@article{oliveira_maggi_woods_breitschwerdt_2010, title={Suspected Needle Stick Transmission of Bartonella vinsonii subspecies berkhoffii to a Veterinarian}, volume={24}, ISSN={0891-6640}, url={http://dx.doi.org/10.1111/j.1939-1676.2010.0563.x}, DOI={10.1111/j.1939-1676.2010.0563.x}, abstractNote={A 7-year-old male Newfoundland dog was referred to the North Carolina State University Veterinary Teaching Hospital (NCSU-VTH) for evaluation of multiple, diffuse cutaneous masses. Fine-needle aspiration yielded a population of round cells consistent with a histiocytic, lymphocytic, or highly undifferentiated epithelial neoplasm. Because of the poor prognosis associated with cutaneous disseminated cancers, the owner declined additional diagnostic evaluation and after discharge from the NCSU-VTH elected euthanasia. The dog was very fractious during aspiration of the cutaneous mass, and the veterinarian experienced a needle stick to the right index finger with the 22 G needle used for aspiration. Because ehrlichiosis was diagnosed serologically in the dog 5 months earlier, the Intracellular Pathogens Research Laboratory (IPRL) was consulted relative to the zoonotic risk of needle stick transmission of an Ehrlichia spp. Ehrlichia canis seroreactive dogs residing in the southeastern United States often are concurrently seroreactive to Bartonella vinsonii subsp. berkhoffii antigens and can be coinfected with both organisms,1 therefore the potential transmission of both of these alpha Proteobacteria was investigated. Bartonella and Ehrlichia spp. antibodies were determined by previously described immunofluorescence antibody assays (IFA), with titers ≥1 : 64 considered seroreactive.2 Using the Bartonella alpha Proteobacteria growth medium (BAPGM) diagnostic platform,3,4 polymerase chain reaction (PCR) was performed after direct extraction from blood and serum, after enrichment culture for at least 14 days (typically at 7 and at 14 days) and from subculture colonies, if visualized. Methods used for DNA extraction, conventional PCR targeting of the Bartonella 16S–23S intergenic spacer region, cloning, and sequencing have been described previously.5–8 Sequences were aligned and compared with GenBank sequences with AlignX software.a,5–8 Historically, the veterinarian had always been healthy. During the 1-year period before, and at the time of the needle stick (day 0), the veterinarian reported no headaches, fatigue, or other medical problems. On postinoculation day (PID) 5, when the 1st blood sample was obtained, there was no serological or molecular evidence to support Ehrlichia or Bartonella spp. exposure or infection. Specifically, B. vinsonii subsp. berkhoffii genotypes I, II, and III, Bartonella henselae, and E. canis antibodies were not detected, Ehrlichia spp. DNA was not amplified from the extracted blood sample and Bartonella spp. DNA was not amplified from any of the samples generated as a component of the BAPGM diagnostic platform (Table 1). By PID 34, the veterinarian reported frequent headaches (near daily for the previous week), fatigue, and intermittent paresthesias in the left arm in focal, nondermatomal areas. B. vinsonii subsp. berkhoffii genotype I was amplified and sequenced after direct extraction of DNA from both PID 34 acid-citrate-dextrose and ethylenediaminetetraacetic acid-anticoagulated blood tubes, but PCR from both BAPGM enrichment cultures and subcultures was negative and the patient had not seroconverted to any Bartonella sp. antigen. Repeated blood culture on PID 81 resulted in amplification and sequencing of B. vinsonii subsp. berkhoffii genotype I from the BAPGM enrichment culture and there was an indication that the patient was generating an antibody response to B. vinsonii subsp. berkhoffii genotypes I and III. Subsequent serological testing confirmed seroconversion to B. vinsonii subsp. berkhoffii genotypes I and III (≥4-fold rise in antibody titer). For serum samples obtained on PIDs 97 and 123, antibody titers remained unchanged, but all BAPGM platform PCR results were negative. Because of continued headaches and fatigue, the patient was evaluated by an infectious disease physician. Serum biochemical profile results were within reference ranges and neither B. henselae or Bartonella quintana IgM or IgG antibodies were detected. On PID 138, treatment with doxycycline (100 mg PO q12h) and rifampin (300 mg PO q12h) for 6 weeks was instituted by the attending physician. During the first 2 weeks after starting antibiotics, the patient reported a substantial increase in the severity and frequency of fatigue and headaches, which were accompanied by occasional episodes of dizziness and a single episode of epistaxis. After completion of the antibiotic treatment, all clinical signs resolved and the patient has remained healthy during a 1-year follow-up period. BAPGM platform PCR results obtained on PIDs 144, 180, 206, and 234 were negative, and antibodies to B. vinsonii subsp. berkhoffii genotypes I decreased to the screening dilution of 1 : 16 and genotype III antibodies were no longer detectable (ie, there was no organism fluorescence at the standard screening dilution of 1 : 16) in the patient's serum samples. The patient did not seroconvert to E. canis antigens and Ehrlichia sp. DNA was not amplified from any blood sample. A single tumor aspiration cytology slide represented the only diagnostic sample obtained from the dog at the time of examination. Therefore, cells were retrospectively scraped with a scalpel blade from the Wright-Giemsa stained microscope slide into a sterile vial, DNA was extracted and B. vinsonii subsp. berkhoffii genotype III was amplified and sequenced. Despite multiple cloning attempts, B. vinsonii subsp. berkhoffii genotype I (the genotype detected in the patient's blood on 2 independent occasions) was not sequenced from the extracted cytological specimen. We cannot rule out the possibility that amplification of B. vinsonii subsp. berkhoffii genotype III DNA from the cytology slide represents DNA carryover during slide processing in the laboratory rather than the presence of this bacteria in the dog at the time of cytological sampling.9 B. vinsonii subsp. berkhoffii initially was isolated from a dog with endocarditis in 1993, after which 4 B. vinsonii subsp. berkhoffii 16S–23S intergenic spacer region genotypes were described in dogs, coyotes, foxes, and humans.10 All 4 genotypes have been reported in dogs in association with endocarditis and a spectrum of other clinical manifestations.11 Dogs are the primary reservoir hosts for B. vinsonii subsp. berkhoffii, which is an important emerging zoonotic pathogen.5–7,12,13 The veterinarian in this study experienced a needle stick while obtaining a fine-needle aspiration sample, after which clinical signs, including headaches, fatigue, and intermittent paresthesias, developed. As the veterinarian had not experienced any of these clinical signs previously and because the clinical signs resolved after antibiotic treatment, it seems possible that infection with B. vinsonii subsp. berkhoffii caused or contributed to the illness. The patient seroconverted to B. vinsonii subsp. berkhoffii genotypes I and III and antibody titers to these 2 genotypes persisted between PIDs 97 through 180. As reported previously for B. vinsonii subsp. berkhoffii infection in dogs, this patient became nonseroreactive after being treated with antibiotics.14 As this individual did not produce detectable antibodies to B. vinsonii subsp. berkhoffii genotype II or to B. henselae antigens, there appeared to be a substantial degree of specificity in the serological response after the needle stick; however, serological cross reactivity between genotypes I and III cannot be ruled out. Previous studies from our laboratory indicate that approximately half of human patients infected with a Bartonella sp. do not have detectable IFA antibodies, whereas other patients are broadly reactive against a panel of Bartonella spp. antigens.5,6 As B. vinsonii subsp. berkhoffii genotype IV has not been successfully isolated, serology could not be performed against this genotype.11 In this veterinarian, active infection with B. vinsonii subsp. berkhoffii genotype I was confirmed by amplification and sequencing of DNA directly from 2 blood samples obtained on PID 34 and subsequently from a BAPGM enrichment blood culture on PID 81, which would reflect growth of the bacteria in the liquid culture medium. B. vinsonii subsp. berkhoffii genotype I most often infects dogs and coyotes,1 but has been reported in 2 human patients from the United States.3 Based upon the cumulative testing experience with animal or human patient samples in the IPRL, genotype II has been most frequently amplified and sequenced from blood samples derived from cats, dogs, and people 5–8,12,13 whereas genotypes I or III are only rarely found. Genotype III was first described in a human patient with endocarditis from France 15 and was subsequently isolated from a military working dog that originated from Germany and was subsequently diagnosed with endocarditis while stationed in Texas.16 Genotype III also has been sequenced from gray foxes from California,17 from dogs in Greece and Italy,18 from dogs in China,19 and from feral swine and foxes in the southeastern United States (A. Beard, R. Maggi, and E. Breitschwerdt, unpublished data). Presumably because of technical limitations, we were only able to confirm infection with genotype III in the small quantity of DNA that was obtained from the dog's cytology specimen. Alternatively, genotype III DNA may have been because of carryover in the laboratory.9 Based upon the patient's pattern of seroconversion, it is possible that 2 genotypes were transmitted to the veterinarian by needle inoculation. Alternatively, we cannot rule out the possibility of a switch from genotype III in the dog to genotype I in the patient, perhaps induced by immune pressure on the bacteria. However, the 16S–23S intergenic spacer region does not code for proteins, and in previous studies from our laboratory sequential passage of B. vinsonii subsp. berkhoffii genotype I in BAPGM did not result in a change in genotype.10 In addition, differences in the amino acid sequence-based alignments of the phage-associated protein 31 of B. vinsonii subsp. berkhoffii genotypes I and III 10 also suggest that a switch of genotype is unlikely. Simultaneous infection with more than 1 Bartonella sp. or genotype has been reported in dogs, rodents, and human beings.5,6,11,20 However, molecular documentation of a single Bartonella sp. in diagnostic samples from dogs and human patients remains technically challenging, and documentation of coinfection with more than 1 Bartonella spp. in patient samples is more difficult.18,20 It is becoming increasingly evident that dogs can serve as a source for human infection with B. vinsonii subsp. berkhoffii. Recently, dog bite transmission of B. vinsonii to a child in France was inferred based upon serological testing.21 Also, B. vinsonii subsp. berkhoffii genotypes I and II have been sequenced from dog saliva.22 Although our patient had cat and dog contact, this veterinarian reported no history of animal scratches or bites in the 24 months before or during the follow-up period after the accidental needle stick. Obviously, other modes of transmission (eg, arthropod exposure) are possible, but this individual rarely participated in outdoor activities. A serosurvey from the southeastern United States identified exposure to fleas, ticks, cattle, and a rural environment as risk factors associated with B. vinsonii subsp. berkhoffii seroreactivity in dogs.1 The dog in this report lived in a rural environment, was regularly infested with ticks and fleas, and had historically been treated for infection with an Ehrlichia sp. In dogs experimentally infected with E. canis, there was no serological cross-reactivity with B. vinsonii subsp. berkhoffii antigens.1 As B. vinsonii subsp. berkhoffii antibodies were found in 36–42% of E. canis seroreactors,1 we investigated the potential transmission of both organisms to this veterinarian. Five days after the needle stick, the veterinarian remained asymptomatic and there was no serologic, microbiologic, or molecular evidence of Bartonella spp. infection. Subsequently, headaches, numbness, and parethesias in the extremities developed. These clinical signs have been described previously in a small number of immunocompetent persons infected with B. vinsonii subsp. berkhoffii or B. henselae.5–7 After initiation of antibiotic treatment, this patient had 1 episode of epistaxis, a rarely reported clinical sign in cat scratch disease patients and a clinical sign associated with bartonellosis in dogs.23 If the mode of transmission was because of the needle stick, this patient required approximately 12 weeks before seroconversion was detected. Although seroconversion may have been delayed because of a low-dose inoculum, there are minimal data that define the pattern of seroconversion after transmission of any Bartonella spp. to a human. In this patient, we were fortunate to have access to 9 serum samples, spanning PID 5 through PID 234, which were all retested blindly by the same research technician at the same time, and with the same conjugate and antigen preparations. This was done to generate the most consistent serological comparisons possible. Because all known Bartonella spp. are vector-transmitted blood-borne pathogens, inadvertent transmission by needle stick or by blood transfusion seems plausible. B. henselae was successfully isolated from units of human blood after experimental inoculation and after storage at 4°C for 35 days.24 On an ultrastructural basis, the same research group implicated infection with a Bartonella spp. in an aplastic anemia patient, who died after receiving multiple red blood cell transfusions.25 At the NCSU-VTH, canine and feline blood donors are routinely screened using the BAPGM platform for evidence of Bartonella sp. infection. As evolving evidence indicates that there are numerous Bartonella spp. that have coevolved as highly adapted intravascular and intraerythrocytic organisms in pets and numerous wildlife species, animal health professionals should exercise special caution when obtaining, handling, or processing diagnostic specimens derived from animals. In conclusion, the clinical history as well as serological and molecular microbiological findings suggest the possibility of needle stick transmission of B. vinsonii subsp. berkhoffii genotypes I and III to this veterinarian. After administration of antibiotics, all clinical signs resolved, B. vinsonii subsp. berkhoffii antibodies decreased to nondiagnostic levels and repeated enrichment blood cultures were PCR negative. a Vector NTI Suite 6.0, InforMax Inc, Bethesda, MD We thank Julie Bradley for serological testing, Barbara Hegarty for preparation of Ehrlichia and Bartonella spp. antigens, and Tonya Lee for editorial assistance. This research was supported in part by a grant from the American College of Veterinary Internal Medicine Foundation and by the State of North Carolina. The data in this report were not presented previously. Disclosure : In conjunction with Dr Sushama Sontakke and North Carolina State University, Dr Breitschwerdt holds US Patent No. 7,115,385; Media and Methods for Cultivation of Microorganisms, which was issued October 3, 2006. He is the chief scientific officer for Galaxy Diagnostics, a newly formed company that provides diagnostic testing for the detection of Bartonella species infection in animals and in human patient samples. Dr Ricardo Maggi is the Scientific Technical Advisor and Laboratory Director for Galaxy Dx.}, number={5}, journal={Journal of Veterinary Internal Medicine}, publisher={Wiley}, author={Oliveira, A.M. and Maggi, R.G. and Woods, C.W. and Breitschwerdt, E.B.}, year={2010}, month={Aug}, pages={1229–1232} }