Careful microbiologic evaluation of patients with various illnesses has led to the discovery of many important pathogens in recent decades, including human immunodeficiency virus (HIV), legionella species, Borrelia burgdorferi (the agent of Lyme disease), human herpesvirus 8 (HHV-8), and numerous others.

Success in these endeavors, however, was critically dependent on the availability of the appropriate technology for both the detection of the microorganism and its characterization to the level necessary to permit clear differentiation from already recognized pathogens.

The delay between the recognition of a particular clinical syndrome and the identification of its causative agent has been highly variable. Whereas HIV, for example, was discovered within 2 to 3 years after the recognition of AIDS, it took more than 120 years to establish that HHV-8 was the cause of Kaposi’s sarcoma.

Bartonella are small, curved, pleomorphic, gram-negative rods. A characteristic feature of these bacteria is their adherence to and invasion of erythrocytes, although this phenomenon is dependent on the erythrocytes’ species of origin. A unique facet of infection with bartonella is the ability of these microorganisms to stimulate neovascular proliferation in tissues, presumably by causing endothelial-cell proliferation and migration. Although highly fastidious, bartonella are often cultivable, and available methods for analyzing the genetic and protein compositions of the isolated microorganism permit very precise molecular characterization. Having used such an approach, Eremeeva et al. present compelling evidence in this issue of the Journal (pages 2381–2387) that a new bartonella species, Bartonella rochalimae sp. nov., should be added to the list of recognized human pathogens.

Of the 19 recognized and extant species and subspecies in the expanding bartonella genus before the report by Eremeeva et al., perhaps 9 had been linked to human infections, but only 3 of them had been implicated in such infections frequently. The spectrum of clinical illness varies with the species causing the infection, but even among patients infected with the same species, the clinical features can be surprisingly variable. At times, the clinical illness caused by these microorganisms is so distinctive that bartonella infection would be at or near the top of the differential diagnosis, whereas in other patients the presentation is completely nonspecific.

B. henselae is now regarded as the principal cause of cat scratch disease, the most frequently recognized bartonella infection in humans. The cause of cat scratch disease was not conclusively elucidated until more than 40 years after its recognition as a clinical entity in 1950. The hallmark of this infection is the prominent enlargement of lymph nodes that drain lymph from cutaneous sites where B. henselae was introduced by the bite or scratch of a cat (the reservoir for this bartonella species) or possibly by the bite of a cat’s fleas. In the majority of cases, either a papule or pustule or residual evidence of a bite or scratch remains visible, though the skin site must be examined carefully to discern it.

B. quintana, transmitted by the human body louse (Pediculus humanus), is the cause of a relapsing febrile illness associated with prominent limb pain. The term “trench fever” was first applied to this condition in a publication describing an outbreak among British military personnel during World War I. Because of its association with human body lice, infection with this bartonella species is closely associated with homelessness in urban areas and with poor personal hygiene.

Both B. quintana and B. henselae are causes of bacillary angiomatosis, which consists of small lesions showing histologic evidence of neovascular proliferation. These lesions are typically found on the skin but can also involve regional lymph nodes and a variety of internal organs, occurring especially (but not exclusively) in immunocompromised patients, such as those with AIDS. Both of these bartonella species are also an increasingly appreciated cause of culture-negative subacute bacterial endocarditis (i.e., cases in which blood cultures are negative for conventional pathogens). Both can cause febrile illnesses in patients with AIDS, but only B. henselae has been linked to bacillary peliosis, an unusual disorder characterized by the development of numerous blood-filled cystic structures as large as several millimeters in diameter; these infected lesions typically occur in the liver, spleen, or lymph nodes of patients with AIDS or other highly immunocompromised patients.

Source Information

Dr. Wormser is chief of the Division of Infectious Diseases at New York Medical College, Valhalla.

Gary P. Wormser, M.D.

Full text