictive value; NPV: negative predictive value.
Using the McNemar-Bowker test to analyze the results obtained only from whole blood samples, there was no difference in the concordance of results between nested PCR (ftsZ) and real-time PCR (gltA) (p value = 0.6547; McNemar-Bowker test, p > 0.05), demonstrating that they have similar diagnostic power in this type of sample (Fig 3A and Tables C and D in S1 Appendix). This result was confirmed by the ROC curve method, which showed very close AUC values for these PCRs (Fig 4A).
A: from blood samples; B: from liquid culture samples. AUC: Area under the curve. In the plot, the yellow 45-degree line marks the cutoff point that maximizes the sum of sensitivity and specificity.
No agreement was found between the PCR results from liquid culture according to the McNemar-Bowker test, suggesting that sensitivity is different between them (Fig 3B and Tables E and F in S1 Appendix). ROC curve method analyses showed that the AUC of real-time PCR was greater than the AUC of other PCR tests, i.e., real-time PCR had greater discriminatory diagnostic power for liquid culture samples (Fig 4B).
Fig 3C shows the Bayesian LCM test combining liquid culture and blood sample PCR results, and together with the McNemar-Bowker test, it shows that the PCRs are different between them in this case (Tables G, H and I in S1 Appendix).
Discussion
B. henselae was detected in 20.4% of donors (102/500) in our project, and comparing these data with the scientific literature, the prevalence found in our study is high. A survey about publications of prevalence of Bartonella sp. in blood donors published from 2009 to 2022 (Table J in S1 Appendix) have showed that only five from 20 studies used PCR as diagnostic method [37,47–65]. Two of these five were developed in Brazil, with prevalence from 1.8% to 3.2%, and in both only one screening PCR was performed with each sample [37,52].
According to our literature review, few studies have been conducted with humans using several combined molecular tests and many samples. We found only ten studies published during the same period (from 2009 to 2022) about Bartonella sp. prevalence using PCR that analyzed at least 30 individuals, excluding blood donors [66–74] (Table K in S1 Appendix). Few of them have used PCR for different regions. Therefore, data found in the literature are not sufficient to determine the exact prevalence of bacteremia caused by Bartonella spp. [75].
A study conducted in Spain by Portillo et al. [72] analyzed samples from 97 sanitary workers using several techniques, such as serology for different species, direct blood extraction, liquid and solid cultures, and different molecular tests of these cultures, such as conventional and real-time PCR, with different primers for the ITS region, in addition to sequencing. With this combination of diagnostic tests, they obtained 83.1% positive results in IFA tests, and the DNA of Bartonella spp. was amplified by 21.6%. This percentage of molecular detection in asymptomatic individuals is very similar to our data.
The present study again analyzed samples from 500 donors previously studied with just one PCR from liquid culture. The results obtained now from whole blood and liquid culture using different PCR methodologies were compared and showed that detection was lower in reactions that used DNA directly extracted from whole blood (6% or 30/500) compared to liquid culture (15.4% or 77/500). In the McNemar-Bowker test, liquid culture showed better sensitivity and negative predictive value when compared to whole blood, which can be explained by the fact that a diagnosis obtained directly from blood is less effective than after enrichment culture [76]. The samples are from asymptomatic blood donors, so they must have low-level bacteremia, i.e., they are below the detection level of diagnostic sensitivity. After enrichment culture, bacterial multiplication may occur to a number above this detection limit. In immunocompetent humans, bacteremia caused by Bartonella spp. is estimated to be 1–10 GE/μL (i.e., 1,000–10,000 GE/mL) [15], which may lead to the real possibility of false negatives. Liquid culture of Bartonella spp. increases the sensitivity of detection of infection caused by these bacteria by molecular methods [76–78].
Twenty-five samples had DNA from Bartonella spp. detected from whole blood but were undetected when tested in liquid culture. This fact can be explained by the ‘dilution effect’ [79]. There was no increase in the number of bacteria in the culture, but dilution occurred in a large volume of culture medium, so the concentration of bacteria in the liquid culture was below the detection level, indicating amplification of nonviable bacteria. The fastidious characteristic of the bacterium combined with the presence of growth inhibitors (such as the use of antimicrobials, even if occasionally used as anti-inflammatories, as seen with sulfone, hydroxychloroquine, etc.) make this hypothesis even more probable [35]. Currently, several techniques must be used concomitantly to avoid false negative results [66].
As with other diagnostic methods, there is no consensus on the best primers and conditions for PCR to detect the DNA of Bartonella spp.. Several studies describe the 16S rRNA gene region, the 16S-23S rRNA intergenic locus (ITS), the citrate synthase gene, or the riboflavin synthase gene, the groEL gene, the ftsZ gene, the gltA gene, or the RNA polymerase beta subunit as the most efficient and promising for the detection and differentiation of the various species of Bartonella [80,81]. In addition to the primers that determine the region to be amplified and therefore the sensitivity of the reaction, the PCR technique also influences the success of the diagnosis. Nested PCR and real-time PCR can greatly increase detection sensitivity [78,82–84]. In this study, the results obtained with statistical analysis show that conventional PCR in the gltA region is the least efficient technique to detect Bartonella sp.-DNA. Real-time PCR (gltA) is the best test for liquid culture, while for blood samples, no difference in sensitivity was found between nested (ftsZ) and real-time PCR (gltA) since confidence intervals for sensitivity overlap.
The great advantage of molecular diagnostic methods such as PCR is the fast result when compared to culture in addition to possible identification of the species causing the infection [85]. More sensitive and specific PCR tests allow quick diagnosis of the infection, even with low-level bacteremia. Maggi et al. developed, optimized, and validated droplet digital PCR (ddPCR), a new molecular technology, for the detection of Bartonella spp. DNA within several sample matrices. The ddPCR sensitivity (53/112) was significantly better than that of qPCR (6/112) when testing patient blood and enrichment blood culture samples [86]. Despite these advantages, PCR has some limitations: the possibility of false positive results due to contamination by control DNA or previously positive samples and false negative results for having less DNA than the detection limit. In addition, finding the pathogen DNA in the sample does not accurately indicate an active infection [79,87].
Several case reports [20,34,88–90] and a previous study with blood samples from cats [36] show that a combination of PCR and different samples increases the chances of detecting the pathogen. The results of this study agree with the literature and reinforce the need to combine several diagnostic tests to avoid false negatives.
In the previous project conducted with the same samples, the DNA of Bartonella spp. was detected in 16 donors [37]. Of these, the DNA was again amplified in only three samples, and only one that had been previously isolated by culture was detected again in real-time PCR. (Table L in S1 Appendix). This divergence in the results from the two studies developed with the samples of 500 blood donors shows how challenging the laboratory diagnosis of Bartonella spp. infection can be, since the samples from which isolates were previously obtained were supposed to have positive reactions, which did not happen in most samples. In the first study with these samples, only one conventional PCR genus-specific reaction was performed for the ITS region using DNA extracted from the liquid culture. Several factors can explain this divergence, including low-level bacteremia (1–10 GE/μL), especially for asymptomatic individuals. Therefore, there must be a small amount of pathogen DNA close to the detection limit. The aliquot used in the reaction may not have the amount required for amplification (Fig A in S1 Appendix).
Additionally, a stochastic (random) variation of the PCR amplification process occurs in the analysis of low amounts of DNA. Stochastic effects are seen as a fluctuation of results between replicated analyses [79]. For this reason, even a combination of several techniques does not prevent false negative results. In a previous study that used the same samples, the DNA of Bartonella spp. was detected in only 3.2% (16/500) of liquid culture samples using conventional PCR, and six of them were isolated in solid culture. In five of these six isolates, we were unable to detect Bartonella sp.-DNA [37].
Edouard et al. argue that to confirm a diagnosis of bartonelloses using exclusively the PCR technique, only samples with the DNA of Bartonella spp. detected in at least two different genome regions [68] should be considered. In this case, the possibility of false positives is reduced, and consequently, the specificity increases, but sensitivity is lost. Even considering this criterion, 35 (7%) samples were positive in our study with reactions in two different genome regions. If the six samples from which isolates were obtained in the previous study were added, we obtained 41 B. henselae DNA-detected samples (8.2%). None of the samples that originated these six isolates met the criterion of two distinct detected regions even in this current study using different PCRs.
As bartonelloses are caused by fastidious bacteria and low-level bacteremia is characteristic of the infection, it would not be advisable to use the criterion of two different gene region detections to confirm the diagnosis. Then, when adding up all the samples with Bartonella sp.-DNA detection in the two studies, 115/500 donors had Bartonella spp. detected, which corresponds to 23%. This result is close to the percentage of bloodstream infection in sanitary workers found in the study by Portillo et al. [72].
Cases reported in the literature [34,91,92] show that the low sensitivity of molecular tests may impact clinical practice. The five most common manifestations related to Bartonella spp. (CSD, bacillary angiomatosis, bacillary peliosis, culture-negative endocarditis and fever of undetermined origin) [93,94] are unquestionable and justify more investments in studies of this kind.
The laboratory diagnosis of Bartonella spp. is a challenge for several reasons: first, the fastidious characteristic of the bacterium, which makes laboratory culture an obstacle; second, the fact that it causes cyclic and low bacteremia; and finally, the lack of specific and sensitive tests for its diagnosis [15,95]. Combining methods is required to reduce false negatives. Further efforts should be dedicated to improving the diagnostic methods and ensuring better sensitivity to screen for infection by Bartonella spp..
The statistical analysis using all results of blood and liquid culture samples showed that, regardless of the sample, the sensitivity differs with the PCR types (conventional, nested and real-time PCR) and targets of gene regions.
The use of three different PCR tests with two types of samples (blood and liquid culture) increased the possibility of detecting Bartonella spp. considering that, in a previous project, 3.2% of blood donors were positive, and in this project, this rate increased to 20.4%. However, the combination of techniques did not prevent false-negative results since 13 donors who were positive in the previous project were not positive again.
Conclusions
More than one-fifth of blood donors had at least one B. henselae DNA detected by a PCR test among the eight molecular reactions performed. Seven percent had the DNA detected for two or more distinct regions.
The statistical analysis using all results of blood and liquid culture samples showed that, regardless of the sample, the sensitivity differs with the PCR types (conventional, nested and real-time PCR) and targets of gene regions.
The results of our study indicate that public health authorities must review the risks and the impact of the transmission of Bartonella spp. through blood transfusions, especially for immunocompromised patients. Low-level bacteremia and the fastidious characteristics of the bacterium are challenges to laboratory diagnosis.
Supporting information
S1 Appendix. Table A: Primers used in this study.
Table B. Statistical analysis (McNemar-Bowker test) of results from blood DNA samples versus liquid culture DNA samples (considering ‘detectable in any PCR’ as the gold standard). Table C. Bayesian latent class model (LCM) statistical analysis comparing the results of three distinct PCR tests for B. henselae from blood samples. Table D. Comparison of the agreement of results between each PCR technique performed with blood samples using the McNemar-Bowker test (p < 0.05). Table E. Bayesian latent class model (LCM) statistical analysis comparing the results of three distinct PCR tests for B. henselae from liquid culture samples. Table F. Comparison of the agreement of results between each PCR technique performed with liquid culture samples using the McNemar-Bowker test (p < 0.05). Table G. Bayesian latent class model (LCM) analysis comparing the results of three different PCRs for B. henselae performed with combined blood and liquid culture samples. Table H. Comparison of agreement of results between each PCR technique, regardless of liquid culture or blood sample, using the McNemar-Bowker test. Table I. Bayesian latent class model (LCM) analysis comparing the results of three different PCR tests for B. henselae regardless of liquid culture or blood sample, considering ‘detectable in any PCR’ as the gold standard. Table J. The Google Scholar, PubMed and Scopus databases were searched from 2009 to June 2022 for articles with ‘blood donors’ and ‘Bartonella’ in the title or abstract. NT: Not tested. Table K. The Google Scholar, PubMed and Scopus databases were searched from 2009 (year of sample collection for our study) to June 2022 for articles with ‘PCR’ and ‘Bartonella’ in the title or abstract (excluding studies that were not conducted with humans and that analyzed fewer than 30 individuals). Table L. Positive samples of PCR from a previous study with blood donors and results obtained in the current study. Fig A. Graphical representation of a variation in the amount of DNA in multiple PCRs.
https://doi.org/10.1371/journal.pntd.0011336.s001
(DOCX)
Acknowledgments
We would like to thank professor Carlos Henrique Inácio Ramos for making available the qPCR equipment (FAPESP 2011/50515-1).
References
- 1. Rolain JM, Brouqui P, Koehler JE, Maguina C, Dolan MJ, Raoult D. Recommendations for treatment of human infections caused by Bartonella species. Antimicrobial agents and chemotherapy. 2004;48(6):1921–33. Epub 2004/05/25. pmid:15155180; PubMed Central PMCID: PMC415619.
- 2. Lynch T, Iverson J, Kosoy M. Combining culture techniques for Bartonella: the best of both worlds. Journal of clinical microbiology. 2011;49(4):1363–8. Epub 2011/02/04. pmid:21289156; PubMed Central PMCID: PMC3122786.
- 3. Maggi RG, Harms CA, Hohn AA, Pabst DA, McLellan WA, Walton WJ, et al. Bartonella henselae in porpoise blood. Emerging infectious diseases. 2005;11(12):1894–8. Epub 2006/02/21. pmid:16485476.
- 4. Pulliainen AT, Dehio C. Persistence of Bartonella spp. stealth pathogens: from subclinical infections to vasoproliferative tumor formation. FEMS microbiology reviews. 2012. Epub 2012/01/11. pmid:22229763.
- 5. Okaro U, Addisu A, Casanas B, Anderson B. Bartonella Species, an Emerging Cause of Blood-Culture-Negative Endocarditis. Clin Microbiol Rev. 2017;30(3):709–46. Epub 2017/05/12. pmid:28490579; PubMed Central PMCID: PMC5475225.
- 6. Billeter SA, Levy MG, Chomel BB, Breitschwerdt EB. Vector transmission of Bartonella species with emphasis on the potential for tick transmission. Medical and veterinary entomology. 2008;22(1):1–15. Epub 2008/04/03. pmid:18380649.
- 7. Breitschwerdt EB. Bartonellosis, One Health and all creatures great and small. Vet Dermatol. 2017;28(1):96–e21. pmid:28133871.
- 8. Abbott RC, Chomel BB, Kasten RW, Floyd-Hawkins KA, Kikuchi Y, Koehler JE, et al. Experimental and natural infection with Bartonella henselae in domestic cats. Comp Immunol Microbiol Infect Dis. 1997;20(1):41–51. pmid:9023040.
- 9. Schultz MG. A history of bartonellosis (Carrion’s disease). The American journal of tropical medicine and hygiene. 1968;17(4):503–15. Epub 1968/07/01. pmid:4876803.
- 10. Drancourt M, Tran-Hung L, Courtin J, Lumley H, Raoult D. Bartonella quintana in a 4000-year-old human tooth. J Infect Dis. 2005;191(4):607–11. pmid:15655785.
- 11. Dehio C. Molecular and cellular basis of bartonella pathogenesis. Annual review of microbiology. 2004;58:365–90. Epub 2004/10/19. pmid:15487942.
- 12. Kaiser PO, Riess T, O’Rourke F, Linke D, Kempf VA. Bartonella spp.: throwing light on uncommon human infections. International journal of medical microbiology: IJMM. 2011;301(1):7–15. Epub 2010/09/14. pmid:20833105.
- 13. Anderson BE, Neuman MA. Bartonella spp. as emerging human pathogens. Clinical microbiology reviews. 1997;10(2):203–19. Epub 1997/04/01. pmid:9105751; PubMed Central PMCID: PMC172916.
- 14. Harms A, Dehio C. Intruders below the radar: molecular pathogenesis of Bartonella spp. Clinical microbiology reviews. 2012;25(1):42–78. Epub 2012/01/11. pmid:22232371; PubMed Central PMCID: PMC3255967.
- 15. Breitschwerdt EB, Maggi RG, Chomel BB, Lappin MR. Bartonellosis: an emerging infectious disease of zoonotic importance to animals and human beings. Journal of veterinary emergency and critical care. 2010;20(1):8–30. Epub 2010/03/17. pmid:20230432.
- 16. Velho PENF, Souza EM, Cintra ML, Mariotto A, Moraes AM. Angiomatose bacilar: revisão da literatura e documentação iconográfica. Anais Brasileiros de Dermatologia. 2003;78(5):601–9.
- 17. Velho PENF, Souza EM, Cintra ML, Moraes AM, Uthida-Tanaka AM. Diagnosis of Bartonella spp. infection: study of a bacillary angiomatosis case. An Bras Dermatol. 2006;81(4):349–53.
- 18. Dehio C. Bartonella-host-cell interactions and vascular tumor formation. Nat Rev Microbiol. 2005;3(8):621–31. pmid:16064054.
- 19. Sanogo YO, Zeaiter Z, Caruso G, Merola F, Shpynov S, Brouqui P, et al. Bartonella henselae in Ixodes ricinus ticks (Acari: Ixodida) removed from humans, Belluno province, Italy. Emerging infectious diseases. 2003;9(3):329–32. Epub 2003/03/20. pmid:12643827; PubMed Central PMCID: PMC2958531.
- 20. Breitschwerdt EB, Maggi RG, Cadenas MB, Diniz PP. A Groundhog, a Novel Bartonella Sequence, and My Father’s Death. Emerg Infect Dis. 2009;15(12):2080–6. pmid:19998749.
- 21. Chomel BB, Boulouis HJ, Maruyama S, Breitschwerdt EB. Bartonella spp. in pets and effect on human health. Emerg Infect Dis. 2006;12(3):389–94. pmid:16704774; PubMed Central PMCID: PMC3291446.
- 22. Mosbacher ME, Klotz S, Klotz J, Pinnas JL. Bartonella henselae and the potential for arthropod vector-borne transmission. Vector borne and zoonotic diseases. 2011;11(5):471–7. Epub 2010/10/27. pmid:20973657.
- 23. Lins KA, Drummond MR, Velho PENF. Cutaneous Manifestations of Bartonellosis. Anais brasileiros de dermatologia. 2019;94(5). pmid:31780437.
- 24. Reis C, Cote M, Le Rhun D, Lecuelle B, Levin ML, Vayssier-Taussat M, et al. Vector competence of the tick Ixodes ricinus for transmission of Bartonella birtlesii. PLoS neglected tropical diseases. 2011;5(5):e1186. Epub 2011/06/10. pmid:21655306; PubMed Central PMCID: PMC3104967.
- 25. Tsai YL, Chang CC, Chuang ST, Chomel BB. Bartonella species and their ectoparasites: selective host adaptation or strain selection between the vector and the mammalian host? Comp Immunol Microbiol Infect Dis. 2011;34(4):299–314. pmid:21616536.
- 26. Scolfaro C, Mignone F, Gennari F, Alfarano A, Veltri A, Romagnoli R, et al. Possible donor-recipient bartonellosis transmission in a pediatric liver transplant. Transplant infectious disease: an official journal of the Transplantation Society. 2008;10(6):431–3. Epub 2008/07/25. pmid:18651873.
- 27. Velho PE. Blood transfusion as an alternative bartonellosis transmission in a pediatric liver transplant. Transplant infectious disease: an official journal of the Transplantation Society. 2009;11(5):474. Epub 2009/10/07. pmid:19804481.
- 28. Breitschwerdt EB, Maggi RG, Farmer P, Mascarelli PE. Molecular evidence of perinatal transmission of Bartonella vinsonii subsp. berkhoffii and Bartonella henselae to a child. Journal of clinical microbiology. 2010;48(6):2289–93. Epub 2010/04/16. pmid:20392912; PubMed Central PMCID: PMC2884525.
- 29. Psarros G, Riddell J, Gandhi T, Kauffman CA, Cinti SK. Bartonella henselae infections in solid organ transplant recipients: report of 5 cases and review of the literature. Medicine (Baltimore). 2012;91(2):111–21. pmid:22391473.
- 30. Pons MJ, Lovato P, Silva J, Urteaga N, del Valle Mendoza J, Ruiz J. Carrion’s disease after blood transfusion. Blood Transfus. 2016;14(6):527–30. pmid:26674821.
- 31. Magalhães RF, Pitassi LH, Salvadego M, de Moraes AM, Barjas-Castro ML, Velho PE. Bartonella henselae survives after the storage period of red blood cell units: is it transmissible by transfusion? Transfus Med. 2008;18(5):287–91. pmid:18937735.
- 32. Ruiz J, Silva W, Pons MJ, Del Valle LJ, Tinco CR, Casabona VD, et al. Long time survival of Bartonella bacilliformis in blood stored at 4°C. A risk for blood transfusions. Blood Transfus. 2012;10(4):563–4. pmid:22507863; PubMed Central PMCID: PMC3496215.
- 33. Silva MN, Vieira-Damiani G, Ericson ME, Gupta K, Gilioli R, de Almeida AR, et al. Bartonella henselae transmission by blood transfusion in mice. Transfusion. 2016;56(6 Pt 2):1556–9. Epub 2016/03/13. pmid:26968530; PubMed Central PMCID: PMC4905798.
- 34. Drummond MR, Dos Santos LS, Silva MND, Almeida AR, Diniz P, Angerami R, et al. False Negative Results in Bartonellosis Diagnosis. Vector Borne Zoonotic Dis. 2019;19(6):453–4. Epub 2019/02/08. pmid:30730266.
- 35. Breitschwerdt EB. Bartonellosis: one health perspectives for an emerging infectious disease. Ilar j. 2014;55(1):46–58. Epub 2014/06/18. pmid:24936029.
- 36. Drummond MR, Lania BG, Diniz PPVdP, Gilioli R, Demolin DMR, Scorpio DG, et al. Improvement of Bartonella henselae DNA Detection in Cat Blood Samples by Combining Molecular and Culture Methods. Journal of Clinical Microbiology. 2018. pmid:29540455
- 37. Pitassi LH, de Paiva Diniz PP, Scorpio DG, Drummond MR, Lania BG, Barjas-Castro ML, et al. Bartonella spp. bacteremia in blood donors from Campinas, Brazil. PLoS Negl Trop Dis. 2015;9(1):e0003467. Epub 2015/01/16. pmid:25590435; PubMed Central PMCID: PMC4295888.
- 38. Vieira-Damiani G, Diniz PP, Pitassi LH, Sowy S, Scorpio DG, Lania BG, et al. Bartonella clarridgeiae bacteremia detected in an asymptomatic blood donor. J Clin Microbiol. 2015;53(1):352–6. pmid:25392353; PubMed Central PMCID: PMC4290971.
- 39. Diniz P, Velho P, Pitassi L, Drummond M, Lania B, Barjas-Castro M, et al. Risk Factors for Bartonella species Infection in Blood Donors from Southeast Brazil. Plos Neglected Tropical Diseases. 2016;10(3). WOS:000373272500041. pmid:26999057
- 40. Birkenheuer AJ, Levy MG, Breitschwerdt EB. Development and evaluation of a seminested PCR for detection and differentiation of Babesia gibsoni (Asian genotype) and B. canis DNA in canine blood samples. J Clin Microbiol. 2003;41(9):4172–7. Epub 2003/09/06. pmid:12958243; PubMed Central PMCID: PMC193857.
- 41. Diniz PP, Maggi RG, Schwartz DS, Cadenas MB, Bradley JM, Hegarty B, et al. Canine bartonellosis: serological and molecular prevalence in Brazil and evidence of coinfection with Bartonella henselae and Bartonella vinsonii subsp. berkhoffii. Vet Res. 2007;38(5):697–710. :2007023. pmid:17583666.
- 42. Staggemeier R, Pilger DA, Spilki FR, Cantarelli VV. MULTIPLEX SYBR GREEN-REAL TIME PCR (qPCR) ASSAY FOR THE DETECTION AND DIFFERENTIATION OF Bartonella henselae AND Bartonella clarridgeiae IN CATS. Rev Inst Med Trop Sao Paulo. 2014;56(2):93–5. pmid:24626408.
- 43. Kawasato KH, de Oliveira LC, Velho PE, Yamamoto L, Del Negro GM, Okay TS. Detection of Bartonella henselae DNA in clinical samples including peripheral blood of immune competent and immune compromised patients by three nested amplifications. Rev Inst Med Trop Sao Paulo. 2013;55(1):1–6. pmid:23328718.
- 44. Lim C, Wannapinij P, White L, Day NP, Cooper BS, Peacock SJ, et al. Using a web-based application to define the accuracy of diagnostic tests when the gold standard is imperfect. PLoS One. 2013;8(11):e79489. Epub 2013/11/23. pmid:24265775; PubMed Central PMCID: PMC3827152.
- 45. Cohen JF, Korevaar DA, Altman DG, Bruns DE, Gatsonis CA, Hooft L, et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open. 2016;6(11):e012799. Epub 2017/02/01. pmid:28137831; PubMed Central PMCID: PMC5128957.
- 46. Hajian-Tilaki K. Receiver Operating Characteristic (ROC) Curve Analysis for Medical Diagnostic Test Evaluation. Caspian J Intern Med. 2013;4(2):627–35. pmid:24009950.
- 47. Yilmaz C, Ergin C, Kaleli I. [Investigation of Bartonella henselae seroprevalence and related risk factors in blood donors admitted to Pamukkale University Blood Center]. Mikrobiyol Bul. 2009;43(3):391–401. pmid:19795614.
- 48. Pandak N, Daković-Rode O, Cabraja I, Kristof Z, Kotarac S. Prevalence of Bartonella henselae antibodies in children and blood donors in Croatia. Infection. 2009;37(2):166–7. pmid:19274430.
- 49. Minadakis G, Chochlakis D, Kokkini S, Gikas A, Tselentis Y, Psaroulaki A. Seroprevalence of Bartonella henselae antibodies in blood donors in Crete. Scand J Infect Dis. 2008;40(10):846–7. pmid:18609206.
- 50. Lamas CC, Mares-Guia MA, Rozental T, Moreira N, Favacho AR, Barreira J, et al. Bartonella spp. infection in HIV positive individuals, their pets and ectoparasites in Rio de Janeiro, Brazil: serological and molecular study. Acta Trop. 2010;115(1–2):137–41. pmid:20206113.
- 51. Sun J, Fu G, Lin J, Song X, Lu L, Liu Q. Seroprevalence of Bartonella in Eastern China and analysis of risk factors. BMC Infect Dis. 2010;10:121. pmid:20482887; PubMed Central PMCID: PMC2886056.
- 52. Corrêa F, Pontes C, Verzola R, Mateos J, Velho P, Schijman A, et al. Association of Bartonella spp bacteremia with Chagas cardiomyopathy, endocarditis and arrhythmias in patients from South America. Braz J Med Biol Res. 2012;45(7):644–51. pmid:22584639; PubMed Central PMCID: PMC3854270.
- 53. Mansueto P, Pepe I, Cillari E, Arcoleo F, Micalizzi A, Bonura F, et al. Prevalence of antibodies anti-Bartonella henselae in western Sicily: children, blood donors, and cats. J Immunoassay Immunochem. 2012;33(1):18–25. pmid:22181817.
- 54. Musso D, Broult J, Parola P, Raoult D, Fournier PE. Absence of antibodies to Rickettsia spp., Bartonella spp., Ehrlichia spp. and Coxiella burnetii in Tahiti, French Polynesia. BMC Infect Dis. 2014;14:255. Epub 2014/06/03. pmid:24885466; PubMed Central PMCID: PMC4022530.
- 55. Noden BH, Tshavuka FI, van der Colf BE, Chipare I, Wilkinson R. Exposure and risk factors to coxiella burnetii, spotted fever group and typhus group Rickettsiae, and Bartonella henselae among volunteer blood donors in Namibia. PLoS One. 2014;9(9):e108674. Epub 2014/09/27. pmid:25259959; PubMed Central PMCID: PMC4178180.
- 56. Aydin N, Bulbul R, Telli M, Gultekin B. Seroprevalence of Bartonella henselae and Bartonella quintana in blood donors in Aydin province, Turkey. Mikrobiyol Bul. 2014;48(3):477–83. Epub 2014/07/24. pmid:25052114.
- 57. Muller A, Reiter M, Schotta AM, Stockinger H, Stanek G. Detection of Bartonella spp. in Ixodes ricinus ticks and Bartonella seroprevalence in human populations. Ticks Tick Borne Dis. 2016;7(5):763–7. Epub 2016/03/22. pmid:26997137.
- 58. Pons MJ, Urteaga N, Alva-Urcia C, Lovato P, Silva J, Ruiz J, et al. Infectious agents, Leptospira spp. and Bartonella spp., in blood donors from Cajamarca, Peru. Blood Transfus. 2015;14(6):504–8. Epub 2015/12/18. pmid:26674831; PubMed Central PMCID: PMC5111371.
- 59. Kwon HY, Im JH, Lee SM, Baek JH, Durey A, Park SG, et al. The seroprevalence of Bartonella henselae in healthy adults in Korea. Korean J Intern Med. 2017;32(3):530–5. pmid:28490714.
- 60. Núñez M, Contreras K, Depix M, Geoffroy E, Villagra N, Mellado S, et al. Prevalence of Bartonella Henselae in Blood Donors and Risk of Blood Transmission in Chile. Revista chilena de infectologia: organo oficial de la Sociedad Chilena de Infectologia. 2017;34(6). pmid:29488546.
- 61. Brydak-Godowska J, Kopacz D, Borkowski P, Fiecek B, Hevelke A, Rabczenko D, et al. Seroprevalence of Bartonella Species in Patients With Ocular Inflammation. Advances in experimental medicine and biology. 2017;1020. pmid:28405890.
- 62. Łysakowska M, Brzezińska O, Szybka M, Konieczka M, Moskwa S, Brauncajs M, et al. The Seroprevalence of Bartonella Spp. In the Blood of Patients With Musculoskeletal Complaints and Blood Donors, Poland: A Pilot Study. Clinical rheumatology. 2019;38(10). pmid:31115789.
- 63. Pawełczyk A, Bednarska M, Kowalska J, Uszyńska-Kałuża B, Radkowski M, Welc-Falęciak R. Seroprevalence of Six Pathogens Transmitted by the Ixodes Ricinus Ticks in Asymptomatic Individuals With HIV Infection and in Blood Donors. Scientific reports. 2019;9(1). pmid:30765826.
- 64. Dichter AA, Schultze TG, Wenigmann A, Ballhorn W, Latz A, Schlüfter E, et al. Identification of immunodominant Bartonella bacilliformis proteins: a combined in-silico and serology approach. Lancet Microbe. 2021;2(12):e685–e94. Epub 20210910. pmid:35544109.
- 65. Łysakowska ME, Szybka M, Olga B, Moskwa S, Konieczka M, Makowska J, et al. Cytokine and LL-37 gene expression levels in Bartonella spp. seropositive and seronegative patients of a rheumatology clinic. Adv Med Sci. 2022;67(1):163–9. Epub 20220310. pmid:35279619.
- 66. Maggi RG, Mascarelli PE, Pultorak EL, Hegarty BC, Bradley JM, Mozayeni BR, et al. Bartonella spp. bacteremia in high-risk immunocompetent patients. Diagnostic microbiology and infectious disease. 2011;71(4):430–7. Epub 2011/10/15. pmid:21996096.
- 67. Maggi RG, Mozayeni BR, Pultorak EL, Hegarty BC, Bradley JM, Correa M, et al. Bartonella spp. bacteremia and rheumatic symptoms in patients from Lyme disease-endemic region. Emerg Infect Dis. 2012;18(5):783–91. pmid:22516098; PubMed Central PMCID: PMC3358077.
- 68. Edouard S, Nabet C, Lepidi H, Fournier PE, Raoult D. Bartonella, a common cause of endocarditis: a report on 106 cases and review. J Clin Microbiol. 2015;53(3):824–9. pmid:25540398; PubMed Central PMCID: PMC4390654.
- 69. Oteo JA, Maggi R, Portillo A, Bradley J, García-Álvarez L, San-Martín M, et al. Prevalence of Bartonella spp. by culture, PCR and serology, in veterinary personnel from Spain. Parasit Vectors. 2017;10. pmid:28061911.
- 70. Hobson C, Le Brun C, Beauruelle C, Maakaroun-Vermesse Z, Mereghetti L, Goudeau A, et al. Detection of Bartonella in cat scratch disease using a single-step PCR assay kit. J Med Microbiol. 2017;66(11):1596–601. Epub 2017/10/27. pmid:29068281.
- 71. Chaudhry R, Kokkayil P, Ghosh A, Bahadur T, Kant K, Sagar T, et al. Bartonella henselae Infection in Diverse Clinical Conditions in a Tertiary Care Hospital in North India. The Indian journal of medical research. 2018;147(2). pmid:29806608.
- 72. Portillo A, Maggi R, Oteo JA, Bradley J, García-Álvarez L, San-Martín M, et al. Bartonella Spp. Prevalence (Serology, Culture, and PCR) in Sanitary Workers in La Rioja Spain. Pathogens (Basel, Switzerland). 2020;9(3). pmid:32143533.
- 73. Soares TCB, Isaias GAB, Almeida AR, Drummond MR, da Silva MN, Lania BG, et al. Prevalence of Bartonella spp. Infection in Patients with Sickle Cell Disease. Vector Borne Zoonotic Dis. 2020;20(7):509–12. Epub 20200203. pmid:32013778; PubMed Central PMCID: PMC7336878.
- 74. Šimeková K, Soják Ľ, Víchová B, Balogová L, Jarošová J, Antolová D. Parasitic and Vector-Borne Infections in HIV-Positive Patients in Slovakia-Evidence of an Unexpectedly High Occurrence of Anaplasma phagocytophilum. Pathogens. 2021;10(12). Epub 20211129. pmid:34959511; PubMed Central PMCID: PMC8704717.
- 75. Chomel BB, Boulouis HJ, Breitschwerdt EB. Cat scratch disease and other zoonotic Bartonella infections. Journal of the American Veterinary Medical Association. 2004;224(8):1270–9. Epub 2004/04/29. pmid:15112775.
- 76. Duncan AW, Maggi RG, Breitschwerdt EB. A combined approach for the enhanced detection and isolation of Bartonella species in dog blood samples: preenrichment liquid culture followed by PCR and subculture onto agar plates. J Microbiol Methods. 2007;69(2):273–81. pmid:17346836.
- 77. Maggi RG, Duncan AW, Breitschwerdt EB. Novel chemically modified liquid medium that will support the growth of seven bartonella species. Journal of clinical microbiology. 2005;43(6):2651–5. Epub 2005/06/16. pmid:15956379; PubMed Central PMCID: PMC1151927.
- 78. Diaz MH, Bai Y, Malania L, Winchell JM, Kosoy MY. Development of a Novel Genus-specific Real-time PCR Assay for Detection and Differentiation of Bartonella Species and Genotypes. Journal of clinical microbiology. 2012. Epub 2012/03/02. pmid:22378904.
- 79. Wilson IG. Inhibition and facilitation of nucleic acid amplification. Appl Environ Microbiol. 1997;63(10):3741–51. Epub 1997/11/05. pmid:9327537; PubMed Central PMCID: PMC168683.
- 80. Maggi RG, Breitschwerdt EB. Potential limitations of the 16S-23S rRNA intergenic region for molecular detection of Bartonella species. Journal of clinical microbiology. 2005;43(3):1171–6. Epub 2005/03/08. pmid:15750079; PubMed Central PMCID: PMC1081238.
- 81. Kosoy M, Bai Y, Lynch T, Kuzmin IV, Niezgoda M, Franka R, et al. Bartonella spp. in bats, Kenya. Emerg Infect Dis. 2010;16(12):1875–81. pmid:21122216; PubMed Central PMCID: PMC3294596.
- 82. Bartonellosis Guptill L. Veterinary microbiology. 2010;140(3–4):347–59. Epub 2009/12/19.
- 83. Guptill L. Feline bartonellosis. The Veterinary clinics of North America Small animal practice. 2010;40(6):1073–90. Epub 2010/10/12. pmid:20933137.
- 84. Pennisi MG, La Camera E, Giacobbe L, Orlandella BM, Lentini V, Zummo S, et al. Molecular detection of Bartonella henselae and Bartonella clarridgeiae in clinical samples of pet cats from Southern Italy. Research in veterinary science. 2010;88(3):379–84. Epub 2009/12/08. pmid:19963231.
- 85. Gutierrez R, Vayssier-Taussat M, Buffet JP, Harrus S. Guidelines for the Isolation, Molecular Detection, and Characterization of Bartonella Species. Vector Borne Zoonotic Dis. 2017;17(1):42–50. Epub 2017/01/06. pmid:28055575.
- 86. Maggi R, Richardson T, Breitschwerdt E, Miller J. Development and validation of a droplet digital PCR assay for the detection and quantification of Bartonella species within human clinical samples. Journal of microbiological methods. 2020;176. pmid:32795640.
- 87. Yang S, Rothman RE. PCR-based diagnostics for infectious diseases: uses, limitations, and future applications in acute-care settings. Lancet Infect Dis. 2004;4(6):337–48. Epub 2004/06/03. pmid:15172342.
- 88. Breitschwerdt EB. Did Bartonella henselae contribute to the deaths of two veterinarians? Parasit Vectors. 2015;8:317. pmid:26062543.
- 89. Velho PE, Pimentel V, Del Negro GM, Okay TS, Diniz PP, Breitschwerdt EB. Severe anemia, panserositis, and cryptogenic hepatitis in an HIV patient infected with Bartonella henselae. Ultrastruct Pathol. 2007;31(6):373–7. pmid:18098054.
- 90. Velho PENF, Bellomo-Brandão MÂ, Drummond MR, Magalhães RF, Hessel G, Barjas-Castro MdL, et al. Bartonella henselae as a putative cause of congenital cholestasis. Revista Instituto de Medicina Tropical de São Paulo. 2016;58. pmid:27410916
- 91. Drummond MR, Visentainer L, Almeida AR, Angerami RN, Aoki FH, Velho PENF. Bartonella henselae bacteremia diagnosed postmortem in a myelodysplastic syndrome patient. Rev Inst Med Trop Sao Paulo. 2019;61:e50. Epub 2019/09/12. pmid:31531628; PubMed Central PMCID: PMC6746197.
- 92. Rodino K, Stone E, Saleh O, Theel E. The Brief Case: Bartonella Henselae Endocarditis-a Case of Delayed Diagnosis. Journal of clinical microbiology. 2019;57(9). pmid:31451567.
- 93. Maguina C, Guerra H, Ventosilla P. Bartonellosis. Clin Dermatol. 2009;27(3):271–80. Epub 2009/04/14. pmid:19362689.
- 94. Florin TA, Zaoutis TE, Zaoutis LB. Beyond cat scratch disease: widening spectrum of Bartonella henselae infection. Pediatrics. 2008;121(5):e1413–25. pmid:18443019.
- 95. Wolf LA, Cherry NA, Maggi RG, Breitschwerdt EB. In Pursuit of a Stealth Pathogen: Laboratory Diagnosis of Bartonellosis. Clinical Microbiology Newsletter. 2014;36(15):33–9.