Bartonella spp. and Ticks | CDC EID

EID Journal Home > Volume 16, Number 3–March 2010

Volume 16, Number 3–March 2010
Perspective
Bartonella spp. Transmission by Ticks Not Established
Sam R. Telford III and Gary P. Wormser
Author affiliations: Tufts University Cummings School of Veterinary Medicine, North Grafton, Massachussetts, USA (S.R. Telford III); and New York Medical College, Valhalla, New York, USA (G.P. Wormser)

Suggested citation for this article

Abstract
Bartonella spp. infect humans and many animal species. Mainly because PCR studies have demonstrated Bartonella DNA in ticks, some healthcare providers believe that these microorganisms are transmitted by ticks. B. henselae, in particular, is regarded as being present in and transmissible by the Ixodes scapularis tick. The presence of a microbial agent within a tick, however, does not imply that the tick might transmit it during the course of blood feeding and does not confer epidemiologic importance. After a critical review of the evidence for and against tick transmission, we conclude that transmission of any Bartonella spp. by ticks, to animals or humans, has not been established. We are unaware of any well-documented case of B. henselae transmission by I. scapularis ticks.
Infections with Bartonella spp. appear to be widespread in many animal species besides cats (1). Some evidence has been advanced in support of the possibility of tick transmission. Such findings have resulted in diagnostic testing and empiric therapies directed at B. henselae infection that are of dubious value with respect to illnesses thought to be caused by deer tick exposure. We critically examined the reported findings regarding tick transmission of Bartonella spp.

Bartonella spp. are common bacterial hemoparasites of mammals; for as long as 100 years, 2 species have been known to cause infections of public health significance. Trench fever, caused by B. quintana (formerly Rochalimaea quintana) and transmitted by body lice, affected hundreds of thousands of soldiers or displaced persons during World War I and to this day affects homeless persons. Oroya fever (and its chronic manifestation verruga peruana), caused by infection with B. bacilliformis and transmitted by phlebotomine sandflies, is a potentially severe febrile disease. Although it is geographically restricted to the high altitudes of the Andes and affects only a relatively small number of persons, the high case-fatality rate brought attention to this apparent anthroponosis as early as the late 1800s.

B. henselae causes cat-scratch disease, the most common Bartonella spp. infection in the United States (2). The hallmark of cat-scratch disease is enlargement and tenderness of lymph nodes draining the site of inoculation of the microorganism (3). In addition, a skin or mucous membrane lesion may be observed at the site of inoculation for 25% to >90% of patients (3,4). Extranodal clinical manifestations (e.g., encephalopathy, neuroretinitis, arthritis, and lytic bone lesions) occur in ≈10% of patients (3–6). Cats are the main reservoir of B. henselae. In a study from San Francisco, 25 (41%) of 61 pet, pound, or stray cats (Felis domesticus) were found to have B. henselae bacteremia (7). Bites or scratches from infected cats are associated with development of cat-scratch disease. The gut of cat fleas is commonly infected, and exposure to feces of infected fleas is the presumed route of transmission to uninfected cats and a possible route of transmission to humans.

Parasitologists focusing on blood parasites have long noted the ubiquity of Bartonella spp. within mammals, particularly rodents, and by the late 1960s nearly 2 dozen species had been described within the genus Grahamella (8). The genera Rochalimaea and Grahamella were subsumed into the genus Bartonella (9), and many of the validly published Grahamella spp. have been excluded from the list of approved bacterial taxa (10). These actions tended to foster ignorance of the history of the diversity of Bartonella spp. and to promote a fallacy in pathogen discovery (11); namely, if a DNA sequence is not present in GenBank, surely it must represent something novel, the extensive classical literature on a likely identical organism known only by morphology notwithstanding. The significance of such a fallacy is that a large body of literature that may provide critical details on the biology of a "novel" agent is completely overlooked or dismissed.

Vector Relationships
Seminal studies by Richard Pearson Strong and the members of the American Red Cross trench fever commission (12) conclusively demonstrated biological as opposed to mechanical transmission of the trench fever agent by body lice. Feeding experiments on human volunteers established that lice may transmit by bite or by fecal contamination of abraded skin; that an infected louse remains infectious for at least 2 weeks; that the agent is not inherited by the progeny of infected lice; and that transmission may be extremely efficient, causing trench fever in 75% of volunteers after 1 exposure to a feeding box containing ≈50 lice that had previously fed on patients with trench fever.

Although initially Oroya fever was epidemiologically associated with ticks (13), it rapidly became evident that phlebotomine sandflies (particularly Lutzomyia verrucarum) were the vectors. Sandflies were the only blood-feeding arthropods that were peridomestic in their habits and occurred in the "bartonella zone," >2,000 m elevation. Experimentally, sandflies acquired infection from blood-smear positive patients and transmitted infection by bite to those without evidence of Bartonella spp. infection (14).

Grahamellae (now bartonellae) of rodents have long been known to be transmitted by fleas (15–17). Such studies have noted the difficulty with which experimental infections may be established by means other than inoculation of flea homogenates, the persistence within the rectal sac of the flea, and the likely mode of perpetuation of the bacteria by larval fleas ingesting dried infected blood. In addition, grahamellae-infected rodents were noted to exist in the absence of ticks, demonstrating that ticks were not required to perpetuate these particular bacteria.

Ticks as Vectors
Ticks are notorious vectors of a variety of agents that cause zoonotic infections (11), including viruses, bacteria, and protozoans. Like all animals, ticks have a diverse microflora. Recent analyses, using cloning and sequencing broad-range 16S rDNA amplification products, have documented a large bacterial flora within northeastern populations of Ixodes scapularis ticks that bite humans as nymphs, hereafter referred to as deer ticks (18,19). Amebas, mycoplasma, fungi, and helminths have been detected in these ticks by microscopy or other standard methods. However, the presence of a microbial agent within a tick does not imply that the tick might transmit it during the course of blood feeding or that it is pathogenic.

During early investigations of the causes of Oroya fever, Noguchi (20) demonstrated that B. bacilliformis could be experimentally transmitted between monkeys by the bites of Dermacentor andersoni ticks. However, the ticks that had been fed for a few days on infected monkeys were removed and allowed to reattach and complete their blood meal on uninfected animals, which became infected. Noguchi concluded that mechanical transmission had been demonstrated (perhaps by contamination of mouthparts or by regurgitation of the infectious partial blood meal), but persistence of viable bacteria or transstadial passage had not, and thus ticks were not biologic vectors.

Based on the volume of studies, the most compelling argument in favor of a tick vector for Bartonella spp. is that these microorganisms are sometimes detected in field-collected ticks (Table 1) (15). Although at least 20 studies have provided evidence for the presence of Bartonella spp. in primarily Ixodes spp. ticks collected at various locations in the United States and Europe, only 1 study has confirmed the presence of Bartonella spp. by culture (15,21,22). Caution is warranted when interpreting such data, however, because acquisition of Bartonella spp. from animal sources through a blood meal would be anticipated given the ubiquity of the microorganism in domestic animals and wildlife. In New England, as many as 60% of white-footed mice are blood-smear positive for Grahamella spp. (now Bartonella), regardless of collection site, including those trapped within the house of 1 of the authors where a tick life cycle was not present (S.R. Telford III, unpub. data); prevalence would probably reach unity if more sensitive modes of detection were used. The mere presence of Bartonella spp. or their DNA in ticks does not prove vector competence or confer epidemiologic significance (15), but it should serve as the impetus to rigorously perform the studies necessary to establish vector competence of ticks. At the least, viability should be established for bartonellae detected within ticks by means of in vitro cultivation.

To date, no report has documented transmission of B. henselae or any other Bartonella spp. to an animal after a tick bite (Table 2). The strongest evidence that ticks might be competent vectors for bartonellae was reported in a recent study in which I. ricinus ticks were infected with B. henselae in spiked (artificially infected) ovine blood by using an artificial feeding system (23). The ticks maintained infection throughout the molt, thereby establishing transstadial transmission. The experimentally infected ticks were also able to transmit B. henselae during a subsequent blood meal, again through the artificial feeding system; the dissected salivary glands from such ticks, when introduced into a cat, produced typical B. henselae infection, proving viability. Serious questions exist, however, as to whether these experiments are relevant to establishing vector competence. The ticks were fed continuously on blood meals with 109 CFU/mL, representing a bacteremia that would rarely be seen in natural infections of cats. Given that Ixodes spp. nymphs ingest a total of ≈15 μL blood (24), each nymph may have ingested 106–107 bacteria, a large dose. In addition, the Houston-1 strain of B. henselae used in this study may not represent strains found in nature. It is highly adapted to the laboratory and readily grows in vitro, whereas primary isolates are extremely fastidious and grow slowly.

A more straightforward experiment to establish vector competence would be to feed an uninfected Ixodes sp. tick on a B. henselae–infected cat and then, after the tick has molted, determine whether B. henselae can be transmitted by tick bite to an uninfected cat. However, even if such an experiment were to prove vector competence, additional data would be needed to conclude that Ixodes spp. ticks are epidemiologically relevant as B. henselae vectors.

Do epidemiologic data that support tick transmission of Bartonella spp. in animals exist? One study correlated canine seropositivity to B. vinsonii subsp. berkhoffii with tick exposure and with seropositivity to other tick-borne pathogens (25). However, the dogs in that study were also heavily exposed to fleas, and according to findings with cats, flea transmission is as likely a possibility as tick transmission in dogs, if not more so (15,25,26). A study in the United Kingdom reported an association between seropositivity to B. henselae and to Borrelia burgdorferi in feral cats (27). The method used to detect antibodies to B. burgdorferi was not precisely described. However, the fact that the rate of seropositivity to B. henselae was nearly the same for domestic and feral cats, despite domestic cats having much less tick exposure than feral cats, raises questions about the epidemiologic relevance of tick transmission. In another study, a "novel" Bartonella subspecies was detected more often in white-footed mice concurrently infected with the tick-borne pathogens B. burgdorferi or Babesia microti (1), but this analysis failed to compare the likelihood that the Bartonella spp. might also commonly co-occur with rodent trypanosomes, which are maintained by fleas. Epidemiologic arguments must carefully control for confounding, and none to date argues convincingly for tick transmission of Bartonella spp.

abrir aquí para acceder al documento CDC EID completo:
Bartonella spp. and Ticks | CDC EID


Suggested Citation for this Article
Telford SR III, Wormser GP. Bartonella spp. transmission by ticks not established. Emerg Infect Dis [serial on the Internet]. 2010 Mar [date cited]. http://www.cdc.gov/EID/content/16/3/379.htm

DOI: 10.3201/eid1603.090443